Association between Total Daily Doses with duration of hospitalization among readmitted patients in a multi-ethnic Asian population.

Increased length of stay (LOS) in the hospital incurs substantial financial costs on the healthcare system. Multiple factors are associated with LOS. However, few studies have been done to associate the impact of Total Daily Doses (TDD) and LOS. Hence, the aim of this study is to examine the association between patients' LOS upon readmission and their TDD before readmission. A retrospective cross-sectional study of readmission cases occurring from 1st January to 31st March 2013 was conducted at a regional hospital. Demographics and clinical variables were collected using electronic medical databases. Univariable and multiple linear regressions were used. Confounders such as comorbidities and drug related problems (DRP) were controlled for in this study. There were 432 patients and 649 readmissions examined. The average TDD and LOS were 18.04 ± 8.16 and 7.63 days ± 7.08 respectively. In the univariable analysis, variables that were significantly associated with the LOS included age above 75 year-old, race, comorbidity, number of comorbidities, number of medications, TDD and thrombocytopenia as DRPs. In the multiple linear regression, there was a statistically significant association between TDD (β = 0.0733, p = 0.030) and LOS. Variables that were found significant were age above 75 year-old (β = 1.5477, p = 0.008), Malay (β = -1.5123, p = 0.033), other races (β = -2.6174, p = 0.007), depression (β = 2.1551, p = 0.031) and thrombocytopenia as a type of DRP (β = 7.5548, p = 0.027). When TDD was replaced with number of medications, number of medications (β = 0.1487, p = 0.021), age of 75 year-old (β = 1.5303, p = 0.009), Malay (β = -1.4687, p = 0.038), race of others (β = -2.6499, p = 0.007), depression (β = 2.1951, p = 0.028) and thrombocytopenia as a type of DRP (β = 7.5260, p = 0.028) were significant. In conclusion, a significant relationship between TDD and number of medications before readmission and the LOS upon readmission was established. This finding highlights the importance of optimizing patients' TDD in the attempt of reducing their LOS.