Patient satisfaction is a subjective measure and depends on the patient’s preferences and perceived expectations. In this small survey, the patients demonstrated an overall high level of satisfaction with counselling services provided by both pharmacists and nurses. Patients expressed greater satisfaction with the information about medication side effects that was provided by pharmacists, and there was a trend toward greater satisfaction with information about side effect management provided by nurses.
Subject selection was nonrandom, but was probably unbiased because subjects for both groups were drawn from the same sample population according to the same inclusion criteria. In addition, recall bias, a common criticism of data obtained by questionnaire, can be discounted because there was a high level of patient recall (see Table 3), and insignificant time lapse between events (2 to 4 weeks). This short data collection period assisted in limiting recall bias but might not have been sufficient in duration to represent patients’ knowledge retention over the long term. An increase in the number of subjects would have maximized the statistical power for the satisfaction survey and knowledge assessment (i.e., by decreasing the chance of type II errors). With regard to the survey format, despite the fact that simple language was used, the participants for whom English was not their first language had more difficulty in completing the questionnaire.
When a Likert scale is used for questionnaire responses, numbers are arbitrarily assigned to categories (e.g., poor = 1 and excellent = 5). Traditionally, the data are treated as real numbers by application of statistical tests to calculate means and standard deviations. However, applying ordinal logistic regression to data in ordered categories is a better method of analyzing the data where the probability of answering within a category can be expressed in terms of an odds ratio. As shown for question 2 (satisfaction with information about adverse effects), pharmacists demonstrated an increase in the odds ratio for receiving a higher response (1.66:1). The internal consistency of the scales measured by Cronbach’s a was well above the expected satisfactory value, which indicates that the generated scale was reliable and the results were reproducible (a = 0.8823, expected value 0.70).
Table 3. Patients’ Knowledge of Medication Regimen
The patients demonstrated a high level of knowledge of their medication regimens, which illustrates the benefit of additional support provided by pharmacists. These results are consistent with results from previous studies showing that pharmacist intervention improves patients’ ability to recall their medication and increases patients’ knowledge of side effects after medication counselling. In addition, the pharmacist’s involvement in the TB counselling program has been beneficial in ensuring that drug histories are more accurate and complete.
Patient satisfaction has practical implications for improving the quality of pharmaceutical care. These results indicate that pharmacists and nurses consulting with patients on medication use can increase overall levels of patient satisfaction. Pharmacists and nurses working cooperatively with the patient can promote the correct use of and access to medicines while emphasizing the importance of medication adherence to achieve the BCCDC mandate to control and eliminate active TB while preventing the emergence of drug resistance. This appropriate collaboration of health care professionals is reflected by the philosophy of teamwork at the BCCDC. The close proximity of the pharmacy to the TB Clinic allows for direct patient consultation and review of antituberculosis medication by the pharmacist. Furthermore, the pharmacist’s participation in counselling for 3 days per week has helped to reduce the nursing workload, which allows nurses more time for contact tracing, skin testing, and reporting.
Future directions suggested by this research may include pharmacist participation in counselling patients with active tuberculosis, who often have concomitant respiratory illnesses, complex drug histories, and the possibility of drug interactions. As well, the pharmacist’s role may be expanded to include follow-up, including summarizing prior drug therapy and observed responses to treatment. There is now a need for further work to determine if there is an impact on clinical outcomes such as improving adherence to antituberculosis therapy and decreasing the risk of treatment failure and drug-induced hepatitis. Prospective studies should incorporate measures of patient recall and rates of adherence to the antituberculosis regimen.
This patient survey demonstrated high levels of satisfaction with counselling services provided by both pharmacists and nurses. There was significantly higher patient satisfaction with the written pamphlet information used as a counselling aid by pharmacists. By providing a medication pamphlet for the patient to review at home, pharmacists can reinforce in writing important counselling points that might be easily missed by the patient during the initial interview. Pharmacists and nurses working cooperatively with the patient can promote the correct use of and access to medicines while emphasizing the importance of medication adherence to ensure a lasting cure of active TB and prevention of acquired drug resistance.