Cervical cancer screening utilization and its predictors among women in bench Sheko Zone, Southwest Ethiopia: using health belief model

A community based cross-sectional study design was employed from February 2021 to April 2021.

The study was conducted in Bench Sheko Zone, which is found in the Southern Nation Nationality and Peoples’ Region (SNNPR) State, Ethiopia. Bench Sheko is bordered on the west by South Sudan, on the north by Sheka Zone, on the northwest by Keffa Zone, and on the east by west Omo Zone. The administrative center of Bench Sheko Zone is Mizan Teferi, which is located 561 km far from Addis Ababa. The zone has a total of two administrative towns and six districts namely; Debub Bench, Guraferda, Semen Bench, Shey Bench, Gide Bench, Sheko, Cize town administration, and Mizan Aman town administration with a total population of 609,588 (2011 Zonal Report). In Bench Sheko Zone, there are 1 teaching hospital, 1 primary hospital, 25 health centers, and 130 health posts with different health services. Currently, only the Mizan Tepi University Teaching hospital and Mizan health center give cervical cancer screening service.

According to Ethiopian cervical cancer prevention and control guideline [37], all women aged between 30–49 years in Bench Sheko Zone were source population. Hence, women aged between 30–49 years and who lived in selected kebeles during the study period were the study population. Furthermore, women in the age range of 30–49 years and who lived for at least 6 months in the study area were included in the study, whereas, those who were critically ill during the data collection period were excluded.

The sample size was calculated using a single population proportion formula of \(n=({z\frac{a}{2})}^{2}\frac{p(1-p)}{{d}^{2}}\), where n = the sample size, d = degree of precision = 0.04, Z α/2 = 1.96 by assuming 95% confidence interval and the population proportion of 0.155 (15.5%), as it was used for the study conducted in Jimma town, Southwest Ethiopia [22]. Then, by adding a 10% non-response rate and design effect of 2, the final sample size was 690.

For this particular study, a multi-stage stratified sampling technique was used to select the study participants. The primary sampling units, the 3 districts; Shey Bench, Gidi Bench, and Debub Bench, and 1 administrative town, Mizan Aman were selected by simple random sampling from the total of 6 districts and two administrative towns in Bench Sheko Zone, respectively.

The secondary sampling units, the 21 kebeles (the smallest administrative unit within the district) were randomly selected from each stratum (urban and rural). Then, the sample size was distributed to each selected kebeles by population proportion to size (PPS) formula. Then, the tertiary sampling units, the households were selected by using systematic random sampling. The sampling interval (k^th interval) of six was used to select the eligible participants using the household lists which were determined after the survey. Then, the next household was selected systematically with a random starting point which was selected from the first 6 households, and the next eligible woman in the household was interviewed every 6^th household. For households with more than one eligible woman, only one study participant was selected using the lottery method. Revisit of three times was made when the eligible respondent was not available at the time of the data collection [Fig. 1].

A structured interviewer administered questionnaire was used to collect the data which was adapted from different reviewed literature [22, 24, 25, 38, 46‐48]. The questionnaire was translated from English to the national language Amharic and to assure its consistency, it was translated back to English by another linguistic professional. The data were collected by eight trained BSc nurses and supervised by four public health professionals. Before doing the actual data collection, three days of training was given regarding the objective of the study on the data collection tool, interviewing approach, and privacy and confidentiality. The questionnaire was also pretested among 5% of the total study participants in Sheka Zone.

The quality of the data was assured by providing three days training for data collectors and using of pretested questionnaire. Moreover, close supervision of the data collection process was done and the questionnaire was checked for consistency, clarity, and completeness.

After data collection was completed, the data was entered into Epi-data manager version 4.0.2.101 and exported to Statistical Package for the Social Sciences (SPSS) version 22.0 for analysis. Descriptive statistics like frequency, percentage, and mean with standard deviation were used. Bivariate logistic regression analysis was performed to select variables for multivariate analysis with a 95% confidence interval and a p-value less than 0.05. Finally, multivariate logistic regression analysis was performed and a p-value of less than 0.05 was considered as strong predictor variables for cervical cancer screening utilization.

From the total of 690 eligible women, 678 women have participated in the study yielding a response rate of 98.2%. The mean ± SD age of the participants was 36.01 (SD ± 5.11) years. More than half, 415 (61.2%) of the participants were urban residents, and 297 (43.8%) were protestant religion followers. From the total participants, more than three-fourth, 527(77.7%) of them were currently married. Concerning the respondents’ educational status, 188 (27.7%) of them were unable to read and write. Of all participants, 265(39.1%) were housewives, 194(28.6%) had partners who able to read and write, and 195(28.8%) had partners whose occupations were merchants [Table 1].

From the total participants, more than two-third, 470(69.3%) of them had the first sexual intercourse before the age of eighteen, whereas, majority or 635(93.7%) of them gave their first child after the age of eighteen with the mean ± SD of 21.8(SD ± 2.6) years. Of all the study participants, about 155(22.9%), 98(14.5%), and 81(11.9) had a history of HIV test, and a history of sexual transmitted diseases (STD), and smoking history, respectively. Concerning contraceptive history, about 427(63.0) of the participants were used modern contraceptives [Table 2].

Ninety-six (14.2%) (95% CI: 11.5, 16.8) of the participants have ever utilized cervical cancer screening at least once in the last five years. Among the total participants who screened for cervical cancer, about 56 (8.3%), 25 (3.7%), and 15(2.2%) of them were screened due to the offer by the health professionals, self-initiated, and know someone screened, respectively. The most common reason for the participants not to utilize cervical cancer screening was feeling healthy, 269 (39.7%) [Fig. 2].

Using bivariate logistic regression, different factors like; the age of participants, partner educational status, age at first sex, ever use of alcohol, smoking history, overall knowledge, overall attitude, perceived susceptibility, perceived severity, and perceived benefit were found to be significantly associated with cervical cancer screening utilization.

After adjusting for possible confounding factors using multivariate logistic regression model; the age of participants, partner educational status, age at first sex, ever use of alcohol, overall knowledge, overall attitude, and perceived benefit were found to be strong predictors of cervical cancer screening utilization.

Participants with the age range of 40–49 years were five times more likely to use cervical cancer screening than those whose ages were 30–39 years (AOR = 5.35, 95% CI = [2.89, 9.90]). Participants with partner educational status of the certificate and above were four times more likely to use cervical cancer screening than those with partner educational status of unable to read and write (AOR = 4.36, 95% CI = [1.65, 11.51]). Those who had first sexual intercourse before eighteen years were almost five times more likely to use cervical cancer screening than those who had first sexual intercourse after eighteen (AOR = 4.85, 95% CI = [2.29, 10.26]). Participants who had ever used alcohol were almost four times more likely to use cervical cancer screening than their counterparts (AOR = 3.99, 95% CI = [1.23, 12.89]). Knowledgeable participants were almost nine times more likely to use cervical cancer screening than their counterparts (AOR = 8.98, 95% CI = [4.06, 19.89]). Those who had a favorable attitude were almost four times more likely to use cervical cancer screening than their counterparts (AOR = 3.56, 95% CI = [1.78, 7.09]). Having perceived benefit of cervical cancer screening utilization were almost four times more likely to use cervical cancer screening than their counterparts (AOR = 2.94, 95% CI = [1.48, 5.84]) [Table 3].