Is screening costeffective

Page 224
Black blue
224 HEALTH PSYCHOLOGY
involve shifting funds from other services. In addition, the ‘inverse care law’ (Hart 1971),
which suggests that those who seek out tests most frequently are often those who
need them the least, when applied to screening, highlights a shifting of finances to the
most healthy individuals in society.
Is screening cost-effective?
The second problem with screening concerns its cost-effectiveness. A cost-effectiveness
analysis involves assessing either how to achieve a set objective at minimum cost or how
to use a fixed resource to produce the best output. In terms of screening, this raises issues
about the objectives of screening (to detect asymptomatic illness, which can be treated)
and the degree of resources required to achieve these objectives (minimum interventions
such as opportunistic weighing versus expensive interventions such as breast screening
clinics). The economic considerations of screening have been analysed for different policies for cervical screening (Smith and Chamberlain 1987). The different policies include:
(1) opportunistic screening (offer a smear test when an individual presents at the surgery); (2) offer a smear test every five years; (3) offer a smear test every three years; and
(4) offer a smear test annually. The results from this analysis are shown in Figure 9.1.
These different policies have been offered as possible solutions to the problem of screening
for cervical cancer. The results suggest that annual screening in England and Wales
would cost £165 million and would potentially prevent 4300 cancers, whereas smears
every five years would cost £34 million and would potentially prevent 3900 cancers.
The problem of cost-effectiveness is also highlighted by a discussion of the OXCHECK
and Family Heart Study results (Muir et al. 1994; Wood et al. 1994). Both studies
indicated that intensive screening, counselling and health checks have only a moderate
effect on risk factors and the authors discuss these results in terms of the implications for
government policies for health promotion through doctor-based interventions.
Fig. 9-1 Costs per potential cancer prevented for different screening policies (after Smith
and Chamberlain 1987)
Page 224
Black blue
Page 225
Black blue
SCREENING
225
The Family Heart Study
The Family Heart Study (Wood et al. 1994) examined the effects of screening and
lifestyle interventions on cardiovascular risk factors in families over a one-year period.
The study involved 26 general practices in 13 towns in Britain and recruited 12,472
individuals aged 40–59 years. The total sample consisted of 7460 men and 5012
women. The practices within each town were paired according to socio-demographic
characteristics and were randomly designated as either the intervention or the comparison practice. Intervention practices were then randomly allocated either to a further
comparison group or to an intervention group. This provided both an internal and
external comparison with the subjects receiving the intervention. All intervention
practices received screening, but only the intervention group of these practices received
lifestyle counselling and follow-up within the one-year period. All subjects from all
practices were followed up at one year. The screening process involved an appointment
with a trained research nurse, who asked about demographic, lifestyle and medical
factors and measured height, weight, carbon monoxide, blood pressure, blood glucose
and blood cholesterol. The subjects in the intervention group also received lifestyle
counselling and repeated follow-up. The counselling used a client-centred family
approach and involved an assessment of the patients’ risk status, educational input and a
booklet for the subject to document their personally negotiated lifestyle changes. All
subjects were then offered follow-up every 1, 2, 3, 4, 6 or 12 months, depending on their
risk status. Outcome was measured at the follow-up in terms of changes in the main risk
factors for coronary heart disease and the Dundee risk score, which is dependent
on serum cholesterol concentration, systolic blood pressure, and previous and current
smoking behaviour. Outcome was compared within the intervention practices, between
the intervention practice and the internal comparison practice, and between the intervention practice and practices in the external comparison group. The results showed
a 16 per cent reduction in overall risk score in the intervention practices at one year, a
4 per cent reduction in smoking, a small reduction in systolic (7 mmHg) and diastolic
(3 mmHg) blood pressure and marginal reductions in weight (1 kg) and cholesterol
concentrations (0.1 m). The results showed no changes in blood glucose levels. In
addition, the greatest changes in risk status were reported in subjects with the highest
risk levels. Although this intensive screening and intervention did result in changes in
risk for coronary heart disease in the correct direction, Wood et al. (1994: 319) concluded that ‘whether these small reductions can be sustained long term is not known,
but even if they were they would correspond only to a 12 per cent lower risk of coronary
heart disease events’. The authors also concluded that the government sponsored health
promotion clinics ‘would probably have achieved considerably less and possibly no
change at all’ (Wood et al. 1994: 319) and that ‘the government’s screening policy
cannot be justified by these results’ (Wood et al. 1994: 313).
The OXCHECK study
The results from the OXCHECK study also produced similarly pessimistic conclusions
(Muir et al. 1994). This study involved 6124 patients recruited from five urban practices
Page 225
Black blue