Patient characteristics that may predict the likelihood of the presence of mental health problems in patients attending the general outpatient clinic of a tertiary hospital in South-South Nigeria.

Background A considerable number of patients seen in general outpatient clinics (GOPC) are known to suffer from psychiatric rather than physical disorders. Studies have shown that doctors working in these clinics have difficulty in making accurate ratings of mental health problems in their patients and have poor knowledge of psychiatric diagnosis. Accurate recognition of psychiatric symptoms in a patient is essential for specific diagnosis and successful management. There is a need for the use of an easy tool such as the12-item General Health Questionnaire (GHQ-12) for screening and identification of psychopathologies especially in a busy clinic setting like the GOPC. Aside from psychometric screening tools, patients' sociodemographic characteristics such as gender, age, marital status, occupation, education etc. have been found to be of value in predicting those at risk.Objectives This study seeks to correlate GHQ 'caseness' with sociodemographic factors and to compare physician diagnosis with GHQ diagnosis.Subjects and method Three-hundred and twenty-two respondents were recruited for the study by a systematic random sampling method. Using a cut off score of three on both the English and Efik translation versions of the GHQ-12, 'cases' and 'non-cases' generated were compared with the same classification as identified by the GOPC doctors. Identification rates for both groups were calculated and the coefficients determined using a two-by-two contingency table. Sociodemographic correlates were determined by statistical comparison of the classifications in both groups.Results Statistically significant differences in sociodemographic characteristics of respondents were found for age (χ(2)=48.97; P <0.05) and education (χ(2)=45.64; P=0.05) using their GHQ-12 scores, and for occupation (χ(2)=37.90; P <0.05) among those seen by the GOPC doctors. A further comparison of identified 'cases' and 'non-cases' by doctors again revealed significant difference for age (χ(2)=7.151; P <0.05). Sex as a sociodemographic characteristic showed no statistically significant difference though a greater percentage of females (57.3%) were observed as 'high scorers' as compared to their male counterparts (42.7%). The GHQ-12 identified 46.6% 'cases' while the GOPC doctors identified 6.8% among the attendees with a diagnostic sensitivity of 8% and a specificity of 94%, respectively.Conclusion Belonging to the 18-39 years age group, being employed and having less than 12 years of education were the patients' characteristics that suggested the likelihood of the presence of mental health problems.This study also revealed that despite the high proportion of psychiatric morbidity (46.6%) in the GOPC of the University of Calabar Teaching Hospital (UCTH) rate of detection by the clinic doctors was low (6.8%).It is recommended that primary care doctors should be alerted to the possibility that clinically significant psychiatric morbidity may be present in GOPC attendees. The correlation between patients' sociodemographic parameters and presence of mental health problems could be informative and should be given adequate attention during consultation.