By the same authors

Economic model of adult smoking related costs and consequences for England

Research output: Book/ReportBook

Author(s)

Department/unit(s)

Publication details

DatePublished - 2011
Number of pages95
PublisherPublic Health Research Consortium
Original languageEnglish

Abstract

Background

Smoking is a major avoidable cause of morbidity and mortality in the United
Kingdom. A recent study suggested that smoking was responsible was 109,164 deaths in the UK during the year 2005-06 (Allender et al 2009). The direct cost to the NHS in the UK due to smoking related conditions was estimated to be £5.2 billion per annum in 2005. The costs were highest in England (£4398.9m, equal to 6.5% of the total health care budget for England), followed by Scotland (£409.4m), Wales (£234.2m) and Northern Ireland (£127.9m).

Doll et al (2004) in their analysis of 50 year observations on British male doctors
found that the age standardised mortality rate (ASMR) per 1,000 men per year for smokers was almost twice that of never-smokers (35.40 versus 19.38). The higher mortality was attributable to higher probability of ischaemic heart diseases, cerebrovascular diseases, chronic obstructive pulmonary disease (COPD), lung cancer and other vascular diseases.

Aims
The aims of the study are:
a) To conduct a literature review of the cost models of adult smoking, and
b) To develop an economic model of adult smoking for England.
This report presents the findings of a literature review of cost models, and the design, methodology and results of the economic model of smoking cessation.

Literature review

A literature review identified studies estimating the health care costs of smoking
related disease. A wide range of studies were identified making point estimates of the annual cost in a static framework. In 2005, the direct annual cost to the NHS in England due to smoking related conditions was estimated to be £4.4 billion, which was equal to 6.5% of the total health care budget for England(Allender et al, 2009). These static models making point estimates are limited in their usefulness, and cannot be used to project potential longer term savings from interventions reducing the prevalence of smoking and consequently smoking related disease.

Few attempts have been made to model the costs of longer term smoking related
disease in a dynamic framework. Dynamic cost models may be less prolific because these projections are particularly data intensive, requiring extensive demographic, smoking prevalence, disease and cost information. Much of this data is not routinely available and has itself to be modelled to generate data in the form that can be used to estimate these long term costs of smoking. In addition to disease and cost data limitations, uncertainty over the timeframe over which to project these costs and the wide range of diseases which may be included make these models difficult to specify.

Design and Methods

The design and methodology of our economic model for adult smokers in England are discussed in detail in the report. The model evaluates four smoking-related conditions, i.e. myocardial infarction, stroke, chronic obstructive pulmonary disease and lung cancer. These conditions are known to have the highest economic and health related consequences associated with smoking. The economic model also takes account of higher risk of mortality from other diseases among smokers and ex-smokers compared to non-smokers. Lifetime costs and consequences were modelled for three population groups, i.e. never smokers, current smokers and ex-smokers. The probability of events in the latter two groups was modelled using relative risk estimates from the published literature. Risk reduction in ex-smokers was modelled as a function of time since quitting smoking. The event probabilities for myocardial infarction pathway are presented in detail.

Results

The model was evaluated using cohorts of 1,000 non-smokers, smokers and quitters by allowing for varying cessation rates from 0 – 100% with intervals of five
percentage points. The model runs separately for men and women and estimates costs for each smoking-related group and the cost savings and life years gained associated with smoking cessation at varying cessation rates. The results estimate that the lifetime health care costs for a cohort of 1,000 non-smokers are £20.7 million (not discounted) or £5.2 million (discounted at 3.5%); the costs for a comparable cohort of current smokers are £29.3 million (not discounted) or £9.3 million (discounted) and for quitters are £24.3 million (not discounted) or £6.7 million (discounted). The figures for women are following: non-smokers (not discounted: £17.5 million; discounted £3.9 million), current smokers (not discounted: £25.4 million; discounted £7.0 million) and quitters (not discounted: £20.5 million; discounted £4.9 million).

We investigated the impact of varying smoking cessation rates on lifetime health care cost savings. The results are presented here: 5% cessation rate in a cohort of 1,000 smokers (men: £246,320; women: £241,365), 10% cessation (men: £492,640; women: £482,730), 20% (men: £985,280; women: £965,459), 30% (men: £1,477,919; women: £1,448,189), 40% (men: £1,970,559; women: £1,930,919) and 50% (men: £2,463,199; women: £2,413,649). Similarly, the difference in the life years lived by the smoker and quitter cohort was estimated as life years gained as a result of smoking cessation. These estimates are summarised here for a cohort of 1,000 smokers: 5% cessation rate in a cohort of 1,000 individuals (men: 213.1 years; women: 226.7 years), 10% (men: 426.2; women: 453.4), 20% (men: 852.5; women: 906.8), 30% (men: 1278.7; women: 1360.2), 40% (men: 1705.0; women: 1813.6) and 50% (men: 2131.2; women: 2267.0). It should be noted that the model may underestimate the costs associated with smoking because it does not explicitly incorporate concurrent risk of multiple conditions. However, the results clearly demonstrate the benefits of smoking cessation in terms of cost savings and life years gained.

The cost of smoking-related diseases and potential cost savings from smoking
cessation were further evaluated for the current prevalent population of England. We estimate that the total lifetime cost of smoking for the prevalent population of
England (35 years of age) is £44.8 billion (or £26.5 billion after discounting at
3.5%). We also estimate that the maximum cost savings as a result of smoking
cessation in this population is £23.3 billion (or £14.7 billion after discounting at
3.5%).

Contribution:

The model enables the user to estimate the costs and life years saved as a result of adult smoking in England. Using these estimated costs and life year gains, we
demonstrate that changes in smoking rates can have an impact upon the cost of
treating smoking related diseases in the population. This model addresses the need for a lifetime economic model of smoking-related costs and consequences that is based on population-specific epidemiological data. We present cost and life year gains as a result of quitting for a cohort of 1,000 individuals in the English population, and also evaluate cost savings for the prevalent population of England.

The model provides a useful ‘bolt-on’ for evaluators, as the cessation rates from
smoking cessation interventions can be inserted into the model to estimate the
potential longer term cost savings following a successful quit. This permits the
evaluator to project health care cost savings using a longer timeframe than the 12
month follow up traditionally employed in the majority of economic evaluations to
date. Therefore a ranking of different interventions driven by effectiveness, health
care cost savings and programme costs, can be constructed in order to demonstrate potential value for money afforded by different strategies.

Discover related content

Find related publications, people, projects, datasets and more using interactive charts.

View graph of relations