This was a cross-sectional study conducted amongst PCDs in private practice from the Federal Territory of Kuala Lumpur (KL), Selangor and Penang from June to December 2016. Private PCDs were studied as a previous study has already reported the association between physical activity intensity and exercise counselling in government primary care settings(6). All PCDs in private practice were included whereas those excluded were doctors who were away on prolonged leave, pregnant or who had conditions or disabilities limiting exercise during the data collection period.
The sample size for this study was calculated from a total population size of PCDs in KL, Selangor and Penang of 842 (728 from Kuala Lumpur and Selangor and 114 from Penang respectively). Considering a 20% dropout rate, the sample size required was 318. Universal sampling was done as systematic sampling could not be carried out in view of the AFPM’s policy of not allowing access to members’ particulars.
The study tool consisted of a self-administered questionnaire divided into two sections. The first section measured physical activity intensity of PCDs using the English version of the International Physical Activity Questionnaire (IPAQ) which has been validated in Malaysia(7). The short version of the IPAQ which is more suitable for self-administration was used and consists of 7 questions assessing time spent in activities of varying intensity which are walking, moderate-intensity and vigorous-intensity activities over the last 7 days. From this a score was obtained in MET-minutes per week. MET stands for Metabolic Equivalent of Task which is a functional measurement of energy use. During data analysis, the IPAQ scores were calculated and categorised into low, moderate and high levels of physical activity intensity as per IPAQ guidelines(8).
The second section consisted of independent variables including sociodemographic data which included age, gender, height, weight, body mass index (BMI), smoking status, presence of chronic medical illnesses, years working, years in primary care, average number of patients seen per day and highest academic qualification (undergraduate/post-graduate) and frequency of exercise counselling.
Outcome measures comprised of initiation of (proactive) exercise counselling to patients with five chronic medical diseases (i.e. cardiovascular diseases, hypertension, type 2 diabetes mellitus, obesity/metabolic syndrome, dyslipidaemia). These conditions were chosen as exercise has been shown to benefit these conditions(1). An example of a question is as follows:
11b) What percentage of your patients on the following chronic disease follow up visits do you initiate counselling on exercise (before a patient asks about it) IN A WEEK?
i) Cardiovascular Diseases (Heart Disease and Stroke)
None
(0% of my patients)
|
Some
(25% of my patients)
|
Half
(50% of my patients)
|
Most
(75% of my patients)
|
All
(100% of my patients)
|
1
|
2
|
3
|
4
|
5
|
The exercise-counselling questions had a 5-point Likert scale and underwent prior content validation from a panel of 5 experts in primary care. Following their feedback, the questions were reconstructed until they were sufficiently clearly understood. Subsequently, face validation was done via a pilot study among 25 primary care doctors at the University Malaya Medical Centre and minor modifications were made to the final version.
This study was approved by the Medical Research Ethics Committee of University Malaya Medical Centre (UMMC MREC ID: 20166-2590).
The Academy of Family Physicians Malaysia (AFPM) agreed to distribute the questionnaires to the 728 PCDs registered in KL and Selangor. Self-addressed, stamped envelopes were sent of which 192 were returned by post (10 were excluded as they did not fulfil the inclusion criteria). To further increase response rates, the researcher attended workshops organized by the AFPM for candidates of the Diploma of Family Medicine programme (DFM) and was able to distribute another 20 questionnaires.
Private PCDs from Penang were recruited via the annual Penang General Practitioner’s Conference organized by the MMA in Penang. At the conference, 114 questionnaires were distributed to participants who fit the inclusion criteria. A total of 78 questionnaires were then collected in person or placed in designated labelled boxes at the end of each conference day. The recruitment process for all final included results is summarized in Figure 1.
Data from this study was subsequently analysed using Statistical Package for Social Sciences (SPSS) software, Version 23.0. Descriptive statistics was used to look at physical activity intensity of private primary care doctors and frequency of initiation of exercise counselling to patients with chronic diseases. With regards to physical activity, participants’ IPAQ scores were calculated to get continuous variables and further categorized into ‘Physically inactive’ (Low IPAQ scores), and Physically active (‘Moderate’ or ‘High’ IPAQ Scores). For exercise counselling frequency, Likert scores from 1-3 were classified as ‘Rarely initiates’ and 4-5 as ‘Often initiates’. To determine associations between physical activity intensity and confounding factors with initiation of exercise counselling for each of the five chronic diseases, bivariate regression was conducted for categorical variables and continuous variables that were normally distributed.
For continuous variables that were not normally distributed, the Kruskall-Wallis test was done. Finally, to look at the independent association between physical activity intensity and exercise counselling, multiple regression analyses was used. For this, all variables with a p- value of ≤ 0.25 were included in the analyses(9).