Polydipsia and water intoxication in a long-term psychiatric hospital.

This cross-sectional survey attempts to establish the prevalence of polydipsia and water intoxication at a state hospital (N = 360) using staff diagnosis, specific gravity of the urine (SPGU), weight changes, and chart review. There were 150 [42%, 95% confidence interval (CI) 37-47%] patients diagnosed as polydipsic by the staff or by SPGU. At least 93 (26%, CI 21-30%) had primary polydipsia not explained by other causes. Chart review identified 17 (5%, CI 3-7%) patients with a history of water intoxication. Using a case-control study design, schizophrenia, extended duration of hospitalization, and heavy smoking were associated with primary polydipsia in a logistic regression analysis (respective odds ratios were 1.6, 1.8, and 3.6). All patients with a history of water intoxication were Caucasian (versus 83% in those without a history) and had significantly more extended hospitalizations (94 vs. 49%). Future case-control studies should combine longitudinal identification of true cases and controls and exhaustive collections of clinical information in a standardized way.