Relationship of nutritional status to length of stay, hospital costs, and discharge status of patients hospitalized in the medicine service

Objective This study tvas conducted to determine the relationship, if any, between nutritional status, length of stay (LOS) in hospital, discharge placement, readmission rates, and hospital costs and charges in patients hospitalized in the medicine service. Design Data regarding medical diagnosis, LOS, hospital costs, charges, discharge destination, and readmission rates were collected prospectively from medical records and through patient interviews on patients admitted to the medical service mho were classified to be at risk or not at risk for malnutrition on the basis of established criteria (weight for height <75% ideal body weight, admission serum albumin level <30 g/L, or greater than or equal to 10% unintentional weight loss within 1 month before admission). Subjects Ail patients admitted directly to any of three medicine units during December 1994 who met study criteria were included in the study. Off-service patients, transfer patients, and patients discharged before screening (usually admitted and discharged within 72 hours) were excluded. Data were collected on 173 patients. Statistical analysis performed At-risk and not at-risk patients were compared for LOS, costs and reimbursement, and discharge placement (to home, to home with home health care services, or to another facility for further care). Two sample t tests and a survival analysis technique were used to compare continuous variables between the two study cohorts. Nonparametric tests were used for LOS and readmission data. chi(2) Tests were used for categoric variables. An a level of .05 was used throughout to determine statistical significance. Results Median LOS in the not-at-risk population (n=56) was significantly greater than in the not-at-risk population (n=117): 6 days (25th percentile=4 days, 75th percentile=8 days) vs 4 days (25th percentile=3 days, 75th percentile=7 days) (P<.01). Mean hospitalization cost per patient was also higher in the at-risk group ($6,196 vs $4,563, P<.02). Readmission rate per month of follow-up was not significantly different. At-risk patients were significantly less likely to be discharged home with self-care (23 [41%] vs 77 [66%], P<.05). At-risk patients were significantly more likely to use home health care service than not-at-risk patients (17 [31%] vs 14 or [12%], P<.001). Applications Patients at risk for malnutrition had significantly higher LOS, costs, and home health care needs, despite the fact that 51, or 91%, received nutrition intervention while hospitalized. Further research should explore the use of nutrition screening and intervention before, during, and after hospitalization to ensure that appropriate nutrition intervention, as indicated by medical patients' clinical condition and nutritional risk status, is initiated and continued.