Study designs of evaluations included in the review
The included studies were randomised controlled trials (RCTs) of at least one antipsychotic drug with double-blind assessment and placebo-control, or comparisons with an active medication. The length of treatment had to be at least 4 weeks.
Specific interventions included in the review
Antipsychotics including acetophenazine, trifluoperazine, haloperidol, thioridazine, thiothixene, loxapine and oxazepam. The comparators included placebo, diazepam, thioridazine, haloperidol, cis(Z)-clopenthixol, chlormethiazole and diphenhydramine. The duration of treatment ranged from 3 to 12 weeks.
Participants included in the review
The participants were typically patients with primary dementia (greater than 70% of the patients) who were being treated for behavioural disorders. The diagnostic descriptors of the patients were as follows:
organic brain syndrome with psychosis (n=64);
schizophrenia (n=7);
depression and psychosis (n=1);
chronic brain syndrome with (senile) psychosis (n=73);
nonpsychotic organic brain syndrome (n=24);
senile dementia (n=77);
presenile dementia (n=7);
multi-infarct dementia (n=65);
senility (n=56);
senile arteriosclerosis (n=18);
primary (degenerative) dementia (n=98);
other organic dementia (n=5);
alcoholic dementia (n=1);
alcoholism (n=8);
Korsakoff (n=4);
head trauma (n=1);
dementia, Alzheimer and arteriosclerotic forms (n=74);
major strokes (n=4);
previous depression (n=1);
unknown (n=2);
unavailable (n=113);
primary (degenerative) dementia or Alzheimer's, multi-infact dementia or senile dementia (n=40).
Outcomes assessed in the review
The outcomes assessed were the proportion of respondents who were responders, had side-effects or dropped-out. In addition, behavioural outcomes were measured using the following scales: Motility Affect Cooperation Communication Scale; Psychotic Reaction Profile; Hamilton Rating Scale for Anxiety; Nurses' Observation Scale for Inpatient Evaluation; global rating; Brief Psychiatric Rating Scale; Clinical Global Impression; psychiatric evaluation form; Gottfries-Cronholm Geriatric Rating Scale; Crichton Geriatric Rating Scale; Sandoz Clinical Assessment-Geriatric; Crichton Geriatric Behavioral Rating Scale; Behavior Rating Scale of the Clifton Assessments Procedures for the Elderly; verbal rating scale for agitation; Alzheimer Disease Assessment Scale; 'agression scale'; Cohen-Mansfield Agitation Inventory; and an unspecified scale.
How were decisions on the relevance of primary studies made?
Three raters were given a sample of the methods section of the 51 collected papers, to assess for adherence to the inclusion criteria. Each rater was given 13 to 15 papers to rate individually plus 10 randomly selected papers to test inter-rater reliability. The raters were blinded to the authors, date, journal and place of publication of the papers.