A NICE example? Variation in provision of bariatric surgery in EnglandBMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f2453 (Published 01 May 2013) Cite this as: BMJ 2013;346:f2453
- Amanda Owen-Smith, research fellow1,
- Ruth Kipping, research fellow and consultant in public health12,
- Jenny Donovan, professor of social medicine1,
- Christine Hine, consultant in public health3,
- Christina Maslen, clinical effectiveness research lead4,
- Joanna Coast, professor of health economics5
- 1School of Social and Community Medicine, University of Bristol, Bristol BS8 2PS, UK
- 2NHS North Somerset, Public Health, Clevedon, UK
- 3NHS Bristol and NHS South Gloucestershire Public Health Directorate, Emerson’s Green, Gloucestershire, UK
- 4NHS Bristol, Bristol, UK
- 5Health Economics Unit, School of Health and Population Sciences, University of Birmingham, Birmingham, UK
- Correspondence to: A Owen-Smith
- Accepted 12 March 2013
The National Institute of Health and Care Excellence (NICE) has been the most ambitious attempt to set healthcare priorities more systematically and explicitly since the inception of the National Health Service in 1948.1 NICE was established in 1999 with the dual aims of reducing regional inequality in access to treatments and making decision making more systematic through explicit consideration of the clinical and cost effectiveness of (mostly) new treatments. Despite controversies related to the time taken to issue recommendations and the defensibility of its cost effectiveness thresholds, the NICE decision making process is respected within the UK and internationally.2 3
Nevertheless, NICE guidance has met with a mixed reaction from NHS commissioners, who have to implement the recommendations, some of which are mandatory within three months. Frustrations have arisen over the lack of consideration given to the affordability of implementation. (NICE recommendations have to be funded from existing allocations, implying disinvestment elsewhere.) Drawing on the example of bariatric surgery for morbidly obese patients, we review the implementation of NICE guidance and regional specialist commissioning guidelines over the past 10 years. We examine how they have affected provision of care and consider what challenges remain for the new clinical commissioning groups and NHS England as they take on their full commissioning roles.
NICE issued mandatory guidance on bariatric surgery in 2002.4 The technology appraisal instructed primary care trusts (PCTs) to provide surgery for morbidly obese patients (body mass index ≥40 or ≥35 in the presence of specified comorbidities) when other interventions for weight loss had failed. Although PCTs were given some flexibility on the three month implementation schedule because of shortages in capacity, the message from NICE was clear: surgery was both effective and cost effective and should be made available for patients meeting the criteria when both they and their clinicians think it is appropriate. In 2006, the appraisal was subsumed within a broader clinical guideline on obesity,5 which effectively removed the mandate on PCTs to provide access to surgery, although it was still recommended as the most cost effective treatment; the new guideline added that surgery should be considered as first line treatment for people with a body mass index >50 (box 1).
Box 1: NICE criteria for bariatric surgery5
Bariatric surgery is recommended if all of the following criteria are fulfilled:
Body mass index ≥40 or 35-39 if other serious disease (eg, type 2 diabetes or high blood pressure) could be improved by weight loss
All appropriate non-surgical measures have been tried and failed to achieve or maintain adequate clinically beneficial weight loss for at least 6 months;
Patient has been receiving or will receive intensive management in a specialist obesity service
Patient is generally fit for anaesthesia and surgery
Patient commits to long term follow-up.
Bariatric surgery is also recommended as a first line option (instead of lifestyle interventions or drug treatment) for adults with body mass index >50 in whom surgical intervention is considered appropriate
Access to surgery
Since NICE published the guideline, evidence has increased that surgery has a positive effect on weight loss and the resolution of obesity related comorbidities (box 2).6 7 Indeed, it is the only treatment currently available for morbid obesity that has a reasonable chance of sustained weight loss.8 9 The number of NHS operations has soared, with 8087 weight loss operations (including revisions) undertaken in 2010-11, a rise of over 300% since 2006-07.10 However, NHS provision is still dwarfed by the rising tide of morbid obesity (body mass index >40) in England, which was estimated at 2.7% in 2010, equivalent to 1.35 million people.11 Of course, not everyone who is morbidly obese would choose, or be suitable for, surgery because of the lifestyle restrictions it imposes or the severity of comorbidities.5 Additionally, the first national audit of bariatric surgery suggests that about 31% of bariatric operations are currently done in the private sector, 86% of which are on patients with a body mass index over 40, implying eligibility for NHS treatment.12 Nevertheless, the fact that less than 0.6% of those who are potentially eligible for bariatric surgery receive treatment on the NHS, coupled with media reports and qualitative research documenting difficulties in accessing care, indicates a shortfall in provision or access.13 14
Box 2: Summary of evidence on effectiveness and cost effectiveness of bariatric surgery
All six randomised controlled trials and cohort studies comparing surgical and non-surgical interventions for obesity showed that participants who had had surgery lost more weight and were able to sustain weight loss for longer6 8
In the only long term cohort study (Swedish Obese Subjects study) weight loss and reduced comorbidity reduction were still apparent 10 years after surgery9
Compared with conservative management, economic evaluations show an incremental cost effective ratio of £2000-£4000 per quality adjusted life year gained for people with a body mass index above 40. (However, estimates are uncertain because of the limited number of long term studies)6
Data for 2010-11 show substantial regional variations remain in access to NHS bariatric surgery, with rates ranging from 6/100 000 population in the North West region to 32/100 000 in the East Midlands (table 1⇓). There are no data available on the prevalence of morbid obesity by region, and although data from the Health Survey for England suggest some regional differences in the prevalence of simple obesity (body mass index >30; 22-26.5%) this is not sufficient to explain the variation observed.10 In addition, inequalities in provision have widened over the past five years, suggesting differences in implementation of NICE guidance.15
Role of regional commissioning policies
As some of the differences in provision might be explained by regional variations in specialist commissioning policies, we systematically compared these policies with the NICE criteria (table 2⇓). When possible, we accessed up to date documents online. If none could be located we contacted relevant bodies until they provided the policies or confirmed that none were available. Of the eight sets of specialist commissioning policies we were able to access, none provided for all of the patient groups that NICE recommends are suitable for bariatric surgery. Only half of regions recommend treatment for those with a body mass index of 35-39 in the presence of comorbidity, and only one (Yorkshire and Humber) complies with the NICE recommendation that bariatric surgery should be first line treatment for those with a body mass index >50.
However, variations in regional policies do not necessarily reflect rates of provision of bariatric surgery. Indeed, the highest providing region, the East Midlands, has some of the strictest criteria for access to surgery, and the North West, which is the lowest provider, has among the most generous policies. Although five of the 10 regions are ranked roughly as might be expected given variations in access criteria, levels of NHS provision are so low throughout England that it seems unlikely that even people fulfilling the strictest criteria would easily be able to access NHS treatment. Our analysis suggests that regional policies for accessing bariatric surgery are not followed consistently, leading to differences in local provision. Other factors that might contribute to the variation include regional differences in prevalence of morbid obesity, the capacity of surgical units, differential local funding pressures and priority setting processes, and differences in the attitudes of commissioners, clinicians, and patients to the acceptability and effectiveness of a surgical solution to the problem of morbid obesity.
Variation at the primary care trust level
The NHS Atlas of Variation data also show considerable variation in provision by individual PCTs within regions, with a 93-fold differential observed between the highest and lowest providing trusts.10 These findings are consistent with patient information websites that list the differing access criteria that people have encountered,16 and with qualitative research that has found that some PCTs are adding additional access criteria to help manage the gap between supply and demand.14 For example, PCTs in one area provided surgery only for those with diabetes, even though it was openly acknowledged that this left other needy patients without access to care. One consultant in the area said, “I’ve got a desk drawer full of patients who aren’t eligible [because they aren’t diabetic] even though they’re severely depressed to the extent that they’ve tried to commit suicide . . . or they’ve got sleep apnoea, or other significant comorbidities.”17
Our analysis raises several questions for patients and providers seeking to improve access to bariatric surgery, and for the new clinical commissioning groups and NHS England. The data show considerable regional variation in the provision of morbid obesity surgery, and local commissioning guidelines suggest that individual access to treatment remains subject to a “postcode lottery.” Regional variation in the availability of care is widely known to be disliked by patients and clinicians, and was one of the problems that NICE guidance was designed to overcome.1 In addition, overall levels of provision clearly fall far short of the number of operations that would be needed if NICE access criteria were to be fully implemented, implying that some people who could be effectively and cost effectively treated are left without access to care.
In this era of austerity, it is unlikely that the investment needed to fund surgery for morbid obesity to the level recommended by NICE will be forthcoming. It will thus fall to more local decision makers to prioritise patients for treatment and decide what (if any) investment should be awarded to morbid obesity surgery amid the many other calls on constrained budgets. For at least the next two years, clinical commissioning groups can be expected to be supported by NHS England, which will issue guidelines on access to specialised treatments and promote regional consistency in access to care.18 However, our analysis suggests that simply issuing further sets of guidelines is unlikely to promote either regional equity in access to bariatric surgery or even access for most patients eligible for treatment.
Clinical commissioning groups are likely to find it difficult to effectively prioritise patients for care in the short term and determine the capacity required in the longer term. Decisions might be aided by using the widely supported accountability for reasonableness ethical framework, which argues that any prioritisation structure must meet the conditions of public accessibility, relevance to the local population, enforceability, and the availability of an appeals process.19 Tools such as programme budgeting and marginal analysis are also available to help commissioners make more informed decisions about the costs and benefits associated with investing or disinvesting in different treatments.20
This analysis is limited by our inability to accurately estimate the number of people who would be eligible for treatment under NICE guidance and the uncertainty surrounding the influence of other factors on regional variations in treatment provision, such as local capacity shortages or different clinician and patient attitudes towards the acceptability of care. Nevertheless, given that there are many other examples of treatment areas where NICE guidance does not seem to be fully implemented, including multiple sclerosis,21 diabetes treatment,22 treatment for sub-fertility,23 and numerous drug interventions,24 it is clear that clinical commissioning groups will struggle to achieve fairness and equity of provision even if they are given more sophisticated support to implement new and existing guidelines. This requires development of good practice in local prioritisation using recognised tools and frameworks so that there is more transparency around what is being invested in and why, together with research to better understand need and demand.
NHS provision of bariatric surgery has risen by 300% over the past six years but less than 0.6% of those potentially eligible receive treatment
Access to surgery varies widely between regions and primary care trusts and bears little relation to stated policy
Access to care risks becoming even more inequitable as clinical commissioning groups take on NHS commissioning in an era of mounting austerity
Better and more transparent approaches to local priority setting are needed
Cite this as: BMJ 2013;346:f2453
Contributors and sources: AOS, JC, and JD have all been involved in primary qualitative research about the implementation of NICE guidance in this area, and RK, CH, and CM have all been involved with the implementation of NICE guidance at the PCT level. All authors contributed to the idea for this article. The initial review of regional policies was undertaken by CM and updated by AOS. AOS undertook the bulk of the literature review, assisted by JC, JD and CH. Quantitative data were largely researched and contributed by RK. AOS wrote the article, which has been reviewed and commented on by all authors. AOS is the guarantor.
Competing interests: All authors have completed the unified competing interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no financial support for the submitted work and no financial relationships with any organisations that might have an interest in the submitted work in the past three years. This study was undertaken as part of AOS’s NIHR postdoctoral fellowship, but the NIHR had no involvement in the work or the decision to submit the paper for publication.
Provenance and peer review: Not commissioned; externally peer reviewed.