Safer Prescribing and Care for the Elderly (SPACE): feasibility of audit and feedback plus practice mail-out to patients with high-risk prescribing

KEYWORDS: Prescription medicines; patient safety; ageing; primary health care

Avoidable adverse drug event (ADE) hospital admissions are common, costing health systems billions of dollars every year.^1–3 Approximately 10% of hospital admissions in older people are medication related, of which more than half are considered preventable.^4,5 Most ADE admissions are caused by commonly prescribed drugs. Non-steroidal anti-inflammatory drugs (NSAIDs), antiplatelet medications and anticoagulants together account for one-third of ADE admissions.^1,5,6

The greatest predictor of ADEs is the number of medicines a person is taking.⁷ Polypharmacy, the taking of multiple medicines, is increasing as more people are living longer with more chronic conditions. Despite evidence to guide safe prescribing, high-risk prescribing is also common in older people, with one in five prescriptions considered potentially inappropriate.^8–10 High-risk prescribing places patients at increased risk of ADEs. While patients’ individual circumstances may justify high-risk prescribing, it is important that medicines are regularly reviewed for on-going appropriateness and discontinued or initiated as appropriate, to minimise harm.¹¹ However, there are many barriers to the regular review of medicines in everyday general practice.¹²

Variation in prescribing between practices and regions suggests prescribing could be improved.^13,14 However, the most effective, cost-effective and practical approach to safer prescribing in everyday practice is not yet known.^15,16 Translating research evidence into practice is difficult. Complex interventions as part of on-going quality improvement programmes show the most promise, in particular interventions that combine audit, feedback, incentives to participation and patient engagement.^15,17–19 The Australian Veterans’ Medicines Advice and Therapeutics Education Service (MATES) quality improvement programme, which delivers four topics per year, has shown promising results in the Australian veterans population.¹⁸ The MATES programme is based on sound theoretical underpinnings; the intervention uses prescribing audits, patient-based feedback and education for GPs, and a practice mail-out to at-risk patients with an information brochure and a letter inviting them to discuss their medicines when they next see their GP. The mail-out to empower patients and promote patient engagement in medicines management is a novel addition that may be key to provoking change in GP prescribing behaviour.²⁰

The MATES programme provides a model that could be adapted for use in New Zealand primary care. In New Zealand, patients are registered with one general practice that is responsible for all on-going prescribing, including of medicines initiated by a specialist; nearly all practices use computer practice management systems capable of generating prescribing data, lists of patients and patient letters; and all practices are organised into Primary Health Organisations (PHOs) that hold funding contracts and organise initiatives to improve patient care. Many larger PHOs already employ clinical advisory pharmacists. An adapted MATES programme could be delivered through PHOs to promote regular medicines review and support safer prescribing in general practice.

We developed the Safer Prescribing and Care for the Elderly (SPACE) intervention by adapting the MATES intervention to the New Zealand primary care context. The aim of this study was to test the feasibility of the SPACE intervention in one Auckland PHO, involving two practices and seven general practitioners (GPs). For the feasibility study, we chose the high-risk prescribing topic of NSAIDs and antiplatelet medications. We chose this topic because these medicines cause many ADE admissions and fatal ADEs,⁵ their high-risk prescribing is common,^21,22 their prescribing can be improved^15,17,18 and patients with risk factors can be identified. Findings from this feasibility study will inform optimisation of the SPACE intervention before testing in a larger trial its effect on high-risk prescribing rates.

The setting was two purposively sampled urban general practices; one medium-sized and one solo practice, in one Auckland PHO. Participants were the PHO clinical advisory pharmacist, all seven GPs in the participating practices, and patients identified as having high-risk prescribing who received the mail-out.

The SPACE intervention comprises a practice audit to identify patients with high-risk prescribing; patient-specific feedback and education to GPs delivered by the PHO clinical advisory pharmacist at an outreach visit; a tick-box for GPs to indicate which patients are to receive the mail-out; and a practice mail-out to patients with information about their medicines and a letter inviting them to discuss their medicines when they next see their GP (Appendix 1).

Feasibility of the intervention was assessed using semi-structured interviews. Patient participants were recruited via letter of invitation that was included in the intervention mail-out. One researcher (RT) conducted all the interviews. Interviews were guided by an interview schedule that included several open-ended questions with flexible prompts (Appendix 2). The interviews explored participants’ views on the acceptability and utility of the intervention. Participants were asked to express their views, and informed that they would not be judged or compared, and that their responses would not affect their relationship with their GP or their usual care. All interviews were audio-recorded with permission and transcribed verbatim. Transcripts were read and coded, and the codes grouped into emergent themes using the general inductive approach.^23,24 A final list of themes and subthemes was developed.

Ethical approval for this interview study was obtained from the University of Auckland Human Participants Ethics Committee (Ref no. 017983).

The practice audits identified 86 patients with high-risk prescribing for the chosen topic. The PHO clinical advisory pharmacist delivered one-on-one feedback and education to all seven GPs. GPs selected 29 patients to receive the mail-out, and 13 (45%) responded via email or mail to indicate willingness to interview. The main reason GPs chose not to send the mail-out was because the high-risk prescribing had already ceased (eg the NSAIDs had been only a short course). Eleven patients were interviewed; the remaining two could not be contacted. The clinical advisory pharmacist and all seven GPs were interviewed. Interviews lasted between half and one hour, depending on how much information participants had to share. Participant characteristics are shown in Table 1. In general, participants said the SPACE intervention was acceptable and useful. Participant quotes are given in Tables 2 and 3.

Table 1. Participant characteristics

Table 2. Participant views on the Safe Prescribing and Care for the Elderly (SPACE) intervention: audit and feedback

Table 3. Participant views on the Safe Prescribing and Care for the Elderly (SPACE) intervention: mail-out to patients identified as having high-risk prescribing

The pharmacist said that the SPACE intervention would appeal to PHOs that already employed clinical advisory pharmacists because there was no extra cost, and the focus on safer prescribing aligned with PHO goals. The SPACE intervention facilitated pharmacists developing relationships with GPs to influence their prescribing behaviour. The intervention provided a structured format for PHO pharmacists to do what they were employed to do, and provided pharmacists with a useful foot-in-the-door with GPs, who were sometimes ‘too busy’ to see the pharmacist. On average, the pharmacist spent ~15 min giving feedback to each GP. Feedback sessions were made more efficient by culling from the list of patients those who had been prescribed only a short course of NSAIDs.

GPs reported that the feedback sessions were educational and they appreciated being prompted to review prescribing. Some GPs liked going through the patient list with the pharmacist, while others preferred to go through the list themselves in their own time. They sometimes added a comment in the patient record to remind themselves a patient had high-risk prescribing. They did not want any alerts added in the practice management system. GPs were concerned about competing demands and time constraints. They said two prescribing topics per year might be do-able, but that four would be too many.

GPs supported the practice mail-out to patients identified as having high-risk prescribing to encourage engagement in their medicines management. However, GPs wanted that the mail-out be carefully designed to avoid upsetting patients or harming the GP–patient relationship. GPs appreciated having control over which patients received the mail-out.

Most patients said the mail-out made them feel cared for; they were reassured to know someone was checking their medicines. No patient reported being upset or worried. Most patients said they trusted their GP to know which medicines were best for them. Some said they would take the letter with them to their next appointment, because the letter told them to. However, some patients said they would not because there were already too many things to discuss in the brief time they had with their GP. They did not want to make a separate appointment to see their GP about their medicines because of the cost. Some patients said the medicines information brochure was confusing; they thought the brochure did not apply to them, or did not know which medicines they were taking.

Both the GPs and the pharmacist said the NSAIDs and antiplatelet medications topic was a good one to choose because these medicines were commonly prescribed. Suggestions for future topics included: serotonin syndrome, inhalers, benzodiazepines, proton pump inhibitors and anticoagulant medications.

The SPACE intervention was designed to promote medicines review and support safer prescribing in everyday general practice. The intervention comprises practice audit to identify patients with high-risk prescribing, patient-specific feedback and education to GPs, and a practice mail-out to at-risk patients with a medicines information brochure and a letter inviting them to discuss their medicines when they next see their GP. The mail-out seeks to empower patients and promote engagement in their medicines management.

Findings suggest it is feasible for PHOs to use the SPACE intervention in general practices to support safer prescribing. However, its use will be curtailed by GP time constraints and competing demands. The SPACE intervention provides a structured format for clinical advisory pharmacists to do what PHOs employ them to do at no extra cost. GPs said they appreciated the education and the prompt to review prescribing. Some GPs were concerned the mail-out might upset patients, but patients said they were reassured to receive the mail-out and to know someone was checking their medicines. Some patients said they would take the letter to their next appointment and use it to prompt a discussion with their GP about their medicines, but other patients said there was not enough time in a consultation to talk to their GP about their medicines.

Findings from this study contribute to the growing body of literature on interventions to promote safer prescribing in general practice, and confirm the acceptability and utility of both audit and feedback, and patient empowerment through practice mail-out.^{15,20,25–29}

Strengths of this study lie in having interviewed participants involved in different aspects of the intervention, including the clinical advisory pharmacist who delivered the patient-specific feedback, all GPs from participating practices and at-risk patients who received the mail-out. A limitation is that we worked with only one PHO. This PHO had well-established, pharmacist-led quality improvement processes in practices and motivated clinical advisory pharmacists with established relationships with practices. A further limitation is that patient participants were a self-selected group; only 13 of the 29 patients who received the mail-out (45%) agreed to interview, introducing a risk of bias. It is possible that the patients who did not come forward for interview were distressed by the mail-out, or disinterested. It is also possible that participants provided responses seeking to please the interviewer, even though we provided reassurance that there were no right or wrong answers and that participants would not be judged or compared.

High-risk prescribing and avoidable adverse drug event hospital admissions are common and costly. The most effective, cost-effective and practical approach to safer prescribing in everyday general practice is not yet known. Findings from this feasibility study suggest that the SPACE intervention could be used by PHO clinical advisory pharmacists to support safer prescribing in everyday practice. Findings suggest changes to optimise the SPACE intervention, including simplification of the medicines information brochure for patients. The next step is to test in a larger trial the effect of the SPACE intervention on high-risk prescribing rates.

None.