Abstract
Background
General practice has been traditionally considered a low-risk healthcare setting, however high patient volumes and high prescribing rates elevate the risk of harm. Māori and Pasifika patients are at greater risk of healthcare harm. The extent of medication-related harm, and harm by ethnicity, is unknown in New Zealand general practice.
Determining the nature and extent of these harms is a first step to addressing these problems. Strategies to reduce harm increasingly involve patients. Providing risk information at an appropriate health literacy level can improve patient engagement.
Aims
1) To identify problems associated with medication use in New Zealand general practice
a) To evaluate the extent of medication-related harm arising from prescribing in NZ general practice
b) To evaluate an automated clinician alert system to see whether there were any inequities in clinician action taken based on patient ethnicity or other demographic factors
2) To explore strategies to improve medication safety in New Zealand general practice
a) To explore what patients and prescribers would like in a decision support and communication tool
b) To explore whether a tailored information package for patients can improve knowledge of NSAIDs and reduce self-reported use of NSAIDs
Methods
This thesis used a pragmatic mixed-methods approach in four linked studies. The first two studies are analyses of general practice record review data. They outlined problems associated with medication use, by estimating the amount of medication-related harm, and by determining whether clinician action varies by patient ethnicity when notified of the harm. Studies three and four explored strategies to improve medication safety.
Patients and prescribers were interviewed in the third study to understand what they would like from a medication decision support and communication tool. The fourth study considered the feasibility of conducting a randomised controlled trial to test whether a tailored information package is acceptable and effective in improving knowledge of non- steroidal anti-inflammatory drugs (NSAIDs) and reducing self-reported NSAID use in patients at risk of renal damage.
Results
1. The estimated incidence rate of all medication-related harms in New Zealand general practice was 73.9 harms per 1000 patient-years, the estimated incidence of preventable or potentially preventable medication-related harms was 15.6 per 1000 patient-years, and the estimated hospitalisation rate was 1.1 per 1000 patient-years. Most harms were minor (1390/1762, 78.9%), but one in five harms were moderate or severe (373/1762, 21.1%); three patients died (3/9076, 0.03%). Most medication-related harms were not preventable (n=1432, 81.3%); the remainder were considered preventable or potentially preventable (n=330, 18.7%).
2. Analysis of whether clinicians took action following a risk alert found no evidence of a difference in the odds of having action taken by patient ethnicity, however the estimated odds for Māori and Pasifika patients were lower compared with Europeans (Māori OR 0.88, 95 %CI 0.63–1.22; Pasifika OR 0.88, 95 %CI 0.52–1.49). Females had significantly lower odds of having action taken compared with males (OR 0.76, 95 %CI 0.59–0.96).
3. Patients want as much information as possible about their medications and risk, but doctors find it difficult to communicate that information. Participants were cautiously optimistic about a prescribing risk assessment and communication tool, but worried about potential harm arising from its use. They also identified requirements for the tool and features to avoid. Culturally safe and trustworthy doctor-patient relationships are required before successful implementation of any tool.
4. Patients at risk of renal damage are willing to participate in a study via email recruitment, and engage with an interactive learning activity about non-steroidal anti-inflammatory drugs online. This randomized feasibility trial demonstrated that this research method is feasible for the purposes of recruiting patients and testing the effects of providing this targeted information package.
Conclusion
Medication-related harm in general practice is common. Patient and prescriber perspectives are needed to inform harm-reduction strategies. Use of a targeted alert system has potential to mitigate risk from medication-related harm. Clinicians typically take action on alerts arising from a general practice electronic alert system, but our study suggested that they appear to take less action for women and Māori and Pasifika patients. Recognising clinician biases may improve the equitability of health care provision; respectful, culturally safe doctor-patient relationships are critical to the successful implementation of any risk-mitigation tool. Delivering medication-risk information to targeted patients online is feasible, and further studies are required to determine determine whether that improves knowledge or changes behaviour.