By Rebecca Lawton, Gillian Janes and Abi Albutt
Why is this necessary?
Within an NHS that has an acute shortage of time and resources, the often heard demand is ‘to do more with less’. However, evidence suggests that as much as 25% of healthcare is unecessary (Grol & Grimshaw, 2003). The NHS is characterised by a tendency to add more initiatives, protocols, interventions in an attempt, at least in part, to make care safer. Such processes and practices are not always evidence based but they may become integrated in a complex healthcare system and embedded within the culture of the institution.
There is increasing recognition that some clinical procedures are unnecessary and can do more harm than good. The Choosing Wisely campaign is actively encouraging health professionals to help patients choose care that is: supported by evidence, not duplicative of other tests or procedures already received, free from harm and truly necessary (http://www.choosingwisely.org/our-mission/). This campaign and much of the work to date has focussed on the removal of health technologies (Haas et al, 2012) and clinical practices such as specific tests and treatments that offer little or no benefit (see for example Bekelis et al 2017). In other words, the focus has been on clinical practices. However, to build a safer healthcare system, reduce costs and improve the process of care we must also consider the necessity of non-clinical safety practices, rules and procedures that do not necessarily result in safety improvements or no longer add value (Norton et al 2017). Whilst there is a lack of published evidence on this topic, making progress on this issue could also help create the space and time needed for healthcare staff to deliver more effective, patient-centred care.
How might it be achieved?
Whilst a common language for the process of removing practices that are no longer useful is still lacking (Davidoff, 2015; Bekelis 2017), terms such as decommissioning and disinvestment are often used. These reflect the top-down strategies most often deployed for this purpose by which external policy makers and commissioners decide what practices are least cost ineffective or evidence based and discourage health professionals from their use, based on funding disincentives and guidelines (Roosenhas et al 2015). This approach to influencing the clinical practice of health professionals is challenging and difficult to implement (Haas et al, 2012) with limited success thus far. It is plausible, however, that the staff themselves might know best which safety practices are not fit-for-purpose, do not result in benefits for safety or are just not possible to implement. While the use of bottom-up processes may prove more promising, there is little evidence on how to do this work or whether staff are able to identify low value safety practices that might represent opportunities for ‘disinvestment’ in the context of safety.
Moreover, if we are able to identify low-value safety practices, we do not yet know how easy it will be for staff to ‘stop’ doing these things. Enrico Coiera (2017) uses the term ‘mindful forgetting’ to highlight the conscious decision-making process that is needed to let go of low-value practices. In other words, concerted effort may be required. For example, there may be awareness that a safety rule, process or practice is not very effective, but this doesn’t mean it will automatically be stopped (Roman and Asch, 2014). The best way to approach the cessation or ‘mindful forgetting’ of inefficient or unhelpful safety practices in healthcare however is little understood (Niven et al 2015; Bekelis et al, 2017) and may be even more difficult for staff than the adoption of new innovations (Ubel et al 2015). It may be that the use of evidence-based behaviour change techniques could be considered to support staff in the cessation of unhelpful safety practices.
How can you get involved?
Researchers in the Yorkshire and Humber Patient Safety Translational Research Centre are working to address this evidence gap. Initially, we are seeking to understand what safety rules, processes and practices frontline healthcare staff perceive as low-value. Where the evidence supports this we will then work with staff and patients to develop interventions to support and evaluate ‘mindful forgetting’ of these in practice.
If you work in frontline patient care and would like to contribute to this new patient safety initiative, tell us your example(s) of low-value safety rules, practices or processes for consideration.
References
Bekelis K; Skinner J, Gottlieb D, Goodney P (2017) et al De-adoption and exnovation in the use of carotid revascularization: retrospective cohort study BMJ 359:j4695 doi: 10.1136/bmj.j4695
Choosing Wisely campaign http://www.choosingwisely.org/our-mission/
Coeira E (2017) The Forgetting Health System Learn Health Sys. 1:e10023 https://doi/org/10/1002/lrh2.10023
Davidoff F (2015) On the Undiffusion of Established Practices JAMA Internal Medicine 175(5): 809-811
Grol R, Grimshaw J (2003) From best evidence to best practice: effective implementation of change in patients’ care Lancet 362(9391): 1225-1230
| Haas M, Hall J, Viney R, Gallego G (2012) Breaking up is hard to do: why disinvestment in medical technology is harder than investment Australian Health Review 36(2): 148-152 |
Roman BR, Asch DA (2014) Faded promises: the challenges of de-adopting low-value care Ann Intern Med 161(2): 149-151
Roosenhas L, Owen-Smith A, Hollingworth W, Badrinath P, Beynon C, Donovan JL (2015) “I won’t call it rationing…”:’ An ethnographic study of healthcare disinvestment in theory and practice Social Science and Medicine 128: 273-281
Ubel PA, Asch DA, Chase C (2015) Commentary: Creating Value in Health by Understanding and Overcoming Resistance to De-Innovation Health Affairs 34(2): 239-244
Personally I would be interested in knowing which safety processes clinical teams think are redundant but laboratory staff think are essential. Examples that I see from my role as Clinical Governance lead for pathology include ward staff ignoring these. :-
1) Checking patient identity details on labels that are about to be attached to blood (& other) samples.
2) The 2 sample rule for blood transfusion. We have regular examples where the two samples are collected at the same time by the same doctor but labelled as being a few minutes apart. Our examples include one where we can prove that the wrong patient was bled! We already had a blood group on record for the patient so could detect the issue and reject the samples rather than potentially kill the patient with a wrong group transfusion.
3) Passing on results to the doctor when the lab staff consider them to be abnormal enough to telephone back to the patient location. In one case the patient had a cardiac arrest in medical imaging 6 hours after we telephoned a very high potassium result and was un-resuscitatable, acting on the result might have prevented the arrest, made it recoverable or led to discussion with the patient and a DNAR order being in place.
Hi Richard – thankyou for your comment and the examples shared, which sound like situations where the Datix or other relevant processes would be used.
I agree, the issue you raise about different staff perceptions of whether certain safety practices are or are not necessary is important and is being addressed at the next stage of the study by involving stakeholders in examining and determining which ideas will be taken forward to controlled testing in practice in order to mitigate this risk.
Does anyone have a bibliography of solid research which shows unequivocally which accepted safety practices are unproductive or counter-productive? It has to be unequivocal to be persuasive. Anecdotes are useful in stimulating research, but they are only anecdotes and we should not change policy until such anecdotes have been checked by robust research, with replicable relevant measurements, on representative samples of patients.
Recall that the level of adverse incidents (some trivial, some serious) is far too high, and we should be doing nothing to lower our guard. We want to increase standards, not reduce them. I agree that there is a danger of there being false positives as well as false negatives (Just read Gerd Gigerenzer’s Reckoning with Risk). However, we shall not be safe until everyone is using terms like “false positive” or “false negative”.
I write not only as a researcher, but also as a former elected hospital governor, and (perhaps more important) as a recent patient. I received excellent care overall, and was quite a willing participant in any activity which would reduced latent risks. I suspect that most patients would feel the same way. We do not need a situation in which patients want patient safety, but in which the local Board appears to wish to dilute efforts In that respect. Administrators need to practice EBM as well.
Having read the NIHR review of the work of Cochrane, I have my doubts. The job of meta analyses is to try to get the right answer. That is not necessarily a quick answer. The NIHR review keeps talking about “time windows” and “relevance”. It looks as though they are saying “never mind the quality, just feel the width”. It reads as though they are damning with faint praise, and trying to find something (anything) wrong with Cochrane, EBM, RCTs, or just about anything that might, just might, uncover inconvenient or expensive truths – however important those truths might be to patient welfare. That is why I have greater doubts about bureaucrats than I do about Cochrane and the researchers who provide their data. I hope that the current exercise will provide a forum for the two to cooperate.
Thanks for your considered response Laurence. We would like to stress that eliciting the views of staff about those safety practices they perceive to be of low value is only the first stage in a programme of research. We will identify those safety practices that are most freequently identified – patterns are already emerging – and we will then review the scientific literature to understand the evidence base for these practices. We anticipate finding that some, but not all of these practices are not underpinned by good evidence. We will update the website with findings as they arise.