This is the first study that has looked at the breakdown of the EI group by a number of health measures in Scotland. This study's main variable of interest is employment status, which distinguishes it from other studies that looked at employment status as a confounding factor and not as the main outcome variable [10, 36].
The EI are a significant proportion of the potential workforce and much of the literature have studied them as one single group [22, 25, 26]. This has resulted in inherent assumptions of homogeneity within this group and therefore important differences between the EI subgroups being undetected and policies not always being appropriate. In this study, the permanently sick made up 36% of the EI population, but the largest group were the LAHF (42%). While the permanently sick did have much worse health than all other economic groups, important results were found for the LAHF. This study highlights that the LAHF, while grouped in with the EI due to their association with the labour market, constitute a rather different group and demonstrate similar traits to the employed and unemployed.
Only more recently has the effect on mental health and psychological wellbeing been investigated in the EI population, however unlike our study they have been treated as one group . Our study revealed a relatively better mental status in the LAHF with respect to the remaining EI subgroups which was consistent to a recent English study using the Adult Psychiatric Morbidity Survey 2007 . Similar to our study, Ford et al. (2010) examined mental health status in the EI by breaking down the group . However, their breakdown was based on benefits received , and may therefore inadvertently exclude many of the EI, in particular the LAHF who do not receive benefits. The majority of the LAHF group in our study were women. This gender disproportion may provide one explanation for our finding. Women tend to carry out the greater share of domestic responsibilities and therefore, when unemployed or LAHF, their multitasking overload may decrease, which in turn could reduce stress and attenuate the risk of poor mental health .
By investigating lung function we were able to include a measure of physical health in addition to perceived health. Our study showed that being in work was not a predictor of good lung function, as the LAHF showed no difference from the employed, whereas the unemployed had slightly worse lung function than those in employment. The results we obtained for the unemployed were comparable to a previous study, where unemployment was associated with increased susceptibility to upper respiratory infections .
The permanently sick, LAHF and unemployed are more likely to be smokers than the employed group which is consistent with the findings of a longitudinal study which showed that men who experienced unemployment were more likely to smoke . Also the non employed do not have the workplace smoking restrictions which have been shown to decrease smoking prevalence . In this study all groups within the EI population were less likely to exhibit heavy alcohol consumption compared to the employed. Economic reasons and under-reporting consumption levels  may provide possible explanations as it has been estimated that population surveys can underestimate true levels of alcohol consumption by approximately 50% in Scotland . This is in contrast to a previous study in Scotland that found that being unemployed, retired or EI was a significant predictor of alcohol-related hospitalisations compared to those in employment . However alcohol-related hospitalisations are a much later event than measuring current alcohol consumption.
The impact of unemployment on health can be modified by a number of socio-economic factors [7, 30]. The role of educational attainment on health and health behaviours is not fully understood. Although potential mechanisms have been proposed [10, 41], this study seeks only to describe educational attainment by economic group. The permanently sick showed much poorer educational attainment. The LAHF had a higher proportion with no qualifications than the unemployed, but were more alike in all other educational categories. This lack in educational qualifications in the permanently sick and LAHF, aside from health issues, may be a consequence of socio-economic circumstances or caring responsibilities in early adulthood, which inhibit them from progressing beyond school qualifications.
Although employment, education skills and training are two of the six domains used to derive the SIMD, the comparison of employment status by SIMD reveals interesting findings. The highest proportion of the permanently sick live in the most deprived areas, yet the LAHF are more evenly distributed across the five quintiles but still live in more deprived areas with respect to the employed. This further illustrates the fact that the EI are not a homogeneous group and reinforces the need to examine their characteristics and needs separately. This divide in the distribution of the EI subgroups by SIMD category can partly be explained by the fact that many of the LAHF are in this group by choice, may be financially secure and not dictated there due to ill health.
Separating the EI group has shown important differences in health. A review study has shown that ‘employment’ is overall good for health and wellbeing . However, the available evidence is mainly about paid employment. But work is much more than this. Work includes unpaid and voluntary work, education and training, family responsibilities and caring [7, 30]. The fact that in this study the LAHF show better health outcomes than the permanently sick but roughly similar to the unemployed demonstrates the benefits of engaging in purposeful activity.
There are limitations of the data used in this study particularly with the LAHF group. The LAHF are predominantly women and although we have discussed the considerable heterogeneity of this group this study only looks at individual status and is not able to consider their marital/partnership status, if they live in a household where their partner’s income is sufficient enough to make their involvement in the labour market unnecessary or if they are lone parents unable to work because of unaffordable childcare costs. Another limitation is that we used data from 2003 which predates the current recession. Evidence suggests psychological ill health, LLI and poor SRH all appear to increase during economic recessions but there is also the suggestion that health behaviours may improve especially tobacco and alcohol consumption . The size of the EI group is therefore more likely to be larger and the more recent Scottish Health Surveys (the survey became continuous in 2008 and now report annually) could now be used to investigate health effects of the recession.
In the past most studies have focussed on the employed and the unemployed. The employed are often the target of health promotion, health protection and vocational rehabilitation efforts  and the unemployed also have compulsory employability and skills training in order to receive benefits. Previously the EI were often a largely ignored part of the labour market. However, there have been major changes in the welfare system in the UK and Scotland and some attempts are being made to assist these groups. Changes in government policy have seen the introduction of measures aimed at helping the permanently sick back into the labour market (e.g. New Deal for Disabled People, Pathways to Work). Further, with the introduction of ESA in 2008, many of the permanently sick now have to engage with employability and rehabilitation services [33, 34], though this reform of the sickness benefit system signifies a dangerous political shift in how ill and disabled patients are seen as either ‘deserving’ or ‘undeserving’ of state support . Similarly, a substantial proportion of EI lone parents are also being targeted by policy reform . Also many of the LAHF group who do not claim benefits will not be required to engage with these new employability programmes and while many may not want to work, improving the health and education/skills of the LAHF group does needs consideration, as this group have an important role of educating and maintaining the health of their own children.
The findings of this study have important policy implications for the health strategies focused on the WAP and in particular the EI. Similar to other studies [27–30] our findings further indicate that the EI group is at higher risk for ill health. They are part of the potential workforce and should not be ignored in public health policies aimed at improving the health of the WAP. Returning to work has proved beneficial for recovery from LLI and poor mental health regardless of socio-economic circumstances [22, 29] and therefore national vocational rehabilitation strategies could help a proportion of the EI become work ready, improve their health and alleviate health inequalities. Even though there are programmes in place in the current policy framework [33, 34], there are still people within the EI group who could benefit from better tailored services and available research evidence suggests that a ‘health first’ approach to welfare reform is potentially the most effective . A better health status of the entire WAP would benefit both the individual and society.