Pathways to care and determinants of first help-seeking among patients with psychotic disorders in Northwestern Madagascar: a multicenter cross-sectional study

¹Faculty of Medicine, University of Mahajanga, Mahajanga, Madagascar, ²Department of Psychiatry and Addiction Medicine, CHU PZAGA, Mahajanga, Madagascar, ³Department of Psychiatry, Analakininina Hospital, Faculty of Medicine, University of Toamasina, Toamasina, Madagascar

^&Corresponding author
Raphaël Fidelis Randrianarivo, Faculty of Medicine, University of Mahajanga, Mahajanga, Madagascar

Introduction    

Mental health represents a major public health issue and a major challenge in African countries such as Madagascar, mainly because of sociocultural barriers affecting the acceptability of mental health care [1]. Access to quality mental health services remains limited in many African settings, including Madagascar. Psychotic disorders affect approximately 1% of the global population [2]. Social representations of mental illness play a crucial role in family decisions regarding the care of affected individuals [3].

Worldwide, numerous studies have explored the pathways to care of patients with psychotic disorders or other mental illnesses. The perceived causes of psychotic disorders vary across regions. In Western countries, these disorders are commonly attributed to scientific explanations such as genetic and environmental factors [4]. The organization of mental health services also plays an important role in shaping care pathways [5], often leading to first consultation with general practitioners in cases of mental disorders [6].

In contrast, in many Asian countries, mental disorders are frequently attributed to supernatural or spiritual causes. Traditional healers, traditional medicine, and religious practices are therefore often preferred as first-line care [7]. Similarly, in Africa, individuals tend to seek these approaches first, including in cases of psychotic disorders, with psychiatric consultation often considered a last resort [8]. Madagascar is characterized by 18 ethnic groups and diverse cultures. Local explanatory models of mental illness commonly include witchcraft, spirit possession, and violation of taboos. Traditional healers and religious rituals therefore occupy an important place in the pathways to care for mental disorders [9]. However, delays in medical management have significant negative consequences on the prognosis of psychoses [10].

In Mahajanga, a previous study focused on all mental disorders observed in hospital settings [11]. However, specific data on the pathways to care of patients with psychotic disorders and the determinants of first help-seeking in Northwestern Madagascar remain scarce. We hypothesize that the pathways to care of patients with psychotic disorders are frequently marked by an initial resort to traditional or religious practices. This choice appears to be related to insufficient coordination between psychiatric, social, and medico-social structures, as well as to certain sociodemographic characteristics of caregivers or decision-makers. Therefore, the present study aimed to analyse the pathways to care of patients with psychotic disorders and to identify factors associated with the choice of first help-seeking source in Northwestern Madagascar.

Methods     

Study design and setting

This was an observational, cross-sectional, analytical, descriptive, and multicenter study conducted in four institutions: the Neuropsychiatry Department of the Centre Hospitalier Universitaire Professeur Zafisaona Gabriel (CHU PZAGA) Androva Mahajanga, and three faith-based healing centers (Toby Fifohezana), including two affiliated with the Church of Jesus Christ in Madagascar (FJKM): Toby Peniela FJKM Antanimasaja Mahajanga and Toby Kristy Fahazavana FJKM Ankazomborona Marovoay, and one affiliated with the Malagasy Lutheran Church (FLM): Toby Betela FLM.

Study period

The study was carried out over ten months, from July 1^st, 2020 to April 30^th, 2021.

Study population

The study included all patients presenting with psychotic symptoms who were hospitalized in the Neuropsychiatry Department and/or admitted to the Toby centers and who agreed to participate.

Inclusion criteria

All psychotic patients hospitalized in the Neuropsychiatry Department and/or admitted to Toby centers during mass consultations were included.

Non-inclusion criteria

Non-psychotic patients hospitalized in the Neuropsychiatry Department and/or admitted to Toby centers were not included.

Exclusion criteria

Psychotic patients without an accompanying decision-maker, patients whose accompanying person did not know the history of the illness, and patients or accompanying persons who refused to participate were excluded.

Sampling method

A non-probabilistic consecutive sampling method was used. All eligible patients encountered during the study period were recruited after informed consent.

Data collection

Data were collected using a standardized structured questionnaire. Interviews were conducted with patients and their accompanying decision-makers. The questionnaire was pre-tested and administered by 8th-year medical interns from the Neuropsychiatry Department under supervision.

Variables

Sociodemographic characteristics of patients (age, sex, marital status, educational level, profession, religion, ethnicity). Sociodemographic characteristics of decision-makers (relationship with patient, marital status, educational level, profession, residence, religion, region of origin). Clinical variables (reason for admission, length of stay, diagnosis: acute psychoses or chronic psychoses). Pathway-related variables (first, second, third, and fourth resort; perception of causes; belief in curability; factors influencing choice; perceived advantages and disadvantages of each type of care).

Data analysis

Data entry and analysis were performed using SPSS version 25.0. Categorical variables were expressed as frequencies and percentages. Pearson´s chi-square test was used to assess associations. A p-value < 0.05 was considered statistically significant.

Ethical considerations

Authorization was obtained from the heads of the Toby centers and from the head of the Neuropsychiatry Department. Verbal informed consent was obtained from all participants. Confidentiality and anonymity were ensured. The study respected the principles of the Declaration of Helsinki.

Results     

Study population and recruitment flow

During the study period, a total of 426 patients were followed either in the Neuropsychiatry Department or in the Toby faith-based healing centers. Among them, 122 patients presented with psychotic symptoms. Fourteen patients were excluded (absence of accompanying decision-maker or refusal to participate), yielding a final sample of 108 patients included in the analysis.

Sociodemographic characteristics of patients

The mean age of patients was 29.85 ± 9.68 years, with ages ranging from 16 to 65 years. The most represented age group was 25-34 years (42.6%), followed by 15-24 years (29.6%). Males predominated, accounting for 72.8% of cases, with a male-to-female ratio of 2.6 (Table 1). Most patients were single (78.7%), while 21.3% were married. Regarding occupation, 47.2% worked in the primary sector, 24.1% were students or pupils, and 23.1% were unemployed. Only small proportions were employed in the secondary (4.6%) and tertiary (0.9%) sectors. Concerning educational level, the majority had a secondary level of education, followed by primary level, whereas illiteracy and university-level education were less frequent (Table 2). With respect to religion, 53.7% of patients were Protestant, 22.2% Catholic, 14.8% traditionalist, 5.6% Evangelical, and 3.7% belonged to other religions. The most represented ethnic groups were Sakalava (21.3%), Merina (16.7%), Tsimihety (13%), Antefasy (13%), Betsileo (12%), and Antandroy (10.2%).

Characteristics of decision-makers

Most patients were accompanied by ascendants (65.8%), followed by collaterals such as siblings or spouses (29.6%), and descendants (4.6%) (Table 3). Three-quarters of decision-makers were married (75%), while 25% were single. Educational level was mainly secondary (44.5%), followed by primary (29.6%), illiterate (18.5%), and university level (7.4%). A large majority of decision-makers worked in the primary sector (88%), whereas 11.2% were in the secondary sector and 1.8% in the tertiary sector. Decision-makers were more frequently from rural areas than urban areas. Regarding religion, 50.9% were Protestant, 22.2% Catholic, 16.7% traditionalist, 4.6% Evangelical, and 5.6% belonged to other religions. The most frequent regions of origin were Boeny (25.7%), Atsimo Atsinanana (17.6%), Sofia (13%), and Analamanga (12%).

Clinical characteristics

Most patients were diagnosed with acute psychoses, while the remainder presented with chronic psychoses. The main reasons for admission were behavioral disturbances, agitation, hallucinations, and delusional symptoms.

First help-seeking source

First help-seeking was mainly from traditional healers (56.5%) and religious practices (22.2%). Only 12.0% of patients consulted psychiatric services as a first resort, and 9.3% consulted general practitioners (Table 4).

Pathways to care

Psychiatric services became more prominent at later stages of the pathway, representing the second resort in 39.8% of cases and the third resort in 22.2% of cases. Religious practices remained frequent at the second (38.9%) and third (16.7%) resorts. In contrast, recourse to traditional healers markedly decreased after the first contact (Table 5).

Length of stay at place of care

Regarding length of stay at the place of care, among patients managed in psychiatric services, 26.8% stayed less than one month and 15.7% stayed between one and three months. No patients managed in psychiatric services stayed between three and six months, between six and twelve months, or more than one year (Table 6). Among patients managed in Toby faith-based healing centers, 5.6% stayed less than one month, 20.4% stayed between one and three months, 8.3% stayed between three and six months, 2.8% stayed between six and twelve months, and 20.4% stayed more than one year. Overall, longer stays (three months or more) were observed exclusively among patients managed in Toby faith-based healing centers, whereas psychiatric services mainly provided short-term care (less than three months).

Determinants of first help-seeking

A significant association was found between educational level of decision-makers and first help-seeking source. Illiterate decision-makers were more likely to choose traditional healers or religious practices, whereas those with university education preferentially used medical services (p < 0.001). Profession of decision-makers was also significantly associated with first help-seeking. Those working in the primary sector more often opted for traditional or religious care, while those in the secondary sector more frequently consulted medical services (p < 0.001) (Table 7).

Decision-makers originating from the Analamanga region were significantly more likely to use medical services as a first resort (p = 0.001). Religion was significantly associated with first help-seeking, with Protestant decision-makers more likely to choose medical services (p = 0.013). Finally, perception of etiology influenced first help-seeking behavior. Belief in witchcraft as the cause of illness was significantly associated with initial consultation of traditional or religious healers (p = 0.013), whereas uncertainty about etiology was more often associated with medical consultation.

Discussion     

This study provides important insights into pathways to care for patients with psychotic disorders in Northwestern Madagascar, highlighting the predominance of non-medical help-seeking at illness onset and the key sociodemographic and cultural determinants influencing access to psychiatric services.

The relatively young mean age of patients reflects the typical early adult onset of psychotic disorders, as consistently reported in previous studies conducted in Africa, Europe, and Asia [12-15]. Early adulthood represents a critical developmental period marked by major social and professional transitions, during which emerging psychotic symptoms may be misinterpreted as behavioral, social, or spiritual disturbances. This misattribution likely contributes to delayed recognition and initiation of appropriate treatment. A marked male predominance was observed, in line with findings from several African and European studies [16-18]. Gender-related social norms may partly explain this pattern, as men may delay help-seeking due to expectations of strength and autonomy, while women may face additional barriers related to stigma, limited decision-making power, or financial dependence. These factors may influence both access to care and continuity of treatment.

Socio-economic vulnerability characterized the study population. Most patients and decision-makers had low educational attainment and were employed predominantly in the primary sector, reflecting limited socioeconomic resources. Similar profiles have been reported in other low-income settings [19, 20]. Low levels of education may reduce mental health literacy and reinforce explanatory models based on supernatural causation, thereby shaping help-seeking behaviour and favouring non-medical pathways.

Traditional healers and religious practices constituted the main first point of contact for nearly four out of five patients. This finding is consistent with studies from Africa and Asia, where traditional and faith-based care often precedes medical consultation [20,21-23]. In contrast, psychiatric services were mainly accessed as second or third resort, frequently following the perceived failure of traditional or religious interventions. Such delayed engagement with psychiatric care may contribute to a prolonged duration of untreated psychosis, which has been associated with poorer clinical and functional outcomes [10].

Educational level, occupation, region of origin, religion, and perceived etiology emerged as significant determinants of first help-seeking behaviour. Decision-makers with higher education and formal employment were more likely to seek medical care initially, whereas belief in witchcraft or supernatural causation strongly favoured traditional or religious care pathways [20,23-25]. These findings underscore the central role of sociocultural representations of mental illness in shaping care trajectories and delaying access to biomedical treatment.

Despite being perceived as the most effective treatment option, psychiatric care was also viewed as costly and less accessible. Financial constraints therefore represent a major barrier to early psychiatric consultation, reinforcing reliance on alternative care pathways. In contrast, traditional and religious care was perceived as more accessible and affordable, albeit less effective, a perception similarly reported in other African contexts [26-28]. This highlights persistent structural and economic challenges affecting mental health service utilization in low-resource settings.

Overall, the findings of this study emphasize the need for integrated mental health strategies that address both sociocultural and structural barriers to care. Strengthening mental health literacy, improving affordability and geographical accessibility of psychiatric services, and promoting structured collaboration between biomedical, traditional, and faith-based care providers may facilitate earlier access to effective treatment and improve outcomes for patients with psychotic disorders in Madagascar and similar settings.

Conclusion     

This multicenter study highlights the complexity of care pathways for patients with psychotic disorders in Northwestern Madagascar and confirms that initial help-seeking is predominantly non-medical. Traditional healers and religious practices constitute the main first points of contact, while psychiatric services are generally accessed at later stages of the care pathway. The choice of first help-seeking source was strongly influenced by sociodemographic and sociocultural factors related to decision-makers, including educational level, occupation, region of origin, religion, and perceived etiology of illness. Belief in supernatural causation and socioeconomic vulnerability favoured non-medical pathways, whereas higher education and formal employment were associated with earlier medical consultation. Despite being perceived as the most effective treatment option, psychiatric care remains limited by financial and structural barriers, reinforcing delays in access to appropriate treatment. Overall, these findings emphasize the need for integrated and culturally sensitive mental health strategies that strengthen mental health literacy, improve accessibility to psychiatric services, and promote collaboration between biomedical, traditional, and faith-based care providers in low-resource settings.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

The authors have read and agreed to the final manuscript.

Tables     

Table 1: socio-demographic characteristics of patients (n = 108)

Table 2: religious affiliation and ethnicity of patients (n = 108)

Table 3: sociodemographic characteristics of decision-makers (n = 108)

Table 4: first help-seeking source (n = 108)

Table 5: pathways to care: sequence of help-seeking (n = 108)

Table 6: length of stay according to place of care (n = 108)

Table 7: determinants of first help-seeking source (bivariate analysis)

References