Health behaviours influence life expectancy much more than quality of medical care

Dahlgreen Whitehead Model of Health (1991)

Our medical course is satisfyingly broad beyond the physical science - they even have us considering the sex lives of the elderly (all part of a healthy lifestyle, but lots of chlamydia & lack of partners tends to hold many of them back though, apparently).  All the talk of public health rang a bell in my head from psychology days about the importance of health-promotion and the limits of medical care to help.  So I wondered, well, since part of the reason we're in this is to have an impact, just what impact does medical care have on health, relatively speaking?

One meta-analysis [LINK] indicates that health-related behaviour accounts for 40% of early deaths in the US, genetics 20%, stressful social circumstances (e.g. income inequality, discrimination, lack of social relationships 20% (LINK1 LINK2 - together approx equivalent to tobacco smoking)), physical environment 10%, while early death is influenced only to a limited extent by the quality of medical care (10% of deaths).  The challenging message for us trainee doctors is 'public health is much more important than your hospital interventions'.  So - I briefly reviewed the research to see if these stats hold up.  They seem to!  

1) Research papers about common diseases, their causes, and historic changes in life expectancy 

(give qualitative support to the statistic)

The diseases which cause the greatest number of deaths - UK

Ischaemic heart disease (12%), Dementia & Alzheimers (~10%), Stroke (~6%), Lung cancer (~6%), Chronic lower respiratory infections (~5%) (together account for 41% of deaths)

http://www.ons.gov.uk/ons/rel/vsob1/mortality-statistics--deaths-registered-in-england-and-wales--series-dr-/2014/sty-what-do-we-die-from.html

The risk factors which cause the greatest number of early deaths - Europe

Tobacco smoking - 15% of all deaths; Other cardiovascular risk factors: hypertension, obesity, low physical activity/obesity, high blood glucose, high cholesterol, low fruit & veg intake - 25% of all deaths (together smoking and cardiovascular risk therefore account for 40% of deaths)

http://www.who.int/healthinfo/global_burden_disease/GlobalHealthRisks_report_full.pdf

The diseases which cause the greatest number of years lived with disability (DALYs) - UK

Cancer (14%), Ischaemic heart disease (10%), Depression (8%), Stroke, (6%), Road Traffic accidents (4%), Direct effects of alcohol use (4%) - together account for 46% of years lived with disability

http://apps.who.int/quantifying_ehimpacts/methods/en/smith.pdf

Contributing non-medical factors to disease/death pre-20th century (life expectancy has increased by 25 years across the 20th century):

- people weren't aware that tobacco smoking was bad for your health

- people weren't aware that a balanced diet was required for health

- people weren't aware that exercise was good for your health (idleness)

- widespread slum housing (squalor)

- deaths at work

    

Contributing medical factors pre-20th century:

- infant and maternal mortality perinatally (which can now be resolved by obstetric care and vaccination)

- hypertension (which can now be modified by exercise, smoking cessation and diet, as well as being assessed and controlled medically)

- infection (which can now be controlled by antibiotics)

http://www.cdc.gov/about/history/tengpha.htm

2) Back-of-envelope calculations on Quality of Life and behaviour:

(give quantitative support to the statistic)

NHS - £110bn budget, NICE guidelines indicate acceptable spends vary between £0-30k per QALY (Quality-adjusted-life-year), therefore 11m QALYs saved per year if £10k per QALY is assumed.

https://www.nice.org.uk/advice/lgb10/chapter/judging-the-cost-effectiveness-of-public-health-activities

AND

Sedentary lifestyle, i.e. lack of exercise: 7 QALYs lost per sedentary patient -> 60% prevalence -> 252m QALYs

Smoking: 10.5 QALYs lost per heavy smoker -> 20% prevalence -> 120m QALYs

Obesity: 5 QALYs lost per obese patient -> 25% prevalence -> 75m QALYs

High alcohol consumption: 4 QALYs lost per heavy drinker -> 15% prevalence -> 34m QALYs

Total on these factors - 470m QALYs (may be some double counting, but this is 40x the amount of QALYs seemingly accounted for by NHS interventions)

Prevalence stats 1) http://www.hscic.gov.uk/catalogue/PUB05131/obes-phys-acti-diet-eng-2012-rep.pdf

Prevalence stats 2) http://www.hscic.gov.uk/catalogue/PUB17712/alc-eng-2015-rep.pdf

QALYs cost (from a Scandinavian study) http://www.ncbi.nlm.nih.gov/pubmed/17852988

3) Conclusion:

Based on these data, health behaviours in the UK have a far greater impact on life expectancy and quality of life than does the quality of medical care.  Challenging!  A good incentive to encourage our patients to get active, stop smoking, and lose weight.

Causes of death in the UK (2014) - ONS

Risk factors for death with high income Europe highlighted - WHO (2009):
Smoking - 15% of the 3.8m total deaths
The other cardiovascular factors are combined elsewhere in the research paper to give the 25% figure