Seven studies (767,200 participants) were included. There were four controlled studies (755,464 participant: range 59 to 731,105). The quality of three controlled studies was good; one was acceptable. The quality of one uncontrolled study was good; two were acceptable.
Appropriateness of medication use (three controlled studies; two uncontrolled studies):
Higher prevalences of potential inappropriate drug use were generally found in participants using automated dose dispensing, compared to standard dispensing procedure. This included a higher risk of using anticholinergic and three or more psychotropic drugs (ORs 1.43 to 4.93, 95% CI 1.40 to 5.17; one controlled study; 731,105 participants). However, the same study showed that women using automated dose dispensing used fewer long-acting benzodiazepines and lower numbers of drug-drug interactions were observed in women and men (ORs 0.69 to 0.80, 95% CI 0.66 to 0.83). Another controlled study (24,146 participants) showed a higher risk of potential inappropriate drug use in recipients of automated dose dispensing (ORs 1.36 to 5.48, 95% CI 1.18 to 6.30). Automated dose dispensing was associated with more instances of unchanged drug regimens (OR 1.66, 95% CI 1.20 to 2.31; one controlled study; 154 participants).
Medication safety (one controlled study; one uncontrolled study):
Discrepancies in medication records between general practitioners and home care services were reduced by 34% (p<0.001) with automated dose dispensing (one controlled study; 59 participants).
The results of uncontrolled studies were reported in the paper.