This was a cross-sectional, retrospective, descriptive study over a five-year period, from 1 January 2017 to 31 December 2021.

Our study population consisted of women hospitalised with an ICD-10 diagnosis in the psychiatric department of the CHNU de Fann.

The study concerned all women hospitalised for psychiatric reasons and registered during the period January 2017-December 2021.

Women hospitalised with medical records that did not provide information on:

· biography and family dynamics to record a significant life event;

· epidemiological data such as age and locality.

The ward register enabled us to identify the various patients hospitalised during our study period. We then went to the various divisions, of which there were 7, to collect the medical records.

We drew up a data collection form to fill in the data relating to the sociodemographic profile, i.e. age, current address, marital status, ethnicity, religion, level of education, occupation and use of psychoactive substances.

The results are presented in the form of averages and standard deviations for quantitative variables and percentages for qualitative variables. Most of the graphs were produced using Excel 2016.

We were able to carry out this study with the permission of the doctors in charge of the various divisions of the psychiatry department. The data were collected anonymously and kept confidential.

The mean age of the patients was 37 years, with a standard deviation of 13 and extremes from 16 to 74 years. The median was 35 years. The age groups [21 - 30] and [31 - 40] were the most represented, with 27.9% and 28.6% respectively (Figure 1).

Most patients live in the Dakar region (79.9%) (Table 1). Ten of the fourteen regions of Senegal were included in our study: Dakar (79.9%), Thiès (4%), Diourbel (3%), Louga (3%), Saint-Louis (2.7%), Matam (2%), Kaolack (1.7%), Fatick (1%), Tambacounda (0.7%), and Ziguinchor (0.2%) (Table 1). 1.6% of our study population came from other countries such as Gambia (0.7%), Mauritania (0.5%), Guinea Conakry (0.2%) and the USA (0.2%) (Table 1).

Most patients were unmarried (56%) [single (32%), divorced (20%) and widowed (4%)] compared with 44% of married women (Figure 2).

The marital regime was only specified for 104 of the married patients. The monogamous regime was in the majority in our study: 56% compared with 44% for the polygamous regime.

Our study population was predominantly Wolof (33.3%), followed by Haalpular (22.9%) and Serer (11.4%). The ‘other’ category was represented by other ethnic groups in Senegal (the Mandingues, the Laobés, the Mandjagues, the Mankagnes, etc.) and foreign ethnic groups (the Soussous from Guinea Conakry, the Hadjaraïs from Chad, the Kabyes from Togo, the Outats from Mauritania, the Sénoufos from Côte d’Ivoire, the Tékés from Congo and the Yorubas from Nigeria). In our study, ethnicity was not provided for 45 patients (11.2%) (Figure 3).

Most patients were Muslim (91%). The remaining 9% were Christian.

In our study, school attendance was recorded for 344 patients (86%). The women attended French schools in 74% of cases. Only 5% of patients did not attend school (Table 2).

- Level of education at French school:

In our study, 32% of patients had reached university level and 26% had reached primary level. Junior and secondary education represented 22% and 21% respectively (Figure 4).

In our study, most patients were unemployed (67%).

Of the unemployed patients, 39% were housewives. Pupils and students accounted for 12% of patients. 14% of these patients were not in employment (Figure 5).

A total of 132 patients (33%) were in employment (Table 3). Occupations included administrative staff (17%), teaching staff (12%), domestic staff (8%), tailor (8%), nursing staff (6%), tradeswomen (5%), hairdressers (5%), restaurant staff (2%) and hotel staff (2%) (Table 3).

Hairdressing salon staff (5%), restaurant staff (2%), hotel staff (2%) (Table 3).

The notion of mourning was found in 214 patients (30.2%). The birth of one or more children marked the lives of 184 patients (26%). Divorce was noted in 103 patients (14.5%). Marital conflicts were experienced by 94 patients (13.3%). Fosterage was noted in 46 patients (6.5%). Other significant life events included remarriage (3.7%), partner abandonment (2.4%), arranged/forced or early marriages (1.7%), job loss (1.4%) and sexual abuse (0.4%).

In our study, 80% of patients did not use psychoactive substances (Figure 6). 20% of patients had used psychoactive substances: 8% for tobacco, 6% for alcohol and 2% for other substances such as opiates (morphine, codeine, tramadol), cocaine, benzodiazepines, and hard drugs (Figure 6).

The number of psychoactive substances taken by the patients varied from zero to six. Patients who used only one substance had a rate of 4.3%; 2.9% used two (Table 4).

The mental disorders concerned the following ICD-10 pathology groups [F00 - 09], [F10 - F19], [F20 - F29], [F30 - F39], [F40 - 48], [F50 - F59] and [F60 - F69].

In our study, the mean age was 37 ± 13 years, with a median of 34 years. Most patients were between 31 and 40 years of age (28.6%). These results agree with those found by Coulibaly and al [13] and Agbir and al. [14] who respectively obtained a mean age of 35.3 years ± 12.8 and 31.78 years ± 11.65. Ouédraogo and al. [15] and Karfo and al [16] also found almost identical results, with a predominance of young women.

In Senegal, the national average age is low at 23.2 years, reflecting a predominantly young female population [17] , as the number of women is higher than that of men. This average may explain the predominance of young women in our study. This result may be evidence of the great vulnerability of young subjects to mental disorders. The slight difference between the 2 average ages could be explained by the delay in seeking care, which can last for years. More often than not, patients or their families consult traditional therapists first, until treatment fails and their financial resources are exhausted. This deviation from the therapeutic circuit means that they arrive in hospital years after the onset of the disease.

Women over 65 accounted for 4.7% of all patients. This result is in line with the national rate.

In our study, we noted a predominance of unmarried patients with a percentage of 56% (single, divorced and widowed) compared with 44% of married patients. Our results are like those obtained in the study by Koundoul and al. [6] where single women predominated with a rate of 59.7%. The studies by Agbir and al [14] from Nigeria and Ouédraogo and al [18] from Burkina Faso also show a predominance of single women with respectively 63.3% and 59.41%. In Kenya, we also found a study in which single women were in the majority (65.1%) [19] . Saarento and al [20] , in Finland, found that unmarried women outnumbered married women (63.3%). However, another study by Ouédraogo and al [15] found a higher rate of married women (56.1%).

In contrast to our study, the rate of single women in Senegal in 2013 (43%) was lower than that of married women in the country [17] .

Marriage, which is considered a protective factor against mental illness by Scott and al [21] , may explain the increase in the number of single women in our study population. In fact, they show that marriage has a positive effect on an individual’s mental health by reducing the likelihood of most mental disorders such as depression, anxiety and drug addiction.

The rate of divorced women in our study population is 20%, a result that is clearly higher than the national rate of 3.3% [17] and the rate of 3.21% found in the studies by Salifou and al. [22] . The study by Agbir and al. [14] , which is more in line with our results, obtained a rate of 22.3% of divorced women.

Nowadays, divorce has become a recurrent phenomenon due to the modernisation of society and women’s knowledge of their rights [23] . Contradictorily, many divorces are not registered with the civil registry, hence the persistently low national rate. Judicial divorce is considered a failure and a disgrace by women and their families [23] . As a result, they remarry at a very early age [23] . However, our results lead us to believe that mental disorders mean that divorce is more tolerated among these women and constitute an obstacle to a probable remarriage.

The matrimonial regime was specified for 104 of the married women and 56% of these women were in a monogamous regime. The national rate of monogamous marriage is the same as in our study [24] . In the studies by Ouédraogou and al. [15] and Salifou and al. [22] , monogamy was in the majority, with 74.4% and 71.4% of married women respectively.

The French school enrolment rate in our study was 74%. Karfo and al. [16] obtained similar results, with 73.25% of women attending school. In contrast, Ouédraogo and al. [15] found an enrolment rate of less than 50%. The national enrolment rate for girls is between 50% and 60% [25] .

University level was achieved by 32% of women. In the studies found, women often did not go beyond secondary school [13] [15] [16] . Our population, which is predominantly urban, explains the high rates we found. These results may also highlight the success in urban areas of campaigns to enrol girls in school, such as the SCOFI project (government programme for the enrolment of girls), set up by the Ministry of Education in 1994.

Despite the high rate of school enrolment and the high level of education achieved by most of the patients in the study, 67% of them are unemployed. Studies carried out in Mali, Burkina Faso and Finland also found higher rates of women with no income [13] [15] [16] [20] .

Housewives were in the majority among the unemployed (39%), followed by women not in employment (14%). These results are comparable with those of Coulibaly and al. [13] . Mental illness is a cause of instability, loss of skills and reduced ability to carry out certain responsibilities [26] . It is therefore an obstacle to obtaining a job or keeping a job.

Studies show that psychosocial factors are implicated in the onset of mental disorders. Mourning was the most common factor in our study (30.2%). Studies by Parkes [27] found rates comparable to those in our study, with 46.2% and 33% respectively. A lower rate (5.05%) was found by Salifou and al [22] . Mourning is a trigger factor in the onset of mental illness [28] . When mourning is unhappy, it becomes pathological bereavement, which is characterised by the onset of a physical or mental illness during the bereavement period [27] . The various types of pathological mourning are characterised by the exacerbation of affects such as guilt, depression, loss of self-esteem, identification with the deceased, and sometimes even delusions [27] .

Marital conflict affected 13.3% of the women in our study. Slightly higher rates were found in the Salifou and al. study [22] , of the order of 25.69%. The study by Ouédraogo and al. [15] found a lower rate than ours (8.6%). Marital conflicts can have a major impact on the psyche and psychiatric disorders can take various forms. According to Lamarre and Lussier [29] , in some couples the conflicts can be so intense, so frequent, so violent, so insidious and so little or poorly resolved that the spouses suffer the consequences for their mental health. According to the latter, the disorders most frequently observed are firstly emotional problems, then depression and finally personality disorders.

On the other hand, Menick [30] has shown that psychiatric and/or psychosomatic manifestations are the preferred mode of expression of these conflicts in Senegal. Depression is the main mode of decompensation [31] . This led Gueye in 1995 [12] to say that “everything is happening as if crazy behaviour is the only way out for this woman who, under normal conditions, has no right to speak”.

A notion of trust was found in 6.5% of the women in our study. Trust is an ancient practice that is deeply rooted in our society. In their study on trust, Camara and al. [31] found that women are more trusting than men.

Arranged/forced or early marriages were found in 1.7% of patients, a much lower percentage than the 21% found in the study by Ouédraogo and al. [15] . These types of marriage are still relatively common in our society [24] . A study carried out in Cameroon showed the emergence of significant emotional overload in the form of sadness, anxiety, pessimism, disgust with life, despondency, and a constant drop in self-esteem, which can lead some women to attempt suicide [32] . The result of these marriages is a devalued social image of women, leading to mental disorders such as depression and anxiety [32] .

Sexual abuse was also reported by patients in our study (0.4%). Garcia and al. [33] found that between 15% and 71% of women worldwide have been physically or sexually abused by a male partner at some point in their lives. Such violence can lead to numerous psychological disorders, in addition to post-traumatic stress disorder [33] . The 1989 Australian National Policy on Women’s Health demonstrated this relationship by stating that “Women who experience sexual and physical violence are more likely to have psychological problems” [34] .

Other studies have shown that victims of sexual violence present more psychiatric symptoms than non-victims, especially depression, anxiety, somatisation, obsessive-compulsive disorders, and paranoid ideation [35] [36] [37] . The study by Winfield and al. [38] showed that recent rape victims were significantly more likely to meet DSM-III criteria for major depression and drug abuse than for other disorders.

In summary, these different life events should lead us to reflect on the link between the existence of stressful socio-environmental factors and the onset of mental disorders in women. As Ferreri and al. [39] have pointed out: “The existence of personal or social event-related difficulties preceding the onset of mental or somatic disorders has long been reported”.