NPC Archive Item: Multifaceted educational programme reduces antibiotic dispensing in primary care

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
MeReC Rapid Review NPC Logo

21st March 2012

A practice based randomised controlled trial in Wales found that use of a multifaceted educational programme (STAR) resulted in a statistically significant 4.2% reduction in antibiotic dispensing compared with control practices who did not receive the programme. The clinical significance of this finding is uncertain.

Level of evidence:
Level 3 (other evidence) according to the SORT criteria.

Antibiotic prescribing remains at a high level with wide variation between general practices. Changing prescribing behaviour is a complex problem (see MeReC Bulletin 22 No.2 December 2011). Although guidance, such as that from NICE on respiratory tract infections and the Health Protection Agency (HPA), are important, their ability to change prescribing behaviour is limited without additional support. An educational programme, such as the Stemming the Tide of Antibiotic Resistance (STAR) programme, might be used to encourage more appropriate prescribing of antibiotics for self-limiting infections in the community. The clinical and cost-effectiveness of this programme have yet to be established. This approach should be of interest to those concerned with antimicrobial stewardship and those involved in improving the quality of prescribing by changing clinician behaviours.

What is the background to this?
Antibiotic resistance poses a significant threat to public health. Emerging resistance against antimicrobials that are currently used, and the upsurge in a number of infections, some of which had previously been well-controlled, are two factors that contribute to this threat. Much of the prescribing of antibiotics in primary care is for self-limiting respiratory tract infections; this is of questionable value, puts patients at risk from side-effects and increases the risk of antibiotic resistance.

Changing prescribing behaviour is a complex problem. Guidelines by themselves such as the NICE guidance on antibiotic prescribing for self-limiting respiratory tract infections have an important role, but may not be sufficient to change practice by themselves. Public awareness campaigns have not been associated in the past with reductions in all cause antibiotic prescribing and there is unexplained wide variation between individual practices.

Recognising these issues, the authors developed the multifaceted STAR educational programme, using a variety of learning methods and topics, based on social learning theory. It attempts to develop clinicians’ sense of importance of change (the ‘why’) and their confidence in their ability to achieve change (the ‘how’). General practices in Wales were randomly assigned to receive the educational programme. The volume of oral antibiotics dispensed from prescriptions arising from these practices was compared with those of practices not provided with the programme. Dispensing, rather than prescribing volume, was chosen as a better proxy for antibiotics actually consumed to take account for the use of ‘delayed’ prescriptions (as per NICE guidance on respiratory tract infections). Re-consultations, hospital admissions for selected causes, and costs were measured as secondary outcomes.

What does this study claim?
In this study, practices exposed to the education programme reduced antibiotic dispensing on average by 14 items per 1000 registered patients, while control practices increased antibiotic dispensing by 12 items per 1000 registered patients (difference 26.1 items per 1000 registered patients). Taking into account baseline dispensing rates, this equated to a reduction of 4.2% (95% confidence interval [CI] 0.6 to 7.7%; p = 0.02) for all age groups in the year after the intervention.

There were no significant differences in re-consultations, admissions to hospital, or costs between the groups. The reduction in antibiotic dispensing was achieved at an average cost of £2893 per practice. Although the costs of dispensing were not significantly reduced, the authors speculate that the cost of the programme might be recouped within 3.5 years from reduced dispensing costs.

So what?
The authors point out that most of the studies that have looked at interventions to reduce unnecessary antibiotic prescribing have only looked at one or two approaches. In contrast to these, and recognising the complex nature of the issue, the authors, explored both the ‘how’ and ‘why’ of change in a flexible learning format , using practice-specific prescribing and resistance data, information about guidelines, reflections in practice based experience, academic outreach, and a web forum. They claim to have included all elements showing promise or effective previously. As pointed out in the MeReC Bulletin: ‘Rather than a ‘scattergun’ approach of employing multiple approaches in an unsystematic way, recent research advocates a consideration of the full range of options and using a rational system for selecting from among them.’

These data add some knowledge about the role of educational programmes for reducing inappropriate antibiotic prescribing. However, we do not know if this multifaceted approach represents a good use of money and other resources compared to other, simpler interventions. The editorial accompanying the article states: ‘The results are similar to (although at the lower end of) reductions seen with other such programmes.’ Although the 4.2% reduction in prescribing is within the range of other studies, meaningful comparisons with other studies is difficult in view of the heterogeneous nature of the study designs, interventions, and outcome measures. The study looked at dispensing across practices, and in many cases not all GPs within the practices receiving the educational intervention participated in the programme, so the reduction found may underestimate the effect of the benefit. Indeed, the benefit was only seen in practices where more than 67% of GPs participated; in those practices with fewer participants the dispensing actually increased.

Few, if any, other studies have gathered information on complications, repeat consultations or costs. The cost of providing the intervention (at about £3000 per practice) may be partly offset by reductions in dispensing costs (on average £830 per practice), but this reduction in dispensing costs was not statistically significant. Although the study showed a difference in antibiotic dispensing between groups, the clinical significance in terms of clinical outcomes is uncertain. The study was not powered to identify small differences in re-consultations or hospital admissions, although no significant increase between groups was identified. The lack of diagnostic data meant that the study was only able to identify the proportion of prescriptions for antibiotics that were clinically inappropriate. The effect on antibiotic resistance was not evaluated, although the accompanying editorial to the article suggests that ‘the small reduction seen in the STAR study is unlikely to lead to a clinically important change in resistance patterns.’

A rapid response to this study, whilst recognising that it is difficult to identify an effective strategy to change behaviour, points out some important further limitations:

‘Firstly, a multifaceted intervention might be a way forward, but “bundling” them into one is difficult to demonstrate which component is the most effective: the audit and feedback, educational activity, printed educational materials, educational outreach visits, or having an opinion leader to spread the gospel; each of these approaches have been assessed by EPOC with modest outcomes at best. Having at least another group with any of the above might improve the evidence base; however, this would have been at the expense of reduced number of practices in each arm and therefore the power of the study. Even then, a much larger trial and length of follow up would be required to answer the question relating to the “dose” and effect of interventions.

Secondly, the online training might be time consuming (average of 238 minutes or nearly four hours) and some GPs and nurse practitioners might not be able to find the time to do this. In the related qualitative study, most of the sample of 31 practitioners found the educational programme useful, but a small number considered it “a waste of time and money” and found it difficult to engage in online learning.

Thirdly, with an overall mean cost of nearly £3000 per practice and annual saving of £830 per year for an average intervention practice, it might take 3.5 years to recoup the cost. In an era of austerity and efficiency savings, this poses a huge financial barrier.’ [Note: cited references to supporting information have been removed from this quote]

Study details
Butler CC, et al. Effectiveness of multifaceted educational programme to reduce antibiotic dispensing in primary care: practice based randomised controlled trial. BMJ 2012;344:d8173

Practice based randomised control trial of multifaceted educational programme (STAR) compared with usual care. Clinicians and observers were blinded to group allocation until after randomisation.

68 general practices in Wales with about 480,000 patients.

Outcomes and results
The rate of oral antibiotic dispensing (items per 1000 registered patients; in the year after the intervention adjusted for the previous year’s dispensing) decreased by 14.1 in the intervention group but increased by 12.1 in the control group, a net difference of 26.1. After adjustment for baseline dispensing rate, this amounted to a 4.2% (95% CI 0.6% to 7.7%) reduction in total oral antibiotic dispensing for the year in the intervention group relative to the control group (p = 0.02). Reductions in dispensing of individual classes of antibiotics were only statistically reduced for phenoxymethylpenicillins (penicillin V) (7.3%, 95%CI 0.4% to 13.7%) and macrolides (7.7%, 95%CI 1.1% to 13.8%), although CIs were wide for all classes.

There were no significant differences between intervention and control practices in the number of admissions to hospital or in re-consultations for a respiratory tract infection within seven days of an index consultation. The difference in the rate of hospital episodes for possible respiratory tract infection and complications between groups (on average increased in the intervention group) was 1.9% (95% –8.2% to 13.2%, p = 0.72). The  difference in the rates of re-consultations between groups (per 1000 registered patients) for respiratory tract infections within 7, 14 and 31 days after the original consultations (on average decreased in the intervention group) were –0.65 (–1.69 to 0.55), –1.33 (–2.12  to 0.74) and –2.32 (–4.76 to 1.95) respectively (all p >0.4).

The mean cost of the programme was £2923 per practice (standard deviation £1187). There was a 5.5% reduction in the cost of dispensed antibiotics in the intervention group compared with the control group (−0.4% to 11.4%), equivalent to a reduction of about £830 a year for an average intervention practice.

Study sponsorship

The study was funded by the UK Medical Research Council.

Further information on common infections can be found on NHS Evidence and in the common infections e-learning section of the NPC website. The NPC e-learning resources include two patient decision aids on common infections, which may be helpful when discussing the benefits and risks of using antibiotics with patients.

Please comment on this rapid review using our feedback form.

Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you.