The specific focus of this paper is the use of the PHQ-9 as the depression screening tool in an acute hospital inpatient setting. Recently published guidelines recommend a model of care that includes screening of coronary heart disease (CHD) patients for depression during their inpatient admission; the aim of screening is to identify depressed patients and facilitate appropriate treatment. The relatively recent publication of these guidelines means that there are few studies to date that can comment on the ‘real world’ feasibility of implementing routine screening in the acute hospital setting. Further, it is not yet clear which model of care based on these guidelines, including screening, will be most effective in reducing the mortality and morbidity of depressed CHD patients. Our study, the first phase of which is due for completion in October 2010 is testing a model of care for the identification and management of CHD patients with depressive symptoms. It is located within an acute tertiary hospital setting and uses a Consultation Liaison Mental Health model.

To date, one hundred and twenty patients have screened positive for depressive symptoms on the PHQ-9 and have been randomised either to control (‘GP Liaison’ only, n = 60), or intervention group (‘GP Liaison, and assertive mental health follow-up’, n = 60). A score of 5 on the PHQ-9 was established as the minimum cut-off to indicate depressive symptoms. One week following discharge from hospital, those in the intervention group have been individually assessed in-depth by a Clinical Psychologist using the Mini-International Neuropsychiatric Interview (M.I.N.I), and recommendations for treatment made according to the outcome of this assessment.

Outcomes of this assessment revealed that of those with ‘mild’ depressive symptoms (PHQ-9 score of 5-9) 22% had a psychiatric diagnosis; of those with ‘moderate’ depressive symptoms (score of 10-14) 46% had a diagnosis; 75% scoring in the ‘moderate-severe’ range had a diagnosis, and 100% of patients in the severe range had a diagnosis. Of those with a psychiatric diagnosis, not all had a diagnosis of depression

Our paper describes and discusses the use the PHQ-9 as a depression screening tool as part of a model of care that begins in the cardiology inpatient setting. We discuss evidence supporting the use of the summed PHQ-9 score over the algorithm scoring method, and the decision to use a score of 5 as the minimum cut-off to indicate depressive symptoms. We discuss the accuracy of the PHQ-9 as a screening tool using the chosen scoring protocol when compared to the in-depth clinical assessments. We also discuss the accuracy of the PHQ-9 using alternative screening and scoring protocols in this sample and the clinical implications of these alternatives.