Each year The Pharmaceutical Journal publishes a supplement containing the full abstracts of all the pharmacy practice research papers accepted for presentation at the British Pharmaceutical Conference. A feature of each abstract this year is a panel of focal points submitted by the authors to help readers to see the main elements of the research. As further guidance for readers, the report below presents a review of all the papers to help give an overall picture of the research presentations and identify particular groupings of the papers that might be of interest
Seventy-six practice research papers were accepted for presentation either orally or as posters at this year's British Pharmaceutical Conference in Birmingham. This review initially looks at some of the research design and methodological aspects of the papers and then at the nature of the research inquiry. Some papers may be mentioned more than once, and reference to some papers may be omitted in a section where the authors would have expected to see them. Some papers may be cited for their methodology and some for their results. This reflects the reviewer's perspective and is not indicative of the perceived value of the research. The papers are referenced by the page number of the practice research supplement issued with this week's Journal, where readers are directed for the full abstract and further details of the authors.
Although submissions to the conference are not necessarily representative of pharmacy practice research in its entirety, there is little evidence of large multicentre collaborative projects. Most of the research projects presented are fairly small-scale studies carried out in isolation of other health services research. The research design and methodologies employed by researchers do, however, appear to be far more robust than was the case a few years ago. Researchers also seem to be more aware of study limitations and more willing to point these out - a helpful feature for the reader unaccustomed to critical reading of research papers.
|
Randomised controlled trials (RCT) were used by several research groups. RCTs are seen as the gold standard for rigorous evaluation in that they potentially eradicate systematic bias and also allow meta-analysis to be performed.
|
Alison Ewing (member of the Royal Pharmaceutical Society's Council) examines a poster on the supervision of methadone consumption in Glasgow pharmacies (R37) |
|
Perhaps one notch down from the RCT in terms of obtaining evidence are the quasi-experimental studies. This by no means lessens their value in practice research; quite often they are the only way to achieve results in an action research approach. A cohort approach was taken by Fowlie et al (R16) in evaluating an electronic prescribing and administration system used in hospital. The authors do indeed recognise the study's limitation of a lack of a control group to address change over time. A quasi-experimental approach to the comparisons of pharmacist-run medication review clinics was conducted by Morgan et al (R28) in their use of different patient selection criteria in two general medical practices. A "within subject comparison" or pre/post evaluation of knowledge was used to assess the effectiveness of a medicines training package for carers by Balachandran and Nathan (R31). Assessments were conducted before and after the intervention and compared with before and after scores from a control group. |
Optimising statin usage was one of the suggested roles for pharmacist intervention in a medication review service in coronary-artery bypass graft patients (R22) |
That stalwart of practice research - the survey - was less in evidence than in past years. Where it was used it was more for explanatory reasons than for purely descriptive purposes. A cross-sectional survey used by McLean et al (R20) showed that diabetes education by practitioners enhanced patient knowledge and may contribute to reduced complications. Boardman et al (R36) used a questionnaire to investigate headache experiences, impact and management to give an insight into advice-seeking behaviour. Important methodological considerations of the pilot study were the way in which the questionnaire was developed; based on literature sources, advice from physicians and recognised standard questionnaires, and the use of validation studies, with testing of repeatability in a sub-sample of respondents after one month.
Several researchers made good use of data collected routinely for another purpose. Information from an automated recording system provided for Medicare/Medicaid nursing homes in the Unites States was analysed by Hughes et al (R30). They identified the influence that facility characteristics, such as staffing, and patient characteristics had on hypnotic and anxiolytic prescribing, showing that legislation alone does not lead to better care.
Prescription Analysis and Costs (PACT) data were used by Dalton et al (R50) to explore the variation in prescribing of oral hypoglycaemic drugs. Other national data sets, such as ethnicity, age, percentage of people who were overweight and deprivation index, were used as independent variables in a multiple regression analysis. Differences in the population factors were not the prime reason for a variation in prescribing of metformin and gliclazide, although the variation could be partly explained.
Semi-structured interviews were used in several of the research papers. Desai et al (R38) used this methodology to explore the use of protocols and guidelines by community pharmacists and NHS Direct nurses. Thematic analysis drew out differences in that pharmacists felt that protocols and guidelines were flexible and allowed them to apply professional judgment whereas the nurses felt that they were rigid and gave no room for flexibility. An exploratory approach using semi-structured interviews by Watson et al (R43) identified community pharmacists' views about the treatment of vaginal symptoms, finding that customer embarrassment was perceived as the main barrier. Semi-structured interviews were also used by El-Beik and Elliott (R61) to explore clinicians' views of medicines management.
A novel method of data collection for qualitative analysis, the internet, was used by Alexander (R70) to gather pharmacists' descriptions of the cognitive processes used in checking. Data collected from online discussion groups was also used by Burton and Anderson (R78) in an evaluative case study approach which showed that as participants spent more time on the internet there appeared to be a growing acceptance of online pharmacy services.
A variety of methods were employed by Bateson and Duggan (R59) to collect evidence on pharmaceutical care provision across the health care interface. Qualitative interviews and observations provided an insight into the people aspects, and content analysis of hospital protocols and guidelines and a literature review gave a perspective of the management, theory and philosophy of pharmaceutical care. In the project from Engová et al (R7), a combination of research methods permitted the collection of comprehensive data to provide an in-depth evaluation of community pharmacy services to the care of people with mental health problems which showed that positive contributions to care can be made.
A comparative case study approach using data collected via document analysis, observation of meetings and staff interactions, and interviews with key staff were used by Cotter and McKee (R62) to answer the question of whether NHS trusts have brought about any changes to the systems of managing prescribing policy. The work was part of a larger evaluative study on the management of prescribing policy.
It was encouraging that some of the presentations had involved users not only as research subjects but also as part of the research process, a move advocated in NHS research. Åström et al (R47) developed a tool to measure the desire for drug and medical information based on a novel method of interfacing qualitative and quantitative data derived from a cohort of general medical inpatients.
Moving on from the methodological aspects to specific subject areas, it is notable that assessing the need for services seemed to play a prominent part in the conference papers. Stakeholder involvement also featured in the needs analysis. A survey completed by primary school teachers identified variation between schools in asthma medication management, and led the authors, Jones et al (R53), to conclude that there are opportunities for pharmacist involvement.
Following on from the publication of the Department of Health guidelines on shared care opportunities for drug users, Jesson et al (R37) carried out a survey to assess the size of the drug misusing population. They found a marked interest from both community pharmacy and medical practitioners in shared care arrangements, but also identified barriers to progress of the different professional cultures, insufficient community drug workers and lack of dedicated funding.
Douglas et al (R6) found that traditional approaches to identifying patients in need of pharmaceutical care may not be valid. A study of 106 patients with type 2 diabetes mellitus showed that care of people with diabetes should be a priority for the pharmacy profession. A survey of patients contributed to the study by Gray et al (R25) which has given a mandate for the production of a community pharmacy-based programme on medicines management for people with Parkinson's disease and their carers. Carers were also the subjects of a paper from Gray et al (R46), which identified the need for pharmacists to seek more opportunities to support carers, and to address the medication-related problems they experience. Unmet needs were also identified in study by Garfield et al (R26) among people beginning courses of antidepressant medication.
Having identified a need it is of value for pharmacists to understand patients' attitudes and beliefs about their illnesses. Varma et al (R48) used a qualitative approach to assess factors which impact on the health-related quality of life of congestive heart failure patients. Most patients demonstrated a poor knowledge of their condition and the authors concluded that they might benefit from counselling and written information.
The effect of parents' beliefs about medicines in general upon their children's medication adherence was examined by Al-Omran et al (R12). The study, which was conducted in Riyadh, Saudi Arabia, showed a statistically significant correlation between maternal educational level and some non-adherent behaviour. It is intended that the study is repeated in the United Kingdom; it will be of interest to observe the effect of the cultural differences between the two countries.
Home visits were used by Lowe and Raynor (R19) in their study of non-adherence in elderly patients. They concluded that, because the elderly have significant barriers to adherence, this can lead to a focus on unintentional adherence. They found that 34 per cent of patients were taking their medicines in a way different from that prescribed as a result of a conscious decision: intentional non-adherence. Patient reports about medicine taking were compared with medical records to gauge non-adherence. A formulaic approach was developed by Sturgess et al (R35), who used refill records from patient medication records (PMRs) as a method of measuring change in compliance. The formula was developed so that it could be used as an outcome measure in practice-based research. The method was sensitive enough to identify changes in medicine-taking behaviour following pharmaceutical intervention.
As would be expected in the research section of a professional conference, new services utilising the expertise of the professional formed a substantive portion of the presentations to the BPC. Many of the papers not only provided a description of the development but also provided some key messages for others contemplating similar changes.
Whittington et al (R8) presented the minor ailments project from Sefton, which offers patients consulting a general medical practitioner the option of a consultation with a community pharmacist who had the ability to prescribe under NHS prescription charges and exemption arrangements. Thirty-eight per cent of minor ailments were transferred; however, the range of transfer rates varied for different minor ailments, just one of the factors that needs to be considered when developing pharmacy-based services for minor ailments. In a more focused service targeted at upper respiratory tract infections, Rugg and Loader (R39) found that 13 per cent of respondents to a survey, following the campaign, said that they would still see the GP for antibiotics next time.
The current Government emphasis on increasing the availability of health information to the public was recognised by Bennett and Jones (R45) in their study to quantify information requests in six community pharmacies. Approximately one-third of requests were made by proxy and 44 per cent of requests resulted in the sale of an over-the-counter medicine. The authors noted variation between the types of request received by the pharmacies. A comparison with the number of requests for information received by NHS Direct found that the pharmacies averaged 12.6 inquiries per pharmacy per day compared with 44.6 inquiries per centre per day received by the NHS pilot sites.
Howarth et al (R49) evaluated the provision of medicines information through a personalised website. They found that patients were more satisfied with personalised information, and suggested that a natural evolution of their system would be to extend the range of patient information via the NHSnet.
The transfer of patient-based information between pharmacists and other health care professionals was studied by Adeturinmo et al (R58) by interviewing and observing a range of health professionals who cared for general medical patients in a London hospital. The lack of documentation of the pharmacist's contribution to patient care, especially after counselling, was thought to hinder effective information transfer.
The barriers that hinder the development of pharmaceutical care practice were recognised in the novel service for patients with asthma described by Shaw et al (R24) in New Zealand community pharmacies. Although both pharmacists and patients supported continuation of the service, which produced improvement in asthma management and the quality of life, the patients said they would be reluctant to pay for it directly.
In secondary care, the role of the pharmacist as a facilitator to improve the care of patients with congestive cardiac failure was described by McLean and Campbell (R23). Suboptimal use of angiotensin converting enzyme inhibitors had prompted a multidisciplinary approach using the pharmacist as the catalyst for change, a model which could be adapted for use on a larger scale.
Pharmacy participation in a novel approach to smoking cessation using tailored interactive computer programmes based on the cycle of change was described by Anderson and Mair (R41). Preliminary data indicate that smokers move through the cycle of change and an increase in cessation rates was also indicated. An open, uncontrolled study was presented by Barnes et al (R42) of the use of an extract of St John's wort as an aid to smoking cessation. It was carried out as exploratory work in order to develop a hypothesis for possible testing by an RCT.
Cunningham et al (R14) audited the extent to which rational prescribing of antihypertensive combinations was being used in one GP practice following the introduction of a formulary and prescribing protocol. They found a high use of monotherapy and future work is suggested to identify if this is justified.
An understanding of prescribing in primary care from a pharmacoepidemiological viewpoint was presented by Offia et al (R9) who used the Welsh general practice morbidity database (GPMD) as a source of information on laxative prescribing in relation to age and gender. They found that when the data were adjusted for their greater numbers, elderly females did not receive significantly more laxatives than elderly males.
Aggregated prescribing data at a locality level was used by Muir et al (R11) to inform the health needs assessment. A basket of prescribed items and prescribing indicators was generated to reflect variation in local health care needs and the difference in socio-economic characteristics. In addition to informing the health needs assessment, these data have potential for planning local pharmacy service initiatives.
The impact of a prescribing change from one statin to another was investigated by Thomas and Walker (R13), who concluded that a switch based on cost alone could not be justified: variable patient response and poor tolerability to side effects were associated with a high discontinuation rate. Optimising statin usage was just one of the suggested roles for pharmacist intervention in a medication review service in coronary-artery bypass graft patients which was assessed in a paper from Alldred et al (R22).
Hughes et al (R34) looked at the potential for improved drug utilisation and reduced costs of a repeat dispensing project. Costs savings of £10 per patient per month were realised from non-dispensed medicines in addition to improvement in patient compliance.
An audit of transdermal fentanyl use in the palliative care setting was conducted by Hirsch et al (R15) The study allowed comparison of usage before and after the issue of updated prescribing guidelines, but little change was detected.
Data from an integrated patient record system were used to carry out an investigation of antibiotic usage and resistance patterns by Curtis et al (R64). Electronic linkage at patient level to microbiological testing records showed that quinolones were frequently prescribed without microbiological justification. An investigation of the impact of national recommendations on antibiotic prescribing by Smith et al (R51) found that while absolute prescription numbers had decreased, the levels of prescribing still remained high and could not be justified by demographic or morbidity factors.
Deprivation was one of the variables found to characterise differences in asthma prescribing investigated by Salamzadeh et al (R52). GP practices with a lower ratio of preventers to bronchodilators had on average a greater workload, were in areas with a higher deprivation rate and with a larger percentage of South Asian patients. It was suggested that these practices could benefit from joint initiatives for disease management of asthma.
A pharmacoeconomic comparison of antiemetics used in established post-operative nausea and vomiting by Al-Ghadeer et al (R56) showed that ondansetron was the most effective; however, the pricing structure for drug acquisition meant the droperidol proved to be more cost-effective within the patient group.
Two papers presented aspects of the use of medication in intensive care units. Beggan et al (R57) identified the need for guidelines for the titration of propofol infusion used for sedation. The paper from Whitfield et al (R18) described the development of a sedation score to assess critically ill children. The research formed part of a larger study investigating the correlation of sedative plasma levels with depth of sedation.
Pharmaceutical science with a practice research focus was in evidence in just one paper looking at the stability of diamorphine and glycopyrrolate in syringes for patient controlled analgesia. Smith et al (R69) recognised the need for reassurance since little was known about the stability of the mixture, which was often administered by subcutaneous injection in palliative care. The study indicated that at the concentrations currently used admixtures would remain stable over the usual period of administration.
Clinical governance requires all health professionals to contribute to ensuring quality of services. A very practical way was demonstrated for the care of people with diabetes by Dixon et al (R21) who found that significant improvement in the technique of blood glucose monitoring could be achieved by educating the operator in a community pharmacy setting. With the support of the pathology laboratory they were also able to implement a quality control scheme.
In Lambeth, Southwark and Lewisham, a community pharmacy quality standards group provided resources to support quality initiatives. Gregory et al (R40) showed how those pharmacies involved in the initiative obtained higher percentage patient satisfaction ratings than non-participants. A paper by Campbell and Fowler (R54) also reflected the current interest in the spread of best practice. In a survey of the attitudes of senior hospital pharmacy managers to benchmarking, they found that while it was believed that benchmarking was generally applicable to clinical services, most felt that there were problems associated with its implication.
In a survey of attitude to and knowledge of evidence-based medicine in secondary care, Edwards and Brophy (R65) found that there was still a need for training, and that pharmacists in particular needed training in change management if they are to promote evidence-based practice.
The contribution of continuing professional development (CPD) as an investment in quality in the new NHS was investigated by Khanderia et al (R72) who found that current systems in primary care do not support or facilitate CPD in independent practice. They are in the process of supporting CPD through the establishment of pharmacy education leaders. Progress to date showed that these leaders were establishing inter- and intra-professional networks. Future plans are for these leaders to facilitate the personal development of independent contractors.
Interprofessional networking via an internet mailing list was described in the evaluation of Private-Rx by Whitaker et al (R73). Balachandran and Savage (R79) showed how intraprofessional links could be achieved by pharmacists shadowing GPs and practice members. They found that both pharmacists and GPs benefited from the pilot work shadowing scheme.
Macrae et al (R75) recognised that CPD required integration of clinical audit and continuing education, and their paper described a scheme in Scotland that has introduced an accreditation process to help achieve this by providing the equivalent hours of CE for talking part in audit. Research skills are also necessary to equip pharmacists to work in an evidence-based primary care environment and Wood et al (R77) described an evaluation of research module as part of a clinical diploma course. A pre-course questionnaire showed a low level of awareness and self-confidence in pharmacy practice research methods.
Educational aspects for support staff and undergraduates were also presented at the BPC. O'Loan et al (R74) presented an evaluation of a self-development programme for hospital pharmacy technicians. The transtheoretical model of change was applied as part of the evaluation. In an assessment of the use of objective structured practicals for third-year undergraduate students, Belcher et al (R76) concluded that, although it was a useful tool in teaching and learning, the method exerted mental and physical pressure upon the candidates.
The clinical governance agenda requires risk to be managed and errors to be minimised. A study in a London teaching hospital conducted by Dean et al (R17) found that human error theory was a useful model for studying prescribing errors. Latent failure was related to gaps in medical school training, staffing policies and misunderstanding about responsibility. Performance failures were influenced by fatigue, interruptions and working in unfamiliar circumstances. Active failures were slips in concentration, lapses in memory and lack of knowledge. Another aspect of risk management in the work of a pharmacist, the process of checking dispensed medicines, was investigated by Alexander (R70). An understanding of human error theory and the procedures for avoidance were applied to the processes which pharmacists described.
An aspect related to risk management was studied by Miller et al (R55) who developed a predictive model for the risk of hospital readmission. Nine variables, both predicative and protective, were used to identify those vulnerable patients on admission who required comprehensive discharge planning, enabling services to be targeted at those in greatest need.
An evaluation of the contribution that pharmacists could make to patient care during a repeat dispensing pilot study was carried out by Wilson et al (R32). They found that although patients recognised an improvement in consumer care there was little evidence of improved professional care. The authors note, however, that although pharmacists were given an opportunity and a financial incentive to be creative and proactive they did not receive extra training or have protocols to work to. They conclude that is unlikely that all pharmacists and pharmacies can make an equal contribution to all aspects of pharmaceutical care. The need for adequate training was also recognised in a paper on developing pharmaceutical care by Krska et al (R33) as one of the factors identified by representatives of the major pharmaceutical organisations. Other factors identified as being of great importance were the nature of the remuneration and relationships with general medical practitioners.
Some of the papers identified theories that may not sit happily with the proposed future for pharmacy. The paper by Bateson and Duggan (R59) on pharmaceutical care provision across the health care interface raised the theory that pharmacists themselves are the biggest barriers to pharmaceutical care provision. Empowerment of pharmacists as autonomous professionals was suggested as the key to moving pharmaceutical care forward. Involving hospital pharmacists in the development of the pharmaceutical care model and providing community pharmacists with more information regarding drug therapy decisions made during the hospital stay is to be tested a route to empowerment.
Geographic information systems (GIS) were used for a study of patronage of community pharmacies in New Zealand by Norris et al (R10). By combining information from the GIS with data from the prescription databases they showed that different types of pharmacies, city centre, suburban and rural, had very different patterns of customer patronage.
As a contribution to the study of workforce issues, Tweddell and Wright (R44) used a questionnaire to survey reasons for leaving community pharmacy. The sample, all community pharmacists leaving a large multiple community pharmacy during 1997-98, identified long hours, lack of professional fulfilment and monotony of workload as major reasons for their decision to leave. The authors note that owing to a low response rate the results may not be representative; however, they conclude that to retain pharmacists, community pharmacy must evolve to provide many of the facets found within hospital pharmacy.
Zayed et al (R68) studied how hospital ward pharmacists occupy their time when different supply systems are used. In hospitals where pharmacists were involved in supply, an average of less than 25 per cent of ward visits was spent on clinical services, whereas when a computerised prescribing system was in place pharmacists spent 68.4 per cent of their time providing a wide range clinical services.
Computer modelling of a dispensary supply service was carried out by Marriott et al (R67). It was found to be of potential in optimising organisational changes such as staffing numbers and skill mix issues. Work currently in progress to evaluate the impact of an automated dispensing service being undertaken by Rutter (R66) provided baseline data on the physical contribution of pharmacists and support staff to dispensing. Pharmacists were found to spend the greatest proportion of their time on labelling compared with dispensers who spent the greatest proportion of time assembling the prescription. Original pack dispensing was found to be significantly quicker than dispensing of items that required counting, pouring or cutting.
Whereas patient packs provide an advantage over loose tablets in traditional dispensing, in the provision of monitored dosage systems they were found significantly to increase the dispensing time. In a time and motion type study of 10 dispensers dispensing 10 prescriptions, Green et al (R71) found that for certain packs the time was doubled. This increase in workload further contributes to the situation that pharmacy contractors have to subsidise monitored dose system dispensing.
Several of the papers presented at the BPC produced the immediate feel of "I could do that now". This aspect of getting research into practice should be developed more within the practice research sessions of the BPC.
An excellent example, which practising pharmacists and manufacturers should take note of, was the audit conducted by Nelson and Haylor (R63) into the ease of use of eye medication in an outpatient clinic. A quarter of patients did not find the labels easy to read and half of the patients did not find the eye-drop containers easy to use. Patients' concerns were knowing if any and how many drops had gone in their eyes. The paper gives some key pharmacy action points: first to advise cooling of the drops prior to use to guide the patient by feeling them go into the eye and, secondly, to offer devices which help position or squeeze the container. For partially sighted patients information needs to be available in a large font on a contrasting background as a separate instruction sheet.
No doubt the needs of patients with poor sight have been researched previously and possibly presented at BPC, yet problems still exist. Practice researchers and those in positions of influence should do all they can to ensure that the lessons from all the research presented at BPC are implemented into changes to the benefit of the patient and the profession.