A total of 250 French-speaking patients who attended the clinic between September 15, 2010 and December 30, 2010 agreed to answer a questionnaire administered face-to-face before their consultation and were included in the study. A total of 5 patients were excluded (3 refused to participate, 1 had a psychiatric disorder and 1 required urgent care). This survey did not fall into the category of biomedical research (as defined by French law) and did not involve collecting any personal identification data. Therefore, it did not require ethical approval.

The questionnaire contained 62 questions (most requiring closed-ended responses) designed to gather sociodemographic data and included the Mini-International Neuropsychiatric Interview (M.I.N.I.) 12 , which contains questions about major depressive episode the type of healthcare sought, and the reasons for not seeking healthcare during depressive episodes. For major depressive episode, the M.I.N.I. 12 uses a set of ten questions to identify depressive symptoms during the two weeks preceding the interview. The responses were used to calculate an individual score which, with appropriate sensitivity and specificity, was used to identify the individuals with depressive symptoms.

As regards the respondents’ sociodemographic status, we considered their age (four categories), their self-declared nationality (two categories: French and foreign immigrants), their sex, their civil status (four categories: single, married or living with partner, divorced, and widow/widower).

As for their socioeconomic status, we considered their educational level (in two categories: high school and postsecondary), their occupational category (initially using the 8-category INSEE nomenclature: farmers, tradespeople and shopkeepers, managerial staff, intermediary occupation, manual staff, clerical staff, retired and unemployed people. However, for statistical reasons, we used two categories: managerial/professional staff and other), their health insurance status (in four categories: standard social security, universal [which is free healthcare for low-income individuals], government medical assistance, and none).

The respondents were also asked whether or not they had a regular physician and about their perceived social isolation.

To analyze healthcare-seeking, the interviewees were asked whether they had talked to a physician about their depression (past or current episode). If the response was negative, a series of closed-ended questions was used to explore why. Lastly, the physician’s involvement in terms of psychological health was investigated by means of a general question (During your visits, has a physician shown an interest in your morale, mental state, or mental well-being?). Lastly, we asked the interviewees if they would like the physician to ask them about this more often.

The chi-squared test (or Fisher’s exact test for small samples) was used for the comparisons of proportions. Logistic regression models were estimated, along with the odds ratio (OR) and their 95% confidence intervals (95% CIs), for the multivariate analysis of factors associated with depression and healthcare-seeking. Models were estimated separately for men and women, since sex was an interaction factor. The statistical analyses were performed using Stata v. 10 software (StataCorp LP, College Station, Texas, USA).

The study participants (mean age: 45 years; 52.0% men) were generally of low socioeconomic status, 52.4% were immigrants, and 31.2% reported having a postsecondary education. Only 40.0% had a job on the day of the survey, 9.6% had managerial/professional positions, 10.0% were unemployed, and 16.0% were retirees (Table 1). Half of the interviewees (55.2%) had the standard health insurance enjoyed by most of the population in France (i.e., basic insurance with the social security system together with private supplemental insurance), 20.0% had a health insurance plan intended for the poor (CMU, or universal health insurance), 18.8% were insured through a specific mechanism for undocumented immigrants (AME, or public medical assistance), and 6.0% were uninsured. More than a third (37.6%) had no supplemental insurance (and were therefore only partially covered, by the basic social security insurance).

*Comparison with the distribution for the total population of the same sex.

Nearly half (43.2%) of the interviewees reported that they were married or living with someone, and 43.6% were single. Only 32.0% of the interviewees reported that they were living alone, 58.0% said that they were living with a family member, and nearly 9.6% were living with a non-relative. Lastly, 82.0% reported having a regular physician (even if he or she was not always officially registered with the appropriate Department of Health office).

The M.I.N.I. indicated an estimated prevalence of major depressive episodes of 56.7% among the people who were attending the Baudelaire Outpatient Clinic (95% CI=[51.4%-63.8%]). The prevalence was 53.5% (95% CI=[44.8%-62.2%]) in men and 62% (95% CI=[53.2%-70.8%] in women (p=0.17).

In univariate analysis, in men and women alike, not having a managerial/professional position and/or the standard health insurance was significantly associated with a risk of depression (Table 1): 56.7% of the men who were not in a managerial/professional position were depressed (p=0.02) versus 67.6% of the women (p<0.001). A majority (70.0%) of the men who had health insurance for the poor and 64.5% of those who were receiving AME were depressed versus 40.0% of the men with the standard health insurance (p=0.04). All the uninsured women were depressed (as were 87.5% of those receiving AME and 75.0% of those with CMU) versus 50.0% of the women with the standard health insurance.

After adjustment for age, the characteristics associated with a risk of depression in univariate analysis were female gender, not being in a managerial/professional position (p<0.001), a low educational level (p=0.05), being single (p=0.03), not having the standard health insurance (p<0.05) and foreign nationality (p=0.002).

Multivariate analysis in women (Table 2) showed that not being in a managerial/professional position was associated with a five-fold greater risk of depression (p=0.02). Living alone was associated with a three-fold greater risk (p=0.05). Although the association was not statistically significant (p=0.17), the women receiving AME had a five-fold greater risk of depression than those with the standard health insurance.

Multivariate analysis in men (Table 2) showed that the risk of depression was three-fold higher in those with health insurance for the poor (p=0.05) than in those with the standard health insurance, and 10 times higher in those who reporting having a strong feeling of isolation (p=0.02). Unlike in women, living alone or not having a managerial/professional position did not seem to be associated with a risk of depression in men.

Altogether, 61.6% of the women and 53.8% of the men had a regular physician (p=0.01). During the interviews, of the 144 people diagnosed as depressed, only 52.8% reported having talked to a physician about their depression. This seemed to apply more often to women, although the difference was not significant (p=0.07, and the number was small). Only 44.3% of the foreigners had talked to a physician about their depression versus 66.1% of the French nationals (p=0.01). Close to 60% of the participants without supplemental health insurance had not talked to a physician about their depression versus 39.2% of those with supplemental insurance and 33.3% of those with CMU (p=0.02). Lastly, all the participants in a managerial/professional position had done so versus half of those in other occupational categories (p=0.03).

In multivariate analysis (Table 3), only supplemental health insurance and French nationality were significantly associated with seeking healthcare. The participants who had supplemental insurance were twice as likely to have consulted a physician than those who did not (p=0.05), and those with CMU were three times more likely (p=0.02). Lastly, foreigners were three times less likely to have seen a physician than the French nationals (p=0.01).

¹The multivariate analysis included the following variables: age, sex, nationality and health insurance.

²The multivariate analysis included the following variables: age, sex and nationality.

³The multivariate analysis included the following variables: age, sex and health insurance.

Our findings concerning the factors associated with depression are consistent with those of previous surveys in the general population 1 2 3 17 18 19 . Among other things, we found that women are at greater risk for depression, although it should be borne in mind that there may have been some underreporting bias, since men may find it more difficult to admit to states such as sadness, a loss of interest, or a lack of self-confidence 18 . In our study population, the women living alone had a three-fold greater risk of a depressive episode. Surveys in the general population have estimated that a person living alone is 1.5 to 2.4 times more likely to experience a depressive episode than someone who is married or living with a partner (for women and men alike) 3 . More generally, we found that, regardless of their relationship with their cohabitant(s), the participants who were not living alone seemed to be at lower risk for a depressive episode. Of course, in our cross-sectional survey, the significance of such an association (and its causal nature) remains open to discussion, since a depressive state may also have a harmful impact on the maintenance of a conjugal relationship. Likewise, the feeling of isolation (significantly associated with depression in multivariate analysis), may be a cause or a consequence of a depressive state.

We also found that depression was more prevalent in the individuals in the least-advantaged social and occupational categories. The SIRS survey, which was conducted in the general population in the Paris metropolitan area, the same urban context as for our study, found, in 2005, that people who were not in managerial/professional positions were at a two-fold greater risk for depression than the others 14 15 . In our study, this risk was five-fold higher in the women who were not in managerial/professional positions and four-fold higher in their male counterparts. This much greater risk might be explained by the fact that we classified the unemployed (58% of the interviewees) together with those in non-managerial/non-professional positions. This result may therefore reflect the risk of depression in individuals in non-managerial/non-professional positions, but, above all, in those who did not have a job. Indeed, the risk of depression in the participants who were unemployed was 2.7-fold higher than that in those who were employed (p<0.001), this for both genders. This result is indicative of the role of the rupture of social links (in this case, in the workplace) in the onset of a depressive episode. Here, too, it should be stressed that the relationship between being unemployed and depression is not one-way (severe depression can also cause one to be out of work, at least temporarily).

In our survey, as in those conducted in the general population in France 1 2 3 15 16 , the prevalence of depression was higher in foreigners than in French nationals. Such a difference is also observed in other European countries 11 20 . This was particularly so for the women in our study, since close to 80% of the immigrant women were depressed compared to less than half of the French women (p<0.001). Generally speaking, most immigrants report worse perceived general 21 22 or mental 20 health status than that of people in their host country (at least after living in the host country for some time, since, for the new migrants, a health migrant effect is often reported 23 ). This can be explained by certain conditions and circumstances associated with migration (violence, dangerous journeys, 24 25 26 ), conditions in the host countries. In France, among other countries, immigrants more often have jobs with difficult working conditions, less security and lower salaries than their French counterparts for the same level of qualification 21 . The break with their country of origin, their family and their culture, and their social isolation in France (especially in a city like Paris) probably explain this greater risk as well. Indeed, most of the interviewees were immigrants, for whom it is known that acculturation is a major risk factor for depression 27 28 29 , and, moreover, close family ties are important in preventing anxiety/depressive disorders 30 .

Only half of the depressed individuals in our survey reported having talked to a physician about their depression compared to 60% of the people in the general population who had consulted a healthcare professional, as estimated by the Baromètre santé survey conducted in France in 2005. This difference is probably due to methodological differences: the Baromètre santé survey was conducted by telephone and considered healthcare seekers those people who had seen a healthcare professional (of any type) and people who were taking psychotropic medications (whether or not they were seeing a healthcare professional), whereas we only took into account medical consultations in our survey. When asked, “During your visits, has a physician shown an interest in your morale, mental state, or mental well-being?”, almost two-thirds of the interviewees answered “no”! The vast majority (78.7%) of these patients indicated that they would have liked the physician to ask them about this more often, and nearly 20% stated that they would like such questioning to be routine. Contrary to what some physicians might think, reticence in addressing these issues does not stem from the people themselves (including socially disadvantaged individuals attending a free clinic for the poor and/or uninsured). In our survey population, depressed foreigners were less likely to have talked to a physician (44% versus 66% of the depressed French nationals, p=0.01). This disparity was also noted in the general population in the 2005 SIRS survey, in which only 34% of the depressed foreigners had sought out a healthcare professional for depression (versus 52% of French nationals, p=0.03). Regardless of the reason for consultation, foreigners in France generally seek healthcare less often than French nationals, as shown in a 2005 survey by the Department of Research, Studies and Statistics (DREES) of the French Ministry of Health 31 . In other European countries as well, for example Spain and the UK, it has been shown that ethnic minorities use secondary care services less than others 32 33 .