Antipsychotics

In the USA, the Nursing Home Reform Amendments saw antipsychotic prescribing fall by a third, and use has since been reported to vary between 20% and 50% (Reference Shorr, Fought and RayShorr et al, 1994). The amendments limit the use of antipsychotic medication to specific clinical conditions and demand regular medication review and documentation of decisions. Homes are fined if they fail to adhere to the amendments.

In Ireland, 19% of older people in nursing homes were reported to be taking phenothiazines (Reference Panmore, Crawford and BeringerPanmore et al, 1995). This figure will probably be lower now, following a tightening recently of the licensed indications for thioridazine. In the UK, a study of south Manchester nursing homes showed that 30% of residents were taking antipsychotics (Reference Furniss, Burns and CraigFurniss et al, 2000).

Two studies have looked at the appropriateness of antipsychotic prescribing in the UK. In Scotland, where antipsychotic use in nursing homes was reported to be 24%, McGrath & Jackson (1996) found that 88% of these prescriptions were inappropriate if the US criteria for use was applied. This study relied solely on the use of written records to define appropriateness. In our study (Reference Furniss, Burns and CraigFurniss et al, 2000) my colleagues and I interviewed staff and found a lower figure of 54% of antipsychotics' prescriptions inappropriate according to the US criteria. Recent work from Denmark has suggested that behavioural problems were a determinant for the use of antipsychotics (and benzodiazepines), irrespective of the psychiatric diagnosis of the resident (Reference Sorensen, Foldspang and GulmannSorensen et al, 2001). The authors conclude that staff perceptions of psychiatric morbidity have a greater impact on the prescription of psychotropics than do standardised clinical criteria.

Atypical antipsychotics

At the time of writing, none of the atypical drugs (as defined in the BNF) are licensed for the treatment of agitation in the older person and antipsychotics should be used cautiously (if at all) for this indication. These newer drugs have a different side-effect profile than the typical ones. They cause fewer movement disorders, but may cause comparatively more of other side-effects such as drowsiness. For example, a study comparing the use of olanzapine with placebo showed that somnolence and gait disturbances were more common with the active treatment, which may lead to increased falls and failure to meet standard 6 of the NSF (Reference Street, Clark and GannonStreet et al, 2000). However, the study was only placebo-controlled, so it could not determine how the drug compares with typical agents in this respect.

Antidepressants

The prevalence of depression in nursing homes varies between 12% and 32% (Reference Heston, Garrard and MakrisHeston et al, 1992). Given its high prevalence and serious consequences, appropriate identification and treatment of depression must be a priority for health professionals. Antidepressants have been reported to be underused in nursing homes, their rate of use being between 6% and 15% (Reference Harrington, Tompkins and CurtisHarrington et al, 1992). However, this figure has increased to 25% in recent years (Reference Furniss, Burns and CraigFurniss et al, 2000).

Hypnotics

A number of studies (e.g. Reference Furniss, Burns and CraigFurniss et al, 2000) over the past decade have found that 25–30% of private nursing home residents regularly take hypnotics. Significantly fewer patients in hospital continuing care wards take these medicines. Traditionally when benzodiazepines have been used, those with a shorter half-life have been recommended. However, recent work has shown that although the risk of falls among patients taking such medicines is less than for the long-acting agents, all benzodiazepines are associated with a materially increased risk of nocturnal falls (Reference Ray, Thapa and GideonRay et al, 2000). If benzodiazepines are to be withdrawn, this should be done gradually and residents may need greater staff involvement during this time.

Unwanted effects of psychotropics

Regular use of psychotropic medication is associated with an increased risk of recurrent falls (Reference Thapa, Gideon and FoughtThapa et al, 1995) and of fractures in the neck of the femur (Reference Lauritzen, McNair and LundLauritzen et al, 1993). Another study used case–control methods to explore the relationship between falls, drugs and diagnosis in elderly residents of a nursing home (Reference Granek, Baker and AbbeyGranek et al, 1987). It found that those taking sedatives and antidepressants were at a greater risk of falling. Medication changes are particularly important in this respect: a change during the previous 2 weeks is a significant risk factor for falling.

The beneficial effects of antipsychotic drugs must be balanced against extrapyramidal and other (e.g. constipation, drowsiness) side-effects. Long-term use is associated with tardive dyskinesia. Studies have found associations between the use of antipsychotic drugs and restlessness, wandering, falls and incontinence of urine (Reference Granek, Baker and AbbeyGranek et al, 1987; Reference Nygaard, Bakke and BreivikNygaard et al, 1990). Antipsychotic drug use may be associated with an increased rate of cognitive decline in dementia (Reference McShane, Keene and GedlingMcShane et al, 1997).

Educational programmes

Although regulations in the USA impose penalties for inappropriate use of antipsychotics in nursing homes, guidance is needed generally on how to address behavioural problems. It is useful here to identify two studies that have looked at the provision of educational programmes to medical and nursing staff employed in nursing homes and the effects of such programmes on prescribing (Reference Avorn, Soumerai and EverittAvorn et al, 1992; Reference Ray, Taylor and MeadorRay et al, 1993). Drug use and clinical status were used as outcome measures. Both studies showed a reduction in the use of antipsychotic drugs with no deterioration in functioning of residents. Neither study showed an increase in disruptive behaviour, but one showed an increase in the level of depression in the group with the greatest drug reduction. This could imply that the sedative effects of psychotropic drugs were masking depressive symptoms or that antidepressants were inappropriately stopped in some instances. However, another study has shown an improvement in the affect of residents whose antipsychotics were discontinued, which suggests that the former explanation is more likely (Reference Thapa, Meador and GideonThapa et al, 1994). The education of physicians alone showed no effect on the prescription of antipsychotic medication. Together these three studies imply the need to involve all the staff in a nursing home in any educational programmes.

USA

In the USA, nursing homes are required by law to employ a consultant pharmacist to review prescribed medication every 1–3 months. They advise medical practitioners on matters such as choice, duration and monitoring of drug therapy and also participate in staff education about medication. Physician acceptance of pharmacist recommendations is high and has been reported to be between 60% and 81% (Reference Dyer, Oles and DavisDyer et al, 1984).

In one study, clinical pharmacists assumed total responsibility for the drug management of a group of elderly residents in a nursing home; this group was then compared with a control group who received traditional care (Reference Thompson, McGhan and RuffaloThompson et al, 1984). The pharmacists' group had improved mortality and morbidity and took approximately two fewer drugs. A potential saving of about $70 000 per year per 100 residents was identified. Other studies (e.g. Reference Harrison, Bootman and CoxHarrison et al, 1998) have shown similar economic benefits and similar reductions in the number of drugs used per patient and also demonstrate the usefulness of the pharmacist in the eradication of unnecessary and potentially harmful medication use. Drawbacks of these studies are that most are not controlled and they do not measure objective resident outcomes.

UK and Europe

In the UK, the pharmacist's role is primarily one of supply and the provision of basic advice to the nursing home about documentation and storage. A more comprehensive clinical service is rarely provided. The situation is similar throughout northern Europe, but it is beginning to change. The targets given in the NSF for Older People regarding individualised medication review (with patient and carer) require that this clinical input be increased (Department of Health, 2001).

One study has shown that 1 hour per week of a pharmacist's time can make a significant contribution to patient care in nursing homes. It found that this input was well received by nursing staff and prescribers and that GPs accepted the pharmacist's advice in 78% of cases (Reference LapsleyLapsley, 1998). Physician acceptance was higher (91%) in south Manchester, where 55% of interventions resulted in actual treatment modifications. In Northamptonshire, community pharmacists analysed prescriptions of nursing home residents and provided prescribing advice to the GPs. The GPs agreed with the advice in 73% of cases and it was estimated that pharmacist involvement could give a 14% reduction in the cost of prescribing.

More recent work has used randomised controlled trials (RCTs) and objective resident outcome measures to strengthen the earlier evidence base. An RCT in 33 nursing homes in Sweden demonstrated that multi-disciplinary meetings (doctor, pharmacist and nurse) reduced prescribing of psychotropic drugs and prevented polypharmacy and therapeutic duplication (Reference Schmidt, Claesson and WesterholmSchmidt et al, 1998). Similar work in England in 14 nursing homes showed that a brief medication review reduced the amount of medication overall with no detriment to the mental and physical functioning of the patients (Reference Furniss, Burns and CraigFurniss et al, 2000). It also showed a reduction in the use of primary and secondary care resources by the pharmacist medication review group. Structured assessment and review of residents, educational interventions and greater multi-disciplinary working have been suggested as ways of improving the care of residents in homes (Reference Hughes, Lapane and MorHughes et al, 1999). Prescribing advice from pharmacists must be supplied on a continual basis if reductions in medication are to be maintained in the nursing home. Although the ideal frequency has not been determined, if the NSF is followed then most residents should be receiving a 6-monthly review.

The recommendations suggested by pharmacists include stopping and starting medicines, generic substitution, changes to another medicine, dose modification, changes in administration frequency, formulation change and requests for laboratory tests or nurse monitoring. Examples of reasons for such recommendations are given in Box 2. In our study, my colleagues and I found that almost half of all recommendations were to stop medication and two-thirds of these were because there was no indication for the drug prescribed (Reference Furniss, Burns and CraigFurniss et al, 2000). This suggests that medication regimes are not reviewed thoroughly. Conversely, initiation of medication made up 8% of recommendations, which implies that indications are present but not always treated. This might have been due to age discrimination, which is not to be tolerated since the publication of the NSF for Older People. Fifteen per cent of recommendations were for laboratory test requests (e.g. blood profiles or therapeutic drug monitoring), suggesting that laboratory test monitoring of drug therapy is not often carried out in nursing homes. It can be seen that the majority of interventions were to improve patient care, but a small number, such as generic prescribing, were for purely economic reasons.

Pharmacists offering specialist clinical services to nursing homes will often have received postgraduate training in therapeutics, usually to Masters degree level, although many interventions require only a basic knowledge of drug treatment in elderly people. Health authorities may enter into a contract for an advisory service with local community pharmacists or it may be purchased from local hospital pharmacy departments, thus using clinical pharmacists. A clinical pharmacist is employed on the Isle of Wight with the sole remit of reviewing medication regimes in nursing homes.

As well as issuing the repeat prescriptions for residents, GPs may themselves carry out detailed medication reviews, and this too has been shown to reduce the prescribing of inappropriate medicines (Reference Khunti and KinsellaKhunti & Kinsella, 2000). It is a requirement of the GP contract in the UK that all patients over 75 receive an annual medication review. Given the additional recommendations within the NSF for Older People concerning 6-monthly review for people taking four or more medicines, there are significant implications here for the workload of GPs. Nursing home residents may benefit from a multi-disciplinary review of their medicines. Indeed, the GPs in our study (Reference Furniss, Burns and CraigFurniss et al, 2000) were keen to discuss with the pharmacist the use of medication in residents. However, they often described a lack of time to carry out adequate medication reviews, both in this population and in other patients on chronic multiple-drug therapy.