Abstract

Identifying the factors that deter or stimulate the women to participate in screening activities is very important in order to design effective education and motivation strategies, particularly in the countries without an organized system. The study employed a case-control design. The participants were recruited in four primary health care institutions in Belgrade over a month. The study group comprised all women aged 18–70 years, who demonstrated an initiative for a PAP- smear. The controls were women with no Pap smears within the last 4 years, matched by age (±2 years), education and marital status with the study group participants. The study instrument was the 62-item self-administered questionnaire. According to multivariate analysis, adherence to cervical cancer screening practices is significantly related to better financial status [odds ratio (OR) = 10.8, P = 0.001], no gender preference for a gynecologist (OR = 3.1, P = 0.015), consultations with a gynecologist (OR = 4.7, P = 0.029), conversation with the women with cervical cancer about that disease (OR = 2.8, P = 0.029) and higher media exposure to information about cervical cancer prevention (OR = 5.0, P = 0.004). Open communication, social networks and improving social–economic status of women in our society are the most prominent factors, most of which are mainly outside the health services’ domain and require multisectoral collaboration to improve women’s reproductive health.

Introduction

Cervical cancer screening is a lifesaving preventive strategy. Different correlation studies of cervical cancer trends in countries in North America and Europe demonstrate dramatic reductions in incidence of invasive cervical cancer and a 20–60% reduction in cervical cancer mortality due to organized preventive programs at the national level [1, 2]. Cervical cancer is the second most common female malignancy in Serbia after breast cancer, with very high age-standardized incidence and mortality rate (27.2/100.000 and 7.2/100.000 women) [3]. Until now, the response of Serbian health authorities at the national or local government level to this issue has not been adequate and systematic. Despite the fact that the first National Organized Screening Program was finalized in 2007 and approved by the Serbian Government in May 2008, the program faces a number of financial and organizational constrains that limit its implementation. Apart from the partial initiation of the organized screening activities (in a northern region of Serbia-Vojvodina and partly in the capital city of Serbia-Belgrade), in general, Serbia still employs opportunistic screening for cervical cancer.

Opportunistic screening for cervical cancer is provided free of charge in community health centers. Pap smears are performed only by gynecologists and women do not require a referral from a general practitioner. Keeping in mind all shortcomings of the opportunistic model (high coverage in younger and low coverage in middle-aged and older women), the attempts of health services to deal with the public health problem of cervical cancer in Serbia have a limited effect on morbidity and mortality. The challenges in starting the organized screening program for cervical cancer are common for countries in the process of transition to market economy and for the new member states of the European Union [4, 5].

In the typology of different health-related behaviors, preventive health behavior is defined as ‘any activity undertaken by an individual who believes himself to be healthy for the purpose of preventing or detecting illness in an asymptomatic state’ [6]. Analyzing screening behavior is equally important both for the organized or opportunistic screening models in order to develop new or improve existing programmes. Screening behavior appears to vary according to many different personal and sociocultural factors, including demographic factors (age, level of education, socioeconomic status, residence and marital status) [7], knowledge of and attitudes toward a Pap smear and awareness about risk factors for cervical cancer [8], emotional factors (modesty, fear and embarrassment in relation to gynecologic presentation) [9] and perceptions of personal risk and benefits of screening [10]. Furthermore, apart from personal factors, there are number of social factors that are also significant: an overall influence of the social network surrounding the woman, media exposure to information on cervical cancer prevention and especially, advice from the GP or gynecologist [11]. In addition, the women are influenced by the organizational context of health care, the conditions of the particular health insurance model and the availability of the organized screening [12].

Until now, little information has been published about cancer screening behaviors and programs in countries from Central and South-Eastern Europe. Most published accounts of screening behaviors and programs come from countries with organized cervical screening at the national level (UK, Canada, Finland and Australia). However, even the best-organized programs are frequently faced with poor screening compliance. Identifying and targeting the factors that deter or stimulate the women to participate in screening activities are even more important in the countries without an organized system. The challenges of establishing the screening programmes are particularly noticeable in low-income countries, and the examples from India or Africa highlight the urgent need for public health education along with the social empowerment of women in their societies [13, 14].

The objective of this study was to develop a model of cervical cancer screening behavior of Serbian women in order to design effective educational and motivational strategies. Accordingly, the most important questions addressed in this study are: who are the women who regularly present for the Pap smear, despite the lack of an organized program and what are the determinants of their preventive behavior?

Methods

Setting and study population

Assessment of women's behavior models concerning cancer screening practices was conducted during the period of 1 month (February 2008 to March 2008) in four public primary health care institutions in the capital city of Serbia, Belgrade. The institutions selected for participation in this study served both urban and suburban populations. The study employed a case-control design.

Our study group comprised women who presented in the gynecological departments of selected public primary health care institutions and demonstrated preventive behavior in relation to cervical cancer. All women aged 18–70 years, who demonstrated an initiative to visit the gynecologists with request for a Pap smear without any symptoms and gynecological complaints, comprised the study group. Continuity in previous gynecological checkups was an additional inclusion criteria (at least two Pap smears in the previous 4 years), which was checked through the patient medical documentation. Women who had undergone hysterectomy, young women who visited the gynecologist for the first time, women who requested Pap smears for the first time and women who refused participation were not included in the study group. The control group comprised women who were patients in the department of general practice during the study period and matched with women in the study group by age (±2 years), education and marital status. Majority of control group participants visited GP due to seasonal acute respiratory infections. The inclusion criterion for the control group was the absence of Pap smears within the last 4 years. The introduction letter was provided to all study participants explaining the purpose of the study and the anonymous nature of their participation. All women provided written informed consent to take part in the research, allowing the researchers to audit medical records. This study was previously approved by the Ethics Committee of the Belgrade School of Medicine.

Data collection

The study instrument was a 62-item self-administered questionnaire, designed and validated during and after the qualitative explorative study of cervical screening behavior of the Serbian female population [15, 16]. The authors conducted the qualitative study in 2003 and 2004, which was based on nine focus group discussions with 62 women from diverse socioeconomic backgrounds and health status recruited in two cities with contrasting social settings, Belgrade and a regional town, Smederevo. The study explored women's knowledge about early detection of cervical cancer, the impact of individual sociodemographic and social factors on women's screening behavior and the perceived barriers to cervical cancer screening. Thematic analysis identified the most salient barriers to screening: poor knowledge about cervical cancer screening, lack of patient-friendly health services, poor communication between women and gynecologists, inadequate counseling, sociocultural health beliefs, gender roles and personal difficulties. These findings were the basis for designing the questionnaire in the present study. The questionnaire collected demographic information and data regarding women’ knowledge, attitudes and practices in relation to cervical cancer and screening. All study participants filled in the questionnaire during their visit to the health institution. Information collected in the questionnaire was supplemented and confirmed through a review of the medical records.

We assessed the following sociodemographic variables: age, formal education, marital status, employment status, occupation, home ownership status and self-reported financial status. The variables addressed practices related to gynecological checkups, particularly Pap smears, and the regularity of screening and participants’ attitudes to compliance with medical recommendations. One variable that was only for women in the control group asked about the reason for their pattern of irregular checkups. We asked participants about the source of information they drew upon to make their decisions regarding screenings: Did they have consultations with a gynecologist about cervical cancer prevention? Did the information come from the friend or relative? Had that friend or relative had cervical cancer? Did the information come from exposure to the media (television, radio or newspapers)? Several questions addressed knowledge, attitudes and beliefs related to cervical cancer screening: Who should receive screening? What symptoms are present when a woman has cervical cancer? How important are regular checkups for older women and for women who are experiencing good health? How satisfied or dissatisfied are women with their current level of information about cervical cancer and screening? Three questions assessed cervical cancer risk perception of the respondents and what reasons or conditions might lead them to believe risk was high. Although we collected information in regard to the participants’ expectations and appraisal of the accuracy and reliability of the Pap smear results, these responses were not included in our results, given the generally very low level of understanding in both groups, as identified in the previous studies [15–17].

Data analysis

The odds ratio (OR) was estimated for each variable with 95% confidence interval using univariate logistic regression analysis. To test interactions among variables, multivariate logistic regression analysis was applied. The minimum significance level for entry was 0.05. The analyses were done by using the statistical software package SPSS 15.

Results

Based on the previously described inclusion criteria, 286 women were recruited into the study group. Of that number, 17 questionnaires were excluded from the data analysis due to incomplete or inconsistent answers and two women approached for the study group refused to participate In the end, 267 women were included in the study group and 267 women were included in the control group, in total, 534 study participants. The overall response rate was 93.4%.

The women who participated in the study group are not representative of the average female population of Belgrade. Compared with women in the census data, our respondents were younger, with a higher level of education and more likely to be married [18]. The study population mainly consisted of middle-aged (age group: 35–54 years) (47.2%) and married women (69%), with the high school education (12 years of formal education)—58%. In both groups, approximately three quarters (72%) of women had children (OR = 0.9, 95% CI: 0.7–1.4). The demographic characteristics of the women are presented in Table I.

Table I.

Demographic characteristic of study participants

Variable Study group (N = 267) Control group (N = 267) OR (95% CI) P
Age (years) 
    18–34 96 96   
    35–54 126 126   
    55–70 45 45 Matched  
Education 
    Elementary school 17 17   
    Trade 11 11   
    High school 155 155   
    University 84 84 Matched  
Marital status 
    Single 51 51   
    Married 184 184   
    Divorced 17 17   
    Widow 15 15 Matched  
Employment 
    Yes 170 185 0.8 (0.5–1.1)  
    No 97 82  ns 
Occupation 
    Housewife, farmer, manual worker, clerk 173 192   
    Professional 40 36 0.1 (0.7–1.7) ns 
Self-reported financial status 
    Excellent, good and average 250 181   
    Poor and very poor 17 86 6.9 (4.0–12.1) 0.001 
Homeownership status 
    Owner of an apartment/house 239 221 1.7 (1.0–3.0) 0.028 
    Lodger/share house 28 46   
Variable Study group (N = 267) Control group (N = 267) OR (95% CI) P
Age (years) 
    18–34 96 96   
    35–54 126 126   
    55–70 45 45 Matched  
Education 
    Elementary school 17 17   
    Trade 11 11   
    High school 155 155   
    University 84 84 Matched  
Marital status 
    Single 51 51   
    Married 184 184   
    Divorced 17 17   
    Widow 15 15 Matched  
Employment 
    Yes 170 185 0.8 (0.5–1.1)  
    No 97 82  ns 
Occupation 
    Housewife, farmer, manual worker, clerk 173 192   
    Professional 40 36 0.1 (0.7–1.7) ns 
Self-reported financial status 
    Excellent, good and average 250 181   
    Poor and very poor 17 86 6.9 (4.0–12.1) 0.001 
Homeownership status 
    Owner of an apartment/house 239 221 1.7 (1.0–3.0) 0.028 
    Lodger/share house 28 46   

ns = not significant.

a

According to univariate logistic regression analysis.

There were no significant differences between the study and control group in regard to employment and occupation status. Univariate logistic regression analysis indicated that women from the study group, when compared with those from the control group, were significantly more likely to report having better financial status (OR = 6.9, P = 0.001) and satisfactory homeownership status (OR = 1.7, P = 0.028) than did the controls.

There were significant differences between the two groups concerning practices of gynecological checkups (Table II). We found that the women from the study group significantly more often presented to the gynecologists from the primary healthcare institution (OR = 1.8, P = 0.046), while the controls chose a variety of settings for their irregular checkups. The study population significantly more often demonstrated no gender bias when selecting a gynecologist, while controls had gender preferences, mainly preferring female gynecologists (OR = 1.5, P = 0.023). A significantly higher proportion of women from the study group believed that the necessary frequency for Pap smears is at least once a year, or more frequently, than did the controls who argued for less frequent presentations (OR = 6.3, P = 0.010).

Table II.

Variables regarding practices of gynecological checkups

Variable Study group (N = 267) Control group (N = 267) OR (95% CI) P
To whom women usually present for the gynecological checkup 
    Gynecologist from the primary healthcare  institution 244 204 1.8 (1.0–3.1) 0.046 
    Private gynecologist, gynecologist from  the hospital, depends on the situation 23 34   
    Missing data 29 29   
Respondent's choice of gynecologist in regard to gender 
    No gender preferences 165 123 1.5 (1.1–2.1)  
    With gender preferences 102 115   
    Missing data  29  0.023 
Respondent's opinion about necessary frequency for PAP smears 
    Once a year and more frequently 255 205 6.3 (3.3–12.0)  
    Less frequently 12 62  0.001 
Variable Study group (N = 267) Control group (N = 267) OR (95% CI) P
To whom women usually present for the gynecological checkup 
    Gynecologist from the primary healthcare  institution 244 204 1.8 (1.0–3.1) 0.046 
    Private gynecologist, gynecologist from  the hospital, depends on the situation 23 34   
    Missing data 29 29   
Respondent's choice of gynecologist in regard to gender 
    No gender preferences 165 123 1.5 (1.1–2.1)  
    With gender preferences 102 115   
    Missing data  29  0.023 
Respondent's opinion about necessary frequency for PAP smears 
    Once a year and more frequently 255 205 6.3 (3.3–12.0)  
    Less frequently 12 62  0.001 

ns = not significant.

a

According to univariate logistic regression analysis.

A second area in which the study group participants differed significantly from the controls was that of knowing and talking with women who had been diagnosed with cervical cancer along with having had the opportunity to consult with a gynecologist (Table III). Namely, women in the study group were significantly more likely to know a woman who had been diagnosed with cervical cancer (OR = 2.7, P = 0.001) and to have discussed the issue with those women about their illness (OR = 3.3, P = 0.001); they were also significantly more likely to have consulted a gynecologist in regard to cervical cancer prevention (OR = 4.7, P = 0.003). Women in the study group were significantly more likely to report that they were exposed to media information about cervical cancer prevention (OR = 5.1, P = 0.002).

Table III.

Different sources of information about cervical cancer prevention and screening

Variable Study group (N = 267) Control group (N = 267) OR (95% CI) P
Acquaintance with the women with cervical cancer 
    Yes 105 51 2.7 (1.8–4.1) 0.001 
    No 162 216   
Conversation with the women with cervical cancer about that disease 
    Yes 65 16 3.3 (1.6–6.7) 0.001 
    No 40 35   
    Not applicable 162 216   
Consultation with a gynecologist about cervical cancer prevention 
    Yes 94 44 4.7 (1.7–9.1)  
    No 173 223  0.003 
Media exposure to the information on cervical cancer and screening 
    Yes 243 180 5.1 (1.8–7.5) 0.002 
    No 24 87   
Variable Study group (N = 267) Control group (N = 267) OR (95% CI) P
Acquaintance with the women with cervical cancer 
    Yes 105 51 2.7 (1.8–4.1) 0.001 
    No 162 216   
Conversation with the women with cervical cancer about that disease 
    Yes 65 16 3.3 (1.6–6.7) 0.001 
    No 40 35   
    Not applicable 162 216   
Consultation with a gynecologist about cervical cancer prevention 
    Yes 94 44 4.7 (1.7–9.1)  
    No 173 223  0.003 
Media exposure to the information on cervical cancer and screening 
    Yes 243 180 5.1 (1.8–7.5) 0.002 
    No 24 87   
a

According to univariate logistic regression analysis.

Women in the study and control groups also differed in their opinion about who should get a regular Pap smear (Table IV). The study group participants more often than the controls recognized who was at risk of developing cervical cancer: all women (OR = 3.6, P = 0.001), middle-aged women (OR = 2.3, P = 0.001), young women (OR = 4.1, P = 0.001), women with the family history of malignant diseases (OR = 1.7, P = 0.001), women who smoke or quit smoking (OR = 2.4, P = 0.001), women who frequently change sexual partners (OR = 1.7, P = 0.003) and women who have had at least one sexual partner (OR = 5.8, P = 0.001). The majority of women in both the study group and the control group (more than 92% in each) reported that they felt that older women no longer needed to get Pap smears, with no significant difference between the groups (OR = 4.1, P = 0.623). Also, there were no significant differences between the study and control groups regarding the percentage of women who reported that they were uncertain which women should be targeted for cervical screening, although there were twice as many indecisive women in the control group (6 versus 3%).

Table IV.

Women's opinions of who should obtain regular PAP smears

Variable Distribution
 
OR (95% CI) P
Study group Control group 
Women who have had at least one sexual intercourse 
    Yes 72 16 5.8 (3.3–10.3) 0.001 
    No 195 251   
Women who frequently change sexual partners 
    Yes 113 80 1.7 (1.2–2.4) 0.003 
    No 154 187   
Women who smoke and ex-smokers 
    Yes 106 57 2.4 (1.6–3.5) 0.001 
    No 161 210   
Women with the family history of malignant diseases 
    Yes 160 123 1.7 (1.2–2.5) 0.001 
    No 107 144   
Young women 
    Yes 33 4.1 (1.9–8.6) 0.001 
    No 234 258   
Middle-aged women 
    Yes 50 24 2.3 (1.4–3.9) 0.001 
    No 217 243   
Older women 
    Yes 22 19   
    No 245 248 1.2 (0.6–2.2) ns 
All women 
    Yes 203 125 3.6 (2.5–5.2) 0.001 
    No 64 142   
I do not know 
    Yes 17 0.4 (0.2–1.1) ns 
    No 259 250   
Variable Distribution
 
OR (95% CI) P
Study group Control group 
Women who have had at least one sexual intercourse 
    Yes 72 16 5.8 (3.3–10.3) 0.001 
    No 195 251   
Women who frequently change sexual partners 
    Yes 113 80 1.7 (1.2–2.4) 0.003 
    No 154 187   
Women who smoke and ex-smokers 
    Yes 106 57 2.4 (1.6–3.5) 0.001 
    No 161 210   
Women with the family history of malignant diseases 
    Yes 160 123 1.7 (1.2–2.5) 0.001 
    No 107 144   
Young women 
    Yes 33 4.1 (1.9–8.6) 0.001 
    No 234 258   
Middle-aged women 
    Yes 50 24 2.3 (1.4–3.9) 0.001 
    No 217 243   
Older women 
    Yes 22 19   
    No 245 248 1.2 (0.6–2.2) ns 
All women 
    Yes 203 125 3.6 (2.5–5.2) 0.001 
    No 64 142   
I do not know 
    Yes 17 0.4 (0.2–1.1) ns 
    No 259 250   

ns = not significant.

a

According to univariate logistic regression analysis.

Women's attitudes about cervical cancer risk perception differed significantly between the study and control group (Table V). Study group participants were significantly more able to estimate their own risk for getting cervical cancer (OR = 3.6, 95% CI: 2.2–5.8). They were more likely than members of the control group to report that their own risk was high or very high (OR = 3.7, 95% CI: 2.1–6.6).

Table V.

Women's attitudes in relation to cervical cancer risk perception

Variable Study group (N = 267) Control group (N = 267) OR (95% CI) P
Ability of respondent to estimate her cervical cancer risk 
    Able to estimate 240 190 3.6 (2.2–5.8) 0.001 
    Unable to estimate 27 77   
How does the respondent estimate her cervical cancer risk? 
    Very small, small and average 176 173 3.7 (2.1–6.6) 0.001 
    High and very high 64 17   
    Not applicable/unable to answer? 27 77   
How does the respondent estimate her cervical cancer risk compared to other women of her age? 
    Smaller or same risk 159 136 0.4 (0.2–1.1) ns 
    Higher risk 16   
    Unable to answer 92 125   
Variable Study group (N = 267) Control group (N = 267) OR (95% CI) P
Ability of respondent to estimate her cervical cancer risk 
    Able to estimate 240 190 3.6 (2.2–5.8) 0.001 
    Unable to estimate 27 77   
How does the respondent estimate her cervical cancer risk? 
    Very small, small and average 176 173 3.7 (2.1–6.6) 0.001 
    High and very high 64 17   
    Not applicable/unable to answer? 27 77   
How does the respondent estimate her cervical cancer risk compared to other women of her age? 
    Smaller or same risk 159 136 0.4 (0.2–1.1) ns 
    Higher risk 16   
    Unable to answer 92 125   

ns = not significant.

a

According to univariate logistic regression analysis.

In analyzing women's knowledge about cervical cancer and screening (Table VI), we found that women in the study group were significantly more likely than the those in the control group to have a greater amount of knowledge, particularly related to screening. For example, they were more likely to report that cervical cancer can be asymptomatic (OR = 1.9, 95% CI: 1.3–2.7), that healthy women should get screened on a regular basis (OR = 3.8, 95% CI: 2.4–5.9) and that if cervical cancer is diagnosed and treated in its early stages, the cancer can be cured (OR = 1.7, 95% CI: 1.2–2.5). Although the majority of women in both groups reported that they felt it was important that women of their age should undergo cervical cancer screening, participants of the study group were more likely to make this claim (OR = 6.6, 95% CI: 3.0–14.2). There was no difference between the two groups in regard to whether or not they felt they were sufficiently informed about cervical cancer, despite differences in their actual behavior. Included in the multivariate model were all variables related to preventive behavior and adherence to cervical cancer screening at the significance level of P ≤ 0.05 by univariate logistic regression analysis (Table VII). According to multivariate analysis, the following factors were significantly related to adherence to cervical cancer screening practices: better self-reported financial status (OR = 10.8, P = 0.001), no gender preference for a gynecologist (OR = 3.1, P = 0.015), consultations with a gynecologist about cervical cancer prevention (OR = 4.7, P = 0.029), talking to women who had been diagnosed with cervical cancer about the disease (OR = 2.8, P = 0.029) and receiving information about cervical cancer and screening through greater media exposure (OR = 5.0, P = 0.004).

Table VI.

Women's knowledge about cervical cancer and screening

Variable Study group (N = 267) Control group (N = 267) OR (95% CI) P
Women must have symptoms if they have cervical cancer 
    I do not agree 104 68 1.9 (1.3–2.7) 0.001 
    I agree 84 109   
It is important to perform cervical cancer screening in my age 
    I agree 259 222 6.6 (3.0–14.2) 0.001 
    I do not agree, not sure 45   
I am insufficiently informed about cervical cancer 
    I agree 130 149 1.1 (0.8–1.3) ns 
    I do not agree, not sure 137 118   
Cervical cancer in early stage could be successfully cured 
    I agree 233 173 1.7 (1.2–2.5) 0.001 
    I do not agree, not sure 34 94   
Healthy women should not regularly perform cervical cancer screening 
    I agree 19 43 3.8 (2.4–5.9) 0.001 
    I do not agree, not sure 248 224   
Variable Study group (N = 267) Control group (N = 267) OR (95% CI) P
Women must have symptoms if they have cervical cancer 
    I do not agree 104 68 1.9 (1.3–2.7) 0.001 
    I agree 84 109   
It is important to perform cervical cancer screening in my age 
    I agree 259 222 6.6 (3.0–14.2) 0.001 
    I do not agree, not sure 45   
I am insufficiently informed about cervical cancer 
    I agree 130 149 1.1 (0.8–1.3) ns 
    I do not agree, not sure 137 118   
Cervical cancer in early stage could be successfully cured 
    I agree 233 173 1.7 (1.2–2.5) 0.001 
    I do not agree, not sure 34 94   
Healthy women should not regularly perform cervical cancer screening 
    I agree 19 43 3.8 (2.4–5.9) 0.001 
    I do not agree, not sure 248 224   

ns = not significant.

a

According to univariate logistic regression analysis.

Table VII.

Model of preventive cervical cancer behavior of Belgrade female population

Variable OR (95% CI) P
Self-reported better financial status 10.8 (2.8–14.5) 0.001 
Choice of gynecologist with no gender preferences 3.1 (1.2–7.2) 0.015 
Consultation with gynecologist about cervical cancer prevention 4.7 (1.6–9.0) 0.029 
Conversation with women who have had cervical cancer about that disease 2.8 (1.1–5.6) 0.029 
Media exposure to the information on cervical cancer and screening 5.05 (1.1–7.8) 0.004 
Variable OR (95% CI) P
Self-reported better financial status 10.8 (2.8–14.5) 0.001 
Choice of gynecologist with no gender preferences 3.1 (1.2–7.2) 0.015 
Consultation with gynecologist about cervical cancer prevention 4.7 (1.6–9.0) 0.029 
Conversation with women who have had cervical cancer about that disease 2.8 (1.1–5.6) 0.029 
Media exposure to the information on cervical cancer and screening 5.05 (1.1–7.8) 0.004 
a

According to multivariate logistic regression analysis.

Discussion

This case-control study has confirmed many statistically significant and, as such, possibly important determinants of preventive behavior related to cervical cancer screening in women living in Belgrade. The study group participants had better knowledge about all important items concerning cervical cancer and screening. Yet, none of these variables was a strong determinant of preventive health behaviors, as we discuss below, in the final model.

Our final model indicated that preventive health behavior in the population of females living in Belgrade was strongly influenced by their exposure to cervical cancer information and communication about screening or illness within their social network. Women from the study group were much more likely to report that they had been exposed to both formal and informal health messages on screening or cervical cancer diagnosis and treatment; these messages came from three distinct sources (the media, gynecologists and other women). Our research indicates that women who received information from these multiple sources were more likely to present for screening, regardless of the level of formal education, employment status or occupation status or ability to estimate their cervical cancer risk. Based on our previous qualitative study, we hypothesize that while the provision of information from multiple sources may have not resulted in accurate understanding of cervical cancer and screening, women nevertheless undertook health promoting behaviors; trust in the gynecologist or friend rather than knowledge may have played a crucial role, an area that requires further investigation. This hypothesis is in accordance with Cialdini who argued that people may appear inconsistent in their choices, but there may be a reason for such behavior, including the (limited/unsupportive) personal network, poor access to services or understanding of information (health literacy) [19]. This survey confirms the importance of good communication between women and gynecologists as well as proper counseling. Furthermore, it demonstrates an urgent need for different educational and organizational strategies to improve the reproductive health of women in Serbia [16].

The results of this study have highlighted possible determinants of cervical cancer screening behaviors that might be used in future efforts to motivate women to seek cervical cancer screening. Our research has suggested that social relationships are of central importance and those individuals, who are in a position to influence a woman's behavior, include not only physicians and other health care providers but also friends and even acquaintances who have had experience with cervical cancer and screening [20]. In addition, the data collected in this study may be useful for future programs that use social marketing to improve cervical cancer screening in Serbia. Social marketing uses marketing techniques to influence the voluntary behavior of target audience members for health benefit [21]. Involvement of celebrities through social marketing may be a useful way of promoting preventive health behavior of women [22]. In 2005, for example, in the 2 weeks following Kylie Minogue's public announcement of her diagnosis, an unprecedented number of women presented for breast cancer screening and overall bookings in Australia rose by 40% [23].

This research also sheds light on poverty as one of the common social factors in transitional economies, which impedes women's ability to single out preventive behavior from other life priorities. Namely, preventive behaviors of women from the study group were significantly more associated with a better financial status (OR = 6.9, P = 0.001) and preferable homeownership status (OR = 1.7, P = 0.028). This was in accordance with the results of other international studies that have shown that women with lower economic status are less likely to undergo cervical cancer screening [24, 25]. Similarly to the results of Lin [26] who found lower economic status to be a barrier even for freely available screening services (covered by the National Health Insurance in the United Kingdom), women in Serbia face the same barrier. There are other possible reasons why poor women may not access free screening, including lack of child care, lack of free transport, insufficient social support and inability to develop a good rapport with her gynecologist due to infrequent presentation [27]. Our results are in accordance with the recent study from Serbia that has demonstrated the socioeconomic inequalities in the utilization of nonpreventive health services [28].

Finally, gender preferences in our model of preventive behavior confirm the ‘silent’ barrier for screening and may help explain why some women postpone the opportunity to have a free and available cervical smear (as revealed in our qualitative study discussed above). Further research is required to uncover if encouragement from people within the woman's social network to see a gynecologist regardless of gender may influence women's presentation to screening. Recent studies on Vietnamese American women and Mexican women confirm that the availability of female physicians seems to play an important role in the screening decisions [29, 30].

This study, like others, has limitations despite our conscientious efforts to utilize sound methodological approaches in our studies concerning cervical cancer screening behavior of Serbian women (case-control study, survey and qualitative study). As an example, we do not as yet have an appropriate ‘sampling framework’ for our research problem—the database of all eligible women for screening, making it impossible to recruit women randomly. While waiting the onset of the first national screening program and finally, the database, we are compelled to produce the pieces of puzzle that can add to our knowledge about preventive behavior. Another limitation is that the control group included women who did not present for screening regularly rather than those who had never been screened; these two types of women may face different barriers to screening, which may be explored in further research.

In conclusion, open communication, availability of female gynecologists, supportive social networks and improving social–economic status of women in our society are the most prominent factors influencing women's presentation to cervical cancer screening. As most of these factors are mainly outside the health services’ domain, they require multisectoral collaboration to improve women's reproductive health.

Funding

Ministry of Science and Technology of the Republic of Serbia (grant no. 145045).

Conflict of interest statement

None declared.

References

1.
International Agency for Research on Cancer (IARC)
Working Group on the Evaluation of Cervical Cancer Screening Programmes. Screening for squamous cervical cancer: duration of low risk after negative results of cervical cytology and its implication for screening policies
Br Med J
 , 
1986
, vol. 
293
 
6548
(pg. 
659
-
664
)
2.
U.S. Preventive Services Task Force
Screening for Cervical Cancer: Recommendations and Rationale. AHRQ Publication No. 03-515A
 , 
2003
Rockville: MD
Agency for Healthcare Research and Quality
 
Available at: http://www.ahrq.gov/clinic/3rduspstf/cervcan/cervcanrr.htm. Accessed: 11 December 2009
3.
Kesic
V
Jovicevic-Bekic
A
Vujnovic
M
Cervical cancer screening in Serbia
Coll Antropol
 , 
2007
, vol. 
31
 
Suppl. 2
(pg. 
31
-
6
)
4.
Todorova
I
Baban
A
Alexandrova-Karamanova
A
, et al.  . 
Inequalities in cervical cancer screening in Eastern Europe: perspectives from Bulgaria and Romania
Int J Public Health
 , 
2009
, vol. 
54
 (pg. 
222
-
32
)
5.
Nicula
FA
Anttila
A
Neamtiu
L
, et al.  . 
Challenges in starting organized screening programmes for cervical cancer in the new member states of the European Union
Eur J Cancer
 , 
2009
, vol. 
45
 (pg. 
2679
-
84
)
6.
Kasl
SV
Cobb
S
Health behavior, illness behavior, and sick role behavior
Arch Environ Health
 , 
1966
, vol. 
12
 (pg. 
246
-
66
531–541
7.
Behbakht
K
Lynch
A
Degeest
K
, et al.  . 
Social and cultural barriers to Papanicolau test screening in an urban population
Obstet Gynecol
 , 
2004
, vol. 
104
 (pg. 
1355
-
61
)
8.
Ogedegbe
G
Cassells
AN
Robinson
CM
, et al.  . 
Perceptions of barriers and facilitators of cancer early detection among low-income minority women in community health centers
J Natl Med Assoc
 , 
2005
, vol. 
97
 (pg. 
162
-
70
)
9.
Holroyd
E
Twinn
S
Adab
P
Socio-cultural influences on Chinese women's attendance for cervical screening
J Adv Nurs
 , 
2004
, vol. 
46
 (pg. 
42
-
52
)
10.
Blomberg
K
Ternestedt
BM
Tőrnberg
S
, et al.  . 
How do women who choose not to participate in population-based cervical cancer screening reason about their decision?
Psychooncology
 , 
2008
, vol. 
17
 (pg. 
561
-
9
)
11.
Frisch
LL
Allen
GD
Padonu
G
, et al.  . 
Social influences on Pap smear screening frequency
Alaska Med
 , 
2000
, vol. 
42
 (pg. 
41
-
5
47
12.
Jones
T
Clarke
V
Fernbach
M
Cervical cancer and cervical cancer screening: a review of literature
PapScreen Victoria Evaluation Report Volume One 1997–2000
 , 
2001
Victoria, Australia
Anti-Cancer Council of Victoria
(pg. 
21
-
31
)
13.
Basu
P
Sarkar
S
Mukherjee
S
, et al.  . 
Women's perceptions and social barriers determine compliance to cervical screening: results from a population based study in India
Cancer Detect Prev
 , 
2006
, vol. 
30
 (pg. 
369
-
74
)
14.
Mangoma
JF
Chirenje
MZ
Chimbari
MJ
, et al.  . 
An assessment of rural women's knowledge, constraints and perceptions on cervical cancer screening: the case of two districts in Zimbabwe
Afr J Reprod Health.
 , 
2006
, vol. 
10
 (pg. 
91
-
103
)
15.
Marković
M
Kesić
V
Topić
L
, et al.  . 
Barriers in cervical cancer screening. A qualitative study with women in Serbia
Soc Sci Med
 , 
2005
, vol. 
61
 (pg. 
2528
-
35
)
16.
Matejić
B
Kesić
V
Marković
M
, et al.  . 
Communications about cervical cancer between women and gynecologists in Serbia
Int J Public Health
 , 
2008
, vol. 
53
 (pg. 
1
-
7
)
17.
Kesić
V
Marković
M
Matejić
B
, et al.  . 
Awareness of cervical cancer screening among women in Serbia
Gynecol Oncol
 , 
2005
, vol. 
99
 
3 Suppl. 1
(pg. 
S222
-
5
)
18.
Republic of Serbia, Republic Statistical Office
Census of the Population, Households and Dwellings
 , 
2002
 
Available at: http://www.statserb.sr.gov.yu/Ter/pop.htm. Accessed: 11 January 2010
19.
Cialdini
RB
Influence: The Psychology of Persuasion. Revised edition.
 , 
1998
Collins
20.
Cabinet Office. Institute for Government
Mindspace: Influencing Behaviour Through Public Policy
 , 
2010
 
21.
U.S. Department of health and human services. National Institutes of Health
Theory at a Glance. A Guide for Health Promotion Practice
 , 
2009
2nd edn
 
22.
Larson
RJ
Woloshin
S
Schwartz
LM
, et al.  . 
Celebrity endorsements of cancer screening
J Natl Cancer Inst
 , 
2005
, vol. 
97
 (pg. 
693
-
5
)
23.
Chapman
S
McLeod
K
Wakefield
M
, et al.  . 
Impact of news of celebrity illness on breast cancer screening: kylie Minogue's breast cancer diagnosis
Med J Aust
 , 
2005
, vol. 
183
 (pg. 
247
-
50
)
24.
Coughlin
SS
King
J
Richards
TB
, et al.  . 
Cervical cancer screening among women in metropolitan areas of the United States by individual-level and area-based measures of socioeconomic status, 2000 to 2002
Cancer Epidemiol Biomarkers Prev
 , 
2006
, vol. 
15
 (pg. 
2154
-
9
)
25.
Tsu
VD
Levin
CE
Making the case for cervical cancer prevention: what about equity?
Reprod Health Matters
 , 
2008
, vol. 
16
 (pg. 
104
-
12
)
26.
Lin
SJ
Factors influencing the uptake of screening services for breast and cervical cancer in Taiwan
J R Soc Promot Health
 , 
2008
, vol. 
128
 (pg. 
327
-
34
)
27.
O'Malley
AS
Forrest
CB
Mandelblatt
J
Adherence of low-income women to cancer screening recommendations
J Gen Intern Med
 , 
2002
, vol. 
17
 (pg. 
144
-
54
)
28.
Jankovic
J
Simic
S
Marinkovic
J
Inequalities that hurt: demographic, socio-economic and health status inequalities in the utilization of health services in Serbia
Eur J Public Health
 , 
2010
, vol. 
20
 (pg. 
389
-
96
)
29.
Tung
WC
Benefits and barriers of pap smear screening: differences in perceptions of Vietnamese American women by stage
J Community Health Nurs
 , 
2010
, vol. 
27
 (pg. 
12
-
22
)
30.
Watkins
MM
Gabali
C
Winkleby
M
, et al.  . 
Barriers to cervical cancer screening in rural Mexico
Int J Gynecol Cancer.
 , 
2002
, vol. 
12
 (pg. 
475
-
9
)

Comments

0 Comments