Do women in rural areas of Serbia rarely apply preventive measures against cervical cancer ?

Background/Aim. The incidence of cervical cancer in Central Serbia has the higher rate as compared with that in other European countries. Considering mortality rate for cervical cancer, the standardized rate in Serbia is 10.1 per 10,000 females, which is the second highest one after that in Romania with 13.0. The aim of this study was to examine application of preventive measures for cervical cancer in women both from rural and urban areas in Serbia and if they are associated with sociodemographic characteristics and sexual behaviour. Methods. We analyzed secondary data of the 2006 National Health Survey of the population of Serbia focused on characteristics of adult females aged 25 to 65 years (5.314 in total) taking into consideration that programme of the organized screening will include female population aged over 25 years. Results. Respondents from rural areas have gynecological examination less than once a year in comparison with those from urban areas (OR = 0.60, 95% Cl 0.54–0.68). Less women from rural areas did Pap test during the last 12 months in comparison with respondents from urban areas (OR = 0.55, 95% Cl 0.48–0.64). Respondents from urban areas less often do the Pap test on doctor's advice in comparison with those from rural one (OR = 0.55, 95% Cl 0.42– 0.62). Conclusion. This study shows that women in rural areas rarely implement preventive gynecological measures againt cervical cancer in comparison with those in urban areas. Implementation of preventive measures among rural women is conditioned by lower levels of education and lower socioeconomic status.


Introduction
Incidence of cervical cancer in central Serbia has the highest rate among the other European countries.Significant regional differences range from the lowest rate (16.6 per 100,000 females) registered in the Machvan region to the highest one in eastern Serbia and the Belgrade region (32.5-38.1 per 100.00 females) 1 .Considering mortality rate for cervical cancer, the standardized rate in Serbia is 10.1 per 100.00 females, which is the second highest one after that in Romania with 13.0 2 .According to the current knowledge, human papillomavirus (HPV) infection has an important role in the development of the cervical cancer 3,4 .Other important risks are poor socioeconomic conditions, 5 chemical agents (smoking), 6 sexual habits (early sexual activities), promiscuity 7,8 factors associated with male partners 9 , abortions and deliveries in adolescent as 8 .There is an increased risk for women taking oral contraceptives 3 .Women from rural areas are at higher risk of cervical cancer associated with factors such a: lower educational level, 10 poorer socioeconomic conditions 11 and insufficient awarness of necessary regular preventive control examinations 12 .Women at high risk of cervical cancer are those unreliable for follow-up, those who have no regular Pap test, those with high parity 13 .
Cervical cancer is preventable and can be effectively treated if early diagnosed.The problem of both high incidence and mortality rate in Serbia can be partly attributed to the lack of awarness about health, but also to the problem in the system of health care approach and lack of the prevention programme.A key reason for higher incidence of cervical cancer in developing countries is the lack of effective screening programs 14 .
So far, in Serbia except for pilot projects in some regions, there was no organized screening for cervical cancer, but only the oportune one 15 .Alarming is the fact that 7.9% of women from rural areas in Serbia have never visited the gynecologist and that only 25.6% of them did the Pap test within a 3-year period 16 .Sociodemographic factors (educational level, occupation, socioeconomic status) are the dominant factors influencing upon application of preventive gynecologic examinations of women 17 .The study by Mateji et al. 18 have shown that the lack of women's knowledge on reproductive health in Serbia, inappropriate gynecologists' attitude and personal problems are associated with negative experience in the primary health care and influence upon low priority of preventive measures both for women and gynecologists 18 .All these factors result each year in about 500 fatal outcomes in Serbia due to cervical cancer.
According to the available literature no study has so far systematically examined differences in behaviour related to reproductive health in our country referring to either regular visits to gynecologist, colposcopic examinations and Pap test, or differences in usage of preventive health services among women living both in rural and urban environment of Serbia and being relevant for the cervical cancer prevention.This study has been designed to complement previous studies and point out women from rural areas as a target group with less probability to respond to organized screening programme for cervical cancer.
The aim of this study was to examine application of preventive measures against cervical cancer in women both from rural and urban areas in Serbia and to determine their association with sociodemographic characteristics and sexual behaviour.

Methods
This study was a secondary data analysis of the 2006 National Health Survey of the population of Serbia (without data concerning Kosovo and Metohia), carried out by the Ministry of Health of the Republic of Serbia with financial and professional support from the World Bank, the WHO regional Office for Europe (Country Office of Serbia) and the Institute of Public Health of Serbia "Dr Milan Jovanovich Batut".This was a cross-sectional study on a randomly selected representative population sample including 14,522 subjects aged 20 years and more 16 .In order to provide statistically reliable estimates of the health indicators on the national level, firstly was formed a stratified two-stage randomized sample of all registered housholds in the 2002 Serbia population census.Out of 7,673 randomly selected housholds, 6,156 were interviewed within the period September to October 2006.The houshold response rate was 80.2%.In these housholds there were 7,664 women aged 20 years and older.Individual response rate was 93.2%.We focused on characteristics of adult females aged 25 to 65 years (5,314 in total) taking into consideration that the programme of organized screening will include female population aged over 25 years 19 .Information about sociodemographic and socioeconomic characteristics as well as about preventive measures were obtained through interviews (face-to-face questionnaire and self-administered questionnaire) administered by the trained interviewers.
The three groups of data relevant to cervical cancer were analyzed: sociodemographic characteristics, sexual behaviour and application of preventive gynecological measures.
Of the analyzed sociodemographic variables there were: age of responders (shown in the ten-year intervals from 25 to 65 years); education (primary, high school and university); socioeconomic status measured by the houshold wealth index); region (Vojvodina, Central Serbia, Belgrade) and the number of children (0, 1, 2, 3 or more).According to the calculated wealth index values respondents were classified into five socioeconomic categories or quintiles: poorest, poor, medium, rich and wealthy 16 .
Of the variables related to sexual behaviour we analyzed: early sexual activity (< 16 years, 17-19, 20-22, 23-25  and > 26 years), the number of sexual partners in the last 12 months but not regular partners (1-3, or more) and the number of abortions (0-3, or more).
Of the variables related to the use of preventive measures we analyzed: frequency of regular gynecological examinations (once a year once every two years less than once in two years never) and time when the last Pap test was done (during the last year 1-3 years ago more than 3 years ago more than 5 years ago I do not remember never).Pap testing reasons (own decision on doctor's advice in the screening) were also analyzed.
All the above mentioned data are presented and compared with each other in relation to place of residence (urban/rural, urban = 0, rural = 1).
To find out whether the place of respondents' residence was a key factor influencing upon women's decision to practice prevention for cervical cancer, the variables (regular gynaecological examination, during the last Pap test and reasons for it) were analyzed as dependent ones and place of the residence as an independent one.Dependent variables were regular gynecological examinations (once a year, less than once a year), time of the last Pap test (in the last year, infrequently) and the reason for the Pap test (on their own initiative, on doctor's advice).
All the data were analyzed in accordance with the methods of descriptive and inferential statistics.The difference in distribution of the mentioned variables among women living in rural and urban areas was tested by the 2 test.For the minimum level of statistical significance p < 0.05 was used, where p < 0.01 was taken as highly statistically significant.Variables shown as highly significant after univariate analysis, were further tested by using multivariate analysis.
The association between gynecological measures preventive application dependent variables: regular gynecologi-cal examinations (once a year, less than once a year), time of the last Pap test (in the last year, infrequently), and reasons for the Pap test (on their own initiative, on doctor's advice) and sociodemographic factors (independent variables ) was tested by both univariate and multivariate logistic regression analysis.Multivariate logistic regression model was created by adding place of living.Associations were expressed by the odds ratio (OR) and 95% of the confidence interval (95% CI).
Analyses were performed by using the SPSS software (version 19).

Results
The study included women aged 25 to 65 years (mean age 44.9 ± 11.3 years).Table 1 shows the frequency and the results of the univariate logistic regression for sociodemographic characteristics of respondents by the place of living.
There were no differences in the age of women living either in urban or in rural areas of Serbia (p = 0.593).The number of examinees of all the age groups was equal both for those living in urban and rural regions (22-28%).With regard to the number of children, the difference is significant: in rural areas there are more women with two, three or more children.As for education, women with secondary or high educational level considerably more often live in urban areas.According to the socioeconomic status more women belonging to the category of the wealthiest live in urban areas in comparison with those living in rural ones.Table 2 shows the number of variables related to women's sexual behaviour in relation to the place of residence.Concerning the number of partners (not regular), there is no difference between respondents from rural and urban areas (p = 0.931).Also, there is no difference in the number of abortions (p = 0.452) among women living either in rutral or urban areas.In our sample, respondents from rural areas are more often younger when start with sexual relationships than those from urban environment (p < 0.000).
Table 3 shows the variables related to applications of preventive measures against cervical cancer in Serbia by the women's place of residence.Respondents from urban areas apply them considerably more often (gynecologycal examination and Pap test), in comparison with those from the rural areas (p < 0.000).Respondents from urban areas considerably more often do Pap test on their own, unlike those from the villages who more often do it owing to the doctor's advice.
Table 4 shows the results of the univariate regression analysis.Whether the respondents' place of residence is the key factor influencing upon women's decision to apply preventive measures for cervical cancer, then variables (regular gynecological examination during the last Pap test and the reasons for it), were all analyzed as dependent, but the place of residence as an independent one.Respondents from the rural areas had gynecological examination less than once a year in comparison with those from the urban areas (OR = 0.60, 95% Cl 0.54-0.68).Less women from rural areas did Pap test during the last 12 months in comparison with respondents from urban areas (OR = 0.55, 95% Cl 0.48-0.64).Respondents from the urban areas less often did the Pap test on doctor's advice in comparison with those from the rural one (OR = 0.55, 95% Cl 0.42-0.62).
Table 5 shows results of multivariate regression analysis for cervical cancer preventive measures (regular gynecological examinations and during the last Pap test).Multivariate logistic regression showed that the difference in implementation of cervical cancer preventive measures present between respondents from urban and rural areas were not caused by the place of living, but by their educational level and financial status being lower in women from the rural areas.The middle aged and older respondents went to preventive examination for cervical cancer less often than the youger ones (aged 25-34 years).College-educated women often go to preventive gynecological examinations in comparison with those with only primary school education (OR = 1.32, 95% Cl 1.10-1.58).The richest women in comparison with the poorest ones went often to control gynecological examinations (OR = 1.29, 95% Cl 1.05-1.52).In the last 12 months the Pap test was more often performed in college-educated women in comparison with those with primary education (OR = 1.42, 95% Cl 1.17-1.72).When compared with the poorest women, the richest ones went to the Pap tests more frequently in the last 12 months (OR = 1.27, 95% Cl 1.04-1.56).Age was also a factor influencing upon women's decision to do preventive controls.The oldest category subjects rarely went to preventive gynecologic examinations in comparison with the youngest ones (25-34 years), (OR = 0.41, 95% Cl 0.4-0.50)and to the Pap test (OR = 0.44, 95% Cl 0.35-0.55).
Table 6 shows the results of the multivariate regression analysis of the reasons to go to the Pap test.This analysis shows that the difference existing between the respondents from urban and rural areas is not only caused by the place of respondents' living, but also by their educational and financial status being lower in women from rural areas.Urban areas respondents did the Pap test less often on doctor's advice than on their own initiative (OR = 0.79, 95% Cl 0.63-1.00).Age was not a factor influencing upon reason for the Pap test.Only those aged 35-44 years did the Pap test less often on the doctor's advice in comparison with younger than 35 years (OR = 0.75, 95% Cl 0.58-1.00).

Discussion
The aim of this study was to examine applications of preventive measures among women in rural and urban areas of Serbia and to identify whether they have any association with women s sociodemographic characteristics and sexual behaviour.
It was found out that women in rural areas rarely apply preventive measures against cervical cancer (gynecological controls and Pap tests) in comparison with those living in urban areas.However, after considering effects of educational level, socioeconomic status and age, this difference was no longer significant.Women from rural areas, less educated, and with lower socioeconomic status, middle-aged and elderly, rarely apply preventive measures against cervical cancer.
Numerous studies have identified demographic and behavioural factors associated with cervical cancer prevention 12,20 .Higher socioeconomic status is associated with more frequent application of preventive measures against cervical cancer in female population on the Belgrade territory 21 .
Our study confirms that girls from the rural areas often have sexual activities before the age of 16 years in comparison with those from urban areas.The majority of our respondents had the first sex at the age of 19.Early sexual activity is a risk factor for reproductive organs health and for sexually transmitted diseases (STDs) including HPV infections 3 .In the study performed by Stankovi et al. 22 , most of girls have sexual activities at the age of 17.9.Asked about the number of not regular partners within the last 12 months, only 2.4% of respondents answered.According to the obtained answers there was no difference between respondents' place of residence.Having more sexual partners is a risk factor for reproductive organs health and numerous studies emphasize the importance of reducing irregular partners number for cervical cancer prevention 7 .As for the number of abortions, 40.3% of respondents gave this information with no significant differences with regard to the place of respondents' residence.These data are inconsistent with the study written by Rasevich and Sedlecki 23 in which the projected number of abortions in Serbia was over 200,000 a year.However, registered data on induced abortions since 1999 are not reliable (the number of registered abortions in 2000 was 42,322 and in 2007 was 24,273).
According to our results, women from urban areas considerably more often go to gynecological examinations in comparison with those from rural areas.Also, the Pap test once a year did significantly more respondents from urban than from rural areas.The fact that every other woman from rural area as well as one-third of those from urban environment have never done a Pap test.Since the same number of women (4,894) in our study gave information about gynecological examination and the Pap test, but because their medical reports were not used for verification of their statements, we could not confirm that any of them did both gynecological examination and the Pap test.The fact that 8.9% of women from rural areas do not know what kind of a test it is, our results may appear to be doubtful.In a study on Serbian population health in 2006, in a sample of women older than 20 years, 6.3% of respondents had never visited a gynecologist.Organized prevention activities and screening cover only a small part of female population and there are also considerable geographic differences and variations by the type of a settlement (5.2%) in Eastern Serbia where the incidence of cervical cancer is highest).Many women do not go to gynecological examination because they are symptomless and have finished the reproductive function, postmenopausal and elderly women from rural areas 16 .
In our study as a part of the organized screening, the Pap test was done only by 4.7% of women in rural areas and by 5.1% of them in urban ones.Women from rural areas were screened more frequently upon gynecologist's advice, as a type of opportune screening.Women from urban areas did more frequent Pap screening independently.The study by Spaczinski et al. 20 as well as our study confirm that women from villages less often than those from urban areas comply to the Pap screening.Some studies indicate that Pap screening is associated with the socioeconomic women's status 11,24 .This study as well as those by Franceschi et al. 5 , Sabates and Feinstein 11 , and Spaczynsky et al. 20 confirm that higher level of education is an important factor for women's decision to initiate the Pap screening test.
Screening was conducted in some other countries such as Hungary until 2003, but it did not offer satisfying results aiming to reduce morbidity and mortality rate from cervical cancer 25 .Efforts to prevent cervical cancer in women are worldwide focused on organized screening and treatment of precancerous lesions.When screening of high quality and coverage were realized, the incidence of invasive forms of cervical cancer in Serbia was reduced even by 90% 26 .Since 2012 in Serbia a classical cytodiagnostics has been applied as a part of the programme of organized cervical cancer screening 19 .American Association of Obstetrics and Gynecology and the European Association for Infectious Diseases in Gynecology and Obstetrics have given different recommendations for screening such as: to initiate screening at the age of 21 with a 2-year intervals up to 30 years of age and afterwards, combination of ligniol based cytology (LBC) and HPV testing at a 3-year intervals.This instruction was aimed to prevent many unnecessary tests for women who are not at risk of cervical cancer 27,28 .Only 16 European Union countries have organized the National Screening Programme for cervical cancer and they applied it for women aged 20-30 and 60-65 years after 3-or 5-year intervals 29 .
In Poland, highly educated women from urban areas more frequently do screening in private health institutions 20 .Numerous studies both from developed countries, Greece, Sweden as well as from China and South Africa confirmed the association between the place of residence and preventive measures against cervical cancer, with more frequent application of preventive measures in urban female population [30][31][32]14 .

Conclusion
Our study shows that women in rural areas rarely implement preventive gynecological measures against cervical cancer in comparison with those in urban areas.Women from rural areas have more risk factors (early sexual relationships, higher parity, lower educational and socioeconomic status).Implementation of preventive measures among rural women is conditioned by lower levels of education and lower socioeconomic status.A particular attention should be paid to reducing identified differences.Education and preventive gynecological practices should be provided and available to rural women.