Determinants and levels of cervical Cancer screening uptake among women of reproductive age in South Africa: evidence from South Africa Demographic and health survey data, 2016

Cervical cancer is a serious public health problem and is one of the second leading causes of cancer-related mortality among women worldwide. Its epidemiology and health impacts are not only affecting women, but also their families, communities and social institutions [1]. Although it is also one of the most preventable disease through proper screening, treatment and follow-up, more than half a million women worldwide develop intrusive cervical cancer each year [2]. Middle- and low-income countries bear an unequal share (83%) of the global cervical cancer burden, but only achieve an average successful screening coverage of 19%, compared to 63% in high-income countries [3]. Studies [4‐6] have accounted for over 275,000 female deaths and approximately 529,000 new diagnoses each year; besides, cervical cancer disease burden is more prevalent in older women who are post-childbearing [7‐12]. In the developing countries, lack of resources limits coverage of cervical screening for women [13‐15], and notably, for those from rural communities compared with urban areas, as the rural population is extensively poor and their access to health services is more difficult [16‐22]. Overall, developed countries such as the United States of America (USA) and United Kingdom, 40 and 42% of the women diagnosed with cervical cancer respectively, die from it [7, 10, 11, 22, 23], while in Africa and South Asia, the equivalent death rates are nearly twice (78%) as high as the rates experienced in high-income countries [23, 24].

In sub-Saharan Africa, cervical cancer encompasses 20–25% of all cancers among women. The estimate is double that of women globally, as its incidence ranges from 30 to 40 per 100,000 women [24]. In South Africa, cervical cancer is the cancer with the most incidents and the leading cause of cancer mortality among women [25]. Contemporary estimates have shown that every year in South Africa, 5743 women are diagnosed with cervical cancer and 3027 die from the disease. Also, it is the second most frequent cancer among women of reproductive age between 15 and 44 years of age, after breast cancer. Studies [26‐28] have shown that young women are at higher risk as they tend to be sexually active and have higher numbers of sexual partners. Several studies [8, 26, 27] have reported that young women are poorly informed about cervical cancer with its associated risk factors, and are unclear about the intent of cervical cancer screening, as well as holding on to negative or inaccurate beliefs or attitudes to Pap testing. Furthermore, one of the known main causative agents in cervical cancer is HPV and there are over 200 recognized serotypes of the HPV virus. Existing studies [29, 30] have documented that about 21.0% of women in the general population of South Africa are estimated to harbor cervical HPV infection in South Africa. These reports have shown that 62.8% of invasive cervical cancers among women in South Africa are attributed to HPVs 16 or 18 [29], which fall in the high-risk HPV serotypes classification, and which are responsible for almost 70% of cervical cancer cases.

Other factors for increasing young women’s susceptibility to cervical dysplasia (dysplasia is defined as medically unusual growth or unusual development or growth of a part of the body such as an organ, bone or cell, including the total absence of such a part) include smoking, oral contraceptive use, and vulnerability of the adolescent cervix to sexually transmitted infections [31‐33]. As a consequence of the sexually transmitted nature of HPV, early exposure to sexual intercourse and multiple sex partners are significant risk factors for cervical cancer [5, 34]. Studies have reported that 80–90% of women will have this sexually transmitted infection at some point in their life, although only 3–4% of them will develop cervical cancer [35, 36]. Earlier studies [29, 37] have indicated that the South African Cancer Association (SACA) report in the year 2006 revealed that the age-standardized incidence rate for cervical cancer was 24.71 per 100,000 population. As a result of SACA reports, the South African Department of Health developed the Cervical Cancer Screening Programme, which allows three Pap smears per lifetime, at 10-year intervals, starting at the age of 30. This screening policy programme was designed for target coverage of at least 70% of women nationally [37].

Even with the implementation of the Cervical Cancer Screening Policy Programme in South Africa, the screening coverage is 20%, nationally low for women over the age of 30 years [38]. A population-based study conducted among rural South African women reported that only 18% of the women had ever had a Pap smear test [39]. Another study conducted among female university students found that 42.9% of the participants had heard of cervical cancer, but only 9.8% of the participants had ever had a Pap smear test [40]. Another study conducted in Vhembe District of South Africa, reported that campaigns and motivational talks on cervical cancer as well as cervical cancer screening services are provided for women in all the clinics in the district [41], yet very few women voluntarily present themselves for cervical cancer screening services.

Moreover, women who utilized the cervical cancer screening services do so because they have been referred by other health facilities, or the women are presenting with symptoms. Factors that are associated with poor uptake of cervical cancer screening services are pain, stigmatization, and fear of embarrassment [42‐44], time constraint [25, 45], the related costs [25, 42, 45], insolence of health workers [46], lack of awareness about where to get screening [25, 47] and delays in hospital cancer screening [25]. Therefore, cervical cancer becomes a public health burden in countries where there are no cervical cancer screening services, or poor utilization of screening services. Cervical cancer screening services could be utilized better if awareness campaigns are sustained and the services are accessible, affordable and available [48]. Early screening is proven to be cost-effective and a form of control strategy of the disease burden [6, 33]. Thus, early screening of cervical cancer is an important preventative strategy for the disease burden. Improving screening services will not be sufficient to result in increased screening uptake among women, unless we understand and address the multidimensional causes that are likely to impel women’s disposition towards cervical cancer screening. Very little is known about determinants and levels of cervical cancer uptake among women of reproductive age in South Africa. No previous study conducted in South Africa has investigated the determinants and levels of cervical cancer uptake among women of reproductive age (15–49 years) in South Africa. The objectives of this study were to examine the association between determinants and cervical cancer uptake at the individual- and community-factor levels using a logistic statistical model, and to assess the extent of the variation in the uptake levels of cervical cancer screening services in South Africa. The findings of this study will provide insight into the provision of appropriate educational interventions for disease risk reduction and effective cervical cancer screening uptake among women of reproductive age in South Africa.

Table 1 shows the socio-demographic characteristics of women of reproductive age with a total of 5903 who participated in the study. From the table below, 35.5% of women indicated that they have had a Pap smear test, and 64.5% of them reported that they have not had a Pap smear test. The mean age at cervical cancer screening among women in South Africa was 40.8 years (SD 18.6, range 15–95 years). Most women were 45⁺ years (39.0%) followed by 15–24 years (23.6%) who participated in the survey. A majority of the women were from the African/Black population group (84.9%), with the Indian/Asian population group having the least (1.3%). Out of all the women interviewed, 15.9% of the respondents were from KwaZulu-Natal province, followed by 14.1% from Limpopo province. More women were from the urban areas (54.6%) than rural areas (45.5%) (Table 1).

The majority of the respondents had secondary education (64.0%) and 17.3% of them have primary education. Among the women interviewed, 69.9% of the women are not working while 17.3% work in the professional/formal sector and 12.8% work in the non-professional/informal sector. Three thousand six hundred and sixty-seven (62.1%) of them have never been married; most (41.2 and 36.5%) were in the poor and rich wealth index categories respectively (Table 1). The majority (86.5%) of the women do not have health insurance cover and 40.4% perceived themselves to have good health status.

Table 2 presents the results of the bivariate analysis of the individual-level characteristics and Pap smear testing. In the bivariate analysis, all individual-level variables (age group, population group, province, geographical type, educational attainment, occupation, marital status, wealth index, health insurance cover and perceived health status) were significantly associated with women’s responses to cervical cancer screening (Pap smear test). Regarding the age group showing the most significant association with Pap smear testing (χ2 = 724.93, p = 0.000), a majority of women (52.5%) in age cohorts of 35–44 years have undertaken a Pap smear test compared to those in other age groups.

A large proportion of white women (76.5%) have undertaken a Pap smear test compared to women in the other population groups, and this shows a significant association between population group and Pap smear uptake (χ2 = 306.76, p = 0.000), while educational attainment was found to be significantly associated with Pap smear testing (χ2 = 306.7635; p = 0.000), and 43.2% of urban women have had a Pap smear test compared to their rural counterparts, at only 26.31%.

It can be deduced from the findings that urban women have higher chances of undertaking Pap smear testing, as geographical type was significantly associated with Pap smear testing (χ2 = 183.80; p = 0.000) (Table 2). Figure 1 depicts the proportion of women of reproductive age who have had Pap smear by province. Western Cape Province had the highest proportion (66.52%) of women who have had a Pap smear, while Limpopo province had the lowest proportion of women who have had a Pap smear (Fig. 1).

However, a majority of women (39.95%) who perceived their health status to be poor had taken a Pap smear test compared to those with perceived excellent health status (27.94%) (Fig. 2).

Table 3 presented the multivariate analysis illustrating both the unadjusted (U) and adjusted (A) Odds Ratios (OR) of population characteristics which were found to be significant predictors of Pap smear uptake among women of reproductive age. Women aged 35–44 years were found to have 14 and 12% (UOR: 12.14: 95% CI = 9.36–15.72; AOR = 13.62: 95% CI = 10.56–17.55) higher odds of having a Pap smear compared to those aged 15–24 years, and this was found to be statistically significant. Population group has a significant association with Pap smear uptake as the odds of having a Pap smear among white respondents were 2.2 and 2.5 times (UOR; 2.21, 95% CI; 1.54–3.15; AOR; 2.47, 95% CI; 1.75–3.50) higher as compared to African/Black women. Province was significantly associated with lesser odds of having Pap smear uptake: Gauteng (UOR: 0.24, 95% CI: 0.17–0.34; AOR: 0.24, 95% CI = 0.17–0.33); Limpopo (UOR; 0.30, 95% CI; 0.21–0.43; AOR; 0.30, 95% CI; 0.21–0.42); KwaZulu-Natal (UOR; 0.30, 95% CI; 0.22–0.42; AOR; 0.30, 95% CI; 0.21–0.41).

Also, educational attainment has a significant association with Pap smear uptake, as the odds of being screened for cervical cancer among women having secondary and higher education were 2.7 and 4.4 times higher as compared to women who have no education (Table 3). Women who had health insurance cover had 13% higher odds of having Pap smear uptake compared to those who had no health insurance cover, and this was found to be statistically significant (UOR; 1.36, 95% CI; 1.11–1.68). For occupational status, women who are in the professional/formal sector had 14% higher odds of taking Pap smear test compared to those who are not working (AOR; 1.38, 95% CI; 1.14–1.69).

This study examined in an all-inclusive manner the background determinants of the use of cervical cancer screening among women of reproductive age in South Africa, with particular emphasis on individual proportion of geographical type demerit (urban and rural), occupation disadvantage (not working, professional and non-professional), wealth index drawback (poor, middle and rich), health insurance cover (yes and no), and educational attainment disadvantage (no education, secondary and higher). The study found that unmeasured irregularity in cervical cancer screening routine across age and population groups was significant through peripheral factors, as population characteristics are explained by significant differences. Such significant differences have been observed in several studies, where population characteristics have accounted for specific background unpredictability factors that determine cervical cancer screening utilization among women of reproductive age [6, 52, 53].

Generally, all the background factors for the study were found to be significantly associated with cervical cancer screening uptake at the individual level, a lattice of other factors (Model II). This observation is not unusual when all individual-level variables are considered simultaneously [8, 54, 55]. This is a key finding showing the significance of the multilevel models in examining classified structured datasets of the 2016 SADHS. Further, access to health facilities have been established to be an important determinant of cervical cancer screening among women [28, 32, 56], as population characteristics seem to be more significant in health matters. The identified significant individual-level factors were age group, population group, province, geographical type, educational attainment, marital status, health insurance, wealth index, occupation, and perceived health status. The consequences of these factors in cervical cancer screening uptake are well established in studies conducted in African countries [5, 57].

Our study contributes to the understanding of determinants and levels associated with cervical cancer screening uptake in South Africa, where the cervical cancer screening prevalence remains low [58, 59]. To the best of the authors’ knowledge, this is the first study that assesses population factors associated with cervical cancer screening uptake in South Africa using the 2016 SADHS. Our findings reveal that a significant number of women (64.46%) did not have an uptake of cancer screening; while the small fraction of the female population who had had a Pap smear were majorly found in Limpopo (25.96%) and KwaZulu-Natal (28.27%) provinces. Thus, lack of knowledge, younger age, lack of income, apprehension of Pap smear testing, and poor accessibility of health facilities’ screening services were significantly associated with low cervical cancer screening outcomes [4, 33, 60]. Consequently, in most industrialized countries, cervical cancer screening programmes have shifted from clinician-sampling to self-sampling for HPV testing, which has been proven to be equally accurate in combination with a follow-up Pap smear test. This implies that various health stakeholders can incorporate self-sampling HPV DNA testing into cervical cancer screening health educational programmes, especially at grassroots’ levels. Present studies have recommended self-sampling HPV DNA testing to be feasible, and may significantly improve cervical cancer screening uptake in South Africa [61‐63].

Our study revealed that the prevalence of women who had Pap smear testing was higher among women with poor perceived health status, and also among those residing in Western Cape and North West provinces, respectively. A possible explanation for the provincial variation observed is that the urban settlements characteristically have higher socioeconomic status, less cultural conservatism, and easier access to health care services [64‐66]. Corroborating the results of similar studies, our results demonstrate that women’s occupation and wealth index were positively associated with cervical cancer screening [5, 31, 57]. The lower outcome of cervical cancer screening among women of reproductive age may indicate a financial burden, which is a barrier to accessing cervical cancer screening services. Women with higher occupational status were more likely to undergo screening because these groups of women are most likely to own health insurance cover [1, 67]. Our study determined that individual-level factors such as women’s wealth index had a positive influence on cervical cancer screening behaviour and attitude, indicating that gender disparity, as assessed through wealth index, can affect screening services uptake.

Studies have indicated that women do not accumulate properties converted as wealth as much as men, resulting in a gender wealth gap [68, 69], as women are often perceived as passive and powerless; in addition, societies have apportioned wealth and assets ownership as mostly in the male domain. Besides, several myths and misinformation related to women that have undergone cervical cancer screening have been accepted generally [34], and consequently have various health implications among such women who will tend towards denial of life-threatening health problems [20, 21]. Empowering women with the right knowledge of the benefits of Pap smear testing is necessary and key to having prior information about health matters, especially with terminal diseases. Women who resided in urban settlements, comprising a higher proportion of women with secondary and higher educational attainment, were more likely to undergo screening. This finding is consistent with the findings of previous studies, suggesting that communities with a high concentration of educated women can increase the utilization of health care facilities, including cervical cancer screening services [70, 71]. Education is frequently associated with increased access to health care services and improved knowledge regarding health matters and behaviours. Increasing the proportion of educated women may facilitate the dissemination of knowledge and awareness to those with lower education, aiding them in accessing health services through informal social networks and contacts within their community space [18, 69].

Supporting our study objectives and previous findings of other studies [33, 59, 72], we found a positive association between the white female population group and cervical cancer screening. The possible explanation for this unexpected result is that the white population group may have higher expectations for health services and accessibility to health facilities. Other structural reasons may be gaining full access to transportation and finance to access cervical cancer screening services, which can influence screening behaviour among the white population group [33, 60]. Urban women were seen to have undertaken more Pap smear testing than rural women, as geographical type was significantly associated with Pap smear testing. It could be inferred that women in urban settlements may have access to sensitization programmes on Pap smear testing, health facilities and health insurance cover. Women in Western Cape Province were found to have undergone more Pap smear tests than women from Limpopo province. The outlying areas in South Africa play a role in sensitization of women on the benefits of cervical cancer, as these provinces are stratified as urban and rural settlements due to infrastructural facilities put in place in these areas.

Further research should include these potential factors associated with cervical cancer screening in the study design and analysis. As anticipated, health insurance coverage was strongly associated with cervical cancer screening outcomes, as our findings regarding the association between health insurance cover and Pap smear testing are consistent with the outcomes of previous studies [58, 59]. Our study findings prove that the adoption of a universal health insurance scheme ensuring equity in access to health care can largely enhance the possibility of cervical cancer screening use [6]. The cost of a Pap smear test may be a major impediment in cervical cancer screening uptake among women with meagre earnings, as this may further explain the low turn-out for cervical cancer screening among rural women [52, 59]. In a geographical area where poverty is high, alternative health services are given greater priority to out-of-pocket payments than preventive services [53, 59], and consequently, health insurance cover may potentially reduce the financial burden for rural women to have access to preventive health care services, including Pap smear testing.

In line with other studies, this study has shown a positive relationship between perceived health status and uptake of Pap smear testing. Women of reproductive age with consistent upkeep and maintenance may have regular dealings with health facilities and health care workers, including going for a Pap smear test when necessary [1, 59]. Specifically, in constrained settings, interaction with community health personnel after visitation to health facilities may increase women’s exposure to undertake preventive measures and encourage adoption of accurate health information on the benefits of cervical cancer screening [4]. Preceding studies have also validated that having a prescription from a doctor and treatment medic’s prescription and medicament was constantly found to be a strong predictor of adherence to cervical cancer screening [6, 53]. Our study has some prospective limitations. First, the cross-sectional study design restricted our capacity to draw underlying deductions for the relationship of individual-level factors with uptake of a Pap smear test, which require a longitudinal designs, but could not be determined. In addition, due to limited number of variables collected by the 2016 SADHS, we could not examine a full array of factors related to cervical cancer screening, particularly cultural factors including the quality of the health service and other factors related to user-friendliness of cancer screening facilities. The study also suggests the use of ethnographic methods that may unravel community factors that may influence the outcomes of cervical cancer screening.

Most women in South Africa demonstrated a low uptake of cervical cancer screening, as utilization of Pap smear testing among women of reproductive age in the study area was 35.5%. Despite that, when compared to previous studies conducted in countries within East Africa (Uganda – 20.6%; Kenya – 12.3%), West Africa (Nigeria – 13.5%; Benin – 0.6%) and Southern Africa (Namibia – 23.6%; Zimbabwe – 17.0%), there is still a high coverage of Pap smear testing among women of reproductive age in South Africa [73‐77]. As implied in the literature, cervical cancer screening is associated with an increase in demographic factors that will influence the utilization of the services [33, 60]. Although in this study, demographic factors (such as age, educational attainment, health insurance cover, occupation, geographical type, and perceived health status) were significantly associated with utilization of cervical cancer screening services, yet in reality cervical cancer screening uptake is relatively low. Thus, these factors that result in cervical cancer screening services usage vary between individuals and from community to community [8, 54, 55]. The most basic conclusion is that there is a significant variation that exists in the population group and provinces in the use of cervical cancer screening services among women in South Africa.

The selected broad range of individual-level factors was not able to explain contextual variation as there is much more variation in cervical cancer screening among factors related to women’s utilization of cervical cancer screening services. Policies that look at women’s individual characteristics, such as promoting women employment, especially in professional occupations, higher educational level attainment and wealth accumulation, should continue to be implemented in order to increase Pap smear test uptake and achieve the universal health coverage goal. To mitigate those problems, there should be consistent monitoring of the cervical cancer detection and preventive actions among women through sensitization of health education, to modify women’s behaviours and attitude towards adopting cervical cancer screening. Health professionals and community health workers should promote educational and health recreational programmes provided by local councilors for individuals in their respective communities to have a prior knowledge of the dangers of not undergoing Pap smear testing.

Since early case detection through screening is the most cost effective activity for reducing the disease burden, reproductive health workers and policy stakeholders are needed to demonstrate more commitment in creating awareness about cervical cancer. Stakeholders in the Ministry of Health are needed to design and make Pap smear tests free for women through the establishment of more screening centres in different geographical areas of the provinces in South Africa. The existing Pap smear testing programmes mostly provided by non-governmental as well as faith-based organizations, which are majorly located in urban settlements, need to be decentralized and harmonized for greater efficiency in South Africa. Also, there is need to integrate the cervical cancer screening exercise into the mainstream health care services in the medical institutions. Women who are at least 15 years of age, particularly those with a family history of cervical cancer, must be encouraged to opt for cervical cancer screening at every available opportunity. Lastly, there is need to increase the number of medical professionals and community health workers with the requisite skills to conduct cervical cancer screening in South Africa.