Knowledge of cervical cancer risk factors among Palestinian women: a national cross-sectional study

A national cross-sectional study was conducted between July 2019 and March 2020 in Palestine. The Palestinian Ministry of Health (MoH) hospitals and primary healthcare centers (PHCs) are the main entry sites for healthcare services in Palestine. These are distributed in two main geographical areas: (1) the Gaza Strip and (2) the WBJ. Therefore, governmental general hospitals with a bed capacity of more than 100 and PHCs with level four services (i.e., providing all primary healthcare services) were targeted to recruit participants into the study. Additionally, public spaces in the same governorates of hospitals and PHCs were involved, including markets, downtowns, mosques, churches, parks, malls, and restaurants.

In 2019, the estimated female population in Palestine was 2.45 million with about half of them in the reproductive age between 15 to 49 years [26]. Therefore, adult women aged 18 years or older were the target population and were invited to participate in the study. Potential participants were excluded if they had a citizenship other than Palestinian, were visiting the oncology departments, or were working or studying in a health-related field.

The data collection process took place in 11 hospitals, 12 PHCs as well as 11 public spaces across Palestine. The hospitals had bed capacities of over 100, while the PHCs offered all services to the general Palestinian public. These sites were located across Palestine in different governorates covering a wide geographical area and were chosen for recruitment of participants by stratified convenience sampling.

A translated-into-Arabic version of the validated Cervical Cancer Awareness Measure (CeCAM) was used [9]. The questionnaire consisted of two sections. The first section included socio-demographic questions. The second section comprised 11 questions based on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree) to assess the knowledge of CC risk factors.

The translation and adaptation of the questionnaire were performed based on World Health Organization recommendations [27]. The questionnaire was translated from English to Arabic by two healthcare professionals fluent in both languages and then back-translated into English by another two healthcare professionals who were also fluent in both languages. All healthcare professionals involved in this process had relevant clinical and research experience in gynecology, public health, and survey design.

A few items were adapted from the original CeCAM and were modified in the Arabic version to make them more culturally accepted in Palestine. ‘Having a sexual partner who is not circumcised’ was modified into ‘having a husband who is not circumcised’. Similarly, ‘having a sexual partner with many previous partners’ was modified into ‘having a husband with many previous partners’. In addition, ‘starting to have sex at a young age (before age 17)’ was modified into ‘being married at a young age (before age 17)’.

The Arabic CeCAM was modified for the purposes of this study. To minimize the possibility of participants answering questions at random, the original questions with yes/no/unknown responses were modified into 5-point Likert scale questions. Meanwhile, the participants’ responses were then converted to correct/incorrect responses similar to what was done in previous studies [28‐31].

A pilot study was conducted with 130 respondents to test the clarity of the items of the Arabic CeCAM version. These responses were not included in the final analysis. The Cronbach’s Alpha showed that the questionnaire had an acceptable internal consistency (α = 0.72).

Well-trained data collectors with a medical background conducted face-to-face interviews with the recruited participants for completion of the Arabic CeCAM. Data were collected utilizing the secure, user-friendly data collection tool ‘Kobo Toolbox’ that is accessed via smartphones [32].

Participant characteristics were summarized utilizing descriptive statistics. Continuous non-normally distributed variables were described using the median and interquartile range. Frequencies and percentages were utilized to summarize categorical variables. To reflect the age-associated risk of CC, age was categorized into three groups: 18–20 years, 21–40 years (at-risk group), and ≥ 41 years [9]. A monthly income of 1450 NIS (about $450) was the minimum wage in Palestine at the time of data collection [33]. Therefore, participants were categorized into two categories: ≥ 1450 NIS and < 1450 NIS. Baseline characteristics of participants from the WBJ vs. the Gaza Strip were compared using Pearson's Chi-square test if they were categorical or Kruskal–Wallis test if they were continuous.

For questions asking about CC risk factors, answering with ‘strongly agree’ and ‘agree’ was considered as a correct answer, whereas answering with ‘strongly disagree’, ‘disagree’, or ‘not sure’ was considered as an incorrect answer. Recognizing each CC risk factor was described using frequencies and percentages with comparisons utilizing Pearson's Chi-square test. This was followed by bivariable and multivariable logistic regression analyses. The model of the multivariable analysis adjusted for factors of socioeconomic status including age, educational level, occupation, monthly income, residency, and marital status. In addition, the model adjusted for other factors including having a chronic disease, knowing someone with cancer, and site of data collection. The model was pre-specified based on previous studies [9, 34‐36]. Results of all bivariable logistic regression analyses were provided in Additional file 1.

To evaluate the knowledge level of CC risk factors, a scoring system was used. Similar scoring systems had been adopted in previous studies [24, 28]. For each correctly identified risk factor, the participant was given one point. The total score was then calculated (ranging from 0 to 11) and was categorized into three categories: poor knowledge (0–3), fair knowledge (4–7), and good knowledge (8–11). The knowledge level between the participants from the Gaza Strip and the WBJ was compared using Pearson's Chi-square test. Bivariable and multivariable logistic regression analyses were used to test the association between participants’ characteristics and having a good knowledge level.

Complete case analysis was used to handle missing data (i.e., cases with incomplete data were excluded from the analysis; a total of 135 cases). The missing data were completely random and unrelated to the study variables. Data were analyzed using Stata software version 16.0 (StataCorp, College Station, Texas, United States).

A total of 7223 participants, out of 8086 approached, completed the questionnaire (response rate = 89.3%). The final analysis included 7058 questionnaires (30 did not meet inclusion criteria and 135 had missing values); 4403 from the WBJ and 2655 from the Gaza Strip.

The median age [interquartile range] for all participants was 32.0 years [24.0, 42.0] (Table 1). Participants recruited from the Gaza Strip were younger, getting lower monthly income, and with less chronic diseases than participants recruited from the WBJ.

Only 1670 women (23.7%) had a good knowledge of CC risk factors (Table 2). Women living in the Gaza Strip were more likely than women living in the WBJ to have good knowledge (25.2% vs 22.7%).

On the multivariable analysis, completing secondary or diploma degree, being employed, and having a monthly income of ≥ 1450 NIS were all associated with a decrease in the odds of having good knowledge of CC risk factors (Table 3). On the other hand, knowing someone with cancer was associated with an increase in the odds of having good knowledge.

Among all participants, the most frequently recognized risk factor was ‘having a weakened immune system’ (n = 5458, 77.3%) followed by ‘infection with a sexually transmitted infection (STI)’ (n = 5388, 76.3%) (Table 4). These risk factors were also the most recognized factors in both the Gaza Strip and WBJ. The least recognized risk factors were ‘having many children’ (n = 1597, 22.6%) and ‘being married at a young age’ (n = 2197, 31.1%).

The Chi-square test showed that participants from the Gaza Strip had a higher likelihood than participants from the WBJ to recognize ‘having a weakened immune system’, ‘infection with a sexually transmitted infection (STI)’, ‘infection with HPV’, ‘not going for regular Pap smears’, and ‘having uncircumcised husband’. On the other hand, participants from the WBJ were more likely to recognize ‘having a relative with CC’, ‘having a husband with many previous partners’, ‘being married at a young age’, and ‘having many children’ as risk factors for CC.

On the multivariable analysis, women with age-related risk of CC (aged 21–40 years) were less likely than younger women (aged 18–20 years) to recognize ‘infection with an STI’ (OR = 0.71, 95% CI 0.56–0.91), ‘infection with HPV’ (OR = 0.76, 95% CI 0.61–0.96), and ‘not going to regular Pap smears’ (OR = 0.75, 95% CI 0.61–0.91) as risk factors for CC (Tables 5, 6).

Participants with a bachelor degree had a higher likelihood than illiterate participants to identify ‘having a weakened immune system’ (OR = 2.92, 95% CI 1.94–4.39), ‘infection with an STI’ (OR = 2.40, 95% CI 1.60–3.60), and ‘infection with HPV’ (OR = 1.69, 95% CI 1.14–2.50) as risk factors for CC. However, participants who had a bachelor degree were less likely to identify ‘having a husband with many previous partners’ (OR = 0.56, 95% CI 0.36–0.83) as a CC risk factor.

Married women were more likely than single women to recognize 7 out of 11 CC risk factors. Moreover, participants with a monthly income of ≥ 1450 NIS had a higher likelihood than participants with a lower monthly income to recognize ‘having a weakened immune system’ (OR = 1.22, 95% CI 1.03–1.44) and ‘infection with an STI’ (OR = 1.19, 95% CI 1.01–1.40) as risk factors for CC. Nonetheless, participants earning ≥ 1450 NIS had a lower likelihood to recognize other CC risk factors including ‘not going to regular Pap smears’, ‘having a husband with many previous partners’, ‘being married at a young age’, and ‘having many children’. In addition, employed women were less likely than unemployed or housewives to recognize ‘infection with an STI’ (OR = 0.72, 95% CI 0.61–0.85), ‘infection with HPV’ (OR = 0.80, 95% CI 0.69–0.93), and ‘not going to regular Pap smears’ (OR = 0.87, 95% CI 0.75–0.99).

Women who knew someone with cancer were more likely than women who did not to identify all CC risk factors except ‘having a husband with many previous partners’, ‘being married at a young age’, and ‘having many children’ for which no differences were found.

Participants visiting hospitals were less likely than participants visiting public spaces to identify ‘infection with an STI’, ‘infection with HPV’, ‘having a relative with CC’, and ‘long term use of the contraceptive pill’. However, hospital visitors were more likely to identify ‘having a husband with many previous partners’ (OR = 1.18, 95% CI 1.03–1.34).

Participants visiting PHCs were less likely than participants visiting public spaces to identify ‘infection with an STI’, ‘infection with HPV’, ‘having a husband who is not circumcised’, ‘having a husband with many previous partners’, and ‘being married at a young age’ as risk factor for CC. However, visitors to PHCs were more likely to identify other CC risk factors including ‘having a weakened immune system’, ‘having a relative with CC’, ‘smoking any cigarettes at all’, and ‘not going for regular Pap smears’.

Good awareness of CC, early detection and treatment remain the cornerstones to improve CC survival outcomes especially in low- and middle-income countries [10, 12, 37, 39]. Only 23.7% of participants in this study had a good level of knowledge of CC risk factors, which is similar to reports from Tunisia, Libya, Qatar, and Oman [34, 35, 40, 41]. The relatively lower incidence and mortality rates in these Arab countries might have driven the health authorities to focus on educating women about other types of cancers that have higher rates (e.g., breast cancer) [42]. Education campaigns can be costly, and their funding is usually limited. However, the long-term investment in raising public awareness of CC risk factors may lead to prevention and early diagnosis of CC reducing the financial burden associated with treatment.

Low and colleagues reported better knowledge of CC risk factors among British women who knew someone with cancer, in concordance with this study and other studies in the United Kingdom [43, 44]. A possible explanation could be that women who know someone with cancer are expected to take care and accompany them during healthcare visits. Therefore, these women may come across more experience and knowledge about health-related topics. Furthermore, women’s concerns about someone’s health might lead them to read more about their diagnosis.

Married women were more likely than single women to recognize most of the CC risk factors in this study, which is in concordance with results of other studies [34, 35, 40]. Married women are expected to be more knowledgeable of topics related to reproductive and sexual health through their visits to healthcare facilities and, thus, also have higher chances to access more accurate information from healthcare professionals. In addition, married women may educate themselves by reading printed health materials distributed in clinics or by using internet resources. On the other hand, single women in conservative communities, such as Palestine, may feel inhibited to read or talk about sexual and reproductive health issues. In fact, the data collectors noticed this when they asked single women about risk factors related to sex, such as having a husband with many previous partners, marrying a husband who is not circumcised, and infection with an STI. Furthermore, although some health-related topics are part of the school curriculum in Palestine, topics around sexual health and CC are not included, potentially increasing the barriers of single women to address these topics. Therefore, education interventions should be tailored to address the emotional barriers of single women to promote their willingness to know more about reproductive health topics including CC.

Low socioeconomic status is one of the CC risk factors, raising the importance of improving the awareness of these factors in this group of women for prevention and early detection of CC [45]. In this study, being employed and having a high monthly income were associated with a decrease in the odds of having good knowledge. The decrease in the likelihood of having good knowledge associated with higher monthly income might only reflect the fact that women in the Gaza Strip know more than those in the WBJ but have less income. Furthermore, employed women were less likely to have a good knowledge, which might reflect that more single women might be employed than married women. In addition, those employed women might have less time to read about health-related topics and less involvement in social interactions where women talk about their own and their relatives’ experiences including those health issues. Another contributing factor might be that unemployment is higher in the Gaza Strip compared with the WBJ and this includes women, so that this might be another reflection of the generally better knowledge amongst women from Gaza compared with those from WBJ [46]. In contrast to this, previous studies showed that employed women and those having a high monthly income were more likely to have a good knowledge level of CC risk factors [35, 40, 47].

Higher education level was shown to be associated with more uptake of CC prevention and early detection strategies [48, 49]. Similar to previous studies on cancer awareness in the Gaza Strip, participants with only a secondary or diploma degree in this study showed lower likelihood to have good knowledge of CC risk factors [28‐30]; highlighting the lack of such topics within the Palestinian school curricula. There is a need to revise school curricula to include a wider range of health-related topics. Kyle and colleagues demonstrated that a school-based educational intervention improved the recall and recognition of most of the cancer signs and symptoms even after six months from the intervention [50]. Raising such awareness among adolescents could be useful as this might shape their health-related behaviors in the future.

The participants from the Gaza Strip were more likely than participants from the WBJ to recognize 8 out of 11 CC risk factors. A possible contributing factor could be that living in extended families is more notable in the Gaza Strip. This could increase the likelihood of sharing and discussing health-related experiences or relatives’ stories, which may help in shaping women’s knowledge. Another form of interaction that could play a role in building women’s knowledge is the interaction with healthcare professionals. Women in the WBJ encounter several challenges in accessing healthcare facilities due to the Israeli checkpoints between geographical areas. These checkpoints restrict their movement and impede access to healthcare services [51, 52]. In contrast to this, movement within the Gaza Strip is easy and unrestricted for women, so that most women in the Gaza Strip can access healthcare facilities easily and shape their knowledge while communicating with healthcare providers [53, 54]. Moreover, women in the Gaza Strip have a relatively higher fertility rate among women of childbearing age (15–49 years) compared with those in the WBJ (4.5 vs 3.7 births per woman), therefore, they may be exposed to more experience in sexual and reproductive healthcare and associated health education [55].

In this study, women recognized ‘infection with an STI’ more than ‘infection with HPV’. This is similar to findings among Libyan and British women [35, 43], which suggests that women are more aware of the link between CC and STIs than causative micro-organisms (e.g., HPV) as reported in the literature [35, 43, 56]. Future educational campaigns should highlight the role of HPV in CC etiology.

Having five or more children was the least risk factor reported in this study. This is similar to findings of other studies conducted in Libya, United States, India, Oman, and Malaysia [35, 47, 57‐59]. A possible explanation for this could be that women’s thoughts of CC risk factors are shaped by the culture of the country where they were raised. Palestinian culture encourages having many children as a source of kinship and wealth. Therefore, this might have prevented Palestinian women to consider negative associations with multiparty, such as it being a risk factor of CC. Moreover, more than 70.0% of study participants were married and, considering the high fertility rate in Palestine, which might also have had contributed to shaping such beliefs about multiparity [55]. Education interventions should focus on ‘having many children’ as a risk factor of CC since this is very relevant to the Palestinian society.

The findings of this study reflect the need to promote educational programs to improve women’s knowledge of CC in Palestine. Enriching school curricula with health-related topics and targeting women in the reproductive age should be prioritized. This could drive these women to adjust their behavioral risk factors, hence, decrease their chance of developing CC. In addition, raising young women’s awareness of CC may make them more confident to talk about any possible CC symptom and less embarrassed to seek medical advice or discuss their concerns with healthcare professionals.

The main strengths of this study included the large sample size and the high response rate. In addition, the stratified approach that may provide a representative view of the target population’s knowledge on different levels of the Palestinian community.

Limitations of this study included the convenience sampling that may limit the generalizability of the findings. However, this may be alleviated by the recruiting a large number of participants while having a high response rate and covering different geographical areas in Palestine. Another limitation could be the exclusion of visitors or patients in the oncology departments and participants with medical backgrounds, possibly decreasing the number of participants with a presumably good level of knowledge. On the other hand, their exclusion was meant to increase the relevancy of this study as a measure of knowledge among the public.