Background: To determine whether nutrition is associated with rehabilitation outcome in the disuse syndrome.
Methods: A retrospective cohort study was performed in 223 inpatients admitted to a university hospital who were diagnosed by physicians in the rehabilitation department as having the disuse syndrome, and subsequently prescribed physical therapy. Malnutrition was defined as a body mass index<18.5 kg/m², hemoglobin level<10.0 g/dl, serum albumin level<3.0 g/dl, or total lymphocyte count<1200 cells/mm³. Rehabilitation outcome was defined as whether or not the ADL score improved during rehabilitation. Nutritional status was assessed at referral using the Onodera's prognostic nutritional index (PNI).
Results: The study cohort included 136 men and 87 women (mean age 67.5 years; median duration between admission and referral 17 days; median rehabilitation duration 32 days). A total of 202 patients (91%) were defined as being malnourished. Mean PNI was 32.9, with the ADL score improving in 135 patients (61%) during rehabilitation. Rehabilitation outcome was better in patients with normal nutrition compared to malnourished patients (relative risk: 0.72, p=0.04). Patients with a hemoglobin level>10.0 g/dl (relative risk: 0.69, p=0.001), total lymphocyte count>1200 cells/mm³ (relative risk: 0.78, p=0.03), or PNI>35.0 (relative risk: 0.74, p=0.01) had a better rehabilitation outcome. Logistic regression analysis showed that hemoglobin level was associated independently with rehabilitation outcome (odds ratio 2.34, p=0.005).
Conclusions: Malnutrition is common in patients with the disuse syndrome. Patients with low hemoglobin level and PNI at referral are more likely to have a poor rehabilitation outcome.

Objective: To develop a weighted disuse risk score rating. Design : Based on specified selection criteria, items of the disuse risk factor were chosen. Then, a conjoint questionnaire for physicians was created using orthogonal cards with ordinal disuse risk grade. Conjoint analysis (SPSS 15.0) was performed, weighting each item and making a formula for calculating the disuse risk score. Setting : The Rehabilitation Department of a University Hospital. Participants : Forty-one physician respondents and 129 consecutive inpatients referred to the rehabilitation department. Main outcome measures : Agreement of the rated disuse risk grade with the calculated disuse risk score, and the reliability of the calculated disuse risk score of the 129 inpatients. Results : Eleven disuse risk factor items were selected ; age, malignant tumor, heart impairment, respiratory impairment, kidney impairment, pain, gait disorder, psychiatric impairment, intellectual disability, dementia, and body mass index. Forty-one physicians completed and returned a postal questionnaire. Cronbach's a of the rated disuse risk grade was 0.948. The correlation coefficient of the average rated disuse risk grade to the calculated disuse risk score was 0.985 (p=0.000). Items of high overall importance (%) were gait disorder (14.0%), malignant tumor (11.5%), and respiratory impairment (11.0%). Those of low overall importance (%) were body mass index (6.8%), kidney impairment (6.6%), and psychiatric impairment (6.1%). For the 129 patients, the calculated disuse risk score of the 11-item was correlated to that of the 9-item weighted disuse risk factor (R=0.930, p=0.000). Conclusions : Both the 11-item disuse risk factor and the weighted disuse risk score rating were considered reliable and useful.