Would you like to help prevent overdiagnosis and overtreatment? GP Julian Treadwell has put together a proposal for the RCGP to take on “Preventing Overdiagnosis and Overtreatment” as one of their next 3-year Clinical Priority areas. Please will you support this by getting in touch and adding your name at: email@example.com
See “Winding back the harms of too much medicine” http://www.bmj.com/content/346/bmj.f1271
RCGP Clinical Priority Programme
Proposal Form for Three-Year Clinical Priorities
1. GENERAL INFORMATION
Proposed Area of Clinical Priority:
Preventing Overdiagnosis and Overtreatment
Lead Individual for Proposal:
Dr Julian Treadwell
Email address & contact number:
Lead Organisation (if relevant):
Support for and endorsement of this proposal:
Please list the names of the organisations and individuals supporting this proposal:
Why is it important for this area to be one of the RCGP clinical priorities?
This February, the BMJ launched its “Too Much Medicine” campaign 1,2 and is organising an international conference this September: “Preventing Overdiagnosis” in conjunction with Dartmouth Institute and Bond University 3. Inspired by, and connected to, this movement I wish to propose “Preventing Overdiagnosis and Overtreatment” as a RCGP Clinical Priority for 2014.
The problem of overdiagnosis and overtreatment has emerged over the last decade as a consequence of improved health technologies and research, systematised EBM/guideline driven chronic disease management, and implementation of population wide screening programmes. The undoubted benefits of this progress for some have come hand-in-hand with undesired consequences for many, who have been diagnosed with illnesses that would never have harmed them or who have taken risk factor modification treatments to prevent end point events they were in fact never going to have 4,5,6. This dilemma arises in all areas of medicine, from histopathology to psychiatry. GPs are involved at many levels, but most obviously in the processes of risk factor modification and implementing or advising our patients about screening programs.
A tension arises when producing a measurable and significant population level benefit but a less impressive potential benefit for the individual. For example, treating populations at less than10% risk of CVD with statins could produce a population level benefit, but the NNTs are very high- conferring a very low potential individual benefit 7. Similarly, screening mammography in the UK has most recently been estimated to save approximately 1,300 breast cancer deaths a year but has NNS of approximately 250 and results in overdiagnosis of approximately 4,000 breast malignancies which would never have caused a problem in the woman’s lifetime 8.
Currently, the population level benefits of these strategies are most heavily promoted. Guidelines and research summaries generally express benefit in terms of RRR or hazard ratios 9-13,Appx1 which cast interventions in the most favourable light. The QOF incentivises treating to population based targets.
There is research to show that when patients are fully informed about the potential benefits and risks of screening processes or treatments, many would choose not to undergo screening or would only be interested in treatment at a very different threshold of benefit than our currently accepted practice would encourage. 14-22
At the same time, it is very difficult for GPs to access really meaningful evidence to enable us to give truly patient centred advice and prevent overdiagnosis and overtreatment (as defined by patients’ preferences). An illustrative clinical example is given in appendix 2. Much work has been done on decision making aids for patients 23 but their value is likely to be limited unless the heath professionals advising them are better equipped with knowledge and feel able to use it.
Therefore a problem exists relating to issues of consent, patient choice and the kind of patient centred medicine we wish to practice. This proposal aims to empower GPs to practice better evidence based, patient centred medicine.
How does this clinical area relate to the current political and policy environment, including UK and devolved country health care strategy and initiatives?
• RCGP Medical Generalism 2012 “seeing the person as a whole” Managing multi-morbidities and chronic illness.
• RCGP,RCP,NHS Alliance 2004 Chronic Disease Management. 3.13, 3.16 multi-morbidity and patient choice.
• nMRCGP Curriculum Area 2 Person Centred Care 2.2.1, 2.2.2, 2.2.3
• nMRCGP Curriculum Area 4 Comprehensive Approach 4.1.1, 4.1.4
• nMRCGP Curriculum Essential Features 3 “..to adopt a critical and evidence based approach”
• GMC Consent Guidance 2008 esp. clauses : 5,7,8,9,19.
• Kings Fund 2011. Making Shared Decision Making a Reality, No Decision About Me Without Me.
How does this clinical area link to RCGP strategic aims e.g. building strong links with stakeholders to support quality improvement and providing leadership to promote clinical excellence?
I would envisage that work on this area would be highly collaborative, drawing in expertise from external bodies – academic, NHS, specialist and patient groups.
There is a lot of work going on in the area already and the aim is to disseminate this to GPs.
AAre you confident that there are funding opportunities to support development work in this clinical area? The RCGP clinical priority programme currently has no specific project funding attached to it.
Activities 1+2 (below) I would anticipate could be done with relatively little funding.
The development of an on-line resource (3, below) would require funding and sources suggested have included The Wellcome Foundation, DoH.
What opportunities do you envisage for leadership and promotion of good practice? This includes a demonstrable need for a senior clinician (clinical champion) to lead a programme of work in this clinical area and the likelihood of attracting a high calibre individual to this role.
I have no doubt that we would attract high calibre individuals to get involved – see the list of supporters. One option would be to have a small group of clinical champions rather than just one.
3. KEY MESSAGES, TANGIBLE IMPACT OF LASTING BENEFIT
Please list the major areas of activity you envisage informing a three-year programme of work in this clinical area and what outcomes are sought (may include, for example, audit, clinical guidance and its implementation, commissioning, education and learning, and evidence-based research).
1) Publication of articles (in BJGP and/or perhaps in collaboration with BMJ) to inform, educate and provoke discussion in primary care.
2) Engagement with other organisations looking at overdiagnosis, overtreatment and patient choice issues.
3) To produce a new, on-line evidence resource and decision making aid aimed at GPs with the goal of providing really useful, clinically meaningful evidence to answer the kinds of questions described in Appendix 2.
The need for this kind of resource has been mooted and partially defined at a recent international conference 24.
This is an ambitious goal, but one I am sure could be realised with collaborative working and modest funding. There is the potential here to create something groundbreaking and transformative.
Please describe what would be the tangible impact and long-term benefits of these proposed activities on the following areas:
a. Benefits for patients:
Improved information and more meaningful choice/consent in making testing and treatment decisions.
b. Benefits for GPs:
Better access to meaningful EMB
A return to providing patient centred care.
“Permission” to make patient centred decisions rather than blind guideline following.
Enhancing the role of the generalist.
Long term – are GPs servants of specialist agendas or can we have more of a voice regarding what we think is best for our patients and what is the best use of an increasingly limited primary care resource?
c. Benefits for the wider primary care community and integration of care (including specialist services if applicable):
Potential financial and time savings generated by better tailored treatments and chronic disease management activity.
5. REFERENCES Please provide relevant supporting references.
1)Moynihan R,Glasziou P,Woolshin S,Schwartz L,Santa J, Godlee F.Winding back the harms of too much medicine. BMJ 2013;346:f1271
2)Godlee F. Too Much Medicine. BMJ 2013;346:f1328
3)Preventing Overdiagnosis Conference Sep 10-12 2013 Dartmouth Institute for Health Policy and Clinical Practice. http://www.preventingoverdiagnosis.net
4)Moynihan R,Doust J, Henry D. Preventing Overdiagnosis: how to stop harming the heathly. BMJ 2012;344:e3502
5)McCartney M .The Patient Paradox. Why sexed up medicine is bad for your health. Pinter & Martin 2012
6) Welch G,Schwartz L,Woolshin S. Overdiagnosed. Making people sick in pursuit of health. Beacon Press 2011
7) Cholesterol Treatment Trialists’ (CTT) Collaborators. The effects of lowering LDL cholesterol with statin therapy in people at low risk of vascular disease: meta-analysis of individual data from 27 randomised trials. The Lancet, Volume 380, Issue 9841, Pages 581 – 590
8) The Independent UK Panel on Breast Cancer Screening. The benefits and garms of breast cancer screening: An Independent Review. Cancer Research UK and DoH; October 2012
9) Sedrakyan A, Shih C Improving depiction of benefits and harms: analyses of studies of well-known therapeutics and review of high impact medical journals. Med Care 2007 Oct;45(10 Supl2):S23-8
10) Nuovo J, Melnikow J, Chang D. Reporting number needed to treat and absolute risk reduction in randomized controlled trials. JAMA 2002 Jun 5;287(21):2813-4
11)Schwartz L,Woolshin S,Dvorin E, Welch HG. Ratio measures in leading medical journals: structured review of accessibility of underlying absolute risks. BMJ 2006;333:1248
12) Schroter S,Glasziou P,Heneghan C .Quality of descriptions of treatments: a review of published randomized controlled trials. BMJ Open 2012;2:e001978
13)Appendix 1 Informal analysis of some current UK guidelines.
14) Leaman, H. and Jackson, P. R. (2002), What benefit do patients expect from adding second and third antihypertensive drugs?. British Journal of Clinical Pharmacology, 53: 93–99.
15) Taylor C, Ward A, Patients’ views of high blood pressure, its treatment and risks.Aust Fam Physician. 2003 Apr;32(4):278-82.
16) Hux JE, Naylor CD.Communicating the benefits of chronic preventive therapy: does the format of efficacy data determine patients’ acceptance of treatment? Med Decis Making. 1995 Apr-Jun;15(2):152-7.
17) Misselbrook D,Armstrong D Patients’ responses to risk information about the benefits of treating hypertension. BJGP 2001,51,276-279
18) Malenka DJ, Baron JA, Johansen S, Wahrenberger JW, Ross JM.
The framing effect of relative and absolute risk. J Gen Intern Med. 1993 Oct;8(10):543-8.
19) Moxley A,O’Connell D,McGettigan P,Henry D.Describing Treatment Effects to Patients.How They Are Expressed Makes a Difference J Gen Intern Med. 2003 November; 18(11): 948–959
20) Hersch J,Jansen J,Barratt A,Houssami N,Howard K,Dhillon H,McCaffery K.Women’s views on overdiagnosis in breast cancer screening: a qualitative study. BMJ 2013; 346:f158
21) Flood AB, Wennberg JE, Nease RF Jr, Fowler FJ Jr, Ding J, Hynes LM. The importance of patient preference in the decision to screen for prostate cancer. Prostate Patient Outcomes Research Team. J Gen Intern Med. 1996 Jun;11(6):342-9
22) Melnikow J, Paterniti D, Azari R, Kuenneth C, Birch S, Kuppermann M, Nuovo J, Keyzer J, Henderson S.
Preferences of Women Evaluating Risks of Tamoxifen (POWER) study of preferences for tamoxifen for breast cancer risk reduction. Cancer. 2005 May 15;103(10):1996-2005.
23) Stacey D, Bennett CL, Barry MJ, Col NF, Eden KB, Holmes-Rovner M, Llewellyn-Thomas H, Lyddiatt A, Légaré F, Thomson R. Decision aids for people facing health treatment or screening decisions. Cochrane Database of Systematic Reviews 2011, Issue 10. Art. No.: CD001431.
24) Gigerenzer G, Muir Gray J A. Strungmann Forum Reports. Better Doctors, Better Patients, Better Decisions .Envisioning Health Care 2020.
MIT Press:2011.204p. 280p.
Informal Survey of Selected UK Guidelines April 2013
Aim: To review how the benefits of pharmacologic interventions for important chronic conditions are presented in widely used guidelines.
Selection method: The chronic conditions chosen are those which feature in the GP contract Quality and Outcomes Framework. NICE guidance is analysed first and then some other guidance from leading bodies.
Intervention examined How are the benefits of the intervention expressed numerically? +comment (mine)
NICE CG 127 Hypertension
Quick ref. guide
BP lowering medication No quantitative benefit reported.
NICE CG 127 Hypertension
Full Guideline 328 pages
BP lowering medication ARR and hazard ratios mainly.
Only at p218 do we get tables with ARR included.
Useful data buried in complex tables/text.
BP lowering medication RRR only – mentioned in just one line.
SIGN 97 (full version)
BP lowering medication Hazard ratios only mentioned, briefly in one statement.
NICE CG 67 Lipid modification
Quick ref guide
Use of statins No quantitative benefit reported.
NICE CG 67 Lipid Modification
Full Guideline 241 pages
Use of statins Mainly RRR
One mention of NNT (95) for ASCOT trial
Use of statins RRR mainly
One ARR re HPS statins in stroke
NICE CG 36 Atrial fibrillation
Quick reference guide
Anticoagulation for stroke prevention in persistent AF
No quantitative benefit reported.
NICE CG 36 Atrial Fibrillation
Full Gudieline 171 pages
Anticoagulation for stroke prevention in persistent AF ARR and Hazard ratios both used.
Better, but meaningful figures hidden in lots of text.
SIGN 129 Antithrombotics: Indications and management.
Anticoagulation for stroke prevention in persistent AF
Quick ref Baseline stroke rate given with CHADS2 stratification.
No figures given regarding effect on risk with anticoagulation.
Anticoagulation for stroke prevention in persistent AF Mixture of Hazard ratio and RRR within text.
One ARR figure given only.
NICE TA160 Osteoporosis- primary prevention.
Quick ref guide.
Pharmcotherapy for fracture prevention. No quantitative benefit reported.
NICE TA 160 Osteoporosis – primary prevention
Pharmcotherapy for fracture prevention. RRR only.
No baseline risks.
Does state 24-30% discontinuation rate for bisphosphonates.
NICE CG 66/87 Diabetes
Pharmacotherapy for prevention of complications Almost all OR and RRR
Much based on surrogate endpoints.
Some absolute risks used: Progression of retinopathy (though returned to RRR as regards blindness), Incidence of proteinuria (but RRR for definite renal outcomes)
SIGN 116 Diabetes
Quick reference guide No quantitative benefit reported
SIGN 116 Diabetes
Pharmacotherapy for prevention of complications Mainly expressed as RRR and HR
NNTs used in subgroup analysis of pioglitazone and once for hypertension treatment in stroke prevention. Both deep within text.
ARR used comparing hypoglaycaemic incidence with different insulins
NICE CG 48 Secondary Prevention of MI. Quick Ref guide No quantitative data
NICE CG48 Secondary Prevention MI. Full version. 211 pages
Pharmocotherapy for prevention of CV events and death Mainly OR,HR, RRR
4 uses ARR – 2 for antiplatelets, one for BBlockers, one for statins (used<24hr post MI).
SIGN 98 Stable Angina
Pharmocotherapy for prevention of CV events and death
47 pages Antiplatelets, BBlockers an dstatins: no numerical data
ACE inhibitors, some RRR, ARR in text
Chromic Heart Failure
Pharmacotherapy for treatment of CHF
163 pages CG5 : no quantitiative data
CG108: Update for HFPEF and A2RA use. Some RRR and trial data only in appendix E
European Society of Cardiology 2012 Heart Failure Guideline
Pharmacotherapy for treatment of CHF
60 pages Good mixture of RRR,ARR,NNT within text.
SIGN 95 Management of Chronic Heart Failure. Quick Ref Guide No quantitative data
SIGN 95 Management of Chronic Heart Failure Full version
Pharmacotherapy for treatment of CHF Mixture of RRR, OR and NNT within text.
What evidence do we need? An illustrative clinical example.
Mary is an 82 y.o woman who is generally in good health.
She had a Colles fracture 3 years ago and has been on alendronate and calcium/vit D since but never had a DEXA scan.
She lives on her own having been widowed 5 years ago and is independent, though her mobility is decreasing slowly from osteoarthritis; she is resisting using a stick so far but knows she probably should.
She comes to see you because she has been found to have hypertension and after a full CVrisk workup has a home BP average reading of 162/94 and fasting TC/HDL ratio of 5.3 She has no other cardiovascular risk factors.
Discussing the options with her, it transpires she is not very keen on taking medication unless it is “really necessary”. Friends of hers have had side effects and problems. She tells you she’s not worried about having a heart attack or trying to stave off death but really fears having a stroke, especially now she’s been told her blood pressure is high.
“By the way” she says, does she really need to take these tablets for her bones? They’re OK, no major side effects but the calcium’s nasty and the other tablets are a bit like horse pills…..
Following our usual guidelines would almost inevitably end up with further drug treatment: probably a statin and two anti-hypertensive drugs in addition to continuing with her osteoprosis medication. What do we need to know to help Mary make a really informed choice?
• What is her actual risk of stroke? (as opposed to all CV events)
• What non pharmacologic interventions might have a real effect on this?
• What is the actual risk reduction for stroke (and overall CV events) to be gained from drug treatment for blood pressure ?
• What is the most appropriate blood pressure target to aim for in the over 80s?
• Is any particular class of antihypertensive better at preventing stroke than others?
• How effective are statins at reducing stroke risk in real terms? (vs all CV events)
• What effect do blood pressure drugs have on falls risk at her age?
• What is the real likelihood of adverse effects from each of the different medications on offer?
• What is her baseline fracture risk?
• What is the actual risk reduction in fracture obtained by her medication?
• How do we judge the benefit (from research) on radiologically detected vertebral fractures vs clinically apparent fractures.
• How long should alendronate be continued?
• Is there any merit in a DEXA scan at all?
• What about the current safety concerns about bisphosphonates? (stress fractures, ONJ, gastric bleeds and malignancy) Are they meaningful?
These are complex questions and require a fairly substantial knowledge base on the part of her doctor. Guidelines currently in use do not address these sorts of meaningful questions in any kind of accessible manner, hence the need for doctors to spend a lot of time and energy searching for the answers or (as I suspect is usually the case) succumb to following prescriptive guidance.
What would a guideline/information resource designed for GPs by GPs look like….?