Prevalence and associated factors of cervical cancer among women in Ethiopia: A systematic review and meta-analysis

Background: cervical cancer is the second commonest disease of women in Ethiopia. Even though, some studies were conducted to assess the prevalence and associated factors of cervical cancer among women in Ethiopia, a variation in reported prevalence across the studies were observed. The aim of this study was to determine the pooled prevalence estimate and associated factors of cervical cancer among women in Ethiopia. Methodology: A systematic review and meta-analysis was performed based on published and gray literatures between 2010 and 2019. Articles were evaluated using Joanna Briggs Institute database guidelines by two independent authors. The pooled prevalence estimate was calculated using MedCalc software-version 19.0.7 and the pooled odd ratios for predictors was calculated using RevMan software version 5.3. A meta-analysis using a mantel Hansen variance random-effects model was performed to assess the amount of variation in between studies. The report was depicted descriptively using pooled estimated prevalence and odd ratios using tables and plots. Result: From 124 retrieved data, 10 full text articles were eligible for the review. The pooled prevalence estimates of cervical cancer among women in Ethiopia using random-effect model was 15.7%, 95%CI (10.8, 21.3%). Statistically significant heterogeneity between studies was detected (I 2 =94.35% (p<0.0001). Among all associated factors measured: numbers of life time sexual partners >1, OR=0.40, 95% Cl (0.21,0.27), being HIV positive, OR=0.41,95%CI(0.21,0.75), having history of STI,OR=1.99,95% CI (1.02,3.87), women getting widowed, separated or divorced), OR=0.32, 95% CI(0.14,0.75),income <1000birr, OR=1.78, 95% CI (1.19,2.65) and women had experience of contraceptive use, OR=2.32, 95% CI (1.75,3.43) were had statistical The pooled prevalence estimate of cervical cancer among women in Ethiopia was high. There was a variation of cervical cancer reports across studies. Reporting of this information in a consistent manner is important for researchers to enhance future studies and also useful for policymakers and practitioners for better understanding of the burden of cervical cancer in Ethiopia for prevention, diagnosis, and early treatment of the disease. Trial registration: This systematic review for registration in PROSPERO was retrospectively registered on June 20,2019.


Background
Cervical cancer is stated as a major public health problem worldwide. Cervical cancer is a sexually transmitted disease caused by Human papillomavirus (HPV) commonly type 16 and 18 which cause 70% of cervical cancers and pre-cancerous cervical lesions (1).
Cervical pre-cancer goes through many stages and takes many years to develop into cervical cancer. It becomes cancer when the abnormal cells spread below the epithelial layer down into the deeper tissues of the cervix. In 2018 cervical cancer is estimated as the fourth most frequent cancer among leading causes of death for women worldwide. It represented 7.5% of all female cancers. More than 85% of these occur in less developed countries (1).
According to the latest WHO data published in 2017, cervical cancer accounts 0.78% of total deaths in Ethiopia. That is about 1death due to cervical cancer of every 128 deaths, 14 people die of Cervical Cancer each day and an average of 1 death occurs every 2 hours (2).
Cervical cancer is preventable disease through periodic screening and early detection of lesions before progress to cancer. Precancerous lesions can be treated easily by cryotherapy using freezing gas (liquid nitrogen) to destroy precancerous cells on the cervix.
The most frequent method for cervical cancer screening is cytology, and there are alternative methods such as Human Papilloma Virus (HPV) Deoxyribonucleic Acid (DNA) tests and Visual Inspection with Acetic Acid (VIA). VIA is an alternative to cytology-based screenings in low resource settings. It is the 'see and treat' approach.
Low level of awareness, lack of effective screening programs, and lack of attention to women's health are the possible factors that leads to higher the prevalence rate of cervical cancers (13.4%) in the Ethiopia. As a result, more than 80% of cervical cancer cases are detected at a late stage due to lack of information and weak preparedness to provide services (3).
Vaccines against HPV 16 and 18 are recommended by WHO and have been approved for use in many countries including Ethiopia(1).
The vaccine is widely administered in rich countries. While countries with the highest burden of cervical cancer in Africa and Asia are covering late (4).

Ethiopia launched Human Papilloma Virus (HPV) vaccination pilot project in
December 2015 targeting adolescent girls in the 9-13 year age groups in Oromia and Tigray regions (3).
Cervical cancer can be cured if the infection is diagnosed and treated at an early stage (girls between the ages of 9 and 14 years) with the alignment of good life styles (4).
Some epidemiological studies are available to assess cervical cancer distribution and associated factors among women in Ethiopia. But individual studies show a wide variation of cervical cancer distribution over time and across geographical areas in the country. Therefore, the main objective of this systematic review and metaanalysis is to provide a pooled prevalence estimate and to assess associated factors of cervical cancer among women in Ethiopia in 2019 to answering the following research questions:

1.
To what extent cervical cancer is distributed among women in Ethiopia?

2.
What are the associated factors of cervical cancer among women in Ethiopia?

3.
To what extent in-between variance among study reports on cervical cancer among women is observed in Ethiopia? . The country has a federal system of governance with nine regional states and two chartered cities (32).

Inclusion criteria
The inclusion criteria for this study was Published and gray literatures with appropriate observational study designs conducted at health facility level on the prevalence and associated factors of cervical cancer among women in Ethiopia.
Articles reported with English language, Odds Ratios (OR) and corresponding 95% confidence intervals (CI). (Figure 2

Data extraction (selection and coding)
This systemic review and meta-analysis were performed based on published and gray literatures between 2010 and 2019. The retrieved data was screened independently by two reviewer authors (DZ and BE) to verify studies that possibly meet inclusion criteria. Any disagreement was resolved through discussion with a third reviewer (KM.

Strategy for data synthesis
In total, 10 studies were eligible to conduct this study, but not all researchers had used the same classification for each predictor at their study level. When this kind of event happened, a better inclusive classification system or category was selected. But for age categories, midpoint calculation was employed to fit it in the best placement. If all these efforts did not work, studies containing unfitted categories for predictors would not be included in the study. Therefore, 6 full text cross-sectional studies, 1 comparative cross-sectional study and 3 case control studies conducted at health facility level among women in Ethiopia reported from 2010 to 2019 were used. Only 7 studies were used to measure the prevalence of cervical cancer screened by any methods whereas all eligible studies (10 full text articles) were chosen to measure associated factors of cervical cancer among women in Ethiopia.

Data analysis
For this study, 2×2 tables summarizing was computed for each outcome measure.
We also calculated a weighted study effect using a random effects model in case of heterogeneity at P ≤ 0.10 for χ 2 test during meta-analysis. The pooled prevalence estimate was calculated using MedCalc software-version 19.0.7 whereas the overall effects estimate of odd ratios for associated factors were calculated using RevMan software version 5.3.(41). The variance of the study was stabilized with Mantel Hensel before pooling the data within a random-effects or fixed effect meta-analysis model (42). In addition to these, Funnel plot was used to assess the presence of publication bias (43).Moreover, heterogeneity was evaluated by using the χ 2 test and I 2 value( 44).

The pooled prevalence estimate of cervical cancer among women in Ethiopia
Seven studies conducted at health facility level with a total simple size of 3951 were included to compute the pooled prevalence estimate of cervical cancer among women in Ethiopia. The highest prevalence was observed in a study done in gynecology referral Clinics in 2018 (39.5%) (45) whereas the lowest point estimate was observed at a study conducted in North Ethiopia in 2017(6.7%)(25). A Statistically significant heterogeneity between studies was also detected (I 2 =94.35% (p<0.0001) ( Table 4).
This study based meta-analysis forest plot with random effects model discovered that more weight was given for a study conducted in Yirgalem General Hospital in 2017(19) (i.e. the bigger the box, the more participants in the study, the smaller 95% confidence intervals and the higher the precision to predict the pooled estimate).As a result, studies given the higher weights were had the more influence on the pooled effect estimate. Therefore; the combined point estimate of cervical cancer was found to be 15.7%, 95%CI (10.78%, 21.31%) ( Figure 3) In figure 4, publication bias had been indicated by funnel plot symmetry or asymmetry. This visual inspection of the funnel plot confirmed that the plot was somewhat symmetrical. This indicates that there was no possible publication bias up on discovering the prevalence of cervical cancer among women in Ethiopia.

Factors associated with cervical cancer among women in Ethiopia
( Therefore, the overall effect estimates of women's age (OR=1.43, 95 %CI (0.65, 3.12) was slightly associated with cervical cancer but it is not statistically supported ( Figure 5).
From this sub group meta-analysis, the study done at Yirgalem General Hospital in 2017 had more effect size and weight on the combined effect estimate (19). Its 95% confidence interval did not overlap 1 (the no effect). The 95% confidence intervals of all other the studies overlap 1. But the overall effect estimate 95% CI did not the no effect line That means marital status had statistically significant association with cervical cancer (Figure 6).
Except one study (19), all other studies did not show a statistically significant association with cervical cancer at a study level (22, 46, 49). As a result, the overall effect estimate touches the line of no effect (Figure 7).
The study conducted at Yirgalem General Hospital in 2017 had more effect, high precision and statistically significant (19). But a study conducted in Addis Ababa in 2018 had low precision and statistically significant at a study level (22). Due to the higher weight (35.8%) and more effect was given for a study done at Yirgalem General Hospital, it pulls the overall effect estimate towards to the better event (favours to Negative) (19). Moreover, studies were had higher heterogeneity (I 2 =96%) at (p<0001). The 95%CI of the overall effect estimate did overlap the no effect line. That means there was no statistical significance ( Figure 8).
In this analysis more effect was observed at a study conducted in Jimma University Specialized Hospital in 2015(47) where as more weight (52.7%) was given for a study conducted at Yirgalem General Hospital in,2017 due to its larger simple size (19). Higher heterogeneity was also observed (I 2 =92%) between these studies.
Income was another important variable to predict cervical cancer distribution among women in Ethiopia. The 95% confidence intervals of all the studies did not overlap 1 and the 95% confidence intervals of the overall effect estimate did not overlap 1.
This indicates that there was statistical significance at the study level ( Figure 10).
As the forest plot indicates that the combined point estimate did not touch the null value. Therefore, there was statistically significant effect difference on cervical cancer between women of modern contraceptive users and non-users (Figure11).
All studies except a study done at Family Guidance Association of Ethiopia in Jimma in 2015(50), the 95% confidence intervals did not touch the line of no effect(46, 47, 49 showed that there was statistical significance effect difference between women having a history of STI and women did not have a history of STI ( Figure 13).
This forest plot shows that the highest effect but the least weight (7.4%) was given for a study done in North Ethiopia in 2016 even though it did not show statistical significance(25). The 95% confidence intervals of studies conducted at Debre Markos referral hospital, Addis Ababa and Yirgalem General Hospital did not overlap 1 (19,46,51). But there was statistical significance effect difference at the metaanalysis level ( Figure 14).
This forest plot shows that the 95% confidence intervals of both studies(46, 47) overlap 1 as a result the 95% confidence intervals of the overall effect estimate also overlaps 1 (Figure 16).

Discussion
Cervical cancer is a major problem that threaten the developing countries which in fact the second commonest disease of women in Ethiopia (2) (15) and higher than studies conducted in Egypt (10.4%)(13) and Qatari (8.1%)(11). These differences might be due to socioeconomic and cultural variations.
A Statistically significant heterogeneity between studies was also detected (I2=94.35% (p<0.0001). This might show that studies were inconsistent due to a reason other than chance (Table 4).
There was a variation of evidences reported on the effect of associated factors on cervical cancer like: women's age, education, occupation, residence and age at first sexual intercourse. Some studies revealed that they had a statistically significant association with cervical cancer while others did not show at their study levels ( Figure 6,8,9,10, 16 & 17). This might be due to difference in sample size, variance, methodology, study populations and reliability of the outcome measures at each study level.
The overall effect estimates of women's age (OR=1.43, 95 %CI (0.65, 3.12) was slightly associated with cervical cancer but it is not statistically supported. This indicate that we did not have enough evidence to say women of ≥40 years were 1.34 times to have an effect of getting cervical cancer than women <40 years of age ( Figure 5).
The odd of cervical cancer was 68 % lower among married, widowed and divorced women than single women. It might be that all single women were not protected them from the risk of cervical cancer infection ( Figure 6).
There was no overall effect difference association with cervical cancer between non formal educated and literate women (Figure 7) as women residing Urban and rural ( Figure 9), women had first sexual intercourse at < 15 years and ≥15 years ( Figure   15) and women ever had cervical cancer screening and women did not ever have cervical cancer screening ( Figure 16).
Being unemployed women did not have clear difference effect on cervical cancer as compared to employed women (Figure 8).
Less episodes of cervical cancer were observed among women who did not have an experience of using modern contraceptives than users (Figure11). But more episode of cervical cancer was observed among women having >1 life time sexual partners than women had 1 life time sexual partner ( Figure 12) The better outcome of cervical cancer was observed among women who did not have a history of STI than women having a history of STI ( Figure 13).
Moreover, cervical cancer occurred less frequently among HIV negative women than HIV positive women (OR<1) (Figure 14).

Limitation of the study
This study was not done without limitations. When we describe some limitations, we hope that no one be considered them as inherently bad rather, they are good for readers, researchers and others to understand where new efforts need to be made or researchers may be inspired by these limitations and consider them as the foundation for their future studies. Therefore, some of the limitations of this study are failure to show sub group analysis due to small number of studies were included. Even if our study has strong design and estimable statistics, it may not be able to collect the most important missing data at a study level that might produce limitations in terms of interpreting the findings.

Conclusion:
The pooled prevalence estimate of cervical cancer among women in Ethiopia was high. The pooled prevalence estimate was had a statistical significance association with some variables like income, being HIV positive, previous STI history, more than one numbers of life time sexual partners and prolonged uses of modern contraceptives. There was also a variation of cervical cancer reports across studies.
Reporting of this information in a consistent manner is important for researchers to enhance future studies and also useful for policymakers and practitioners for better understanding of the burden of cervical cancer in Ethiopia for prevention, diagnosis, and treatment of the disease.

Consent for publication
Not applicable

Availability of data and materials
This research data is available at the hand of the corresponding Author.

Competing interests
We declare that we have no competing interests.

Funding
No funding was given to support this study            Forest plot showing the association between occupation and cervical cancer among women in Figure 9 Forest plot showing the association between residence and cervical cancer among women in Figure 10 Forest plot showing the association between income and cervical cancer among women in Et Figure 11 Forest plot showing the association between contraceptive use and cervical cancer among w Figure 12 Forest plot showing the association between contraceptive use and cervical cancer among w Figure 13 Forest plot showing the association between history of STI and cervical cancer among women Figure 14 Forest plot showing the association between HIV Status and cervical cancer among women in Figure 15 Forest plot showing the association between age at first sexual intercourse and cervical canc Figure 16 Forest plot showing the association between ever had cervical cancer screening and cervical