Anthony G Threlfall

Cost effectiveness of shortening screening interval or extending age range of NHS breast screening programme: computer simulation study

Abstract Objective : To compare the cost effectiveness of two possible modifications to the current UK screening programme: shortening the screening interval from three to two years and extending the age of invitation to a final screen from 64 to 69. Design : Computer simulation model which first simulates life histories for women in the absence of a screening programme for breast cancer and then assesses how these life histories would be changed by introducing different screening policies. The model was informed by screening and cost data from the NHS breast screening programme. Setting : North West region of England. Main outcome measures : Numbers of deaths prevented, life years gained, and costs. Results : Compared with the current breast screening programme both modifications would increase the number of deaths prevented and the number of life years saved. The current screening policy costs £2522 per life year gained; extending the age range of the programme would cost £2612 and shortening the interval £2709 per life year gained. The marginal cost per life year gained of extending the age range of the screening programme is £2990 and of shortening the screening interval is £3545. Conclusions : If the budget for the NHS breast screening programme were to allow for two more invitations per woman, substantial mortality reductions would follow from extending the age range screened or reducing the screening interval. The difference between the two policies is so small that either could be chosen.

Reliability of Underlying Incidence Rates for Estimating the Effect and Efficiency of Screening for Breast Cancer

The process of setting screening performance targets requires an estimate of what the incidence of breast cancer would have been in the population invited for screening if there had not been a screening programme. Before the introduction of the National Health Service breast screening programme in 1988 the incidence of breast cancer was already increasing in the population targeted for screening. To establish the incidence before screening the most recent complete data from all the regional cancer registries were collated. An age-period model was constructed to predict what these incidence rates might now have been if the screening programme had not been introduced. The model predicted that if prescreening trends continued (between 1987 and 1995) underlying incidence over this period would increase by 2.3 per 10 000 in women aged 50–54, 2.6 per 10 000 in women aged 55–59, and 2.9 per 10 000 in women aged 60–64. If the prescreening trends have continued then the use of a universal rate, applied across all calendar years and age groups, would seem to be inappropriate when setting targets.

Survival rates from interval cancer in NHS breast screening programme

The NHS breast screening programme invites women aged 50-64 for screening every 3 years. In this programme the term interval cancer is applied to a breast cancer occurring within 3 years of a screening test with negative results. Substantially higher than anticipated rates of interval cancers have already been reported from the NHS breast screening programme, 1 2 and there is conflicting evidence on whether the survival rates of women with interval cancers are different from those of women with breast cancer occurring in an unscreened population. 3 4 Were interval cancers to have a worse prognosis than cancers in an unscreened population, the reduction in mortality from breast cancer in the screened population might be substantially less than predicted.nnTo interpret survival estimates for women with interval cancers requires identification of a suitable group of unscreened women for compari- son. In the context of a national screening programme this …

The appraisal of public health interventions: the use of theory

BACKGROUNDnPublic health decision-making is hampered by inappropriate adherence to underpowered randomized controlled trials (RCTs) which give inconclusive results and lead to decision-makers being loath to recommend interventions with strong theoretical and observational support.nnnMETHODSnWe outline situations in which robust decisions about health interventions can be made without trial evidence. We present a new approach in which theory, causal models and past observations are given proper regard in the decision-making process.nnnRESULTSnUsing our approach, we provide examples where the use of causal theories and observations in areas, such as salt reduction, smoking cessation and gardening to improve mental health, is sufficient for deciding that such interventions are effective for improving health without needing the support of underpowered RCTs. Particularly where RCT evidence is inconclusive, our approach may provide similar aggregate health outcomes for society for vastly lower cost.nnnCONCLUSIONSnWhen knowledge and theoretical understanding are unable sufficiently to reduce doubt about the direction of effect from an intervention, decisions should be made using evidence-based medicine approaches. There are, however, many cases where the combination of robust theory, causal understanding and observation are able to provide sufficient evidence of the direction of effect from an intervention that current practice should be altered.

Breast screening programme: should the interval between tests depend on age?

The NHS breast screening programme was introduced in 1988 following the report of a committee, which recommended that women aged 50–64 years should be screened every 3 years, but also concluded that for different age groups the optimum screening interval may not be the same and that this should have a high research priority. We now describe the frequency with which cancers present between screening appointments (interval cancers) in relation to age at time of screening. Interval cancers provide an early indication of the appropriateness of a screening interval and recent reports show that the overall occurrence of these cancers in the screening programme is substantially higher than predicted, especially in the third year after screening. The occurrence of interval cancers in 79 918 women screened with single-view mammography at three screening units in northwestern England between 1988 and March 31, 1991, is shown in the table. Interval cancers are presented for each year following a negative screen in women above and below the age of 60 at the time of screening. The number of cancers expected to occur in these women, assuming they had not been screened, was estimated by extrapolating from an age-period model, allowing for ageing of the cohort and a small general increase in incidence over a calendar year. The relative risk (observed interval cancers/expected number of cancers in the absence of screening) of a breast cancer occurring within 3 years of a negative screen was 0·523 in women aged 50–59 but was 0·317 in the older age group (xtest; p<0·001). Interval cancer targets for the NHS breast screening programme are based on the results of the Swedish Two County Trial, because this large randomised trial achieved a 31% reduction in mortality from breast cancer. In northwestern England interval cancer rates are unacceptably high in younger women but are substantially lower in older women, and only in this latter age group are similar to those achieved in the Swedish Two County Trial. If our findings are replicated elsewhere and financial considerations continue to limit the number of tests which the NHS breast screening programme can offer then a different scheduling of visits might produce a greater return on the same investment.

The appraisal of public health interventions: an overview

BACKGROUNDnThe approach currently used to appraise public health interventions is close to that of health technology appraisal for drugs. This approach is not appropriate for many public health interventions, however, when extremely small individual level benefits are delivered to extremely large populations. In many such situations, randomized controlled trials with sufficient size and power to determine individual level effects are impractical. Such interventions may be cost-effective, even in the absence of traditional evidence to demonstrate this.nnnMETHODSnWe outline an alternative approach based on decision theory. We apply it to cases where prior beliefs are sufficiently strong and well grounded to allow decision-makers to assume the direction of change of the interventions outcome, within the context of a transparent and deliberative decision-making process. Decision theory also assumes that decision-makers are risk neutral, implying that they should make decisions based on an interventions mean cost-effectiveness, and should therefore disregard variance except when deciding to wait for more information. However, they must allow for biases.nnnRESULTSnA framework is presented which has the potential to achieve large health gains at no additional cost.nnnCONCLUSIONSnThis analysis provides a rigorous theoretical framework for decision-makers in public health. The implied paradigm shift also applies to some clinically based areas.

The appraisal of public health interventions

Abstract Background Existing methods for the appraisal of public health interventions have been borrowed from those developed for the appraisal of health technology. They give precedence to randomised controlled trials (RCTs) and use a two-stage approach: a hypothesis-testing framework to establish an interventions effectiveness, followed by the use of decision theory to establish its cost effectiveness. This approach is not always appropriate for public health interventions, for which RCTs often do not exist, and those RCTs that do are often very underpowered. We propose instead the use of a (Bayesian) decision-theoretic approach for both effectiveness and cost effectiveness. Methods When evidence exists that includes RCTs that are sufficiently powered, our proposed approach will not change existing methods: the approach used in health technology appraisal would remain the same. However, in public health the role of theory in the formation of previous beliefs, as a means of establishing the direction of change of an outcome, becomes an important element in our proposed calculus. Another important element of our approach is that public heath decision makers should be risk-neutral. This aspect of decision theory, which has far-reaching consequences for appraisal, has not received the prominence it deserves, and does not appear to have been discussed previously. Findings Basing recommendations for public health on a restricted evidence base that requires the demonstration of a particular level of significance for its outcomes can lead to the rejection of interventions that would provide cost-effective improvements in population health. We will use case studies to show that interventions with small, non-significant effect size at the individual level, such as the imposition of 20 mph speed limits, can be cost effective at population level. Such interventions would, however, not be recommended with the standard approach of statistical significance from RCT evidence as the primary arbiter of effectiveness. Interpretation The changes we propose show that a decision-theoretic approach has great value for public health, just as it has in many other fields. Funding None.