The effect of employment services on employment and health among unemployed persons: an IV approach

Abstract Background In a large city in the Netherlands, professionals of one department of the municipality focus on rapid return to work, even if that work is not directly in line with education, experience or preference of the client. The first aim was to investigate the effects of this work-first strategy of the municipality on entering paid employment and on health among unemployed persons. The second aim was to determine whether these effects differed between persons, depending on age, sex, education, health, and employment history. Methods To identify the effectiveness of employment services of the municipality, a judge leniency instrumental variable (IV) design was used, based on referral behaviour of case workers. In total, 3272 persons who applied for welfare in a large city in the Netherlands in the year 2015, were included in the study. Information on employment services was derived from the municipality. Information on individual characteristics and employment status was derived from register data from Statistics Netherlands. Information on diseases was derived from a medication register. IV regression models and ordinary least squares (OLS) regression models were performed. Results The work-first strategy increased the likelihood of entering paid employment by 27% points (b = 0.26, se = 0.07) and had a positive influence on health by 4% points (b = 0.06, se 0.03) of welfare recipients compared to other employment services. Although standard errors were larger for subgroups, the positive effect of the work-first strategy on employment was driven by relatively young (18-30 yrs) and old (45-65 yrs) welfare recipients, with low and intermediate education level. Conclusions The instrumental variable approach is a valuable approach to investigate the effect of employment services on entering paid employment and health. To improve health of unemployed persons, it is important to promote entering paid employment. Key messages • Active labour market policies, such as the work-first approach, are beneficial for individuals, by increasing employment participation and health, and society, by increasing the employment rate. • The IV approach is a promising approach to study causal relations in a quasi-experimental study. By using large datasets, differential effects across different groups of participants can be studied.


Background:
Common mental disorders (CMDs) are an extensive problem in the society and attributed to stigma. Prevention of CMD at work is advocated but few studies have investigated what kind of preventive actions managers take. We investigated managers' attitudes to depression and two managerial preventive actions (MPA): 'reviewing assignments and the work situation' (MPA-review) and 'taking initiative to talk about depression and anxiety at the workplace' (MPA-talk). We hypothesized that managers' negative attitudes towards depression would be negatively associated with both MPAs.

Methods:
An on-line survey was sent in 2017 to 4737 managers, answer rate 71% (n = 3358), of which 2 899 were included in this study. Negative attitudes were measured through the 12-item instrument ''Managerial stigma towards employee depression'' (scores 12-72), a cut-off at the 3rd quartile was used as an indicator for having negative attitudes. MPAs were measured with two single questions. Negative attitudes to depression were analyzed in relation to MPA-review and MPA-talk using binary logistic regression analysis with adjustments for sex, education, managerial experience and training, lived experiences of CMD, work organizational context and general preventive actions in the organization towards CMD.

Results:
The proportion of managers with negative attitudes to depression was 20%, performing MPA-review and MPA-talk was 50% and 57% respectively. Adjusted for all co-variates, managers with negative attitudes towards employees with depression were less likely to do both MPA-review (OR 0.71; 95% CI, 0.57-0.89) and MPA-talk (OR 0.53; 95% CI, 0.42-0.66).

Conclusions:
Managers with negative attitudes to depression were less likely to take actions to prevent CMD among their employees which confirmed the study hypotheses. The study suggests that initiatives to reduce stigma among managers could be a way forward to prevent CMD at work. Key messages: Stigma to depression hampers managers' prevention of CMD and needs to be addressed.
To increase managers prevention of CMD, managerial training to reduce stigma towards depression is essential.

Background:
In a large city in the Netherlands, professionals of one department of the municipality focus on rapid return to work, even if that work is not directly in line with education, experience or preference of the client. The first aim was to investigate the effects of this work-first strategy of the municipality on entering paid employment and on health among unemployed persons. The second aim was to determine whether these effects differed between persons, depending on age, sex, education, health, and employment history.

Methods:
To identify the effectiveness of employment services of the municipality, a judge leniency instrumental variable (IV) design was used, based on referral behaviour of case workers. In total, 3272 persons who applied for welfare in a large city in the Netherlands in the year 2015, were included in the study. Information on employment services was derived from the municipality. Information on individual characteristics and employment status was derived from register data from Statistics Netherlands. Information on diseases was derived from a medication register. IV regression models and ordinary least squares (OLS) regression models were performed.

Results:
The work-first strategy increased the likelihood of entering paid employment by 27% points (b = 0.26, se = 0.07) and had a positive influence on health by 4% points (b = 0.06, se 0.03) of welfare recipients compared to other employment services. Although standard errors were larger for subgroups, the positive effect of the work-first strategy on employment was driven by relatively young (18-30 yrs) and old (45-65 yrs) welfare recipients, with low and intermediate education level.

Conclusions:
The instrumental variable approach is a valuable approach to investigate the effect of employment services on entering paid employment and health. To improve health of unemployed persons, it is important to promote entering paid employment. Key messages: Active labour market policies, such as the work-first approach, are beneficial for individuals, by increasing employment participation and health, and society, by increasing the employment rate. The IV approach is a promising approach to study causal relations in a quasi-experimental study. By using large datasets, differential effects across different groups of participants can be studied.