A recent study by the MaGPIe Research Group1 demonstrated high GP recognition rates (70.3%) of mental health problems. The findings may, however, be biased by the fact that GP case identification was based on a historical 12-month observation period whereas the final diagnosis (CIDI) was a present state assessment (within 1 month). This could lead to an overestimation of GP recognition rates. The conclusion that there is little gain from screening due to high detection rates may therefore be erroneous.
In a study using a composite screening questionnaire2,3 we found that 43% of patients had high screening scores. Among patients with high screening scores we found an NNT (number needed to test) of 5, indicating that feedback on high scorers has the potential to identify 20 new cases per 100 cases. The effectiveness of feeding back high scorers' results on depression and anxiety scales is supported by a systematic review by Gilbody and colleagues.4 A clinically useful description of this high score group could provide us with suggestions for making rational use of rating scales in general practice. While there is little evidence to support routine mental health screening, case-finding seems most promising.5
- © British Journal of General Practice, 2005.