Background The effectiveness of interprofessional collaboration (IPC) in primary care is unclear. It may have health benefits for patients with cardiovascular or mental health problems. The increase in the prevalence of mental disorders (particularly depressive episodes) in France over the last decade will mobilize both primary care and mental health actors in the future. They have a vested interest in working together to meet the growing needs of the French population. A consultation-liaison with a private psychiatrist was tested over 4 years within a French multi-professional health center (MHC) located in a French rural town. The overall aim of the study was to assess the care needs of GPs and their patients who benefited from this IPC. A study using a mixed methods approach was carried out with the following specific aims: (1) the main objective of the quantitative study was to describe the reasons for referral to psychiatrist by GP. The secondary objectives were to describe the responses given by the psychiatrist and to estimate the association between reasons for referral and patients characteristics; (2) the main objective of the qualitative study was to explore how patients perceived the GP-psychiatrist IPC, and the collaboration they experienced in this MHC. The secondary objectives were to explore their actual experiences in the French mental health system and their perceived needs. Methods A convergent parallel design study was set up by combining a retrospective cross-sectional study by analyzing data from the medical records of patients, a qualitative study using semi-directed individual interviews and a non-participant observation with volunteer patients, and an integrative analysis phase to mirror the results of the two substudies. The analysis of data from the quantitative study was descriptive, followed by multivariate logistic regression analyses. The analysis of data from the qualitative study was inspired by Grounded Theory. The target population was adult patients who had visited at least one GP in the MHC. Results One hundred patients were included in the quantitative study (women: 65%, mean age: 47.2 years), who were seen by the psychiatrist in 117 consultations. Three types of request were made by GPs to the psychiatrist: therapeutic requests (83.7%), diagnostic requests (35.9%), and administrative requests related to work absence (4.2%). After adjustment, patients were more likely to be referred to the psychiatrist for a therapeutic reason if they had depressive or anxiety disorder (AOR = 4.46, 95% CI: 1.57–12.69). Patients with bipolar disorder were more likely to be referred for diagnostic advice (AOR = 10.59, 95% CI: 1.88–59.72). The psychiatrist's response was mainly therapeutic (91.5%): pharmacological in 74.3% of cases and psychotherapeutic in 50.4%. A diagnostic response was given in 41.9% of consultations. Of these diagnostic responses, 48.9% were diagnostic confirmations and 22.4% were alternative diagnoses. Ten patients participated in the qualitative study. They perceived the GP as the coordinator of care and the psychiatrist as the expert. This pair, supported by other actors or approaches, guided the patient with a mental disorder towards holistic rehabilitation. Patients also found the organization of mental health care in France difficult to understand and inadequate (short consultations, too-long intervals between consultations). Four changes in this mental health care system were identified by patients as priorities: proximity and flexibility of mental health care, more interaction between mental health professionals and GPs, patient involvement in theIPC, and integration of other actors or approaches in care. GPs sought support from psychiatrists in situations of diagnostic or therapeutic uncertainty, and when mental health was interfering with work. GPs and their patients agreed that the first provider of mental health care should be the GP, and patients also felt that this IPC could improve emergency management. Conclusions GPs in this MHC were involved in the mental health care pathway of their patients who recognized it. However, IPC remained necessary to obtain psychiatric expertise as a second resort in identifiable primary care situations.< Réduire