Select Committee on Health Written Evidence

Memorandum by the National Institute for Health and Clinical Excellence (PS 57)



  1.1  Patient safety is important to NICE and is an integral part of the guidance issued to the NHS. In addition, at the request of the Department of Health, NICE has issued guidance on the reduction of risk of transmission of Creutzfeldt-Jakob disease (CJD) via interventional procedures. Working with the NPSA, again at the request of the Department of Health, NICE has issued guidance on strategies to prevent ventilator-acquired pneumonia and guidance on medicines reconciliation on admission. This Memorandum provides insight, gained from NICE's experience, into the Health Select Committee's questions as to what the NHS should do next regarding patient safety. A number of suggestions for future development are made.


  2.1  NICE is responsible for providing national guidance on the promotion of good health and the prevention and treatment of ill health, in three areas:

    —  Health technologies—guidance on the use of new and existing medicines, treatments and procedures, including interventional procedures used in the NHS.

    —  Clinical practice—guidance on the appropriate treatment and care of people with specific diseases and conditions within the NHS.

    —  Public health—guidance on the promotion of good health and the prevention of ill health for those working in the NHS, local authorities and the wider public and voluntary sector.

  2.2  The Institute and its remit have grown rapidly since its establishment in 1999 and it is now the primary source of clinical standards, based on clinical and cost effectiveness, in England, Wales and Northern Ireland. The applicability of NICE guidance in the UK, as a whole, is summarised in Table 1.

Table 1: Applicability of NICE guidance in the UK

Country Technology appraisals Clinical guidelines Interventional procedures Public health guidance
England Yes Yes Yes Yes
Wales Yes Yes Yes No
Scotland Yes[347] No Yes No
N Ireland Yes[348] YesYes No

  2.3  Established in April 1999 to set clinical standards as part of a comprehensive quality framework for the NHS, NICE's role has since been extended. The public health white paper Choosing Health, published in November 2004, confirmed the Institute's new role in providing the NHS and the wider community with guidance on effective public health practice. NICE merged with the Health Development Agency in April 2005 and by the end of 2006 systems to deliver public health interventions and programme guidance were fully established and beginning to provide guidance in a wide range of areas, including physical activity, smoking cessation, sexually transmitted infection and drug misuse.

  2.4  Patient safety is important to NICE and is an integral part of the guidance that it issues. In November 2006, NICE issued its first patient safety guidance on "patient safety and reduction of risk of transmission of Creutzfeldt-Jakob disease (CJD) via interventional procedures".[349] This guidance was developed following a specific request from the Chief Medical Officers of England, Scotland, Wales and Northern Ireland.

  2.5  Safety First, a review of the organisational arrangements for patient safety in the NHS, made a number of recommendations about the future of patient safety, including the establishment of a Patient Safety Forum, which includes NICE in its membership[350]. In February 2007, the Institute co-signed the Patient Safety Charter alongside the National Patient Safety Agency (NPSA), the Healthcare Commission and other national bodies to emphasise organisational commitment to improving patient safety.

  2.6  One of the key recommendations of Safety First was that "a pilot should be established to examine the option of the National Institute for Health and Clinical Excellence (NICE) developing technical patient safety solutions". As a result, NICE and the NPSA were asked to develop and evaluate a pilot project for the production of a technical solution to address specific patient safety issues, which would form guidance for the NHS. They were also asked to take advantage of international experience, such as that of the WHO World Alliance for Patient Safety, so that duplication of effort would be minimised.

  2.7  The pilot began in April 2007.[351] Two topics were chosen, evaluating a total of seven potential patient safety solutions (PSS). NICE and the NPSA issued joint safety guidance to the NHS on three potential solutions for errors in medicines reconciliation on admission of adults to hospital in December 2007; and on four potential solutions for the prevention of ventilator-associated pneumonia in adults in August 2008.The evaluation was completed in February 2008.[352]

  2.8  NICE's Citizens' Council, a representative group of 30 members of the general public, considered patient safety issues in June 2007. Its findings are presented in Section 5 and the full report can be found in on our website.[353]


  3.1  For the purposes of the pilot study, a technical PSS was defined as "a cost-effective intervention to prevent or mitigate patient harm stemming from processes of healthcare based on the best available evidence". The terminology is derived from the WHO's definition of a safety solution as "any system design or intervention that has demonstrated the ability to prevent or mitigate patient harm stemming from the processes of healthcare".

  3.2  Patient safety solutions can vary greatly in their complexity, function and application and can assist with patient monitoring, treatment management and service delivery. However, safety solutions may introduce new risks and increase existing risks as well as eliminating and reducing other risks.

  3.3  The WHO Collaborating Centre for Patient Safety Solutions states that the basic purpose of the solutions is to guide the re-design of care processes to prevent inevitable human errors from actually reaching patients[354]. An individual solution will present the problem, the strength of evidence supporting the solution, potential barriers to adoption, risks of unintended consequences created by the solution, patient and family roles in the solution, and references and other resources[355]. Solutions need to be practical, deliverable, and minimise risk without themselves introducing risk, and they need to take into account the mobility of the workforce.

  3.4  The pilot's methodology and process were designed to meet the Institute's core principles of guidance development: a comprehensive evidence base; independent advisory committees; clinical and patient expert input; transparent process and decision making; genuine consultation; effective dissemination and implementation; and regular review.

  3.5  For each of the pilot topics, the Institute applied its methods of assessing clinical and cost-effectiveness within a project-specific frame work.[356] The methodology chosen was intended to complement the existing Root Cause Analysis and Risk Assessment methodology used by the NPSA. Elements of existing processes were used, with shortened intervals to ensure that the guidance could be produced within a six months timeframe.


4.1  A flexible system is required

  4.1.1  Patient safety comprises a range of errors and system failures associated with the delivery of patient care. This can include situations such as mistakes and delays in diagnosis, medication and treatment errors, problems with equipment, infections acquired in hospitals and accidents such as slips and falls. Each event is unique although there may be similarities in the patterns and sources of risk. As a consequence, there are a range of potential PSS types and complexities.

  4.1.2  We believe that mechanisms for evaluating potential PSS need to be adaptable enough to examine each different type using the most appropriate method and process. In some circumstances, for example the prevention of ventilator-associated pneumonia, a guideline approach might be most appropriate. This would allow a large number of PSS to be evaluated simultaneously and permit consensus-based methodology to supplement health technology assessment (HTA) methodology to enable recommendations about entire pathways of care, even in the absence of robust evidence. In other circumstances, a focused intervention-based approach would be more appropriate, for example directly comparing kinetic beds with manual rotation of the patient in order to prevent ventilator-associated pneumonia.

4.2  Determination of best practice

  4.2.1  The NHS is potentially a rich source of information on measures to improve patient safety. However, many locally developed measures may never be published and the literature is therefore unlikely to reflect current NHS practice. We suggest that for each patient safety issue a survey of current practice could identify potential solutions and inform the development of new solutions.

  4.2.2  In addition to potential PSS identified through incident reporting, there are a number of PSS that have already been developed, including those launched by the WHO patient safety solution initiative. The possibility of adapting existing PSS has been raised, and we suggest that an approach similar to ADAPTE,[357] which has been developed for guidelines, could be developed. Further methodological exploration is required.

  4.3  Evidence base for measures to improve patient safety

  4.3.1  A key finding from NICE's pilot was that the published evidence base for most of the seven PSS examined was extremely poor. The paucity of the evidence base is related to both the quality and quantity of the research that had been undertaken. Much of the research that had been undertaken used non-randomised controlled trial (RCT) methodology that is inherently at greater risk of bias and requires appropriate critical appraisal and interpretation. It is likely that much research undertaken by the NHS is never published. Further development of the infrastructure is required to raise awareness about methodology and also encourage research to be undertaken and disseminated.

  4.3.2  It was also extremely difficult to identify the relevant literature. It is unclear whether patient safety research is adequately represented in the medical literature databases that are commonly searched. For example Medline and Embase. In addition, it is unclear to what extent existing MESH[358] headings adequately catalogue patient safety issues. If the studies are not catalogued properly then they cannot be identified.

  4.3.3  The grey literature (conference proceedings, research reports and theses) could be another potentially rich source of evidence. However there has been no research to investigate the extent to which patient safety issues are addressed in the grey literature. The validity of such grey literature will however need to be rigorously evaluated because it has not been through the publication process, which affords a degree of quality assurance.

  4.3.4  We therefore recommend further exploration of methods to overcome the inadequacies in the evidence-base summarised in this section and including, for example, formal consensus methodology (see 4.1.2).

4.4  Involving patients and the public

  4.4.1  One of NICE's core principles of guidance development is that a genuine consultation is undertaken. In order to ensure this a common process has been developed that identifies the relevant stakeholders. The stakeholders are then asked for their views on the question being asked, the evaluation exercise being undertaken, and the proposed guidance.

  4.4.2  NICE's experience was that it was difficult to engage with stakeholders during the pilot projects. The relevant stakeholder group differs slightly from those that are used to working with NICE and there was therefore a lack of awareness about the Institute and its guidance.

  4.4.3  In particular it is crucial that patients and groups representing patients are fully engaged, as they hold vital perspectives on the patient safety issues and can therefore have a key role in being part of the solution. NICE routinely consults with a wide range of patient experts, primarily through patient-support groups and umbrella organisations that are condition-specific. We suggest, however, that a different approach is required when identifying individuals who have had first-hand experience of the patient safety issue.

4.5  What should be measured and assessed?—the role of clinical and cost-effectiveness

  4.5.1  The existing Technology Appraisals-based methodology that was used in the pilot is well suited to assessment of the clinical and cost-effectiveness of existing PSS that are underpinned by an evidence base. The methodology is however not suitable for the development of new solutions to address a patient safety issue.

  4.5.2  For each patient safety issue there are likely to be a number of potential solutions. Careful consideration needs to be given to whether alternative PSS are directly compared or considered in isolation. In addition to the issues with the evidence-base already identified, it is unlikely that direct comparisons of PSS will have been undertaken.

  4.5.3  As highlighted in the pilot the cost-effectiveness of PSS can be evaluated. However, in many circumstances there will be a lack of appropriate evidence, which is likely to increase the uncertainty surrounding the results. For PSS that require service reconfiguration the evaluations are likely to be complex. It is unlikely that stakeholders will develop alternative economic models for evaluation due to lack of ownership of many PSS. In addition, formal modelling may not be needed in some circumstances.

  4.5.4  Current cost effectiveness evaluation methodology could be used to explore the cost effectiveness of PSS in general, but such evaluations are likely to be complex. Furthermore there is likely to be considerable uncertainty surrounding the results due to the lack of appropriate evidence. This is particularly the case for PSS systems involving multiple parameters and service reconfiguration. In some circumstances it will be evident that other considerations would become paramount, for example availability of relevant human expertise. A decision whether to appraise cost-effectiveness would need to be made on a case by case basis.

  4.5.5  Establishing cost-effectiveness using cost-utility methodology would allow interventions for patient safety to be evaluated on the same basis as other health technologies. For PSS that cover a range of diseases or the complete management of a condition, multiple models would be required to determine cost-effectiveness. There has to be sufficient flexibility in the system to accommodate this range. There has been very little research into applying HTA methodology to patient safety issues. We suggest that further methodological development is required including a review of the existing literature.

4.6  What data should be published?—Cost of implementation of patient safety solutions

  4.6.1  To assist the NHS with financial planning NICE develops costing tools to help organizations estimate the costs and savings associated with implementing our guidance.

  4.6.2  Estimating the national cost or saving associated with implementation can be challenging and subject to a degree of uncertainty, particularly if limited data are available and significant variation in current practice exists across service providers.

  4.6.3  In the pilot project, the lack of good quality national data concerning baseline practice for medicines reconciliation on admission to hospital was overcome by collecting a random sample and extrapolating the results to estimate wider practice. This pragmatic solution allows an estimate to be made where no published data are available.

  4.6.4  At a local level direct cost of provision of some PSS should be easier to estimate, depending on information being available on local circumstances. Estimating costs such as staff training and potential changes to number and grade of staff are relatively straight forward; estimating the effect of consequences, such as bed days avoided through safer practices preventing adverse events, is subject to greater uncertainty.


  5.1  NICE's Citizens Council brings the views of the public to NICE decision-making about guidance on the promotion of good health and the prevention and treatment of ill health. A group of 30 people drawn from all walks of life, the Citizens Council tackles challenging questions about values—such as fairness and need. The Citizens Council met from June 7- 9 2007 at the NICE headquarters in London. The Citizens Council was asked to consider the following questions

    —  Does the Citizens Council accept that it is appropriate when developing "patient safety solutions" that NICE takes the costs, as well as the benefits, into account?

    —  If the answer to question 1 is "yes," what principles of cost- effectiveness should apply?

    —  If the answer to question 1 is "no," what criteria should NICE apply in deciding whether or not it should recommend a particular safety solution to the NHS?

  5.2  At the three-day meeting the Council heard evidence on the extent of medical error in the NHS, on the case for and against taking cost-effectiveness into account when designing new measures, and how this could be done in practice. The Council heard from patients' representatives, doctors, nurses, health managers and economists. Three case studies were also considered in which risk reduction was needed or had been attempted, and the Council performed exercises intended to place them in the position of those whose task it is to make and justify decisions on what should be done. The Council questioned the experts and discussed the issues among themselves, collectively and in small sub-groups. To monitor changes in the responses to NICE's basic questions, tracking questionnaires were completed by Council members at the beginning and end of the meeting, and twice during its course.

  5.3  After three days of presentations from experts and debate, the Citizens Council concluded that it was appropriate for NICE to take costs as well as benefits into account when developing guidelines on the improvement of safety.

  5.4  The Council highlighted the fact that the economic methodology used by NICE in its other guidance programmes might not lend itself well to making decisions on cost-effectiveness in the area of patient safety. In particular, the fact that the perspective used by NICE does not include certain costs including litigation, cost to carers and those left behind following a death.

  5.5  The Council also pointed out some of the methodological weaknesses of the outcome measure used by NICE- the quality-adjusted life year (QALY). Given however that the QALY continues to be widely used in work of this kind and that currently there is no other adequate cost-effectiveness tool, the majority of the Council felt it would not be helpful to NICE simply to dismiss QALYs out of hand.

  5.6  So, with this absence of a suitable alternative, the Council suggested that if NICE is to use QALYs in assessing safety solutions, it should do so with a degree of flexibility greater than is normally the case when setting limits (the threshold) on how much the NHS should pay for each additional QALY equivalent gained as a result of an intervention. The Council recognised that departing from a single standard limit could create problems of its own—not least in consistency, However it also noted that NICE does on occasion move outside its own self-imposed limits when particular circumstances seem to justify such action.

  5.7  With respect to the circumstances that might be relevant when making decisions on the cost of equipment or practices intended to avoid error, the Council envisaged a sliding threshold limit. Factors that might contribute to the case for moving a threshold included:

    —  the severity to an individual of any likely injury or harm resulting from the error;

    —  the wider cost to society of coping with the aftermath of the error—cost to those left caring or bereaved, cost of litigation;

    —  the extent to which the error is unique to the medical environment (falls can happen anywhere; only in operating theatres do people have the wrong kidney removed); and

    —   the possibility that failure to address the safety issue in question could have a severely damaging effect on public confidence in the NHS.

5.8 We suggest that further consideration of economic methodology in the assessment of PSS should take account of the issues raised by NICE's Citizens Council.

September 2008

347   With advice on implementation in Scotland from NHS Quality Improvement Scotland Back

348   In Northern Ireland, from the DHSSPNI. Back

349 Back

350   Department of Health. Safety First: a report for patients, clinicians and healthcare managers. London 2006. Back

351 Back

352; and Back

353 Back

354 Back

355   ibid Back

356 Back

357 Back

358   Medical Subject Headings (MESH) are a standardised set of subject terms and an associated thesaurus that is used to catalogue medical research. It is maintained by the United States National Library of Medicine. Back

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2008
Prepared 30 October 2008