FACTORS ASSOCIATED WITH COGNITIVE IMPAIRMENT AMONG OLDER ITALIAN INPATIENTS

OBJECTIVE: To examine the association of cognitive impairment with edu cational, demographic, and nutritional variables in older hospitalized people. DESIGN: Survey of older patients admitted consecutively to a hospital during two 2-month periods in 1993. SETTING: Patients admitte d for general medical care at 35 hospitals participating in the GIFA s tudy throughout Italy. PARTICIPANTS: A total of 3628 patients aged 65 or older were studied. MEASUREMENTS: The Hodkinson Abbreviated Mental Test (HAMT) was used as a screening method to assess the patients' bas ic cognitive function. Multiple logistic regression was used to examin e the association between cognitive impairment and demographic, educat ional or nutritional variables. RESULTS: Twenty-nine percent of older inpatients were classified as having cognitive impairment, with Simila r distribution of HAMT score found in both genders. Educational attain ment has a highly significant inverse relationship with cognitive impa irment (highest education: OR 0.32; 95% CI 0.20-0.52). Moreover, cogni tive impairment decreased with increasing body mass index (3rd tertile : OR 0.69; 95% CI: 0.51-0.93), cholesterol serum level (highest values : OR 0.59; 95% CI 0.37-0.93), circulating lymphocytes (highest values: OR 0.55; 95% CI 0.45-0.69), and serum albumin (highest values: OR 0.6 0; 95% CI 0.47-0.76), with a gradient of influence for each variable. CONCLUSIONS: Educational attainment affects cognitive function in olde r inpatients. The strong association between cognitive impairment and nutritional variables suggests that every effort to improve nutritiona l status is needed in approaching cognitive impairment in older patien ts.