The first author conducted the study after informing the purpose of the study to each of the study participants and obtaining an informed written consent from the study participants, as the study addresses some of the sensitive personal issues of the participants.

Study Design: A community based case control study was conducted among the breast cancer cases with age matched controls in Arpookara Panchayat of Kottayam district in Kerala, India.

Study Area: Arpookara is a Grama Panchayat and is one of the field practice areas of the Government Medical College, Kottayam, Kerala, with a total population of 22,942 with females constituting about more than fifty percent of the population. It is located 9.58°N latitude and 76.52° E longitude.

Study Period: October 2012 – December 2012.

Cases

Inclusion Criteria

Exclusion Criteria

Controls

Inclusion Criteria

Exclusion Criteria

Breast cancer cases in Arpookara Panchayat who have been registered in the cancer register of the Government Medical College, Kottayam irrespective of their time of diagnosis were considered for this study and the 20 breast cancer cases belonging to this locality were included in the study. According to the Family register in Arpookara Panchayat, there were 5692 families in Arpookara Panchayat with the population of 22942. From this Family register, Controls belonging to age group between 30–70 years from Arpookara panchatyat itself, were selected by simple random technique with the help of random number table. Before being enrolled in to the study, it is ensured that the controls were free from breast cancer. In our study for each case, four age matched controls free from breast cancer were selected from the same Panchayat hence 20 cases and 80 controls in the ratio of 1:4 were included in the study thus making a total of 100 subjects.

The questionnaire containing particulars with regard to socio demographic factors like age, marital status, education, occupation, family income, religion, menstrual and reproductive health history viz., age at menarche, age at first child birth, abortions, menopausal status, family history of breast cancer, previous history of benign tumours, oral contraceptive intake and awareness regarding symptoms of breast cancer and breast self examination, initially drafted in English, translated to Malayalam, the local language of the study participants and then pre–tested to assess the suitability with regard to duration, language appropriateness and question comprehensibility. With the help of the grass root level workers of the health care delivery system viz. Accredited Social Health Activitist (ASHA) and Anganwadi workers of the Panchayat, all the study participants were interviewed at their residence.

Analysis was done using SPSS version 16.0. Quantitative variables were compared using student t-test. Chi-square test was used to find significance of categorical variables. The risks were estimated by using odds ratio (OR) & their 95% confidence interval.

Age group of cases ranged from 32-70 years. The mean age of cases and controls were 49.7 and 49.6 years respectively [Table/Fig-1]. More than 95 % of study subjects were literate with at least 8 years of schooling. In the present study almost 72% of cases belong to lower socio–economic status [OR =14.47 (1.28 -162) p = 0.03].

[Table/Fig-2] shows that 55% of the cases attained menarche before 13 years of life which was significantly associated with increased risk of breast cancer [OR =3.2 (1.18 -8.8 ) (p=0.03). Marital status showed a statistically significant difference (p=0.001) between cases and controls suggesting that being single or unmarried as risk factor for breast cancer. 98% of study subjects delivered their first child before 30 years which reduces the risk of breast cancer [OR =0.302 (0.1 -0.4) p < 0.03]. 4 cases and 27 controls had abortion in the past. 13 cases (65%) and 39 controls (48.8%) had attained menopause. Out of that 40% of cases and 37.5% of controls attained menopause after 45 years. However the above variables were not statistically significant.

From the [Table/Fig-3], it is evident that 45% of cases and 18 % of controls reported to have positive family history of breast cancer. Increase in breast cancer risk was seen among cases with family history of breast cancer [OR =3.5 (1.27 -10.8) p < 0.02]. 30 % of cases had previous history of benign tumours which was highly significant [OR =8.14 (2.03 -32.62) p = 0.04]. 88% of study subjects had breast fed their child for more than two years with an average of 4.5 and 5.6 years for cases and controls. Breast feeding for less than two years or absence of breast feeding as in nulliparous women increased the risk of breast cancer [OR =2.28 (1. 48 -18.8) p = 0.01]. Out of 3% of the study subjects had a history of OCP intake in the past and 65% of cases were overweight and obese based on BMI and these two variables did not show any significance with breast cancer. 47% of study subjects had significantly less awareness about symptoms of breast cancer and self examination of the breast [OR =11.6 (1.2-112) p = 0.03].

And the [Table/Fig-4] shows the risk factors associated with breast cancer from the present study are early menarche, being single or unmarried, family history of breast cancer, previous history of benign breast tumours, breast feeding less than two years or absence of breast feeding and poor awareness of the symptoms of breast cancer. Giving birth to first child before 30 years was a protective factor against breast cancer in our study.

Breast cancer is associated with various risk factors. Despite having far advanced health indices among the states in India, the breast cancer is a public health problem in Kerala with increasing incidence. This study was carried out to identify the various risk factors associated with breast cancer in Arpookara Panchayat of Kottayam district in Kerala, as identifying the risk factors helps to plan the future course of action and framing the policies with regard to the prevention of the breast cancer.

Age is an important risk factor for breast cancer. The risk of breast cancer increases as the age progresses [7]. The median age at diagnosis for cancer of the breast in our study was 43 years, which is consistent with other studies done in India [8–10] and the median age at diagnosis for cancer breast was 61 years of age in the western world [11]. The reason for the above difference could be due to the various ethnic, socio-demographic, geographical and life style factors which needs further exploration.

We found that the early menarche before attaining 13 years was significantly associated with increased risk of breast cancer. Many studies have shown that breast cancer risk is more for women whose menarche occur at an early age, in India and other countries [12,13]. Early age at menarche is associated with increased risk of breast cancer and there appears to be a 20% decrease in breast cancer risk for each year if menarche is delayed [14,15]. Studies reported that women who began menstruating at an early age (before age 12) and those who reach menopause after age 55 years had an increased risk of breast cancer [16,17].

Marital status was found to be a significant risk factor for breast cancer in our study. The same finding that the single, unmarried women compared to married women had 4–5 fold higher risk for development of breast cancer in the age group of 40 years and above was noted in the studies done in India and outside India [17,18]. In most studies single and nulliparous married women were found to have a similar increased risk for cancer of the breast as compared with parous women of the same age [19] and early first full-term pregnancy might attribute to this protective effect.

In our study, delivering the first child before 30 years was found to be a protective factor against the breast cancer. The increased risk for breast cancer in those women who were delivering their first child after 30 years has been found in many Indian and western studies [20, 21]. There is another study which found that each live birth reduced life time risk of breast cancer by 7% [22]. In present study both the study groups had average of two children and parity was not accounted to be a risk for breast cancer.

Breast feeding for less than two years or absence of breast feeding as in nulliparous women was significantly associated with the risk of breast cancer. The protective effect of prolonged duration of breast feeding reducing the risk of breast cancer is well established in studies conducted in India and abroad [17, 23, 24].

A statistically significant association was found in those women having a family history of breast cancer and the risk of acquiring the same in our study and similar results were seen in other studies also [17, 25]. Also in the study, it was found that there existed a significant association between the risk of breast cancer and those who had history of benign breast tumours and this is in accordance with other studies [26].

Women in higher socio–economic status have an increased risk of breast cancer compared to those belonging to the low socioeconomic status. Socio–economic status per se does not increase the risk but other factors like delayed age at marriage and first child birth after 30 years may play a significant role [24, 27]. In our study almost 60% of cases belong to lower socio–economic status and this could also be due to the various socio-demographic, geographical and life style factors and needs further studies.

Despite high literacy status among women, our study revealed a statistically significant poor awareness regarding breast self examination and the symptoms of breast cancer.

There are studies suggesting increased risk of breast cancer with increased body mass index [28] but no statistical significance was observed in our study with regard to overweight and breast cancer. Similarly history of oral contraceptive pills intake, history of abortions and menopause showed no significant association with the breast cancer in the study subjects.

The study being a case control, the question of bias always arises. Though the selection bias was minimized with appropriate matching of the cases and controls, recall bias remained a challenge particularly when remembering the past events viz. Age at menarche, history of oral contraceptive pills intake, which might have influenced the outcome of these study variables.

Another point which has to be pondered is that the woman starts living in the husband’s place after the marriage. So the geographical area, diet pattern and radiation exposure if any, before marriage when she was residing in her matriarchal residence have to be kept in mind before arriving at a conclusion.

Higher incidence of breast cancer particularly in Kerala among the Indian states which is accompanied by poor awareness regarding self examination of the breast and symptoms of breast cancer brings forth the need for the following possible remedial actions.

Active steps should be taken to train and sensitize the medical and paramedical personnel about identifying symptoms of breast cancer in the primary care settings itself and periodical follow up should be given.

Vulnerable groups must be identified and appropriate preventive measures like awareness creation with regard to breast self examination and symptoms of breast cancer. Suitable IEC (Information, Education and Communication) activities must be formulated to sensitize the vulnerable population for seeking immediate health care as and when the symptoms arise.

Policy makers can consider encouraging community participation by involving the Non-Governmental Organizations, Women Self Help Groups and Public Private Partnerships with the support of the grass root level health workers in spreading the awareness of breast cancer.