We thank Drs Christensen and Fink for their interest in our work. The results of our initial validation deserve replication in a larger unselected GP population, as we highlight in our discussion.1 However, for the PHQ–9, our results are broadly in line with the other PHQ9 validation studies (based upon 5026 patients) which are in the public domain at the time of writing. We have recently subjected these data to a systematic review and diagnostic meta-analysis.2
Our BJGP paper represents the first UK validation of the PHQ–9 and CORE instruments and we felt it important to place these data in the public domain, given the recent emphasis on routine depression assessment under the Quality and Outcomes Framework.
The performance of any instrument will vary between populations and studies, and sensitivity and specificity are especially influenced by baseline prevalence.3 However, the baseline prevalence of depression in our study is of a similar magnitude to that found in ‘high risk’ populations such as those with coronary heart disease and diabetes (where the use of brief instruments is rewarded under the QOF). We therefore also reported likelihood ratios,4 which are relatively insensitive to baseline risk and are much more informative to clinicians in their clinical decision making.5 Likelihood ratios are ‘portable’ and can be readily used to establish post-test probability of a disorder within a plausible range of baseline prevalence estimates. We presented one such estimation using figures commonly encountered in primary care in our paper.
- © British Journal of General Practice, 2007.
