With an increase in the global prevalence of dementia, there is also an increase in behavioural and psychological symptoms of dementia (BPSD) for which antipsychotic drugs are often used. Despite several safety warnings on antipsychotic use in dementia, there is little evidence to support the efficacy of antipsychotics in individual BPSD symptoms or to evaluate the drug safety profile by individual antipsychotic drug. There is emerging but scarce evidence that suggests an inter-drug variability between antipsychotic safety outcomes in BPSD. The objective of this review was to examine the existing literature on antipsychotic drug use in dementia patients; in particular to see whether inter-drug differences regarding antipsychotic safety were reported. A literature search was conducted for observational studies published in the English language from 2004 to 2014 that reported the risk of all-cause mortality, cerebrovascular events, pneumonia and other outcomes such as hip/femur fracture, deep vein thrombosis (DVT) and hyperglycaemia. Six of 16 mortality studies (38%), 7 of 28 stroke studies (25%), 1 of 6 pneumonia (17%) studies and 2 of 6 fracture studies (33%) investigated inter-drug safety outcomes in elderly patients/dementia patients, while to our knowledge, there are no studies investigating the inter-drug variation of deep-vein thrombosis and hyperglycaemia risk. The results of the observational studies provide mixed results on the safety of antipsychotics in BPSD but it is clear that there are differences between the safety profiles of antipsychotic drugs. Robust evidence of such inter-drug variability could significantly improve patient safety as antipsychotics become more targeted to clinical risk factors.