|Ahead of print
Risk factors and prevalence of cervical squamous intraepithelial lesions among women in south India: A community-based cross-sectional study
MC Kalavathy1, Aleyamma Mathew1, KM Jagathnath Krishna1, VN Saritha2, K Sujathan2
1 Divisions of Cancer Epidemiology and Biostatistics, Regional Cancer Centre, Trivandrum, Kerala, India
2 Cancer Research, Regional Cancer Centre, Trivandrum, Kerala, India
|Date of Submission||03-Aug-2019|
|Date of Decision||04-Aug-2019|
|Date of Acceptance||06-Jul-2020|
|Date of Web Publication||27-Jan-2021|
Divisions of Cancer Epidemiology and Biostatistics, Regional Cancer Centre, Trivandrum, Kerala
Source of Support: None, Conflict of Interest: None
Background: Cervical cancer in India accounts for one-fifth of the global burden. Well-defined precancerous stages help early detection of the disease. Apart from human papillomavirus, the risk factors include age, education, occupation, early age at marriage and first delivery, abortions, and multiple sexual partners. Prevalence and risk factors for cervical squamous intraepithelial lesions (SIL) among women by Pap smear screening in south India were analyzed through a cross-sectional study.
Methods: Women from rural and urban area were motivated by local accredited social health activists to attend pre-fixed Pap smear clinics in government hospitals. Pap smears collected in these clinics were taken to the Regional Cancer Centre, Thiruvananthapuram, processed, and cytology reports were prepared. Multiple logistic regression analysis was used to identify risk factors for SIL and high-grade SIL (HSIL).
Results: The number of SIL was 67 out of 10,580 and HSIL was 39. Having higher education (Odds Ratio, OR:0.05(95% Confidence Interval, CI: 0.01-0.2), being married but living single (OR : 5.3, 95%CI:2.4-11.5), Having >2 abortions (OR:21, 95% CI:4.5-24), having younger age at delivery (OR : 0.1, 95% CI:0.01-0.3) and having unhealthy cervix (OR: 16.4, 95% CI:6.2-42.7) were the factors found to be the associated risk factors in multiple regression analysis.
Conclusion: Pap smear screening can be focused on women with risk factors such as low education, married but living single, having >2 abortions, younger age at delivery, and unhealthy cervix on per speculum examination.
Keywords: Cervical cancer, HSIL, LSIL, Pap smearKey Message Lower education, early age of delivery, living seperated from spouse, having more abortions and unhealthy cervix are associated with low and high grade squamous intraepithelial lesions .Cervical cancer control programmes to be focused on women with these risk factors.
|How to cite this URL:|
Kalavathy M C, Mathew A, Jagathnath Krishna K M, Saritha V N, Sujathan K. Risk factors and prevalence of cervical squamous intraepithelial lesions among women in south India: A community-based cross-sectional study. Indian J Cancer [Epub ahead of print] [cited 2021 Jul 25]. Available from: https://www.indianjcancer.com/preprintarticle.asp?id=308050
| » Introduction|| |
Globally, around 567,000 new cases of cervix-uteri cancers and around 266,000 deaths from this disease occur annually, the same being 96,900 new cases and 67,000 deaths in India, accounting for one-fifth of the global burden of this disease. According to the Indian National Cancer Registry Programme Report 2016, rural cancer registries like Barshi, Maharashtra showed cervix-uteri cancer incidence rate to be 16 per 1,00,000 while Trivandrum, Kerala, showed an incidence of 7 per 100,000. The incidence in Trivandrum came down from 9.6 in 2005 to 6.1 in 2015 per 100,000 women. Cervix-uteri cancer incidence has been declining globally and nationally. Still, the disease continues to occupy second or third positions in different Indian registries. The disease is associated with well-defined precancerous stages starting from mild cellular atypia to low-grade squamous intraepithelial lesion (LSIL) and then high-grade squamous intraepithelial lesions (HSIL), which may progress to cervix-uteri cancer if left untreated. The disease can be detected in the precancerous stages by using a Pap smear test. The introduction of systematic population-based Pap smear screening has been shown to bring a reduction in the burden of cervical cancer in many developed countries.
Based on a pooled analysis of three studies done in West Bengal, Basu et al. have reported a population prevalence of 2.3% and 0.9% for LSIL and HSIL, respectively. A study from Haryana reported one LSIL and HSIL each out of a total of 1615 women. In a population-based study done in Kerala, the prevalence of both LSIL and HSIL were 0.18% and 0.15%, respectively.
A study from China reported a difference in the prevalence of precancerous cervical intraepithelial neoplasia (CIN) 2 and 3 among rural and urban populations. In rural and urban women, age-standardized CIN2 prevalence was 1.5% (95% confidence interval [CI]: 1.4–1.6%) and 0.7% (95% CI: 0.7–0.8%) and CIN3 prevalence was 1.2% (95% CI: 1.2–1.3%) and 0.6% (95% CI: 0.5–0.7%), respectively.
Infection with human papillomavirus (HPV), a DNA virus, is supposed to be associated with the development of CIN and cervical cancer. Of more than 300 strains of HPV identified, nearly 10 high-risk strains (HRHPV) have been found to be the necessary cause of cervical neoplasia. These HRHPV infections may result in the development of atypia in cervical epithelium, and the persistence of this infection can lead to the development of SIL, both low- and high-grade. Other than the HPV, the risk factors associated with the development of this cancers/precancers include early age at first sexual intercourse, early age at first delivery, more number of abortions, and multiple sexual partners. According to a study by Wang et al. in 2017, the variables such as age, education, and occupation differed significantly between the CIN group and the control group. Even though the incidence has been declining, cervical cancer continues to be an important public health issue in India. Since cervical precancer registry data are not available in India, information about the prevalence and risk factors of cervical precancers can be assessed through focused screening programs.
The specific objectives of the present study are (i) to assess the prevalence of SIL including low- and high-grade SIL (including LSIL and HSIL) and HSIL alone in a community-based Pap smear screening program in urban and rural areas in Thiruvananthapuram district, south India and (ii) to identify the risk factors for the development of the above mentioned conditions.
| » Materials and Methods|| |
The study was conducted in a cross-sectional manner during 2010–2014 (48 months). The study participants were women from urban and rural areas in Thiruvananthapuram district reported to Pap smear screening. A total of 127 primary health centers (PHCs) in the district (rural and urban) were selected for the Pap smear screening program. Parous or married women of age 30–64 from the specific district, who were at least on the 11th day of their menstrual cycle and who opted to attend the Pap smear screening clinic only were eligible for the study. Unmarried women, pregnant women, women with history of hysterectomy, and women with preexisting malignancy or growth per cervix or vaginum were excluded from the study. The key persons in this programme were Accredited Social Health Activists (ASHAs), who conducted the house-to-house survey for the identification of women to be screened from the general public. These ASHAs were given training on the signs, symptoms, etiology, early detection, etc. of uterine cervical cancer in simple vernacular language. They visited each and every house and invited women in the 30–64 years age group for a Pap smear examination, conducted in the local PHC on prefixed dates. This Pap smear clinic was run by the Regional Cancer Centre (RCC), Thiruvananthapuram. All women who attended the clinic and fulfilled the eligibility criteria and provided informed consent were included in the study.
Women were subjected for an interview using a pre-tested questionnaire which collects information on the demographic, socioeconomic, reproductive, and obstetric details. A per speculum examination of cervix was done under good light source and women were subjected for Pap smear collection. The methodology adopted is the traditional Pap smear using Cusco's speculum and Ayres spatula. The collected samples were assigned specific identification number which helped in keeping easy track of patients. Pap smears collected in these clinics were taken to RCC, Thiruvananthapuram, processed, and cytology reports were prepared. The study was approved by the institutional review board of RCC.
Women with abnormal Pap smear were actively followed up, directed for further investigations, and referred for further treatment as indicated.
The analysis was done with SIL (including LSIL and HSIL) and HSIL alone as outcomes. Analysis with LSIL alone as outcome was not done due to the small numbers. Two trained social investigators were involved in data collection throughout the study. The exposures or variables analyzed included age, educational status, occupation, economic status, current marital status (living with husband or not), age at first marriage, age at first delivery, number of pregnancy, number of abortions, type of delivery, and clinical appearance of cervix per speculum. These variables were grouped into two or three categories. Age was grouped into two categories—upto 35 and above 35 years. Age at marriage and age at first delivery were grouped into three categories—below 20, 20–24, and above 24 years. Statistical package for social sciences (SPSS) software (version 20) was used for this categorization and further analysis.
Data were summarized using descriptive statistics such as frequency, mean along with standard deviation (SD). Initial analysis for association of Pap smear abnormality with each variable was done using Chi-square test and for small samples, Fisher's exact test was employed. The risk for each variable was estimated using logistic regression. Variables found to be significant in univariate logistic regression with a P value <0.05 were subjected for multiple logistic regression analysis.
| » Results|| |
The total samples included 10,580 women who had undergone Pap smear examination, urban: 3939 (175, 884, 1234, and 1046 year wise), rural: 6641 (1120, 2070, 1790, and 1661 year wise). The mean age of the participants was 46.49 (SD = 9.1, range: 20-86). Of the total population, 5671 (53%) had secondary education, 5004 (47.3%) were from lower income category, 9050 (77.6%) were manual laborers, 85.2% were living with husbands. The mean age at marriage and mean age at first delivery of the participants were 21.61 (SD: 2.9, range: 11-51) and 23 (SD: 4.3, range: 13-48) years, respectively. The mean number of pregnancy was 2.55 (SD: 1.3, range: 1-12) and the mean number of abortion was 0.55 (SD: 0.7, range: 1-9); 8580 (81%) had normal vaginal deliveries and 6872 (65%) had normal appearing cervix per speculum.
The total number of SIL was 67 (0.8%), urban: 15 (0.4%) and rural: 52 (0.8%). The mean age at diagnosis of SIL was 50.2 years (SD: 9.9, range: 30-64 years). There were 39 cases of HSIL (0.36%) of which 6 (0.15%) were in urban and 33 (0.5%) in rural [Table 1]. The mean age at diagnosis of HSIL was 54.6 years (SD: 8.6, range: 40-64 years).
A very high proportion of women with HSIL had lower levels of education (illiterate: 17, 44%) compared to normal women (illiterate: 520, 9.4%). All women with SIL and HSIL were manual laborers. However, only 87.9% of normal women were manual laborers. In the univariate regression analysis of SIL, the significant risk factors were (i) having more than two abortions (Odds Ratio, OR 71.8), (ii) not living with husband (OR: 7.6), and (iii) having an unhealthy appearing cervix (OR: 12.4). The corresponding significant protective factors were (i) having primary/middle level (OR: 0.18) or middle/higher level of education (OR: 0.03), (ii) being in the middle or higher economic strata (OR: 0.48), (iii) age of marriage being 20–24 years (OR: 0.42), (iv) age of first delivery being >24 (OR: 0.10), and having cesarean section (OR: 0.30).
In the multiple logistic regression analysis, having >2 abortions (OR: 94.3) and an unhealthy appearing cervix (OR: 14.4) were associated with the high-risk for developing SIL. Having education of primary/middle (OR: 0.26) and age of first delivery being >24 (OR: 0.15) were the corresponding protective factors [Table 2].
In the univariate logistic regression analysis, the risk factors for HSIL were: (i) having abortions (OR: 13.0), (ii) not living with husband (OR: 7.6), and (iii) a clinical finding of unhealthy cervix (OR: 15.0). The protective factors against the development of HSIL were (i) having middle and above level education (OR: 0.2), (ii) age at first marriage above 20 (OR: 0.1), (iii) having age at first delivery 20–24 years (OR: 0.3) or above 24 (OR: 0.1), and (iv) delivery by cesarean section (OR: 0.1).
In the multiple logistic regression analysis, the risk factors for HSIL were not living with husband (OR: 5.3), having 1–2 abortions (OR: 20.0) or more than 2 abortions (OR: 21.0), and a clinical finding of unhealthy cervix (OR: 16.0). The corresponding protective factor was higher education, OR and 95% CI being, 0.2 and 0.1–0.5, respectively [Table 3].
| » Discussion|| |
In the present study, the overall prevalence of SIL was 0.4% and HSIL was 0.36%. The prevalence of SIL was low compared to a few similar studies in India. In a population-based cohort study done in Taiwan, the prevalence of HSIL was 0.45%. Another population-based study in Nepal, conducted according to their national guidelines for cervix cancer in 2006–2007, reported the prevalence of HSIL to be 0.6%. Cervix-uteri cancer incidence rates in Trivandrum were low and this could be the reason for the low prevalence of HSIL in this population.
In the present study, the prevalence of both SIL and HSIL were more in the rural population compared to the urban population. The high prevalence of cervical lesions in rural sample is reported in other studies also.[12,13] Urban-rural differentiation was done for the estimation of the prevalence of SIL and risk estimation could not be done due to small numbers.
In the present study, occupation, marital status, method of delivery, and per speculum finding of cervix were grouped into two categories and all other variables were grouped into three categories to enable statistical analysis. The number of missing cases were: age in 28 cases, education in 55 cases, occupation in 201 cases, economic status in 182 cases, marital status in 182 cases, number of pregnancy in 89 cases, method of delivery in 99 cases, age at marriage in 118 cases, age at first delivery in 153 cases, number of delivery in 214 cases, number of abortions in 181 cases and per speculum finding of cervix in 167 cases, and the over-all missing data were below 2%.
In the present study, higher level of education had a negative association with both SIL and HSIL. This is in accordance with many studies conducted worldwide. This could be because of the fact that lower level of education is usually associated with low economic background with restricted access to health care, etc. The study conducted in Beijing reported that having college education and above compared to lower level education had a protective effect against the development of CIN II/III (OR = 0.79; 95%: CI = 0.72–0.90, P = 0.033). In a study in China, women with education of junior/middle school or below had an OR of 1.62 for the development of HSIL (P = 0.016) compared to women with the education of senior middle school or above. In a study reported in Hong Kong, a total of 44,219 women attended the cervical cancer screening and the significant risk factors of cervical abnormalities included having received primary school education only (OR 1.5).
In the present study, women who were not living with their husband showed a high-risk for developing HSIL. This kind of an association might be due to the fact that they are at high-risk of being in the lower socioeconomic status and in many situations there might be a higher chance of having sexual promiscuity. The role of their first husband's promiscuity is also to be studied. In a study reported from Chennai in 2003, which was a case-control study, the OR for cervical cancer among widowed women was 8.3 (95% CI: 4.2–17.6) and all the seven women who were either separated or divorced were of cervical cancer cases. Makuza et al. (2015) reported from Rwanda that being living single (married but divorced and widowed) (OR 3.29; 95% CI 1.26–8.60) is a risk factor for developing CIN. Many times married women living single give least preference to their gynecologic care, pointing the importance of marital status with regard to the gynecological health of women. Even though this was reported from among Arab women in United States, the same is true for majority of the rest of the world. They have many other priorities in life like supporting family, healthcare of the rest of the family, and above all the financial concerns. So they are detected in the precancerous stage when they take part in Pap smear programs. The possibility of increased chance of these women attending such physically and financially accessible screening programmes also has to be taken into consideration. In this study, this kind of an association was not evident in case of SIL because the effect of this variable in the development of LSIL (SIL being combination of LSIL and HSIL), was not significant.
In the present study, history of abortions had a strong positive association with HSIL and SIL. Tao et al. reported that in China the OR for developing HSIL was 1.53. Chankapa et al. in 2011 found that abortion and still birth were significantly related with cervical lesion (P = 0.0287, RR: 1.6 (95%CI: 0.96–2.56). As part of the European Prospective Investigation into Cancer and Nutrition (EPIC) study in Europe in 2016, an increased risk for cervical cancer was seen among women who had at least one induced abortion.
Almost in 98% of study samples, both the age at first pregnancy and age at first child birth were the same; hence, in the present study, we considered only age at first delivery. Here we have seen that the young age at first childbirth was statistically associated with the development of SIL. In a pooled analysis of case-control studies on invasive cervical cancer (ICC) done from different countries with varied ICC rates, Louie et al. reported that women who had their first pregnancy before or at the age of 16, had an OR of 2.36 (95% CI: 1.82–3.07) for developing ICC compared to women who had their first pregnancy between 17 and 20. Zhang et al. in 2016 reported the relationship of age of first delivery upto 20 with increased incidence of CIN (OR of 3.75) (95% CI: 1.19-11.84). In this study, this variable was not significant in case of HSIL may be because of small numbers.
The unhealthy appearance of cervix on unaided visual inspection has shown a strong association with both SIL and HSIL in the present study. Misra et al. reported from Lucknow that the frequency of both SIL and carcinoma cervix was significantly high in women with unhealthy and abnormal cervix (P < 0.01). It was seen that the clinical abnormalities of cervix is associated with a higher proportion of cervical neoplasia (14.1%). Hence, subjecting these women to routine Pap smear evaluation will help in reducing the burden of carcinoma cervix.
The results in the present study may not be replicated in other populations with different characteristics as the beneficiaries were mainly from lower socioeconomic strata, 7739 (76%) being in the lower and lower middle class category and may not be externally valid in a community with more upper economic strata.
The main limitation of the study is that the data collected were mainly representative of lower and lower middle income category (only 24% in the upper middle and upper class) and thus the results might not be a true representation of the general population. Factors like age, unwillingness to attend the PHCs, and denied access due to physical and logistic reasons might have affected the outcomes. Reporting of wrong data either accidentally or wilfully also might have affected results, especially in crowded clinics. Symptomatic women might have over responded to many questions, but this could be managed to a great extent by the trained investigators.
In conclusion, the study brings light into significant risk factors associated with the development of SIL. This helps to shape the cervical cancer prevention policy, which has to focus on women who are illiterate/just literate, women who are living single (married but living single due to divorce, separation, and widowed) and who had abortions and they can be referred for routine Pap smear screening. A per speculum examination of cervix is also important.
We thank the Trivandrum Jilla Panchayath, Trivandrum Corporation Dr. Preethi Sara George, Associate Professor, for the help and support. We also thank Mrs. Nanda and Mrs. Ragini, the project staff, Division of Cancer Epidemiology and Biostatistics, Regional Cancer Centre, Trivandrum, for their effort in data collection.
Financial support and sponsorship
The study is part of the many screening programs being conducted in the RCC, Thiruvananthapuram funded by the Government of Kerala.
Conflicts of interest
There are no conflicts of interest.
| » References|| |
Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global Cancer Statistics 2018: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries, CA Cancer J Clin 2018;68:394-424.
Elmslie TJ. The Pap smear and cervical cancer screening. Can Fam Physician 1987;33:131–7.
Van der Aa MA, Pukkala E, Coebergh JW, Anttila A, Siesling S. Mass screening programmes and trends in cervix cancer in Finland and the Netherlands. Int J Cancer 2007;122:1854-8.
Basu P, Mittal S, Bhaumik S, Mandal SS, Samaddar A, Ray C, et al
. Prevalence of high-risk human papillomavirus and cervical intraepithelial neoplasias in a previously unscreened population- A pooled analysis from three studies. Int J Cancer 2013;132:1693–9.
Varghese C, Kalavathy MC, Amma NS, Nair MK. A Cohort study to address the natural history of HPV and Cervical dysplasia in Trivandrum, South India: Methodological Issues and Initial Results. Asian Pac J Cancer p 2001;2:63-7.
Zhao F, Lewkowitz AK, Hu S, Chen F, Li L, Zhang Q, et al
. Prevalence of human papillomavirus and cervical intraepithelial neoplasia in China: A pooled analysis of 17 population-based studies. Int J Cancer 2012;131:2929–38.
Tao L, Han L, Li X, Gao Q, Pan L, Wu L, et al
. Prevalence and risk factors for cervical neoplasia: A cervical cancer screening program in Beijing. BMC Public Health 2014;14:1185.
Khan MJ, Castle PE, Lorincz AT, Wacholder S, Sherman M, Scott DR, et al
. The elevated 10-year risk of cervical precancer and cancer in women with human papillomavirus (HPV) type 16 or 18 and the possible utility of type-specific HPV testing in clinical practice. J Natl Cancer Inst. 2005;97:1072–9.
Wang Z, Wang J, Fan J, Zhao W, Yang X, Wu L, et al
. Risk factors for cervical intraepithelial neoplasia and cervical cancer in Chinese women: Large study in Jiexiu, Shanxi Province, China. J Cancer 2017;8:924–32.
Chao A, Hsu KH, Lai CH, Huang HJ, Hsueh S, Lin SR, et al
. Cervix cancer screening program integrating Pap smear and HPV DNA testing: A population-based study. Int J Cancer 2008;122:2835-41.
Wong HY, Loke AY, Chan NH. Risk factors for cervical abnormalities among Hong Kong Chinese women: A large-scale community-based cervical screening program. J Womens Health (Larchmt) 2011;20:53-9.
Franceschi S, Rajkumar T, Vaccarella S, Gajalakshmi V, Sharmila A, Snijders PJF, et al
. Human Papillomavirus and risk factors for cervical cancer in Chennai, India: A case-control study. Int J Cancer
Makuza JD, Nsanzimana S, Muhimpundu MA, Pace LE, Ntaganira J, Riedel DJ. Prevalence and risk factors for cervix cancer and pre-cancerous lesions in Rwanda. The Pan Afr Med J 2015;22:26.
Abboud S, Penning E D, Brawner B M, Menon U, Glanz K, Sommers MS. Cervical cancer screening among Arab women in the United States: An integrative review. Oncol Nurs Forum 2017;44:E20-33.
Chankapa YD, Pal R, Tsering D. Correlates of cervical cancer screening among underserved women. Indian J Cancer 2011;48:40-6.
] [Full text]
Roura E, Travier N, Waterboer T, de Sanjosé S, Bosch FX, Pawlita M, et al
. The influence of hormonal factors on the risk of developing cervical cancer and pre-cancer: Results from the EPIC cohort. PLoS One 2016;11:e0147029.
Louie K S, de Sanjose S, Diaz M, Castellsagué X, Herrero R, Meijer CJ, et al
. Early age at first sexual intercourse and early pregnancy are risk factors for cervix cancer in developing countries. Br J Cancer 2009;100:1191–7.
Zhang Q, Xie W, Wang F, Li RH, Cui L, Wang H, et al
. Epidemiological investigation and risk factors for cervical lesions: Cervix cancer screening among women in rural areas of Henan Province China. Med Sci Monit 2016;22:1858-65.
Misra JS, Srivastava S, Singh U, Srivastava AN. Risk-factors and strategies for control of carcinoma cervix in India: Hospital based cytological screening experience of 35 years. Indian J Cancer 2009;46:155-9.
] [Full text]
[Table 1], [Table 2], [Table 3]