Commentary: Predicting and preventing premature mortality

BMJ 1996; 313 doi: (Published 21 September 1996) Cite this as: BMJ 1996;313:715
  1. David P Strachan, reader in epidemiologya
  1. a Department of Public Health Sciences, St George's Hospital Medical School, London SW17 0RE

    Established risk factors for premature mortality may be fixed or modifiable. Prevention in clinical practice naturally concentrates on identifying and attempting to correct modifiable factors such as smoking, hypertension, and hyperlipidaemia. In this context, fixed predictors such as age, sex, and family history of disease may be useful in guiding decisions about whom and when to treat.1 Under the common (but rarely tested) assumption that the relative benefits of treatment are the same for different patient groups, greater absolute benefits may be expected for patients at higher underlying risk of disease. The balance of benefits and side effects may thus depend on the level of fixed risk factors. Pharmacological treatment for mild hypertension, for instance, may be justified at a lower level of initial blood pressure among older patients or those with a family history of cardiovascular disease.2

    What are clinicians to make of the observation by Hole and colleagues that forced expiratory volume in one second (FEV1) is a powerful clinical predictor of premature mortality among both men and women? FEV1 may be regarded as a partially modifiable risk factor, in so far as its rate of decline may be slowed by stopping smoking. However, enquiry about smoking habits is quicker and cheaper than spirometry, and advice to stop smoking should form part of the preventive package offered to all patients in primary or secondary care.

    Intriguingly, Hole and colleagues and others3 have found that the association of FEV1 with premature mortality also holds for lifelong non-smokers, independent of other cardiovascular risk factors. This suggests that FEV1 should be considered as a “fixed” clinical indicator of underlying risk to guide treatment decisions in hypertensive or hyperlipidaemic patients. It is likely, but unproved, that peak expiratory flow rate, which is more cheaply measured than FEV1 but well correlated with it, would serve the same purpose. Spirometric results are conventionally expressed as a percentage of predicted valuse for a given age and height. Short stature itself, however, is a risk factor for early death, and prediction of subsequent mortality is strengthened if age adjusted FEV1 is used, without controlling for height.3 Further work is needed to clarify the interpretation of forced expiratory or peak flow measurements as risk indicators in day to day clinical practice.

    What are the public health implications? Diminished lung function may be a cumulative indicator of environmental influences on mortality or may have direct effects on survival after myocardial infarction or stroke. Forced expiratory volume is influenced both by lung development in childhood and by destructive insults to the lung tissues during adult life. At its peak in early adult life, it is related to both prenatal and postnatal growth4 and thus may be a better integrated measure of developmental influences on survival than adult height.3 Arguably, taking a long term public health perspective, FEV1 should be considered as a potentially modifiable risk factor, its association with premature mortality indicating a plausible mechanism for causal relations between family circumstances in childhood and the chances of surviving through middle age.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    View Abstract