Clinical assessment and management of multimorbidity: summary of NICE guidanceBMJ 2016; 354 doi: https://doi.org/10.1136/bmj.i4843 (Published 21 September 2016) Cite this as: BMJ 2016;354:i4843
All rapid responses
Re: Clinical assessment and management of multimorbidity: summary of NICE guidance - response to Professor Harvey and colleagues
We thank Professor Harvey and colleagues for their positive response to the new NICE guidance on the assessment and management of multimorbidity, and would like to respond to the specific issues raised with the recommendation relating to bisphosphonates, namely:
“Tell a person who has been taking bisphosphonate for osteoporosis for at least 3 years that there is no consistent evidence of:
• further benefit from continuing bisphosphonate for another 3 years
• harms from stopping bisphosphonate after 3 years of treatment.
Discuss stopping bisphosphonate after 3 years and include patient choice, fracture risk and life expectancy in the discussion.”
NICE guidelines are created using a consistent process for the evaluation of evidence which is applied to all review questions . The Guideline Development Group (GDG) considered the evidence for stopping antihypertensive drugs, statins and bisphosphonates, and the difference between the recommendations made is because the evidence available was different. Randomised trials of stopping drugs were sought to answer this question. As Professor Harvey and colleagues point out, “the evidence for long-term benefits is frequently assumed and the risks of stopping therapy under studied”. It is to the credit of osteoporosis researchers that they have carried out randomised trials in this regard when that has only rarely been the case for other treatments.
Based on six cited studies they assert that the guideline “ignores evidence for efficacy of bisphosphonates beyond three years”. Two of the studies cited [3, 4] were included in the guideline systematic review and meta-analysis because they were randomised trials examining the effect of continuing or discontinuing bisphosphonates after a period of initial treatment (three other trials not cited in the response were also included in the guideline review). The remaining studies were not included in the guideline review as they were not eligible. Two are trials which do not examine the effects of discontinuation after a period of treatment,[5, 6] and one is an interim analysis of a study included in the review . This final study is a post-hoc analysis of a trial included in the review in which six subgroups were analysed in terms of two fracture outcomes, with OR for fracture in those continuing treatment vs stopping >1 for five comparisons and <1 for seven, with one statistically significant finding of reduced non-vertebral fracture risk in women with a non-vertebral fracture at baseline and very low bone mineral density . Such a finding is clearly hypothesis generating rather than definitive.
Two studies are cited in support of a statement that there is an “increased risk of vertebral fractures following discontinuation of treatment”, one of which is a trial included in the guideline review  and the other one is an observational study which compares discontinuation after 12 months of treatment with discontinuation after treatment of even shorter durations so is not directly comparable to the trial data in the guideline which examined discontinuation after longer-term initial treatment . The cited trial  shows a statistically significant reduction in morphometric vertebral fracture but not clinical vertebral fracture, non-vertebral fracture or hip fracture . This with other trials included in the review contributed to the GDG’s assessment of the evidence.
Based on the systematic review and meta-analysis of all included trials across all outcomes, the GDG concluded that there was not consistent evidence of either benefit or harm from continuing treatment with bisphosphonates, and the recommendation made reflects that .
Professor Harvey and colleagues make two further points. First, that the recommendation strongly implies that treatment is discontinued in all patients and that it may be misinterpreted as applicable to all osteoporosis therapies. The recommendation clearly says neither but encourages provision of information and discussion of risks and benefits with the patient. Second, that there is asymmetry in the specific recommendation made for bisphosphonates compared to a recommendation for further research for antihypertensives and statins. A specific recommendation was made for bisphosphonates because there was evidence to support it, whereas the GDG considered there was not sufficient evidence to make any recommendation for antihypertensives and statins. Importantly, the research recommendation made is not condition specific. We reproduce it below and would encourage osteoporosis (and other) researchers to consider proposing such trials in appropriate patient groups, and research funders to consider commissioning them.
“The guideline committee considered that 1 or more large, well-designed trials of stopping preventive medicine in people with multimorbidity would be of value in defined patient groups in the community (for example, people in nursing homes, people who are housebound, people with well-defined frailty, people with high levels of multimorbidity or polypharmacy, people with limited life expectancy). Discontinuation could either be complete (all relevant medicines) or partial (for example, reduced intensity of hypotensive or hypoglycaemic treatment). Such trials have to be sufficiently powered to provide evidence of clinically important effects of interventions on outcomes that are relevant to patients and health and social care systems (for example, quality of life, hospital and care home admission and mortality). The committee believed that given the existing evidence, it would be of greater value to evaluate the effects of stopping discrete medicines or drug classes, rather than stopping all preventive medicines at the same time. The committee also believed that no single trial could likely address this research need, since there are many medicines that could be stopped and many defined populations in which this might be of value.” 
In summary, we welcome the general approval of the intent of the guideline and believe that the recommendation made in relation to bisphosphonates is an appropriate reflection of the current evidence, and wish to highlight that the research recommendation made is as applicable to bisphosphonates and other treatments for osteoporosis as it is to treatments for other conditions.
Norma O’Flynn, chief operating officer
National Guideline Centre, Royal College of Physicians, London NW1 4LE, UK
Bruce Guthrie, professor of primary care medicine
Population Health Sciences Division, University of Dundee, Dundee DD2 4BF, UK
1. National Institute for Health and Care Excellence, NG56 Multimorbidity: clinical assessment and management, 2016, National Institute for Health and Care Excellence: London.
2. National Institute for Health and Care Excellence, Developing NICE guidelines: the manual, 2014, National Institute for Health and Care Excellence: London.
3. Black, D.M., et al., Effects of continuing or stopping alendronate after 5 years of treatment: The fracture intervention trial long-term extension (flex): a randomized trial. JAMA, 2006. 296(24): p. 2927-2938.
4. Black, D.M., et al., The Effect of 3 Versus 6 Years of Zoledronic Acid Treatment of Osteoporosis: A Randomized Extension to the HORIZON-Pivotal Fracture Trial (PFT). Journal of Bone and Mineral Research, 2012. 27(2): p. 243-254.
5. Cummings, S.R., et al., Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: Results from the fracture intervention trial. JAMA, 1998. 280(24): p. 2077-2082.
6. Sorensen, O.H., et al., Long-term efficacy of risedronate: a 5-year placebo-controlled clinical experience. Bone, 2003. 32(2): p. 120-126.
7. Ensrud, K.E., et al., Randomized Trial of Effect of Alendronate Continuation Versus Discontinuation in Women With Low BMD: Results From the Fracture Intervention Trial Long-Term Extension. Journal of Bone and Mineral Research, 2004. 19(8): p. 1259-1269.
8. Schwartz, A.V., et al., Efficacy of continued alendronate for fractures in women with and without prevalent vertebral fracture: The FLEX Trial. Journal of Bone and Mineral Research, 2010. 25(5): p. 976-982.
9. Ström, O., E. Landfeldt, and G. Garellick, Residual effect after oral bisphosphonate treatment and healthy adherer effects—the Swedish adherence register analysis (SARA). Osteoporosis International, 2015. 26(1): p. 315-325.
Competing interests: No competing interests
We read your article ‘Clinical Assessment & Management of Multimorbidity: Summary of NICE Guidance’ with great interest. In response, we believe it is important to consider Rheumatoid Arthritis (RA) as a condition with a significant comorbidity burden. RA is a chronic debilitating condition affecting approximately 386,600 patients in the UK, equating to 0.81% of the population . It is characterised by inflammation of the synovial joints resulting in peri-articular tissue destruction and widespread extra-articular features.
Common conditions that co-exist with RA include coronary heart disease and heart failure. A meta-analysis by Aviña‐Zubieta et al found that there was a 50% increased risk of cardiovascular-disease-related death in patients with RA, secondary to ischaemic heart disease and cerebrovascular accidents . Indeed, the incidence of coronary artery disease and stroke in RA patients is the same as that in patients with Diabetes Mellitus [3,4]. Other common comorbidities include chronic pain  and depression. In a systematic review conducted by Matcham et al, the prevalence of major depressive disorder in patients with RA was found to be 16.8%, with the prevalence of depression as 38.8% .
Polypharmacy is also a significant problem in RA patients. A study by Treharne et al (2008) found that the mean total number of medications per patient was 5.39, with only 2.41 of these being directly for RA, on average . Many patients with RA are on disease-modifying-anti-rheumatic drugs (DMARDs), such as methotrexate. In the past, this medication was stopped when patients were having elective surgery performed. However, a study by Grennan et al found that continuation of methotrexate treatment did not increase the risk of infection or surgical complications in patients undergoing elective orthopaedic surgery, and thus it should not be stopped in patients whose disease is controlled by the drug . All RA patients undergoing surgery should have pre-operative blood tests to check bone marrow suppression and liver function.
When treating patients with rheumatoid arthritis, it is important to consider the full comorbidity burden in order to provide a holistic approach to management. The guidelines that have been set out in your article would therefore be particularly pertinent to this selection of patients.
1. D. Symmons et al, The prevalence of rheumatoid arthritis in the United Kingdom: new estimates for a new century Rheumatology (2002) 41 (7): 793-800 doi:10.1093/rheumatology/41.7.793
2. Aviña‐Zubieta JA, Choi HK, Sadatsafavi M, Etminan M, Esdaile JM, Lacaille D. Risk of cardiovascular mortality in patients with rheumatoid arthritis: A meta‐analysis of observational studies. Arthritis Care & Research. 2008 Dec 15;59(12):1690-7.
3. del Rincon, I., Williams, K., Stern, M.P., Freeman, G.L. and Escalante, A., 2001. High incidence of cardiovascular events in a rheumatoid arthritis cohort not explained by traditional cardiac risk factors. Arthritis & Rheumatism, 44(12), pp.2737-2745.
4. Van Halm, V.P., Peters, M.J.L., Voskuyl, A.E., Boers, M., Lems, W.F., Visser, M., Stehouwer, C.D.A., Spijkerman, A.M.W., Dekker, J.M., Nijpels, G. and Heine, R.J., 2009. Rheumatoid arthritis versus diabetes as a risk factor for cardiovascular disease: a cross-sectional study, the CARRE Investigation. Annals of the rheumatic diseases, 68(9), pp.1395-1400.
5. Heiberg, T. and Kvien, T.K., 2002. Preferences for improved health examined in 1,024 patients with rheumatoid arthritis: pain has highest priority. Arthritis Care & Research, 47(4), pp.391-397.
6. Matcham, F., Rayner, L., Steer, S. and Hotopf, M., 2013. The prevalence of depression in rheumatoid arthritis: a systematic review and meta-analysis. Rheumatology, 52(12), pp.2136-2148.
7. Treharne, G.J., Douglas, K.M.J., Iwaszko, J., Panoulas, V.F., Hale, E.D., Mitton, D.L., Piper, H., Erb, N. and Kitas, G.D., 2007. Polypharmacy among people with rheumatoid arthritis: the role of age, disease duration and comorbidity. Musculoskeletal Care, 5(4), pp.175-190.
8. Grennan, D.M., Gray, J., Loudon, J. and Fear, S., 2001. Methotrexate and early postoperative complications in patients with rheumatoid arthritis undergoing elective orthopaedic surgery. Annals of the rheumatic diseases, 60(3), pp.214-217.
Competing interests: No competing interests
The recent NICE guidance on assessment and management of multimorbidity, summarised by Farmer et al. in the BMJ, offers much laudable advice to the clinician caring for elderly patients. The scenario of ever increasing medication load, some of which is instituted to counter side-effects of other medications, is unfortunately common, and few would argue with the overall NICE approach, aiming to actively evaluate and optimise prescription lists. Such a complex document necessitates review of numerous disease areas, with the implication that similar evidential criteria will have been applied across all sections. Sadly, we would contend that the requisite uniformity of approach is absent from the section entitled “Review medicines and other treatments”, specifically in the recommendations pertaining to bisphosphonates.
The requirements of drug development and registration ensure that all drugs currently prescribed have been trialled over a finite length of time (usually relatively short compared with durations of use for chronic conditions commonly encountered in clinical practice). Further, for regulatory purposes across all specialties, the majority of pivotal randomised controlled trials have focused on initiating rather than stopping treatment. The evidence for long-term benefits is frequently assumed and the risks of stopping therapy under-studied. Given the perhaps unique ability of bisphosphonates to persist within the skeleton, several studies have examined the effects of withdrawing these agents. Although there is variation in endpoints, statistical power and approaches to controls across these investigations, the recommendation by NICE that bisphosphonates should be reviewed at 3 years, with the strong implication that treatment is discontinued in all patients, fails to appreciate the harms that will clearly ensue in patients at higher risk of fracture. The flaws of this recommendation, which ignores evidence for efficacy of bisphosphonates beyond 3 years[2-7] and for the increased risk of vertebral fractures following discontinuation of treatment[7, 8], were drawn to the attention of NICE during the consultation process, but did not apparently lead to any re-evaluation of the evidence (https://www.nice.org.uk/guidance/NG56/documents/consultation-comments-an...). Since then the BMJ itself has published data, not based on RCTs but on well-powered population cohort studies, showing a favourable effect of long-term bisphosphonate use on fracture risk and a very favourable risk-benefit ratio for prevention of outcomes such as hip fractures compared to potential iatrogenic femoral shaft fractures.
Whilst the emphasis on bisphosphonates is perhaps not surprising, it highlights the dearth of data on other therapies and demonstrates an obvious imbalance in the criteria applied across diseases (https://www.nice.org.uk/guidance/ng56/evidence/full-guideline-2615543101). For example, the Guideline Development Group identified only one non-blinded trial of cessation of statin therapy containing only 381 individuals, which the group graded as low quality (p263). Similarly, evidence from the 3 trials of antihypertensive cessation was graded as very low quality, other than for an increase in blood pressure (p249). For both these interventions, the recommendation is further research; for bisphosphonates, with arguably a stronger evidence base supporting longer term treatment, the recommendation is clinical review, and suggested cessation of treatment, at 3 years. This clearly does not represent an entirely equitable approach across the entirety of the NICE guidance.
When viewed from a global perspective, it is clear that bisphosphonates, and treatments for osteoporosis in general, are vastly under-prescribed relative to the burden of osteoporotic fractures[10-13]. A further danger introduced by this guidance is that the bisphosphonate recommendations, inappropriate even in their current context, may be misinterpreted as applicable to all osteoporosis therapies. Such an inference would clearly be misplaced for medications such as raloxifene, denosumab and teriparatide, which have substantially shorter durations of action than do bisphosphonates. Major osteoporotic fractures reduce survival by around 20%, increase morbidity, dependence on care and cause a huge economic impact, for example €39 billion per year across Europe[11, 14]. Whilst supporting the general sentiment of the NICE guidance, we strongly refute the bisphosphonate specific recommendations, which if followed are likely to further increase the treatment gap in osteoporosis, ironically making a substantial contribution to the multimorbidity that NICE is attempting to address.
Nicholas C Harvey1,2, Eugene McCloskey3,4, John A Kanis3,5, Cyrus Cooper1,2,6
1.MRC Lifecourse Epidemiology Unit, University of Southampton, Southampton, UK; 2.NIHR Southampton Biomedical Research Centre, University of Southampton and University Hospital Southampton NHS Foundation Trust, Tremona Road, Southampton, UK; 3.Centre for Metabolic Bone Diseases, University of Sheffield, Sheffield, UK; 4.Centre for Integrated research in Musculoskeletal Ageing (CIMA), Mellanby Centre for Bone Research, University of Sheffield, Sheffield, UK; 5.Institute for Health and Ageing, Catholic University of Australia, Melbourne; 6.NIHR Musculoskeletal Biomedical Research Unit, University of Oxford, Oxford, UK
1. Farmer C, Fenu E, O'Flynn N, et al. Clinical assessment and management of multimorbidity: summary of NICE guidance. BMJ 2016;354:i4843. doi: 10.1136/bmj.i4843 [published Online First: 2016/09/23]
2. Cummings SR, Black DM, Thompson DE, et al. Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: results from the Fracture Intervention Trial [see comments]. JAMA 1998;280(24):2077-82.
3. Sorensen OH, Crawford GM, Mulder H, et al. Long-term efficacy of risedronate: a 5-year placebo-controlled clinical experience. Bone 2003;32(2):120-6. [published Online First: 2003/03/14]
4. Ensrud KE, Barrett-Connor EL, Schwartz A, et al. Randomized trial of effect of alendronate continuation versus discontinuation in women with low BMD: results from the Fracture Intervention Trial long-term extension. J Bone Miner Res 2004;19(8):1259-69. doi: 10.1359/jbmr.040326 [published Online First: 2004/07/03]
5. Black DM, Schwartz AV, Ensrud KE, et al. Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial. JAMA 2006;296(24):2927-38. doi: 10.1001/jama.296.24.2927 [published Online First: 2006/12/28]
6. Schwartz AV, Bauer DC, Cummings SR, et al. Efficacy of continued alendronate for fractures in women with and without prevalent vertebral fracture: the FLEX trial. J Bone Miner Res 2010;25(5):976-82. doi: 10.1002/jbmr.11 [published Online First: 2010/03/05]
7. Black DM, Reid IR, Boonen S, et al. The effect of 3 versus 6 years of zoledronic acid treatment of osteoporosis: a randomized extension to the HORIZON-Pivotal Fracture Trial (PFT). J Bone Miner Res 2012;27(2):243-54. doi: 10.1002/jbmr.1494 [published Online First: 2011/12/14]
8. Strom O, Landfeldt E, Garellick G. Residual effect after oral bisphosphonate treatment and healthy adherer effects--the Swedish Adherence Register Analysis (SARA). Osteoporos Int 2015;26(1):315-25. doi: 10.1007/s00198-014-2900-5 [published Online First: 2014/10/10]
9. Abrahamsen B, Eiken P, Prieto-Alhambra D, et al. Risk of hip, subtrochanteric, and femoral shaft fractures among mid and long term users of alendronate: nationwide cohort and nested case-control study. Bmj 2016;353:i3365. doi: 10.1136/ bmj.i3365 [published Online First: 2016/06/30]
10. Kanis JA, Svedbom A, Harvey N, et al. The osteoporosis treatment gap. J Bone Miner Res 2014;29(9):1926-8. doi: 10.1002/ jbmr.2301 [published Online First: 2014/06/24]
11. Svedbom A, Hernlund E, Ivergard M, et al. Osteoporosis in the European Union: a compendium of country-specific reports. Archives of osteoporosis 2013;8(1-2):137. doi: 10.1007/s11657-013-0137-0 [published Online First: 2013/10/12]
12. Klop C, Gibson-Smith D, Elders PJ, et al. Anti-osteoporosis drug prescribing after hip fracture in the UK: 2000-2010. Osteoporos Int 2015;26(7):1919-28. doi: 10.1007/s00198-015-3098-x [published Online First: 2015/05/13]
13. Kanis JA, Borgstrom F, Compston J, et al. SCOPE: a scorecard for osteoporosis in Europe. Archives of Osteoporosis 2013;8:144. doi: 10.1007/s11657-013-0144-1 [published Online First: 2013/09/14]
14. Harvey N, Dennison E, Cooper C. Osteoporosis: impact on health and economics. Nat Rev Rheumatol 2010;6(2):99-105.
Competing interests: All authors have no disclosures in relation to this work. Outside the current work: NCH reports consultancy, lecture fees and honoraria from Alliance for Better Bone Health, AMGEN, MSD, Eli Lilly, Servier, Shire, Consilient Healthcare and Internis Pharma; EVM reports research funding/ honoraria/ advisory boards for Amgen, Alliance for Better Bone Health, Bayer, Boehringer Ingelheim, GE Lunar, GSK, Hologic, Lilly, MSD, Novartis, Pfizer, Roche, Sanofi, Servier, Tethys, UCB, Warner Chilcott; JAK reports grants/ non-financial support from Amgen, Lilly, Medimaps, Unigene, Asahi, Radius Health, AgNovos, and is the architect of FRAX but has no financial interest; CC reports consultancy, lecture fees and honoraria from AMGEN, GSK, Alliance for Better Bone Health, MSD, Eli Lilly, Pfizer, Novartis, Servier, Medtronic and Roche.
Re: Clinical assessment and management of multimorbidity: summary of NICE guidance. NICE AND THE BLAME GAME
"CLINICIANS CAUSE uncertainty about the balance of benefit and harm of treatments in people with multimorbidity because evidence is largely based on trials of interventions for single conditions, from which people with multimorbidities are often excluded" (my capitals) is a statement to which I object.
Has a subeditor mistakenly inserted the word "cause" instead of using the word "suffer"? Or are the NICE affiliated authors stuck with mindsets that consider clinicians are the problem?
I think the statement would be far more accurate if the word "Clinicians" was removed and replaced with the phrase "Some researchers, NICE, QoF and sundry parties lacking wisdom and perspective..."
My point may seem petty but at a time when morale in the medical profession is low maybe it is not.
Competing interests: I am a clinician.
I wonder how many patients were involved in the selection of the graphic on the cover of this week's BMJ illustrating the article on guidelines for managing multimorbidity, Where is the dignity for our patients that we are all enjoined, quite rightly, to respect, in a 'joky' caricature implying that all people with multi-morbidities are bumbling crazy 'bag ladies' shambling about in ill fitting slippers.
If this is the view of the profession then we really need very tough and new guidelines!
Dr. J. Fisken
Competing interests: No competing interests