The dumbing down of Medicine?
Robyn McDermott
DRS Annual Conference Adelaide 10 September 2006
DALYs, by broad cause group 1990 - 2020 in Developing Countries (baseline scenario)
%
50
1990 2020
25 DALY = Disability adjusted life-year
Communicable diseases, maternal and perinatal conditions and nutritional deficiencies
Injuries
Noncommunicable conditions
Source: WHO, Evidence, Information and Policy, 2000
Compression-decompression of morbidity
• Early “good news” of the health transition: Improved length and QoL “nearly everywhere” (Caldwell,1989)
• But in developed countries, less than 50% of CVD mortality reduction due to decreased incidence • Role and cost of “medical prevention” and medical rescue • “Diabesity” in industrialised countries (1980-2000), then everywhere • Longer life for Australians, but also longer with disability (AIHW 2006)
Life expectancy increased in Australia, but with more years of disability
In 2002-3, 27,831 coronary Angioplasty and 26,110 coronary stenting procedures performed
Large increase in lipid lowering drugs accompanied by only modest reduction in AMI, England 1996-2002
Source: BMJ 2004;329:645
Decline in coronary heart deaths England & Wales 1981-2000
Source:Unal et al BMJ 2005
• 54% fall in CH Deaths (68,230 fewer deaths in 2000) – IMPACT methodology • Smoking decline of 35% (29,715 fewer deaths: 5,035 in known CHD and 44,675 in non-CHD) • Population total cholesterol fell by 4.2% (5,770 fewer deaths due to diet and 2,135 fewer deaths due to statin treatment) • Mean pop BP fell by 7.7% (5,870 fewer deaths, including 1,890 due to HT treatment) • Overall, 81% of deaths avoided occurred in people without CHD (19% in those with CHD)
Decline in CHD England & Wales: Contribution of primary vs secondary prevention
(Unal et al 2005)
45000 40000 35000 30000 25000 20000 15000 10000 5000 0 Tobacco Chol BP control Known CHD No CHD
The Rose effect: a small change in a large number has a bigger impact than a larger change in a smaller number of high risk individuals
Projected prevalence of overweight males and females aged 30 years or more, 2005 and 2015: Shrinking “normality”
Men 2005
26% 20%
Men 2015
74%
Overweight Not Overweight Overweight Not Overweight
80%
Women 2005
31%
25%
Women 2015
69%
Overweight Not Overweight
Overweight Not Overweight
75%
Increasing prevalence of overweight among Australian children 1985 - 1995
30 25 20 15 10 5 0
7 - 15 years 12 - 14 years 7 - 15 years 12 - 14 years
1985 ACPHER 1995 NNS 1985 ACPHER 1996 NNS
Girls
Boys
Obesity: the tipping point
• Predicted upturn in CVD mortality due to “diabesity” (Zimmet 1996) • NZ CVD mortality increasing among young and middle-aged adults (2006)
Coronary mortality (deaths per 100,000) as a function of saturated fat intake
Source: Kromhout et al Seven Countries Study, 1995 Prev Med
1400 1200 1000 800 600 400 200 0 0 5 10 15 20 25
What should we be doing in prevention – where’s the evidence?
Segal 2006 Monash www.monash.edu.au/centres/che
• Aims: How best - most efficiently - reduce burden of harmful lifestyle behaviours:
– – – – physical inactivity poor nutrition alcohol misuse tobacco smoking
• Client: Dept of Health and Ageing (DoHA) • Advisory Panel: DoHA, DHS, Vichealth, Epidemiologists, Experts in risk factor areas
Broad Research Tasks
1 Identify intervention options to reduce burden from harmful lifestyle behaviours
Identified 438 articles and reports and collated data for 168 interventions
2 Select interventions for economic analysis
3 Compare performance of options through cost-effectiveness & cost-utility analysis
Criteria for selecting interventions
Breadth of scope • modalities – population/public health, clinical • target group (low & high risk) Objective evidence available • behaviours, clinical, health Suitable for marginal analysis - not structural/system wide Health sector (mostly)
Interventions Analysed & Quality of Evidence
Interventions Analysed: 34
Key Components of Quality of evidence:
• Valid Trial results – behaviour, clinical, health • Report of maintenance of behaviour change • Reln b/w Behaviour change & clinical & LT health
Overview of quality of evidence
16 Poor quality # - no control, ST F-U, behaviour only 8 Mediocre ## 10 Good ### 1 Excellent #### - RCT, LT F-U, health outcomes
Multi Risk Factor - Adult
1. Large scale media + community – UK FFFF 2. Large scale media + community, 5 yr intensive for CVD risk factors. Stanford 5 city
3. Workplace. Group O‟weight males Gutbusters 4. Workplace – European collaborative trial multifactorial prevention of CHD 5. Primary care: health checks by practice nurse. The OXCHECK study
# #
# # ###
Multi Risk Factor – Children (Schools)
6. 18 lessons + hardware to TV viewing, mean age 9.5yrs (Robinson) 7. Intensive 32 lessons, interdisciplinary program to reduce obesity, by TV, activity, mean 11.7yrs (Gortmaker)
# ##
8. Lessons on exercise, nutrition, smoking + PhysEd classes, mean age 9 yrs (Harrell)
9. 20-lessons, multiple risk-reduction via education + Phs Ed mean 15 yrs (Killen)
#
#
Physical Activity
10. GP active script – Australia (Hwang) 11. GP active script + leisure centre contact – New Zealand (Elley) 12. GP Referral to ½ price supervised exercise for patients w. CVD risk factors UK (Taylor) 13. Access to free community based exercise programs, for the elderly (Munro) 14. Access to Exercise specialist in primary care for elderly Australia (Halbert)
# ## ## # ###
Nutrition
15. Intensive dietician +/- GP counselling re nutrition in primary care (Pritchard)
#
16. Mediterranean diet for persons post AMI by cardiologist and dietician (DeLorgeril) 5 year
17. Reduced fat diet/group education persons w. IGT, 5 years (Swinburne)
### # ### ## ###
# ### #
18. Diet +/- orlistat for obesity
19. Nutritionist consult for persons with IGT The Finnish Diabetes Prevention study (Eriksson) 20. Clinician advice using IT – Talking Computer 21. Advice by practice nurse to increase fruit and veg (Steptoe) 22. State wide media campaign„2 fruit 5 veg‟
Smoking
23. Mass media aimed at entire population. Massachusetts Tobacco Control Program (Beiner) 24. Mass media campaign primarily at smokers 18-40 yrs Aust. National Tobacco Campaign (Wakefield)
#
#
## ## ## ##
25a Minimal Advice by clinicians re smoking cessation
25b More Intensive advice from clinicians re smoking cessation (Silagy) 26. Behavioural and NRT 27. Phone counselling + NRT (Zhu) 28. Pharmacotherapy; buproprion + counselling
## #
Approach to Determining Performance
• Record outcomes from high quality trials
intermediate - behaviours, clinical parameters, final - mortality, events, quality of life
• Model relationship b/w reported outcomes and
Current health and wellbeing Future health and wellbeing
• Estimate costs
of program - from description (or as reported) downstream costs – based on health service use (as reported or modelled.)
Primary outcome measure
QALY (QoL x time in health state + Δ LYs)
• Combine morbidity and mortality • Preference based instrument (perceptions) • Utility values reported in clinical trials • Published algorithm to derive utility value from SF-36 • Focus on efficiency defined as $ / QALY
$ / QALY benchmark for “good value” in Australia
Current Australian norms • < $40,000/QALY threshold for 100% listing
drugs on PBS • given high quality evidence
Overall great performance $/QALY <$5,000
Mediterranean diet post-AMI
(60% reduction in major CVD events & mortality, 6 yr follow-up)
$340
####
Lifestyle for persons with IGT
(53% reduction in incidence of T2DM)
$1,900
###
Brief intervention (GP advice) for alcohol misuse
$200700
#
$5,000-$15,000 / QALY
Reduced fat diet for IGT (20% reduced incidence of T2DM over 3 years) Nutritional counselling by nurse in GP Minimal GP quit smoking advice Intensive GP quit smoking advice NRT + counselling to quit smoking Buproprion (Zyban) + counselling to quit smoking
$10,000 $10,600
M=$5,300 F=$8,600 M=$6,600 F=$10,700 M=$11,800 M=$10,500 F=$14,000
# # # # # ##
Naltrexone + CBT for alcohol dependence
$5,200$13,000
#
<$20,000 / QALY, but weaker evidence
Fighting Fat Fighting Fit (FFFF; BBC 2001) Mass media to fight obesity Stanford 5 City Gutbusters 2 fruit 5 veg (Vic health, 98) US mass media smoking Aust Nat Tobacco Campaign Brief interventions in primary care for heavy drinkers
$5,600 $12,600 $20,000 $50 $133 $33 $35-$250
# ## # # # # #
>$20,000 / QALY mixed quality evidence
School (Robinson) School (Gortmaker) School (Killen) Oxcheck (nurse check for high risk in GP) NZ Green prescription Physical activity in the elderly (Halbert) Orlistat $75,000 $51,000 $37,000 $63,000 $29,000 $40,000 $83,000
# ## ## ### ## # #
NRT + counselling (women)
$20,000
#
Summary of Decisions Concerning the Listing of drugs on the PBS (Pharmaceutical Benefits Schedule)
$/QALY or $/LY N
% listing recommended at requested price % listing recommen ded at lower price
% rejected
<$40,000 $40,000-$80,000 >$80,000
19 8 7
89 50 0
11 0 29
0 50 71
Source George et al (2001), Tables 1 and 2 p1106, 1107 Notes: # Incremental cost per QALY or Life Year gain , n = number of submissions that expressed outcomes in QALYs or LYs between 1991 to 1996 with $/QALY or $/LY falling within the nominated range
100%
20%
40%
60%
80%
0%
Ph ar m ac eu tic al s (n =5 2)
86%
Li fe st yl e (n =6 8)
2%
Al lie d he al th (n =2 9)
9%
Va cc i na t io n (n =1 7) en t Te r ti
47%
Tr ea tm
(n =1 19 ) ar y ca re
Likelihood of funding
35%
(n =4 8)
36%
Health Expenditure Australia 20034 (AIHW)
Health Activity
Hospitals
Medical services Pharmaceuticals High level residential care $’000 % budget
26,183
12,961 10,935 4,985
33.4
16.5 14.0 6.3
Dental
Other clinical Public Health Total
4,694
3,378 1,265 78,369
6.0
4.3 1.6 100.0
In summary
• Much primary prevention, especially nutrition, is cheap and highly cost effective • We spend a huge amount on big pharma and have high costs in adverse events especially in the elderly • Need to invest +++ in children‟s nutrition and fitness • Some legislative measures cost nothing yet are not considered (tobacco vs food policy)
What about Equity?
• These interventions improve average outcomes, but may widen social disparity eg tobacco • Random sample of 95 Cochrane reviews, only 1 included differences in treatment effect by SES • Future research should include analysis of impact by SES and other social indicators • There is generally high willingness to trade some health for health equity (Wagstaff, J Hlth Econ 1991)
What about the doctors and health services?
The rise and rise of chronic disease: Primacy of Primary Care?
Primary care has evolved over time in most countries, but only a few have been able to improve essential features since the 1970‟s… Understanding that the main chronic care providers are the patients, 3 things follow…
• Need to invest in increasing capacity of patients to be experts (cf investment in training health professionals) • Integrated rather than fragmented community-based care (eg Kaiser Permanante model) • Funding model should reward quality and continuity of care for chronic conditions, rather than the “billable 6-minutes”
Unintended consequences of 30 years of health care reform
• Behavioural impact of fee-for-service medicine • Behavioural impact of high cost medical education • Domination of drugs • Procedural skills deficit and the marginalisation of education • Flight to private specialist practice
Past and projected full-time equivalent medical workforce per 100 000 persons in Australia, 1995–2012
Source: Joyce et al, MJA 2006;184:441-6.
Evidence-based medicine: Who produces what evidence?
• Hierarchy of evidence, with meta-analysis of RCTs at the top • Drug trials are the gold standard • Complex community-level prevention interventions (with non-commercial outcomes) score low in methodological rigor, are expensive, time-consuming and logistically difficult
Increasing proportion of major studies funded by industry, 1994-2002
Source: Patsopoulos et al, BMJ 2006;332:1061-4
Choosing the expensive option
• Increased use of gabapentin (now > US$2.6 billion in sales globally) • Cox-2i vs ibuprofin vs Fish Oil for arthritis • Angiotensin receptor blockers vs thiazides for hypertension • Increased use of statins (high dose atorvastatin for “new” treatment goals) vs Mediterranean diet
The year the total SA budget will be spent on health care
What to do?
1. Significant role extension for non-doctors, especially for chronic care and procedural work 2. New funding model rewarding prevention, quality and hospital avoidance 3. Large investment in obesity prevention trials, including significant industry regulation and/or partnership 4. Establish a “Prevention Benefits Schedule” along the same lines as PBS, with similar cost per QALY standards
How will this make Medicine smarter?
• Growing cognitive dissonance between caseload and service model • Fee for service is increasingly burdensome administratively and distracts from the “real” work of medicine • Liberation from big pharma • Doctors‟ knowledge of non-drug therapy and prevention would improve • Teamwork and reflective practice is good for doctors too
Medicine: A brief history