Factors Affecting Compliance with Clinical Practice Guidelines for Pap Smear Screening among Healthcare Providers in Africa: Systematic Review and Meta-Summary of 2045 Individuals

Background Although the importance of the Pap smear in reducing cancer incidence and mortality is known, many countries in Africa have not initiated yet widespread national cervical cancer screening programs. The World Health Organization (WHO) has published Clinical Practice Guidelines (CPGs) on cervical cancer screening in developing countries; however, there is a gap between expectations and clinical performance. Thus, the aim of this study was to conduct a systematic review and meta-summary to identify factors affecting compliance with CPGs for Pap screening among healthcare providers in Africa. Methods And Findings: MEDLINE, Scirus, Opengate and EMBASE databases were searched in January 2012. Studies involving medical personnel practicing in Africa, whose outcome measured any factors that affect medical personnel from using a Pap smear to screen for cervical cancer, were included. Two reviewers independently evaluated titles and abstracts, then full-texts, extracted data and assessed quality of the included studies. A descriptive analysis of the included studies was conducted. We calculated Frequency effect sizes (FES) for each finding and Intensity effect sizes (IES) for each article to represent their magnitudes in the analyses. Of 1011 studies retrieved, 11 studies were included (2045 individuals). Six different themes related to the factors affecting compliance with CPGs were identified: Insufficient Knowledge/Lack of awareness (FES = 82%), Negligence/Misbeliefs (FES = 82%), Psychological Reasons (FES = 73%), Time/Cost Constraint (FES = 36%), Insufficient infrastructure/training (FES = 45%) and also no reason given (FES = 36%). IES for articles ranged between 33 and 83%. Conclusions These results suggest that prevention initiatives should be comprehensive to include education and resources needs assessments and improvement, Pap smear test training, strategies on costing, and practitioner time studies.


Introduction
Cervical cancer is the second most common cancer among women worldwide, and the leading cause of cancer-related deaths among women in developing countries. In 2008, 453,000 new cases and 242,000 deaths occurred from cervical cancer, with 83% of the cases of cervical cancer occurring in developing countries [1,2]. Although the effectiveness of the Pap smear in reducing cervical cancer incidence and mortality has already been demonstrated in many developed countries [3,4], there is a wide disparity in rates of screening for cervical cancer in developing countries with the average screening coverage rate in developed countries at 63% compared to 19% in developing countries [5]. In the developing world, women at highest risk for developing cervical cancer are among the least likely to be screened [5]. In spite of the World Health Organization (WHO) published clinical practice guidelines (CPGs) on cervical cancer screening in developing countries, most developing African countries have not initiated widespread national cervical cancer screening programs. A WHO report on cervical cancer screening in Sub-Saharan Africa noted that while this region was the most affected by cervical cancer, it has access to less than 5% of the global resources for cervical cancer prevention [6]. Additionally, health system strengthening, in order to promote, restore, or maintain health, requires the translation of WHO guidelines into national guidelines and the development of implementation strategies [7].
In African countries, not only is the incidence of cervical cancer high but a large proportion of patients present with advanced disease (> stage IIb) [8,9]. These data suggest that in African countries there is a lack of a successful large scale screening programs. Despite the significant lack of screening on the health of women in Africa, there have been no previous systematic reviews to identify the factors leading to compliance or non-compliance with CPGs on Pap smear testing for cervical cancer screening in Africa.
Physician behavior, specialty and gender have all shown to play an important role in patient compliance [10,11]. In a systematic review of physician non-adherence to general clinical practice guidelines, the authors noted many factors could limit adherence like lack of awareness, familiarity, agreement, self-efficacy, and outcome expectancy as well as inertia of previous practice. However, the authors emphasize that, since barriers may differ from setting to setting, these results may not be generalizable [12].
In Africa, factors that have been shown to affect physicians' compliance include but are not limited to: busy clinics and lack of manpower, lack of access to care, lack of transport to care and opposition to care by men [13]. Another review investigating guideline compliance concluded that there is a considerable gap between expectations and clinical performance but did not investigate the reasons for this disparity [14]. These studies did not address the compliance with clinical guidelines on Pap smear testing. The objective of this systematic review is to identify factors affecting compliance with Clinical Practice Guidelines for Pap screening among healthcare providers in Africa.

Protocol and Registration
This systematic review is reported in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) statement [15]. The protocol of the study can be found in File S1, and the PRISMA check list in Table  S1.

Eligibility Criteria
The following inclusion criteria were considered: 1) Studies involving medical personnel; 2) Studies whose outcome measures included any factors that affect medical personnel from using a Pap smear to screen for cervical cancer, 3) Studies conducted in Africa, and 4) Observational design studies.
We excluded studies that retrospectively analyzed clinical trial data, unpublished articles, dissertations, and abstracts without full text. In addition, manuscripts in languages other than English, French or Portuguese were not included in the review.

Information Sources
We searched the following electronic databases for published literature up until January 2012: PubMed, Scirus, Opengate, Directory of Open Access Journals (DOAJ), and EMBASE. We did not use limits for language when searching the databases. The references of the included articles were reviewed, as well as we performed citation analysis of the included studies using Google Scholar, and also sought experts' suggestions through email communications.

Search
The initial search comprised the following Mesh terms "Vaginal Smears", and "Africa" and the related entry terms. The complete search strategy used for the PubMed database is shown in Table S2.

Study Selection
Titles and abstracts of the retrieved articles were independently evaluated by two reviewers (E.A. and M.V.). Abstracts that did not provide enough information regarding the eligibility criteria were kept for full-text evaluation. Reviewers independently evaluated full-text articles and determined study eligibility. Disagreements were solved by consensus and if disagreement persisted, we sought a third reviewer's opinion (A.J.).

Quality of Studies
Several tools have been proposed for evaluation of methodological quality of observational epidemiological studies [16]. In this study, quality assessment was based on a checklist specific for evaluating cross-sectional studies [17], which is based on 15 items. We chose this checklist based on a systematic review previously published [18], which recommends to use checklists rather than scales, as well as to use a tool as specific as possible, considering study design and area.

Data Extraction
Two reviewers (E.A. and M.V.) independently conducted data extraction and disagreements were also solved by the third reviewer (A.J.). The data extraction spreadsheet was pretested by the two reviewers and alterations were made as necessary. General characteristics of the studies were collected, such as: year of publication, location and setting where the study took place, number of health care providers, health providers characteristics. In addition, we collected from each study the factors cited as affecting compliance with CPGs for Pap smear screening among healthcare providers in Africa.

Data Analysis
We performed a descriptive analysis of factors affecting healthcare provider compliance with CPGs for Pap smear in Africa, described in the included studies. Factors affecting compliance were grouped by similarity, and within each group, brief descriptions of the findings were generated. To represent the magnitude of each finding, we adapted the methodology proposed by Sandelowski M, et al (2007) [19]. Frequency effect sizes (FES) were generated by dividing the number of studies citing a particular theme by the total number of studies in our final list and multiplied by 100. The themes with higher FES are the ones more prevalent across the included papers. To represent the magnitude of each report, intensity effect sizes (IES) were calculated. Specifically, for each study, the number of findings with a FES >25% was divided by the total number of findings with frequency effect size >25%. Additionally, for each study, the number of themes it cited was divided by the total number of themes overall [19]. The studies with higher IES are the ones presenting more themes. As the lowest FES identified was 36%, and both IES were the same, we report only one value for IES for each study.

Characteristics of included studies
A total of 1011 records were identified and screened with 94 full-text articles assessed for eligibility based on the inclusion and exclusion criteria. Ultimately, 11 surveys were included in the qualitative synthesis and meta-summary. Figure 1 illustrates the search and article selection process as well as the numbers of articles retrieved and included during the article selection phases of this project.

Studies Characteristics
Included studies had a total of 2045 individuals. Sample sizes varied from 60 to 483 with a mean of 185.9 of which two studies included both females and male respondents. The characteristics for the studies included in this analysis are shown in Table 1. Table 2 demonstrates the risk of bias assessment of the included studies. There was prevalence of "Yes", which means low risk of bias (93/165), however many studies were 'Unclear' in their presentation of some items, representing moderate risk of bias (33/165). Also, some studies did not present come items at all ("No"), being classified as high risk of bias (39/165).

Synthesis of Results
Factors affecting compliance with CPGs for Pap smear screening identified in the articles are presented in Table 3. The systematic review uncovered multiple main themes for factors affecting non-compliance with CPGs for Pap screen including: insufficient knowledge/awareness, negligence/    [21] Reasons for non-use of Pap smear by female medical practitioners: poor health consciousness -2 (3%), do not feel susceptible to cervical cancer -6 (9%), scared of the outcome -4 (6%), too busy The results of the meta-summary demonstrated all findings reported a FES > 25, ranging range from 82% to a low of 36%. Effect sizes greater than 25% are considered more relevant. Non-compliance factors had the highest FES, meaning the most common findings amongst studies were Insufficient Knowledge/Lack of Awareness and Negligence/Misbeliefs. Each of these themes and their corresponding FES are displayed in Table 4. The IES for each of the included articles is displayed in Table  5. The results demonstrate IES ranging from 50 to 83%. The study of Olaniyan OB, et al (2002) contains the largest number of findings, in contrast to the study of Tarwireyi F, et al (2003), which had the lowest number of findings.

Discussion
To our knowledge, this is the first systematic review describing the factors that affect compliance with CPGs for Pap smear screening among healthcare providers in Africa. Our  The factors found in our study to be most cited and therefore that had the highest FES were Insufficient Knowledge/Lack of Awareness and Negligence/Misbelief for not complying with Pap smear guidelines. Surprisingly, even among professionals with knowledge on Pap smear and cervical cancer, most of them have never had a Pap smear. This result is in agreement with another study that found that improved awareness of Pap smear may not affect its use in Nigeria, in which all respondents were aware of the Pap smear but only 18% had used it before [31]. In South Africa, a national study evaluating 20,603 women in public health service around the country found that 80% had never had a Pap smear [32]. In this South African context, barriers to effective screening programs are    further complicated by the legacy of racial and geographic inequity in education and literacy levels, health infrastructure and access to health services, thus representing a considerable challenge [33]. Previous studies have evaluated cervical cancer and Pap smear awareness in different populations [34][35][36][37][38][39][40][41][42]. A recent study conducted in Nigeria aimed to determine the level of awareness of cervical cancer and Pap smear test, and factors associated with the utilization of Pap test among female civil servants in Jos. In this study, 51% of participants had cervical cancer awareness while 39% were aware of Pap smear testing. In addition, the study demonstrated that 50% of the participants cited the media and hospitals as their source of information about cervical cancer and Pap smear, thus showing their important role in information dissemination and education. The authors highlight that a health education program about cervical cancer that incorporates media might be very impactful in that context [34]. Another study assessing cervical cancer and Pap smear awareness among undergraduate students showed that 71% of the students were aware of cervical cancer, however, only 33% were aware of Pap smear [36].
In this context, previous study carried out in Nigeria among sexually active woman demonstrated that only 26% of the respondents were aware of cervical cancer screening, and in addition, only 47% of the aware group knew that the test was to screen cervical cancer. The authors report an association between the educational status and the knowledge of Pap test, but not in relation to the test utilization [39].
Another study determining the level of awareness and uptake of cervical screening in Owerri, Nigeria has demonstrated as major finding that 52.8% of the respondents were aware of cervical screening and that 7.1% had ever done the test. The main reasons for not doing it were "no need for it", lack of knowledge that it could be done locally, and fear and anxiety over a positive result [35].
Although our results demonstrate high IES throughout the studies (greater than 25%), meaning that the themes identified were consistent across the studies, the study quality assessments demonstrated that not all studies presented some of the assessed items or it was unclear, representing high and moderate risk of bias, respectively. Specifically, most studies did not have a representative sample and blinding of the outcome assessment. These limitations might have affected the quality of the information extracted from these studies, the size of the effect.
Findings from our study suggest though that while Insufficient Knowledge/ Lack of Awareness is a large limitation, to CPG adherence, interventions for improving adherence must be more comprehensive to further address Negligence/ Misbeliefs and Psychological Reasons and well as the Time/ Cost Constraints and Insufficient Infrastructure/Training. Given the high incidence of advanced stage cervical cancer in Africa, there is an urgent need for successful strategies for improving clinical practice guideline compliance. These findings support adopting a multifaceted approach, not only addressing provider education, but also patient education, cost, timing, infrastructure and psychosocial concerns. In addition, future work in this area should be performed to further understand the extent of the actual versus perceived psychosocial reasons and the validity of these reasons from the perspective of the patient. Secondly, further understanding of cost, time and infrastructure limitations, whether perceived or actual, should be undertaken.
Our data coincide with a previous study that reported implementing multifaceted approaches are needed to achieve optimal screening and treatment for cervical cancer [33]. Such approaches should involve improvement of health workers and community knowledge by implementing then evaluating education, and communication strategies, as well as establishing appropriate structure and guidelines, and developing new technologies [33]. Research has been conducted to evaluate alternative cervical cancer screening technologies in low-resource setting, which would require fewer resources and infrastructure, thus having the potential of increasing the feasibility of cervical screening implementation [43]. However, it has been demonstrated that technological interventions and innovations alone are not sufficient to improve cervical screening programs. Other concerns intrinsic to health systems in Africa include human resources concerns such as training, increasing demands on personnel, attrition, and skills mix. These should be addressed concurrently within a comprehensive workforce development strategy, alongside work to make the health care delivery system functional, otherwise the task shifting would be limited [44].
In conclusion, studies that have evaluated factors affecting healthcare provider clinical practice guideline adherence for Pap smears in Africa had high effect sizes, demonstrating that the themes identified were consistent across the studies. The studies suggest provider non-compliance factors include lack of awareness and knowledge about protocols, time and cost constraints, and lack of infrastructure or training on performing Pap smears. These results suggest that prevention initiatives should be multifaceted including education, resource needs assessments and improvement, Pap smear testing training as well as strategies on costing and practitioner time studies.