Introduction
Self-efficacy is a critical element in social cognitive theory and a social foundation of thoughts and action, defined by Bandura as “The belief in one’s capabilities to accomplish a certain level of performance” [
1]. This belief depends on psychosocial interactions between four constructs: individual performance accomplishments, vicarious experience, verbal persuasion, and physiological information [
2]. Sufficient research self-efficacy indicates physicians’ confidence in demonstrating research skills, enabling them to be critical of the published information and implement evidence-based medicine. Conversely, insufficient research self-efficacy distorts certainty in clinical decisions and blocks continuous professional development [
3‐
5].
In a study conducted in the United States of America (USA) among academic physicians’ residents and fellows, self-efficacy was assessed through the online 92-item Clinical Research Appraisal Inventory. Fellows had higher self-efficacy than physicians’ residents due to their higher years of research experience [
6]. Consistent results were revealed in another paper, where previous experience and mentoring are influential sources of research self-efficacy [
7]. Both studies were conducted among residents still under training in their medical education path [
6,
7]. Notably, the whole body of literature focuses on developing research skills among medical students and academic settings [
8,
9]. However, no studies have discussed research self-efficacy among physicians in practice, even though low research self-efficacy affects generating and justifying new knowledge essential in the practice of evidence-based medicine [
10].
The Qatar National Strategy prioritizes research productivity, provides funds, and considers publication one of the enabling requirements for promotion. Thus, motivating physicians to work on research [
11].
No previous research addressed research self-efficacy and its determinant among Qatar’s primary healthcare physicians. We seek to address this gap by assessing research self-efficacy levels among physicians in primary health care and identifying significant determinants.
Purpose
Examining the level of research self-efficacy among physicians working at primary health care centres (PHCC) in Qatar will enable us to understand the baseline and the determinants of low research self-efficacy to implement specific interventions in continuous professional development (CPD), which will reflect on the quality of care at primary health care level than at the health care system.
Methods
Ethical approval, relevant guidelines, and informed consent to participate
All methods were performed according to the relevant guidelines and regulations of Primary Health Care Corporation Research Committee approval (PHCC/RC/18/09/001).
Study design and setting
An analytical cross-sectional design was employed among primary health care physicians. We sent a web-based survey to all physicians registered at PHCC via email with an information letter and a link to the electronic version of the questionnaire. The data collection took place from the 21st of October to the 23rd of November 2018.
PHCC provides preventive and free-of-charge services to the whole community in Qatar through 23 primary health care centres distributed across the country (total number of physicians registered at PHCC at time of data collection = 634).
Sampling technique
We extracted the list of all the physician’s emails registered in primary health centres from primary health care operations. Again, a convenience sample was utilized, as we aimed to include all the physicians registered.
Sample size and participants’ enrolment
Sample size calculation was not used in the study as we are using a convenience sample.
Eligible participants involved all physicians registered at primary health care who were willing to participate and communicate in English during the study period. Therefore, we did not have specific exclusion criteria.
Data collection
The open survey was developed through DACIMA Software, an advanced web-based electronic data capture (EDC) software for collecting, managing, and reporting research data.
The survey was conducted only in English as all the physicians are knowledgeable and comfortable answering surveys and communicating in English. The PHCC research delegates emailed the physicians with an information letter and a link to the electronic version of the survey. There was no randomization of items or surveys. The average time of each participant was 15 min. We ensured data collection completion through two reminders sent each week.
The survey was advertised through the PHCC website and official emails. JavaScript checked the consistency and completeness of the submitted surveys, and it enforced one response option. The open poll did not utilize adaptive questioning. The number of items ranged between 18 to 33 per page, and the number of screen pages was 15 pages. We enabled a review check of the respondent to their answers and prevented multiple entries from the same individual through the record IP address of the individual.
Materials
Only complete questionnaires were considered in the data analysis phase. We utilized the CHERRIES (Checklist for Reporting Results of Internet Electronic Surveys) criteria during writing the protocol [
12].
We conducted piloting on 10% of the total sample that allowed testing of the usability and technical functionality of the electronic survey. Piloting did not result in any changes to the tool, so we included it in our total sample. In addition, we measured the time needed to complete each questionnaire.
The dependent variable is our primary outcome that relates to research self-efficacy among primary health care physicians [
13]. This was defined as one’s ability, capabilities, and confidence to perform research [
1].
Independent variables are variables that self-regulate and may have an impact or stimuli on the dependent variable [
13]. The independent variables included sociodemographic characteristics and professional development factors.
We utilized Self-Efficacy in Research Measure (SERM) developed by Phillips and Russell. It is a 33-item English questionnaire that measures research efficacy among professionals in various career fields. It focused on four dimensions of research self-efficacy, including (a) research design skills (8 items), (b) practical research skills (8 items), (c) quantitative and computer skills (8 items), and (d) writing skills (9 items). In addition, the respondent indicates how confident they are to perform each research: 0 (belief of inability) to 9 (belief of performing in the full item ability). The total score for each participant varies from 0 to 297. The higher the scores, the more confident the participant in the research. The Coefficient alpha of total SERM scores was reliable at 0.96 [
4,
13,
14].
The structured questionnaire was developed through extensive literature. It included questions regarding sociodemographic (age, gender, year of graduation, and nationality) and professional development factors (clinical experience in PHCC in years, the average number of patients seen per day, speciality, academic position, computer literacy, experience developing the clinical guideline, previous training in research methodology, and previous publications).
Computer literacy was assessed through one question (Not at all, basic, good, excellent), and self-efficacy in using guidelines was assessed through five items, and it was questioning confidence in utilizing guidelines (I strongly agree, I disagree, I strongly disagree).
An expert panel established the questionnaire’s face validity (English) and relevance to the study objectives. Additionally, we verified the content validity through an extensive literature review to ensure the consistency of the contents and scale level. Finally, using Lawshe’s method, each item was rated for its importance and relevance by applying a three-point scale: (1) not necessary, (2) useful but not essential, and (3) essential. The universal agreement between the three evaluators was 80%.
Analysis
We analyzed data through Statistical Package for the Social Sciences (SPSS), Version 20.0 (IBM Corp, Chicago, Illinois, USA). The statistical analysis involved descriptive summarization of the variables: categorical variables in frequency and percentages and continuous variables in mean ± standard deviation (SD).
We conducted a bivariate analysis to test associations between dependent and independent variables through Pearson’s chi-squared test (odds ratio (OR) and 95% CI). In addition, we included all determinants in multivariable logistic regression analysis. We computed the adjusted odds ratio (entry method Logistic Regression). We also performed reliability testing and principal component analysis of the ‘Self-Efficacy in Research Measure’ (SERM) tool.
The Domain score was calculated by summing all feasible items in the domain named Index. The Index is the summation of all the scores of each item and multiplying it by a total number of items, 33, then dividing them by 297 (which is the highest code, 9, multiplied by the number of items, 33). We then multiplied the Index by 100. A standardized score was obtained to the total score, then cut-off points of 75 percentile discriminated high self-efficacy from low-self efficacy; several papers inspire domain calculation [
15‐
17].
P-values ≤ 0.05 were considered statistically significant. As data were collected first in Microsoft Excel and then extracted to SPSS, to ensure the quality of data entered, we performed an audit on 10% of the data entered by another researcher.
Discussion
This study investigated research self-efficacy among physicians working in primary health care clinics in Qatar. Almost three-quarters (n = 92, 75.4%) showed insufficient research self-efficacy. In addition, a significant association was found between insufficient research self-efficacy level and lack of participation in clinical guideline development.
Our results reveal that the factor’ participation in clinical guidelines’ increases the likelihood of sufficient research self-efficacy by threefold. Furthermore, the logistic regression model proved that this factor significantly predicts insufficient research self-efficacy. Thus, recommending participation in clinical guidelines will positively reflect their research self-efficacy level and eventually lead to better practice.
Gender was a non-significant factor. This finding is consistent with another study conducted among physicians [
6].
We reported total scales rather than the subscales as, in a previous publication, the confirmatory factor analyses did not support the sub-analysis [
13].
This research is the first to examine research self-efficacy status among physicians in the Arab Gulf region. The regional body of research has focused on medical students only. The strength of this study is that it aims to explore the research skills of primary care physicians. Research competency is an essential skill for physicians as it contributes to evidence-based medicine and improves patient care. It is an original topic that evaluates physicians’ research skills and allows them to develop research capacity-building activities.
The ‘Self-Efficacy in Research Measure (SERM)’ is a valid instrument. In addition, we delivered evidence for a three-factor structure tool (short version) that could be used in time-critical situations for physicians to ensure high response rates.
Several strategies were employed to decrease the measurement bias, including selecting a reliable tool (SERM) with good internal consistency (Cronbach’s alpha = 0.98). Similarly, the content and face validity of the questionnaire were assured through appropriate techniques. The CHERRIES (Checklist for Reporting Results of Internet Electronic Surveys) criteria were utilized in reporting.
The study’s cross-sectional design compromises causation as it specifically lacks temporality. Furthermore, external validity is affected by utilizing a non-probability sampling technique. Fortunately, the selected sample revealed heterogeneity and included variable sociodemographic subgroups, reflecting positively on the population’s accurate account [
18]. However, another limitation was that the survey was long, leading to a low sample size that affects external validity and generalizability with a response rate of 19.2%.
Conclusion
This study indicates that insufficient research
self-efficacy is common among physicians in Qatar. Therefore, health officials
should design and implement targeted interventions to promote research
self-efficacy by mandating physicians to be trained in developing clinical
guidelines. The next step would be a confirmatory,
controlled study. Once that is done, firm recommendations might
be available.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.