Cognitive dysfunction in older subjects with diabetes mellitus: impact on diabetes self-management and use of care services

Objective: To determine whether cognitive impairment is associated with cha nges in self-care behaviour and use of health and social services in older subjects with diabetes mellitus. Research design and methods: This was a co mmunity based, case-control study of subjects registered with general pract ices participating in the All Wales Research into Elderly (AWARE) Diabetes Study. The 396 patients aged 65 years or older with known diabetes mellitus were compared with 393 age- and sex-matched, non-diabetic controls. Adjust ed odds ratio estimates of normal performance on Mini-Mental State Examinat ion (MMSE) and Clock Drawing Test (numbers and hands) were determined. Info rmation on self-care behaviours and use of services was obtained. Results: A total of 283 (71%) diabetic subjects scored 24 or more on MMSE, compared with 323 (88%) of controls (OR 0.54, P < 0.0005). The mean (S.D.) scores we re 24.5 (5.1) and 25.7 (4.3), respectively (difference between means 1.22; 95% CI 0.56, 1.88; P < 0.001). Clock testing demonstrated that 257 (65%) an d 286 (72%) diabetic subjects correctly placed the numbers and hands, respe ctively, compared with 299 (76%) and 329 (84%) of controls (OR 0.59, P < 0. 001 and P < 0.52, P < 0.0005, respectively). Both test scores declined with increasing age, earlier school leaving age and deteriorating visual acuity . Of other variables examined, only need for oral hypoglycaemic drugs or in sulin, history of stroke, dementia or Parkinson's disease and symptoms of a utonomic neuropathy significantly impaired one or more cognitive test score s. The odds ratios (95% CT) for normal cognitive test results in subjects w ith diabetes after adjusting for all significant variables was 0.74 (0.56, 0.97), P = 0.029 for MMSE scores and 0.63 (0.44, 0.93), P = 0.019, and 0.58 (0.38, 0.89), P = 0.013, for the numbers and hands parts of the clock test , respectively. In comparison with diabetic subjects with no evidence of co gnitive impairment, diabetic subjects with an MMSE score <: 23 were signifi cantly less likely to be involved in diabetes self-care (P<0.001) and diabe tes monitoring (P<0.001). A low MMSE score was also significantly associate d with higher hospitalisation in the previous year (P = 0.001), reduced ADL (activities of daily living) ability (P < 0.001)and increased need for ass istance in personal care (P = 0.001). Conclusions: Elderly subjects with pr edominantly Type 2 diabetes mellitus display significant excess of cognitiv e dysfunction, associated with poorer ability in diabetes self-care and gre ater dependency. Routine screening of cognition in older subjects with diab etes is recommended. (C) 2000 Elsevier Science Ireland Ltd. All rights rese rved.