Although pharmacists have potentially important roles in the prevention of medication errors, evidence for the effectiveness of pharmacists’ interventions is conflicting, explained Tony Avery (Professor of Primary Health Care, University of Nottingham, UK). However, the PINCER trial has provided some robust evidence. PINCER was a cluster-randomised trial comparing the effectiveness of a pharmacist-led IT-based intervention with simple feedback in reducing rates of clinically important errors in medicines management in general practices.
Medication errors in both primary and secondary care are an important cause of morbidity and mortality. Reducing error rates is important to prevent patients from suffering from avoidable harm. In the PINCER trial, general practitioners (GPs – family doctors) were randomised to receive either simple feedback about patients at risk from hazardous prescribing, or simple feedback combined with educational outreach and dedicated support to correct and prevent potentially hazardous prescribing.
The primary outcome measures were based on three common, hazardous prescribing scenarios:
- prescribing of non-selective NSAIDs to patients with a history of peptic ulcer without a proton pump inhibitor
- prescribing of beta-blockers to patients with a diagnosis of asthma
- prescribing of ACE inhibitors or loop diuretics to patients aged 75 years and older without an assessment of renal function and electrolytes in the preceding 15 months.
In the simple feedback group, a computer-generated report of affected patients was sent together with a summary of the evidence base for each of the problems. In the intervention group, simple feedback was supported by additional input from the pharmacist. The pharmacist was available for two days per week per practice over a period of 12 weeks. A total of 36 practices were recruited to each arm of the study. The results showed that, at six and twelve months, the error rates in both groups fell but the reductions in the intervention group were significantly larger. The apparent improvements in the simple feedback group could have been due to secular trends, commented Professor Avery. The reduction in error rates was modest in the asthma/beta blockers group, and this group was difficult to tackle because of resistance from both patients and their doctors, he noted.
The researchers concluded that the pharmacist-led complex intervention was effective at reducing proportions of patients at risk of hazardous prescribing at six months. The differences were less marked at 12 months, suggesting that a one-off pharmacist intervention may not lead to long-term changes in the prevalence of medication errors and underlining the need for a sustained input. The PINCER trial provides further support for the effectiveness of pharmacy-led interventions in primary care, said Professor Avery. Importantly, the interventions tackled an issue that was important to professionals and their teams and was delivered in a way that was acceptable to practices with minimum disruption of normal work. Careful project design and the intensity of the pharmacy input had both been crucial to the success of the project, acknowledged Professor Avery. Now that ‘proof of principle’ had been demonstrated, this approach could be extended to use other prescribing safety indicators – indeed, about 60 further indicators have already been developed and agreed with GPs, he added.
Professor Avery concluded that what is needed is “a treatment dose and a maintenance dose of a pharmacist”. Ideally, there should be a half- or full-time pharmacist in each GP practice, and more of their time should be devoted to safety improvement measures rather than cost-control activities.
When assessing the effectiveness of clinical pharmacy services, humanistic, clinical and economic outcome measures may need to be considered, Christine Bond (Professor of Primary Care: Pharmacy, University of Aberdeen, Scotland) told the audience. Humanistic measures include quality of life assessments, patient satisfaction and knowledge. Clinical measures include physiological indices, such as blood pressure or plasma lipids, together with process measures such as the number of medicines prescribed or length of stay in hospital. Pharmacy studies have tended to focus strongly on process measures, noted Professor Bond. Other important aspects of effectiveness studies are choosing the right comparator and considering the questions, ‘What can pharmacists change?’ and ‘What matters to patients?’
The highest quality evidence, usually from systematic reviews, is needed to demonstrate the effectiveness of pharmacy interventions. Cohort studies can also be useful for ongoing monitoring and evaluation, she added. A systematic review of community pharmacy interventions for smoking cessation had found 21 studies, 19 of which were excluded, mainly because of failure to include control groups. In one study, 62 community pharmacies were involved, 31 in each group. For the intervention group, pharmacists and their staff attended one two-hour workshop on the theory of behavioural change.
Customers starting new attempts to stop smoking were recruited and registered in all the participating pharmacies. The results showed that the training made a significant difference to pharmacists’ knowledge and to the numbers of patients who stopped smoking. Patients reported that they received better support than on previous attempts to stop smoking. The cost of intensive pharmacist support per quitter was £300 or £83 per year of life gained. All community pharmacists can help people to stop smoking but training improves the results considerably, commented Professor Bond.
Another study had evaluated the impact of pharmacist prescribing on patients with chronic pain in a randomised, controlled trial. The results showed that there was a consistent trend for improved pain control when pharmacists prescribed the treatment.
There is a need for more and better quality evidence to support pharmacy interventions, said Professor Bond. In addition, we need the confidence to say that there is a ‘class effect’ – if pharmacists can improve the management of blood pressure why not other things too? We must stop thinking that our interventions will change hospital admission and death rates, because these can be influenced by many other factors. Pharmacists need to demonstrate practice changes that have an impact, and they must be responsive to the needs of the health service, for example, by working on interventions such as administration of influenza vaccine.
The primary focus of health economics is optimising care, according to Steven Simoens (Professor, Research Centre for Pharmaceutical Care and Pharmacoeconomics, Catholic University of Leuven, Belgium). Economic evaluation assesses cost effectiveness of an intervention as compared with an alternative. Establishing the cost effectiveness of clinical pharmacy interventions is important as it allows pharmacists to demonstrate the value of providing such interventions. Such information may be used to inform those responsible for the reimbursement of clinical pharmacy interventions.
Many benefits are claimed for clinical pharmacy services such as improving outcomes and avoidance of adverse drug events. However, interventions are often neither well described nor standardised, and so it is difficult to aggregate information. Cost effectiveness studies usually report savings made as a result of changes in drug therapy or adverse drug events avoided. Many studies do not include the cost of the pharmacist who makes the interventions and therefore overestimate the benefits. On the other hand, some studies do not include the cost impact of preventable adverse drug events and therefore underestimate the benefits.
Seven cost effectiveness analyses of seamless care interventions have been published and most show that compliance with treatment was increased and hospital admission rates were decreased. Although they are likely to reduce healthcare costs, it is not possible to derive any strong conclusions, said Professor Simoens.
A simple tool is available to help readers to assess and interpret health economic studies. It takes into account the methodological quality of the study and the ‘transferability’ to local settings – and gives an overall score between
0 and 1. Scores above 0.6 indicate good studies with a high probability of transferability (see Resources).
If clinical pharmacy is to develop further, it must be shown to be a cost-effective use of scarce resources, said Professor Simoens. The subject should be included in pharmacy curricula and methodological guidelines should be developed for economic evaluation of clinical pharmacy interventions. In addition, researchers should take the opportunity to gather economic data when any intervention study is performed. “We do not advocate the use of clinical pharmacy enough – we should follow the example of the pharmaceutical industry and use the available evidence more aggressively”, he concluded.
Pharmacists in Portugal now have four years’ experience of providing influenza vaccination, explained Suzete Costa (Executive Director, Centre for Health Research and Evaluation (CEFAR)). In 2007, a change in legislation allowed pharmacists in Portugal to expand the scope of their activities into new areas, including immunisation. In order to provide influenza vaccination services, it was necessary either to hire nurses or to train pharmacists, said Dr Costa. In 2008, the Portuguese National Association of Pharmacies (ANF) set up a training programme modelled on a similar programme that had been developed in the US and accredited by the Centers for Disease Control. The programme comprised 10.5 hours of education covering vaccines, the pharmacy-based immunisation service, anaphylaxis and vaccine administration techniques, using live individuals and dummy vaccine. In addition, a four-hour course in basic cardiac life support was provided. A total of 53 training sessions were delivered in 12 weeks, at the end of which nearly 2000 pharmacists had been trained and there was at least one trained pharmacist in 48% of pharmacies. Now, more than 3000 have been trained, added Dr Costa. The ANF provided leaflets about influenza vaccination, vaccination record cards, anaphylaxis management flow charts and a poster to display in the pharmacy.
Pharmacists targeted patients with prescriptions for influenza vaccine and suggested vaccination in the pharmacy. They also identified at-risk patients and referred them to doctors to obtain prescriptions. Over the past four influenza seasons, the number of patients receiving vaccinations in pharmacies has steadily increased, so that, by the 2011–12 season, 49% of influenza vaccinations dispensed were administered in pharmacies and this had now become the preferred place for vaccination. Patient satisfaction with the service was high, partly because appointments were not required and waiting times were short. Despite the good results, the proportion of over-65s vaccinated still fell below the 75% target level recommended by the World Health Organization, said Dr Costa. For the 2012–13 influenza season, the Portuguese Government has decided to provide free influenza vaccinations via health centres for the over-65s, regardless of income. This might not be the most efficient way to increase the level of vaccination. Patients have reported problems such as health centres running out of stock and the need to make appointments, explained Dr Costa. European health authorities should consider allowing qualified community pharmacists to administer influenza vaccinations to high-risk patient groups (for example, aged 65 years and older) without a prescription, under protocol, in the same way as in the USA, she suggested.
Currently, pharmacists in Portugal are not paid by the Government for the influenza vaccination service. Some make a charge of about €2.0, but many do not, said Dr Costa. They are also able to administer other vaccines including HPV (human papilloma virus) vaccine and travel vaccines, she added.
Medication reconciliation is important in preventing medication errors when patients move from one setting to another, explained Tommy Eriksson (Professor and Director MSc Pharmacy Program, Lund University, Sweden). Two recent systematic reviews have concluded that clinical pharmacists are able to make successful interventions in this area but that studies of higher quality are required to demonstrate the most effective approaches.
In Lund, an integrated medicines management model has been developed to individualise and optimise drug treatment for patients. Studies have shown that the introduction of the process improved the proportion of correct medications listed from 34% to 68%, reduced the mean number of errors per patient from 2.2 to 0.97 and decreased the frequency of health care contacts within three months of discharge from 8.9% to 4.4%. Further studies showed that inappropriate drug treatment was reduced by up to 60% and hospital revisits due to drug-related problems were reduced by 50%. There were also time-savings for physicians and nurses in both primary and secondary care. Overall the service was well-received and the pharmacists’ contribution to quality and safety was recognised.
- Pharmacoeconomics for European hospital pharmacists. De Rijdt T et al. A practical instrument for assessing the quality of pharmacoeconomic studies. In: Rutten F, Vulto A (eds). Pharmacoeconomics for European hospital pharmacists. PPME: Mol, 2010: 60–8 (available at http://wip.ppme.eu/).
European Society of Clinical Pharmacy (ESCP)
The 41st ESCP Symposium on Clinical Pharmacy was held in Barcelona, Spain, on 29–31 October 2012.