Impact of job insecurity on sexual desire : an exploratory analysis

a Institute of Public Communication and Education (ICIeF), Institute of Microeconomics and Public Economics (MECOP), University of Lugano (USI), Lugano, Switzerland b Institute of Public Communication and Education (ICIeF), University of Lugano (USI), Lugano, Switzerland c Institute for Social and Preventive Medicine, Centre Hospitalier Universitaire Vaudois, University of Lausanne, Lausanne, Switzerland d Institute of Microeconomics and Public Economics (MECOP), University of Lugano (USI), Lugano, Switzerland e Cantonal Health Office, Bellinzona, Switzerland


Introduction
Funding/support: none.

Study population
Data was used from the Swiss Health Survey (SHS), a 5 year cross-sectional population based study and random-digit-dialled telephone and written questionnaire surveys carried out by the Swiss Federal Statistical Office.
The SHS collects data on health care, physical and mental health status, lifestyles and health-relevant practices and behaviours, social and working status and conditions among the permanent resident population in Switzerland.In order to obtain representative data the SHS implements a two step sampling procedure (household level and individual level) stratified by Cantons.The current analysis refers to the SHS carried out during 2002 including 19706 individuals aged 15 and older.T he response rate was 60%.Erosion of the original gross sample of 32868 households was due to invalid household addresses (n = 2039) and refusal/non participation to the survey (n = 11123).Nevertheless, according to Swiss Federal Statistical Office, the final SHS sample is representative of the Swiss non institutionalised population [8].
For the purpose of the current study, a further subsample was selected targeting only employees working full and part-time aged 20-64 years (n = 8043).Of these, 796 individuals (9.9%) did not answer the first or the second (or to both) main questions of this study, 579 individuals (7.2%) did not answer the question on job insecurity and 266 individuals (3.3%) did not answer to the question on sexual desire.The analysis refers to a final sub-sample of 7247 cases.

Measurements and statistical analysis
Exposure and outcome variables were measured with two self reported questions.Levels of job insecurity were assessed by the question asked during the telephone survey: "Are you afraid of losing your present job?(a) not at all; (b) rather no; (c) rather yes; (d) yes a lot".Levels of decrease in sexual desire were checked by the following question included in the written questionnaire survey: "During the last two weeks have you experienced a decrease of sexual desire or a total absence of it?(a) very often; (b) sometimes; (c) never".
In order to show a possible relationship between perceived job insecurity and levels of sexual desire, crude prevalence rate for each sex and age group (20-49 versus 50-64) were firstly calculated.The significance of the association was assessed by the chi-square test and by Gamma coefficient.Then the bivariate associations of decrease in sexual desire with the principal independent variable and the selected covariates were calculated.The strength of the association between the exposure variable (job insecurity: Not at All + Rather Low versus Rather High + High) and the outcome variable (decrease of sexual desire: Never versus Sometimes + Ve ry Often) was measured by performing logistic regression models including covariates that were significantly associated with a decrease of sexual desire.
The 95% confidence intervals (CIs) were calculated for each OR.p values are two tailed.Statistical analyses were performed with the SPSS 15 software package.

Results
Ta ble 1 shows the crude prevalence rates of job insecurity and sexual desire for the total sample, and by sex and age groups.In Switzerland, 11.5% of employees have a high or quite high fear of job loss while 35% mentioned a decrease in sexual desire.
For men only, rates of job insecurity increased significantly (p <0.001) according to age (from 11 to 15%).In contrast, prevalence rates of loss of sexual desire significantly increased (p <0.001) with age for both men and women.In particular, compared to the same age groups, women have a higher prevalence of loss of sexual desire than men ["sometimes/very often": women, age group 20-49: 39.6% / men: 25.1% (p <0.001); women, age group 50-64: 46.1% / men: 39.6% (p <0.001)].
Ta ble 2 shows the crude prevalence rates for each level of decrease in sexual desire by levels of job insecurity for the whole sample.A significant positive dose-response association between these variables is confirmed by Chi-square test (p <0.001) and Gamma Coefficient (p <0.001).
Ta bles 3 and 4 refer to the crude prevalence rates stratified by sex and age groups.
For men, a dose-response relationship be-  tween job insecurity and decrease of sexual desire only exists for the age group 20-49 years (Chisquare: p <0.001; Gamma Coefficient: p <0.001), while the relationship is not significant for the age group 50-64 years (Chi-square: p = 0.865; Gamma Coefficient: p = 0.279).
Ta ble 5 shows, for men and women, the bivariate associations of decrease in sexual desire with the principal independent variable and other covariates.
men aged 20-49 years, decrease in sexual desire was significantly associated with depression, fear of job loss and hypertension, while in the age group 50-64 years only depression was associated with the outcome variable.
For women aged 20-49 years, depression, job insecurity and civil status were associated with decrease in sexual desire.In the age group 50-64, in addition to fear of job loss and depression, profession and body mass index were significantly associated with the outcome variable.Ta bles 6 and 7 refer to the results of logistic regression analysis according to sex and age groups.
For men, a significant odds-ratio of 1.53 (CI 1.16-2.01)only exists for the age group 20-49 years, meaning that workers perceiving high levels of job insecurity are exposed to a 53% higher risk of having a decrease in sexual desire compared to the reference group (table 6).
For women, 20-49 years old respondents, who stated they were afraid of losing their job, 47% were more likely to have a significant decrease in sexual desire with respect to those without problems of job insecurity (OR: 1.47; CI 1.13-1.91),while in the logistic regression analysis performed for the age group 50-64 the exposure variable is no more associated with a decrease in sexual desire (table 7).
According to these findings, job insecurity is therefore a relevant risk factor with respect to decrease in sexual desire only for employees in the age group 20-49, regardless of gender.Moreover, the difference of the strength of the association between the exposure and the outcome variables show that fear of job loss has a slight potentially greater negative impact on men than on women.
Regarding covariates included in the logistic regression models, it is important to point out that the variable "depression" plays an important role on sexual desire regardless of gender and age.The strength of the association is always even greater than the association to job insecurity.On the contrary, other covariates are more age and gender specific: in the age groups 20-49 years, hypertension is a relevant risk factor for men, while for women civil status plays a significant role.Women: logistic regression models of predictors of decrease in sexual desire.

Discussion
Although there are some methodological weaknesses (cross-sectional analysis, truthfulness of self-reported intimate behaviour during a questionnaire survey), to our knowledge this is the first population based exploratory analysis which provides statistical evidence of an association between job insecurity and sexual desire.The results show that an increasing fear of job loss is associated with a deterioration in sexual desire among employees aged 20-49.The robustness of these findings seems to be confirmed by the substantial reduction of the potential selection bias due to the low rates of non answer to both the self reported questions regarding job insecurity and decrease in sexual desire.In particular, although this last question would be considered highly unacceptable in some countries, in Switzerland the non response rate was only 3.3%.Furthermore the robustness of the association between the exposure and the outcome variables is also sustained by the low levels of missing data in the logistic regression models performed for people aged 20-49 years that are respectively 1.2% for the women's group and 9.1% for the men's group.
Our results are not surprising since emotional stress has been identified as the most important predictor of low sexual desire [22].Thus, a decrease of sexual desire could be viewed as one of the by-products or outcomes of the increase of emotional stress related to job insecurity.Moreover, it is important to point out, as shown by the logistic regression analy-sis, the role played by depression.In fact, other studies have shown that the most prevalent product of fear of job loss is depressive symptoms [1, 2].The question remains to what extent decrease of sexual desire could be viewed as a direct consequence of job insecurity or a by-product of depression [13].In fact employment stressors might affect family functioning via depression [23].
The current results also confirm the role of unemployment status as an important factor among determinants of decrease in sexual desire [24,25].The current data was obtained during the year 2002, a period of low unemployment rate and of good economic growth for Switzerland.The crude rate of fear of job loss among employees, at the time of SHS (see table 1), was 11.5%.At the current period, it has reached a rate of 23% according to a survey carried out in November 2008 by Gallup International among 1000 Swiss citizens [26,27].
These preliminary findings should promote further epidemiological and clinical prospective studies on the impact of job insecurity and work stressors on intimate life and sexual dysfunctions.In fact, although preliminary, this analysis also provides indications that stressful events such as restructuring, downsizing, privatization, mergers, and closures could also affect the family or intimate relationship life of workers who might consequently face an additional health problem.
Correspondence: Gianfranco Domenighetti Università della Svizzera italiana Via Buffi 13 CH-6900 Lugano E-Mail: gianfranco.domenighetti@lu.unisi.ch Bivariate associations of decrease in sexual desire with the principal independent variable and other covariates