Impact of individual level interventions – should we go for more individual level, or for policies that prevent the need for pills. Some thoughts
“Prevention” interventions at the level of the individual are important, but “prevention” interventions that affect whole populations are arguable more important from a population level. Here’s some thoughts on the impact of individual level prevention interventions at population level.
Policy interventions versus individual level interventions
Individual level interventions are important
Interventions need to shape the behaviour of individuals. The NHS has a critical role in this – this is mainly at the level of “preventive” interventions delivered at the level of the individual – smoking cessation, alcohol brief intervention, weight management. These can be termed “clinical preventive services”.
With the exception of stop smoking and alcohol screening and brief advice (which are spectacularly good value for money and under implemented across the system), most of the approaches to healthy lifestyles are based on individual level interventions – weight management etc these have marginal individual level effects (other than bariatric surgery – but an expensive surgical solution to a public health problem) and limited to no population impact.
Interventions implemented individual by individual at may be effective and cost effective (and sometimes cost SAVING) and thus good value for money. However they:
- might not make a big impact on population level outcomes
- might not be implementable at scale, or affordable at scale – even if we had the capacity and skills (which we don’t)
- are arguably inequitable unless very carefully targeted.
Some examples are appended relevant to weight management, stop smoking and alcohol services.
Population level policy interventions – policies prevent the need for pills
More powerful population health policy interventions have a more powerful effect[i]. There is much to be learned from national approaches to both tobacco and alcohol control – targeting availability, acceptability and affordability.
Weak and insipid PH policies will have weak and insipid effects. This is illustrated by a recent paper in the BMJ[ii] neatly quantified how “interventionist” approaches are significantly more effective (by a magnitude of 50 times) than less intrusive intervention. Such interventions are arguably more equitable, with some evidence in the case of minimum unit prices.
Policy interventions shape the exposure to risk and or the environmental and economic cues that drive behaviour, in a large population, and are arguably more progressive and equitable. Govt (local and national) has critical role in prevention through population wide policies. Perhaps this can be framed as “a key role of local govt in NHS then is to prevent stuff in order to protect the social institution that is the NHS”.
Policy interventions are FREE for the NHS (for eg in alcohol the cost is to the alcohol industry – thus from an economists perspective re internalising the externality). However investment might be needed to bring a policy to the point of decision – any strong policy will be met by opposition.
Some examples and some evidence
- Obesity in children
MEND and children’s obesity. Trial recently reported[iii].
- this is about the best evidence based programme for obesity “treatment” I think we have. Thus its hard to not support.
- programme for 7-13yr olds
- 0.5% of those eligible were referred to, particpated in or completed the programme. Thus 99.5% of those eligible (overweight 7-13 yr olds) didnt. Scale up is a key issue.
- referrals, n = 18,289
- starters, n = 13,998
- completers n = 8311
- on average, BMI reduced by 0.76 kg/m2 over the period of the programme (10-week follow-up). Whether this is clinically relevant in the long term is debatable.
- cost per programme was around £4000. The mean cost per starter is £463, mean cost per completer is £773. Scale up is key – 99.5% of overweight 7-13 year olds didn’t get.
- Adult obesity
- The clinical effectiveness of long-term weight management schemes for adults
- NICE evidence update – Review 1a 11 February 2013
- At 12 to 18 months, the meta-analysis showed a statistically significant effect of BWMPs on weight loss when compared to control (mean difference -2.59 kg, with 95% confidence intervals (CI) -2.78 to -2.41). continues, though attenuated out to 36 months – where the mean difference was -2.21, 95% CI -2.66 to -1.75).
- there were few trials of UK-based weight loss programmes so the conclusions are tentative.
- The quality of the evidence is not strong. The quality of evidence on cost effectivness is weak.
- It is likely (though I haven’t calculated it) that the population impact is negligible
Recent work by Carl Heghnegan has highlighted that 10 in 1000 people referred to weight management (Weightwatchers in this case) attain and maintain their target weight out to 5 years. The population benefit of such intervention and scalability is debatable.
Recent summary of NIHR systematic reviews[iv]reinforced the view that individual weight loss and/or educational interventions are the least effective ways to reduce socio-inequalities in obesity, whereas community and environmental interventions were the most effective. For adults, the review found that primary care delivered weight loss programmes targeted at low-income groups can have positive short term effects (up to 9 months) but these are not sustained over the longer term
Smoking cessation remains one of the most cost effective ways to spend money for health gain. However there are more powerful population level interventions that have a far bigger impact on smoking population prevalence (as opposed to simply encouraging quits). For example overwhelming evidence that introduction of smoke free laws has had substantial impact on population level MI rate[v], respiratory admission rate.
Witness the fierce opposition (ideological and commercial) to plain packs, continued opposition to tax and price initiatives, and continued opposition to any effort to further limit smoking in public places (de-normalisation of smoking) – this gives strong clues as to the views of tobacco industry on the most effective tobacco control policies.
This is carefully documented by Glantz, Chapman and others in an international context.
Arguably the Cinderella of behavioural risk factors. Over consumption leads to avoidable harm, morbidity and mortality. There is a personal and social / population cost of this. Alcohol is an important modifiable risk factor for hypertension (and the downstream CVD consequences such as stroke), it is a key risk for liver disease (watch out for this being a big deal in the not too distant future) and a risk for psychological harm, social harms associated with violence.
Addressing the preventable is important in averting future and current cost. Agreeing the area of focus is interesting. Should we focus on psychosocial intervention[vi]such as? or pharmacotherapy.
Which pharmacotherapy – most seem marginally effective at very best[vii], and are arguably medicalising a social problem.
Expensive pharmacotherapy[viii] is simply not affordable on the scale it would need to be used to effect population good.
- interventions in diabetes prevention
options for preventing diabetes
NNT = 12 or so (based on this study) / cost = £5.5k per case/ so got to operate on 12, at £5.5k each to prevent one incident case at whatever timeframe was quoted in the study
seems pretty durable over time
maybe only option for the morbidly obese. See above.
ICER – ?
some efficacy profile NNT = 9 or so (from memory)
cost = £30odd
but major issues with medicalising vast populations…….
tremendously beneficial. NNT = 6 or so. cost = ??? don’t know. ICER of c£4k per QALY or so
major issues with implementing this at scale – we are bad at this form of behaviour change intervention in NHS I feel (again – happy to be proven wrong).
Would need to be done on a massive scale to effect population change.
Issues re long term effectiveness???
big P population policies
- Nanny etc
- cant express effectiveness in same way
- arguably more about overweight prevention than diabetes prevention
- probably considerably more beneficial and efficacious (population wise) in the medium to long term
- issues re economics and political acceptability of this….
4.6m people in Ireland
a conservative scenario: reductions in dietary salt by 1 g/day, trans fat by 0.5% of energy intake, saturated fat by 1% energy intake and increasing F/V intake by 1 portion/day. approximately 395 fewer cardiovascular deaths per year – 10% reuctuction in CVD mortality.
a more substantial but politically feasible scenario: reductions in dietary salt by 3 g/day, trans fat by 1% of energy intake, saturated fat by 3% of energy intake and increasing F/V intake by 3 portions/day. CVD mortality could be reduced by up to 1070 deaths/year, representing an overall 26% decline in CVD mortality.
[v] http://tobaccocontrol.bmj.com/content/23/6/471?etoc. Uraguay evaluation of smoke free – an intervention that is free for the health sector that results in 17-% reduction in MI admits
Evidence of cost effectiveness of smoking ban on reduced need for health cre (100 fold savings compared to investments)
[vi] NICE CG on alcohol interventions
[vii] AHRQ Evidence summary on pharmacotherapy for alcohol use
[viii] Nalmefene TA (NICE)