Handling missing items in the Hospital Anxiety and Depression Scale (HADS)

A simulation study Public Health

Melanie L Bell, Diane L. Fairclough, Mallorie H. Fiero, Phyllis N. Butow

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

Background: The Hospital Anxiety and Depression Scale (HADS) is a widely used questionnaire in health research, but there is little guidance on how to handle missing items. We aimed to investigate approaches to handling item non-response, varying sample size, proportion of subjects with missing items, proportion of missing items per subject, and the missingness mechanism. Methods: We performed a simulation study based on anxiety and depression data among cancer survivors and patients. Item level data were deleted according to random, demographic, and subscale dependent missingness mechanisms. Seven methods for handling missing items were assessed for bias and imprecision. Imputation, imputation conditional on the number of non-missing items, and complete case approaches were used. One thousand datasets were simulated for each parameter combination. Results: All methods were most sensitive when missingness was dependent on the subscale (i.e., higher values of depression leads to higher levels of missingness). The worst performing approach was to analyze only individuals with complete data. The best performing imputation methods depended on whether inference was targeted at the individual or at the population. Conclusions: We recommend the 'half rule' using individual subscale means when using the HADS scores at the individual level (e.g. screening). For population inference, we recommend relaxing the requirement that at least half the items be answered to minimize missing scores.

Original languageEnglish (US)
Article number2284
JournalBMC Research Notes
Volume9
Issue number1
DOIs
StatePublished - Oct 22 2016

Fingerprint

Public health
Anxiety
Public Health
Depression
Screening
Health
Sample Size
Population
Survivors
Demography
Research
Neoplasms

Keywords

  • Anxiety
  • Depression
  • Distress
  • Imputation
  • Missing data
  • Questionnaires
  • Simulation

ASJC Scopus subject areas

  • Medicine(all)
  • Biochemistry, Genetics and Molecular Biology(all)

Cite this

Handling missing items in the Hospital Anxiety and Depression Scale (HADS) : A simulation study Public Health. / Bell, Melanie L; Fairclough, Diane L.; Fiero, Mallorie H.; Butow, Phyllis N.

In: BMC Research Notes, Vol. 9, No. 1, 2284, 22.10.2016.

Research output: Contribution to journalArticle

Bell, Melanie L ; Fairclough, Diane L. ; Fiero, Mallorie H. ; Butow, Phyllis N. / Handling missing items in the Hospital Anxiety and Depression Scale (HADS) : A simulation study Public Health. In: BMC Research Notes. 2016 ; Vol. 9, No. 1.
@article{922e25ade7374b72b9a43f4e3a84493b,
title = "Handling missing items in the Hospital Anxiety and Depression Scale (HADS): A simulation study Public Health",
abstract = "Background: The Hospital Anxiety and Depression Scale (HADS) is a widely used questionnaire in health research, but there is little guidance on how to handle missing items. We aimed to investigate approaches to handling item non-response, varying sample size, proportion of subjects with missing items, proportion of missing items per subject, and the missingness mechanism. Methods: We performed a simulation study based on anxiety and depression data among cancer survivors and patients. Item level data were deleted according to random, demographic, and subscale dependent missingness mechanisms. Seven methods for handling missing items were assessed for bias and imprecision. Imputation, imputation conditional on the number of non-missing items, and complete case approaches were used. One thousand datasets were simulated for each parameter combination. Results: All methods were most sensitive when missingness was dependent on the subscale (i.e., higher values of depression leads to higher levels of missingness). The worst performing approach was to analyze only individuals with complete data. The best performing imputation methods depended on whether inference was targeted at the individual or at the population. Conclusions: We recommend the 'half rule' using individual subscale means when using the HADS scores at the individual level (e.g. screening). For population inference, we recommend relaxing the requirement that at least half the items be answered to minimize missing scores.",
keywords = "Anxiety, Depression, Distress, Imputation, Missing data, Questionnaires, Simulation",
author = "Bell, {Melanie L} and Fairclough, {Diane L.} and Fiero, {Mallorie H.} and Butow, {Phyllis N.}",
year = "2016",
month = "10",
day = "22",
doi = "10.1186/s13104-016-2284-z",
language = "English (US)",
volume = "9",
journal = "BMC Research Notes",
issn = "1756-0500",
publisher = "BioMed Central",
number = "1",

}

TY - JOUR

T1 - Handling missing items in the Hospital Anxiety and Depression Scale (HADS)

T2 - A simulation study Public Health

AU - Bell, Melanie L

AU - Fairclough, Diane L.

AU - Fiero, Mallorie H.

AU - Butow, Phyllis N.

PY - 2016/10/22

Y1 - 2016/10/22

N2 - Background: The Hospital Anxiety and Depression Scale (HADS) is a widely used questionnaire in health research, but there is little guidance on how to handle missing items. We aimed to investigate approaches to handling item non-response, varying sample size, proportion of subjects with missing items, proportion of missing items per subject, and the missingness mechanism. Methods: We performed a simulation study based on anxiety and depression data among cancer survivors and patients. Item level data were deleted according to random, demographic, and subscale dependent missingness mechanisms. Seven methods for handling missing items were assessed for bias and imprecision. Imputation, imputation conditional on the number of non-missing items, and complete case approaches were used. One thousand datasets were simulated for each parameter combination. Results: All methods were most sensitive when missingness was dependent on the subscale (i.e., higher values of depression leads to higher levels of missingness). The worst performing approach was to analyze only individuals with complete data. The best performing imputation methods depended on whether inference was targeted at the individual or at the population. Conclusions: We recommend the 'half rule' using individual subscale means when using the HADS scores at the individual level (e.g. screening). For population inference, we recommend relaxing the requirement that at least half the items be answered to minimize missing scores.

AB - Background: The Hospital Anxiety and Depression Scale (HADS) is a widely used questionnaire in health research, but there is little guidance on how to handle missing items. We aimed to investigate approaches to handling item non-response, varying sample size, proportion of subjects with missing items, proportion of missing items per subject, and the missingness mechanism. Methods: We performed a simulation study based on anxiety and depression data among cancer survivors and patients. Item level data were deleted according to random, demographic, and subscale dependent missingness mechanisms. Seven methods for handling missing items were assessed for bias and imprecision. Imputation, imputation conditional on the number of non-missing items, and complete case approaches were used. One thousand datasets were simulated for each parameter combination. Results: All methods were most sensitive when missingness was dependent on the subscale (i.e., higher values of depression leads to higher levels of missingness). The worst performing approach was to analyze only individuals with complete data. The best performing imputation methods depended on whether inference was targeted at the individual or at the population. Conclusions: We recommend the 'half rule' using individual subscale means when using the HADS scores at the individual level (e.g. screening). For population inference, we recommend relaxing the requirement that at least half the items be answered to minimize missing scores.

KW - Anxiety

KW - Depression

KW - Distress

KW - Imputation

KW - Missing data

KW - Questionnaires

KW - Simulation

UR - http://www.scopus.com/inward/record.url?scp=84992195751&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84992195751&partnerID=8YFLogxK

U2 - 10.1186/s13104-016-2284-z

DO - 10.1186/s13104-016-2284-z

M3 - Article

VL - 9

JO - BMC Research Notes

JF - BMC Research Notes

SN - 1756-0500

IS - 1

M1 - 2284

ER -