- Top of page
Aim: To investigate the prevalence of medication-related problems (MRPs) in patients attending aged care and memory disorder clinics and explore the potential role of a clinical pharmacist to obtain medication histories and identify unresolved MRPs.
Methods: The clinical pharmacist interviewed patients and reviewed their medication regimens in the outpatient clinics. Clinical significance of pharmacist-identified MRPs was rated by an independent expert panel using validated criteria.
Results: Forty-six patients (mean age 82 years) were reviewed. They took a median of nine medications, of which three were not recorded in the medical record. One hundred and thirteen MRPs (median 2.0 per patient) were identified by the pharmacist. Independent review rated 35% of MRPs as high or extreme risk. Thirty-seven (33%) MRPs related to medications not recorded in the medical record.
Conclusions: Medication-related problems were present for most patients. Involvement of a clinical pharmacist resulted in a more comprehensive medication history and identified unresolved MRPs.
- Top of page
At least 40 000 older Australians are hospitalised each year as a result of medication-related problems (MRPs), representing 20–30% of unplanned hospital admissions for this age group [1,2]. The prevalence of adverse drug reactions (ADRs) in older people with chronic illness is 15–20% [3,4].
Patients referred to specialist aged care and memory disorder clinics are likely to be at high risk for MRPs due to factors such as polypharmacy and cognitive dysfunction [5–11]. Although they are not usually referred primarily for a comprehensive medication review, this might be an opportune time to conduct such a review . Several studies have demonstrated clinical and cost benefits associated with clinical pharmacy services in outpatient clinics for older people [13–18], but there has been no Australian study conducted in the aged care outpatient setting and no study internationally conducted in a memory disorder clinic.
The aims of this study were to investigate the prevalence of clinically important MRPs in older patients referred to the aged care assessment or memory disorder clinics at a tertiary care hospital, and explore the potential role of an outpatient clinical pharmacist to obtain medication histories and identify unresolved MRPs.
- Top of page
This was a cross-sectional study involving a convenience sample of patients who attended either the aged care assessment or memory clinic between July 2007 and January 2008. Patients were eligible for inclusion if they were older than 65 years, taking regularly scheduled medications, not residing in an aged care home and presented to clinic on a day when the pharmacist was in attendance (75% of clinic sessions over the study period). Both new and review patients were included but no patient was included more than once.
Eligible patients were invited by the pharmacist to participate in the study (some eligible patients were not approached, if they arrived when the pharmacist was interviewing another patient). Patients (and their carers, where applicable) who agreed to participate were interviewed by the pharmacist in the clinic after their medical consultation. The pharmacist took a medication history and administered a structured questionnaire, based on validated questionnaires [19,20], to obtain consistent and valid information about medication management, medication adherence and self-reported ADRs. Patients' community pharmacies were contacted to clarify medication history details and/or confirm medication adherence in cases where the patient or carer was unable to provide a medication history or where there were doubts about the accuracy of the history provided. After the interview the pharmacist reviewed the patients' medication regimen in conjunction with their medical history (obtained from the hospital medical record and clinic referral correspondence) to identify potential MRPs.
The MRPs were classified by the outpatient pharmacist into nine categories based on the Strand et al. classification system: potentially unnecessary medication, inappropriate medication choice, dose too low, dose too high, ADR, drug interaction, non-adherence, untreated indication and medication management problem .
Risk of an adverse outcome resulting from MRPs identified by the pharmacist was assessed independently by a consultant geriatrician and a senior clinical pharmacist who were not involved in providing services to the participating clinics. They were provided with an extract of the case notes and a description of the MRPs, and rated the risk using a validated system based on Australian Standards for Risk Management, where risk is based on an estimate of the likelihood and severity of an adverse outcome if no intervention was made [22,23]. They also reviewed the pharmacist recommendations to assess whether the benefit of the recommended intervention was likely to outweigh the risk of an adverse outcome from the intervention. When there was disagreement, the case was discussed and consensus reached. Potential ADRs were assessed for causality using the World Health Organization's system for standardised case causality assessment .
Outcomes measures for the study were:
Sample size was based on the hypothesis that the pharmacist would identify at least one MRP per patient that was not addressed by the clinic doctor (with standard deviation of 2.0 for mean number of MRPs). It was calculated that 44 patients would provide 90% power to detect a difference that was significant at the 5% level.
The study was approved by the Austin Health Human Research Ethics Committee.
- Top of page
Over the study period, on the days the pharmacist was in attendance, 55 patients attended the memory clinic and 31 attended the aged care assessment clinic. Forty-six patients (23 from each clinic) were eligible, able to be seen by the pharmacist and agreed to participate (Table 1). A larger proportion of patients were excluded from the memory clinic group because of the larger number of patients attending each session and presence of patients younger than 65 years. The primary reason for referral to the clinics was most commonly cognitive assessment or cognitive review (80%), followed by management of behavioural and psychological symptoms of dementia (6.5%). Other reasons for referral were pain, impaired mobility, dyspnoea, medication review and completion of paperwork for respite care. Twenty-two (48%) patients were new to the clinics; the remainder were attending for follow-up or review.
Table 1. Study patients and MRPs
|Study patients||Aged care assessment clinic (n= 23) Mean/Median (range)||Memory disorders clinic (n= 23) Mean/Median (range)||All patients (n= 46) Mean/Median (range)|
|Age (years)||86 (73–93)||79 (69–88)*||82 (69–93)|
|MMSE||20 (6–28)||22 (11–29)||21 (6–29)|
|Comorbidities||5.8 (3–8)||5.0 (3–8)||5.4 (3–8)|
|Medications – regular||6.0 (3–12)||8.0 (2–18)||7.0 (3–18)|
|Medications – when required||2.0 (0–7)||2.0 (0–6)||2.0 (0–7)|
|Pharmacist-identified MRP||2.0 (0–6)||2.0 (0–8)||2.0 (0–8)**|
|Doctor-identified MRP||1.0 (0–5)||0 (0–1)*||0 (0–5)†|
|Total MRP||3.0 (0–9)||2.0 (0–9)||2.5 (0–9)|
|Home medicines review‡ in past 12 months||2 (8.7%)||1 (4.3%)||3 (6.5%)|
According to the pharmacist's medication history, patients were using a median of 9.0 medications (range 3–21). In comparison, the median number of medications recorded in the patients' medical record was 6.0 (P < 0.0001). At least one medication was omitted for 39 (85%) patients. There was also at least one dose discrepancy compared with the pharmacist medication history for 20 (43%) patients (median 1.0 per patient; range 0–8).
One hundred and forty-seven MRPs were identified either by the clinic doctor or pharmacist (median 2.5 per patient; range 0–9). Thirty-four MRPs were addressed by the clinic doctor (median 0 per patient, range 0–5), the most common of which was need for an additional medication (untreated indication; 44%), followed by unnecessary medication, dose too low and ADR (12% each). In most cases (27/34; 79%), these issues were directly related to the reason for referral to the clinic. One hundred and thirteen additional MRPs were identified by the pharmacist (median 2.0 per patient; range 0–8; Table 2). The most common was potentially unnecessary medication, which made up 21% of all MRPs and affected 32% of patients. Other common MRPs were: inappropriate drug choice (41% patients), untreated indication (35% patients) and non-adherence (33% patients).
Table 2. Types of medication-related problem identified by the pharmacist and associated clinical risk
|Medication-related problem||Extreme risk||High risk||Moderate risk||Low risk||Total (%)|
|Potentially unnecessary medication†|| ||2||16||6||24 (21)|
|Inappropriate medication choice‡|| ||3||18||2||23 (20)|
|Untreated indication|| ||13||7|| ||20 (18)|
|Medication management problem§|| ||7||3|| ||10 (8.8)|
|Dose too low|| ||2||6||1||9 (8.0)|
|Adverse drug reaction||1||3||1|| ||5 (4.4)|
|Dose too high|| || ||4|| ||4 (3.5)|
|Drug interaction|| ||1||0|| ||1 (0.9)|
The number of MRPs addressed by clinic doctors was significantly lower in the memory clinic than the aged care assessment clinic (P= 0.01), but the number of MRPs identified by the pharmacist did not differ significantly between clinics (Table 1).
Independent expert review of pharmacist-identified MRPs rated 39 (35%) as high or extreme risk and 64 (57%) as moderate risk. Twenty-nine (63%) patients had at least one high- or extreme-risk MRP (mean 0.85 per patient). The types of MRP identified by the pharmacist, and their clinical risk ratings, are described in Table 2.
Among the pharmacist-identified MRPs, the most common untreated indication was osteoporosis or vitamin D deficiency (24% patients), whereas the most common inappropriate medication was benzodiazepines (15% patients). The most common potentially unnecessary medications were frusemide (8.7% patients), lipid-lowering drugs (6.5%), acid-suppressing drugs (6.5%) and folate (6.5%).
Thirty-seven (33%) of the 113 pharmacist-identified MRPs were related to medications not recorded in the medical record. Of these, one was rated as extreme risk, eight (22%) high risk, 24 (65%) moderate risk and four (11%) low risk.
Eighteen suspected current ADRs were identified in 12 (26%) patients (Table 3). Seven were documented by the clinic doctor and 11 were identified by the pharmacist. All ADRs were rated as ‘possible’ and none as ‘certain’ or ‘probable’ because outcomes following drug withdrawal and/or re-challenge were not available. Most pharmacist-identified ADRs were rated as moderate or high risk; one was rated as extreme risk. Three were related to medications not recorded in the clinic notes.
Table 3. Suspected ADR†
|Alendronate§||Gastroesophageal reflux disease|
|Alendronate||Gastroesophageal reflux disease|
|Buprenorphine||Cognitive impairment; constipation|
|Diazepam§||Cognitive impairment; falls|
|Tolteradine||Cognitive impairment; constipation|
A total of 33 medication-related recommendations or interventions were made by clinic doctors, and a further 144 by the pharmacist (Table 4). For one pharmacist recommendation, the potential risk of intervention was judged by the independent geriatrician to outweigh the potential benefit (addition of low-dose aspirin in a functionally independent 86-year-old woman with a past history of ischaemic heart disease and coronary artery bypass grafts who was also taking warfarin for atrial fibrillation). All other recommendations were judged to have potential to reduce medication-related risk.
Table 4. Recommendations or interventions to address identified medication-related problems
|Recommendation/intervention||No. pharmacist (%)||No. doctor (%)||Total (%)|
|Stop drug||36 (25)||8 (24)||44 (25)|
|Add drug||22 (15)||15 (45)||37 (21)|
|Adjust dose||13 (9.0)||8 (24)||21 (12)|
|Switch drug||16 (11)||0||16 (9.0)|
|Additional monitoring (e.g. blood pressure, blood test)||13 (9.0)||0||13 (7.3)|
|Medical assessment required†||11 (7.6)||1 (3.0)||12 (6.8)|
|Switch dose form||9 (6.3)||0||9 (5.1)|
|Provide education to patient or carer||9 (6.3)||0||9 (5.1)|
|Provide medication management service or aid (e.g. HMR or DAA)||9 (6.3)||0||9 (5.1)|
|Adjust dose timing||6 (4.2)||1 (3.0)||7 (4.0)|
The average time spent by the pharmacist on each patient was approximately 90 minutes (10–20 minutes for medical record review, 20–30 minutes for patient/carer interview and 30–60 minutes for medication review and reporting).
- Top of page
This is the first study to investigate the prevalence of MRPs in older patients referred to a memory disorder clinic, and the first to explore the potential role of a clinical pharmacist in an aged care or memory clinic in Australia. Our findings show that patients referred to these clinics are at high risk for MRPs. Polypharmacy (defined in this study as five or more regular medications), which is considered to be the strongest independent risk factor for adverse medication events , was present in more than 80% of the study sample. One in four patients were suspected to be experiencing an ADR, and almost two-thirds had at least one MRP associated with high or extreme risk of adverse outcome. Consistent with other studies [25–27], the accuracy of medication histories recorded by prescribers was poor, with at least one medication omission or dose discrepancy in 91% of cases.
Although only half of the participants in this study were recruited from the memory clinic, 87% were attending the clinics for assessment of a potential cognitive disorder. Therefore, these findings are relevant and potentially generalisable to memory disorder clinics for older people and also to older clients presenting to any clinic with the main issue of cognitive impairment.
As the prevalence of MRPs increases with the number of medications prescribed , it is not surprising that in this population, with an average of nine medications per patient, almost all patients had at least one MRP. The number of MRPs identified in our sample is consistent with other studies involving multidisciplinary medication review [4,29]. In our study, pharmacist medication review identified an additional two MRPs per patient over and above MRPs addressed by usual clinic practice, and about one-third of these (0.85 per patient) were considered by an independent geriatrician and senior pharmacist to be associated with high or extreme risk to the patient.
In terms of the type of MRP identified, our findings are similar to other studies, with ADR, untreated indication, unnecessary medication, inappropriate or suboptimal medication and non-adherence each affecting at least one in four patients . Untreated indications along with medication management and adherence problems made up the majority (69%) of the high-risk MRPs identified by the pharmacist. Potentially unnecessary medication (the most common pharmacist-identified MRP) was usually rated as moderate or low risk; however, any reduction in the total number of medications a patient needs to take has the potential to reduce risk of non-adherence, errors and adverse outcomes .
There are a number of potential reasons for the higher number of MRPs identified by the pharmacist compared with the clinic doctor:
The medication history available to the clinic doctor was usually incomplete, so that one-third of MRPs would not have been detectable.
In most cases, patients presented to the clinic with a specific problem (most commonly cognitive impairment) and this was the focus of the medical review. A comprehensive medication review may have been beyond the scope of the consultation (as evidenced by the fact that almost 80% of doctor-addressed MRPs related directly to the reasons for clinic referral).
The pharmacist assessed medication management and adherence for all patients, whereas doctors were less likely to address these issues (24% of pharmacist-identified MRPs were related to medication management or adherence issues, compared with only 2.9% doctor-identified MRP).
In some cases, MRPs may have been identified by the clinic doctor but not documented.
It is not surprising that the number of MRPs documented by medical staff in the memory clinic was significantly lower than in the aged care assessment clinic, as the role of memory clinics is diagnosis and management of cognitive disorders only (there were no doctor-addressed MRP unrelated to the reason for clinic referral noted in this group). However, the fact that the number of MRPs identified by the pharmacist did not differ significantly between clinics suggests that both groups of patients could potentially benefit from a medication review.
The importance of regular comprehensive medication review (every 6–12 months), especially for the frail older people and those with cognitive impairment is widely recognised [28,31–34]. In Australia, consumers of multiple medications can receive a government (Medicare) funded pharmacist ‘home medicines review’ (HMR) annually upon referral from their general practitioner to an accredited pharmacist. However, the referral rate for this service is low, with fewer than 10% of eligible Australians receiving an HMR . In our study sample, despite the presence of multiple risk factors including polypharmacy and cognitive impairment, only 6.5% of patients had received an HMR in the past year. The low uptake of HMRs in community-dwelling Australians reinforces the importance of performing a medication review for patients attending specialist aged care clinics.
There are some limitations with this study. First, it is a relatively small, non-random sample from a single hospital, so the findings may not be representative of all older patients attending aged care and memory clinics. Although the findings are consistent with larger studies conducted in other settings, and the study was adequately powered for its primary outcome measures, the results may need to be confirmed by a larger study involving multiple hospital sites. Second, although the most common MRP types identified in these patients (under-prescribing, inappropriate medication choice, unnecessary medications and non-adherence) have been shown to increase risk of adverse outcomes [2,6,36], due to the cross-sectional study design we have no data on actual clinical outcomes in this population. However, the independent expert review of each pharmacist-identified MRP, with assignment of clinical risk ratings, indicates that most MRPs had potential for preventable adverse outcomes. Prescriber acceptance of pharmacist recommendations was not investigated; so the actual impact of a pharmacist medication review in this setting is not known. We have not performed a cost-effectiveness analysis, but it is noteworthy that the cost of a medication review conducted by a senior clinical pharmacist in the outpatient setting (approximately $AUD80) was substantially less than the current Medicare payment for a pharmacist HMR in the community ($AUD190).
In summary, medication histories recorded in the aged care and memory clinics were usually incomplete. The clinical pharmacist was able to obtain a more comprehensive medication history and identify at least one unresolved high- or extreme-risk MRP in almost two-thirds of patients. A clinical pharmacist consultation prior to medical review in the clinics has potential to improve quality of care. The impact of an outpatient clinical pharmacy service on resolution of MRPs and clinical and economic outcomes needs to be confirmed in a prospective controlled intervention study.