Results

A total of 411 patients were included in the study. The mean age of our patients were 83·8 ± 8·0 years, 40·6% male. Table 1 summarises patient demographics and characteristics. 38·4% of our patients were admitted with a fall. The median age was 86·0 (Q1 79·0; Q3 90·0) and 84.0 years (Q1 78·0; Q3 90·0) respectively for patients admitted with and without a fall (difference: 2·0 , p=0·125); (table 2a and 2b for univariate and bivariate analysis). Age was associated with an increased rate of patients admitted with a fall per each 1-unit rise (OR=1·03, [1·0 ; 1·05], p= 0·0494).
The mean CCI of our study group was 5·7 ± 1·7. The mean CCI in patients admitted with and without a fall was 5·65 ± 1·66 and 5·74 ± 1·76 respectively. The median CCI were respectively 5·0 (Q1 5·0; Q3 6·0) and 5.0 (Q1 5·0; Q3 7·0) for patients admitted with and without fall (difference: 0.0, p=0·549); (table 3a and 3b for univariate and bivariate analysis). CCI was not associated with the rate of patients admitted with a fall with each 1-unit rise, (OR=0·92, [0·81 ; 1·04], p= 0·1723).
Overall incidence of polypharmacotherapy of our study group was 80·8%. Polypharmacotherapy in patients admitted without and with a fall were 76·3% and 88·0% respectively. There was an association between polypharmacotherapy and an increased risk of fall (OR=2·27 [1·3;3·98],p=0·005); (table 4a and 4b for univariate and bivariate analysis of polypharmacotherapy). Table 5 lists commonly prescribed drugs (incidence of >5%) with ACB within our study group. Incidence of ACB scores of 0, 1, 2, ≥3 was 38·7%, 20·9%, 14·6% and 25·8% respectively. The median ACB scores were respectively 2·0 (Q1 0·0; Q3 3·0) and 1·0 (Q1 0·0; Q3 2·0) for patients admitted with and without fall (difference: 1·0 , p=0·009); (table 6a and 6b for univariate and bivariate analysis of ACB score). ACB score was associated with an increased rate of patients admitted with a fall per each 1-unit rise in ACB, (OR=1·15, [1·02 ; 1·29], p= 0·0245). In our linear regression model, a patient with an ACB score of 3 compared to a patient with an ACB score of 0 would have a >50% higher chance of falling.
Of the patients who were admitted with fall (38·4% of total study group), 29·8% was found to have drug-related orthostatic hypotension, 24·7% had drug-related bradycardia, 37·3% were taking regular centrally acting agents and 12·0% were taking inappropriate hypoglycaemic agents. Table 7 summarises the incidence of secondary outcomes in patients admitted with fall. Table 8 summarises the regression results of the study.