comment “The ‘hospital bad/home good’ narrative is disingenuous” DAVID OLIVER “Hancock should stop trying to distract us with new technology” HELEN SALISBURY PLUS Respecting patients’ health beliefs; managing pandemic panics

WOUNDED HEALER Clare Gerada The GMC is no longer a bogeyman to fear or years I’ve been caring for doctors a doctor’s health problem will not lead to a GMC with mental illness. This has involved sanction if the doctor is receiving the necessary numerous interactions with the GMC. treatment and is managing any risks to patients. Years ago, together with others, I Even if a health case does get referred to the regulator raised concerns about the impact of its the chance of a doctor receiving a serious sanction Finvestigations on doctors’ mental health, including is immensely small. Over a five year period around links to suicide. 400 doctors were investigated for issues relating to Even I was frightened entering the GMC’s London substance misuse, resulting in three erasures and 20 headquarters (which I had to do fairly often)—but this suspensions. These would almost certainly include didn’t compare to the fear my patients felt when the issues other than health (especially conduct), and white envelope marked “confidential” landed on their the GMC will never erase doctors solely for matters doorstep or when they had to attend meetings or relating to their health. hearings at the GMC. It’s been nearly 20 years since Janet Smith chaired How things have changed. Over the past five years the Shipman inquiry. Her recommendations changed the GMC has undergone significant transformation our relationship with the GMC and essentially drove in how it interacts with and deals with registrants. It a wedge between us. But this is now changing, and has changed how it communicates and has amended I welcome it. By working together with the GMC, we its processes to include provisional inquiries and can improve the lives of our doctors and, in turn, the ability to pause investigations. The emphasis improve the care these doctors deliver on “local first” (local resolution to support efficient to patients. handling of cases) reduces the impact on doctors and Clare Gerada is GP partner, Hurley Group, achieves more timely resolution of complaints for London [email protected] patients. The GMC’s encouragement of employment Cite this as: BMJ 2020;368:m647 liaison officers and its training of staff, including how to connect with registrants and to recognise those The ethos is now who are distressed, has improved communication. The ethos is no longer simply “protecting patients”: “supporting instead, it is “supporting doctors to protect doctors to protect patients”—a massive shift of emphasis. patients” The GMC has also produced a suite of papers and independent reports through its Supporting a Profession Under Pressure initiative. These publications are of an incredibly high standard, containing data and analysis that would put most academic publications to shame. It is committed to implementing the recommendations of these independent reviews. Doctors still fear the regulator, and this is understandable. But, as it changes, maybe we should as well. We must start challenging our deeply held views and look at the facts. For example, disclosing the bmj | 29 February 2020 323 BAD MEDICINE Des Spence Why #healthbeliefs should be top of all our communications In the modern world doctors do not have a monopoly on truth

t was 1986—before the preferred atheism has no monopoly on truth: truth cultural elements of health beliefs are pronoun, when # was an obscure is mercury. But our beliefs blind us, and conditioned directly through the actions wannabe on life’s keyboard, and when challenging them invariably leads to of a country’s medical profession: doctors avocados were never crushed but conflict. Best always to respect others’ attempting to extort as much money were eaten unripe at Christmas, with beliefs. as possible through over-investigation, Iprawns and Marie Rose sauce. Doctors are intelligent people—no, we overtreatment, and needless surgery. I reached for my near empty packet of truly are! Medicine is but a belief system Medicine’s core business plan is to make Marlboro beside a mug full of cigarette too, rife with false deities, leaps of faith, even well people think that they’re sick. ends. I struck a match, being careful not learnt incantations, and physical rituals. By our standards, North Americans often to torch my “big hair” laden with product, I don’t believe in quite a lot of it: most have aberrant medical beliefs (though and twisted my earring anxiously, thinking clinical examination; primary prevention normal to them). Antibiotics and scans at how it didn’t suit me. I clasped my cigarette treatment of cholesterol; bisphosphonates; every consultation, annual blood tests, and stained hands together and pressed them to migraine treatments; and, of course, all foot pre-emptive cancer surgery. Doctors have my head, praying to God that the brachial and shoulder surgery. All just post-truth, fixed health beliefs too: even suggesting plexus wouldn’t come up in my exam. fake scientific nonsense. (Getting angry?) that homeopathy is an effective placebo Then I raised my head and laughed. turns both sides of the medical divide into Had my anatomy revision come to this? My Cultural elements frothing banshees. And health beliefs are atheism had been usurped by Christian This model suggests that everyone has a set ubiquitous, in every consultation. Doctors conditioning. The thing is, beliefs defy of health beliefs, passed down in families frequently fail to understand patients’ reason. Many scientific people believe and from wider society too. Think of the health beliefs, dismissing those not in one of the many competing gods. And families who endlessly attend clinics. The congruent with their own. Predictably, this

BMJ OPINION Robert Peckham Covid-19 has shown we need strategies to manage panic My local supermarket recently made the news. Panic is a word that’s been widely used There’s also a thin line between The grey metallic shelves, emptied of food, during this outbreak. We’ve heard the misinformation and uncertainty, were a stark reminder of the panic that covid- oft repeated injunction by experts and particularly in the midst of an epidemic 19 has sparked in Hong Kong and across the media commentators for people to stop region. People are bulk buying provisions, panicking. Throughout history people have us. Panic, in other words, can be a hair’s fearing the worst, and it’s contagious. The sight feared the spread of disease, but global breadth away from prudent concern. of a foodless food store is enough to make connectivity creates new possibilities for There’s also a thin line between anyone sweat. disruption. As the World Health Organization misinformation and uncertainty, particularly has noted, covid-19 is driving a social in the midst of an epidemic. The identity of media “infodemic.” In the tweet-a-second the causal pathogen may be unknown, its 21st century, information overload makes it epidemiology obscure, and the spectrum of increasingly difficult for people to distinguish clinical manifestations confusing. All of this is fact from fiction. grist to the panic mill. Communication, which is so central to In Hong Kong, long queues for face masks public health, also turns out to be key to form outside pharmacies from the early panic. Circuits of information can all too easily hours. Shops are emptied of rice, toilet paper, flip to become conduits of misinformation. and disinfectants. In part, this is triggered Public health messages that seek to jolt us by a self-fulfilling pre-panic panic: people out of complacency can all too readily alarm rush to forestall the effects of a future panic.

324 29 February 2020 | the bmj Medicine is ACUTE PERSPECTIVE David Oliver but a belief system too, Fixing the older inpatient narrative rife with false deities, knelt by the patient’s bed. Admitted among developed OECD nations, and leaps of four days earlier when acutely unwell, England has fewer still. Our hospitals are faith, learnt confused, and immobile, he was now too full, and we’ve not only closed too incantations, improving. “How are you today?” many beds but have put more out of action and physical “Doctor, I must say, I’m because of stranded patients who can’t rituals. I Iconsiderably transformed and substantially transfer to under-resourced community don’t believe restored. I almost feel ready to go home.” services. The problem is not that people in quite a lot Right now, in the language used around shouldn’t be admitted or can’t benefit from of it acute care, admission is relentlessly admission—it’s that they stay too long. portrayed as a bad thing, especially We should care about research evidence leads to dissatisfaction on all sides, and for older patients. It suggests they’d as much as ideology and policy pushes. doctors are seen as arrogant. always want to stay at home or get home Systematic review and meta-analysis of Thirty years on from my studies the sooner—though this barely considers the multidisciplinary, specialist led, geriatric fashions and habits have changed, but we patients who welcome admission or a assessment of frail older people in hospital still cram students’ heads full of irrelevant slightly longer stay, or family providing has repeatedly shown benefits lasting for factual details while avoiding the more care, who are often stressed, burnt out, months after patients leave. The comparable important concepts and ideas that make and unsupported. This narrative says evidence for this assessment in community us better doctors. My health belief is this: inpatients risk hospital acquired infection, settings, including “hospital at home” in the coming decades imaging, laboratory poor nutrition, deconditioning, delirium, models, is less convincing. Meanwhile, testing, and diagnostic algorithms will depersonalisation, iatrogenic harm, and the major research reviews led by the Nuffield usurp the profession’s position. perils of bed rest. There are data to support Trust of integrated community models Only those of us with high emotional such arguments. But many of these patients and approaches have shown no consistent intelligence will survive, and core to have progressive, long term conditions or evidence for reducing admissions or bed this is respecting patients’ differing frailty, and would do as badly, or worse, at use or the associated costs. health beliefs and agendas. But home or in community settings. Of course, we want more older people to #imightbecompletelywrong, and The rhetoric is meaningless unless you remain in their homes and return there. And #godmightexistalthoughthisseemshighly look at the counterfactual—what would admission—especially if prolonged beyond unlikely. have happened if the patient hadn’t come the point of adding value—can lead to risks Des Spence, GP, Glasgow in. And, despite frequent reports suggesting and make it harder to return home. But we [email protected] many inpatients have needs that don’t need to stop badmouthing hospitals on very

Cite this as: BMJ 2020;368:m695 require acute hospital beds, we know from selective evidence just because we have a reports on intermediate care, community perennial bed crisis. BMJ OPINION Robert Peckham services, primary care, and district nursing Many patients are “considerably that adequate capacity is hypothetical, not transformed and substantially restored” Conspiracies abound online, spurred by actual, so audits of bed use can mislead. after admission. They and their families are repeated calls to stop panicking, and by The “hospital bad/home good” glad of it in the short term, and often the public’s widespread mistrust of the narrative, which makes it a thought the alternatives just aren’t there. government. crime to make a pragmatic clinical David Oliver, consultant in geriatrics and The ongoing covid-19 outbreak has decision to admit patients, is acute general medicine, Berkshire underlined the need for more research on the disingenuous. The real drive [email protected] nature of panic and its drivers. We need to behind it is that the UK is one of the Twitter @mancunianmedic better understand the extent to which panic is lowest for beds per 1000 population Cite this as: BMJ 2020;368:m652 universal or culturally specific. How precisely is panic linked to trust—whether that is trust in the government, media, or experts? We should care Panic is still too little studied and too about research little understood. It is easily dismissed as a distraction to the main task of containing an evidence epidemic. And yet the management of panic as much as is likely to be key to managing infectious ideology and diseases in an ever more connected world. policy pushes Robert Peckham is MB Lee professor in the humanities and medicine, chair of the Department of History, and founding director of the Centre for the Humanities and Medicine at the University of Hong Kong

the bmj | 15 February 2020 325 PRIMARY COLOUR Helen Salisbury LATEST PODCAST AND VIDEO Hancock as conjuror

his week Matt Hancock, Recently, The BMJ published a secretary of state for health systematic review of AI based skin and social care, welcomed checking apps currently available, the introduction of the which found that they were dangerously Medicines and Medical inaccurate and inadequately regulated. Devices Bill, tweeting a Daily Express AI may seem like magic: you feed data

T SMITH H RICHARD headline that proclaimed, “NHS red tape and results into a black box and it spots slashed to boost patient care.” While I’d previously unseen patterns, allowing it to Born equal: launch of the hate to be branded a Luddite, I admit to make predictions from new information. Racism in Medicine issue scepticism when it comes to medicine’s Although in the future we may come to rely Earlier this month, The BMJ published a digital revolution. In the rush to approve on such technical achievements, what’s special edition which tackled subjects from new devices and apps, who defines what available now seems to have been rushed differential attainment in medical school to the is red tape and what is due diligence? to the market with inadequate testing or physiological effects of experiencing everyday Some patients are keen to take control regulation. The datasets used in machine discrimination, matters also discussed by of their health and to avoid bothering learning pose a particular problem, with a speakers at a launch event. Here Mala Rao, a their GP, especially as they’re constantly recognised risk of AI compounding racial professor of public health at Imperial College reminded about how hard it is to get an bias and prejudice. London, talks about why it affects everyone: appointment. So, it seems logical and My concern is that the current digital “Race equality matters to us all . We must responsible to use all available tools, offer is not real wizardry but something remember that if we have valued, respected, which might include an online symptom more akin to prestidigitation—a conjuring motivated, engaged staff, then it's good for the checker or a smartphone app to analyse trick. The word suggests that the art of 84% of our population which is white and for a photo of a funny looking mole. stage magic lies in the speed with which the 16% that’s our black and minority ethnic This would be sensible if the technology you can move your digits, but in reality population. So the business case for the NHS is was good enough, but we have steadily it’s all about misdirecting attention. well made.” accumulating evidence that the Perhaps we could ask Hancock to stop algorithms aren’t yet up to the job. October trying to distract us with new technology 2019 saw a flurry of comments about a and to concentrate on the real and Building an evidence symptom checker app that had concluded deadly problems facing our NHS. While base for covid-19 that a 60 year old man with chest pain was ever increasing numbers of seriously ill The latest episode of Talk Evidence pieces having a heart attack but that a 60 year patients languish on trolleys in emergency together all the emerging evidence we have old woman with identical symptoms was departments, where is Hancock’s about covid-19. Raina MacIntyre, a professor having a panic attack. Multiple examples promised plan for social care? Has he even of global biosecurity at the Kirby Institute, of bizarre and unlikely conclusions planted the magic nurse and GP tree? University of New South Wales, Australia, have been demonstrated, including Helen Salisbury, GP, Oxford discusses what areas of evidence need to be lower leg pain and breathlessness [email protected] prioritised in this outbreak: being diagnosed as a Charcot’s Twitter @HelenRSalisbury “It’s really important to get a good evidence joint and asthma. Cite this as: BMJ 2020;368:m648 base for a number of different parameters around covid-19. One is to understand the The digital offer is transmission dynamics. And there’s still not real wizardry uncertainty about, for example, how much but something asymptomatic transmission is happening. Obviously, we need data on the immunology of more akin to the virus so that we can develop vaccines. Then, prestidigitation of course, we need to understand the clinical picture and the treatment. We need to see the clinical trials emerge so we can actually assess the efficacy of therapeutic options. “In two months, the number of cases and deaths from covid-19 has exceeded what

Listen and subscribe to The BMJ podcast on Apple Podcasts, Spotify, and other major podcast apps

Edited by Kelly Brendel, deputy digital content editor, The BMJ

326 29 February 2020 | the bmj QUALITY IMPROVEMENT Revitalising audit and feedback to improve patient care Robbie Foy and colleagues argue that the full potential of clinical audits can be unleashed through a more evidence based and imaginative approach

ealthcare systems face challenges Cumulative Box 1 | Questions for audit programmes and healthcare organisations in tackling variations in patient care incremental to consider in designing, implementing, and responding to audit and 1 2 and outcomes. Audit and feedback gains through feedback8 aim to improve patient care by repeated Nature of the desired action reviewing clinical performance audit cycles Can you recommend actions that are consistent with established goals and Hagainst explicit standards and directing action priorities? 3 can deliver towards areas not meeting those standards. It is Can you recommend actions that can improve and are under the recipient’s a widely used foundational component of quality transformative change control? improvement, included in around 60 national Can you recommend specific actions? clinical audit programmes in the United Kingdom. Nature of the data available for feedback There is currently a gap between what audit Can you provide multiple instances of feedback? and feedback can achieve and what they actually Can you provide feedback as soon as possible and data frequency informed deliver, whether led locally or nationally. Several by the number of new patient cases? national audits have been successful in driving Can you provide individual rather than general data? improvement and reducing variations in care, Can you choose comparators that reinforce desired behaviour change? such as for stroke and lung cancer, but progress Feedback display is also slower than hoped for in other aspects of Can you closely link the visual display and summary message? care (table).4 5 Clinicians might feel threatened Can you provide feedback in more than one way? rather than supported by top-down feedback and Have you minimised extraneous cognitive load for feedback recipients? rightly question whether rewards outweigh efforts Delivering feedback invested in poorly designed audit. Healthcare Have you addressed barriers to feedback use? organisations have limited resources to support Can you provide short, actionable messages followed by optional detail? and act on audit and feedback. Dysfunctional Have you addressed credibility of the information? clinical and managerial relationships undermine Can you prevent defensive reactions to feedback? effective responses to feedback, particularly when it is not clearly part of an integrated Can you construct feedback through social interaction? approach to quality assurance and improvement. Unsurprisingly, the full potential of audit and Apply what is already known feedback has not been realised. There are ways to maximise returns from the Audit and feedback generally work. A Cochrane review of 140 considerable resources, including clinician time, randomised trials found that they produced a median 4.3% invested in audit programmes. These include absolute improvement (interquartile range 0.5% to 16%) in applying what is already known, paying attention healthcare professionals’ compliance with desired practice, such as to the whole audit cycle, getting the right message recommended investigations or prescribing.3 This is a modest effect, to the right recipients, making more out of less but cumulative incremental gains through repeated audit cycles can data, embedding research to improve impact, and deliver transformative change. Audit and feedback also influence harnessing public and patient involvement. reach and population through scaled up national programmes, which other quality improvement approaches (such as financial KEY MESSAGES incentives or educational outreach visits) might not achieve with similar resources. Clinical audit and feedback entail reviewing clinical • The interquartile range in the Cochrane review indicates that a performance against explicit standards and delivering quarter of audit and feedback interventions had a relatively large, feedback to enable data driven improvement positive effect of up to 16% on patient care, whereas a quarter had The impact of audit could be maximised by applying • a negative or null effect. The effects of feedback can be amplified by implementation science, considering the needs of clinicians ensuring that it is given by a supervisor or colleague, provided more and patients, and emphasising action over measurement than once, delivered in both verbal and written formats, and includes Embedding research on how to improve audit and feedback • both explicit targets for change and action plans.3 A synthesis of in large scale programmes can further enhance their expert interviews and systematic reviews identified 15 “state of the effectiveness and efficiency science,” theory informed suggestions for effective feedback (box 1).8 the bmj | 29 February 2020 327 Examples of national clinical audit programmes and randomised trials evaluating audit and feedback Objective Methods Illustrative findings National clinical audit programmes To measure and improve the structure, The National Clinical Audit for Stroke operates a prospective, continuous audit of the Stroke unit performance in key aspects of care improved over processes, and outcomes of stroke care processes and outcomes of NHS funded stroke care and rehabilitation in acute and five years; eg, the proportion of patients assessed by a stroke post-acute settings in England and Wales. It also reviews care at six months and beyond specialist consultant physician within 24 hours rose from 74% to assess how longer term needs are met to 83%, whereas the proportion of applicable patients screened for nutrition and seen by a dietitian by discharge rose from 66% to 81%.4 However, significant gaps in provision remain; eg, fewer than one in three patients receive a six month review To measure and improve care and For the National Lung Cancer Audit, secondary and tertiary care NHS hospitals in The proportion of patients alive at least one year after diagnosis outcomes for lung cancer England and Wales submit data via the National Cancer Registration and Analysis rose from 31% in 2010 to 37% in 2017.5 However, almost Service as part of the Cancer Outcomes and Services Dataset. The data are linked to a third of patients still lack access to the benefits of specialist Hospital Episode Statistics, the National Radiotherapy Dataset, the Systemic Anti-Cancer nursing support Dataset, pathology reports, and death certificate data Randomised trials of audit and feedback To assess the effect of adding an action 21 Dutch intensive care units were randomly assigned to receive usual electronic Over six months, the proportion of patient shifts with adequate implementation toolbox to electronic audit feedback only or to feedback with an implementation toolbox suggesting practical pain management increased by 14.8% compared with 4.8% in and feedback targeting quality of pain actions staff could take to improve pain management the feedback only group management in intensive care units16 To assess the effects of feedback including 1581 English general practices whose prescribing rate for antibiotics was in the top Over six months, the rate of antibiotic items dispensed per 1000 “social norm” persuasive messaging and 20% for their locality were randomly assigned to receive feedback including a letter population was 127 in the feedback intervention group and 131 patient focused information on antibiotic from England’s chief medical officer highlighting the higher rate of antibiotic prescribing in the control group, representing an estimated 73 406 fewer prescribing in higher prescribing general or to no communication. They were then randomly assigned to receive patient focused antibiotic items dispensed. The patient focused intervention did practices7 26 information promoting reduced use of antibiotics or to no communication not significantly affect prescribing

Pay attention to the whole cycle Box 2 | Questions Get the right message to the that healthcare right recipients The audit and feedback process comprises one or more cycles organisations can ask of establishing best practice criteria, measuring current themselves about Feedback comparing performance among different performance13 practice, feeding back findings, implementing changes, and healthcare organisations and clinicians can leverage further monitoring. This chain is only as strong as its weakest Do we know how competitive instincts. This might not always work as link. Feedback effects can be weakened by information- good we are? intended. Nobody likes being told they are getting intention gaps (feedback fails to convince recipients that Do we know where we it wrong, repeatedly. Low baseline performance stand relative to the change is necessary), intention-behaviour gaps (intentions are is associated with greater improvement after best? not translated into action), or behaviour-impact gaps (actions feedback3 but can elicit defensive reactions (“I don’t Do we know where do not yield the desired effect on patient care).9 The success of believe these data”), especially if feedback does and understand why national audit programmes depends on local arrangements not align with recipient perceptions (”My patients 10 variation exists in our that promote action as well as measurement. organisation? are different”). Such responses are not uncommon A synthesis of 65 qualitative evaluations proposed ways of Over time, where given that clinicians tend to overestimate their 15 designing audit programmes to better align with local capacity, are the gaps in our own performance. Continued negative feedback identity, and culture and to promote greater changes in clinical practice that indicate perceived as punitive can also be demotivating and behaviour.11 Clinician beliefs about what constitutes best a need for change? risk creating burnout (“What else can I do?”). practice can influence how they respond to feedback, so audit In our efforts to Giving feedback to professionals who take pride programmes need to consider these while also challenging improve, what’s in their work requires careful thought. Given the law the status quo. All aspects of audit programmes should be working? of diminishing returns, attempts to improve already designed with a focus on the desired changes in behaviour; for high levels of performance might be less fruitful example, feedback tackling unnecessary blood transfusions than switching attention to other priorities. Many could include suggested alternative approaches to minimise clinical actions have a “ceiling” beyond which blood loss during surgery.12 We need to understand existing improvement is restricted because healthcare barriers to desired change and have a plan for how feedback organisations or clinicians are functioning at or helps to tackle those barriers. near their maximum capabilities. Without functioning local networks and systems, national A range of approaches can help tailor feedback audit programmes can become echo chambers, where good to recipients’ needs. First, feedback can include intentions and blame for limited progress reverberate. Audit comparators that show like for like and set realistic and feedback will flounder if local quality improvement goals for change relative to performance levels. is based on repeated, unconnected, and inappropriately Second, feedback can be delivered alongside delegated projects conducted in isolation from mainstream a range of tangible action plans to support pursuits and if any learning is dissipated in collective amnesia. improvement; for example, an implementation Clinical and managerial leaders should ask questions about toolbox improved pain management in intensive their organisational performance in response to feedback care units.16 17 Third, new audit criteria need to be (box 2)13 and set clear goals, mobilise resources, and promote convincing, based on robust evidence and with continuous improvement.14 scope for patient and population benefit.

328 29 February 2020 | the bmj Make more out of less data

Healthcare organisations and clinicians can struggle to act on all feedback from national and local audit programmes. A 2012 snapshot identified 107 NICE clinical guidelines relevant to primary care, resulting in 2365 recommendations.18 Audit programmes can help to identify which recommendations have the greatest potential to benefit patients and populations. One of the highest costs associated with audit programmes is the time and effort involved in data collection, particularly the manual review of patient records. This burden can be compounded by temptations to add in more variables for analyses that marginally improve precision.19 The resulting feedback might reinforce the credibility of data and enable recipients to explore associations in the data. Providing larger amounts of complex data, however, risks cognitive Patients and Harness public and patient involvement overload and distracting recipients from key messages. the public The increasing availability of electronic patient record represent Healthcare providers and researchers are still learning how systems and routinely collected data on quality of care offer an untapped to work meaningfully with patients and the public, and opportunities for large scale, efficient feedback programmes. there are opportunities in audit programmes. This means force for Such approaches offer greater population coverage, which moving beyond current models of involvement—typically can overcome risks of biased sampling associated with change advisory group roles to ensure accountability and contribute manual review, such as the loss of records of patients with to strategy—towards active participation in feedback and poorer outcomes. Routine data can also be collected and service improvement. analysed in real time, thereby enabling faster, continuous Patients and the public are often surprised by the extent feedback and countering objections voiced by clinicians of unwarranted variations in healthcare delivery, which (“These data are out of date”). is the core business of audit programmes.25 They express Data quality is only as good as coding at the point frustration at the difficulties in routinely measuring less of care. Validity checks and quality control of the data technical aspects of care, such as consultation skills and might compound the burden on clinical teams. Data patient centredness. Involving patients and the public, linkage and extraction across different information including seldom heard communities, early in the process requires compliance with data protection and information of developing indicators is important. Audit programmes governance requirements. Even with all this in place, we can be at the forefront of innovating and evaluating different must acknowledge Einstein’s advice that not everything approaches to involvement. Patients and the public that counts can be counted, and not everything that can be represent an underexplored and untapped force for change, counted counts. which audit programmes can learn to harness.

Embed research to improve impact Conclusion

Implementation science aims to translate research evidence Audit and feedback are widely used, sometimes abused, into routine practice and policy but is also affected by and often under-realised in healthcare. More imaginative research waste. A cumulative meta-analysis of the Cochrane design and responses are overdue; these require evidence review of audit and feedback indicated that the effect size informed conversations between clinicians, patients, and stabilised in 2003 after 30 trials.21 By 2011, 47 more trials academic communities. It is time to fully leverage national of audit and feedback versus control were published that audits to accelerate data guided improvement and reduce did not substantially advance knowledge, many omitting unwarranted variations in healthcare. The status quo is no feedback features likely to enhance effectiveness. longer ethical. Implementation laboratories offer a means of enhancing Robbie Foy, professor of primary care, Leeds Institute of Health Sciences the impact of audit and feedback while also producing Mirek Skrypak, associate director for quality and development, Healthcare generalisable knowledge about how to optimise Quality Improvement Partnership, London effectiveness.22 A “radical incrementalist” approach entails [email protected] making serial, small changes, supported by tightly focused Sarah Alderson, clinical lecturer in primary care and Wellcome ISSF fellow, evaluations to cumulatively improve outcomes.23 It can help Leeds Institute of Health Sciences Noah Michael Ivers, clinician scientist, Women’s College Hospital, Toronto answer many questions about how best to organise and Bren McInerney, community volunteer, Gloucerstershire deliver feedback. Embedding sequential head-to-head trials Jill Stoddart, director of operations testing different feedback methods in an audit programme Jane Ingham, chief executive officer provides a robust empirical driver for change. Modifications Danny Keenan, medical director , Healthcare Quality Improvement Partnership, identified as more effective than the current standard London become the new standard; those that are not are discarded. Cite this as: BMJ 2020;368:m213 the bmj | 29 February 2020 329 LETTERS Selected from rapid responses on bmj.com

The STAMPEDE trial evaluated LETTER OF THE WEEK docetaxel in newly diagnosed metastatic Deprescribing prostate cancer. The patients in this trial were much younger and healthier than dementia drugs the real world population because of Twenty years ago Prescrire inclusion and exclusion criteria. The International said that the effects median age of trial participants, for of cholinesterase inhibitors were example, was 65, whereas the highest age “only moderate and of doubtful specific prostate cancer incidence rates clinical significance.” Nothing much in the population are in the 75-79 has changed, judging by the French age group. government’s decision to stop DEMENTIA DRUGS IN FRANCE The median survival of patients with reimbursing this group of drugs, as discussed by Reimburse for the and without docetaxel in the trial was Walsh and colleagues (Editorial, 18 January). sake of responders 81 months and 71 months, respectively—a The cholinesterase inhibitors and memantine net difference of 10 months. But docetaxel seem effective for some people, but their The reasoning behind France’s decision to chemotherapy is more toxic in the benefits seem negligible in the population. stop funding dementia drugs is incomplete real world population than in the trial The guardians of the public purse justify (Editorial, 18 January). population. withdrawing support on the grounds of Firstly, pharmacological treatments We should not be surprised that weak evidence, whereas advocates—some for dementia are not curative and clinicians are individualising treatment clinicians, some people with dementia, and cannot substitute non-pharmacological and not exposing all older men the manufacturers—describe a narratively rich, interventions. In Sweden, healthcare with comorbidities to highly toxic heartfelt range of behavioural and psychological personnel already work hard to develop chemotherapy by uncritically following the symptoms stabilised or reduced. and integrate non-pharmacological guidelines. Underlying this conflict are two problems with interventions; withdrawing reimbursement Santhanam Sundar, consultant oncologist, scientific knowledge and one problem for policy would not promote this further. Nottingham makers. The first knowledge problem is that Secondly, claiming inefficacy based on Cite this as: BMJ 2020;368:m519 basic science has been barking up the wrong a population mean could lead to hasty tree; a single cure for Alzheimer’s disease is negative conclusions and neglect the GENDER PAY GAP unlikely to be found, so we need to reconsider clinically relevant benefits for groups Be careful what you wish our approach to drug development for the within the population. Non-responders dementias. decrease the mean, disguising the true for in general practice The second problem is that a strategy benefits for responders. A doctor’s regular I am concerned about the inference of the to couple our deepening knowledge of clinical evaluation of effects and side article about the gender pay gap in general subcellular biology with information effects is a safeguard to minimise this practice (This Week, 18 January). I have technology has failed to transform human error. If the pharmacological treatment been a GP partner for 28 years. We have health. The problem for policy makers is that for dementia is effective for a patient with always encouraged doctors to join us in making judgments about cure versus care little to no side effects, then the patient, a partnership role, but very few want the investments can provoke hostile responses healthcare provider, relatives, and society additional responsibilities and perceived from a public that sees spending on cure would benefit greatly. risks. research as virtue economics. Finally, in Sweden full dose paracetamol We have been offering golden hellos of The French government hopes to shift costs more than a maintenance dose £10 000 and increased sessional profits to emphasis from a medical model of earlier of donepezil. So the economic benefits incoming doctors for the past three years. diagnosis and treatment to a more person of withdrawing reimbursement are They all, however, initially chose to be centred approach to dementia. This would fit negligible. salaried. with a social model of dementia, in which it is Tobias Damgaard, pharmacist, This article feels like another a disability that unfolds over the life course, Kalmar, Sweden encouragement for us older male GPs to shaped by exposures to harmful and protective Cite this as: BMJ 2020;368:m651 take our pensions and retire rather than factors. The world will be watching to see if the support the new GP workforce. It makes French government’s bold policy works. PROSTATE CANCER me feel naive for remaining full time for the Walsh and colleagues know what the vested Trial data meet the real world sake of patients and the practice. After five interests in the global dementia economy are with real patients years of struggle I will start part time work like and are wise to be pragmatic. in April, thus reducing the gender pay gap. Steve Iliffe, emeritus professor of primary care for The National Prostate Audit figures that Be careful what you wish for: it may have older people; Jill Manthorpe, professor of social work, show apparently lower uptake of docetaxel unintended consequences. London chemotherapy should be interpreted with Stephen Timothy Lytton, GP, Eastbourne Cite this as: BMJ 2020;368:m650 caution (This Week, 18 January). Cite this as: BMJ 2020;368:m515

330 29 February 2020 | the bmj GABAPENTINOIDS VALUE OF KINDNESS Restoring hope with alternative pain management Using our imaginations to move Hamilton need not be so downcast about alternatives for reducing gabapentinoid use beyond kindness to empathy (News Analysis, 18 January). Kairos Rehabilitation is a pain management clinic led by a GP with specialist interest. We provide gentle therapies, ease of access, continuity of care, Klaber and Bailey promote and the personal touch. Patients’ despair is not just related to their pain, but to being “relationships, unacknowledged as individuals and being held back from their personal development. connections, challenge, When hope is restored, inherent creativity engaged, and new friends made— and trust” and “gentle accompanied by a modest reduction in symptoms—it becomes obvious that increasing honesty in discussions” in analgesics will bring no benefit. medical practice (Editorial, A preliminary retrospective evaluation of the project’s first 30 patients shows that, 21 December-4 January). This despite initially higher than average levels of disability, there was statistically significant is not simply kindness, but improvement in mean EQ5D-3L health status, median average pain intensity, and an a necessary condition overall reduction in clinically significant depression. Repeat analgesic and psychotropic of good clinical care. prescriptions fell by 46%. Lack of attention to the experience David McGavin, GP ; Ingrid Hermansen, eurythmy therapist and trust coordinator , of another—whether colleague or Kairos Rehabilitation Trust , London patient—deprives them of a sense of Cite this as: BMJ 2020;368:m534 being noticed. “I went to the doctor, and he ACTIVISM FOR HEALTH didn’t even look at me,” a senior paediatrician said of her consultation If we’re serious about change, think about surgical waste with a man more engaged with his This year has seen an unprecedented focus on our climate change emergency (Fight the computer than his patient. Perhaps Power, 21 December-4 January). being patients ourselves is the best An estimated 57 000 tonsillectomies are carried out each year in the UK; we performed way to discover the fundamental need 1067 in our hospital last year. The surgical consumables from a tonsillectomy or for recognition. adenotonsillectomy are divided into orange or black bags. We weighed the orange bag Doctors are not simply diagnostic waste, which is directly attributable to the ear, nose, and throat surgeon, finding it ranged algorithms: we are humans who can from 1.07 kg to 2.30 kg (mean 1.86 kg) across 10 procedures. If this was consistent identify with patients’ experience. This throughout the year, it would lead to 1984 kg of incinerated waste from adenotonsillar effort of imagination is at the core of procedures in our hospital alone. Extrapolated to the whole country, this would be all care. Our duty of care is not only a 106 020 kg (106 tonnes) of incinerated waste a year. moral obligation but a skill that has We hope this prompts surgeons and theatre staff to think, act, and advocate for the to be learnt in training and relearnt change that everyone wishes to see. throughout our careers. Amit S Gill, foundation year 1 doctor ; Thomas Hampton, specialty trainee year 6 ear, nose, and throat ; Sebastian Kraemer, honorary consultant , Ravi Sharma, ear, nose, and throat consultant , Liverpool London Cite this as: BMJ 2020;368:m225 Cite this as: BMJ 2020;368:m444 “Hello, my name is” helps bring RELIGION IN MEDICINE kindness to busy A&Es Faith is no reason to d ismiss I agree that we should all try to be patient autonomy kinder. How one achieves that in a busy emergency department, however, Sokol’s article on religion in medicine remains a challenge. Staff are usually is unnecessarily provocative (Daniel short on time and resources, and often Sokol, 18 January). He notes, as if it is they haven’t worked together before— a surprising revelation, that religious the perfect recipe for uncivil behaviour. ethicists of the same religion are not Introduction of the “Hello, my name entirely in agreement. Millennia old is” badges has helped in a small but sacred texts do not hold set-in-stone important way, allowing you to feel rules on 21st century palliative care. that your colleagues are human beings The statement that “it is naive to believe religion and medicine can always coexist in rather than faceless healthcare workers harmony” veers towards a paternalistic dismissal of patient autonomy. All decisions who you pass in a never ending haze of should be patient centred—personal choice would not be disputed, so why is a decision shifts. based on religious belief segregated in this manner? Sokol is correct that no doctor Perhaps efforts like these are the way should feel forced to prescribe an inappropriate treatment option, but this is the same for forward. all patients, regardless of the reasons for which they are requesting them. Samuel E Mercer, geriatrics specialist registrar , Christopher G S Gilmartin, medical student , Cambridge Leeds Cite this as: BMJ 2020;368:m526 Cite this as: BMJ 2020;368:m228 the bmj | 29 February 2020 331

Longer versions are on bmj.com. Submit obituaries with a contact telephone number to [email protected]

OBITUARIES Arthur Graham Blyth John George Coxon Lenox Jardine Millar Former general Consultant urologist General practitioner practitioner Andover, Halifax, west Yorkshire Newgate Medical Group, Hampshire (b 1926; (b 1920; q St Thomas’ Worksop (b 1948, q Barts q Guy’s Hospital, London, 1943; FRCS), died 1971; DRCOG), died from 1950; DObst RCOG, peacefully from gall bladder cancer on MRCGP), died from old pneumonia and acute 7 October 2019 age on 30 December 2019 kidney disease on 2 July After working in Arthur Graham 2019 Nottingham City Blyth was house surgeon in obstetrics John George Coxon qualified during the Hospital, Lenox Jardine Millar gained his at Winchester Hospital before joining a war, and so his graduation ceremony was first experience of general practice in practice in Andover. He was also medical postponed until 1992. He completed his Australia, with his wife, Jo, also a doctor. officer to Andover War Memorial Hospital training in Sheffield, where he met Ann, and He started in practice in Worksop in 1974, and to four private schools. He was a careful they married in 1954. In that year, he was became a senior partner in 1990, and diagnostician, willing home visitor, and much appointed as a consultant general surgeon at retired in 2010. He was a popular, much loved family doctor with a special interest in Dewsbury and Halifax hospitals. In 1965 he loved GP, with an ability to connect with mental health problems. He was an examiner moved to work entirely in Halifax, and later people. He became a GP trainer in 1977. in first aid for the Red Cross and a member specialised in urology. He was active in the On retirement he became captain of of the Lord Chancellor’s committee for the British Association of Urological Surgeons Lindrick Golf Club, walked the south west selection of magistrates. A keen naturalist, and the local BMA, and served on the hospital coastal path, and learnt Italian. Diagnosed musician, and theatre lover, he was also a management board. He was involved with the with gall bladder cancer in 2016, he sidesman and occasional organist at his local local branch of the Royal Medical Benevolent responded well to treatment and lived a church. From 1957 to 1999 he was on the Fund. Ann and he were part of the organisation full and active life, walking in Scotland Andover Music Club committee. Predeceased to help set up the first local hospice. until six weeks before his death. He is by his wife in 2013, he leaves five children, 11 Predeceased by Ann in 2010, John leaves a sadly missed by Jo, three children, two grandchildren, and five great grandchildren. son, two daughters, and four grandchildren. grandchildren, and friends. Fiona Blyth, on behalf of the Blyth family Charles Coxon Josephine Millar Cite this as: BMJ 2020;368:m293 Cite this as: BMJ 2020;368:m300 Cite this as: BMJ 2020;368:m296 Michael Hamilton Lisa Jane Newton Abbaraju Mohan Rao Consultant physician Consultant Consultant anaesthetist (b 1923; q St Mary’s haematologist (b 1953; q Gandhi Hospital Medical School, Bradford Royal Infirmary Medical College/ 1945; OBE, MD, FRCP), (b 1967; q Sheffield; Osmania University, died from heart failure FRCP, FRCPath), died from Hyderabad, India, 1978), on17 November 2019 metastatic ovarian cancer died from idiopathic Michael Hamilton was on 17 November 2019 pulmonary fibrosis on an early pioneer of the Having trained in 6 January 2020 clinical research into effective management Sheffield and Leeds, Lisa Jane Newton was Abbaraju Mohan Rao moved to the UK of essential hypertension, making the appointed to a consultant post at Bradford in 1981. Having chosen to specialise in case for the efficacy of trials based within a Royal Infirmary. She was also the service lead anaesthetics, he worked initially in King’s community rather than a teaching hospital for the busy lymphoma service and involved Lynn and Stockton on Tees, before training setting. A Nuffield research fellowship in several national trials. Lisa is remembered in the Mersey region and Belfast. He became enabled him to spend 1954 at the Alfred by staff and patients for her great fashion consultant in anaesthesia at Arrowe Park Hospital in Melbourne, followed by a year sense, her happy cheery manner, and her Hospital in 1997. He retired in 2015 on at the University of Otago, New Zealand, striking looks. Diagnosed with ovarian cancer health grounds. A keen cricketer in his in 1955. In 1956 Michael was appointed seven years ago she underwent radical younger days, he trained as an umpire and as physician to the Chelmsford Group of surgery and intense chemotherapy and regularly officiated in local league games in Hospitals in Essex. He established the radiotherapy. She gave up work in 2016 and north Wales. A connoisseur of Hindustani academic medical unit in Chelmsford underwent trial therapy at the Royal Marsden classical music, he learnt to play the sitar in the 1960s. Michael declined several Hospital, but her cancer returned. She still and played it regularly. He indulged his prestigious academic appointments in managed to organise a wedding to her long passion for photography on his extensive the UK and Australia because of his love term partner, Mark, and ensured her funeral travels around the world. A sincere, kind, and of clinical practice and a dedication to his arrangements were taken care of so her generous man, he had a wonderful family and patients in Essex. He leaves two children; six family would not have to worry. She leaves a wide circle of friends. He leaves his wife, grandchildren; and eight great grandchildren. Mark, her mother, and her sister. Saritha, and two children. Peter Kopelman Majid Kazmi Murti Gollapudi Cite this as: BMJ 2020;368:m294 Cite this as: BMJ 2020;368:m299 Cite this as: BMJ 2020;368:m297 the bmj | 29 February 2020 343

OBITUARIES Brian Frost-Smith Physician, author, and horologist

Brian Frost-Smith (b 1926; q Welsh National A Country Doctor personally for tests. The treatment of School of Medicine, University of Wales, In his entertaining memoir, A Country tuberculosis with streptomycin had just Cardiff, 1950), died from metastatic colon Doctor, Frost-Smith describes how different started, but powerful antibiotics had not yet and prostate cancer on 29 October 2019 it was working as a GP 60 years ago. “I arrived. When they did, in the mid-1950s, completed six months in hospital before huge numbers of sanatoriums closed down. Brian Frost-Smith studied at the Welsh entering a locum practice on my own. I National School of Medicine in Cardiff, where had complete control and could prescribe Clocks he met Marjorie Ion, a fellow medical student. anything, and the patient would receive Frost-Smith’s close friend Kenneth Boyd, They both qualified and were married in it free. One pharmacist said at the time: professor emeritus of medical ethics at 1950. They remained together until her death ‘Anything you write on that prescription the University of Edinburgh, recalled, “My in November 2016. After their honeymoon, form, I must deliver.’ Most drugs were most abiding memories of Brian are of his they were both appointed to junior house alkaloids—organic, plant based compounds conversation and his clocks. No subject jobs in Walton Hospital, Liverpool. Soon containing nitrogen—and there were was off the table: sex, health and medicine, afterwards, Marjorie discovered that she was very few synthetics. I had a patient with religion, philosophy, politics. expecting their first child. Frost-Smith joined pernicious anaemia and I treated her with Brian and Marjorie were founders of the Royal Air Force to serve two years as a anahaemin, an injectable extract of raw a discussion group, mostly of members national service medical officer and received liver. Only a few years before, the treatment of their village church, St Mary Thrimby, a home posting to Cheshire to enable him to was raw liver, which the patient had to eat! called the Thrimby Thinkers, and which remain near Marjorie. There were no steroids, few antibiotics, and met regularly to anatomise life, the His daughter, Rachel Buchanan (also a the only diuretic was a mercury compound universe, and everything with great doctor), said, “After finishing his national called mersalyl. We used arsenic to treat enthusiasm, and total disregard of religious service, my father worked as a GP assistant syphilis and calomel (another mercury or political correctness. in Coningsby, Lincolnshire, and then as compound) and phenobarbital (an “Visits were sometimes interrupted by a a partner in Huyton, Liverpool. However, antiepileptic drug) for a good night’s sleep patient ringing the door bell out of hours, he had always wanted to work in a rural on hospital admission. Chest x rays were but more regularly by the chiming of the practice, so eventually he moved to work in available in mobile units for the diagnosis of numerous clocks of all kinds scattered Shap, living in Little Strickland. He enjoyed tuberculosis, but nothing else.” around the house. Brian was not only life in Cumbria and became involved with As a GP, Frost-Smith could not send blood a skilled and compassionate general community and village life and with the samples for biochemical or haematology practitioner, but also a technically skilled local church.” analysis—patients had to attend hospital horologist, who was generous with his knowledge and expertise. Our parents’ grandfather clock was regularly cleaned, maintained, and repaired by him, and so were those of many other neighbours. “At his funeral, his son, Paul, told the story of what happened when Brian, sometime after retiring, agreed to do a short locum in the new health centre of his former practice. As he was getting ready in the In his entertaining consulting room, he heard the waiting room open and close many times. Brian became memoir Frost-Smith anxious: it had been a few years, was his describes how different medical knowledge sufficiently up to date? it was working as a At last he opened the consulting room door, GP 60 years ago only to discover that it was full of his former patients—all sitting there awaiting his advice, with their clocks on their knees.” Brian Frost-Smith died at Winters Park Care Home in Penrith, where he chose to spend the last two weeks of his life. He leaves two daughters and a son. Rebecca Wallersteiner , London [email protected] Cite this as: BMJ 2020;368:m289

344 29 February 2020 | the bmj