Article Text

Download PDFPDF

Adding a question that asks whether help is wanted improves sensitivity of a depression screening tool
Free
  1. Bernd Löwe, MD, PhD
  1. University of Heidelberg, School of Medicine, Department of
    Psychosomatic and General Internal Medicine, Heidelberg, Germany

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

OpenUrlAbstract/FREE Full Text

Q Does adding a question asking whether help is needed to a two-question screening tool assist doctors to diagnose people with depression?

METHODS

Embedded ImageDesign:

Prospective cohort study.

Embedded ImageSetting:

Six general practice clinic waiting rooms, New Zealand.

Embedded ImagePatients:

1025 consecutive people waiting to see a doctor for no pre-specified clinical reason. Exclusion: people using psychotropic medication.

Embedded ImageTest:

The screening tool consisted of three questions; two developed from the Prime MD-questionnaire which asked “During the past month have you often been bothered by feeling down, depressed or hopeless?” and “During the past month have you often been bothered by little interest or pleasure in doing things?” An additional question was added asking “Is this something with which you would like help?” to which there were three possible responses; “No”, “Yes”, and “Yes but not today.”

Embedded ImageDiagnostic standard:

Mood module of Composite International Diagnostic Interview (CIDI) for diagnosing depression.

Embedded ImageOutcomes:

Diagnosis of depression; sensitivity, specificity, and likelihood ratios of screening questions; sensitivity, specificity, and likelihood ratios of doctor’s diagnosis after seeing the written response to the screening and help questions.

MAIN RESULTS

Adding a help question to screening questions for depression improved their sensitivity and specificity (see http://www.ebmentalhealth.com/supplemental for table). Doctor’s recognition of depression was improved by the modified screening tool when validated against the mood module of the CIDI (sensitivity: 79%, specificity: 94%).

CONCLUSIONS

Adding a help question to a two-question screening tool improves sensitivity and specificity of diagnosis of people with depression in primary care. This modified screening tool is a simple and effective way for doctors to identify people who are suffering from depression.

Commentary

It is well established that for depression screening, two questions asking for the core symptoms of depression—depressed mood and loss of interest—are sufficient.1,4 The current study by Arroll et al is important because it investigates the operating characteristics of these two depression items together with one of the clinically most relevant questions; whether the patient would like help with his or her problems. The study results suggest that compared to the original two screening items, the additional “help” question substantially increases both specificity (89% v 78%) and positive likelihood ratio (9.1 v 4.4), whereas sensitivity remains unchanged (96% v 96%).

For the patient, the effort to answer the additional “help” question is negligible; whereas the benefit for the treating physician is remarkable. First, it elicits information on the patient’s subjective need for treatment and second, the improved operating characteristics markedly reduce the number of patients screened false positive for depression. The fewer patients screened false positive means less time spent sorting out false positive cases. All in all, the inclusion of the additional “help” question is feasible, clinically relevant, and improves criterion validity of the two-item depression screener.

However, due to its dichotomous (yes/no) response format, the two depression items used in this study do not go beyond case finding—that is, they are not suitable for grading depression severity or for assessing depression change over time.5 A further development of this screener is the two-item Patient Health Questionnaire (PHQ-2) that has a four-point response format for each of its two items with total scores ranging from 0 to 6.3 In two studies, the PHQ-2 accurately reflected depression severity, performed favourably with respect to standard diagnostic interviews, and proved sensitive to change.3,4 Thus, the combination of the PHQ-2 and the “help” question clearly appears as the next step in creating a practical multipurpose measure that may be used for detecting depression, grading its severity, assessing the need for treatment, and for monitoring outcome over time. Naturally, the “help” question could contribute substantially to any other depression questionnaire as well as to questionnaires assessing anxiety, somatisation, and other mental disorders.

References

Footnotes

  • For correspondence: Professor B Arroll, Department of General Practice and Primary Health Care, School of Population Health, University of Auckland, Private Bag 92019, Auckland, New Zealand; b.arroll{at}auckland.ac.nz

  • Source of funding: Oakley Mental Health Foundation, New Zealand.