Regular Research Articles
Comparative Validation of Proxy-Based Montgomery-Åsberg Depression Rating Scale and Cornell Scale for Depression in Dementia in Nursing Home Residents With Dementia

https://doi.org/10.1097/JGP.0b013e318233152bGet rights and content

Objective

To 1) compare the accuracy of the Montgomery-Åsberg Depression Rating Scale (MADRS) and the Cornell Scale for Depression in Dementia (CSDD) in nursing home residents with dementia when professional caregivers are the only available source of information and 2) explore different methods to account for missing items.

Design

Cross-sectional design.

Setting

Nursing home (NH).

Participants

One hundred one residents with dementia.

Measurements

NH residents with dementia were assessed on the presence of clinical depression using Provisional Diagnostic Criteria for Depression of Alzheimer's Disease. The MADRS and CSDD were administered in a structured interview with professional primary caregivers.

Results

Receiver operating characteristic analyses revealed no significant differences between areas under the empirical curve for MADRS and CSDD. Imputation of a lowest possible item score for missing items revealed larger areas than three other methods (significant result only for CSDD). A MADRS cutoff score of “>13” yielded the highest sum of sensitivity (78%) and specificity (66%). A CSDD cutoff score of “>6” yielded the highest sum of sensitivity (94%) and specificity (49%). Both scales showed high negative predictive values up to 100% and low positive predictive values not exceeding 50%.

Conclusion

The proxy-based MADRS and CSDD did not differ in distinguishing depressed from nondepressed NH residents and may be used for screening purposes. For missing items, imputation of a lowest possib le item score may be applied. The MADRS and CSDD may be better used for ruling out rather than for ruling in depression.

Section snippets

Participants and Design

The sample for the present cross-sectional study was drawn from a larger, ongoing effectiveness study of the multidisciplinary care program, “Act in Case of Depression” (AID) study, aimed to identify and treat depression in NH residents.26 All data reflect baseline and preintervention assessments. The residents were recruited from seven of the 16 cluster-randomized dementia special care units participating in the AID study. Inclusion criteria were written consent by the legal representative on

RESULTS

Of the 117 residents recruited for the current study, 16 residents died or moved to another unit before they could be assessed. The study sample consisted of 28 male (27.7%) and 73 female residents (Table 1). Of the 101 residents from seven care units (8–22 residents per unit), 18 (17, 8%) were diagnosed as depressed. Dementia type, as specified in personal file, was as follows (first diagnosis N [n depressed]): Alzheimer's disease N = 25 (6), vascular dementia N = 24 (1), other type N = 7 (2),

DISCUSSION

With regard to the first aim of this study, both MADRS and CSDD performed moderately in distinguishing depressed from nondepressed NH residents with dementia when only one source of information was available, namely that of professional caregivers. The ROC analyses revealed that the scales did not significantly differ in their performance against the diagnostic criteria for depression in dementia. Although concurrent validity was acceptable, both the MADRS and CSDD had a combination of a high

CONCLUSIONS

The fact that interviewing the resident was not possible in at least half of our sample underlines the need for the validation of proxy-based depression scales. The results of our study showed that the CSDD and MADRS may be used for screening purposes in NH residents with dementia when only one source of information is available, namely professional caregivers. In addition, it was demonstrated that different imputation methods for missing items might lead to different optimal cutoff scores,

References (40)

  • TL Mark et al.

    Antidepressant Use in Geriatric Populations: The Burden of Side Effects and Interactions and Their Impact on Adherence and Costs

    Am J Geriatr Psychiatry

    (2011)
  • DW O'Connor et al.

    Psychosocial treatments of psychological symptoms in dementia: a systematic review of reports meeting quality standards

    Int Psychogeriatr

    (2009)
  • AL Gruber-Baldini et al.

    Characteristics associated with depression in long-term care residents with dementia

    Gerontologist

    (2005)
  • J Snowdon et al.

    Recognising depression in residential facilities: an Australian challenge

    Int J Geriatr Psychiatry

    (2008)
  • SL Wright et al.

    Distinguishing between depression and dementia in older persons: Neuropsychological and neuropathological correlates

    J Geriatr Psychiatry Neurol

    (2007)
  • RB Wetzels et al.

    Determinants of quality of life in nursing home residents with dementia

    Dement Geriatr Cogn Disord

    (2010)
  • T Gonzalez-Salvador et al.

    Quality of life in dementia patients in long-term care

    Int J Geriatr Psychiatry

    (2000)
  • M Smalbrugge et al.

    The impact of depression and anxiety on well-being, disability and use of health care services in nursing home patients

    Int J Geriatr Psychiatry

    (2006)
  • MA Rapp et al.

    Cognitive decline in patients with dementia as a function of depression

    Am J Geriatr Psychiatry

    (2010)
  • GH Suh et al.

    Mortality in Alzheimer's disease: a comparative prospective Korean study in the community and nursing homes

    Int J Geriatr Psychiatry

    (2005)
  • Cited by (35)

    • Predictive validity of the Cornell Scale for depression in dementia among older adults with and without dementia: A systematic review and meta-analysis

      2022, Psychiatry Research
      Citation Excerpt :

      The selected studies that verified the predictive validity of CSDD in older adults comprised 3499 participants (Table 1). Seven studies (Lam et al., 2004; Leontjevas et al., 2009; 2012 Lim et al., 2012; Mougias et al., 2017; Porta-Etessam et al., 2011; Vida et al., 1994) included participants with dementia, and one study (Karam et al., 2018) included only older adults without dementia. The remaining 12 studies included both older adults with and without dementia.

    • The course of depressive symptoms as measured by the Cornell scale for depression in dementia over 74 months in 1158 nursing home residents

      2015, Journal of Affective Disorders
      Citation Excerpt :

      Before calculating the sum scores of the CSDD (total, mood and non-mood CSDD scores), the rating “item unable to evaluate” was replaced by a score of 0. This method of imputation is in line with what has been recommended in previous research (Leontjevas et al., 2012). However, the CSDD scores were only calculated where at most 20% of the CSDD items had the rating “item unable to evaluate.”

    View all citing articles on Scopus

    The authors thank the research assistant, Alexandra Evers, M.Sc.; the AID-study team; and all of the nursing home personnel who contributed to the study.

    The authors declare no conflicts of interest.

    This study was supported by The Netherlands Organization for Health Research and Development (ZonMW), contract/grant number: 170992801.

    View full text