Target population and sampling

Patients under the age of 18 are considered a vulnerable population in research because of their limited intellectual and emotional capacity, and obtaining proper informed consent is difficult [34]. The target population for the study therefore included patients above the age of 18, both new and follow-up, presenting to the PSD of CHBAH, Gauteng.

To attend the clinic, patients must have a referral, that is, they cannot self-present for consultation. Referrals are either internal, from other outpatient clinics in the hospital, or external. There is only one PSD at this hospital, with an outpatient clinic held every Monday starting at 7 a.m. The target population (N) is approximately 120 patients seen monthly at the PSD of CHBAH.

Target sample size (n) was determined using the single population proportion formula [35] considering a patient satisfaction rate of 84% (P) in Gauteng Province [36], 5% sampling error (d), and a 95% confidence interval (Zα/2). Therefore, the calculated minimum sample size, assuming a nonresponse rate of 10%, was found to be 77.

The sampling strategy employed in this study was systematic sampling [37]. This is a probability sampling method, in which researchers select a member of the population at regular intervals (k) determined in advance. A list of the population is not available in advance. The population order is random; therefore, a representative sample is expected using the systematic sampling method, allowing for conclusions to be made about the population [37]. Systematic sampling remains the preferred method over simple random sampling when the risk of data manipulation is low and the relevant data do not show patterns [38]. There were no discernible patterns in the research population of this study.

The first patient was chosen using the lottery method. The interval (k) is calculated as the target population divided by the target sample size – N/n (120/77 = 1.55); therefore, every second patient was selected. Applying this approach, the number of respondents obtained was 142 patients, which exceeded the target population and sample size. The high number of respondents could be explained by the higher than average number of patients attending the clinic at the beginning of the year, as no outpatient clinics were held during December of the previous year.