Can the 12-item general health questionnaire be used to identify medical students who might ‘struggle’ on the medical course? A prospective study on two cohorts

BACKGROUND Students who fail to thrive on the Nottingham undergraduate medical course frequently suffer from anxiety, depression or other mental health problems. These difficulties may be the cause, or the result of, academic struggling. Early detection of vulnerable students might direct pastoral care and remedial support to where it is needed. We investigated the use of the short-form General Health Questionnaire (GHQ-12) as a possible screening tool. METHODS Two consecutive cohorts (2006 and 2007) were invited to complete the GHQ-12. The questionnaire was administered online, during the second semester (after semester 1 exams) for the 2006 cohort and during the first semester for the 2007 cohort. All data were held securely and confidentially. At the end of the course, GHQ scores were examined in relation to course progress. RESULTS 251 students entered the course in 2006 and 254 in 2007; 164 (65%) and 160 (63%), respectively, completed the GHQ-12. In both cohorts, the study and non-study groups were very similar in terms of pre-admission socio-demographic characteristics and overall course marks. In the 2006 study group, the GHQ Likert score obtained part-way through the first year was negatively correlated with exam marks during Years 1 and 2, but the average exam mark in semester 1 was the sole independent predictor of marks in semester 2 and Year 2. No correlations were found for the 2007 study group but the GHQ score was a weak positive predictor of marks in semester 2, with semester 1 average exam mark again being the strongest predictor. A post-hoc moderated-mediation analysis suggested that significant negative associations of GHQ scores with semester 1 and 2 exams applied only to those who completed the GHQ after their semester 1 exams. Students who were identified as GHQ 'cases' in the 2006 group were statistically less likely to complete the course on time (OR=4.74, p 0.002). There was a non-significant trend in the same direction in the 2007 group. CONCLUSIONS Results from two cohorts provide insufficient evidence to recommend the routine use of the GHQ-12 as a screening tool. The timing of administration could have a critical influence on the results, and the theoretical and practical implications of this finding are discussed. Low marks in semester 1 examinations seem be the best single indicator of students at risk for subsequent poor performance.