Docstoc

CAPTION project final report

Document Sample
CAPTION project final report Powered By Docstoc
					     ‘COMMUNITY-ACQUIRED PNEUMONIA: TOWARDS
         IMPROVING OUTCOMES NATIONALLY’


      The CAPTION Project - A Multi-centre Drug Usage
                    Evaluation Study


                  Final report to the National Prescribing Service

                                    April 2006

            Prepared by the Victorian Drug Usage Evaluation Group




Project Leader:      Ms Marion B Robertson, Royal Melbourne Hospital

Project Officers:    Dr Kylie A McIntosh (Feb ’04 – current)
                     Dr Nazila Jamshidi (Oct ’03 – Oct ’04)
CAPTION Victoria – Final Report                                                                                  April, 2006




Table of Contents

1. Executive Summary ......................................................................................................... 3
2. Introduction ...................................................................................................................... 6
3. Methodology .................................................................................................................... 8
     3.1 State Consultative Process.................................................................................... 8
     3.2 Recruitment of Hospitals........................................................................................ 8
     3.3 Baseline data collection ......................................................................................... 9
     3.4 Academic detailing training .................................................................................... 9
     3.5 Intervention strategy and tools............................................................................. 10
     3.6 Data management ............................................................................................... 11
     3.7 Data analysis and evaluation ............................................................................... 11
     3.8 Communication strategy ...................................................................................... 12
4. Results........................................................................................................................... 13
     4.1 Hospital recruitment ............................................................................................. 13
     4.2 ED doctors’ survey............................................................................................... 13
     4.3 Audits ................................................................................................................... 13
     4.4 Interventions ........................................................................................................ 14
5. Discussion ..................................................................................................................... 16
     5.1 Main findings and limitations................................................................................ 16
     5.2 Recruitment and start-up ..................................................................................... 17
     5.3 Audits ................................................................................................................... 17
     5.4 Interventions and key messages ......................................................................... 17
     5.5 Project management and evaluation ................................................................... 18
6. Conclusion ..................................................................................................................... 19
7. Recommendations......................................................................................................... 20
     7.1 Support of hospitals post-CAPTION .................................................................... 20
     7.2 Multi-centre project work...................................................................................... 20
     7.3 Future national DUE activities.............................................................................. 20
List of Tables ........................................................................................................................ 21
Abbreviations........................................................................................................................ 43
Acknowledgements .............................................................................................................. 44
References ........................................................................................................................... 45
List of Appendices ................................................................................................................ 47




                                                                                                                                         2
CAPTION Victoria – Final Report                                              April, 2006


1.    Executive Summary

Background
As a result of both the low concordance with the Therapeutic Guidelines: Antibiotic (11th
version) for lower respiratory tract infections observed in the 2001 NPS-funded hospital study
by the VDUEG, and the release in 2003 of the 12th revision of the national antibiotic
guidelines (Guidelines), the NPS commissioned the Victorian DUE Group along with NSW,
QLD, SA and Tasmanian DUE groups to undertake a project to implement recommendations
of the new Guidelines in the emergency departments (EDs) of Australian hospitals. The
CAPTION (Community-Acquired Pneumonia: Towards Improving Outcomes Nationally)
study was carried out by the VDUEG in Victoria.

Objectives
• To introduce and implement the CAP recommendations of the Guidelines in Victorian
  hospital EDs
• To train health professionals in the use of social marketing techniques to influence and
  improve prescribing practices for CAP, as one of a suite of interventions

Methods
A total of 17 Victorian hospitals were approached to volunteer to participate in CAPTION.
After initial contact, formal letters of invitation were sent via the hospital Chief Executive
Officer (CEO) and sign-off for the project was required from the CEO and the ED Director.
The responsibility for obtaining approval (or exemption from approval requirements) from the
local Human Research and Ethics Committee lay with the participating hospital(s). Project
teams were formed within each hospital and a person/persons were nominated to coordinate
CAPTION at an institutional level and liaise with the state project officer.

The national project team, in consultation with stakeholders at both a state and national level,
highlighted four key messages for CAPTION –

      1.    “Use a systematic approach to assessing severity”
      2.    “Select antibiotic therapy according to severity”
      3.    “Penicillins are first choice for non-severe CAP”
      4.    “Consider atypical pneumonias when deciding on antibiotic regimen”

These key messages were incorporated into educational intervention tools developed for use
in participating hospitals. The suite of intervention tools included:

1.   An academic detailing card and accompanying script. The card was used in all
     detailing sessions and presented to the healthcare professional being detailed at the
     completion of the session as a gift. The script provided additional supportive material
     and was for use by the detailer only.
2.   Laminated A3 size posters describing calculation of the Pneumonia Severity Index
     (PSI) and the antibiotics recommended for each of the five PSI Classes by the
     Guidelines. Posters were typically displayed in the ED.
3.   PSI calculator stickers of A5 size for use in patient histories. Stickers were left in the
     ED to facilitate ready determination of the PSI score for CAP patients.
4.   Laminated ID card size prompts for PSI calculation and antibiotic recommendations.
     These were distributed to ED medical staff by the hospital coordinators.
5.   A generic letter to prescribers describing the CAPTION project was provided to hospital
     coordinators for use in informing prescribers outside of the ED e.g. medical registrars,
     respiratory and infectious diseases (ID) physicians etc.


                                                                                              3
CAPTION Victoria – Final Report                                                           April, 2006


6.    A generic PowerPoint feedback presentation was developed for use by hospital
      coordinators to inform hospital staff of audit and education results at their hospital and
      also to provide feedback on state and national data from CAPTION.
7.    Automated feedback reports were also available from the Auditmaker software. These
      were used to assist local project coordinators adapt the generic feedback presentation
      for use on site.

Training in academic detailing was provided to at least one representative from each
participating hospital in August 2004 in Melbourne, Victoria. Of the eleven Victorian hospital
staff trained, seven were doctors, three were pharmacists and one was a clinical research
nurse. All hospitals intended to use academic detailing as an intervention in the study
although one was later unable to implement this owing to a conflict of interest with a local
project. Hospital coordinators together with other local team members conducted all the
interventions at a local level, although the state project officer provided some assistance with
academic detailing at three sites.

Prior to the commencement of the project in Victorian hospitals a survey of ED doctors was
carried out in order to ascertain the level of awareness of the current Guidelines and CAP
management recommendations. The survey was also used to determine the doctors’
preferred educational intervention methods for CAP management.

A baseline audit (Audit I) was performed to gauge current CAP management practices at
each of the hospitals. Data were collected on 20 adult CAP patients presenting to the
hospital ED at each site. Information from the first (baseline) audit was utilised to assist in the
intervention phase(s) of the project. Two intervention phases (Intervention Phase I,
Intervention Phase II) and subsequent follow-up audits (Audit II, Audit III) were carried out in
Victoria. All hospitals participated in these activities. Data collected included patient
demographics, penicillin allergies, results of relevant investigations/examinations such that
the PSI could be calculated, antibiotics prescribed (drug, dose, route of administration),
inpatient mortality and length of stay.

Effectiveness of the interventions utilised in the project was measured by determining the
proportion of patients who were prescribed antibiotics that were concordant with the
Guidelines for drug and route of administration1, the proportion of patients who were
prescribed antibiotics that adequately covered the likely causative organisms of CAP, and
the proportion of patients with a documented PSI score in their medical record. ‘Adequate’
antibiotic therapy included antibiotic regimens that were concordant with the Guidelines,
antibiotic regimens that included an intravenous beta-lactam instead of oral amoxycillin for
Class I and II patients and antibiotic regimens for Class III and IV patients that included IV
ceftriaxone or cefotaxime for non-penicillin allergic patients.

Results
A total of eight hospitals took part in the project in Victoria; these included seven major
metropolitan public hospitals and one metropolitan private hospital. All hospitals participated
in the three audits and the two intervention phases.

The survey of prescribers in the ED indicated that most (74%) were aware of the Guidelines’
recommendations surrounding antibiotic prescribing for CAP and the use of the PSI. Two
CAP case studies were described in the survey, with doctors’ responses for suggested
treatment indicating that concordance with the Guidelines was approximately 45-50%. When
asked what form of interventions they would prefer to assist prescribing in CAP, the most
popular was a laminated PSI scoring ID card (36%), followed by improved access to

1
 Intravenous azithromycin was also considered an acceptable alternative to intravenous erythromycin in Class V
patients.
                                                                                                                 4
CAPTION Victoria – Final Report                                                April, 2006


Therapeutic Guidelines (22%), a PSI calculator form (21%), a web-based antibiotic approval
system (12%) and academic detailing (10%).

The baseline audit (Audit I) comprised 160 patients, with 20 patient presentations contributed
to the audit from each of the eight participating hospitals. Although the inclusion criteria
specified only a presumptive admission diagnosis of CAP, audit data throughout the study
revealed that approximately 85% of all patient presentations had X-ray confirmation of
pneumonia documented. ED doctors were typically responsible for around 80% of
prescribing, with antibiotics most frequently prescribed by registrars. Concordance of
antibiotic prescribing with the Guidelines and documentation of the PSI were determined
from Audit I data to be 22% and 6%, respectively. Adequate empiric cover for CAP was
prescribed to 69% of patients audited.

Concordance with the Guidelines improved in Audit II (n=160) by approximately one third
following the first intervention phase to 29%, although this difference was not statistically
significant (p=0.124). Documentation of the PSI was significantly increased almost four-fold
to 22% (p<0.0001). Both indicators remained at a similar level in the final audit (Audit III,
n=147), which was performed following the second intervention phase, with concordance
determined to be 29% (p=0.138) and PSI documented in 18% (p=0.0005) of CAP patient
presentations in Audit III.

For the first intervention phase all hospitals employed the generic feedback PowerPoint
presentation for group sessions together with the posters and ID cards. Seven out of the
eight hospitals employed academic detailing as an intervention and also the PSI stickers. A
total of 359 staff members attended group education sessions, while academic detailing
sessions (1:1) were conducted with 304 staff members. The number of interventions in the
second intervention phase was markedly less than the first.

Conclusions
Despite an initial survey of ED doctors indicating a good awareness of the Guidelines’
recommendations regarding prescribing for CAP and the use of the PSI, baseline audit data
for CAPTION in Victoria demonstrated a fairly low concordance with the Guidelines and
minimal documentation of PSI scores. Participating hospitals subsequently implemented a
suite of educational interventions, including academic detailing, to promote improved CAP
management in hospital EDs. Follow-up audits showed improvements in both concordance
and PSI score documentation although only the increase in PSI score documentation was
statistically significant. Given that a range of interventions was introduced at the same time, it
is difficult to attribute any changes to one intervention in particular. The sustainability of the
project and its key messages should be considered especially in light of the imminent
publication of the updated Guidelines.




                                                                                                5
CAPTION Victoria – Final Report                                               April, 2006


2.    Introduction
Respiratory tract infections, including community-acquired pneumonia (CAP), have been
estimated to represent 10% of global morbidity and mortality (Ball et al, 2002). While little
Australian data are available on the incidence of CAP, its occurrence has been estimated at
approximately two per one thousand of the adult population per year and that CAP
represents 2% of overnight hospital admissions (Tsirgiotis and Ruffin, 2000). Mortality
associated with CAP hospital admissions ranges from 7-10%, a figure rising to 40% for
patients admitted to an intensive care unit (Niederman et al, 2001).

There is considerable evidence that prescribing of antibiotics for lower respiratory tract
infections (LRTI) is varied and inconsistent with national guidelines (McKellar, 2005; Victorian
Drug Usage Evaluation Group (VDUEG), 2003; Lawford, 2003; Robertson et al, 2002;
Dobbin et al, 2001; Radford et al, 1999; Paterson and Playford, 1998). The 2001 VDUEG
study investigating the antibiotic treatment of LRTI in hospital emergency departments found
that the concordance of antibiotic therapy with the recommendations of the Therapeutic
Guidelines: Antibiotic, version 11, was only 7% (Victorian Drug Usage Evaluation Group,
2003). Of the 603 patients enrolled in the study, 59% were diagnosed with CAP, with disease
severity ranging from mild-moderate to severe. Reasons for discordant antibiotic therapy
included inappropriate use of the broad-spectrum cephalosporins, ceftriaxone and
cefotaxime, and the use of oral rather than intravenous antibiotics in severe CAP.

In view of these findings and the release in 2003 of a new edition of the Therapeutic
Guidelines: Antibiotic (version 12, ‘Guidelines’), the NPS agreed to fund and support a
national initiative, ‘CAPTION’, to promote and implement the updated Guidelines in 40
Australian hospital emergency departments, with the VDUEG engaged to perform this project
in Victoria. The other state-based groups engaged in the study were the New South Wales
Therapeutics Advisory Group (NSW TAG), South Australian Department of Human Services,
the University of Tasmania and the University of Queensland/Queensland Health.

The new Guidelines recommended the use of a new CAP severity assessment tool, the
Pneumonia Severity Index (PSI), a scoring system which was developed by an American
physician in the mid to late 1990s (Fine et al, 1997). Based on their PSI score, patients were
assigned to one of five classes, ranging from class I (mild CAP) to class V (severe CAP) and
treated accordingly; for example, a class I patient may generally be discharged home on oral
antibiotics, or a class V patient may be admitted to an intensive care unit and given a course
of intravenous antibiotic therapy. While concerns have been raised regarding the limitations
of the PSI (Buising et al, 2004), data emerging from the ongoing Australian Community-
Acquired Pneumonia Study (ACAPS) indicates that the PSI is superior to either of the British
Thoracic Society’s or modified American Thoracic Society’s criteria in predicting CAP
Severity (Charles et al, 2005). Importantly, it should also be noted that the Guidelines advise
that the PSI is a ‘guide only’ and that clinical judgment remains paramount.

The Guidelines recommend increasing antibiotic coverage with increasing disease severity
(i.e. higher PSI classes) and include the use of dual therapy (a beta-lactam and atypical
cover) for all CAP patients. The empiric antibiotic recommendations of the Guidelines are
based on the likely key bacterial pathogens: Streptococcus pneumoniae, Mycoplasma
pneumoniae, Chlamydophila pneumoniae and Legionella species. Patients with severe
disease may require broad-spectrum antibiotic therapy to cover Legionella pneumophila,
Staphylococcus aureus and enteric Gram-negative bacilli. Where possible however, empiric
therapy is restricted to narrow spectrum antibiotics, such as amoxycillin or benzylpenicillin, in
order to prevent the ‘collateral damage’ associated with the inappropriate overuse of broad
spectrum agents such as the broad-spectrum cephalosporins (Paterson, 2004 and Dancer,
2001).

                                                                                               6
CAPTION Victoria – Final Report                                             April, 2006


It was agreed by the national project group that the CAP recommendations of the Guidelines
should be implemented by CAPTION through the use of a suite of educational intervention
tools, including academic detailing. The objectives agreed on for the project in Victoria were
therefore:

• to introduce and implement the CAP recommendations of the Guidelines in Victorian
  hospital EDs
• to train health professionals in the use of social marketing techniques (academic detailing)
  to influence and improve prescribing practices for CAP, as one of a suite of interventions




                                                                                            7
CAPTION Victoria – Final Report                                                April, 2006


3.    Methodology
3.1 State Consultative Process
A Consultative Committee (Appendix 1) was developed following invitations for expressions
of interest in the proposed project to clinicians from pharmacy, infectious diseases,
respiratory medicine and emergency medicine backgrounds. A meeting of this newly formed
committee was held in October 2003. The committee discussed possible intervention
strategies that could be developed to implement the Guidelines, the methodology of hospital
recruitment and a steering group was formed to serve as the Project Committee (Appendix
1). It was decided that the Consultative Committee would serve as a sounding board for the
Project Committee on intervention processes developed for the project. Documents such as
the project plan and intervention tools were circulated to the Consultative Group for
comments before being finalised.

The Project Committee met as a group for the first time in November 2003. At this meeting
the role of the Project Committee was defined - the group agreed to:
•    Discuss and approve the project plan,
•    Consider details of audits, academic detailing and intervention tools devised by the
project officer,
•    Approve correspondence for distributions to hospitals and individuals targeted for
recruitment,
•    Oversee the co-ordination of the project in Victoria and its collaboration with other states,
•    Advise the project officer and endorse the activities of the project officer.
The committee then agreed on the project’s objective and a set of outcome measures. A
project description (Appendix 2) and plan (separate attachment) were developed along with
letters for the recruitment of EDs (Appendix 3), the ED response form (Appendix 4), letters to
the pharmacy departments (Appendix 5) and letters for the CEO (Appendix 6) to complete in
order to initiate the recruitment process.


3.2 Recruitment of Hospitals
Hospitals were targeted for recruitment on the basis that they had an ED, that they were
interested in the project and that they agreed with the recommendations of the Guidelines.
Altogether, nineteen hospitals were approached to take part in VicCAPTION, including
medium and large public metropolitan hospitals, two private metropolitan hospitals and
several rural hospitals.

Sign-off on the project was required from hospital CEOs and ED directors and the
responsibility for seeking and obtaining approval/endorsement for CAPTION from the local
Human Research and Ethics Committee was assigned to the individual hospitals. A project
starter pack was sent to each hospital and contained:

- a copy of the project description (Appendix 2)
- the project plan with a tentative time-line of major project events (separate attachment)
- a standard ethics letter that could be edited and submitted to the local ethics committee if
required (Appendix 7)
- a form for the ED director and for the hospital CEO to sign as approval to take part in the
project (Appendix 4, 6)
- a suggested letter from the ED director to the person recruited as the local project officer
(Appendix 8)
- the package on a diskette for the hospital to adapt according to the needs of the ethics
and/or drug and therapeutics committee.



                                                                                                8
CAPTION Victoria – Final Report                                                April, 2006


It was suggested that each hospital should form a CAPTION team led by the nominated
hospital coordinator. The suggested composition of the team was a clinical champion (a
consultant or senior registrar from ED, ID or respiratory) and one or two local project officers
who may be an ED doctor or pharmacist, a DUE pharmacist or a clinical nurse educator, who
may share the role. The role of clinical champion and the role of the project officer may be
undertaken by the same person in the smaller hospitals.


3.3 Baseline data collection
The detailed information contained within the CAPTION Project Manual for Hospitals
(“Explanatory notes for BASELINE data collection”, pg 22-24) was promoted to Victorian
hospital coordinators. Briefly, data collection for Audit I (baseline) was to be undertaken
retrospectively, with twenty consecutive ED CAP presentations to be identified.

It was suggested that identification of patients might be accomplished using existing
electronic reporting systems within the ED such as the HASS system available in Victoria.
Alternative suggested methods of identifying patients included liaising directly with relevant
ED staff members e.g. health information manager, ward clerk etc to obtain reports from
specific computer systems utilised within individual hospital EDs and the use of ICD-10
codes. The relevant codes required for this purpose are: J13, J14, J15, J16, J17.8 and J18.
Patient histories were reviewed to determine whether the project eligibility criteria had been
met. Inclusion and exclusion criteria are described in Tables 2 and 3, respectively. A data
collection form (available from the CAPTION Project Manual for Hospitals) was to be
completed for all patients fulfilling the inclusion criteria. One Victorian hospital was involved
in piloting the data collection form.

A coding system was developed to de-identify individual patients to ensure confidentiality.
Each hospital was allocated a two-letter code, beginning with the first letter ‘A’ for Victorian
hospitals. A coding form (Appendix 9) was provided to hospital coordinators for the audit to
link patient medical record numbers with CAPTION study numbers. This form was not to be
forwarded to the state project officer, rather to be kept on site at each hospital.


3.4 Academic detailing training
The state project officers and the project leader attended a two-day academic detailing
training workshop, funded by the NPS, in Sydney in June 2004. The purpose of this
workshop was to assist in the development of the academic detailing training package to be
used for the local project officers nationally and to equip the state project officers and leaders
with skills in academic detailing.

An academic detailing training workshop was subsequently held in Victoria on 11-12 August
2004. Ten hospital coordinators participated in the training workshop, including one from
Tasmania. Victorian participants included six doctors, two pharmacists and one clinical
research nurse. An additional Victorian hospital coordinator later attended training at the
Queensland academic detailing workshop in September 2004. CAPTION hospital
coordinators were trained in the techniques of academic detailing and in the use of the
detailing resources developed specifically for the project by the state project officers. During
the two-day course, a therapeutic briefing on CAP was presented by Dr Tony Korman, Head
of Infectious Diseases at Southern Health. It was expected that the hospital coordinators
would start detailing by approximately the end of October 2004 when the printing of the
detailing cards was expected to be completed. Surveys conducted post-training were
retained by the NPS, so these data are not available on hand to Vic CAPTION although
follow-up telephone conversations between the state project officer and the hospital
coordinators indicated that all were sufficiently confident in academic detailing and
knowledge of CAP to carry out the intervention(s) successfully. Finally, a document was
                                                                                                9
CAPTION Victoria – Final Report                                               April, 2006


drafted by the Victorian state project team for submission to the NPS regarding the proposed
support to hospital coordinators when undertaking academic detailing (Appendix 10).


3.5 Intervention strategy and tools
Prior to the commencement of CAPTION in Victorian hospitals a survey of ED doctors
(Appendix 11) was carried out in order to ascertain the level of awareness and knowledge of
the current Guidelines and CAP management recommendations. The survey comprised six
questions, with five questions specifically relating to Guideline recommendations, including
two CAP case studies, and one question requiring doctors to select their preferred
intervention. Surveys were distributed to ED doctors in the eight Victorian CAPTION
hospitals by study coordinators and returned to the CAPTION state project team for analysis.

Four key messages were highlighted for CAPTION by the national project team, in
consultation with stakeholders at both a state and national level -

      1.    “Use a systematic approach to assessing severity”
      2.    “Select antibiotic therapy according to severity”
      3.    “Penicillins are first choice for non-severe CAP”
      4.    “Consider atypical pneumonias when deciding on antibiotic regimen”

These key messages were incorporated into the suite of educational intervention tools
developed for the project. The tools developed by the national project team were also
reviewed and endorsed by the state-based committees before then being made available for
use in participating hospitals. The toolkit included:

1.   An academic detailing card and accompanying script. The card was used in all
     detailing sessions and presented to the healthcare professional being detailed at the
     completion of the session as a gift. The script provided additional supportive material
     and relevant reference material and was for use by the detailer only.
2.   Laminated A3 size posters describing calculation of the Pneumonia Severity Index
     (PSI) and the antibiotics recommended for each of the five PSI Classes by the
     Guidelines. Posters were to be displayed in the ED.
3.   PSI calculator stickers of A5 size for use in patient histories. Stickers were to be left in
     the ED to facilitate ready determination of the PSI score for CAP patients.
4.   Laminated ID card size prompts for PSI calculation and antibiotic recommendations.
     These were distributed to ED medical staff by the hospital coordinators.
5.   A generic letter to prescribers describing the CAPTION project was provided to hospital
     coordinators for use in informing prescribers outside of the ED e.g. medical registrars,
     respiratory and infectious diseases (ID) physicians etc. For large EDs this could also be
     used to reach ED medical staff who were unable to be reached with an academic
     detailing session.
6.   A generic PowerPoint feedback presentation was developed for use by hospital
     coordinators to inform hospital staff of audit and education results at their hospital and
     also to provide feedback on state and national data from CAPTION. Presentations
     were localised through the addition of hospital logos and local hospital data. An
     explanation was given regarding the interventions to be undertaken, including
     academic detailing.
7.   Automated feedback reports were also available from the Auditmaker software. These
     were used to assist local project coordinators adapt the generic feedback presentation
     for use on site.

A considerable amount of information was made available to project coordinators in the
CAPTION ‘Manual for hospitals’. This contained project information, state project officers’
contact details, project methodology, frequently asked questions, literature review,
                                                                                              10
CAPTION Victoria – Final Report                                              April, 2006


instructions for using the Auditmaker® software, the baseline data collection form and
accompanying explanatory notes and key references. Two updates to the manual were
provided throughout the project. Details of the support strategy for hospital coordinators
conducting academic detailing are described in Appendix 10. Timing of the interventions and
audits were carefully considered; since audits were conducted retrospectively, if a time lag
occurred between the interventions and the follow-up audit this should not have caused a
problem, provided CAP presentations included in the audit were admitted to the ED during
the appropriate timeframe. All of the eight Victorian hospitals participated in three audits
(Audit I, Audit II, Audit III) and accompanying intervention phases (Intervention Phase I,
Intervention Phase II). Data collection logs and intervention activity logs were completed by
hospitals in order to document project activities and feed back this information at both a state
and national level.


3.6 Data management
As described in the section on baseline data collection, the two-letter project coding system
ensured confidentiality for hospitals. A coding form (Appendix 9) was used by hospital
coordinators to link patient medical record numbers with CAPTION study numbers. These
forms were held by the participating hospitals.

Audit data were typically collected using the paper form and then transferred into the
Auditmaker® software program, an auditing tool specifically tailored for use in CAPTION.
Auditmaker® facilitated the calculation of a PSI score for each patient and evaluated
prescribed antibiotic therapy against Guideline recommendations to determine concordance.
These and other data including patient demographics, length of stay, mortality, prior antibiotic
use, penicillin hypersensitivity, concordance by class and antibiotic monotherapy, were
incorporated into the Auditmaker® automated feedback report.


3.7   Data analysis and evaluation
The Auditmaker® software provided the principal means of analysing the project data,
although selected manual analyses were also conducted. Concordance of antibiotic therapy
(drug selection, route of administration) with Guideline recommendations and documentation
of a PSI score in patient notes were the two main indicators monitored in the project. The
statistical comparisons of post-intervention data and baseline data were carried out using the
chi-squared test.

Antibiotic therapy, both concordant and discordant, was closely examined for all CAP
presentations, with reasons for discordant therapy classified under: antibiotics not
recommended by the Guidelines, incorrect route of administration, inadequate empiric cover
and incorrect alternative prescribed for penicillin allergy. ‘Adequate empiric cover’ included
antibiotic regimens that were concordant with the Guidelines, antibiotic regimens that
included an intravenous beta-lactam instead of oral amoxycillin for Class I and II patients and
antibiotic regimens for Class III and IV patients that included IV ceftriaxone or cefotaxime for
non-penicillin allergic patients; oral atypical cover for Class V patients was not considered
adequate empiric therapy for Legionella spp. Therapies were grouped according to the
method presented in the baseline data publication (Maxwell et al, 2005) i.e. monotherapy
(includes moxifloxacin), dual therapy and ‘other’. It should be noted that if two antibiotics of
the same class were prescribed (e.g. two beta-lactams) these were classified as
monotherapy.

Patient outcomes were also assessed in relation to concordance of antibiotic therapy. While
patient mortality for the Victorian dataset was likely to be too small from which to extract
meaningful information, data comparing length of stay to concordance, with a particular focus
                                                                                             11
CAPTION Victoria – Final Report                                              April, 2006


on PSI Class I and II (mild CAP) patients, was reviewed for any trends of note. The
proportion of patients with pneumonia confirmed on X-ray was also recorded.

Nationally produced evaluation forms describing data collection at hospitals, intervention
activities and hospital profiles were distributed to all participating hospitals. Submitted data
were analysed at both state and national levels.


3.8 Communication strategy
Every effort was made to ensure regular communication between the state project team and
participating hospitals, and also between the state and national project teams.

In the development phase of the project in Victoria, three newsletters were distributed to
participants and interested stakeholders to provide updates on CAPTION. In addition to this,
regular phone calls, emails (both individual and group) were also maintained both in the
development phase and throughout the project.

A number of face-to-face meetings were held by the Victorian state project committees: the
first consultative committee meeting was held on 7th October 2003, and following the
establishment of its offshoot project committee two further face-to-face planning meetings
were held on 6th November 2003 and 11th December 2003. Following the commencement of
hospital recruitment, hospital coordinators were also invited to attend project meetings if
available, with these larger group meetings held on: 17th March 2004, 3rd June 2004, 19th
August 2004, 6th October 2004, 9th February 2005, 13th April 2005, 26th July 2005 and the
final wrap-up meeting on 25th October 2005.

Project updates were published on the website of the Victorian Drug Usage Evaluation
Group at http://www.health.vic.gov.au/vmac/groups/vdueg.htm and were also tabled at
meetings of the Victorian Therapeutics Advisory Group (VicTAG) and the Victorian Drug
Usage Advisory Committee (VDUAC). Regular contact was also kept with the Therapeutic
Guidelines Limited (TG) and project feedback and information provided to TG. It should be
noted that TG was also represented on the state project committee (Dr Jonathan Dartnell,
Production Manager). CAPTION project information was also shared with two other
concurrent CAP initiatives, the Australian Community-Acquired Pneumonia Study (ACAPS)
and the CAP project managed by the Victorian Rural Physicians’ Network (VRPN).

At a national level, the NPS developed and maintained the password-protected CAPTION
project website at http://www.nps.org.au/caption which provided a valuable resource to
hospital coordinators. Electronic copies of the hospital manual for the project, evaluation
forms, state project officer contact details, PowerPoint presentations and copies of the
Auditmaker® software were made available on the website. The use of this website was
promoted to hospitals by the state project officer.

Regular teleconferences were held by the NPS with the state project teams and a number of
face-to-face meetings were organised throughout the project to facilitate planning,
development of interventions, training in academic detailing etc (17th December 2003, 21st
April 2004, 17th/18th June 2004, 3rd December 2004, 28th September 2005). Regular
telephone and email contact was also maintained between the state and NPS.

Publications and presentations from the CAPTION project that Victoria have been involved in
are listed in Appendix 14.




                                                                                             12
CAPTION Victoria – Final Report                                               April, 2006


4.    Results
4.1 Hospital recruitment
Altogether, nineteen hospitals were approached to take part in VicCAPTION, including
medium and large public metropolitan hospitals, two private metropolitan hospitals and
several rural hospitals. While at least twelve hospitals were keen to participate, owing
primarily to a lack of resources i.e. limited funding and staff with sufficient time to commit to
the project, only eight hospitals signed up to participate in the project. Additionally
VicCAPTION was unable to pursue any of the rural hospitals owing to a perceived conflict of
interest with another CAP initiative being conducted in Victoria by the Victorian Rural
Physicians Network. Of the eight Victorian hospitals, seven were major metropolitan public
hospitals and one was a metropolitan private hospital.


4.2 ED doctors’ survey
Results from the ED doctors’ survey (see Appendix 11 for survey) are presented in Table 1.
A total of 118 surveys were completed. The majority of doctors (88%) agreed that their
information source if unfamiliar with a prescribing regimen were the Therapeutic Guidelines
publications and 74% indicated that they were familiar with the Guidelines’ recommendations
on CAP management. More than half of prescribers (57%) correctly identified the
recommended antibiotic therapy for mild CAP. They reported that factors determining use of
broad-spectrum cephalosporins included disease severity, co-morbidities and penicillin
allergy. Less than half of respondents correctly identified antibiotic therapy concordant with
Guidelines’ recommendations for the 2 CAP case studies described in the survey (44% and
49%). The most popular interventions to assist CAP prescribing were a PSI (Pneumonia
Severity Index) scoring ID card (36%), improved access to Therapeutic Guidelines (22%), a
PSI calculator (21%), a web-based antibiotic approval system (12%) and academic detailing
(10%).


4.3 Audits
Audit I comprised 160 patients, with 20 patient presentations contributed from each of the
eight participating hospitals. Baseline audit data revealed that approximately 86% of all
baseline audit patient presentations had X-ray confirmation of pneumonia documented, while
62% had both X-ray confirmation of disease and at least two symptoms of pneumonia
documented (Table 5). Patient demographic data, prescriber details, disease severity,
mortality and length of stay are summarised in Table 4. More than half of the presentations
were men (58%), most were aged over 60 years (64%) and approximately a third of patients
had been treated with antibiotics prior to admission. ED doctors were responsible for 84% of
prescribing, with antibiotics most frequently prescribed by registrars (40%). The median
length of stay was five days and inpatient mortality was 4%.

Empiric antibiotic therapy is detailed in Table 6. Seventy-six percent of patients were
prescribed therapy comprising at least a beta-lactam together with a macrolide or
doxycycline, while 22% of patients received monotherapy. Concordance of antibiotic
prescribing (drug choice and route of administration) with the Guidelines and documentation
of the PSI were determined from Audit I data to be 22% and 6%, respectively (Table 7).
Concordance was also assessed according to PSI Class (Table 8) and was found to be
highest in PSI Class III (48%) and lowest in PSI Classes I and II (7% and 6%, respectively).
Adequate empiric cover was provided to 69% of patients audited. Reasons for discordant
prescribing are presented in Table 9 and discordant antibiotic therapy is detailed in Table 10.
The main reason for discordance in PSI Classes I and II was incorrect route of
administration, with antibiotics administered intravenously rather than orally in patients
assessed as having mild CAP. In PSI Classes III and IV the use of ceftriaxone/cefotaxime for
                                                                                              13
CAPTION Victoria – Final Report                                                 April, 2006


patients who did not have a history of penicillin allergy was the predominant reason for
discordant therapy, while the use of oral atypical cover (roxithromycin rather than intravenous
erythromycin) was the main reason for discordance in PSI Class V patients. Length of stay
for patients prescribed concordant therapy and patients prescribed discordant therapy is
presented in Table 15. Given the small numbers of patients involved in these comparisons
however, a statistical analysis to assess significant differences was not conducted.

Information from the data collection logs for Audit I is presented in Table 16. Most hospitals
experienced difficulty retrieving patient records for the retrospective audit and most reviewed
more than 40 histories to identify 20 eligible patients.

Patient demographics in Audit II were generally similar to those recorded in the baseline
audit (Audit I, Table 4). The majority of patients (78%) in Audit II were prescribed therapy
comprising at least a beta-lactam together with a macrolide or doxycycline, while 17% of
patients received monotherapy (Table 6). Overall concordance with the Guidelines was 29%
(p=0.124, Table 7). Concordance for PSI Class IV patients appeared to markedly improve,
increasing from 30% to 46% (Table 8, statistical comparisons not conducted). Reasons for
discordance and details of discordant therapy recorded in Audit II are presented in Tables 11
and 12. Documentation of the PSI increased almost four-fold to 22% (p<0.0001). Data
collection logs for Audit II indicated fewer problems with record review and shorter times
required per record (Table 17).

Patient demographics for Audit III (n=147) were broadly consistent with data from Audits I
and II, although there was a slightly higher number of PSI Class V patients and a
corresponding reduction in Class IV patients (Table 4). Most patients (76%) received dual
therapy comprising a beta-lactam together with a macrolide or doxycycline while 21% of
patients received monotherapy (Table 6). Concordance with the Guidelines was the same as
that observed in Audit II, measuring 29% (p=0.138, Table 7). Reasons for discordance and
details of discordant therapy recorded in Audit III are presented in Tables 13 and 14.
Documentation of the PSI remained elevated and significantly higher than baseline (18%,
p=0.0005). Hospitals were not required to complete data collection logs in Audit III.


4.4 Interventions
All hospitals utilised the generic PowerPoint feedback presentation for group sessions as the
first part of Intervention Phase I. Posters, ID cards and stickers were distributed in EDs.
Seven out of the eight hospitals employed academic detailing as an intervention and while
the eighth hospital intended to do so, a conflict of interest with an internal initiative precluded
detailing from taking place at this site. Prescribers (interns, residents, registrars, physicians)
were typically targeted first for detailing although some detailing of nursing staff was
conducted in two of the eight Victorian hospitals. The generic letter to prescribers was used
in some hospitals where not all the medical staff could be reached with group or individual
education sessions. Educational interventions were timed such that staff exposed to the
interventions would be working in the ED throughout the post-intervention audit period.

Details of group education sessions and academic detailing (1:1) sessions are presented in
Tables 18 and 19. A total of 359 staff attended 19 group education sessions across the eight
hospitals, with around 70% of attending staff being doctors. Academic detailing sessions
were conducted with 304 staff across seven sites; approximately 90% of staff detailed were
doctors.

The number of interventions in Intervention Phase II was markedly less than the first. Details
of group education sessions and academic detailing (1:1) sessions from Intervention Phase II
are presented in Tables 20 and 21. A total of 122 staff attended seven group education
sessions across four of the eight hospitals, with doctors comprising 57% of attending staff.
                                                                                                14
CAPTION Victoria – Final Report                                            April, 2006


Academic detailing sessions were conducted with 142 staff across five sites; approximately
75% of staff detailed were doctors. Hospitals continued to display CAPTION posters and to
distribute ID cards to prescribers, and discussions with hospital coordinators indicated that
CAPTION and best practices in CAP management were again promoted.




                                                                                          15
CAPTION Victoria – Final Report                                              April, 2006


5.    Discussion
5.1 Main findings and limitations
Seven public hospitals and one private hospital participated in the CAPTION project in
Victoria. All hospitals agreed to implement the Guideline recommendations using the
CAPTION intervention tools and to complete Audits I, II and III. Victorian hospital
coordinators were mostly ED doctors and CAPTION project activities were mostly carried out
by project coordinators in addition to their normal workload.

While results of a survey of ED doctors conducted prior to the project indicated that most
were aware of the Guideline recommendations, Audit I (baseline, n=160) data demonstrated
that documentation of the PSI was poor (6%) and concordance of antibiotic prescribing with
Guidelines was a moderately low 22%. Subsequent audits demonstrated a significant,
sustained improvement in the use (documentation) of the PSI and hospitals reported a
greater awareness of CAP management and prescribing in ED following project
implementation. Anecdotal reports from a number of hospital coordinators also suggested
that the actual percentage of CAP patients for whom PSI scores were calculated may be
considerably higher since prescribers sometimes calculated the PSI but did not record this in
the notes. For the purposes of this study, concordance of prescribing with the Guidelines was
defined as the correct selection of both the antibiotic agent and route of administration. This
proved to be a rather stringent definition, for example, incorrect route of administration was
determined to be a major reason for discordance, so if a doctor prescribed the first dose of
an antibiotic as an intravenous dose for a Class I or II patient, the prescribing episode was
considered discordant. A minor improvement in concordance was recorded in Audits II and
III, measuring 29% in both audits, although this change was not statistically significant.

The impact of the interventions on patient outcomes such as inpatient mortality and length of
stay was considered. At a state level, numbers for mortality and length of stay of concordant
patients were too small to allow valid statistical comparisons, although there was a trend
towards discordant mild CAP patients (PSI classes I and II) having an increased length of
stay. It is possible that these discordant ‘mild CAP’ patients cannot be adequately assessed
using the PSI given that, in each of the three audits, their average age was less than fifty
years old and given the acknowledged age bias of the PSI. This issue was emphasised in
the CAPTION project educational information, with the academic detailing card noting that
the PSI may potentially underestimate severity in younger patients.

Barriers to performing the academic detailing intervention included the labour intensive
nature of this activity, difficulty with timing this in the busy environment of ED and the large
number of staff in the larger metropolitan hospitals. Furthermore, medical staff turnover in the
public hospitals was quite rapid, therefore it was sometimes difficult to time the detailing of
staff such that these same staff members remained in the ED over the period of the follow-up
CAPTION audit. Despite these difficulties, over 300 ED staff members were detailed across
seven of the eight hospitals in Intervention Phase I and 140 staff in Intervention Phase II.
Interventions were generally well received by hospital staff and credit should be given to the
committed and conscientious local hospital coordinators for their considerable efforts,
especially with the more labour intensive academic detailing activities. Concerns of ED
prescribers were noted to include a lack of confidence in the PSI as an effective severity
assessment tool and the time taken to calculate the PSI. However sustainability of
interventions, especially academic detailing, should be carefully considered, with a marked
decrease in intervention activities observed in Intervention Phase II.

Specific barriers and enablers are described below; these have been categorised into
recruitment and start-up, audits, interventions and key messages, and project management
and evaluation.

                                                                                             16
CAPTION Victoria – Final Report                                             April, 2006


5.2 Recruitment and start-up
Enablers
• project was based on good evidence
• Therapeutic Guidelines was perceived as a credible and reputable source of information
  and recommendations
• interest in following up on previous Victorian study on lower respiratory tract infections in
  ED, recruitment of hospitals who had participated in this earlier study
• agreement/support for project early from hospital opinion leaders
• hospital coordinators were mostly ED staff or teamed up with ED staff who were well-
  placed to organise the timing of educational interventions (both group and 1:1 sessions)
• participation in a national project, and one perceived as not another initiative just
  focussed on cost-savings

Barriers
• lack of confidence in the PSI, time taken to calculate the PSI and hence an unwillingness
  to adopt it as a routine measure in CAP assessment in ED
• lack of agreement between different units within the hospital (e.g. ID and respiratory
  departments) lead to some hospitals being unable to participate
• lengthy ethics committee approvals
• difficulties in recruiting a hospital coordinator given the considerable time commitment
  required and limited funding to provide compensation to hospitals for the coordinator’s
  time
• hospitals unfamiliar with NPS

5.3 Audits
Enablers
• a team approach for conducting audits, such that this task was not left to just one person
   to accomplish
• use of Auditmaker for data recording and analysis greatly simplified the overall collation
   and analysis of data by the state project officer

Barriers
• difficulties in retrieving medical records
• relatively complex exclusion criteria
• amount of data and detail required
• one person only available to carry out the auditing
• recruitment of sufficient numbers of eligible CAP patients in the warmer seasons

5.4 Interventions and key messages
Enablers
• project materials well received, perceived as useful, informative and professionally
   presented
• generic PowerPoint presentation helpful to hospital coordinators, saved considerable
   preparation time
• ID card size PSI calculators were popular and found to be useful
• academic detailing course well received and found to be useful by hospital coordinators
• the concurrent implementation of the Australian Community-Acquired Pneumonia Study
   (ACAPS), which provided supportive data to the CAPTION key messages including
   endorsing the use of the PSI for most cases of CAP and the use of benzylpenicillin as
   empiric therapy to treat Streptococcus pneumoniae infections (clinical resistance to
   penicillin was not observed in the study)
• ED protocols for CAP management recommending use of the PSI for all suspected CAP
   patients. One health service introduced an ED protocol requiring the calculation of the PSI
                                                                                            17
CAPTION Victoria – Final Report                                              April, 2006


   for all suspected CAP patients following a Victorian coroner’s report into the death of a
   young CAP patient at a large Victorian metropolitan hospital. The report recommended
   the implementation of a pneumonia severity assessment tool in the ED (State Coroner,
   Victoria, 2005).

Barriers
• hospital coordinators were not aware of what academic detailing actually was and what
  was involved in this intervention prior to the training course i.e. hospitals did not have a
  clear understanding what they had taken on regarding this intervention
• long lag time between academic detailing course and availability of project materials
  including detailing cards etc.
• time required to perform academic detailing i.e. very labour intensive
• the busy and constantly changing environment of ED made the provision of academic
  detailing very difficult at times
• challenging to conduct and maintain interventions where the detailing was left up to one
  person only
• concerns regarding resistance of Streptococcus pneumoniae to penicillin
• prescribing ceftriaxone perceived as ‘doing no harm’ to the patient; it covers the targeted
  pathogens and may be dosed once daily compared with benzylpenicillin which requires
  four to six hourly administration; long term benefits of using narrow spectrum antibiotics
  therefore difficult to promote ‘now’
• several of the recommended antibiotics, most notably cefuroxime and moxifloxacin, not
  available in all EDs, and some doctors unfamiliar with prescribing these agents
• elderly CAP patients often treated by geriatricians who were not specifically engaged
  with/familiar with CAPTION

5.5 Project management and evaluation
Enablers
• regular communication between NPS and state project teams e.g. via telephone
   conferences
• collaboration/liaison with other participating state project teams
• evaluation of project nationally planned and coordinated by the NPS

Barriers
• staggered start of the project across the different states
• evaluation of the effectiveness of academic detailing in the ED setting not possible from
  this project given its utility as one of a suite of tools




                                                                                               18
CAPTION Victoria – Final Report                                              April, 2006


6.    Conclusion
Victorian baseline audit data for CAPTION indicated that antibiotic prescribing for CAP in
participating hospital EDs was frequently discordant with Guideline recommendations and
that documented use of the PSI scoring system was minimal. Two intervention phases and
follow-up audits were subsequently conducted, incorporating a suite of intervention tools
including academic detailing, group feedback presentations, posters, ID-card size PSI
calculators, PSI calculator stickers and generic letters to doctors. While not statistically
significant, a consistent moderate improvement was observed in concordance relative to
baseline following both intervention phases. In addition to this, a highly significant four-fold
improvement in the documentation of the PSI was observed.

Academic detailing was reportedly well received by staff, but since this was only one of a
suite of intervention tools, it was not possible to determine its effectiveness in isolation in
promoting a change in ED CAP management practices. Furthermore, in the absence
additional resources e.g. training additional staff to assist with academic detailing, it is
difficult to see how such a labour intensive intervention could be sustained beyond the
specified study period.




                                                                                             19
CAPTION Victoria – Final Report                                            April, 2006


7.    Recommendations
7.1 Support of hospitals post-CAPTION
• revise and redevelop project materials incorporating updates from the January 2006
   update of the Therapeutic Guidelines: Antibiotic
• offer to support a follow-up audit six to twelve months after the official completion of
   project activities
• suggest incorporating a recommendation for routine use of the PSI in CAP management
   into ED/hospital protocol (if this is not already the case)
• encourage and assist where possible publication/presentation of the project in relevant
   forums e.g. emergency medicine or ID conferences
• disseminate project materials to general practitioners via the Divisions of General
   Practice in order to extend the message surrounding appropriate antibiotic selection in
   CAP management to the community
• provide feedback from the project to Therapeutic Guidelines Ltd to inform the ongoing
   review process of this publication

7.2 Multi-centre project work
• offer increased funding for participants, especially where there is a considerable
   workload involved, sufficient to support a project officer for four hours per week
• endeavour to ensure that all sites commence and complete project activities at a similar
   time
• conduct at least one visit (minimum) to all participating sites
• an effective communication strategy should be in place from the beginning of the project
• ensure as far as possible that recommended drug therapies are available on
   participating hospital formularies
• ensure that there is agreement and support between relevant hospital opinion leaders at
   each site e.g. respiratory and ID departments for CAP management
• keep audit process and data requirements as simple as possible
• promote a team approach at sites to avoid the situation where workload is left entirely to
   one person

7.3 Future national DUE activities
• ensure that all states start together to enhance national collaborative and project
   development activities
• promote the National Prescribing Service within the hospital system so that hospitals are
   more familiar with them
• continue to use the Auditmaker software to facilitate data collection and analysis; use of
   this audit tool saved considerable time of what can otherwise be a tedious process!
• ensure at least two staff members per hospital are trained in academic detailing rather
   than relying on one
• build in evaluation of the effectiveness of academic detailing in hospitals where possible
   especially given that very significant time and resources are put into this intervention




                                                                                          20
CAPTION Victoria – Final Report                                               April, 2006


List of Tables

Table 1.    Results of doctor survey conducted in EDs of participating hospitals prior to
            commencement of the CAPTION study (see Appendix 11 for survey). There were
            a total of 118 respondents.
Table 2.    CAPTION study inclusion criteria.
Table 3.    CAPTION study exclusion criteria.
Table 4.    Patient demographics from Audits I, II and III.
Table 5.    Diagnosis of Community-Acquired Pneumonia (CAP); documentation of X-ray
            confirmation and symptoms of CAP from Audits I, II and III.
Table 6.    Empiric antibiotic therapy prescribed for CAP patients in ED in Audits I, II and III.
Table 7.    Concordance of antibiotic prescribing for CAP with guideline recommendations
            and documentation of the use of the Pneumonia Severity Index (PSI);
            comparisons of post-intervention audits with baseline.
Table 8.    Concordance of antibiotic prescribing for CAP with guideline recommendations
            according to Pneumonia Severity Index class for Audits I, II and III.
Table 9.    Reasons for discordant antibiotic therapy for CAP, according to Pneumonia
            Severity Index class, for Audit I (baseline).
Table 10.   Discordant antibiotic therapy prescribed for CAP patients in Audit I, according to
            Pneumonia Severity Index class.
Table 11.   Reasons for discordant antibiotic therapy for CAP, according to Pneumonia
            Severity Index class, for Audit II.
Table 12.   Discordant antibiotic therapy prescribed for CAP patients in Audit II, according to
            Pneumonia Severity Index class.
Table 13.   Reasons for discordant antibiotic therapy for CAP, according to Pneumonia
            Severity Index class, for Audit III.
Table 14.   Discordant antibiotic therapy prescribed for CAP patients in Audit III, according to
            Pneumonia Severity Index class.
Table 15.   Relationship between concordance of antibiotic therapy prescribed for CAP
            patients and length of stay (LOS).
Table 16.   Data collection for Audit I of the CAPTION study at the eight participating
            Victorian hospitals.
Table 17.   Data collection for Audit II of the CAPTION study at the eight participating
            Victorian hospitals.
Table 18.   First intervention phase activity – group sessions.
Table 19.   First intervention phase activity – academic detailing (1:1) sessions.
Table 20.   Second intervention phase activity - group sessions.
Table 21.   Second intervention phase activity – academic detailing (1:1) sessions.




                                                                                              21
CAPTION Victoria – Final Report                                                       April, 2006


Table 1.       Results of doctor survey conducted in EDs of participating hospitals prior to
               commencement of the CAPTION study (see Appendix 11 for survey). There were
               a total of 118 respondents.


Q1. First choice empiric antibiotic therapy for treatment of mild CAP?

        Penicillin/amoxycillin              26%
        Roxithromycin or doxycycline         9%
        Both                               57%
        Other                               9%

Q2. Factors determining prescription of ceftriaxone/cefotaxime?

        Severity         45%
        Co-morbidity      6%
        Both            33%
        Other           16%

Q3. Familiarity with the Guidelines recommendations regarding the use of the PSI and antibiotic
prescribing for CAP?

       74% of respondents indicated that they were familiar with these.

Q4. First source of information if unfamiliar with a prescribing regimen?

       88% of respondents indicated that the Therapeutic Guidelines was their first choice.

Q5a. CAP case study of 42 year old farmer with moderate to severe disease (PSI Class IV)

       44% of respondents suggested therapy concordant with the Guidelines.

Q5b. CAP case study of 38 year old mother with mild disease (PSI Class I)

       49% of respondents suggested therapy concordant with the Guidelines.

Q6. First (or equal first) preference for type of intervention in ED to assist in CAP prescribing?

       Laminated PSI scoring ID card                   36%
       Easier access to Therapeutic Guidelines         22%
       PSI calculator form/stamp/windows based         21%
       Web-based antibiotic approval system            12%
       Academic detailing                              10%
       Posters                                          6%
       Care plan                                       <1%




                                                                                                     22
CAPTION Victoria – Final Report                                                              April, 2006


Table 2.      CAPTION study inclusion criteria.


Inclusion criteria
- Age > 18 years
- Presumptive diagnosis of CAP recorded by ED doctors in patient medical record
- Diagnoses may include ‘lobar pneumonia’, ‘bronchopneumonia’, ‘pneumonia’, ‘community-
acquired pneumonia’
- Patient must have been prescribed antibiotics for CAP while in the ED*

* this was specifically noted in the CAPTION Manual for Hospitals only after baseline data collection




                                                                                                           23
CAPTION Victoria – Final Report                                            April, 2006


Table 3.    CAPTION study exclusion criteria.


Exclusion criteria
- Age < 18 years
- HIV positive
- Cystic fibrosis
- Suspected or confirmed tuberculosis
- Bronchiectasis
- Aspiration pneumonia
- Hospital-acquired pneumonia
- Concurrent chemotherapy/immunosuppressants (within the last 6 months)
- Taking oral prednisolone at home
- Organ transplant recipients
- Patients discharged from hospital in the previous 14 days
- Patients admitted from another hospital (except where patients have been in the ED of the
original hospital for less than 4 hours)




                                                                                          24
CAPTION Victoria – Final Report                                                  April, 2006


Table 4.     Patient demographics from Audits I, II and III.


                                          Audit I                 Audit II           Audit III
Number of patients                          160                     160                147

Age:
18 – 35 years                            29 (18%)                 15 (9%)            17 (12%)
36 – 60 years                            28 (18%)                47 (29%)            32 (22%)
> 60 years                               103 (64%)               98 (61%)            98 (67%)

Gender:
Male                                        58%                     58%                53%
Female                                      42%                     42%                47%

Admitted to:
Ward                                        68%                     73%                67%
ICU                                         1%                      2%                  4%

Prior antibiotic therapy                    32%                     36%                31%

Median length of stay (range)          5 days (0 - 47)         6 days (0 - 57)    5 days (0 – 34)

Antibiotics in ED started by:
ED medical staff                            84%                     81%                82%
Admitting unit                              11%                     12%                14%
Unknown                                     4%                        -                 2%
Other                                       <1%                     8%                  1%

Seniority of prescriber (%):
Intern                                      4%                      6%                  7%
Resident                                    23%                     20%                18%
Registrar                                   40%                     42%                39%
Physician                                   23%                     13%                24%
Other                                       1%                        -                <1%
Unknown                                     10%                     19%                12%

Disease severity (% patients):
Class I                                  30 (19%)                27 (17%)            23 (16%)
Class II                                 34 (21%)                30 (19%)            32 (22%)
Class III                                29 (18%)                34 (21%)            32 (22%)
Class IV                                 47 (29%)                46 (29%)            34 (23%)
Class V                                  20 (13%)                23 (14%)            26 (18%)

Inpatient mortality                         4%                      4%                  3%




                                                                                                    25
CAPTION Victoria – Final Report                                       April, 2006


Table 5.    Diagnosis of Community-Acquired Pneumonia (CAP); documentation of X-ray
            confirmation and symptoms of CAP from Audits I, II and III.



                                     Audit I           Audit II          Audit III
                                     (n=160)           (n=160)           (n=147)

  X-ray confirmation of pneumonia
                                    137 (86%)         134 (84%)         128 (87%)
            documented

    X-ray confirmation AND ≥ 2
     symptoms of pneumonia          99 (62%)          103 (64%)          91 (62%)
           documented




                                                                                      26
CAPTION Victoria – Final Report                                                April, 2006


Table 6.    Empiric antibiotic therapy prescribed for CAP patients in ED in Audits I, II
            and III.


Antibiotics prescribed                    Audit I              Audit II            Audit III
                                          (n=160)              (n=160)             (n=147)
Monotherapy
β−Lactam*                                26 (16%)             23 (14%)             22 (15%)
Macrolide or doxycycline                  9 (6%)               3 (2%)               9 (6%)
Moxifloxacin                                 -                 2 (1%)                  -

Dual therapy
β−Lactam + macrolide or                 116 (73%)            124 (78%)            110 (75%)
doxycycline
β−Lactam + gentamicin                     2 (1%)               2 (1%)               2 (1%)
β−Lactam + metronidazole                  2 (1%)               2 (1%)               2 (1%)

β−Lactam + vancomycin                        -                1 (<1%)                  -

β−Lactam + aciclovir                         -                    -                 1 (1%)

Macrolide or doxycycline +                   -                1 (<1%)                  -
vancomycin
Other
β−Lactam + gentamicin +                   3 (2%)              1 (<1%)                  -
macrolide or doxycycline
β−Lactam + metronidazole +                2 (1%)                  -                    -
macrolide or doxycycline
β−Lactam + gentamicin +                      -                1 (<1%)                  -
metronidazole
β−Lactam + vancomycin +                      -                    -                 1 (1%)
macrolide or doxycycline

*β−Lactams identified: ampicillin, amoxycillin, amoxycillin/clavulanate, benzylpenicillin,
ticarcillin/clavulanate, flucloxacillin, cephazolin, cefaclor, cephalothin, cefepime, cefotaxime
and ceftriaxone
 Macrolides identified: erythromycin, roxithromycin and azithromycin.




                                                                                               27
CAPTION Victoria – Final Report                                            April, 2006


Table 7.    Concordance of antibiotic prescribing for CAP with guideline recommendations
            and documentation of the use of the Pneumonia Severity Index (PSI);
            comparisons of post-intervention audits with baseline.



                                 Audit I       Audit II   p value    Audit III      p value
                                (n=160)        (n=160)               (n=147)

Concordant antibiotic             22%            29%       0.124       29%           0.138
prescribing

Documented PSI use                6%             22%      <0.0001*     18%          0.0005*

* difference to baseline is statistically significant




                                                                                             28
CAPTION Victoria – Final Report                                        April, 2006


Table 8.    Concordance of antibiotic prescribing for CAP with guideline recommendations
            according to Pneumonia Severity Index class for Audits I, II and III.



                                  Audit I         Audit II           Audit III
                                  (n=160)         (n=160)            (n=147)

       Class I                     2 (7%)         3 (11%)            3 (13%)

       Class II                    2 (6%)         3 (10%)            5 (16%)

       Class III                  14 (48%)       16 (47%)           15 (47%)

       Class IV                   14 (30%)       21 (46%)           14 (41%)

       Class V                    3 (15%)         4 (17%)            6 (23%)




                                                                                     29
CAPTION Victoria – Final Report                                                              April, 2006


Table 9.      Reasons for discordant antibiotic therapy for CAP, according to Pneumonia
              Severity Index class, for Audit I (baseline).



                                                         Pneumonia severity index class

Key points of discordance1                            I and II          III and IV             V
                                                      (n=64)              (n=76)             (n=20)

            not
Antibiotics 2 recommended by                         53 (83%)           42 (55%)           11 (55%)
Guidelines

                              Benzylpenicillin       18 (34%)                -                   -

                     Ceftriaxone/cefotaxime          21 (40%)            34 (81%)                -

                               Roxithromycin              -                  -              10 (91%)

Incorrect route of administration                   49 (77%)3           8 (11%)4            9 (45%)5
according to Guidelines

Inadequate empiric cover                             15 (23%)           18 (24%)           16 (80%)6

Incorrect alternative for penicillin                  3 (5%)              1 (1%)                0
allergy

1
  Each presentation may have more than one reason.
2
  More detailed information on discordant antibiotic therapy provided in Table 10.
3
  Intravenous antibiotics were prescribed.
  An oral β-lactam and/or an intravenous macrolide or fluoroquinolone was prescribed.
4
5
  Oral antibiotics were prescribed.
6
  Use of roxithromycin and clarithromycin in Class V patients was classified as inadequate for empiric treatment of
Legionella spp.




                                                                                                                30
CAPTION Victoria – Final Report                                             April, 2006


Table 10. Discordant antibiotic therapy prescribed for CAP patients in Audit I, according to
          Pneumonia Severity Index class.


                                               Pneumonia severity index class
Antibiotic, route of administration     I and II         III and IV             V
Amoxycillin, oral                          -                 2                  -
Amoxycillin, IV                            4                 -                  -
Amoxycillin/clavulanate, oral              4                 3                  -
Ampicillin, IV                             6                 -                  -
Azithromycin, IV                           1                 -                  -
Benzylpenicillin, IV                      18                 -                  -
Cefaclor, oral                             1                 -                  -
Cefotaxime, IV                             -                 5                  -
Ceftriaxone, IV                           21                29                  -
Cephazolin, IV                             -                 2                  -
Cephalothin, IV                            1                 -                  1
Erythromycin, oral                         -                 5                  -
Erythromycin, IV                           1                 -                  -
Flucloxacillin, IV                         -                 1                  -
Gentamicin, IV                             1                 -                  -
Metronidazole, IV                          -                 3                  1
Roxithromycin, oral                        -                 -                  10




                                                                                           31
CAPTION Victoria – Final Report                                                              April, 2006


Table 11. Reasons for discordant antibiotic therapy for CAP, according to Pneumonia
          Severity Index class, for Audit II.



                                                             Pneumonia severity index class

Key points of discordance                                 I and II           III and IV              V
                                                          (n=57)               (n=80)              (n=23)

Antibiotics not recommended by                           47 (82%)            42 (56%)            16 (70%)
Guidelines2

                                 Benzylpenicillin        11 (23%)             13 (2%)                 -

                         Ceftriaxone/cefotaxime          20 (43%)            27 (64%)                 -

                                   Roxithromycin              -                   -               10 (63%)

Incorrect route of administration                       42 (74%)4             6 (8%)5            13 (57%)6
according to Guidelines

Inadequate empiric cover                                 15 (26%)            11 (14%)            18 (78%)7

Incorrect alternative for penicillin allergy              3 (5%)               3 (4%)                 0

1
  Each presentation may have more than one reason.
2
  More detailed information on discordant antibiotic therapy provided in Table 12.
3
  Patient with penicillin hypersensitivity.
4
  Intravenous antibiotics were prescribed.
  An oral β-lactam and/or an intravenous macrolide or fluoroquinolone was prescribed.
5
6
  Oral antibiotics were prescribed.
7
  Use of roxithromycin and clarithromycin in Class V patients was classified as inadequate for empiric treatment of
Legionella spp.




                                                                                                                32
CAPTION Victoria – Final Report                                             April, 2006


Table 12. Discordant antibiotic therapy prescribed for CAP patients in Audit II, according to
          Pneumonia Severity Index class.


                                               Pneumonia severity index class
Antibiotic, route of administration     I and II         III and IV             V
Amoxycillin, oral                          -                 2                  1
Amoxycillin, IV                            5                 -                  3
Amoxycillin/clavulanate, oral              5                 1                  -
Ampicillin, IV                             6                 2                  3
Azithromycin, oral                         -                 -                  1
Azithromycin, IV                           1                 -                  -
Benzylpenicillin, IV                      11                 1                  -
Cefotaxime, IV                             4                 4                  -
Ceftriaxone, IV                           16                23                  -
Ceftazidime, IV                            -                 1                  -
Cefuroxime, oral                           -                 1                  -
Cephazolin, IV                             -                 1                  1
Cephalothin, IV                            -                 2                  -
Doxycycline, oral                          -                 -                  2
Erythromycin, oral                         1                 -                  -
Erythromycin, IV                           1                 -                  -
Gentamicin, IV                             1                 -                  -
Metronidazole, IV                          1                 1                  1
Moxifloxacin, IV                           -                 2                  -
Roxithromycin, oral                        -                 -                  10
Vancomycin, IV                             1                 1                  -




                                                                                           33
CAPTION Victoria – Final Report                                                              April, 2006


Table 13. Reasons for discordant antibiotic therapy for CAP, according to Pneumonia
          Severity Index class, for Audit III.



                                                              Pneumonia severity index class

Key points of discordance                                 I and II           III and IV              V
                                                          (n=55)               (n=66)              (n=26)

Antibiotics not recommended by                           40 (73%)            31 (47%)            11 (42%)
Guidelines2

                                  Benzylpenicillin       16 (40%)                 -                    -

                         Ceftriaxone/cefotaxime          15 (38%)             18 (58%)             13 (9%)

                                   Roxithromycin              -                   -                6 (55%)

Incorrect route of administration                       40 (73%)4            10 (15%)5            9 (35%)6
according to Guidelines

Inadequate empiric cover                                 15 (27%)            11 (17%)            20 (77%)7

Incorrect alternative for penicillin allergy                  0                1 (2%)              1 (4%)

1
  Each presentation may have more than one reason.
2
  More detailed information on discordant antibiotic therapy provided in Table 14.
3
  Patient with immediate penicillin hypersensitivity.
4
  Intravenous antibiotics were prescribed.
  An oral β-lactam and/or an intravenous macrolide or fluoroquinolone was prescribed.
5
6
  Oral antibiotics were prescribed.
7
  Use of oral macrolides in Class V patients was classified as inadequate for empiric treatment of Legionella spp.




                                                                                                                34
CAPTION Victoria – Final Report                                              April, 2006


Table 14. Discordant antibiotic therapy prescribed for CAP patients in Audit III, according to
          Pneumonia Severity Index class.


                                                Pneumonia severity index class
Antibiotic, route of administration      I and II         III and IV             V
Aciclovir, IV                               1                 -                  -
Amoxycillin, oral                           -                 7                  -
Amoxycillin, IV                             7                 1                  1
Amoxycillin/clavulanate, oral               1                 2                  -
Ampicillin, IV                              3                 -                  -
Azithromycin, oral                          2                 2                  1
Azithromycin, IV                            -                 2                  -
Benzylpenicillin, IV                       16                 -                  -
Cefepime, IV                                -                 1                  -
Ceftriaxone, IV                            15                17                  1
Cephazolin, IV                              -                 1                  -
Doxycycline, oral                           -                 -                  2
Erythromycin, IV                            1                 -                  -
Flucloxacillin, IV                          -                 1                  -
Metronidazole, IV                           -                 1                  1
Roxithromycin, oral                         -                 -                  6
Ticarcillin/clavulanate, IV                 -                 1                  -
Vancomycin, IV                              1                 -                  -




                                                                                           35
CAPTION Victoria – Final Report                                         April, 2006


Table 15. Relationship between concordance of antibiotic therapy prescribed for CAP
          patients and length of stay (LOS).



                                          Audit I       Audit II       Audit III

Total                                     (n=160)       (n=160)        (n=147)

LOS for concordant therapy (days)           5.1            5.6           4.6

LOS for discordant therapy (days)           5.0            5.8           4.8

PSI Classes I and II                      (n=63)         (n=57)         (n=55)

LOS for concordant therapy (days)          2.0a           0.0b           0.1c

LOS for discordant therapy (days)           3.5            3.8           2.7

a
  from n=4 patients
b
  from n=6 patients
c
  from n=8 patients




                                                                                      36
CAPTION Victoria – Final Report                                                                 April, 2006


Table 16. Data collection for Audit I of the CAPTION study at the eight participating
          Victorian hospitals.


Hospital code                  AA          AB          AC          AE         AF         AG         AH          AJ
                a
Data collector                 Ph        Ph/Pre        Dr          Ph          Dr       Dr/Dr        Dr         Dr

Data collection start         2/6/04     1/4/04      20/7/04     20/7/04     9/04      6/7/04      1/6/04     1/4/04
Data collection finish       31/8/04     15/4/04    13/10/04     31/8/04     10/04     2/1/05     30/6/04     29/4/04

No. of records reviewed         69         54          22          200        50         30          47         46
No. of records discarded
                                49         34           2          180        30         10          27         26

Method of data                P-AM        P-AM         AM         P-AM       P-AM         P        P-AM        P-AM
collectionb

Avg time / record (min)         45         30         10-15        20         10         20          20         20

Auditmaker ease of use
                                1           1           3           4          2          -          3              2
Scale 1 (easy) to 5
(difficult)

Problems with record         “Obtaining records – had to be ordered in advance”
review
                             “Incomplete records”
                             “Access to records”
                             “Some difficulty accessing multiple records at one time due to Medical Record’s
                             workload”
                             “Inadequate information re follow-up in patient notes”
                             “Slow access to records. Inaccurate computer data entry”
                             “Major problems accessing records – 2 week delay between request & receipt of
                             records, many unavailable & requiring multiple requests”

Other                        “Unable to download Auditmaker from website – showing error messages”
problems/comments
                             “Because of incomplete data in records and mandatory fields in Auditmaker –
                             difficulty. Need missing data default code etc.”
                             “Auditmaker: difficulty in opening file with Access until security feature disabled”
                             “Collecting primary data very time consuming & frustrating dealing with medical
                             records. Transposing data to Auditmaker reasonably easy.”

a
 Dr = Doctor, Ph = Pharmacist, Pre = Pharmacy Pre-registrant
b
 P-AM = paper entry then entry into Auditmaker, AM = directly to Auditmaker, P = data entry on paper only;
submitted to state project officer for entry into Auditmaker




                                                                                                                        37
CAPTION Victoria – Final Report                                                                April, 2006


Table 17. Data collection for Audit II of the CAPTION study at the eight participating
          Victorian hospitals.


Hospital code              AA         AB         AC          AE         AF            AG        AH        AJ
                a
Data collector            Dr/Ph     Ph/Pre       Dr          Ph         Dr            Dr/Dr   Dr/Dr       Dr

Data collection start    20/5/05    1/4/05     19/5/05    20/1/05     2/3/05      7/8/05      18/4/05   20/5/05
Data collection          3/8/05    23/5/05     9/6/05     30/1/05     21/4/05    11/10/05     6/7/05    31/8/05
finish

No. of records             81         59         22          45         20             40       39        50
reviewed
No. of records
                           61         39          2          25          0             20       19        30
discarded

Method of data            P-AM       AM          AM        P-AM         AM             P      P-AM           P
collectionb

Avg time / record          30         15        5-10         10         10             8        15       ND
(min)

Auditmaker ease of
use                         1          1          3           3          1              -       2            -

Scale 1 (easy) to 5
(difficult)

Problems with           “Incomplete medical records”
record review
                        “Difficult to sometimes find results of lab investigations”
                        “Not easy to find HCT and glucose levels in medical records”
                        “Slow access to medical records”
                        “Access to records improved by allowing Medical Records Dept to bill for requests”

Other
problems/comments

a
 Dr = Doctor, Ph = Pharmacist, Pre = Pharmacy Pre-registrant
b
 P-AM = paper entry then entry into Auditmaker, AM = directly to Auditmaker, P = data entry on paper only;
submitted to state project officer for entry into Auditmaker
ND = Not documented




                                                                                                                  38
CAPTION Victoria – Final Report                                                                                    April, 2006


Table 18. First intervention phase activity – group sessions.



             Session no.




                                                                                                                       Pharmacist
                                                           Physician




                                                                                          CNC/CNE
                                               Registrar
                                    Resident




                                                                            Locum
                           Intern




                                                                                                                                    Other


                                                                                                                                            Total
                                                                                    NUM




                                                                                                    CNS
Hospital




                                                                                                          RN
                                                                       GP




                                                                                                               EN
 code



   AA        1              3         6        10            2         -      -      -      4       3     -    -           -          -     28
             2               -         -         6            -        -      -      -       -       -    -    -           -          -      6

   AB        1               -         -          -          5         -      -      -       -       -    -    -           -          -      5
             2               -         -         7           1         -      -      -       -       -    -    -           -          -      8
             3               -         -          -          2         -      -      -       -       -    -    -           -          -      2
             4               -         -          -           -        -      -     1        -      7     -    -           -          -      8
             5               -         -          -           -        -      -      -       -       -    -    -          7           -      7

   AC        1              2         4           -          1         -      -      -       -       -    -    -           -          -      7

   AE        1               -         -         5           5         -      -      -       -       -    -    -           -          -     10
            2*             20       20         25          25          10     -      -       -       -    -    -           -          -     100
  AF**       1              1         2          4           4         -     2      1       2       2     6    -           -         3      27
             2              1         2          4           4         -     2      1       2       2     6    -           -         3      27

   AG        1               -         -          -           -        -      -      -      1        -    3    3           -          -      7
             2               -         -          -          6         -      -      -       -       -    -    -           -          -      6

   AH        1              2         8          3           2         -      -      -       -       -    -    -           -          -     15
             2              2       10           4            -        -      -      -       -       -    6    -           -          -     22

   AJ        1               -         -         8           3         -      -      -      2       3     15   -           -          -     31
             2               -         -         9           2         -      -      -      1       2     14   -           -          -     28
             3             15          -          -           -        -      -      -       -       -    -    -           -          -     15
  Total     19             46       52         85          62          10    4      3     12        19    50   3          7          6      359

* Estimate only: presentation given at a hospital Division of Medicine meeting with
approximately 100 doctors in attendance
** Estimates only: approximately 27 staff attended each group session




                                                                                                                                            39
CAPTION Victoria – Final Report                                                                                   April, 2006


Table 19. First intervention phase activity – academic detailing (1:1) sessions.




                                                                                                                                Total no. sessions
                                                                                                     Pharmacist
                                            Physician




                                                                           CNC/CNE
                                Registrar
                     Resident




                                                             Locum
            Intern




                                                                                                                     Other
                                                                     NUM




                                                                                     CNS
Hospital




                                                        GP




                                                                                           RN

                                                                                                EN
 code




   AA       16       30         12            8         1     3      1       5       7     3    -        -             -        86
   AB         -         -          -           -        -      -      -       -       -    -    -        -             -             0

   AC        1         8           -          5         -      -      -       -       -    -    -        -             -        14
   AE        1          -         7           4         -      -      -       -       -    -    -        -             -        12
   AF        5         7        14          12          -      -      -       -       -    -    -        -             -        38

   AG         -         -         2           8         2      -      -       -       -    -    -        -             -        12

   AH        8       21         16            7         2      -      -       -       -    -    -        -             -        54
   AJ       15       12         30          12          -      -     3       2        -    14   -        -             -        88

  Total     46       78         81          56          5     3      4       7       7     17   -        -             -     304




                                                                                                                                                     40
CAPTION Victoria – Final Report                                                                                        April, 2006


Table 20. Second intervention phase activity - group sessions.



             Session no.




                                                                                                                           Pharmacist
                                                           Physician




                                                                                          CNC/CNE
                                               Registrar
                                    Resident




                                                                            Locum
                           Intern




                                                                                                                                        Other


                                                                                                                                                Total
                                                                                    NUM




                                                                                                        CNS
Hospital




                                                                                                              RN
                                                                       GP




                                                                                                                   EN
 code



   AA        0               -         -          -           -        -      -      -       -           -    -    -           -          -      0
   AB        0               -         -          -           -        -      -      -       -           -    -    -           -          -      0

   AC        0               -         -          -           -        -      -      -       -           -    -    -           -          -      0
   AE        0               -         -          -           -        -      -      -       -           -    -    -           -          -      0
   AF*       1              1         1          3           3         -      -      -      2           2     6    -           -          -     18
             2              1         1          3           3         -      -      -      2           2     6    -           -          -     18
   AG        1               -         -         1           7         -      -                     7              -           -          -     15
             2               -         -         2         16          -      -                     -              -           -         2      20
             3               -         -         1           5         -      -                     -              -           -          -      6

   AH        1              4         8          2            -        -      -      -       -           -    -    -           -          -     14

   AJ        1               -         -       11            3         -      -     1       1           1     14   -           -          -     31

  Total      7              6       10         23          37          -      -     1       5           5     33   -           -         2      122

* Estimates only: approximately 18 staff attended each group session




                                                                                                                                                41
CAPTION Victoria – Final Report                                                                                   April, 2006


Table 21. Second intervention phase activity – academic detailing (1:1) sessions.




                                                                                                                                Total no. sessions
                                                                                                     Pharmacist
                                            Physician




                                                                           CNC/CNE
                                Registrar
                     Resident




                                                             Locum
            Intern




                                                                                                                     Other
                                                                     NUM




                                                                                     CNS
Hospital




                                                        GP




                                                                                           RN

                                                                                                EN
 code




   AA        3       16         16            9         -     4      1       2       20    6    2        -             -        79
   AB         -         -          -           -        -      -      -       -       -    -    -        -             -             0

   AC         -         -          -           -        -      -      -       -       -    -    -        -             -             0
   AE        6          -         4            -        -      -      -       -       -    -    -        -             -        10
   AF        5         5        10             -        -      -      -       -       -    -    -        -             -        20

   AG         -         -          -           -        -      -      -       -       -    -    -        -             -             0

   AH        4         6          2            -        -      -      -       -       -    -    -        -             -        12
   AJ        5         7          5            -        -      -      -      2       2     -    -        -             -        21

  Total     23       34         37            9         -     4      1       4       22    6    2        -             -     142




                                                                                                                                                     42
CAPTION Victoria – Final Report                                   April, 2006


Abbreviations

ACAPS      Australian Community-Acquired Pneumonia Study
ASA        Australian Society for Antimicrobials
CAP        Community-Acquired Pneumonia
CAPTION    Community-Acquired Pneumonia: Towards Improving Outcomes
           Nationally
CEO        Chief Executive Officer
DUE        Drug Usage Evaluation
ED         Emergency Department(s)
Guidelines Therapeutic Guidelines: Antibiotic, version 12, 2003
ID         Infectious Diseases
NPS        National Prescribing Service
PSI        Pneumonia Severity Index
QI         Quality Improvement
TG         Therapeutic Guidelines Limited
SHPA       Society of Hospital Pharmacists of Australia
VDUAC      Victorian Drug Usage Advisory Committee
VDUEG      Victorian Drug Usage Evaluation Group
VicTAG     Victorian Therapeutics Advisory Group
VMAC       Victorian Medicines Advisory Committee
VRPN       Victorian Rural Physicians’ Network




                                                                                43
CAPTION Victoria – Final Report                                           April, 2006


Acknowledgements

The VDUEG CAPTION team gratefully acknowledges the support and advice provided by the
state project steering committee:

Dr Patrick Charles (Infectious Diseases, Austin Health), Dr Jonathan Dartnell (Therapeutic
Guidelines Limited), Dr Andrew Dent (Emergency, St Vincent’s Hospital), Prof Lindsay
Grayson (Infectious Diseases, Austin Health), Prof Lou Irving (Respiratory Diseases, Royal
Melbourne Hospital), Ms Sue Kirsa (formerly Chair, VDUEG), Dr Tony Korman (Infectious
Diseases, Monash Medical Centre), Dr Mary O’Reilly (Infectious Diseases, Box Hill Hospital),
Mr Bill Thomson (formerly Executive Officer, VDUAC)


The project could not have progressed nor have enjoyed the successes it has without the
significant time and effort contributed by the local hospital coordinators and clinical
champions:

Angliss Hospital              Dr Wilson Phiri
Box Hill Hospital             Dr Mary O’Reilly
                              Ms Paula Lee
Dandenong Hospital            Dr Robert Meek
                              Dr Ken Hii
Epworth Hospital              Dr Allen Yuen
                              Dr Carmel Crock
                              Dr Nigel Beck
Maroondah Hospital            Dr Min Hin Chong
Monash Medical Centre         Dr Tony Korman
                              Dr Andre Vanzyl
                              Dr Andrew Block
Peninsula Health              Dr Darsim Haji
                              Ms Frances Caplygin
Royal Melbourne Hospital      Dr Lou Irving
                              Ms Marion Robertson
                              Ms Ann Marie Sherman


Thank you also to the other state-based CAPTION project teams for their assistance and the
opportunity to collaborate throughout the project and to the NPS CAPTION team, Kylie
Easton, Angela Wai, Judith Mackson and Fiona Horn, for their guidance and support. We
would also like to thank Dr Lynn Weekes for funding and supporting this national initiative.

Finally, we would like to thank the Victorian Drug Usage Advisory Committee for their
assistance in the project administration. Mr Bill Thomson for his wisdom, guidance and much
other assistance, Mrs Kay Dunkley and more recently, Ms Helen Leach and the Quality and
Safety Branch, Victorian Department of Human Services, for their support and
encouragement with CAPTION.




                                                                                         44
CAPTION Victoria – Final Report                                             April, 2006


References

Antibiotic Writing Group. Therapeutic Guidelines: Antibiotic, version 12. Melbourne:
Therapeutic Guidelines, 2003:407.

Ball P, Baquero F, Cars O, et al. Antibiotic therapy of community respiratory tract infections:
strategies for optimal outcomes and minimized resistance emergence. J Antimicrob
Chemother 2002; 49:31-40.

Buising K, Thursky K, Black J, Brown G. Are the Australian guidelines asking too much of the
Pneumonia Severity Index (PSI)? [letter] Med J Aust 2004; 180: 486-487.

Charles PGP, Whitby M, Fuller A, et al. The aetiology and severity of community-acquired
pneumonia (CAP) - early results from the Australian community-acquired pneumonia study
(ACAPS). The Australasian Society of Infectious Diseases Inc Annual Scientific Meeting,
Busselton, Western Australia, Australia, 2005.

Dobbin CJ, Duggan CJ, Barnes DJ. The efficacy of an antibiotic protocol for community-
acquired pneumonia. Med J Aust 2001; 174:333-7.

Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify low-risk patients with
community-acquired pneumonia. N Engl J Med 1997; 336:243-250.

Johnson PD, Irving LB, Turnidge JD. 3: Community-acquired pneumonia. Med J Aust 2002;
176:341-7.

Lawford S. Appropriate prescribing in community-acquired pneumonia (APICAP) pilot study.
Aust Pharm 2003; 22:630-634.

McKellar A. Does antibiotic choice affect length of stay for community-acquired pneumonia?
J Pharm Pract Res 2005; 35:25-7.

National Centre for Classification in Health. The international statistical classification of
diseases and related health problems, 10th revision, Australian modification. Sydney: NCCH,
2002.

Niederman MS, Mandell LA, Anzueto A, et al. Guidelines for the management of adults with
community-acquired pneumonia. Diagnosis, assessment of severity, antimicrobial therapy,
and prevention. Am J Respir Crit Care Med. 2001; 163:1730-1754.

Paterson DL. "Collateral damage" from cephalosporin or quinolone antibiotic therapy. Clin
Infect Dis 2004; 38:S341-5.

Paterson DL, Playford EG. Should third-generation cephalosporins be the empirical
treatment of choice for severe community-acquired pneumonia in adults? Med J Aust 1998;
168:344-8.

Radford J, Cardiff L, Pillans P, Fielding D, Looke D. Drug usage evaluation of antimicrobial
therapy for community-acquired pneumonia. Aust J Hosp Pharm 1999; 29:317-20.

Robertson MB, Korman TM, Dartnell JG, et al. Ceftriaxone and cefotaxime use in Victorian
hospitals. Med J Aust 2002; 176:524-9.


                                                                                            45
CAPTION Victoria – Final Report                                            April, 2006


Robertson MB, Korman TM, Dartnell JGA, Ioannides-Demos LL, Kirsa SW, Lord JAV.
Treatment of lower respiratory tract infections in emergency departments., Australian Society
for Antimicrobials 3rd Annual Scientific Meeting, Sydney, 2002.

State Coroner, Victoria. Beware of the ‘well’ young patient. Coronial Communique 2005 3(1):
2. Available at: http://www.vifm.org/attachments/o389.pdf (accessed April 2006)

Tsirgiotis E, Ruffin R. Community acquired pneumonia. A perspective for general practice.
Aust Fam Phys 2000; 29:639-45.

Victorian Drug Usage Evaluation Group. Antibiotic treatment of lower respiratory tract
infection in emergency departments. Melbourne: VDUEG, 2003.




                                                                                          46
CAPTION Victoria – Final Report                                        April, 2006


List of Appendices

Appendix 1.       Victorian CAPTION committee members
Appendix 2.       Project description
Appendix 3.       Letter to ED directors
Appendix 4.       ED response form for CAPTION
Appendix 5.       Letter to pharmacy directors
Appendix 6.       Letter to CEO including study participation form
Appendix 7.       Ethics package
Appendix 8.       Recruitment letter for hospital project officer
Appendix 9.       Patient record form
Appendix 10.      Strategy to support academic detailing in Victoria
Appendix 11.      CAPTION ED doctors’ survey
Appendix 12.      Laminated PSI scoring ID card
Appendix 13.      Generic letter to prescribers
Appendix 14.      Publications




                                                                                     47
CAPTION Victoria – Final Report                                           April, 2006


Appendix 1. Victorian CAPTION committee members

1. Project committee

Marion Robertson (Chair VDUEG - Drug Usage Evaluation pharmacist RMH)
Sue Kirsa (Former Chair VDUEG – Peter MacCallum Cancer Centre)
Jonathan Dartnell (Therapeutic Guidelines Limited)
Kylie McIntosh (Project Officer)
Nazila Jamshidi (Project Officer)
Bill Thomson (Executive Officer, VDUAC)
Lou Irving (Respiratory Diseases, Royal Melbourne Hospital)
Tony Korman (Infectious Diseases, Monash Medical Centre)
Mary O’Reilly (Infectious Diseases, Box Hill Hospital)
Andrew Dent (Emergency, St Vincent’s Hospital)
Pat Charles (Infectious Diseases Royal Melbourne Hospital)
Lindsay Grayson (Infectious Diseases, Austin)


2. Consultative committee

Pharmacists:
David Kong (Drug Usage Evaluation, Alfred Hospital)
Susan Luu (Drug Usage Evaluation, Western Hospital)
Julie Lord (Drug Usage Evaluation/Drug Information, St Vincent’s Hospital)
Jennifer Marriott (Senior Lecturer, Victorian College of Pharmacy)
Sharmila Melvani (Drug Usage Evaluation, Austin Hospital/Therapeutic Guidelines)
Brian Dillon (Director of Pharmacy, Warrnambool Base Hospital)

Physicians:
Kirsty Buising (Infectious Diseases, Royal Melbourne Hospital)
Michele Levinson (Physician, Royal Melbourne Hospital/Cabrini Hospital)
Christine MacDonald (Respiratory Diseases, Austin Hospital)
Denis Spelman (Infectious Diseases, Alfred Hospital)
Alistair Wright (Respiratory Physician, West Gippsland Hospital)




                                                                                        48
CAPTION Victoria – Final Report                                                         April, 2006


Appendix 2. Project description

 CAPTION Study – Community Acquired Pneumonia: Towards Improving
                     Outcomes Nationally

Project Background:
As a result of the low concordance with the Therapeutic Guidelines, Antibiotic (AG) (Version 11) for
lower respiratory tract infections observed in a recent hospital DUE (Robertson et al., 2002a; Robertson et
al., 2002b) and the revision of the national antibiotic guidelines, the National Prescribing Service has
commissioned the Victorian DUE Group along with NSW, QLD, SA and Tasmanian DUE groups to
undertake a project to improve the implementation of recommendations in the AG (Version 12) in
hospital emergency departments (ED) nationally.
Since these studies were performed, the AG Version 12 have adopted a scoring system—the Pneumonia
Severity Index (PSI) which was developed and validated by a number of North American studies (Fine et
al., 1997; Fine et al., 1996). The PSI stratifies patients with community-acquired pneumonia according to
the risk of mortality. The PSI score allows for greater confidence in patient triage and can help direct the
appropriate choice of antibiotics. It is calculated according to individual patient demographics and clinical
signs/symptoms upon presentation. The final score is used to classify the patient to a PSI risk class ranked
from 1 to 5, with class 1 being the lowest risk of mortality and class 5 the highest. A management
algorithm is then applied, using the PSI severity class, to determine the most appropriate antibiotic choice
and the most suitable place for management i.e. at home, at a general ward level or an intensive care unit.
Drug choice is also influenced by the potential for tropical or non-tropical sources of infection.

There are a number of potential benefits of improving concordance with AG Version 12:
• Increase in use of narrower spectrum antibiotics
• Broad-spectrum antibiotics being reserved for more acutely unwell patients
• More appropriate drug choice in the management of community acquired pneumonia (CAP) and
   bronchitis
• Reduction in development of resistant organisms in the hospital setting
• Patients being offered treatment that represents the most up to date evidence based therapy, resulting
   in the highest standard of medical care and service

Project Aim:
This project aims to:
• Introduce and implement the CAP and bronchitis management guidelines (AG Version 12.) into the
    ED
• Influence and improve the prescribing practice in the management of CAP and bronchitis in the ED
• Train hospital health care professionals in appropriate techniques for influencing and improving
    prescribing practice

Project Methodology:
Victorian hospitals with an ED, public or private, will be eligible to participate. A maximum of 10
hospitals will be included in the final study group. The project will employ established DUE methodology,
which has been successfully implemented by VDUEG previously. This involves data collection, evaluation
of data against the AG Version 12, feedback of evaluated data and targeted educational interventions.
Multiple DUE cycles will be implemented during the course of the 2-year project.

A steering committee has been convened comprising emergency, respiratory and infectious diseases
physicians, clinical pharmacologists, pharmacists and a representative from the Therapeutic Guidelines
Limited. The group will provide advice for project direction, development of feedback and educational
materials and mechanisms for on-going sustainability of the program.


                                                                                                          49
CAPTION Victoria – Final Report                                                        April, 2006


A clinical champion has been nominated in each participating hospital and together with the director of
ED has been asked to nominate a local project officer (PO). The hospital PO will be responsible for
liaison with ED staff, hospital Drug Committee and other authoritative hospital committees and for
coordinating the data collection and feedback programs in the hospital. The VDUEG project officer will
provide information and support to hospital coordinators to facilitate these processes.

VDUEG will provide hospitals with:
    1. Staff training in social marketing (academic detailing) – An integral part of the study is to provide
       one staff member from each participating site the opportunity to attend a 3 day course in
       developing skills associated with social marketing. This course is organised by the National
       Prescribing Service and run by the DATIS group from South Australia. The DATIS group has
       been able to demonstrate sustained prescribing change in the group of General Practitioners and
       Specialists that their program covers (May et al., 1999).
    2. Audit and feedback tools for empiric treatment of community-acquired pneumonia (CAP)
       treatment: - this will enable your ED to demonstrate the level of concordance with the
       recommendations in the AG Version 12, and to feedback data on individual performance in this
       area to ED doctors.
    3. Involvement in a quality improvement project: The ACHS Equip accreditation system now
       requires hospitals to demonstrate improvements in patient outcomes. Participation in this project
       could serve to demonstrate improvements in CAP treatment. In addition, it will allow
       benchmarking with other hospitals participating in the project.
    4. Provide support to hospital coordinators at each stage of the project
    5. Provide guidance regarding data collection and management
    6. Prepare templates for data collection and feedback
    7. Prepare and distribute materials to facilitate and support educational interventions
    8. Facilitate on-line and teleconference discussion forums for hospital coordinators to allow sharing
       of information (e.g. regarding successful interventions and strategies)

Your hospital will need to:
1. Allocate a hospital Project Officer (PO), who will become a member of the Consultative group, to
coordinate the project at a local level. It is expected that the project will require a commitment of around
one day a week to begin with within the larger metropolitan hospitals. Some hospitals may wish to assign
this to their ED pharmacist, some may prefer to allocate the task to a research nurse or to a registrar
enrolled in the ED training program, or a combination of all these
2. Allocate a hospital facilitator or clinical champion to participate in the education sessions at the
hospital

The PO will:
•   Act as the point of contact between the hospital and VDUEG
•   Be part of the local project team (see below) and report to the VDUEG project group
•   Coordinate and facilitate DUE cycles within their institution
•   Collate and send all relevant data to VDUEG
•   Commit to completing the required tasks for the project duration
•   The PO would be expected to attend the academic detailing training course offered through the
project.


NPS will:
1. Oversee all aspects of the national project
2. Provide education to all hospital coordinators in the areas of social marketing and academic detailing
3. Facilitate communication between state-based groups coordinating DUE activity

Project Group
This collaborative project will be managed under the auspices of the National Prescribing Service
(NPS) and will be conducted by the VDUAC through the VDUEG.

                                                                                                         50
CAPTION Victoria – Final Report                                                           April, 2006


Appendix 3. Letter to ED directors

                                    Re: CAPTION
           Community-Acquired Pneumonia: Towards Improving Outcomes Nationally

<Hospital name & address>

<Date>

Dear Dr <ED director>,
As I mentioned to you recently I have included a starter pack for <Hospital name> to sign on and take
part in the CAPTION project due to start mid March.

Please find enclosed the following information:

    -   The project description
    -   The project plan with a tentative time-line of major project events
    -   Standard ethics letter that can be edited and submitted to your ethics committee if required
    -   A form for yourself and the hospital CEO to sign as approval to take part in the project (these
        forms are due back by the 20th of February)
    -   A suggested letter from you to the person you have or will be promoting as the local project
        officer.
    -   The package on a hard disk for you to adapt according to the needs of your ethics and/or drug
        and therapeutics committee.
Please note that the director of your pharmacy department has also received a copy of the project
description, the project plan with a tentative time-line of major project events and the standard ethics letter
for the ethics committee. They have been notified of the expected start date in March and as I have
mentioned to you, are keen to take part and assist your department in making the implementation of the
CAPTION project a smooth transition within the department.
At this stage the local project officer at your hospital remains to be identified. But once I receive the
inclosed form a meeting will be organised between yourself, (the local project officer if they have been
recruited), the director of pharmacy and key clinical physicians to plan out how CAPTION will be rolled
out in <Hospital name>.
Until then, should you require any assistance with this stage of the CAPTION, please do not hesitate to
contact me.
Yours Sincerely


Nazila Jamshidi
B. Pharm PhD
Project Officer, VDUEG
email: Nazila.Jamshidi@vduac.org.au
03 9342 7225

                                                                                                            51
CAPTION Victoria – Final Report                                                         April, 2006


Appendix 4. ED response form for CAPTION

RESPONSE FORM
Please complete and check the details in this table and fax it to:
Ms S Kirsa
Austin and Repatriation Medical Centre
Fax: 03 9459 4546            Tel: 03 9496 5846
                         Hospital

                            Address


                         Telephone

                                Fax

                           Director

                      Email address

       Our hospital would like to
   participate in the Multi-Centre
                                       Yes / No
 CAPTION Project in Emergency
                     Departments.
   The Project Champion contact
                        person is:
          His/her email address is:

  The number of patients we treat
         for lower respiratory tract
        infection, in a week, in the
          emergency department is
 We will be arranging approval for     Drug Committee
                 the undertaking of
this quality improvement program       Quality Improvement Committee
               in our hospital with:
                                       Other

   Our hospital does not want to
   participate in the Multi-Centre
 CAPTION Project in Emergency
            Departments because:
           If your hospital wishes to participate in this study, please send copies of the
          hospital antibiotic policy and any emergency department protocols for treating
                           lower respiratory tract infection, with this form.
                               PLEASE RETURN THIS FORM BY
                                          23 February 2004




                                                                                                      52
CAPTION Victoria – Final Report                                                                    April, 2006


Appendix 5. Letter to pharmacy directors

<Hospital name & address>

<Date>


Dear <Pharmacy Director>,


In April - July 2001 several Emergency Departments in three states participated in a study of the antibiotic
treatment of lower respiratory tract infections. This study was funded by the National Prescribing Service and
carried out by the Victorian Drug Usage Evaluation Group (VDUEG). The report of the study was sent to
the National Prescribing Service and participating hospitals in May 2003.


The study identified a particularly low rate of concordance with the national antibiotic prescribing guidelines
(Therapeutic Guidelines: Antibiotic 11th Version). In addition to this, since the time of the study there have
been major changes to the recommendations for treatment of community-acquired pneumonia in the new
edition of the guidelines involving more precise assessment of the severity of the illness using the Pneumonia
Severity Index (Therapeutic Guidelines: Antibiotic 12th Version). The PSI is a scoring system that involves
assigning a numerical value to elements of clinical history and simple physical signs that are commonly
assessed when patients present with suspected community-acquired pneumonia (CAP). The advantages of
using the PSI are many including better placement of patients (home, ward, ICU), consistency of drug choice
for empiric treatment, but calculation of the PSI is an additional step and as with guidelines generally it is
unlikely to be taken up widely within an institution or an emergency department without the help of specific
intervention. As has been demonstrated2, guidelines are unlikely to be implemented within an institution unless
with the help of specific intervention. For these reasons the NPS is supporting an Australia-wide project to
implement the new Antibiotic Guidelines 12 (AG 12) for the treatment of CAP and for treatment of
exacerbations of chronic and acute bronchitis in emergency departments over a two-year period. The project is
titled CAPTION. The VDUEG has been commissioned to carry out this project in 10 Victorian hospitals as
part of the national study in 40 hospitals.


Through a planned intervention process VDUEG aims to simplify the use of PSI for prescribing purposes and
hence to increase physician confidence in assessment and prescribing of antibiotics for CAP. As a result of this
approach it is hoped to not only improve patient placement (home, ward, ICU) and diagnosis but also
potentially decrease length of stay and the length of therapy with in the hospital as a result of greater use of the
guidelines.


It is expected that the study will occur in three phases. Phase one will consist of a pre-intervention snapshot
audit (two to three weeks in duration). Phase two is the intervention phase for CAP guidelines and PSI
calculation, which is interspersed with one-day audits through out the implementation. Phase three is optional
to all participating hospitals and will commence after phase two. It involves the implementation of AG 12 for
acute and chronic bronchitis.


We hope to implement such interventions with the assistance of a staff member recommended by the
emergency department and recruited within the hospital. The recruited staff member, who may be the ED
pharmacist, an ED nurse or an ED doctor will be trained in social marketing skills, also known as academic
detailing, by the NPS and will have access to a range of nationally-developed intervention tools that can be
implemented as a part of phase two. Such tools include computer-based systems for antibiotic decision

1
 GRIMSHAW, J.M. & RUSSELL, I.T. (1993). Effect of clinical guidelines on medical practice: a systematic review of rigorous
evaluations. Lancet, 342, 1317-22.
                                                                                                                       53
CAPTION Victoria – Final Report                                                        April, 2006


support or simply for calculating PSI, laminated scoring cards or worksheets. Each hospital will have the
opportunity to choose and/or customise one or more of the intervention tools. A small payment will be made
annually to participating hospitals/individuals and may vary depending on the project budget at the time.


We are writing to let you know about the project and to ask for your assistance in collecting data on antibiotic
usage for this exciting new project. Nazila Jamshidi is the project officer for CAPTION and is available to
discuss the project with you in greater detail.


As you know change of practice in any setting is difficult to achieve, the VDUEG is committed to consulting
widely with emergency department doctors, nurses and pharmacists. As part of the initial planning process we
have consulted emergency department, infectious diseases, respiratory and general physicians and pharmacists.
A list of the Project Group and the Consultative Group is attached for your interest. We would welcome any
input from you or your department ranging from informal suggestions to representation on the broader
Consultative Group.


If you wish to participate but have concerns or questions with regards to the project please contact me on
94965879.


Yours sincerely




Sue Kirsa
Chair, Victorian DUE Group
Austin and Repatriation Medical Centre
Studley Road
Heidelberg, 3084
Phone 03 94965879
Fax 03 94594546




                                                                                                        54
CAPTION Victoria – Final Report                                                                     April, 2006


Appendix 6. Letter to CEO including study participation form

<Hospital name & address>
<Date>
                                      Re: CAPTION
             Community-Acquired Pneumonia: Towards Improving Outcomes Nationally
Dear <CEO name>,
The Victorian Drug Usage Evaluation Group (VDUEG) in conjunction with the National Prescribing
Service (NPS) would like to invite <Hospital name> to participate in this exciting project aimed at
improving patient care in the management of Community-Acquired Pneumonia (CAP). The project is also
known as the ‘CAPTION Study’ – Community-Acquired Pneumonia: Towards Improving Outcomes
Nationally.
The primary aim of CAPTION is to incorporate the introduction and implementation of the Therapeutic
Guidelines: – Antibiotics (AG) Version 12 management strategies for CAP in to hospital emergency
departments. The objective of this study is in line with and reinforces the approach of the Australian
Safety and Quality Council in Health Care on a national strategy to address healthcare associated
infections. The Council emphasises the use of the AG to guide clinical practice, in order to ensure an
assertive approach to the governance of antibiotic policy. Further, in the report the Council recommends
that the collection of antibiotic utilisation data should be pursued in both the public and the private sector
as an important aspect in the management of multi-resistant organisms3. For further details on the project
aims and processes please refer to the enclosed booklet titled “Project plan”.
This project is a national collaborative quality improvement project involving ten hospitals in Victoria
and up to 40 hospitals nationally. Eligible hospitals are from both the public (metropolitan and regional)
and private sector, and have an emergency department.
To date, we have approached the director of your emergency department (ED), who is keen for your
hospital to take part in the CAPTION project.
The recruitment process has so far involved:
 • Recruitment of a clinical champion within the hospital to assist in co-ordinating the project locally
 • Obtaining the initial expression of interest from ED
 • Investigation of possible candidates as a local project officer
 • Discussions with the director of pharmacy to ensure pharmacy support

Our next step is to initiate the project by obtaining:
• Sign-off from you, the Chief Executive Officer (CEO)
• Sign-off from other appropriate departmental heads including ED, Infectious Diseases (ID) and
   Respiratory.
• Sign-off from hospital Ethics cooperation and/or Drug and Therapeutics Committee.
• Assigning a project officer locally

Please complete the attached form and return to Sue Kirsa, Chair VDUEG, fax: 03 9459 4546 or email:
sue.kirsa@austin.org.au by 20-02-2004. However, if you wish to participate but have concerns or
questions with regards to the project please contact me on 03 94965846.
Yours Sincerely,



Su
Sue Kirsa

3
 YOUNGMAN, J. (2003). National strategy to address health care associated infections. pp. 17: Australian Council for Safety
and Quality in Health Care.
                                                                                                                         55
CAPTION Victoria – Final Report                                                   April, 2006




RESPONSE FORM
     Please complete and check the details in this table and fax or post to:
     Ms Sue Kirsa
     Austin and Repatriation Medical Centre
     Fax: 03 9459 4546        Tel: 03 9496 5846


                                 Hospital

                                 Address

                               Telephone

                                      Fax

                                    CEO

                           Email address
               Our hospital would like to
           participate in the Multi-Centre
                    CAPTION Project in
                 Emergency Departments
          CEO’s signature of approval to
                               participate
          The Clinical Champion contact
                                person is:
                His/her email address is:
           We will be arranging approval Ethics Committee
                   for the undertaking of Drug and Therapeutics Committee
                this quality improvement Other
            program in our hospital with:
         If your hospital wishes to participate in this study, please send a copy of this form to your
                                           Emergency Department
                                PLEASE RETURN THIS FORM BY 23-02-2004




                                                                                                  56
CAPTION Victoria – Final Report                                                           April, 2006


Appendix 7. Ethics package



                                         To Whom It May Concern:


           Re: A Multicentre Drug Use Evaluation in Hospitals – The CAPTION Study


This project is a national collaborative quality improvement project involving ten hospitals in Victoria and
up to 40 hospitals nationally. It aims to introduce and implement the new Therapeutic Guidelines:
Antibiotics (AG) Version 12 for the management of Community Acquired Pneumonia (CAP) and
bronchitis in the Emergency Department. The introduction and implementation of the guidelines will be
facilitated through local hospital employees who will coordinate a number of education interventions
within their institution. Audit and feedback processes will be used at intervals throughout the project
period as part of the intervention program and to assist in evaluating the impact of the project.

The project is funded by the National Prescribing Service (NPS) and is coordinated by the Victorian Drug Usage
Evaluation Group (VDUEG) in collaboration with the Victorian Drug Usage Advisory Committee
(VDUAC). VDUEG is a collaborative interest network of clinical pharmacologists, pharmacists and clinicians
committed to promoting quality drug use both in Victorian hospitals and the community. Its members are from
hospitals within Victoria. Therapeutic Guidelines Ltd is also represented. The NPS is an independent, non-profit
organisation funded by the Commonwealth government, which aims to improve the health of Australians
through quality prescribing of medicines.

The project will utilise iterative Drug Use Evaluation (DUE) methodology, which is a well-established
methodology for facilitating implementation of best practice. With the assistance of pharmacy
departments, hospital personnel normally involved in patient care will undertake the data collection
required for the project. The provisions of the National Privacy Act will be maintained at all times. As this
is a quality improvement project, patient consent to review prescribing records will not be sought.

All data will be de-identified before being sent to the VDUEG project officer for aggregation. Hospitals
will be de-identified in aggregated reports. A list of participating hospitals will be included in the final
report, but data will not be published without permission from each participating hospital.

Please find attached a project description, the proposed project reporting structure and a copy of the data
collection instrument for your information. The data collection instrument may be updated after pilot
testing and through revision by the project Steering Committee, and all, if any, alterations will be
forwarded to your committee for approval prior to utilisation.

Thank you for considering this project for approval by your committee.

Yours Sincerely



Phone
Fax




                                                                                                            57
CAPTION Victoria – Final Report                                                          April, 2006


Does CAPTION require local ethics approval?

In a recently published document by the National Health and Medical Research Council (NHMRC) the
Australian Health and Ethics Committee (AHEC) distributed advice to institutions, and their human
research ethics committees (HREC) on whether or not audits and quality assurance (QA) studies would
require ethics approval.
The AHEC advises that a study can proceed without review by an HREC if:
“Both
(a) The activity is undertaken with the consent of the patients, carers, health care providers or institutions
involved;
or
is consistent with National Privacy Principle 2.1(a), which states:
‘An organisation must not use or disclose personal information about an individual for a purpose (the
secondary purpose) other than the primary purpose of collection unless’ … ‘both of the following apply:
(i) The secondary purpose is related to the primary purpose of collection and, if the personal information
is sensitive information, directly related to the primary purpose of collection;
(ii) The individual would reasonably expect the organization to use or disclose the information for the
secondary purpose’;
And
(b) It is an activity where participants, including patients, carers, health care providers or institutions are
unlikely to suffer burden or harm (physical, mental, psychological, spiritual or social).”

The CAPTION project has been planned to ensure that these guidelines are met and the project can be
carried out with minimal setbacks/delays.




                                                                                                            58
CAPTION Victoria – Final Report                                                                April, 2006


CAPTION Study: Information for Hospital Project officers

The following questions may be included in Ethics Committee application forms. Suggested answers are provided for your
information.
Has this project been submitted to other Ethics Committees already? If so indicate the status of
each application.
This multi-centre quality improvement project is in the process of being submitted to other ethics
committees. The outcomes of the other ethics committees’ deliberations are not yet known.

How will the results of this project be disseminated?
The results of this project will be disseminated via a report to the National Prescribing Service (NPS). A
manuscript will be prepared for publication in a peer-reviewed journal. All data will be de-identified.
Participating hospitals will only be acknowledged if they agree to do so.

What is the proposed storage of, and access to, files etc during the study? How long will the data files etc
be retained after the study and how will they be disposed of?
Data collection forms will only be accessible to the hospital staff members involved in the project. The
hospital staff involved in the project will code and/or de-identify all patient data before recording. De-
identified data may be submitted to Victorian Drug Usage Evaluation Group (VDUEG) for aggregation,
with participating hospitals coded for de-identification purposes. VDUEG will not be provided with codes
used by the hospital. All de-identified data forwarded to VDUEG will be stored in a password protected
Microsoft Access Database. As this is a quality improvement project, all data will be destroyed by the
hospital staff members involved in the project once the results have been reported.

Will this research be undertaken on behalf of (or at the request of) a pharmaceutical company, or
other commercial entity, or any other sponsor?
This project will be undertaken by VDUEG and is funded by the NPS. VDUEG is a committee of clinical
pharmacologists, pharmacists and clinicians that promote quality drug use in hospitals and the community.
It is an independent collaborative interest network of clinical pharmacologists, pharmacists and clinicians
committed to promoting quality drug use both in Victorian hospitals and the community whose members
represent Victorian teaching hospitals. The NPS is an independent, non-profit organisation funded by the
Commonwealth government, which aims to improve the health of Australians through quality prescribing
of medicines.

Will the entity undertake in writing to indemnify the institution, the HREC(s) and the
researchers?
No.

Does the sponsor hold a current insurance policy to cover the project?
No. This is a quality improvement project, with no research involved.

Do the researchers have any affiliation with, or financial involvement in, any organisation or
entity with direct or indirect interests in the subject matter or materials of this research?
Therapeutic Guidelines Limited is a member of the Victorian Drug Usage Evaluation Group. Therapeutic
Guidelines Limited is a not for profit, independent organisation.


       Prepared by David Maxwell/Nazila Jamshidi, NSW TAG/VDUEG, January 2004




                                                                                                                  59
CAPTION Victoria – Final Report                                                   April, 2006


References
FINE, M.J., AUBLE, T.E., YEALY, D.M., HANUSA, B.H., WEISSFELD, L.A., SINGER, D.E., COLEY,
       C.M., MARRIE, T.J. & KAPOOR, W.N. (1997). A prediction rule to identify low-risk
       patients with community-acquired pneumonia. N Engl J Med, 336, 243-50.
FINE, M.J., SMITH, M.A., CARSON, C.A., MUTHA, S.S., SANKEY, S.S., WEISSFELD, L.A. &
       KAPOOR, W.N. (1996). Prognosis and outcomes of patients with community-acquired
       pneumonia. A meta-analysis. JAMA, 275, 134-41.
MAY, F.W., ROWETT, D.S., GILBERT, A.L., MCNEECE, J.I. & HURLEY, E. (1999). Outcomes of
       an educational-outreach service for community medical practitioners: non-steroidal anti-
       inflammatory drugs. Med J Aust, 170, 471-4.
ROBERTSON, M.B., KORMAN, T.M., DARTNELL, J.G., IOANNIDES-DEMOS, L.L., KIRSA, S.W.,
       LORD, J.A., MUNAFO, L. & BYRNES, G.B. (2002a). Ceftriaxone and cefotaxime use in
       Victorian hospitals. Med J Aust, 176, 524-9.
ROBERTSON, M.B., KORMAN, T.M., DARTNELL, J.G.A., IOANNIDES-DEMOS, L.L., KIRSA, S.W.
       & LORD, J.A.V. (2002b). Treatment of lower respiratory tract infection in emergency
       departments. In: Proceedings of Australian Society for Antimicrobials 3rd Annual
       Scientific Meeting 2002. In Australian Society for Antimicrobials 3rd Annual Scientific Meeting
       2002. pp. 10. Sydney.




                                                                                                   60
CAPTION Victoria – Final Report                                                           April, 2006


Appendix 8. Recruitment letter for hospital project officer


<Hospital name & address>

<Date>


Dear

I am writing to invite you to participate in a national project carried out by the Victorian Drug Usage
Evaluation Group (VDUEG) in conjunction with the National Prescribing Service (NPS) aimed at
implementing the use of the Pneumonia Severity Index and recommendations for treating Community-
Acquired Pneumonia (CAP).

Background:
There is evidence that concordance with the CAP and bronchitis management guidelines from the
Therapeutic Guidelines – Antibiotics (AG) in Victorian hospitals is low4. It has been demonstrated that
prescribing of 3rd generation cephalosporins in the management of respiratory tract infections (RTIs)
complied with the AG in only 25% of cases reviewed5. Since the time of the study there have been major
changes to the recommendations for treatment of CAP in the new edition of the guidelines involving
more precise assessment of the severity of the illness using the Pneumonia Severity Index6. The PSI is a
scoring system that involves assigning a numerical value to elements of clinical history and simple physical
signs that are commonly assessed when patients present with suspected CAP.

The advantages of using the PSI are many including:
•  reducing uncertainty in the diagnosis
•  better placement of patients (home, ward, ICU)
•  consistency of drug choice for empiric treatment.

Calculation of the PSI is an additional step in the management of the patient and as with guidelines
generally it is unlikely to be taken up widely within an institution or an emergency department without the
help of specific intervention.

For these reasons the NPS is supporting an Australia-wide project to implement the new AG Version 12
for the treatment of community-acquired pneumonia and for treatment of exacerbations of chronic and
acute bronchitis in ED. The VDUEG has been commissioned to carry out this project in 10 Victorian
hospitals as part of a national study in 40 hospitals.

Project details:
The aim of this project is to help emergency departments to implement the calculation of the PSI as a
routine aspect of management of community-acquired pneumonia in adults. The expected outcomes of
the project will be more consistency in placement of patients, increased concordance of antibiotic
prescribing with the national guidelines, decreased use of intravenous antibiotics and decreased length of
stay in hospital. The study will occur in three phases.



4
  ROBERTSON, M.B., KORMAN, T.M., DARTNELL, J.G.A., IOANNIDES-DEMOS, L.L., KIRSA, S.W. & LORD, J.A.V.
(2002b). Treatment of lower respiratory tract infection in emergency departments. In: Proceedings of Australian
Society for Antimicrobials 3rd Annual Scientific Meeting 2002. In Australian Society for Antimicrobials 3rd
Annual Scientific Meeting 2002a. pp. 10. Sydney.
5
  ROBERTSON, M.B., KORMAN, T.M., DARTNELL, J.G., IOANNIDES-DEMOS, L.L., KIRSA, S.W., LORD, J.A.,
MUNAFO, L. & BYRNES, G.B. 2002. Ceftriaxone and cefotaxime use in Victorian hospitals. Med J Aust, 176,
524-9.
6
  Therapeutic Guidelines Limited, (2003). Antibiotics 12. Melbourne.
                                                                                                            61
CAPTION Victoria – Final Report                                                      April, 2006


•   Phase 1 - a pre-intervention questionnaire of doctor knowledge and acceptance of the
recommendations for CAP and an audit of use of the PSI and empiric prescribing for CAP.
• Phase 2 is the intervention phase – education about the CAP guidelines and PSI calculation by an in-
house academic detailer, and implementation of tools to assist in the calculation of the PSI and choice of
antibiotic. Phase 2 will also involve regular audit and feedback to prescribers.
• Phase 3 is optional to all participating hospitals and will commence after phase two. It involves the
implementation of the Antibiotic Guidelines (Version 12) for acute and chronic bronchitis.

Within each participating hospital these activities will be coordinated and carried out by a local project
officer with the support of myself, our clinical champion <Clinical champion name> and the Victorian
Project Officer.

You have been nominated by myself and consultants in Infectious Diseases and Respiratory, to help
implement the interventions within the hospitals. You will be trained in academic detailing by the NPS and
will have access to a range of nationally-developed intervention tools the ED has chosen to implement as
part of Phase 2 of CAPTION. These tools provide an antibiotic decision support that helps simplify the
calculation and use of the PSI scoring system. We have the opportunity to customise one or more of the
intervention tools to our needs with the assistance of the VDUEG.

Please do not hesitate to contact me should you have any further questions on the matter.


Yours Sincerely,



<ED Director>




                                                                                                       62
CAPTION Victoria – Final Report                                                         April, 2006


Appendix 9. Patient record form




                                    CAPTION PROJECT
           Community-Acquired Pneumonia: Towards Improving Outcomes Nationally



                        Patient record form (confidential)
This form is for use at individual participating hospitals only and is highly confidential. Only patients
meeting the inclusion criteria for the CAPTION project should be recorded.

Your hospital code is: ____________ (provided by your state hospital coordinator)

Write the medical record number (MRN) of patients meeting inclusion criteria in column 2. The patient
study number assigned to each eligible patient is then found in column 1. This individual “patient study
number” is to be entered as the answer to question 1 on the data collection form for each patient.

    Do not send this form to CAPTION team; keep for your
                       future reference
Patient study number              Medical Record Number                                Date

               1
               2
               3
               4
               5
               6
               7
               8
               9
              10
              11
              12
              13
              14
              15
              16
              17
              18
              19
              20


                                                                                                            63
CAPTION Victoria – Final Report                                                                    April, 2006


Appendix 10. Strategy to support academic detailing in Victoria


Victorian approach for supporting CAPTION Local Project Officers when undertaking academic detailing:

After having met with each local project officer face to face, the Victorian state project officer plans to provide
support to each hospital in the following manner to ensure that they feel encouraged and supported through out this
phase of the project. Assistance will be tailored to each hospital’s specific needs, also highlighted below.

1- Meetings as a forum of group discussion:
Initially a consultative meeting will be held (6th October) to provide everyone with the opportunity to discuss his or
her concerns and provide a brief outline of when they plan to start undertaking detailing. Local Project Officers
have been contacted and asked to provide State Project Officer with a plan for starting educational visits.
At this meeting we will discuss with the Local Project Officers if they would like contact information for other
detailers or that of expert detailers.
Future meetings will be organised either half way through the detailing process or after the detailing process to
discuss concerns and issues (options will be discussed with Local Project Officers on the 6th of October). We will
also ask if Local Project Officers would like their questions and answers circulated to all the other detailers for
assistance and information, when detailing starts.

2- Telephone contact:
Local Project Officers can call the State Project Officer with regards to any problems with detailing that they
encounter. We have also put in place a voice mailing system so that they can leave messages and be contacted
ASAP.
Local Project Officers have been encouraged to contact the State Project Officer(s) to practice detailing role-plays.

3- Debriefing support:
Be available to de-brief with the detailer after the visits and to provide further literature support as requested. At this
point difficult questions that the detailers were asked and were difficult to answer will be addressed.

4- Email contact:
State Project Officer is available on email before and after visits occur, to check in with detailer as close as possible
after the visits.

5- Other support:
For Hospital AF: provide the feedback presentation and have offered to assist in academic detailing visits. Plan to
visit for one full day and help in detailing and undertaking presentations.
For Hospital AJ have been asked to assist with academic detailing and have trained a second project officer in
detailing. Have approached pharmacy to provide further support with data collection- this will be followed up
further.




                                                                                                                       64
CAPTION Victoria – Final Report                                                     April, 2006


Appendix 11. CAPTION ED doctors’ survey
1. What is your first choice empirical antibiotic therapy for mild (e.g. short of breath on exertion, O2
Saturation > 95%) to moderate (eg short of breath at rest, O2 Sats 91-95%) Community-Acquired
Pneumonia (CAP)?

2. What factors would determine the prescription of Ceftriaxone/Cefotaxime in your regime, why?

3. Are you familiar with the new Antibiotic Guidelines recommendations for the use of the Pneumonia
Severity Index (PSI) and antibiotic prescribing in CAP?

4. What is your first source of information if you are unfamiliar with a prescribing regimen? (Please
number in order of use)
- Clinicians Health Channel
- Mims
- Australian Medicines Handbook
- Therapeutic Guidelines
- Ward Pharmacist
- Care Plan
- On line literature reviews

5. Please choose the antibiotic of choice in the following case study/ Please outline the procedure you
would undertake to carry out prescribing for the following case study
a) 42 yr farmer with sudden fever, rigors, dyspnoea
- Non-smoker with no previous illness
- T39.5 HR130 BP 105/70 RR 30/min
- SaO2 on air 86%
- Consolidation left lower lobe
- WCC 19,000 Na 128, Cr0.13

b) Mrs CRP is a previously well 38-year mother of three who works as a part time teacher. She
smokes. She saw her GP with a three-day history of cough and left sided pleuritic chest pain. The
night before presentation she had an episode of high fever, sweats and was shivering and shaking. Her
GP measured her temperature at 37.8. Clinical signs of a left lower lobe consolidation are confirmed
on Chest X-ray. PR 96, O2 Sats 92%, BP 95/60. Resting respiratory rate is 20. Her GP is not sure if
she requires admission or not and has referred her for assessment and opinion.


6. Please number (1 to 3) what you prefer as a form of intervention in you Emergency Department in
order to assist you in better prescribing for CAP.
- Academic detailing (one-on-one education sessions with opinion leaders and clinical educators)

-   PSI calculator form/Stamp/windows based
-   Web-based Antibiotic approval system
-   Laminated PSI scoring ID card
-   Easier access to the Therapeutic Guidelines
-   Care plans
-   Information posters
-   Other (if so please list)   ___________________________




                                                                                                     65
CAPTION Victoria – Final Report              April, 2006


Appendix 12. Laminated PSI scoring ID card




                                                           66
CAPTION Victoria – Final Report                                              April, 2006


Appendix 13. Generic letter to prescribers



                                Re: CAPTION
      Community-Acquired Pneumonia: Towards Improving Outcomes Nationally


<Hospital name & address>


<Date>


Dear Dr _____________

The Victorian Drug Usage Evaluation Group (VDUEG) in conjunction with the National
Prescribing Service (NPS) would like to make you aware of an exciting new community-
acquired pneumonia (CAP) project that <Hospital name> is participating in. The CAPTION
project is aimed at implementing the use of the Pneumonia Severity Index (PSI) and
Therapeutic Guidelines: Antibiotic, Version 12, 2003, recommendations for treating CAP.

Background:
A study of the antibiotic treatment of lower respiratory tract infections in emergency
departments was conducted in April - July 2001 by the VDUEG, with funding provided by
the NPS. Results indicated a particularly low rate of concordance with the national antibiotic
prescribing guidelines (Therapeutic Guidelines: Antibiotic, Version 11, 2000).

Since the time of this study there have been major changes to the recommendations for
treatment of CAP in the current edition of the guidelines involving more precise assessment
of the severity of the illness using the PSI (Therapeutic Guidelines: Antibiotic, Version 12,
2003). The PSI is a scoring system that involves assigning a numerical value to elements of
clinical history and simple physical signs that are commonly assessed when patients present
with suspected CAP.

The advantages of using the PSI are many including:

      Reducing uncertainty in the assessment of illness severity
      Better placement of patients (home, ward, ICU)
      Consistency of drug choice for empiric treatment.

Calculation of the PSI is an additional step in the management of the patient and as with
guidelines generally it is unlikely to be taken up widely within an institution or an emergency
department without the help of specific intervention.

For these reasons the NPS is supporting an Australia-wide project to implement the
Therapeutic Guidelines: Antibiotic, Version 12, 2003, for treatment of CAP in emergency
departments over a two-year period. The VDUEG has been commissioned to carry out this
project in up to 10 Victorian hospitals as part of a national study in 40 hospitals.

Project details:
Through a planned intervention process the VDUEG aims to simplify the use of PSI for
prescribing purposes and hence to increase physician confidence in assessment and
prescribing of antibiotics for CAP. As a result of this approach it is hoped to not only improve
                                                                                             67
CAPTION Victoria – Final Report                                          April, 2006


patient placement and diagnosis but also potentially decrease length of stay and the length
of therapy within the hospital as a result of greater use of the guidelines.
It is expected that the study will occur in two parts.

      Part 1 (already completed) - a pre-intervention questionnaire of emergency department
doctor knowledge and acceptance of the recommendations for CAP and an audit of use of
the PSI and empiric prescribing for CAP.
      Part 2 is the intervention phase – education about the CAP guidelines and PSI
calculation by an in-house academic detailer, and implementation of tools to assist in the
calculation of the PSI and choice of antibiotic. Part 2 will also involve regular audit and
feedback to prescribers.

The interventions are being implemented with the assistance of the <Hospital name> clinical
champion/project officer, <name(s)>. CAPTION hospital project officers have been trained in
social marketing skills, also known as academic detailing, by the NPS and have access to a
range of nationally-developed educational tools that will be implemented as a component of
Part 2. Such tools include computer-based systems for antibiotic decision support or simply
for calculating the PSI, posters and ID card size PSI calculators. While we would like to
extend the academic detailing process to relevant medical staff external to the ED, e.g.
infectious diseases and respiratory consultants and medical registrars, owing to limited
resources it may not be possible to do so, hence the provision of an introductory CAPTION
project letter.

Please do not hesitate to contact me or <name(s)> (ph: ________, email ________) if you have
any queries regarding the CAPTION project,

Yours Sincerely,




Marion Robertson
Chair, Victorian DUE Group
Drug Usage Evaluation Pharmacist
The Royal Melbourne Hospital
Phone: 03 9342 7612
Email: Marion.Robertson@mh.org.au




                                                                                        68
CAPTION Victoria – Final Report                                         April, 2006


Appendix 14. Publications

Research paper
Maxwell DJ, McIntosh KA, Pulver L, Easton KL on behalf of the CAPTION Study Group.
Empiric management of community-acquired pneumonia in Australian hospital emergency
departments. Med J Aust 2005, 183(10):520-524.


Letters to the editor
KA McIntosh, N Jamshidi, DJ Maxwell, KI Kaye, KL Easton, LA Stanton, GM Peterson, WB
Dollman, PJ O’Connor, L Pulver, MB Robertson. Optimising community-acquired pneumonia
management. Fiji Medical Journal, 2005, 24(1):31-34.


Abstracts from conference proceedings
Easton KL, Mackson JM, Stanton LA, Peterson GM, Maxwell DJ, Kaye KI, McIntosh KA,
Jamshidi N, Kirsa SW, Dollman WB, O’Connor PJ, Pulver L, Robertson MB. Community-
acquired pneumonia: towards improving outcomes nationally (the CAPTION project). Clinical
and Experimental Pharmacology and Physiology (CEPP) Conference, Brisbane,
2004;31:A92.

McIntosh KA, Jamshidi N, Maxwell DJ, Kaye KI, Kirsa SW, Easton KL, Stanton LA, Peterson
GM, Dollman WB, O’Connor PJ, Pulver L, Robertson MB. Management of Community-
Acquired Pneumonia in Emergency Departments – CAPTION. Australian Society for
Antimicrobials Annual Scientific Meeting, Lorne, Victoria, 2005.

Maxwell DJ, Kaye KI, McIntosh KA, Pulver L, Stanton LA, Marwood A, Wai A. For starters:
the NSW/ACT arm of CAPTION. A taste of things to come. Society of Hospital Pharmacists
of Australia National Conference, Brisbane, November 2005.

Maxwell DJ, Kaye KI, McIntosh KA, Robertson MB, Stanton LA, Peterson GM, Marwood AC,
Dollman WB, Pulver LK, Tett SE, Wai A, Horn F, Mackson JM. Influencing antibiotic
prescribing in the management of community-acquired pneumonia in hospital emergency
departments - The CAPTION project. Australian Society for Antimicrobials Annual Scientific
Meeting, Sydney, New South Wales, February 23-25, 2006.


Local Victorian hospital presentations
Robertson MB, Irving LB, Harrod R, McIntosh KA, Jamshidi N, Maxwell DJ, Kaye KI, Kirsa
SW, Easton KL, Mackson JM, Wai A, Stanton LA, Peterson GM, Dollman WB, O'Connor PJ,
Pulver LK. The CAPTION Project. Management of community-acquired pneumonia in
emergency departments. How does the Royal Melbourne Hospital compare with other
Australian hospitals? 30 June – 6 July 2005, Melbourne Health Research Week, Melbourne,
Victoria.

Haji D, Caplygin F, Goh L, Lam S and Broughton S. Community-acquired pneumonia:
towards improving outcomes nationally, CAPTION PROJECT and the Peninsula Health
experience. November 2005, Peninsula Health Research Week, Melbourne, Victoria. (note:
this presentation won a prize for best scientific poster)




                                                                                       69