Are two items sufficient to screen for depression within the hospital context?

to determine the ability of the reduced form of a screening instrument, the Patient Health Questionnaire-2 (PHQ-2), to assess the presence of depressive disorders in patients admitted to a general hospital.

A sample of 227 patients admitted to the clinical wards of a Brazilian general university hospital were assessed with Module A of the Diagnostic Structured Interview for the DSM-IV (SCID-IV) and filled out the PHQ-9 and PHQ-2.

The PHQ-2 demonstrated an area under the ROC curve of 0.89 (p < 0.0001), with a cutoff point of three or more being the one that best equilibrated the sensitivity (0.86) and specificity (0.75) values. The agreement index between the PHQ-2 and module A of SCID-IV was 78.4% and the Kappa value was 0.51. Regarding reliability, the Cronbach alpha value obtained was 0.64 and the intraclass correlation coefficient was 0.52.

PHQ-2 proved to be an instrument with good psychometric properties comparable to those of PHQ-9, being superior to the latter regarding the rate of false-positive results. In addition, it is a brief instrument that elicits little resistance on the part of the patient, being inexpensive and requiring little time, thus being of important help to the treatment teams for the detection of depressive disorder, being suitable for incorporation into hospital admission protocols and thus possibly favoring more immediate interventions.