Performance Improvement CME Model for the by nikeborome


									Performance Improvement CME Model for the
           Community Hospital

         Karen Scheidt, CCMEP, MS Ed
          Waukesha Memorial Hospital
             725 American Avenue
             Waukesha, WI 53188

Introduction: Intersection of QI and CME at the community hospital level is
novel, and carries its own unique challenges and rewards. This study provides a
PI CME model attempted by the Emergency Department physicians at a
community hospital through a sepsis management improvement project.
Discussed are how the PI CME team was assembled, what novel educational
tools were created, what barriers were overcome, and what unique
circumstances exist for PI CME at the community hospital level.

Methods: Initial baseline data was collected retrospectively from review of 154
medical charts for a defined set of parameters from a six-month period. Data was
presented both for the individuals and collectively for the group, goals were
identified, and interventions planned by the group. Learning from assessing the
current gaps occurred, educational and process interventions were developed,
and data was re-assessed via a second set of 150 chart reviews. Physicians also
self-reflected on their own cases where data fell out of desired parameters.

Results: Four physicians (36 % of group) from the ED completed all three stages
of the project, which was the first PI CME attempted at this hospital. Practice
impact was evident as lactate testing i ncreased 13.4%, antibiotics administered
within three hours increased 7%, and outcomes were attained that were not

Discussion: This project required integration of practice and performance
improvement processes with CME, not previously achieved at this organization.
The group stated a desire to identify a future reportable quality initiative that
could be combined with PI CME.

Key Words: Continuing Medical Education, community hospital model,
performance improvement, sepsis management improvement.
                                    Project content

Project description:

        The primary goal of this project was to improve the use of the identified
early goal directed therapy measures (identified by the IHI) that are part of the
sepsis screening tool in place in the Emergency Department. Early treatment for
sepsis is necessary for best potential outcomes for the patient, but the diagnosis
is often difficult as the symptoms can be mimicked by other diseases.

The following competencies were addressed:
    Patient care
    Medical knowledge
    Practice-Based learning and improvement
    Systems–Based practice

The level of educational outcomes intended included:
    Participation
    Knowledge/ skill/ attitude
    Competence
    Performance

Team Members:
Madelyn Sieraski, MD: Physician Champion, Vice Chair Education and Library Committee
Steven Kulick, MD: Medical Director of Quality, ED Physician, Project Physician Champion
Heidi Young, RN: Medical Staff Quality Reviewer
Karen Scheidt, CCMEP, MS Ed: CME Coordinator

There was no commercial support for this project, and no ongoing costs other
than the departmental support of personnel time diverted from other work to this

Target Audience:
        The Emergency Department physicians were the targeted group for this
project. They are a contracted group of eleven physicians supported by seven
mid-level providers that serve a level III ED, having approximately 40,000 ED
visits per year.

Methodology (process):
   Needs identification- 2007 IHI Sepsis Campaign identified new goals for
     early goal directed therapy, and the quality department along with the ED
     staff identified the need to bring the current processes in line with the new
     IHI goals.
   Objective identification- Medical Director of Quality along with ED Medical
     Director identified the four most crucial parameters in the IHI early goal
     directed therapy, also part of a newly created sepsis screening tool.
       Parameter goals were set to be in line with IHI recommendations, with
       input and agreement from ED participants.
      Intervention process- included the following:
           o PowerPoint presentation of IHI early goal directed therapy with
               evidence-based research included, with group discussion
               afterward. Participants requested a webcast of the presentation,
               and were required to review again individually for Stage A credit.
           o Individual self-reflective review of charts in both baseline and follow
               up data which contained parameters not meeting goal
           o Work station posted and pocket size flow charts of the early goal
               directed therapy
      Educational evaluation:
           o Self-reflective chart reviews included questions on self-evaluation,
               self and group reflection and discussion on pre-and post data,
               group discussion on barriers and keys to improvement
      Sample size of the project: We included 100% of the 154 chart data for all
       identified sepsis cases meeting the screening criteria for all participants for
       the six month period prior to the beginning of the project for the baseline
       data. We then included similar data for the 150 charts reviewed six
       months post implementation for improvement measurement.
      Statistical Procedure used to evaluate the data: simple data spreadsheet
       in Excel, entered and tallied by our Medical Staff Quality Reviewer. Group
       and individual data summary charts created form Excel.
      IRB involvement- we did not have IRB involvement

Results (outcomes) :
    Summary of data results for the project-
          o Lactate testing with a goal of 100% improved from 82% compliance
             to 93% compliance, a 13.4 % increase.
          o Antibiotics administered within 3 hours with a goal of 100%
             increased from 71% compliance to 76% compliance, a 7.0%
          o Blood cultures drawn before antibiotics administered with a goal of
             100% started at 98% compliance and the group maintained this
             level of compliance, which was the goal.
          o The fourth parameter (fluid resuscitation) was dropped from official
             measurement due to difficulties encountered in measurement.
    Brief discussion of the data
          o We felt the more dramatic results with the lactate (and the ability of
             the physician to maintain the high level of blood cultures drawn
             before antibiotics) were due to this parameter having more
             individual physician control, unlike the other parameters which are
             much more affected by system issues.
          o There were additional gains not specifically described in our goals:
                    Nursing processes improved; they were placing the
                     screening tool on the chart more appropriately and doctors
                     were using them.
                    Central lines were placed more appropriately and frequently;
                     CVP monitoring noted more often than previous, leading to
                     better hypotension management.
                    Documentation by physicians in ED dictated notes reflected
                     more awareness and use of sepsis screening tool.

Implications for future Education
    Limitations of the project
           o Number of parameters was still too large; some were interrelated
              and hard to isolate impact of one factor.
           o As barriers were identified, who would be responsible for follow up
              was not clear, nor how it would be accomplished. We needed to
              have more department representation at the planning table up front.
           o Some systems issues were too large to address in this project; i.e.
              lab turn around time for WBC results.
           o Using current meeting structure was effective, but can also make
              the discussion rushed.
           o Interim information was labor intensive to provide.
    Benefits of the project for future educational activities
           o We learned structure and guidance were key for assisting physician
              to complete any or all of the stages. They needed to be walked
              through every detail.
           o Physicians were exposed to a group effort in performance
           o Physicians initially were very focused on barriers to improvement
              and success, but as the project continued, they increased their self-
              examination on what they could do to improve their won practices.

The Community Hospital Difference
        o The culture tends to be one of independence; group efforts just on
          the horizon. Physician used to a lot of individual control.
        o Can be difficult to find a physician champion, fewer options.
        o Can be a lot of preparation to try to get buy in; bottom up approach
          is more work that a top down approach.
        o Most physicians are still uneducated about PI CME.
        o There is little overlap of quality and CME at a formal level, although
          for us discussions are beginning.
        o The Community Hospital environment is generally limited in
          resources at all levels.
        o The CME office (usually one person) will face the bulk of the
          planning, paperwork, communication, and oversight of the project.
        o A small team can be much burdened; One (maybe two) project(s)
          per year would be feasible.
Appendix (1-4 pages-optional) – include actual data collection instruments for
needs assessment and or educational outcomes measurement

1. Reference created by Karen Scheidt to educate participants on how PI CME
models Quality processes, Pre-Stage A.

2. Blinded Baseline Data (group and individual) provided by Heidi Young for
review prior to Stage A.
3. Form we created for recording of Stage A discussion and planning. Since
this time, the Wisconsin State Medical Society has provided us with templates
that we will use in future programs. (see attachments in the email)

                 Physician Participation Form - Stage A: (5 Credits)
                                 November 17, 2008

Present :
Dr. Kulick, Dr. Sore, Dr. Berger, , Dr. Mateer, Dr. Mausner, Dr. Schultz, Dr. Moehring, Dr. Tonsfeldt, Dr.
Szatkowski, Dr. Hucksdorf,

Di scussion Verified by: Karen Bazzetta, CME Coordinator, Steve Kulick MD ED, Heidi
Young, Quality Reviewer
 Dr. Kulick presented support data on seriousness of sepsis, mortality similar to MI, but not
commonly viewed that way. Especially important; management of hy potension, and door to
hang time of antibiotics in preventing deat h.

Summary/review of patient data:
Lactate drawn: group average for Jan – June 08: 82%. E ven though the last month measured
the average was 100%, lactate will continued to be monitored to keep awareness high. Goal
of 100% for Jan-June of 09.

Door to hang time; antibiotics- group average Jan – Jun 08: 154 minutes. Most team
members had an individual average less than the previous standard of 180 mins. The new
goal from IHI is now 60 mins, so much more aggressive action is needed to get antibiotics
hung more quickly. It was mentioned that we should examine ot her organizations data for
best practice trans formation of sepsis; to help us on the path of accomplishing the new goal
of one hour.

A verage minutes of hypotension: group average Jan -June 08: 75 minutes. There are no
actual standards for this value, but the group goal is to increase awareness of the need to
manage hypotension aggressively and quickly.

Blood cultures drawn before antibiotics given – group average Jan –June 08: 97%.
Recommended to continue to be aware and keep up the good work. No monitoring needed at
this time.

Identify problems and hypothe size on issue s/barriers relating to problems: di scu ssi on
     It was mentioned that when a patient is obviously septic upon admission, lactate is not
        necessary to confirm diagnosis (aft er-t hought). But discussion followed that the how
        increased the lactate value is can help determine severity and cours e of acti on. A lactate
        of 4 or above should lead to a central line being placed before transfer to the ICU.
     Discussion included that sometimes it is difficult to get antibiotics to the ED, timing is not
        always good; example of Zosyn, which must be refrigerated and must currently be sent
        from pharmacy. Sometimes patients don’t look septic or sick when they arrive, but
        wors en as they stay in ED, and antibiotics are then ordered, causing a “delay”.
        Discussion ensued about the philosophy of prophylacticly starting antibiotics, as in
        pneumonia, for suspected cases of sepsis, before all diagnostic studies are completed.
     Discussion was centered on the fact that since there are no firm goals regarding duration
        of hypotension, physician response (in terms of amount of fluid and pressor support, and
        how quickly) should be the marker.
List non-education strategies to enhance change as an adjunct to activitie s/educational
interventions (e.g., reminders, patient feedback).
     It was asked if the feedback on data could be more often, and not just at the end of
        the six months, by the time data is reviewed, the physician can’t remember the
        cases. More feedback, more often will allow the physician to re-access any cases
        more effectively. List (by individual) of all the charts reviewed for the data was

Identify factors outside the physician’ s control that impact on patient outcome s.
     Pharmacy issues: related to how quickly they get the antibiotic and/or pressors to the ED
        after they are ordered.
     Triage issues; it was discussed that if nursing can pull and us e screening tool, it could
        save muc h time in assessing sepsis more quickly. This tool was designed for nursing and
        physicians, and MD does not have to sign it. Could nursing do more of the paperwork to
        save time?
     Pump issues: highest setting for the pump reads out 999, and this doesn’t always seem
        to be delivering the fluids fast enough. Questions were asked about a new type of pump
        that is coming out, better? P ressure bags were mentioned as a way to push the fluids
        faster, and it was mentioned that fluid overload should be seen as a less serious issue
        than under-resuscitation.
     Lab issues; turnaround time for important tests can cause delay; esp. WBC. Lactate can
        be done on I-Stat, but test needs improvement, and possibly repair. The lab needs to be
        involved to be able to meet one hour goal for door to hang time.

Identify potential educational strategies to remove, overcome or address barriers to
practice change.
     Participate in a group review of the data on outcomes related to these parameters, to
        be available via a webcast. To include recommendations from the IHI.
     Review of case data individually, where own data had fallen out of the recommended

List other stakeholders for collaboration and cooperation .
     Laboratory, Pharmacy, Nursing

GROUP statement of process and results of assessment of current practice
When examining the data on sepsis, I recognized that our combined practice varies from:
   Ordering blood cultures prior to antibiotic administration (Goal 100% )
   Ordering lactate level for suspected sepsis patients (Goal 100%)
   A verage Door to Hang Time- antibiotics (Stretch Goal one hour per new Surviving Sepsis
      Goals) ___154 mins _______
   A verage Minutes of Hypotension (Goal- decrease in time)
      ________75 mins____________

Waukesha M emorial Hospital is accredited by the Wisconsin M edical Society to provide Continuing M edical
Education for physicians.

Waukesha M emorial Hospital designates this educational activity for a maximum of 5.00 AM A PRA Category 1
Credit(s)TM. Physicians should only claim credit commensurate with the extent of their participations in the activity.
4. Stage B interim data charts supplied by Heidi Young:

   5. Letter and form used for Stage B chart reviews supplied by Karen Scheidt:

                           Stage B: (5 Credits)
                       Complete for January 19, 2009

Performance Improvement CME is a three stage educational process, and
any case review as part of this process is strictly for educational
purposes only; not for any committee or peer review.

Currently, sepsis management is not a publically reportable quality data.

As part of this Stage B, Learning from Application of Performance
Improvement to Patient Care, a link to a short webcast created uniquely for
this program will be sent to you for your review. You can click on the “links”
box under the speaker screen, and submit your evaluation for record of
review of the webcast.

Secondly, attached is a list of the charts reviewed from Jan 2008- June 2008
used to assimilate the base data presented to you at the last division meeting.
Charts are highlighted that would be applicable for review of outliers. We
request you choose the highlighted charts for review, and complete the
questions listed below. The end portion of the webcast will contain some
additional suggestions on participating in focused chart reviews.

Upon completion of these two educational activities, and the accompanying
reflection on practice improvement, you will be awarded the Stage B credit of
5 CME.

Please bring completed chart review forms with you to the January 19,
2009 ED Division Meeting.


Case Number Reviewed: __________________________________

   1. What factors played in to any delays in management? For
      example, prolonged hypertension, delays in time to antibiotic
      administration, etc.

   2. How could the management of this patient be optimized?

   3. What new insights have you gained in sepsis management from
      case review?
6. Blinded Post-Intervention Data (group and individual) provided by Heidi
   Young for review prior to Stage C discussion.
   7. Letter and form used for Stage B chart reviews supplied by Karen Scheidt:

To the ED Physicians,

Please find enclosed your chart data from your sepsis cases January 09-June 09
as part of the Performance Improvement CME Sepsis program. You are
Physician _______ (confidential). We will be discussing this blinded
information at the ED Division meeting Monday September 21, 2009.

Highlighted are the parameters selected to be reviewed as part of the sepsis
management project:
    Lactate level drawn
    Admit to antibiotic hang time (in minutes)
    Antibiotics given before Blood cultures drawn
    Hypotension episodes (if present), and hypotension management
        CVP line inserted in ED (when indicated)

Stage C in the PI/CME process, learning from evaluation of PI effort, includes
the following:
     Review of re-collected chart data (from Jan-June 2009)
     Reflect on practice impact, and patient outcomes
     Identify further issues/barriers that need addressing

Performance Improvement CME is a three stage educational process, and any ca se review
as part of thi s proce ss i s strictl y for educational purpose s only; not for any committee or
peer review. Currently, sepsis management is not publicly reportable quality data.

Please review this data and complete chart reviews from cases containing
a parameter(s) not meeting goal and complete the questions. If you do not
have cases with outliers, please choose cases from your list to review for
further learning opportunities, and share strategies that have worked for
you in meeting sepsis goals.

Please contact me with any further questions you may have.

Karen Scheidt (Bazzetta)
CME Coordinator
Physician Name______________________________________ __________
                                     Stage C: (5 Credits)
Chart Number Reviewed: ________________________________________

Identify Parameter(s) that did not meet goal on this patient:
 Lactate not drawn
 Hypotension minute s too long
 Admit time to antibiotic too long
 CVP line should have been inserted in the ED
 Blood Cultures not drawn before Antibiotics

    4.   What factor(s) played in to this parameter not meeting goal? Add any further
         comments or situations involved in this ca se or thi s parameter.

    5.   How could the management of this patient be optimized? Include strategies
         that could impact sepsi s management.

    6.   What problems and issue s/barriers still exist in managing a septic patient?
7.   What new insights have you gained in sepsi s management from case review ?

To top