1 Department of Statistics, University of California, Berkeley, California.
2 PRIISM Applied Statistics Center, New York University, New York, New York.
3 Department of Applied Statistics, Social Science, and Humanities, New York University, New York, New York.
4 Lady Davis Institute for Medical Research, Jewish General Hospital, Montréal, Québec, Canada.
5 Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Montréal, Québec, Canada.
6 Department of Psychology, McGill University, Montréal, Québec, Canada.
7 Department of Clinical, Neuro and Developmental Psychology, EMGO Institute, VU University, Amsterdam, The Netherlands.
8 Department of Health Sciences, Hull York Medical School, University of York, Heslington, York, UK.
9 Department of Medicine, Stanford University, Stanford, California.
10 Department of Health Research and Policy, Stanford University, Stanford, California.
11 Department of Biomedical Data Science, Stanford University, Stanford, California.
12 Department of Statistics, Stanford University, Stanford, California.
13 Library, Concordia University, Montréal, Québec, Canada.
14 Department of Community Health Sciences, University of Calgary, Calgary, Alberta, Canada.
15 Hotchkiss Brain Institute, University of Calgary, Calgary, Alberta, Canada.
16 O'Brien Institute for Public Health, University of Calgary, Calgary, Alberta, Canada.
17 Department of General Practice and Primary Health Care, University of Auckland, New Zealand.
18 Department of Rehabilitation Medicine, University of Washington, Seattle, Washington.
19 Centre for Research on Ageing, Health and Wellbeing, Research School of Population Health, Australian National University, Canberra, Australia.
20 Centre for Mental Health, Melbourne School of Population and Global Health, The University of Melbourne, Melbourne, Australia.
21 Melbourne Institute of Applied Economic and Social Research, The University of Melbourne, Melbourne, Australia.
22 Centre for Translational Neuroscience and Mental Health, University of Newcastle, New South Wales, Australia.
23 Psycho-Oncology Service, Calvary Mater Newcastle, New South Wales, Australia.
24 Department of Psychiatry, University of Rochester Medical Center, Rochester, New York.
25 School of Social Work, University of Pittsburgh, Pittsburgh, Pennsylvania.
26 Liaison Psychiatry Department, John Hunter Hospital, Newcastle, Australia.
27 Minneapolis Health Department, Minneapolis, Minnesota.
28 Department of Medicine, McGill University, Montréal, Québec, Canada.
29 Department of Clinical Neurosciences, University of Calgary, Calgary, Alberta, Canada.
30 Department of Nursing, St. Joseph's College, Standish, Maine.
31 Department of Epidemiology, Gillings School of Global Public Health, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina.
32 Division of Psychiatry, Royal Edinburgh Hospital, University of Edinburgh, Edinburgh, Scotland, UK.
33 Allina Health, Minneapolis, Minnesota.
34 School of Medicine and Public Health, University of Newcastle, New South Wales, Newcastle, Australia.
35 IMPACT Strategic Research Centre, School of Medicine, Deakin University, Geelong, Victoria, Australia.
36 Monash

Background: The objective of this study was to develop and validate a short form of the Patient Health Questionnaire-9 (PHQ-9), a self-report questionnaire for assessing depressive symptomatology, using objective criteria.

Methods: Responses on the PHQ-9 were obtained from 7,850 English-speaking participants enrolled in 20 primary diagnostic test accuracy studies. PHQ unidimensionality was verified using confirmatory factor analysis, and an item response theory model was fit. Optimal test assembly (OTA) methods identified a maximally precise short form for each possible length between one and eight items, including and excluding the ninth item. The final short form was selected based on prespecified validity, reliability, and diagnostic accuracy criteria.

Results: A four-item short form of the PHQ (PHQ-Dep-4) was selected. The PHQ-Dep-4 had a Cronbach's alpha of 0.805. Sensitivity and specificity of the PHQ-Dep-4 were 0.788 and 0.837, respectively, and were statistically equivalent to the PHQ-9 (sensitivity = 0.761, specificity = 0.866). The correlation of total scores with the full PHQ-9 was high (r = 0.919).

Conclusion: The PHQ-Dep-4 is a valid short form with minimal loss of information of scores when compared to the full-length PHQ-9. Although OTA methods have been used to shorten patient-reported outcome measures based on objective, prespecified criteria, further studies are required to validate this general procedure for broader use in health research. Furthermore, due to unexamined heterogeneity, there is a need to replicate the results of this study in different patient populations.