Making evidence based medicine work for individual patients
BMJ 2016; 353 doi: https://doi.org/10.1136/bmj.i2452 (Published 16 May 2016) Cite this as: BMJ 2016;353:i2452
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
This article is timely when technocratic models of health care, decision making, feminism (in the context of child birth), evidence and patient’s human rights are being discussed in the context of healthcare in more innovative ways than at any other point in time in the history.
The judgement of Montgomery v Lanarkshire Health Board [2015] UKSC 11 was a reflection of the legal and societal change of attitude in this area, as well as developments in the European Court of Human Rights in Glass v United Kingdom [2004] 39 EHRR 15 and Tysiac v Poland [2007] 45 EHRR 42.
An integral part of patient autonomy is the ability to make decisions over bodily integrity. But this is dependent on the patient being given the information in a manner that they understand. This is one of the issues, among others, commented on in Montgomery where the obstetrician failed to advise Mrs Montgomery of the risk of shoulder dystocia and to discuss with her alternatives.
Having cited NICE (2011) guidelines on offering “evidence-based information and support to enable them to make informed decisions about their care and treatment”, Lady Hale in Montgomery goes on to say “Gone are the days when it was thought that, on becoming pregnant, a woman lost, not only her capacity, but also her right to act as a genuinely autonomous human being” highlighting the change of attitude of the patient doctor relationship. Montgomery also highlights “social and legal developments... point away from a model of... medical paternalism” as well as “point away from a model based upon a view of the patient as being entirely dependent on information provided by the doctor” and to treat “them so far as possible as adults who are capable of understanding that medical treatment is uncertain of success and may involve risks, accepting responsibility for the taking of risks affecting their own lives, and living with the consequences of their choices.” Patients are entitled to decide whether or not to incur a medical risk and exercising of it “does not depend exclusively on medical considerations...”. In addition the “Responsibility for determining the nature and extent of a person's rights rests with the courts, not with the medical professions.”
While Stone’s references to the Mental Capacity Act 2005 are noted, the act itself came into force on 1 October 2007. However as Nadine Montgomery gave birth to a child with disability on 1 October 1999, Act would unlikely to have had an effect. However the issue of mental capacity was addressed in earlier cases including Re C (Adult: Refusal of Medical Treatment) [1994] 1 W.L.R. 290 and Re M.B. (Medical Treatment) [1997] 8 Med. L.R. 217 were available to the court. Relevant here is also the Human Rights Act 1998 which requires compliance with the European Convention on Human Rights.
Montgomery is good law in the UK and is likely to remain so. However what is relevant is that Mental Capacity Act 2005 is not used to circumvent the patient’s consent in order to force the doctor’s decisions upon them. This is because a patent would consider other factors such as quality of life in addition to medical expert opinion when deciding on a course of action. What is however interesting is that for some time the medical establishment has been willing to adopt a holistic approach to palliative care as compared to other sorts of medical care.
While I agree with Stone’s recent response on obtaining consent, providing relevant information in an understandable manner can be burdensome and time consuming. However society has come far enough to regard the right to bodily integrity and obtaining consent as important. Society is unlikely to part with these concepts as with parting with the concept of democracy, in favour of despotism or even technocratic models. Should time for consultation not be adequate, then is it not the consultation times that need to be increased in order to meet the standard of care required and reduce the burden on care providers?
Competing interests: No competing interests
Gratified as I am that Amali Lokugamage (08 June 2016) agrees with my earlier response, I would point out that while someone who works in Obstetrics and Gynaecology might understandably quote the Montgomery ruling, all of the law around 'consent' and 'pseudo consent' ('best interests' decision-making, if the patient lacks mental capacity [and when the patient had not already made and expressed the decision]) for adult patients is within the Mental Capacity Act. And section 3(1) of the MCA, which covers consultations with mentally-capable patients, clearly describes 'Informed Consent':
3(1) For the purposes of section 2, a person is unable to make a decision for himself if he is unable—
(a) to understand the information relevant to the decision,
(b) to retain that information,
(c) to use or weigh that information as part of the process of making the decision, or
(d) to communicate his decision (whether by talking, using sign language or any other means).
So L Sam Lewis (03 June 2016) is incorrect when he writes:
'We are in imminent danger of agreeing that evidence-based recommendations should be embedded into daily practice without becoming burdensome, whilst placing patients at the helm of decisions!'
We are already there, with the possible exception of the 'without becoming burdensome' part – England and Wales have been in that situation ever since the MCA embedded patient autonomy as the fundamental aspect of the consultation process, almost a decade ago, unless it is asserted that evidence-based recommendations should not be used. The Montgomery ruling, was for a case in Scotland where the MCA is not the law, and which dated from before the MCA was enacted: so the Montgomery judges could not base their ruling on the MCA .
I myself, would be inclined to suggest that EBM as the guide for recommended treatment, combined with patient autonomy for the decision-making and its requirement of obtaining genuine (not ‘contrived’) informed consent from patients, is sometimes ‘burdensome’: especially if there is also a ‘target orientated culture’ imposed on the professionals.
Competing interests: No competing interests
This type of analysis is most welcome and I agree with Stone’s rapid response, that ultimately the patient gets to make the final decision, and that those decisions take precedence over everything. It is not just the inclusion of patient's values and beliefs in shared decision making within the construct of evidence based medicine. So the term shared decision making is not correct when a human rights caveat is applied to clinical guidelines. This applies to avoiding medical interventions as well as requesting medical interventions because of perceived risks as seen in the Montgomery v Lanarkshire Supreme Court Ruling March [2015] in the UK (1). This shows respect for bodily integrity and fulfils the FREDA criteria for human rights in healthcare (fairness, respect, equality, dignity and autonomy). Speaking from a human rights in childbirth perspective, overarching analyses of the quality of evidence behind RCOG, ACOG, SOGC guidelines indicate that a minority of recommendations have high quality evidence behind them or vice versa majority of recommendations have lower quality evidence behind them (2-4). So where there is scientific uncertainty in the recommendation this needs to be highlighted to the patient, especially if it is thought that the patient would attach significance to it in light of the Montgomery v Lanarkshire case. If a clinician is truly patient centred rather than institution centred or guidelines centred this should present no problem, especially with good respectful rapport with the patient.
References
(1) United Kingdom Supreme Court. Montgomery v Lanarkshire Health Board. 2015. UKSC 11, United Kingdom.
(2) Ghui R, Bansal JK, McLaughlin C, Kotaska A, Lokugamage A. An evaluation of the guidelines of the Society of Obstetricians and Gynaecologists of Canada. J Obstet Gynaecol 2016; 36(5):658-662.
(3) Prusova K, Churcher L, Tyler A, Lokugamage AU. Royal College of Obstetricians and Gynaecologists guidelines: how evidence-based are they? J Obstet Gynaecol 2014; 34(8):706-711.
(4) Wright JD, Pawar N, Gonzalez JS, Lewin SN, Burke WM, Simpson LL et al. Scientific evidence underlying the american college of obstetricians and gynecologists' practice bulletins. Obstet Gynecol 2011; 118(3):505-512.
Competing interests: On the Advisory Board of Human Rights in Childbirth (NGO) and on the Board of Directors of the International MotherBaby Childbirth Organisation (UN recognised NGO). Both are charities.
My husband is in hospital and is receiving excellent care.
"I have some new questions that I have GOT TO ask you" said the delightful nurse admitting him. "What matters to you?"
Great I thought - person centred care.
Reading from her clipboard she continued....."No care about me without me" she states with an upward inflection at the end to turn it into a question.
"What does that mean?" asked my husband.
"I don't know" she replied, but I have got to ask it.
Margaret McCartney and colleagues, together with Fiona Godlee in her linked editorial argue that we need better "tools" to encourage questions like "what are the options?". But when will the NHS learn that well-intentioned, mechanistic, top-down diktats (tools, models, aids) do not result in meaningful caring person-centred discourse in a neat linear way!
Competing interests: No competing interests
We are in imminent danger of agreeing that evidence-based recommendations should be embedded into daily practice without becoming burdensome, whilst placing patients at the helm of decisions!
Stephen Chapman cited the anticoagulation choices in atrial fibrillation. But the tool as implemented at http://www.anticoagulation-dst.co.uk/ requires data-entry box-ticking for some 5 minutes online, and if contraindications, interactions, or licensing requirements dictate, the process aborts. The NICE AF decision supporting tool sets out the logic, but runs to 36 pages! NICE’s statin decision support fares little better at 24 pages. Multiply this unwieldy methodology a dozen times for the average consulting patient, and it becomes unbearable. Remember the lessons of QoF, whereby groaning GPs would quickly sweep the boxed targets aside with a peremptory “see the nurse”.
Nano-nagging might have us throw this EBM baby out with the burdensome bathwater!
There is a simple solution – to embed EBM in the patient computerised record.
GP software suppliers have developed a number of tools which embedded QoF requirements into a ‘prompts box’, after automatically processing the patient data record. Building on this, I merged many of the decisions (eg: statin or not, ACEI or not ) into single lines, and added 99 other evidence-based prompts such as drug interactions (eg. Amlodipine vs Statin , or Metformin vs low eGFR), in a programmable user-defined audit software tool.
It is then a short step to integrate these computed recommendations into the generic Patient Care Summary, which contains all significant diagnoses, problems, allergies, medications, contraindications, latest test results, referrals and patient preferences.
A patient should be able to access their care summary online, already completed. If not, it can be printed off, with a “thankyou for improving your care” compliment slip, ready for the patient to expand, correct and annotate whilst sitting in the GP waiting-room, out-of-hours centre, or A&E. For example, if CVD risk is above the NICE threshold, statin would be recommended, unless the record showed it had been prescribed, was contraindicated, or had been declined by the patient If AF was recorded without contraindications or rejections, the entire algorithm cited by Stephen Chapman would be short-circuited to a recommendation to anticoagulate, together with a simple graphic showing size of benefits and bleeding risks. If Aspirin had been prescribed without an indication, an alert would display. Cancer risk computations could be incorporated, triggering investigation and referral prompts. National priorities in post-hoc prescription-switching could similarly be incorporated, releasing a phalanx of pharmacist advisors to better employment as prescribing associate professionals to handle all the resulting consultations.
I managed to incorporate QoF requirements and 99 other EBM recommendations, blood result alerts, and drug interaction prompts. Only relevant and undecided prompts would display within a small on-screen ‘floating box’, computed in the time taken for the patient to get from the waiting room to the consulting room. Each prompt can have a yes/no tickbox, which launched the prescribing or referral process, or recorded an appropriate patient refusal such as ‘statin declined’. Further development of such a single-screen prompt tool, generating practice-wide audit, should be a national and CCG priority, supplanting QoF. Clinical users should urgently contribute to the specification of risk and display preferences, before they are overcome by reams of 36-page smiley-faced decision aids!
Competing interests: Proof of concept and Intellectual property rights reserved.
Margaret McCartney and colleagues make very pertinent and important points about the tension between population based guidelines and the needs of the individual and an evidence based recommendation and patient preference. This tension between evidence and patient preference has existed since the beginning of modern medicine. Potential technical solutions to resolve such dissonance may lie in the appropriate use of assistive technology such as computerised decision support systems (CDSS). CDSS – easily accessed via a GP’s computer – can be used to re-purpose a static, population based guideline into an interactive and individualised decision aid which guides the prescriber to an evidence based recommendation .
We have recently used a joint working agreement between NICE, Boehringer Ingelheim and ourselves at Keele University to design one such system to aid the implementation of NICE CG 180 for the management of patients with atrial fibrillation. Importantly, and in line with the thinking in Margaret McCartney’s essay, the tool’s prescribing recommendation then prompts the use of an associated patient decision aid which re-fashions the lengthy paper based tool from the original guideline into easy to use and dynamic graphics using Cates plots.
The tool can be downloaded free at http://www.anticoagulation-dst.co.uk/. A formal evaluation of the tool’s usefulness to prescribers and patients is ongoing, but the early anecdotal evidence is very positive. Prescribers value decision support and the opportunity for shared decision-making, but we can’t underestimate the time constraints placed on practitioners and the need for it to be implemented seamlessly into their routine/normal workflow. This suggests the need for relevant tools (such as our own) to be integrated into clinical systems.. This is not without challenges, but negotiations are underway and we are hopeful of a positive outcome.
Competing interests: The AF CDS tool described in the letter was created under a joint working agreement with NICE, Keele University and Boehringer Ingelheim. The BU contribution was an unrestricted educational grant.
Margaret McCartney et al write clearly and eloquently on this recurring problem we front line clinicians face daily.
As colleagues have alluded to earlier, but it will not go amiss to say it again, all this means we will need more time with patients.
How do we inculcate all this best practice into the already very constrained 10 minute consultation time that is primary care??
Competing interests: No competing interests
In their box 1, McCartney et al list the problems they perceive when applying population based evidence to individuals
Because, quite reasonably, randomised trials often exclude patients with comorbidities, and guidelines describe the evidence for single conditions; real patients often have several comorbidities. But a review of best evidence, as for example set out in the Quality and Outcomes Framework (QOF) the interaction of co-morbidity more often than not implied greater interventional benefits for the individual. For example, the CVD risk is greater after Stroke, myocardial infarction or diabetes, and consequently the trial benefits of statins are greater. In hypertension management the same CVD risk interaction is true, and lowering the thresholds at which BP treatment is triggered, as well as BP treatment targets. It is rare that evidence can be cited to show that comorbidity means the individual will gain less than when the single disease obtains. If the individual refused the treatment offer, however well evidenced, then the ‘informed dissent’ exception could be made. Thus McCartney et al’s concern for the individual patients different values and preferences was incorporated. . If the doctor judged that benefit would not accrue ( for example, comorbidities dominated, or the patients expected lifespan was limited) then a ‘patient unsuitable’ exception from the QOF recommendation was permitted. Of course, a few days of life gained might be very highly valued by a patient knowing himself to be terminally ill, and ‘shared decision-making’ might prove difficult, since guidance may “not cover aspects of care important to patients”. Valuable research is awaited as to who decided these ‘exceptional’ matters under QOF. I am unaware of sound evidence that comorbidity should invalidate the single-disease recommendation, in all but a few cases.
I would prefer the embedding of QOF-type recommendations within the GP software systems, rather than the multiple printed shared decision tools to solve the issues listed in Box 2 of this paper. The single ‘potential gains’ screen should automatically display expected benefits, targets and treatments missed, from the fully known data record. A graphic display could show where the patient sat in population range, the size and type of harms prevented, and how much life did they stand to gain. On-screen adjustment of BP, cholesterol, HbA1c could show by how much, how quickly and with what treatments, and it should all print out by hitting ‘Print Screen’.
Competing interests: No competing interests
My wife and I read this article with great interest and agree with much of what Margaret McCartney and colleagues say. We thought it might be of interest to readers to know the impact that rigidly applied guidelines can have on a patient - that patient is my wife.
Please take the time to read her account of what happened after she had an abnormal mammogram. This was published in the BMJ in October 2015. Here is the link:
http://www.bmj.com/content/351/bmj.h4630/rr-7
As a result of her challenging the local system, she has succeeded in changing the way guidelines are applied in Leicester. Much more weight is now given to clinician and patient decision-making but, in my view, this only came about because the status quo was challenged by a brave and articulate woman.
Competing interests: No competing interests
Feasibility is also relevant
This article expressed a very refreshing alternative to the efficiency of the Holborn escalator model proposed by David Oliver recently (1) and promoting a “public collectivist system that prioritises the needs of a group over the wants of the individual”. Both views are polarised and reflect the to-and-fro arguments about the delivery of medical care over the past 40 years. Recently , guidelines have become protocols even when the evidence is relatively poor and insufficient weight is given to the requirements of the individual in the interests of efficiency.. Some of the perceived needs of the group (for example routine “health checks”) may even be seen as having a political rather than medical basis. With many of our cancer referral guidelines , some patients - particularly the elderly or frail - may not want to be fed through a rapid access referral system based purely on a particular symptom - and an ever decreasing statistical possibility of a significant finding - and not on the person . Besides , the referral often involves a tube being inserted into some orifice with attendant discomfort and risk.. I speak from my own recent experiences with a“red flag” symptom but , knowing the odds and my own opportunity costs etc.. , decided to procrastinate. Eventually , after an ultra-sound , I had an evening telephone call from the on duty GP (who had never met me) telling me I needed a rapid access referral . This was commendably efficient , but it might have been a bit of a shock if I had not known what to expect.
It was this pressure to process patients that was a factor in my retirement . When working subsequently doing waiting list initiative endoscopy clinics I found many of the patients opted out of having the procedure when the pros and cons were discussed as part of informed consent and taking into consideration factors such as advanced age and likely benefit beyond a diagnostic label.. Reverting to the escalator model, some of us who choose the stairs , practically or figuratively , may just want an alternative approach , for whatever reason and even if it takes a bit longer. Jumping the queue may well not be the main motive. We must not lose sight of individual needs in the pursuit of overall efficiency but just how far we can or should cater for all individual demands will often remain a tricky and difficult area. But when did anyone suggest that good clinical care will always be straightforward ?
1. BMJ 2016,353: i12127
Competing interests: No competing interests