Key messages

It is uncertain whether older, first-generation or ‘typical’ antipsychotic medicines such as haloperidol have an effect on agitated behaviour (for example, restlessness and aggression); the effect is moderate at best. Typical antipsychotic medicines may decrease delusions and hallucinations slightly in people with dementia. 

Newer, second-generation ‘atypical’ antipsychotic medicines, such as risperidone, probably reduce agitated behaviour slightly. Atypical antipsychotic medicines probably have no effect on psychotic symptoms. 

Both first- and second-generation antipsychotic medicines increase the risk of drowsiness and other unwanted events. When patients’ symptoms improve after antipsychotics have been prescribed, this is probably largely due to natural improvement in symptoms over time.

What are antipsychotic medicines?

Antipsychotics are medicines prescribed to treat psychotic symptoms and severely disturbed behaviour in some mental illnesses, such as schizophrenia, bipolar disorder and severe depression. Psychotic symptoms are delusions (very strongly held beliefs in something which is not true) and hallucinations (sensing – usually seeing or hearing - things which are not really there).

Antipsychotic medicines are often divided into two groups:

1. first-generation (older) or ‘typical’ antipsychotics, for example haloperidol;

2. second-generation (newer) or ‘atypical’ antipsychotics, for example risperidone.

Both types can cause unwanted effects, such as drowsiness, movement disorders (for example, involuntary or uncontrollable movements, tremors, muscle contractions) and weight gain.

Why do people with dementia need antipsychotics?

People with dementia quite often experience hallucinations and delusions during their illness for some time. Particularly in the later stages of the illness, they may also show agitated behaviours such as restlessness, shouting out or aggression towards others. It is important to try to understand what is driving these behaviours and there are many ways to manage them which do not involve drugs. However, antipsychotic medicines have often been prescribed to people with dementia for these problems. In many countries, they are prescribed less often than in the past but are still used when the symptoms are severe.

What did we want to find out?

We wanted to know how well antipsychotic medicines reduce the severity of agitation and psychotic symptoms in people with the two commonest types of dementia, namely dementia due to Alzheimer’s disease and vascular dementia. We also wanted to know how many people experienced unwanted effects.

What did we do?

We searched for studies that investigated antipsychotic medicines currently available in the USA or European Union by comparing them with placebo (a ’dummy’ pill), for treatment of persistent agitation or psychotic symptoms. People in the studies had to have Alzheimer’s disease or vascular dementia. They could be any age and reside in a care home, a hospital, or the community. Most of the people in the studies had to be experiencing agitation (including aggression) or psychotic symptoms, or both, at the start of the study.

We compared and summarised the results of the studies and rated our confidence in the evidence, based on factors such as study methods and sizes.

What did we find?

We found 24 studies with a total of 6090 people:

- six studies tested typical antipsychotics, mostly haloperidol; 

- 20 studies tested atypical antipsychotics, such as risperidone, olanzapine, and aripiprazole; and 

- two studies tested both typical and atypical antipsychotics.

All the studies compared antipsychotics with placebo. The people were living in institutions, hospitals, the community, or a combination of these settings.

Main results

Typical antipsychotics (haloperidol, thiothixene) compared with placebo:

- may improve symptoms of psychosis slightly (2 studies, 240 people), but we are uncertain about their effect on agitation (4 studies, 361 people); 

- probably increase the risk of drowsiness (3 studies, 466 people), and movement disorders (3 studies, 467 people); 

- may slightly increase the risk of serious unwanted effects (1 study, 193 people) and of death (6 studies, 578 people). 

There was no evidence about the risk of non-serious and serious unwanted effects combined.

Atypical antipsychotics (risperidone, olanzapine, aripiprazole, quetiapine) compared with placebo:

- probably slightly reduce agitation (7 studies, 1971 people) and may slightly reduce aggression (1 study, 301 people), but probably make no important difference to symptoms of psychosis (12 studies, 3364 people);

- increase the risk of drowsiness (13 studies, 2878 people) and probably slightly increase movement disorders (15 studies, 4180 people);

- probably slightly increase the risk of experiencing any non-serious or serious unwanted effect combined, the risk of serious unwanted effects, and the risk of death (17 studies, 5032 people).

What are the limitations of the evidence?

Overall, our confidence in the evidence about typical antipsychotics is limited and our confidence in the evidence about atypical antipsychotics moderate. Typical antipsychotics have been investigated in just a few studies. In addition, the studies about typical and atypical antipsychotics did not always use the best methods to carry out their investigations, or did not report the results. Consequently, the effects on agitation or psychosis may have been overestimated, and the effects on adverse events underestimated.

How up to date is this evidence?

The evidence is up-to-date to 7 January 2021.