OBJECTIVE: This study aimed to investigate the relationship between serum albumin levels, drugs used, length of hospitalization and number of diseases with the prognosis of hospitalized elderly. METHODS: A descriptive cross-sectional and retrospective study was carried out using a review of medical charts. Variables collected were disease motivating admission, albumin, drugs used and their affinity for albumin and length of stay in an elderly population of a Brazilian public hospital. We used multiple regression to assess whether age, length of stay, medication with or without affinity to albumin and number of drugs motivating admission were causes of death. The relation between dependent and independent variables was analyzed by two-way ANOVA. RESULTS: The sample consisted of 277 elderly, 151 with hypoalbuminemia and those using an average of 6.6 +/- 3.80 different prescription drugs most of which had affinity for albumin (4.96 +/- 2.64). Results showed that the number of drugs with affinity for albumin and hospital stay were predictors of death (r = 0.47, R2 = 0.224), (F(8.170) = 6.13; p=0.001). The relation between dependent and independent variables resulted in the mathematical equation: (death= 0.545 - 0.374 (albumin) - 0.195 (days of hospitalization) + 0.175 (number of drugs with affinity for albumin)). CONCLUSION: The mathematical equation obtained by this study demonstrated that albumin and days of hospitalization were inversely correlated with death and directly correlated with the number of drugs bound to albumin.
OBJECTIVE: This study aimed to investigate the relationship between serum albumin levels, drugs used, length of hospitalization and number of diseases with the prognosis of hospitalized elderly. METHODS: A descriptive cross-sectional and retrospective study was carried out using a review of medical charts. Variables collected were disease motivating admission, albumin, drugs used and their affinity for albumin and length of stay in an elderly population of a Brazilian public hospital. We used multiple regression to assess whether age, length of stay, medication with or without affinity to albumin and number of drugs motivating admission were causes of death. The relation between dependent and independent variables was analyzed by two-way ANOVA. RESULTS: The sample consisted of 277 elderly, 151 with hypoalbuminemia and those using an average of 6.6 +/- 3.80 different prescription drugs most of which had affinity for albumin (4.96 +/- 2.64). Results showed that the number of drugs with affinity for albumin and hospital stay were predictors of death (r = 0.47, R2 = 0.224), (F(8.170) = 6.13; p=0.001). The relation between dependent and independent variables resulted in the mathematical equation: (death= 0.545 - 0.374 (albumin) - 0.195 (days of hospitalization) + 0.175 (number of drugs with affinity for albumin)). CONCLUSION: The mathematical equation obtained by this study demonstrated that albumin and days of hospitalization were inversely correlated with death and directly correlated with the number of drugs bound to albumin.