Docstoc

Root Cause Analysis Investigation Tools

Document Sample
Root Cause Analysis Investigation Tools Powered By Docstoc
					Root Cause Analysis Investigation Tools

   Guide to investigation report writing
          following Root Cause Analysis
              of patient safety incidents

www.npsa.nhs.uk/nrls
Root Cause Analysis Investigation Tools                                      www.npsa.nhs.uk/nrls




Contents
Section 1: Introduction                                                 3
      1.1 How to use this guide                                         3

Section 2: Overview                                                     4
      2.1 The purpose of the investigation report                       4
      2.2 Principles of investigation                                   4
      2.3 Hindsight and outcome bias                                    4
      2.4 Audiences                                                     5
      2.5 Report writing and presentation style                         5
      2.6 Record keeping and information security                       5

Section 3: Guidance on the sections needed in a report                   6
3a    Cover page                                                         6
3b    Contents                                                           6
      3.1 Executive summary                                              6
3c    Main report                                                        6
      3.2 Incident description and its consequences                      6
      3.3 Pre-investigation risk assessment                              6
      3.4 Background and context of the incident                         6
      3.5 Terms of reference                                             6
      3.6 The investigation team                                         7
      3.7 The scope and level of the investigation                       7
      3.8 The investigation type, process and methods used               8
      3.9 Involvement and support of the patient, relatives or carers    8
      3.10 Involvement and support for staff involved in the incident    8
      3.11 Information and evidence gathered                             9
      3.12 Chronology of events leading up to the incident               9
      3.13 Detection of incident                                         9
      3.14 Notable practice within the case                              9
      3.15 Care and service delivery problems                           10
      3.16 Contributory factors                                         10
      3.17 Root causes                                                  10
      3.18 Lessons learned                                              10
      3.19 Recommendations                                              11
      3.20 Arrangements for shared learning                             11
      3.21 Distribution list                                            11
      3.22 Investigation report appendices                              11

Section 4: Next steps                                                   12
      4.1    Action planning and solutions development                  12
      4.2    Action plan / impact assessment                            12
      4.3    Implementation, monitoring and evaluation arrangements     12

Section 5: Appendices                                                   13
Appendix 1 Prompts for terms of reference                               13
Appendix 2 Prompts for investigation report distribution list           14

Section 6: Bibliography                                                 15

Section 7: Acknowledgements                                             16




                                                                                                    2
www.npsa.nhs.uk/nrls                                                                               Root Cause Analysis Investigation Tools




Section 1: Introduction
1.1 How to use this guide

The purpose of this guide is to provide practical help and support to those writing patient
safety Root Cause Analysis (RCA) investigation reports. It may also prove useful to those
writing Significant Event Audit (SEA) reports.


The guidance is provided as:
•	 an	overview	and	general	background	advice;
•	 guidance	on	the	sections	needed	in	an	investigation	report.
It is useful background reading for those new to investigation. The guidance is also designed for quick reference, providing
notes	on	eachsection	of	the	investigation	report;	for	access	as	and	when	required.
The following associated documents will also be useful and are available from www.npsa.nhs.uk/rca


•	 Guide to investigation report writing following RCA: Guidance on completing an investigation report.
•	 Three levels of RCA investigation: Designed to help when considering what level of detail is appropriate and propor-
   tionate when investigating an incident.
•	 Concise RCA investigation report examples: For demonstration purposes.




3
Root Cause Analysis Investigation Tools                                                                                        www.npsa.nhs.uk/nrls




Section 2: Overview
2.1 The purpose of the                      2.2 Principles of investigation                               2.3 Hindsight and
investigation report                        A key purpose of the patient safety RCA                       outcome bias
The investigation report presents the       investigation and subsequent report is                        It is important when analysing
culmination of all the work undertaken      to share learning from patient safety                         investigation	findings	to	be	aware	
by the investigation team. It conveys       incidents, claims and complaints.                             of, and try to avoid, hindsight and
all necessary information about the         Before writing the report, it is useful                       outcome bias.
incident, the investigation process and     for the investigator or team to
the outcome of the investigation.                                                                         Hindsight bias
                                            consider whether general principles
The audience will use the investigation     of investigation have been followed.
                                                                                                              Hindsight bias is the tendency for
report as the basis for judging the         Include the following in your                                     people	with	the	‘benefit	of	hindsight’	to	
quality of the investigation process,       considerations:                                                   falsely believe, once all the facts become
the	findings,	conclusions	and	                                                                                clear, that the actions that should have
                                            Some general principles of
recommendations. The audience will                                                                            been taken to prevent an incident
                                            investigation                                                     seem obvious, or that they could have
also judge the competence of the
                                                                                                              predicted the outcome of an event2.
investigation team by the content,              Was the investigation
style and tone of the report.                   process conducted with                                        Although considered a serious pitfall
                                                                                        Yes      No
                                                the appropriate level of                                      in investigation terms, hindsight bias
The purpose of the report is to provide:        independence?1                                                has been documented as a potentially
   a
•	 	 	formal	record	of	the	investigation	                                                                     useful mechanism in terms of the
   process;	                                    Was the investigation                                         specific	focus	of	learning	from	incidents.	
•	 a	means	of	sharing	the	learning.             process proportionate                                         Hoffrage, et al3 argue that it is a
                                                                                        Yes      No           by-product of an adaptive mechanism
                                                to the incident and any
The report should explain                       associated risks?                                             that can make human memory more
1. what happened (i.e. chronology of                                                                          efficient.	The	basic	idea	of	this	‘RAFT’	
   events);	                                    Did the investigation                                         model (Reconstruction After Feedback
2. who	it	happened	to;                          begin and end in a timely               Yes      No           with Take the Best) is that any feedback
3. when	it	happened;                            manner?                                                       or correct information received (in
4. where	it	happened;	                                                                                        this case in the form of a now known,
                                                Was the investigation
5. how it happened (i.e. what went                                                                            but previously unpredicted, incident
                                                process open and                        Yes      No
                                                                                                              outcome) is used to automatically
   wrong);	                                     transparent?
                                                                                                              update	a	person’s	knowledge	base.
6. why it happened (i.e. what
   underlying, contributory or deep-            Did the investigation                                         It is important to remember, however,
   rooted factors caused things to go           team kept relevant parties              Yes      No           that failure to recognise hindsight bias
   wrong).                                      appropriately informed?                                       in incident investigation can result in
                                                                                                              misinterpretation	of	findings	and	may	
The report should be clear and logical,         Was the investigation                                         ultimately mask the real lessons to be
and demonstrate that an open and fair                                                   Yes      No           learned.
                                                based on evidence?
approach has taken place.
Help box                                        Did the investigation look
                                                for improvements and not                Yes      No
                                                to apportion blame?
  Unless there are specific exceptions,
  the patient or family of a patient
  have a right to the full investigation                                                                  2
                                                                                                              Fischoff, B. (1975). ‘Hindsight is not foresight: the
  report as defined in the Data                                                                               effect of outcome knowledge on judgement under
  Protection Act 1998 (available from                                                                         uncertainty.’	Journal of Experimental Psychology,
  www.ico.gov.uk).                                                                                            Human Perception and Performance 1(3): 288-299.
                                                                                                              Henriksen, K. and H. Kaplan (2003). ‘Hindsight bias,
                                                                                                              outcome	knowledge	and	adaptive	learning.’	Quality &
                                            1
                                                An element of independence to the ward, unit or
                                                                                                              Safety in Health Care 12 Suppl 2: ii46-50
                                                service where the incident took place. This is distinct
                                                from an independent inquiry which meets HSG(94)27             3
                                                                                                                  Ulrich Hoffrage, Ph.D. , Ralph Hertwig, Ph.D., and
                                                criteria and Investigation of adverse events in mental
                                                                                                                  Gerd Gigerenzer, Ph.D., Max Planck, Institute for
                                                health services (June 2005 - amends paras 33-36
                                                                                                                  Human Development. (2000). ‘Hindsight Bias: A
                                                of circular).
                                                                                                                  By-Product	of	Knowledge	Updating?,’	Journal of
                                                                                                                  Experimental Psychology: Learning, Memory and
                                                                                                                  Cognition, Vol. 26, No. 3: 566-581



                                                                                                                                                                       4
www.npsa.nhs.uk/nrls                                                                                     Root Cause Analysis Investigation Tools




Outcome bias                                    2.5 Report writing and                      Whilst a report must be evidence-
                                                                                            based, the lead investigator/
    Outcome bias is the tendency to judge a     presentation style                          investigating team are required to do
    past decision or action by its success or
    failure, instead of based on the quality    The report should be written in the         more	than	simply	summarise	findings,	
    of the decision made at the time. No        organisation style and font (a clear,       but must not move into speculation.
    decision maker knows for sure whether       good print size gives best access to        Terms such as ‘from the evidence it
    or not the future will turn out for the     everyone). Presentation style pointers      was	observed/	concluded’	are	useful	
    best following any decision they make.      are listed below.                           for distinguishing analysis from
    Individuals whose judgments are
                                                                                            evidence. Source material, evidence,
                                                Report presentation style checklist         and theories which back up analysis
    influenced	by	outcome	bias	can	hold	
    decision makers responsible for events       •	 organisation	name;                      should be appropriately referenced.
    beyond their control.                        •	 date	of	the	incident;
                                                                                            Prior	to	final	release,	the	report	author	
                                                 •	 incident	code	or	number;
    Similarly, if an incident leads to death     •	 author(s)	of	the	report;                must	arrange	for	the	final	draft	to	be	
    it is often considered very differently      •	 date	of	the	report;                     proof-read and checked for factual
    and critically, compared to an incident      •	 annotated	page	numbers;                 accuracy, grammar and spelling.
    that results in no harm, even where the      •	 	 umbered	paragraphs	to	aid	
                                                    n
    incident or error is exactly the same.          referencing;
    When people are judged one way when             c
                                                 •	 	 omputer	file	path	to	indicate	        2.6 Record keeping and
    the outcome is good, and another when           where	the	report	is	stored;
                                                 •	 	 tatus,	for	example	‘draft’	(with	
                                                    s                                       information security
    the outcome is poor, accountability
    levels become inconsistent and unfair.          number	of	draft	version)	or	‘final’.
                                                                                            Working documents, such as timelines
    To	avoid	the	influence	of	outcome	bias,	                                                and analytical work used in the
    one should evaluate the decision or         Bullet points are appropriate for           investigation,	should	be	filed	and	
    action taken at the time it was taken       sections of the report conveying lists of   stored safely, and clearly labelled with
    and given what was known or going on        facts	or	findings,	but	free	text	is	more	   the investigation code and number. If
    at that time, irrespective of the success   appropriate elsewhere.                      there is outside scrutiny or a further
    or failure of the outcome.                                                              investigation, working records and
                                                Reports should be written in the third      evidence may be needed and should
                                                person	e.g.	refer	to	‘the	patient’,	        be	easy	to	find.
2.4 Audiences                                   ‘the	doctor’,	‘the	organisation’,	‘the	
                                                investigating	team’	rather	than	‘I’,	       Documentation should be stored in a
A discussion should have already                ‘we’	or	‘you’.                              lettered	or	numbered	index	file,	with	
taken place with the commissioners                                                          each item of evidence given an
of the investigation to agree who               Names of staff should not typically         individual reference.
the	audience	of	the	final	report	will	          feature in the investigation report.
be. Knowing who the report is being             Location, exact title or gender, e.g.
shared with and who will read it helps          ‘Charge	Nurse	Y	in	ITU’,	can	identify	
the investigation team to decide on the         individuals, particularly in specialist
style of the report.                            departments or roles. More general
                                                terms	such	as	‘the	nurse	in	charge’,	
As a rule, keep the report clear, free          or	‘Ms	Y’	or	‘Dr	X’	may	be	more	
of jargon, acronyms and names, and              acceptable. A key to these terms
use plain English. Where technical              must be retained as part of the
terms are necessary, a glossary may be          investigation	file.
required.
                                                An acceptable pseudonym for the
The report should not assume the                patient is best agreed with the patient
reader understands normal processes             or family themselves. Sometimes the
in the department or the normal                 family	may	prefer	a	real	first	or	full	
progress	of	the	patient’s	condition;	           name to be used.
these need to be clearly explained in
a way lay people can understand in              The report should ensure it presents
order to put the incident in context.           the patient(s) or staff involved
                                                as individuals, without being
                                                overly personal or compromising
                                                confidentiality.




5
Root Cause Analysis Investigation Tools                                                                      www.npsa.nhs.uk/nrls




Section 3: Guidance on the sections
needed in a report
3a Cover page                              What to include in the description of          These should be agreed between the
                                           the patient safety incident                    commissioner and the investigation
Contents of a cover page                                                                  lead prior to the investigation.
                                               a
                                            •	 	 	concise	description	of	the	
 •	 organisation	name;
                                                                                          Have regard for previous internal
                                               incident;
 •	 	 itle	and/or	brief	outline	of	the	
    t                                       •	 incident	date;                             investigations	findings,	and	identify:
    incident;                               •	 incident	type;                                s
                                                                                          •	 	 pecific	problem	or	issues	to	
 •	 incident	date;                          •	 	 ealthcare	specialty	in	which	the	
                                               h                                             be	addressed;
 •	 incident	number;                           incident	occurred;                            w
                                                                                          •	 	 ho	commissioned	the	
 •	 author(s)	of	the	report;	               •	 	 ctual	effect	of	the	incident	on	
                                               a                                             investigation (and at which level in
 •	 date	of	the	report;                        patient	and/or	service	and	others;            the	organisation);
 •	 annotated	page	numbers;                    a
                                            •	 	 ctual	severity	rating	of	the	incident	   •	 Investigation	lead	and	team;
 •	 version	number	(draft	or	final);           (consequences).                               a
                                                                                          •	 	 ims	and	objectives	of	the	
 •	 computer	electronic	file	pathway.                                                        investigation and desired outputs
                                           The impact and consequences                       (see	examples	below);
3b          Contents                       described should only be those                    s
                                                                                          •	 	 cope	(see	3.7)	and	boundaries	
                                           relevant to the incident and may not              beyond which the investigation
3.1         Executive summary              solely be based on physical harm. For             should not go (e.g. disciplinary
                                           example, psychological injury, social or          process);
There should always be an executive        political consequences, or reputation
summary at the beginning of any full                                                         t
                                                                                          •	 	 imescales	for	the	report	and	
                                           of service or individuals might also be           for reviewing progress on the
report. This should comprise one or        considered:
two pages only, listing key points,                                                          action	plan;
                                               a
                                           •	 	 void	emotional,	judgemental	                 p
                                                                                          •	 	 roject	administration	arrange-
under these headings:                          or value laden words to describe
•	 	ncident	description	and	conse-
   I                                                                                         ments (including accountability,
                                               events;                                       meetings, resources, reporting and
   quences. A summary including the            c
                                           •	 	 onsider	the	careful	and	limited	
   following:                                                                                monitoring	arrangements);
                                               use of photographs or diagrams to          •	 timescales;
	 o	 brief	incident	description;               support the description.
	 o	 incident	date;                                                                          a
                                                                                          •	 	 ctual	or	potential	for	involvement	
	 o	 incident	type;                        3.3 Pre-investigation risk                        of the police, Health and Safety
   o healthcare specialty where                                                              Executive and plans for this to be
                                           assessment                                        addressed and managed effectively
       incident	occurred;
   o actual effect on the patient          A baseline assessment of the                      at the earliest point.5
       and/or	service;                     incident should be conducted by the            Help box
   o actual incident severity.             investigation team to estimate the
   L
•	 	 evel	of	investigation	(level	1:	      realistic likelihood and consequence               North and East Yorkshire and
   concise;	level	2:	comprehensive;	       of recurrence (prior to preventative               Northern Lincolnshire Trusts have
   level 3: independent).4                 action), and to help assess the level/             identified a useful list of prompts
                                           detail of investigation indicated and              which may help to develop more
   I
•	 	nvolvement	of	the	patient	and/or	
                                                                                              detailed terms of reference. This is
   relatives;                              more immediate action required.
                                                                                              available in Appendix 1.
•	 Care	and	service	delivery	problems;
•	 Detection	of	incident;                  3.4 Background and context of the
                                                                                          Concise terms of reference should
•	 Contributory	factors;                   incident
                                                                                          be included in the report. If a long,
•	 Root	causes;                            This section should be used to set             detailed terms of reference document
•	 Lessons	learned;                        the scene.                                     exists this should be added to the
•	 Recommendations;                                                                       appendix.
   A
•	 	 rrangements	for	sharing	and	          A brief description should be given
   learning lessons.                       of the type of care and/or treatment
                                           being provided. Information on the
3c Main report                             size of the service, how long this type
                                           of service has been provided and the
3.2 Incident description and its           make up of the clinical team will help
consequences                               the reader understand the context of
                                           the incident.                                   T
                                                                                          		 aken	from	‘Three	levels	of	RCA	investigation’.	
                                                                                          4

Provide a clear, concise description                                                          Available at: www.npsa.nhs.uk/rca
of the incident and its effect on (or      3.5 Terms of reference
outcome for) the patient, the staff, the                                                  5
                                                                                              Memorandum of understanding: Investigating patient
service and any other stakeholders.        Terms of reference describe the                    safety incidents involving unexpected death or serious
                                           plan and latitude allowed to those                 untoward harm. www.dh.gov.uk/en/Consultations/
                                           conducting the investigation.                      Closedconsultations/DH_4090170



                                                                                                                                                   6
www.npsa.nhs.uk/nrls                                                                                      Root Cause Analysis Investigation Tools




It may be appropriate and possible to             The following four types of incidents,     3.6 The investigation team
set organisation-wide aims for patient            or	mid-investigation	findings,	should	
safety investigations, for inclusion in           be referred to alternative investigating   The level of investigation undertaken
the terms of reference section.                   bodies or processes for resolution,        will dictate the degree of leadership,
                                                  for example human resources,               overview and strategic review required.
Possible organisational aims,                     professional regulatory body, the
objectives and outcomes for the                                                              The table below shows the headings
                                                  police etc:                                you should use in this section, to list
investigation                                     1. Events thought to be the result         the core investigation team members
    •	 establish the facts:                           of a criminal act by care              and any chair, facilitators, service users,
         o    what happened (the chronology           providers/staff.                       experts, or other individuals that joined
              and effect)?                        2. Purposefully unsafe (malicious) acts    the extended team.
         o to whom?                                   by care providers intending to
         o when?                                      cause harm.                            Capturing the details of the
         o where?                                 3. Acts related to substance abuse by      investigation team
         o how? (what went wrong)
                                                      care providers/staff.
         o why? (contributory factors and                                                     Name and title                    MR C Jones
              root causes)                        4. Events involving suspected patient
    •	   e
         	 stablish	whether	failings	occurred	        abuse of any kind.                                                        Risk
                                                                                              Job title
         in care and/or treatment (care and                                                                                     Manager
                                                  As the Secretary of State and NHS
         service delivery problems);
    •	   	ook	for	learning	points	and	
         l                                        bodies have a duty to secure the            Qualifications                    EXAMPLE
         improvements rather than                 safety and well being of patients, the
         apportion	blame;                         investigation to determine the root         Background experience             EXAMPLE
    •	   	 stablish	how	recurrence	may	be	
         e                                        causes and learning points should
         effectively	reduced	or	eliminated;       still be progressed in parallel with        Investigation team role           EXAMPLE
    •	   	 ormulate	realistic	recommenda-
         f                                        other investigations, ensuring early
         tions which address root causes,         and robust solutions are put in place       Internal department
         and learning points to improve                                                       or reference to their
                                                  as necessary to reduce the likelihood       independence from the
                                                                                                                                EXAMPLE
         systems	and	services;                    of recurrence. A memorandum of
    •	   p
         	 resent	the	key	findings	in	a	report	                                               service
                                                  understanding (www.dh.gov.uk)
         as a record of the investigation
         process;                                 exists to assist organisations with the    3.7 The scope and level of the
    •	   	 rovide	a	consistent	means	of	
         p                                        planning and scoping of investigations     investigation
         sharing learning from the incident.      where the police or the HSE are
                                                  also involved.                             Explain how far back you decided to
It is important to protect the integrity                                                     go with the investigation and the level
of the RCA investigation process                  In the event of any referral to            of investigation conducted (see ‘Three
from situations where there is the                alternative bodies or processes arising    levels	of	RCA	investigation’),	and	justify	
probability of disciplinary action, or            from the patient safety investigation,     why. For independent investigations,
criminal charges.                                 it is open and transparent practice        the scope of the investigation could be
                                                  to make an anonymised reference            included under terms of reference.
Help box                                          to this in the lessons learned or
 The Incident Decision Tree is a key              recommendations section of                 Help box
 component of work to move away                   the report.
                                                                                              Sometimes the investigation team
 from asking ‘Who was to blame?’ to                                                           might find it needs to amend the
                                                  Root cause analysis techniques are
 asking ‘Why did the individual act in                                                        level of investigation once data
 this way?’ when things went wrong.               used by professionals conducting
                                                  other types of investigation, but it        gathering has commenced.
 The Incident Decision Tree has been              is important that all participants are
 created to help NHS managers and                                                            An explanation as to which relevant
                                                  aware of the clear distinction between
 senior clinicians decide whether                                                            services or other agencies have or have
                                                  the aims and boundaries of patient
 they need to suspend (exclude) staff                                                        not been included in the investigation
                                                  safety investigations, which are
 involved in a serious patient safety                                                        and why, should be provided.
                                                  solely	for	the	identification	and	
 incident and to identify appropriate
 management action. The aim is to                 reporting of learning points,              At the start of the investigation a
 promote fair and consistent staff                compared with disciplinary, regulatory     lessons learned log should be set up to
 treatment within and between                     or criminal processes.                     capture learning points (see 3.17).
 healthcare organisations.
                                                                                             Lessons learned may be described
 Learn more about the Incident                                                               as ‘key safety and practice issues
 Decision Tree at: www.npsa.nhs.uk/                                                          identified	which	did	not	materially	
 idt                                                                                         contribute	to	the	incident’.	They	may	
                                                                                             be related to the:
                                                                                             •	 incident	itself;
                                                                                             •	 investigation	process;
                                                                                                 i
                                                                                             •	 	mplementation	of	
                                                                                                 recommendations/action plans.



7
Root Cause Analysis Investigation Tools                                                                  www.npsa.nhs.uk/nrls




3.8 The investigation type, process        3.9 Involvement and support of the         3.10 Involvement and support for
and methods used                           patient, relatives or carers               staff involved in the incident
Describe the investigation type            In line with NPSA Safer practice           It is important to keep staff informed
undertaken (for example, single RCA,       notice 10: Being open when patients        at all times, and to provide advice,
RCA aggregation or multi-incident          are harmed7, the report should             support and opportunities for
investigation) and describe the process    demonstrate the extent to which those      involvement in the process.
and methods used.6                         affected have:
                                           •	 	 een	given	an	accurate,	open,	
                                                b                                     In the report, acknowledge help
Help box:                                       timely and clear explanation of       received from staff. Names of staff
                                                what has happened, regardless         should not feature in the RCA
  •	 Gather	information,	for	example:
     o interviews (for example                  of, but with sensitivity to, the      investigation other than in the
         cognitive	interviewing);               distressing	nature	of	the	incident;   archived master, and staff should be
  	 o	 brainstorming/writing;              •	 	 eceived	an	apology	in	the	form	
                                                r                                     advised that the report will be made
  	 o	 retrospective	clinical	records;          of a sincere expression of sorrow     anonymous before any circulation
  	 o	 multidisciplinary	team	reviews;          or regret for the harm that           or publication.
     o photographs, diagrams or                 has resulted from a patient
         drawings.                                                                    Outline any support given or offered
  •	 Map	the	incident,	for	example:
                                                safety	incident;                      to staff after the incident and during
  	 o	 narrative	chronology;                    b
                                           •	 	 een	informed	of	plans	regarding	      the investigation, such as counselling,
  	 o	 timeline/tabular	timeline;               what can be done medically to         support during interviews, or
  	 o	 time	person	grid;                        repair	or	redress	the	harm	done;      debriefing.	Refer	to	informal	support	
     o cause and effect chart.                  b
                                           •	 	 een	given	a	clear	statement	of	       from colleagues, as well as formal
  •	 	dentify	care	and	service	delivery	
     I                                          what is going to happen regarding     support, written materials or access to
     problems, for example:                     any investigation.                    support networks. Good practice in
     o multidisciplinary review
         meeting;                          The report should also explain to          this	regard	might	include	debriefing	
  	 o	 brainstorming/brainwriting;         what extent the patient, relatives         sessions. Consider support for all staff
  	 o	 nominal	group	technique;            and/or carers were involved in the         involved in the process including,
     o change analysis.                    investigation. This might include detail   for example, students, contractors
  •	 	 nalyse	problems	to	identify	
     A                                     on whether the patient or family were:     and investigators.
     contributory factors and root            a
                                           •	 	 sked how much involvement
     causes,
     for example:
                                              they want;
  	 o	 fishbone;                              i
                                           •	 	nterviewed	to	establish	the	
     o contributory factors                   questions they hope the
         classification/framework;            investigation will address and to
     o 5 whys.                                hear	their	recollection	of	events;
  •	 	 enerate	solutions	and	
     G                                        a
                                           •	 	 sked	how	they	would	like	their	
     recommendations, for example:            involvement and/or names referred
  	 o	 barrier	analysis;                      to	in	the	report;
  	 o	 risk	benefit	analysis.                 o
                                           •	 	 ffered	a	point	of	contact	
                                              (family liaison person) with regard
                                              to	the	investigation;
                                              g
                                           •	 	 iven	information	on	sources	of	
                                              independent	support/advocacy;
                                              i
                                           •	 	nformed	and	kept	up	to	date	with	
                                              the investigation process, including
                                              agreeing the frequency with which
                                              they	wanted	to	be	updated;
                                              a
                                           •	 	 dvised	that	the	report	and/or	
                                              findings	will	be	shared	with	them	
                                              as they wish, and that it will be
                                              written	in	plain	English;
                                              a
                                           •	 	 dvised	of	whom	they	can	contact	
                                              in the future (job title), should
                                              they want information on
                                              implementation of
                                              recommendations.

                                                                                      6
                                                                                          A guide to aggregated and multi-incident
                                                                                      investigations: www.npsa.nhs.uk/rca

                                                                                      7
                                                                                          Being open when patients are harmed
                                                                                      www.npsa.nhs.uk/nrls/alerts-and-directives/notices/
                                                                                      disclosure



                                                                                                                                            8
www.npsa.nhs.uk/nrls                                                                                      Root Cause Analysis Investigation Tools




3.11 Information and                            Help box                                        professional	assessment,	opinion);	
evidence gathered                                                                               d
                                                                                             •	 	 erivation	or	background	to	the	
                                                 Formal signed witness statements
                                                 would not normally form part of a Root         data (corroborated or in line with
The report (or appendix) should                                                                 best practice guidance).
                                                 Cause Analysis investigation report
contain a summary list of all evidence           produced for learning purposes. Staff
gathered from people, documentation,                                                         This full working document should be
                                                 may wish to write factual reflective
plant/equipment and site visits.                 notes, but if these are shared with         saved within the master investigation
                                                 the organisation, they can become           file	and	any	final,	full	mapped	
    Evidence can include the following items:    discoverable.                               chronology of events should be
    o	   interview	notes;                                                                    included as an appendix to the report.
                                                 Formal, signed witness statements
    o    letters:                                are more relevant and appropriate to        3.13 Detection of incident
    o	   e-mails;                                disciplinary or criminal investigations
    o	   equipment;                              (see ‘Investigative interview guidance’:    It is useful to identify at what stage
    o	   equipment	fault	reports;                www.npsa.nhs.uk/rca).                       in	the	patient’s	treatment	the	error	
    o	   	iterature	review	findings,	such	as	
         l                                                                                   was detected. This gives important
         National	Service	Frameworks;            Witnesses should be made aware that
                                                                                             information on how far the problem
                                                 documents referred to in any interview
    o    NICE and/or other good practice         or multidisciplinary review meeting         progressed	without	identification,	
         guidance;                               may be disclosed in future (this may        indicating how effective existing
    o	   national	alerts;                        include reflective practice documents,      controls/barriers were. It may also
    o	   legislation	policies;                   personal and professional diaries, etc).    add insight into where best to invest
    o	   procedures;                                                                         effort and resources to generate the
    o	   site	plans;                            Whilst staff directly involved in an         most effective solutions. Examples
    o	   photos;                                investigation should have the chance         may include:
    o	   training	records;	                     to correct factual inaccuracies or                a
                                                                                             •	 	 t	risk assessment of new or
    o    maintenance records etc, both in       comment on recommendations before                 changed	service;
         place and in use at the time of        a	report	is	finalised,	it	should	be	clear	        a
                                                                                             •	 	 t	pre-treatment	patient	
         the incident.                          they do not have a right of veto.                 assessment;
    o    contextual data such as local or       Rather than risk situations where the             e
                                                                                             •	 	 rror	recognition	pre-care/
         national audits may also               lead investigator/ investigating team             treatment;
         be valuable.                           might	be	pressured	or	influenced,	this	           e
                                                                                             •	 	 rror	recognition	post-care/
                                                discussion may require support by an              treatment;
An outline summary or list of this              impartial third party.                            b
                                                                                             •	 	 y	machine/system/environment	
information	review	will	suffice,	rather	                                                          change/alarm;
than including copies in the report             3.12 Chronology of events leading            •	 by	a	count/audit/query/review;
(copies of the salient documents                up to the incident                           •	 by	change	in	patient’s	condition.
belong	in	the	investigation	file).	List	the	
version and date, as well as the actual         The report should include a summary          3.14 Notable practice within
document title.                                 of the key points of the mapped              the case
                                                chronology, so that the reader can gain
Copies of key documents can                     a clear understanding of the events          It is important to record, with
be included in the appendix as                  leading up to the incident. This is          appropriate sensitivity, points in the
appropriate	and	useful.	Confidential	           ideally presented in visual format, for      incident or patient journey where
or highly detailed documents should             example a summary timeline, ‘tabular         care and/or practice had an important
be retained as part of the master               timeline’,	or	as	part	of	a	‘cause	and	       positive impact and may provide
investigation	file	only.	                       effect	chart’	(see	‘RCA	toolkit’	at	         valuable learning opportunities.
                                                www.npsa.nhs.uk/rcatoolkit)
To	enable	investigation	report	findings	                                                     Use this section to comment on the
to be shared for learning purposes,             The chronology or tabular timeline           co-operation and openness of staff
investigators should ensure that                included in the report should be             during the investigation.
consent to access/utilise and publish           derived	and	summarised	from	the	final	
information from patient records has            document, rather than including the
been obtained.                                  entire or working document which,
                                                during analysis, will have more detailed
All evidence should be included                 notes identifying gaps in information
whether it supports or contradicts your         and distinguishing between different
conclusions. Record an audit trail of key       types of evidence. For example:
decisions made, and provide reasons for             t
                                                •	 	 he	source	of	the	information	
discounting any facts which contradict              (first-hand,	based	on	memory,	
your	conclusion.	If	there	is	a	conflict	of	         contemporaneous);
facts, explain why one version is more              a
                                                •	 	 ttribution	or	acknowledgement	
credible than the other.                            (who said what or provided the
                                                    information);	
                                                    f
                                                •	 	 oundation	or	basis	of	the	
                                                    information (fact, evidence,


9
Root Cause Analysis Investigation Tools                                                                           www.npsa.nhs.uk/nrls




3.15 Care and service delivery                           3.17 Root causes                            3.18 Lessons learned
problems                                                                                             (see also 3.7)
                                                         This section of the report should
The report should detail how care                        demonstrate a direct link between           There may be occasions when nothing
delivery problems (CDPs) and service                     cause and effect.                           could have prevented the incident and
delivery problems (SDPs) were found                                                                  no	root	cause(s)	are	identified.
                                                         These descriptions of root causes
and which RCA analytical tools were
                                                         (conclusions) should:                       There are always lessons to learn
used to identify them.
                                                         •	 be	numbered;                             and key safer practice issues may be
CDPs and SDPs are points in the                             b
                                                         •	 	 e	clearly	linked	by	analysis	to	the	   identified	which	did	not	materially	
timeline at which:                                          evidence	found;                          contribute to the incident.
•	 	 omething	happened	that	
   s                                                        a
                                                         •	 	 void	blame	and	not	include	
   shouldn’t	have	happened;	                                inflammatory	statements	or	              Lessons learned from the incident and
OR                                                          negative descriptors (e.g. poor/         the	investigation	should	be	identified,	
•	 	 omething	that	should have
   s                                                        careless/inadequate/reckless).           numbered and addressed by the
   happened	didn’t.                                                                                  recommendations, alongside any
                                                                                                     root causes.
The CDPs and SDPs should then be
prioritised for analysis.                                                                            3.19 Recommendations

Help box:                                                                                            Recommendations and solutions
                                                                                                     should be designed to address
 Issue                                    Wrong                       Right                          the root causes (conclusions). For
                                                                                                     shorter, less complex investigations,
 Wording of CDPs and SDPs                 ‘communication	failure’     ‘paramedic did not inform      recommendations and solutions
 needs	to	be	specific                                                 A&E that patient was           may be developed at the same time.
                                                                      confused’                      For more detailed investigations,
 CDPs and SDPs must                       ‘not enough training on     ‘staff members rarely          recommendations may inform action
 describe what happened,                  hand	hygiene’               washed	their	hands’            planning and solutions development
 not why it happened                                                                                 carried out at a later date by a different
                                                                                                     or reconstituted team.
                                                                                                     Designing recommendations and
3.16 Contributory factors                                In essence, the report                      solutions to address the root causes
The	contributory	factors	identified	                     should show a clear thread                   Recommendations should:
for each prioritised care and service                                                                    b
                                                                                                      •	 	 e	clearly	linked	to	identified	root	
                                                         connecting:                                     cause(s) or key learning point(s)
delivery problem should be listed.
                                                                                                         (to address the problems rather
This analysis should highlight                           1. the root cause(s) (in                        than	the	symptoms);
contributory factors taken from                                                                          a
                                                                                                      •	 	 ddress	all	of	the	root	causes	and	
                                                             organisational processes);                  key	learning	points;
the NPSA contributory factors
classification/framework	(see	                           2. how these directly                           b
                                                                                                      •	 	 e	designed	to	significantly	reduce	
                                                                                                         the likelihood of recurrence and/or
www.msnpsa.nhs.uk/rcatoolkit/                                resulted in the specific                    severity	of	outcome;
resources/word_docs/Guidance/                                                                            b
                                                                                                      •	 	 e	clear	and	concise	and	kept	to	a	
Guidance_Factors_Framework_                                  care and service delivery                   minimum	wherever	possible;
Checklist.doc) and should                                    problems;                                   b
                                                                                                      •	 	 e	Specific,	Measurable,	Achiev-
not include negative descriptors, e.g.                                                                   able, Realistic and Timed (SMART)
poor/inadequate/ careless/complacent.                    3. how these led to the                         so that changes and improvements

Clearly explain how these were
                                                             documented actual                           can	be	evaluated;
                                                                                                      •	 be	prioritised	wherever	possible;
identified	and	which	tools	were	used	                        or potential effect on                   •	 be	categorised	as:
to carry out the analysis.                                   the patient.                                o those specific to the area
                                                                                                              where	the	incident	happened;
Although	fishbone	diagrams	are	                                                                          o those that are common only to
often used to identify contributory                                                                           the	organisation	involved;
factors, you may choose to analyse and                                                                   o those that are universal to
display these in a full report using a                                                                        all and, as such, have national
‘contributory	factors	grid’	(available	at	                                                                    significance.
www.npsa.nhs.uk/rca).                                                                                 Recommendations might also
                                                                                                      include:
                                                                                                          p
                                                                                                      •	 	 rovision	of	ongoing	support	of	
                                                                                                          patients and staff affected by the
                                                                                                          incident.




                                                                                                                                                  10
www.npsa.nhs.uk/nrls                                                                                           Root Cause Analysis Investigation Tools




Clearly explain in the investigation               It is important to detail the                  The process of sharing investigation
report how recommendations were                    mechanisms by which lessons have               findings	and	reports	with	the	patient	
developed, which tools were used                   been or will be:                               and/or family should comply with
(for example, barrier analysis to assess           1.	 learned/implemented;                       ‘Being	Open’	principles	(www.npsa.
effectiveness of controls in place), and           2. shared/disseminated.                        nhs.uk/nrls/alerts-and-directives/
who was invited to help (for example,                                                             notices/disclosure/)
system designers, or those involved in             Lessons learned are most effectively
the incident).                                     captured by completing a lessons               Unless	there	are	specific	exceptions,	
                                                   learned log during the investigation           the patient/family of a patient have
What to consider when developing                   and action plan implementation                 a right to the full investigation
recommendations                                    stages.	Key	findings	can	then	be	              report under the requirements of
                                                   shared locally, with Patient Safety            the Data Protection Act. (available at
 o  understand that retraining is not
    always	the	right	solution;                     Action Teams and/or the NPSA as                www.ico.gov.uk).
 o intelligent use of checklists, policies         part of future investigation reporting
    and	protocols;                                 processes.                                     Organisations need to support
 o minimal dependency on short-term                                                               investigators to ensure this does not
    memory	and	attention	span;                     3.21 Distribution list                         inhibit them from identifying areas
 o	 	 implification	of	tasks	and	
    s                                              A discussion should have taken place           of concern. It is important to discuss
    processes;
                                                   with the commissioners of the                  preferences with patients or relatives,
 o standardisation of tasks and                                                                   however, as many may prefer to
    processes;                                     investigation to agree who the
                                                   audience	of	the	final	report	is.	That	list	    receive a shorter executive summary or
 o avoidance of fatigue (review of                                                                just a copy of the recommendations
    working	hours/patterns);                       should be included in this section.
                                                                                                  from the report.
 o alignment with evidence-based
    practice;                                      Help box
                                                                                                  Staff should not disclose in the report
 o alignment with organisational                    North East Yorkshire and North                any health or personal issues of a
    priorities and risk registers.                  Lincolnshire Trusts have identified           patient that the patient may have
                                                    a useful list of other potential              previously chosen not to disclose to
3.20 Arrangements for shared                        stakeholders (Appendix 2) who may             their family or others.
learning                                            routinely or exceptionally request or
                                                    require sight of investigation reports.       If an organisation takes varying
Record in this section the degree                                                                 approaches on how much of the
to which sharing is required                                                                      investigation report they share, the
(see guide below).                                                                                justification	for	this	needs	to	be	clear,	
Guide to sharing learning                                                                         explicit	and	in	line	with	the	patient’s	
                                                                                                  wishes.
 Learning potential     Significance         Sharing
                                                                                                  3.22 Investigation report
 Specific               Local                Shared within the area where the incident            appendices
                                             happened.
                                                                                                  The appendices should include key
 Common                 Organisational       Shared across the organisation involved.             explanatory documents including:
 Broad / universal      National             Shared across organisation involved and with            f
                                                                                                  •	 	 ull	terms	of	reference	(where	
                                             other	organisations/specific	services/specialties/      applicable);
                                             directorates – via patient safety networks,          •	 list	of	literature	reviewed;
                                             Patient Safety Action Teams, NPSA etc.                  s
                                                                                                  •	 	 ummary	list	of	evidence	gathered	
                                                                                                     (if this is too lengthy to be included
                                                                                                     in	the	report);
Note that it is common to assume                                                                     c
                                                                                                  •	 	 opies	of	key	documents,	site	
incidents are unique or only relevant                                                                plans, photographs etc (all others
to your own organisation. In reality,                                                                in	archived	master);
patterns recur. If in doubt, it is always                                                         •	 final	chronology	or	timeline;
wise to share.                                                                                       t
                                                                                                  •	 	 emplates	used	for	analysis,	for	
                                                                                                     example	fishbones,	run	charts,	
                                                                                                     change/	barrier	analyses;
                                                                                                  •	 lessons	learned	log;
                                                                                                     a
                                                                                                  •	 	 cknowledgements	(if	part	of	NHS	
                                                                                                     organisation style and format).




11
Root Cause Analysis Investigation Tools                                                                        www.npsa.nhs.uk/nrls




Section 4: Next steps
The following actions are not normally included as part of the RCA investigation report itself, but must be conducted as
essential next steps once the report and its recommendations have been approved by the overseeing committee.

4.1 Action planning and                              a
                                                  •	 	 voidance	of	fatigue	(review	of	         the potential to produce adverse effects
                                                     working	hours);                           elsewhere in a system as complex as
solutions development                                c
                                                  •	 	 ost	benefit	analysis	and	risk	          healthcare. The risk assessment should:
Action plans should set out how each                 assessment;                                   i
                                                                                               •	 	dentify	and	address	any	material	
recommendation will be implemented,                  a
                                                  •	 	 lignment	with	evidence-based	               downsides to recommendations
with named leads responsible for each                practice;                                     or	solutions;
action point or solution. To ensure                  a
                                                  •	 	 lignment	with	organisational	               d
                                                                                               •	 	 emonstrate	the	expected	impact	
solutions are realistic, accepted, and               priorities and risk registers.                of	solutions,	or	decisions	not	to	act;8
owned by the service, it is essential                                                              i
                                                                                               •	 	dentify	any	priority	in	terms	of	
                                                  As with recommendations, action points           expected effectiveness and ease
that frontline staff and those with
                                                  and/or solutions should:                         of implementation.
appropriate local knowledge are
                                                     b
                                                  •	 	 e	clearly	linked	to	identified	root	
heavily involved in, or consulted on
                                                     cause(s)	or	key	learning	point(s);	
this process.
                                                     (addressing the problems rather than
                                                                                               4.3 Implementation,
Actions taken following a patient                    the	symptoms);                            monitoring and evaluation
safety incident                                      a
                                                  •	 	 ddress	all	of	the	root	causes	and	      arrangements
                                                     key	learning	points;
 Action               Explanation                 •	 	 e	designed	to	significantly	reduce	
                                                     b                                         This section should demonstrate clearly
                                                     the likelihood of recurrence and/or       the arrangements in place to successfully
 Immediate            Taken to prevent               severity	of	outcome;                      deliver the action plan.
 response             or moderate the
 and                  progression (severity or
                                                     b
                                                  •	 	 e	clear	and	concise	and	kept	to	a	
                                                                                               Ideally, overseeing committees should
 recovery             likelihood of impact) of       minimum	wherever	possible;
                                                                                               plan	and	request	final	review	or	risk	
 actions              an	incident;	or	to	treat/      b
                                                  •	 	 e	Specific,	Measurable,	Achievable,	
                                                                                               assessment to be conducted at around
                      compensate for harm            Realistic and Timed (SMART) so that
                                                                                               one year post-implementation, to ensure
                      after an incident. These       changes and improvements can
                                                                                               recommendations and solutions have
                      are often recorded as          be	evaluated;
                      part of the incident                                                     been adopted and that changes designed
                                                     b
                                                  •	 	 e	assessed	for	resource	needs,	risks	
                      report, but may also be                                                  to reduce risk have been successful.
                                                     and	impact;8
                      included in an action          b
                                                  •	 	 e	prioritised	wherever	possible	(for	   Activities for the action plan
                      plan.                          example	following	risk/cost	benefit	
 Preventative         Taken to address the           analysis);                                 Activity           Associated actions
 or risk-             cause(s) of the incident    •	 be	categorised	as:
 reducing             and robustly reduce,                                                      Implement          For example, piloted,
                                                     o those specific to the area where
 actions or           manage or control future                                                                     roll-out, phased,
                                                        the	incident	happened;                                     championed).10
 solutions            risk of harm. These
                                                     o those that are common only to
                      should be logged in the                                                   Monitor            For example, monthly
                      action plan.                      the	organisation	involved;
                                                     o those that are universal to all                             monitoring by the
                                                                                                                   organisation governance
Clearly explain in the investigation                    and, as such, have national
                                                                                                                   committee or progress
report how action plans and solutions                   significance.                                              report complied by risk
were developed, which tools were used,                                                                             manager.
if any, (for example, barrier analysis to         Solutions might also include:
assess effectiveness of controls in place            p
                                                  •	 	 rovision	of	ongoing	support	of	          Evaluate           For example, assessing
                                                     patients and staff effected by                                the impact of changes/
and to design new or more robust                                                                                   solutions introduced (this
controls/solutions), and who was invited             the incident.
                                                                                                                   could include conducting
to help (for example, system designers,           The action plan should be in a format                            an impact analysis,8
those involved in the incident).                  that can be presented to the Board and                           reviewing incidence/
                                                  attached to an executive summary for                             severity of recurrence).
Development of solutions might usefully
include consideration of the following:           internal circulation following approval of
•	 	 nderstanding	that	retraining	is	not	
    u                                             recommendations.9
    always	the	right	solution;
•	 	 ise	use	of	checklists,	policies	and	
    w                                             4.2 Action plan - risk/impact                8 See 4.2 and the ‘Solutions option appraisal and
    protocols;                                    assessment                                    impact	analysis	template’	at	www.npsa.nhs.uk/rca
    m
•	 	 inimised	dependency	on	short	
    term	memory	and	attention	span;               Risk assessments (using your organisation    9	See	‘Action	plan	template’	at	www.npsa.nhs.uk/rca

•	 simplification	of	tasks	and	processes;         risk assessment processes) conducted
                                                                                                  N
                                                                                               10		 HS	Sustainability	Model	and	Guide	2002’	at:
•	 	 tandardisation	of	tasks	and	
    s                                             during the recommendation or action
                                                                                                  www.institute.nhs.uk
    processes;                                    planning process should be included in
                                                  the report. Even positive changes have

                                                                                                                                                   12
www.npsa.nhs.uk/nrls                                                                           Root Cause Analysis Investigation Tools




Section 5: Appendices
Appendix 1: Prompts for terms of reference
The following are thoughts which may prompt or inspire development of standard or individual terms of reference for
investigations. They have been provided to allow you to make notes of your own thoughts next to them.

 Prompt                                                         Notes

 Accountability

 Agreeing recommendations and action plan

 Assurance frameworks

 Awareness

 Building	confidence	in	organisation

 Closure

 Communication

 Encouraging reporting

 Explanation

 Identifying system failures

 Identifying trends

 Improving practice

 Informing patients/carers/families

 Issuing an apology

 Learning and sharing

 Litigation

 Ongoing review – audit/monitoring

 Ownership

 Prevention

 Promoting an open and fair culture

 Quality assurance
 Referral on, if potential disciplinary or performance issues
 are	flagged
 Remedial action

 Resolving complaints

 Support staff

 Transparency

 Valuing staff



13
Root Cause Analysis Investigation Tools                                                                  www.npsa.nhs.uk/nrls




Appendix 2: Prompts for investigation report distribution list
This list will help with ideas when developing an investigation report distribution list.


 Action plan implementers                                            Monitor

 Care Standards Improvement Partnership                              National Clinical Assessment Service

 Clinical Governance and Complaints Committees                       National	Confidential	Enquiry	into	Patient	Outcome	and	
                                                                     National	Confidential	Inquiry	into	Suicide	and	Homicide	by	
 Clinical Risk Group                                                 People with Mental Illness
 Clinical team members involved                                      National Institute for Health and Clinical Excellence

 Commissioners                                                       National Institute for Mental Health Executive

 Confidential	Enquiry	into	Maternal	and	Child	Health                 NHS Litigation Authority (CNST & RNST)
                                                                     NPSA & National Reporting and Learning
 Commissioning Primary Care Organisation                             System
 Coroner                                                             Overview and Scrutiny Committees

 Counter Fraud and Security Management Service                       PALS and PPI Forums

 Department of Health Investigations Unit                            Patient Experience Committee

 Environmental Health agencies                                       Patient/carers/family

 Executive Directors                                                 Police

 General Medical Council                                             Prison Service

 Legal services – claims                                             Protection of Vulnerable Adults

 Nursing Midwifery Council                                           Regional	Office	(Wales)
 Health & Safety Executive Reporting of Injuries, Diseases &         Risk Management Team
 Dangerous Occurrences Regulation
 Health Protection Agency                                            Royal Colleges and professional bodies

 Health Service Ombudsman                                            Service Directors/Managers

 Healthcare Commission                                               Social Services (Child Protection/Mental Health)

 Commissioning Primary Care Organisation                             Staff newsletter

 Local Authority                                                     Staff

 Local	Supervisory	Authority	(Midwifery	Officer)                     Strategic Health Authority/Patient Safety Action Team

 Media                                                               	Trade	Unions	representing	staff	(NB:	issues	of	confidentiality)

 Medical Director/accountable Director for Risk                      Trust/Local Health Board

 Medicines and Healthcare products Regulatory Agency                 Mental Health Act Commission




Good practice principles from North and East Yorkshire and Northern Lincolnshire Trusts (NEYNL)




                                                                                                                                    14
www.npsa.nhs.uk/nrls                                                                              Root Cause Analysis Investigation Tools




Section 6: Bibliography
National Patient Safety Agency. Being open when patients are harmed: www.npsa.nhs.uk/patientsafety/alerts-and-
directives/notices/disclosure/
Canadian root cause analysis framework – A tool for identifying and addressing the root causes of critical incidents in
healthcare. Canadian Patient Safety Institute, Institute For Safe Medication Practices Canada, and Saskatchewan Health.
ISBN 0-9739270-1-1
CNST standards and assessment: www.nhsla.com/RiskManagement/CnstStandards/
Data Protection Act: www.ico.gov.uk/what_we_cover/data_protection.aspx
Department of Health. Independent investigation of adverse events in mental health services. June 2005
www.dh.gov.uk/en/Publicationsandstatistics/Lettersandcirculars/Dearcolleagueletters/DH_076535
Good practice principles from North and East Yorkshire and Northern Lincolnshire Trusts (NEYNL)
HSG (94)27: Guidance on the discharge of mentally disordered people and their continuing care in the community and
Investigation of adverse events in mental health services (June 05 - amends paras 33-36 of circular) www.dh.gov.uk
Incident Decision Tree www.npsa.nhs.uk/idt
Independent investigations of serious patient safety incidents in mental health services: Good practice guide. www.npsa.
nhs.uk/nrls/alerts-and-directives/directives-guidance/mental-health/
International	Classification	of	Patient	Safety	Events	(ICPSE)	-	World	Health	Organization	www.who.int
Learning through action to reduce infection. www.npsa.nhs.uk/nrls/improvingpatientsafety/humanfactors/infection-control
Memorandum of understanding: Investigating patient safety incidents involving unexpected death or serious untoward
harm. www.dh.gov.uk/en/Consultations/Closedconsultations/DH_409017
Modernisation	Agency’s	NHS	Sustainability	model	and	guide	2002	NHS	Institute	for	Innovation	and	Improvement	
www.institute.nhs.uk
National Institute for Health and Clinical Excellence. How to change practice: understand, identify and overcome barriers to
change. Available at: www.nice.org.uk/usingguidance/implementationtools/howtoguide/barrierstochange.jsp
RCA Toolkit. www.npsa.nhs.uk/rcatoolkit/
Risk assessment made easy. www.npsa.nhs.uk/nrls/improvingpatientsafety/patient-safety-tools-and-guidance/risk-
assessment-guides/
Risk matrix for risk managers www.npsa.nhs.uk/nrls/improvingpatientsafety/patient-safety-tools-and-guidance/risk-
assessment-guides/risk-matrix-for-risk-managers/
Safer care www.institute.nhs.uk/safer_care
National Patient Safety Agency. Seven steps to patient safety. www.npsa.nhs.uk/sevensteps/
The Joint Commission Sentinel events policy www.jointcommission.org




15
Root Cause Analysis Investigation Tools                                                              www.npsa.nhs.uk/nrls




Section 7: Acknowledgements
This report was compiled with valuable information, help and advice provided by:
•	          Donna	Forsyth,	Head	of	Reporting	and	Local	Improvement,	National	Patient	Safety	Agency
•	          Alison	Prizeman,	Patient	Safety	Consultant,	NHS	South	East	Coast
•	          Bridget	James,	Patient	Safety	Manager,	NHS	South	West
•	          John	Morrison,	Patient	Safety	Manager,	NHS	East	of	England
•	          North	and	East	Yorkshire	and	Northern	Lincolnshire	Trusts	(NEYNL)
•	          Dr	Christine	Jorm,	Senior	Medical	Advisor,	Australian	Commission	on	Safety	and	Quality	in	Health	Care
•	          	 ryce	Cassin,	Manager	Clinical	Safety	and	Quality	Projects,	Australian	Commission	on	Safety	and	Quality	in	
            B
            Health Care
•	          Colin	Phillips,	Head	of	Investigations,	Department	of	Health
•	          Peter	Walsh,	Chief	Executive,	Action	against	Medical	Accidents	(AvMA)
•	          Malcolm	Alexander,	National	Association	of	LINks	Members
•	          Katherine	Murphy,	Director	of	Communications,	The	Patients	Association
•	          Sue	Line,	National	Association	for	Patient	Participation
•	          Brian	Toft,	Professor	of	Patient	Safety,	Applied	Research	Centre,	Coventry	University	
•	          Chris	Brougham,	Training	Consultant,	Verita




                                                                                                                            16

				
DOCUMENT INFO