Gender-Specific Career Characteristics and Health Trajectories in France: A Long-Term Harm of Unskilled First Jobs, Career Disruptions, and Downward Mobility

Emmanuelle Cambois, INED
Ariane Pailhé, INED
Clémentine Garrouste, Université Paris Dauphine

Due to the well-documented gender occupational career divide, related to the persistent sexual division of labor, key career characteristics are unevenly distributed in men and women: low-skilled first job, downward occupational trajectory, career’s interruptions are overrepresented in women. In this study, we investigated whether these unequally distributed characteristics have an independent and long-lasting impact on men’s and women’s health trajectories. We used the French population survey "Health and Occupational Trajectories" comprising 2 waves (2006 and 2010). We focused on the 45-74 year-old individuals who were present at both waves (n = 5,904). Multinomial logistic regressions assessed whether the impact of past careers characteristics on self-perceived health (SPH) reported in 2006 and 2010. We found that the unskilled first jobs, interrupted or downward careers prior to 2006 impacted the 2006-2010 SPH trajectories in women; downward and stationary trajectories, and shift from self-employed to employed occupations impacted men’s health trajectories. The gender inequalities in first job opportunities and career discontinuity have long term health consequences for women. These results encourage further investigating how policies promoting equity in the labor force could help improving health and reducing the women’s health disadvantage.


Large inequalities in health and mortality are related to both occupational classes and occupational trajectories throughout the life course. Withdrawals from the labor force, unemployment periods, downward trajectories, disrupted careers or work histories of weak ties to the employment have been identified as critical career characteristics that increase health and mortality risks.

Men’s and women’s careers differ significantly in most countries due to gender-specific differences in commitments to work and family. This context induces gender differences in different key career characteristics which are potentially associated with health: qualification at early stage of the career; work interruptions; the direction of the career. These characteristics are interrelated but each of them might have a specific relationship with health. However, it is unclear whether each of the career characteristics independently exposes men and women to health risks. This study aimed at assessing how each of the career characteristics was associated with health trajectories for men and women.


The Health and Occupational Trajectory survey "Santé et Itinéraire Professionnel" (SIP) was conducted in 2006 by the French statistical institute in metropolitan France. The sample was representative of the population of private households aged 19 to 74 years old (76% participation rate) (N=11 016). This survey collected information on current and past health problems, and on current and past occupations, using a retrospective grid and allowed characterizing the respondents’ careers with several indicators. For the second wave in 2010, about 80% of participants were re-interviewed; the questionnaire was identical to that of the first wave, with supplementary questions on events that had occurred since 2006. Our analytical sample was made up of 2933 men and 2971 women aged 45 to 74 years old in 2006 reporting working, being unemployed or retired in 2006. We did not include the inactive men and women (other than retired) to get rid of the healthy worker effect. This restricted sample allowed accounting for long term detrimental effect of past careers based on those who were not yet out of the labor force due to poor health.

Health was measured by the indicator of Poor self-perceived health (SPH): “How is your general state of health?” poor SPH was measured by the answers “poor” and “very poor” (vs. very good, good, fair). Careers were qualified by (1) Qualification at first job. (2) Career types. (3) Career episodes of inactivity. (4) Career span and episodes of unemployment, and part-time work. (5) Current activity status. We account for the activity status reported in 2006: working, unemployed, retired.

We assessed the relationship between career characteristics observed prior to 2006 and SPH in 2010, using multinomial logistic regressions showing the impact of career characteristics on the health trajectories between 2006 and 2010; we derived the corresponding predicted probabilities associated with 1st job qualification, time spent inactive and the career types. Models were adjusted on health problems in the past; Sociodemographic variables. (Education, partnership, number of children ever had, age included). Finally, using nested logistic models, we checked that the effect of past career on health trajectories remained significant event when considering the mediating effect of recent career experience.


Among men and women who reported being in the labour force or retired in 2006 (excluding those reporting other inactive status in 2006 – sickness, housekeeping…), 33% of men and 37% of women reported poor SPH. Regarding career characteristics, men and women differed by a larger female unemployment in 2006. Women were also less likely to have highly skilled first jobs and promotion over their career and more likely to have experienced episodes of inactivity than men.

We identified a long lasting detrimental effect of a number of past career characteristics on self-perceived health: the unskilled first jobs, interrupted or downward careers prior to 2006 led to unfavorable SPH trajectories between 2006 and 2010 in women; detrimental characteristics in men were downward, stationary trajectories, as well as shift from self-employed to employed. Past career affected recent occupational changes, and recent occupational changes were associated with health trajectories. Nevertheless, accounting for these recent changes did not offset the impact of past career on health in 2010. Among men and women in good SPH in 2006, the detrimental effect of a number of past career characteristics remained.

In spite of several limitations, we found that a number of career characteristics, which are over-represented in women in France, have a long-lasting detrimental impact. The over-representation of these career characteristics partly results from gender inequalities in the labour market. Therefore, our results encourage further exploring whether policies and schemes promoting more equity, in particular regarding opportunities at early stages of the career, career continuity and chances of progression might have positive health return.

Presented in Session 1171: Health, Wellbeing, and Morbidity