Correspondence (letter to the editor): Limitations of the Study

In our opinion, the study (1) has substantial limitations/shortcomings—but these were not mentioned in the text. This makes it almost impossible to interpret the results.

• The data came from only 5.8% of the physicians who had been contacted—consequently they are anything but representative.

• The Depression Screening Questionnaire (DSQ) was developed for the purpose of screening; it cannot be used to make a clinical diagnosis. This means that a substantial proportion of patients whose test results were positive do not actually have depression—as is made clear by the low specificity of the test (2). Furthermore, the test reflects only what happens in those few weeks that the questions relate to. This makes it impossible to make a diagnosis of depression, as in some cases it is only situation-specific low mood that is being measured (3).

• Defining a mean value that includes patients with suspected mild to moderate depression, is misleading, especially as far as assessing the quality of treatment is concerned. Medication treatment for mild, even moderate, depression is explicitly stated to be not sufficiently evidence-based in the current version of the national disease management guideline (4), which also applied at the time when the study was conducted. The same is true for international guidelines.

Apart from the—flawed—study, it deserves to be said that the treatment of the depressive patients (who had been identified by the GPs themselves, without using any questionnaire) seemed very good indeed. Therapy rates in severe, moderate, and mild depression were 97%, 87.5%, and 79.1%. To infer inadequate treatment from such results and to deduce from that the need to “improve general practitioners’ ability to diagnose these conditions and determine the indication for treatment” is unsupportable. Rather, it shows the authors’ biases.