Diagnosing dementia in a developing nation: an evaluation of the GMS-AGECAT algorithm in an older Korean population

Int J Geriatr Psychiatry. 2003 Apr;18(4):331-6. doi: 10.1002/gps.834.

Abstract

Background: Numbers of older people are increasing rapidly in many developing nations and there is a pressing need for epidemiological studies of psychiatric morbidity, particularly dementia. Valid methods for case identification are an important first step. The Geriatric Mental State (GMS) is a widely used diagnostic instrument but has received little formal evaluation in developing nations. The objective of this study was to investigate the screening properties of GMS organic disorder with respect to clinically diagnosed dementia in an older Korean population, about half of whom have received no formal education.

Methods: 746 people aged 65+ in Kwangju, South Korea who were participants in a community survey of psychiatic morbidity, completed a GMS interview (with diagnoses assigned through the AGECAT computerised algorithm) and, independently, received a clinical assessment for dementia.

Results: GMS organic disorder was diagnosed in 262 participants (35%) and DSM-IV dementia in 110 (15%). Increased age, female gender and lower education independently predicted disagreement between the two diagnoses. In these groups, GMS sensitivity to a clinical diagnosis of dementia remained high but specificity was markedly reduced.

Conclusions: Dementia may be overestimated in developing nations if the GMS-AGECAT diagnosis of organic disorder is used alone. The validity of other diagnoses, such as affective disorder, may also be affected if the hierarchical diagnostic algorithm is used (i.e. where the diagnosis of organic disorder affects the likelihood of other diagnoses).

Publication types

  • Evaluation Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Aged, 80 and over
  • Algorithms
  • Dementia / diagnosis*
  • Developing Countries*
  • Female
  • Humans
  • Interview, Psychological
  • Korea
  • Male
  • Mass Screening / methods
  • Odds Ratio
  • Psychiatric Status Rating Scales*
  • Reproducibility of Results