Audit in practice Medication compliance and dispensing among psychogeriatric patients

Compliance with medication regimes is fundamental to successful treatment and hence familiarity with factors influencing the level of patient compliance is essential. In relation to the psychogeriatric popu lation poor compliance is a major and worrying problem largely neglected by researchers.


The study
Nineteen patients attending a psychogeriatric day hospital, and prescribed tablets, were included in a seven day compliance study. Sixsuffered from chronic functional mental illness, five from depression and one had chronic schizophrenia. The remaining 13 patients suffered from dementia and were living in the community, with input from a home help, a community psychiatric nurse and a social worker.
An estimation of compliance was made by count ing each patient's tablets on two occasions one week apart. Patients' relatives and home carers were not told that a tablet count was to be performed or that a compliance study was to be undertaken. They were merely told on two occasions one week apart that the doctor wanted to see the tablets. No new advice was given regarding medication compliance over the study period.
The following variables were considered: the number of medications prescribed: the number of times that a particular medication was prescribed each day; whether the prescription was issued before or more recently than four months prior to the study; whether the medication taking was supervised by relatives or home carers; the use of day labelled blister packs; whether the medication was for physi cal or mental problems and whether the patients were suffering from functional or organic mental illness.
Enquiries were made of staff members to ask for examples of where they had implemented any specific measures or offered advice to either relatives or carers regarding medication supervision or other issues relating to medication compliance.

Findings
The 19 patients had a total of 51 prescribed medi cations. Only one patient took his medication in the exact quantity prescribed and the average medication percentage compliance was 72.1%.
In analysing the results, a compliant patient was considered to be one taking the prescribed medi cation with 90% accuracy (Wandless et al. 1979). Medication compliance was poorer for medication prescribed more often than twice daily (Fisher's exact test P<0.0\), and for those medications taken for less than four months (/><0.01). Patients with functional mental illness had a significantly better level of compliance than those with organic mental illness (P<0.05). The day labelled blister packs showed a non-significant trend to improve patient compliance.
There was little difference in compliance noted with increasing numbers of prescriptions which is contrary to research evidence in the mentally well elderly (Hurd, 1986).Neither the origins of the persons super vising the medication administration nor whether the medication was for physical or psychiatric problems affected the level of compliance.
Several medications which had been discontinued with all parties informed continued to be taken regu larly. One of these was a beta blocker stopped because of peripheral vascular disease. Two patients had bottles of tablets rather inappropriately labelled "Take as directed by your doctor". Several patients had bottles containing more than 200 tablets and non-compliance was exaggerated in three patients because they failed to renew prescriptions.
No examples were given where relatives or home carers had received direct advice aimed to enhance medication compliance. Home care schedules had not been specifically arranged for the supervision of medication taking specifically for any patients and when this had occurred it was at the initiative of the home carer or by chance. Eight patients had their medication changed by the general practitioner without the awareness of psychiatric services.

Comment
This was a small study and it is possible that several of the non significant results were type 2 errors. This might explain why neither day labelled blister packs nor the number of prescribed medication had a sig nificant impact on compliance. The study, however, suggests that several commonsense factors such as tailoring medication regimes to allow supervision, avoiding unnecessary changes in medication and a good communication between the GP and psychogeriatric services play a major part in enhancing com pliance. Interestingly, the overall level of compliance for the psychogeriatric sample was similar to that found in a community sample of elderly people (Cartwright, 1988).
Several potentially dangerous situations were highlighted which included unclear labelling of bottles, the failure to renew prescriptions and patients continuing to take discontinued medication. This latter situation could only be prevented by physical removal of discontinued prescriptions from patients' houses.
Difficulties such as these are widespread and indi cate that this is an important area for audit.