In this paper we have estimated the potential impact of diabetes prevention programs on labour force participation and income generation among people at high risk of developing diabetes. The results show that considerable benefits, in terms of both additional working years and increased personal income, could be made by introducing either a lifestyle or metformin intervention to prevent diabetes. Up to an additional $113 million in total personal income between the years 1993 and 2003 could have been generated if a lifestyle intervention was implemented in 1983. While this amount is small relative to total income for this age group (income for this age group was $172 billion in 2003--results not shown), for the individuals involved, the impact is considerable.
Our results should be considered a minimal estimate of the impact of diabetes prevention interventions due to the conservative approach taken. Firstly, we have only captured the lost working years and income among those who retire. Among those who continue in the workforce, diabetes is likely to further reduce productivity through a shift to part-time or lower paid work as a result of the illness and its complications [26
]. Indeed Schofield et al.
have shown that those with diabetes have significantly lower incomes than those with no long term health condition regardless of labour force status [11
]. Secondly, the assumptions used for estimating reductions in prevalence were based on conservative levels of participation in the screening program by both GPs and consumers. If participation rates could be increased, there would be greater reductions in the incidence of diabetes with consequent increases in workforce participation. Thirdly, in the case of the lifestyle intervention there are likely to be additional benefits through reductions in obesity and other diseases such as cardiovascular disease and cancer, which have not been considered in our modelling.
A limitation of the approach taken is that we have assumed that the reduction in self-reported prevalence of diabetes from a diabetes intervention would be similar to the reduction in biochemically confirmed diabetes. Although this seems plausible, it has not been validated. A further limitation is that our approach implies a causal relationship between having diabetes and the reduced rate of labour force participation among those with diabetes, relative to those without. It is possible that some other factor is confounding the relationship and may be reducing the labour force participation of those with diabetes. Diabetes, as well as risk factors for diabetes (such as obesity, physical inactivity and smoking) are more common in lower socio-economic groups [4
]. By controlling for education level in the analysis, we aimed to minimise the potential for confounding by other socio-economic factors, or by other diseases which are also prevalent among lower socio-economic groups. However, the possibility of confounding by unknown factors cannot be excluded. Additionally, the possibility of reverse causality cannot be completely excluded--that is, that early retirement has led to development of diabetes. However, the French GAZEL study provides interesting results which suggest this is unlikely. This longitudinal, repeat-measures study over 15 years has found that having diabetes increases the risk of transition from employment to disability and retirement [27
], but that retirement did not change the risk of developing a number of chronic diseases, including diabetes [28
Several other studies have estimated the indirect cost of diabetes attributed to lost labour force participation, both through temporary absenteeism and permanent retirement due to ill health [29
]. A US study estimated lost productivity costs of US$2.6 billion due to diabetes related absenteeism from work, and US $7.9 billion due to unemployment from diabetes related disability in 2007 [33
]. Similarly, the indirect costs of diabetes were estimated to be 70.1 million EURO in Norway in 2005 [31
]. Studies that have analysed the indirect costs of diabetes have mainly used the human capital approach [34
], which is consistent with the approach we have taken in our study. However, these previous studies assessed the cost of illness, and did not consider the financial benefits of treatment. Our study estimates the potential productivity benefits of interventions to prevent diabetes using a microsimulation model to estimate the reduction in the prevalence of diabetes likely to be achieved through these interventions and linking this to models of the impact of diabetes on labour force participation.
The benefit of increased labour force participation as a result of preventing diabetes is likely to extend beyond the individual, and provide some medium-term benefit to government also. Previous studies have shown that in the 45 to 64 year age group diabetics pay 67% less tax than non-diabetics, and receive 112% more in social security payments each week than non-diabetics [11
]. This is in addition to the $989 million that the Australian government allocates to health expenditure for diabetes (based on figures for the 2004-05 period) and the additional $7 billion that is projected to be spent between 2003 and 2033 [7
]. Preventing diabetes is therefore likely to have considerable financial benefits to governments in the medium-term. While, the future costs to government of caring for subsequent illness among people whose death has been delayed are not considered in this analysis, it is the accepted responsibility of government to optimise the wellbeing of the population [36
In the future, the benefits of diabetes prevention programs are likely to be greater, as diabetes is both increasing in prevalence worldwide, and becoming more common among the working-age population, particularly those aged 45 to 64 years [38
]. Keeping greater numbers of experienced, older workers in the labour force is increasingly becoming a priority of governments. Key government reports have highlighted population ageing and labour shortages as potential pressures threatening the Australian economy [17
]. Such a situation is likely to be reflected internationally in most developed countries, with the ageing of the global population [39
The modelling undertaken assumes that the individuals who benefit from diabetes prevention would participate in the labour force at the same rate as other individuals without diabetes, controlling for education. Unlike many industrialised countries, Australia has low unemployment and labour shortages in a number of industries. In 2009, Australia's unemployment rate was 5.8%--close to the accepted 'full employment' rate of 5% [40
]. In his 2011 budget speech, the treasurer emphasised the high employment rate: "Over 300,000 jobs have been created in the past year and the unemployment rate is forecast to fall further, to 4 1/2 per cent by mid 2013, creating another half a million jobs.... We believe our economy can't afford to waste a single pair of capable hands." [41
] It is therefore likely that those who benefit from diabetes prevention would be able to find employment as easily as others.
Early retirement due to ill health is a problem throughout the world, and in Australia it is estimated that 58% of men and 26% of women who retire from full-time work early (that is, before the age of 55 years--from 55 years of age Australian citizens can access preserved superannuation and are entitled to some social security pensions) do so because of ill health [42
]. Maintaining the health of the workforce is seen as a vital step in securing the economic activity of the nation [17
]. The government has promoted deferred or gradual retirement as a solution, and given the numbers retiring due to illness, the prevention and treatment of long-term health conditions may be critical in helping older Australians to remain in the workforce longer [42
]. There is also likely to be a widening gap between Australians who are able to work and those who retire prematurely when they become ill, in terms of both income earned and assets accumulated.