Article Text

Download PDFPDF

An assessment of the quality and impact of NPSA medication safety outputs issued to the NHS in England and Wales
  1. Annette Lankshear1,
  2. Karin Lowson2,
  3. Saul N Weingart3
  1. 1Cardiff School of Nursing and Midwifery Studies, Eastgate House, City Road, Cardiff, UK
  2. 2York Health Economics Consortium, Market Square, University of York, UK
  3. 3Harvard Medical School and Dana-Farber Cancer Institute, Boston, USA
  1. Correspondence to Dr Annette Lankshear, Reader in Health Policy, Cardiff School of Nursing and Midwifery Studies, Cardiff University, Eastgate House, City Road, Cardiff CF24 0AB, UK; lankshearaj{at}cardiff.ac.uk

Abstract

Objectives To assess the quality and impact of medication safety outputs issued by the National Patient Safety Agency (NPSA) to the NHS in England and Wales.

Methods A multi-method study comprising (1) focus groups and interviews with NHS Chief Pharmacists and (2) an electronic survey of medical, nursing and clinical governance directors.

Results Acute sector respondents agreed that the medication outputs had a major impact on patient safety. Pharmacists welcomed national support for medication safety improvement, despite the resulting workload. Medical Directors were much less likely to be aware of alerts and Rapid Response Reports (RRRs) than their nursing and clinical governance colleagues. One key finding was the inability of around half of NHS trusts to communicate effectively and reliably with their junior doctors.

Conclusion Medication alerts issued by the NPSA have stimulated significant work to improve medication safety and are believed to have had an important impact on patient safety.

  • Healthcare quality improvement
  • medication safety
  • governance
  • patient safety
  • governance
  • healthcare quality improvement

Statistics from Altmetric.com

Background

Medication errors constitute a major threat to patient safety and account for a substantial proportion of the adverse events (those causing actual or potential harm to patients) reported to vigilance organisations across the world.1 2 In the calendar year 2007, the UK Reporting and Learning System (RLS) recorded nearly 86 085 adverse events attributed to medication incidents, a figure that constitutes 9% of all reports.2

Since its inception in 2002, the National Patient Safety Agency (NPSA) has issued a number of safety alerts designed to improve medication safety. Originally, these Patient Safety Alerts were published together with supporting documentations such as patient information and record booklets (methotrexate) or training materials and staff competencies (anticoagulants). Since the Chief Medical Officer's call for faster dissemination of warnings to the NHS,3 most of the outputs have taken the form of one page Rapid Response Reports (RRRs) which are normally disseminated electronically within weeks of notification of the trigger incident.

In January 2009, the NPSA invited the authors to undertake a project with the aim of assessing the impact of their medication alerts and other outputs on the NHS in England and Wales (table 1), with a reporting timescale of 6 months. This paper reports on the awareness amongst key stakeholders in acute NHS organisations and their assessment of quality, appropriateness and impact.

Table 1

Summary of safer medication team outputs (http://www.npsa.nhs.uk/nrls/alerts-and-directives/)

Methodology

We employed a mixed methodology. The major components of the review and the subject of this paper comprised individual and group interviews with, and surveys of, NHS stakeholders. We chose respondents who could represent staff dealing with prescribing of medication (Medical Directors), dispensing (Chief Pharmacists), administration (Directors of Nursing) and those assessing the impact on risk reduction (Clinical Governance Directors).

Data collection

Interviews and focus groups

We invited 95 Chief Pharmacists to one of the three focus groups held in London, York and Wrexham, respectively. Questions were designed to stimulate discussion in which the moderators played no part, other than to make occasional requests for clarification. Four of them who could not attend were interviewed face to face (3) or by telephone (1).

Surveys

A single survey designed to assess awareness of, and experiences with, the medication outputs was developed for use by Medical, Nursing and Clinical Governance Directors (hereafter referred to as the ‘Directors’ survey'). The Medical Directors were made aware of the survey by means of the Chief Medical Officer's monthly Bulletin. Clinical Governance Directors and Nursing Directors were emailed copies of the survey which was also made accessible via a web link. Three reminders were sent out at fortnightly intervals. A second separate survey, which asked more detailed questions about alert implementation, was mailed to 381 Chief Pharmacists. The surveys, which included a mix of questions with Likert response scales and open-ended responses, were pilot tested with representatives from each constituency, and changes were made to clarify the questions and limit their number.

Analysis

Focus groups and face-to-face interviews were digitally recorded and transcribed, and then analysed using thematic analysis techniques by AJL.4 Survey data were entered into an electronic database for analysis using double-entry methods to ensure accuracy. Descriptive quantitative analyses were undertaken by KVL using SPSS v 18.

Findings

Twenty six of the 96 invited Chief Pharmacists attended the focus groups (at which attendance varied from six to ten) and a further four (who could not attend the events) were interviewed individually. The survey response rates are shown in table 2. In 38 cases, Medical or Nursing Directors passed the survey to another individual. In 27 cases, this was a pharmacist and these responses are analysed separately as ‘others (pharmacists)’. Only one Chief Pharmacist completed both surveys.

Table 2

Survey response rates

It should be noted that some of the questions were common to both Directors and Pharmacists' surveys but the latter contained additional questions which explains the variation in the presentation of results.

Topic selection

Within the NHS, there was a high degree of satisfaction with the medication safety topics addressed which were, with few exceptions, perceived to pose a high risk to patients. Overall, 81% (135) of respondents to the Directors' survey were of the view that the topics chosen ‘definitely’ or ‘probably’ addressed issues presenting the highest risk of harm to patients (see table 3).

Table 3

Focus and quality of medication safety alerts

It was said that the publications had empowered pharmacists to achieve change in medication management and practice. One pharmacist stated:

‘I think … it's been a very effective way of getting medication safety up there and high on the agenda in all the trusts … Many of the things that the NPSA have asked us to implement are things we've all known we had to do, been trying to do for years but been unable to do because there's been no bit of national guidance that says you must do it.’ (Chief Pharmacist, FG1)

Dissemination

Responsibility for the appropriate dissemination of safety outputs lies within the recipient organisations. Our survey demonstrated significantly lower awareness of documents amongst Medical Directors (54%, n=25) than their Nurse Director colleagues (81%, n=32). See table 4.

Table 4

Analysis of awareness of Rapid Response Reports (RRRs)

In focus group discussions, pharmacists expressed concern about their ability to communicate medication safety information to junior doctors, which prompted the insertion of a question to this effect into the Directors' survey. Only 43% of Medical Directors (n=20) stated that they could, if the need arose, ‘communicate reliably and effectively’ with all junior doctors working within their trust. Part of the problem, which underscores several challenges, lies in rotating assignments and frequent turnover, but other difficulties reflect unreliable communication methods such as inactive email accounts and out of date contact information.

Implementation

Inevitably, it was the more complex and challenging guidance that received the greatest amount of air time, whether in the interviews and focus groups or within the comments section of the survey. Large and complex alerts, particularly those for anticoagulants and injectable medicines, have given rise to a significant (although necessary) workload in acute trusts and this was exacerbated by their simultaneous publication with three other alerts. At the time at which the study was undertaken, about 50% of NHS organisations in which the alert was relevant had not completed the injectable medicines alert, which had required a full risk assessment of all such medicines in each clinical area. Typical comments included the following:

‘Injectable medicines was like trying to boil the ocean…. The policy was hugely difficult to write – especially as we tried to incorporate theatres which many trusts did not.’ (Chief Pharmacist interview 3)

‘When I got stuck I went to one of the trusts that had signed it off, but they hadn't resolved that issue either, so I would take the signing off with a pinch of salt.’ (Chief Pharmacist, FG2)

Participants in all three focus groups were of the opinion that a more efficient way to have handled the injectables alert would have been to focus the assessment in a few pilot trusts and it was suggested that these complex alerts might, in the future, be better handled as ‘campaigns’ rather than guidance that has to be signed off as complete within a given timescale. Chief Pharmacists pointed out the significant difficulties involved in assessing staff competencies and argued that, if timescales are not consistent with full implementation, trusts are placed in an invidious position with regard to quality standards.

The comments of all respondents reflected the challenges inherent in managing the complex requirements of key guidance; including identifying the point at which they could, with integrity, be said to be complete; the need to devise policies that covered widely disparate clinical environments and the pressure of the given timescales, especially in the light of the training requirements.

‘The challenge is getting 100% compliance. Its fine when you can control supply (eg, potassium chloride or midazolam) harder when it's a question of prescribing practice (eg, anticoagulants).’ (Director of Nursing, survey)

Clinician engagement in implementation was cited as an ongoing problem acknowledged by a wide range of respondents and to address this medical respondents recommended closer working between the NPSA and medical Royal Colleges in order to avoid contrary guidance. However, when the NPSA explicitly involved such a body in the development of the Midazolam alert this had unintended consequences:

‘You may get what happened with the midazolam alert because I think the College of Anaesthetists was involved. They go down one line because that is, if you like, their turf and they forget there's as much (sic) people doing conscious sedation that don't actually use anaesthetists at all.’ (Chief Pharmacist, FG1)

Impact

A total of 84% of respondents (including 89% of chief pharmacists) confirmed that the NPSA medication outputs had ‘definitely’ or ‘probably’ led to meaningful improvements in patient safety. This unambiguous endorsement was reflected in the finding that 76% of respondents (including 72% of Medical Directors) reported that all grades of clinical staff were more aware of medication safety issues as a result of the outputs (table 5).

Table 5

Perceived impact of alerts

Chief Pharmacists concurred with these remarks, citing those deemed to have had the greatest impact. Amongst acute trust Chief Pharmacists, the potassium chloride and methotrexate alerts were a popular choice and illustrative comments included the following:

‘the (Potassium chloride) alert … was very useful, very sensible and encompassed everything. And it went outside of pharmacy. And I think that's important, because pharmacy have always tried to solve these problems.’ (Chief Pharmacist, FG3)

‘Methotrexate across the community's been certainly highlighted when it was first implemented. And we continue to review and audit that across the health community.’ (Chief Pharmacist, FG3)

Other guidance mentioned included anticoagulants, injectable medicines and opioids. One of the respondents commented:

‘Probably the one dealing with anticoagulants was the most useful and it came at the right time because, obviously, VTE [venous thromboembolism] was an issue through the House of Commons reports.’ (Chief Pharmacist, FG1)

Discussion

Concern has been expressed both in the UK and elsewhere that safety alerts issued by the government and quasi-governmental agencies are not necessarily exerting the influence that they should.5–9 A recent report issued by the Association of Victims of Medical Accidents (AVMA)10 led to allegations that alerts were being ignored by a significant proportion of NHS organisations, but although we have previously9 expressed some concern about the management of alerts these current findings suggest that at least some of the delay in signing them off (the basis of the AVMA's reported non-compliance) may lie in the complexity of the requirements themselves. Many of the trusts in our study reported fierce levels of activity combined with a rigorous interpretation of the challenging criteria for signing off the alert as complete.

The study has a number of limitations, mainly resulting from low response rates to the electronic survey (despite its inclusion in Chief Medical and Nursing Officer newsletters) and the very brief timescale for the study which did not permit an assessment of actual changes in practice. Nevertheless, it does confirm a widespread perception amongst senior NHS managers that medication alerts issued by the NRLS have addressed the major areas of risk in acute care organisations, have been well disseminated and implemented and have had a major impact on medication safety.

Within the conceptual framework first suggested by Rogers11 and expanded by Greenalgh et al12 for successful spread of innovation, it is clear that, with very few exceptions, the alerts have addressed observable defects in the system, many of which have been the subject of concern for many years and have contained good advice that gives clear benefits (relative advantage). Additionally, the earlier alerts met the requirement for augmentation support in their accompanying publication of training materials and patient information. Respondents recommended piloting to ensure trialability and to reduce the cost of implementation. The aforementioned authors also predict that the high complexity of some of the key alerts (anticoagulants and injectable medicines) might have reduced the likelihood of successful implementation and this proved to be the case, certainly within the timescales set, although it is questionable whether there was any alternative simple or incremental solution to either problem. Finally, the difficulties of engaging clinicians in the implementation, well recognised in the international risk literature,13 hints at a lack of compatibility between the alerts and the values of the medical profession. One of the possible causes of this disconnect may be the lack of a robust evidence base on which to base recommendations.7 Another may be the provenance and route of dissemination of the alerts (from a government agency distributed via managers) rather than having the stamp of a professional organisation.14

In the light of recent evidence published by the General Medical Council,15 the lack of ability of many NHS trusts to communicate directly and reliably with junior doctors on the subject of medication safety (or anything else) must be a cause for concern, given their key role in the management of patients and the possibility of prescribing error.16–18 If the many quality and safety initiatives currently taking place in England and Wales are bypassing this group, as they have in the US,19 20 it is questionable whether the benefits will be sustainable. The key role of foundation year directors may be under-appreciated by those within trusts with a responsibility for disseminating alerts and ensuring medication safety.

Conclusion

The medication outputs of the NRLS are judged by NHS leaders to address appropriate, high-risk targets. The various alerts have been widely disseminated to senior staff and are consistently perceived to have improved medication safety in the UK.

References

Footnotes

  • Funding National Patient Safety Agency, 4-8 Maple Street, London W1T 5HD, UK. Other Funders: National Patient Safety Agency.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.