Population and design
Population
Physicians practicing in Cameroon (including general practitioners and specialist) will be included in the study. Doctors currently working in five out of Cameroon’s ten regions will be recruited by convenience sampling. The following regions will be included: Adamawa, Centre, Littoral, North West and South West regions. The selected regions were chosen due to easy accessibility and they cut across the major linguistic and religious divide of Cameroon. Amongst the five selected regions are three predominantly French-speaking regions and two English-speaking regions (of which one is a Muslim region and four are predominantly Christian). The choice of regions was mainly by convenience for investigators, and also to account for the diversity of Cameroon.
In the selected regions, rural/urban health institutions will be randomly selected and all consenting physicians will be recruited. Physicians working in the three different levels of health care delivery; the tertiary, the secondary, and the primary health services will be recruited as well as physicians in confessional and private hospitals.
Design
This will be a cross-sectional study that will run for a period of 2 months. Physicians working in the target regions will be visited in the hospital and informed consent sought. Once the physician signs the consent, a printed questionnaire will be handed to them, which they will be asked to fill in at their convenience and hand in. Doctors will receive either an English or a French copy of the questionnaire, depending on their preferred language of expression. All medical doctors will be recruited until the minimum sample size is attained.
Instrument
Data will be collected using a printed structured questionnaire, which comprises two sections. Section one will include: socio-demographic characteristics; work environment and leisure time activities. The following variables will be assessed: age; sex; marital status; presence of dependents; duration of medical practice; location of practice (rural vs urban); level of healthcare practice (primary vs secondary vs tertiary); number of patients seen; remuneration; number of hours of work; number of night calls; working relationship with colleagues; annual occupational leaves; average night’s sleep; practice of physical exercise; alcohol use; smoking; and history of any chronic disease.
Section two will assess burnout using the Maslach Burnout Inventory-Human Services Survey (MBI-HSS) [13], which is a reliable, widely used and a validated tool for assessment of burnout in African healthcare personnel [14]. The MBI-HSS is a self-administered questionnaire consisting of 22 statements on job-related feelings which the respondent answers according to their perceived frequency of occurrence, ranging from never to every day (on corresponding scale from 0 to 6). The three main subsets—emotional exhaustion; depersonalization and lack of personal achievement—are scored separately. Each of these tenets of burnout are further categorized according to severity (low, moderate or high level of burnout) [13].
Sampling
Physicians will be recruited by convenience sampling.
Sample size
Using the following formula [15]:
$$n = \frac{{Z^{2} P \left( {1 - P} \right)}}{{d^{2} }}$$
where n = sample size, Z = Z statistic for a level of confidence, p = expected prevalence or proportion (in proportion of one; 32%, p = 0.32) [16], and d = precision (in proportion of one; if 5%, d = 0.05). Z statistic (Z): for the level of confidence of 95%, which is conventional, Z value is 1.96. For a 95% confidence intervals (CI).
A minimum of 335 physicians will be required for this study.
Data management and statistical analysis
Data will be collected using a printed data collection form. Data entry will be done using EPI Info version 7 (CDC, Atlanta). To ensure accurate data entry, a random sample of 10% of the data will be crosschecked. Data back-up will be done daily and data analysis will be carried out using Epi info software.
Results will be presented as count (percentages), mean and standard deviation (SD) or median and 25th–75th percentiles as appropriate. The reliability and validity of the MBI-HSS will be determined by means of Cronbach alpha coefficients and factor analyses. Pearson correlation will be used to assess univariate associations between continuous variables. Multivariable logistic regressions will be used to identify determinants of burnout amongst doctors; odd ratio (OR) with 95% confidence intervals (CI) will be presented. A p value < 0.05 will be considered significant.
Objective one
To determine the prevalence of burnout syndrome amongst Cameroonian doctors. The three subscales of the MBI (emotional exhaustion; depersonalisation and personal accomplishment) will be used to define burnout. Accordingly, those who score high on the emotional exhaustion (score of 27 or higher) and/or depersonalization (score of 10 or higher) domains of burnout, will be considered to have manifestations of professional burnout [17].
Objective two
To identify potential determinants of burnout among Cameroonian doctors. Standard descriptive statistics and Chi square test or Mann–Whitney U test/two-sample Students t test procedures, as appropriate, will be used for univariate analyses to characterize and compare potential determinants such as: relationship status, sex, age, duration of practice, site of practice (rural vs urban), career level (specialist vs generalist).
Objective three
To compare the prevalence and determinants of burnout among specialist physicians and general practitioners in Cameroon.
Multivariate logistic regression analyses will be done for general practitioners and specialist physicians to identify socio-demographic and professional characteristics associated with burnout.
A p value < 0.05 will be considered statistically significant.
