A community‐based study of reporting demographic and clinical information concordance between informant and cognitively impaired participants

Abstract Background We studied concordance between informants’ and cognitively impaired participants’ reporting of demographic and clinical information in a community‐based cohort study. Method As part of the community‐based Brain Attack Surveillance in Corpus Christi‐Cognitive (BASIC‐C) project, households in Nueces County, Texas, USA, were randomly identified. Participants ≥ age 65 were recruited using door‐to‐door (5/1/2018‐3/15/2020) and phone (04/20/2020‐current) recruitment. Individuals with possible cognitive impairment were identified using the Montreal Cognitive Assessment (MoCA) during door‐to‐door recruitment and Telephone Montreal Cognitive Assessment (T‐MoCA) during phone recruitment. Participants who scored ≤25 and ≤18 on the MoCA and T‐MoCA respectively were eligible for participation. Named informants and participants both answered questions regarding the participant’s demographics and health status. Models were generated to examine the predictors that influence concordance between informant and participant answers. Predictors in the model included participants’ age, gender, ethnicity, degree of cognitive function (estimated by MoCA/T‐MoCA score), and relationship to informant. Result Table 1 provides the concordance between participant and informant answers. Overall concordance was high. Female participants were two and a half times more likely to have concordant answers about date of birth with informants than non‐spouses and male participants, and spouses were three and a half times more likely to agree with participants on date of birth than non‐spouses. (Table 2). Degree of cognitive function of the study participant also showed a large effect on concordance based on the question of whether the participant was diagnosed with dementia (p < .001) (Table 4). Other relationships in comparison to child, including other family member and friend/neighbor/other, were associated with concordance of answers to participant educational attainment and high blood pressure diagnosis (Table 3 and 5). Questions about participant diagnosis of diabetes, heart disease, stroke, and alcohol consumption, had no significant predictors. Ethnicity was not a significant predictor for any question. Conclusion Though most of our concordance rates were >80%, in studies that ask demographic and health history questions of cognitively impaired participants, the gold standard for “true” information remains uncertain. Our results, along with future research, may indicate that in these scenarios, studies should consider supplementing participant responses with informant contributions.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.