HEPRO survey reveals similar patterns in 6 countries on clear social differences in long-term sickness absence

One of the important target themes of the HEPROGRESS project is the prevention of health related exclusion and marginalisation from the labour market and other important spheres of social life.  Many studies have shown that long term sickness absence can be a precursor of future marginalisation or social exclusion. Therefore analyses of sickness absence can contribute to a better understanding of the factors, which are active in the marginalisation processes, and therefore can be subject to preventive interventions.

 This working note presents some overall findings form the analyses on the topic elaborated on the basis of the results from HEPRO survey. The data was collected 2006 and allow for a range of analyses of social disparities and inequalities in health, determinants of health, and consequences of poor health[1]. Citizens aged 15+ in selected municipalities in Norway, Poland, Lithuania, Latvia, Estonia and Denmark took part in the survey.

 It should be noted that the samples used in the analysis are not aimed to be nationally representative samples, as the primary sampling unit were municipalities. But it is assumed that the differences found between countries can be generalised, with a possible exception of Lithuania, as only one municipality from Lithuania participated.
The main purpose of the analyses at this stage is to examine the potentials of the data for further and more elaborate studies, which can support the targeting of HEPROGRESS activities and developments.

 The analysis consists of two parts. The first part is a simple description of the average number of sickness absence days during the past year among the respondents and the distribution among men and women. Based on these distributions, groups with high levels of sickness absence are defined. For each country the highest decentiles of the cumulated distributions of sickness absence days is calculated.[2] The second part of the analysis is a logistic regression analysis of the effect of age, gender and education on the frequency of relative high level of sickness absence (those 10% of the working population with the highest number of sickness days.

RESULTS
According to the HEPRO study, the percentage of those who reports no sickness days during the last year varies considerably: from app. 40% (Denmark) to 65% (Poland). On the other hand, the level of reporting many sickness days is lowest in Denmark and highest in Lithuania and Estonia, whereas Latvia and Poland exhibit a very similar medium pattern. Norway has a pattern similar to Latvia and Poland in the lower end of the range, whereas it exhibits the same pattern as Estonia in the high end of the range of sickness days.

The pattern also varies between men and women and between the age groups – with different patterns in different countries, see table A in the annex (Age and gender specific distributions of sickness absence days).

The table below show the average number of sickness absence days during the last year in age and gender groups days in the participating countries. The group aged 65+ is not included here, as the proportion being occupational active is rather small, and they will tend to be atypical healthy compared to those in same ages not working any more.
The table shows large variations between the countries, and between the age groups.
Many country specific factors together with individually connected factors interplay resulting in these variations and country specific levels of sickness absence rates. The country specific factors can be the level of social benefits received during sickness absence, quarantine rules, social class specific and culturally and company specific attitudes and tolerance towards sickness absence, the level of poor health, poverty levels, unemployment rates, the nature of the employee’s psycho-social and physical work environment etc.

Because of these contextual factors the next analyses will not use the absolute number of sickness days to distinguish between groups with many or few sickness days, but a measure of nation specific high or low sickness absence based on cumulated percentile calculations.

The graph below (Figure 1) shows the cumulative frequency of respondents with specified numbers of sickness days. The dots for each of the x-axis points represent the percentage of those who report the number of absence days displayed at the y-axis, or fewer. That smoothes random variation or the peaks at round figures (e.g. 20 days or 30 days).

For further analyses the cut points between high or lower level of sickness absence is chosen as displayed below:
The number of sickness days among the 10% with the highest level of sickness absence has been used as cut point in a logistic regression analysis separate for each country, with age, gender and education as independent variables.

Obtained results are presented in a table Table 3 below. Estimates marked in bold italics are significantly higher than the comparison group (indicated by “1”). In Denmark men have a lower level of long term sickness absence than women (0.67 compared with 1). The difference is statistical significant. For the lower and medium educational groups there is a significantly higher frequency of sickness absence than among those with higher educational levels (1.86 and 1.30 compared to 1). There are no statistically significant differences between the age groups.

CONCLUSIONS
When comparing the different countries, it is striking, that educational differences consistently exhibit clear and significant differences. Part of the explanation is, of course, that disease and poor health is much more common among citizens with low or no education, but other analyses have shown that even when the analyses adjust for this fact – social differences still prevail.

The difference in educational status per se is not an explanation of the differences in sickness absences. The educational status indicates differences in the life chances and living conditions of the bearer of the education, which again contributes to the explanation of the differences in sickness absence.

Education is nowadays a necessary asset in order to get a desirable job with good salary, less negative work environmental exposures, security and other social advantages. Also a high education is often an indication of the persons social and family background. It is still very common - even in countries with financial free access and no material barriers to higher education - that only very few and disproportionately small part of the offspring of citizens with low income and no education, receives a higher education.

In order to translate social differences according to socio-economic and educational status to practical preventive interventions and measures, it is necessary to study the mechanisms through which educational status and differences manifests in health differences.
Further in-depth analyses of long term sickness absence on HEPRO data will focus on: a) the impact of work environmental factors on sickness absence; b) the influence of social network and social capital; c) relation to empowerment; d) participation in social activities and community life.

Annexes