Medication adherence in patients with chronic diseases has been reported to be approximately 50% in developed countries. This is of particular concern in patients post myocardial infarction (MI) as there is evidence that the medications prescribed help to prolong life and reduce the risk of further cardiovascular events.
There is no single intervention which consistently results in improved adherence although, a common thread among interventions which have led to improved medication adherence is regular follow up of patients by healthcare professionals. This may be stemmed by a number of factors known to impact adherence behaviour, one of which is beliefs about medicines. Current questionnaires in the literature to elicit patient’s medication beliefs do not identify the patient’s individualised medication beliefs. The repertory grid technique is an interview method which may be used to determine individual patient’s beliefs about medications.
This thesis aimed to explore if community pharmacists discussing patients’ beliefs about medicines for secondary prevention of MI was related to adherent behaviour. The following were hypothesised and investigated:
Ho(1):Patients receiving regular interviews by their community pharmacist were more likely to have better adherence than those who did not have this intervention.
Ho(2):Patients receiving regular interviews by their community pharmacist were more likely to have different constructs and medication beliefs than those who did not have this intervention.
Ho(3):Patients with good adherence would have different constructs and beliefs towards their medication than those with poor adherence.
Ho(4):Patients receiving regular interviews by their community pharmacist were more likely to be prescribed guideline recommended medicines at optimised doses for secondary prevention of MI.
Two hundred patients discharged from hospital following a MI were randomised into intervention (n=100) and control groups (n=100). Patients were interviewed at 5-6 weeks, 6 and 12 months post discharge to determine the medications prescribed, medication adherence and elicit their beliefs about medicines for secondary prevention of MI using the repertory grid technique. In the intervention group, patient’s beliefs about medicines were communicated by the researcher to their community pharmacist, who used this information to discuss the patient’s medication beliefs at designated interviews (3, 6 and 12 months). The control group was provided with usual care by their community pharmacist. Patients also completed the Beliefs about Medicines Questionnaire (BMQ) Specific at 6 and 12 months post discharge. Medication adherence was measured using the Medication Adherence Reporting Scale (MARS®) (a score of ≤23 was categorised as non-adherent) and a Medication Possession Ratio (MPR) of the lipid lowering agent, (where a MPR <80% was categorised as non-adherent). Constructs and medication beliefs elicited from the repertory grid and beliefs elicited from the BMQ Specific were compared between the intervention and control groups and also between patients categorised as adherent and non-adherent.
There were 137 patients remaining in the study (intervention group n=72, control group n=65) at twelve months.
The results addressing the four hypotheses at twelve months post MI were as follows:
1.) There were no significant differences in medication adherence between patients in the intervention and control groups. Measuring adherence using MARS®, in the intervention group 15 (21%) patients were categorised as non-adherent compared to 9 (14%) patients in the control group, (p=0.283). Using the MPR of the lipid lowering agent to measure adherence, there were 20 (29%) patients in the intervention group and 16 (25%) patients in the control group that were non-adherent, (p=0.605).
2.) More patients in the intervention group generated constructs themed, ‘improve blood flow’, whereas patients in the control group generated more constructs themed ‘regular medicine’. Beliefs about medicines elicited from the repertory grid technique, revealed patients in the intervention group believed their lipid lowering was a benefit to the heart whereas this association was weaker for patients in the control group. There were no significant differences in necessity and concerns beliefs between patients in the intervention and control groups, (p=0.09), as determined using the BMQ Specific.
3.) Patients categorised as adherent more frequently reported their medicines as a ‘benefit to heart’ whereas those categorised as non-adherent more frequently reported constructs relating to ‘medication taking’ and ‘benefit to me’, (p=0.009). Beliefs about medicines elicited from the repertory grid technique revealed patients in the adherent group believed their lipid lowering affected cholesterol and was a benefit to the heart whereas non-adherent patients did not report this. Adherent and non-adherent patients had similar levels of necessity and concerns beliefs as determined using the BMQ Specific, (p=0.884).
4.) There were no significant differences in the prescription of guideline recommended medicines or doses between the intervention and control groups.
Discussing patients’ medication beliefs did not improve medication adherence. The lack of difference in adherence between the intervention and control group may be a reflection of factors other than beliefs affecting adherence. Further research should target patients where it is known that beliefs are impacting on adherence.