|Home | About | Journals | Submit | Contact Us | Français|
Breast cancer is the most common cancer in women. Its prevalence is increasing annually by 2%. The determination of modifiable risk factors has been the subject of various studies. The aim of this study was to determine risk factors of breast cancer in women in Golestan Province.
This case-control study was conducted among women with breast cancer recorded in the cancer registry system between 2004 and 2006 (n = 134), and their age-matched healthy neighbors (n = 133). Data were statistically analyzed.
Age at marriage, menarche and pregnancy, breast feeding, positive family history, marital status, and educational level were not significantly correlated with risk of breast cancer, but age at menopause (< 46.6 years) was significantly correlated (95% confidence interval 1.15–7.37; p = 0.021). Live births, still births, and infant deaths were not significantly different between the 2 groups. For other variables, such as smoking history, no odds ratio was calculated.
Results show that there is no significant correlation between variables and risk of breast cancer in our population, except for age at menopause. A large cohort study is recommended.
Breast cancer is the most prevalent cancer in women worldwide, with an annual increase in prevalence of about 2%. In 2000, the breast cancer incidence increased to 1 in 8 women. In the 7th decade of life, it will affect 1 in 13 American women, and in the 8th decade 1 in 11 women [1, 2]. There are significant variations between geographic areas with regard to breast cancer incidence . Over the next decade, as women adapt to western lifestyles by marrying and bearing children later in life, developing countries may face a potential breast cancer epidemic , and due to these lifestyle changes and the lack of resources to create mass awareness and provide early detection, the incidence is increasing across countries with lower health care facilities. Iran is amongst the medium prevalence countries, but its incidence and prevalence are increasing . Similar to other developing countries, the mean age-specific incidence of patients in Iran is 10–20 years less than other countries (approximately 47 years) . Iran is estimated to face a serious breast cancer burden in the coming years.
As the Golestan cancer registry shows, breast cancer is the 3rd most common cancer in our province, after esophageal and stomach cancers, with an age standardized rate of 13.5 per 100,000 . Considering the incidence and prevalence of breast cancer, the high cost of treatment, and the young age at disease onset, it is important to urge governments to aggressively schedule strategies for early detection such as providing mammography machines in health care facilities and hospitals [3, 7]. Several risk factors are considered for breast cancer some of which are confirmed, while others are controversial and may be rejected. In Iran, few studies about breast cancer risk factors exist. This study was designed to detect the risk factors for the Golestan Province, Northeast Iran.
This case-control study included all patients who were registered as having breast cancer between 2004 and 2006. Age-matched neighbors were used as controls. Names and addresses were derived from the cancer registry system of Golestan Province. The database contained 300 registered cases some of which were repeated and therefore deleted from the list, some addresses were incomplete, and others had moved to a different location, leaving a total of 134 cases. The population-based cancer registry gathers data relating to cancer patients of all pathologies from laboratories, hospitals, local health offices, clinics, and physician offices. However, despite strong recommendations by the coordinator, personnel who fill in the forms frequently fail to write down a complete and straight forward home address, which causes limitations for our kind of study. Furthermore, due to the lack of community-wide screening programs and no specific attention being paid to routine simple screening tests, patients are usually referred at high stages of cancer. We used trained personnel from the Golestan Health Network to complete valid and reliable questionnaires through interview. Patients who had passed away were excluded and no control was considered for them. One supervisor controlled all questionnaires of a particular district and posted the completed forms to the main office of the Health Deputy. The questionnaire included demo graphic data such as age, residency, ethnicity, marital status, profession, educational level, and risk factors which were discussed in previous studies such as age at menarche, menopause and first pregnancy, number of children, and number of abortions. The coded data were entered into a computer and analyzed using SPSS-13 (SPSS Inc., Chicago, IL, USA); statistical tests included chi square and independent t-test as well as non-parametric tests; 95% confidence interval and odds ratio (OR) were calculated.
Of the registered cases, 52 were dead and 162 could not be found for various reasons; 134 cases and 133 controls were enrolled. The mean age was 47.15 ± 10.36 years (cases) and 42.96 ± 11.93 years (controls). All demographic data and risk factors addressed in the questionnaire are reported in table table1.1. Those factors which had a p value of less than 0.2 were ineluded into the regression analysis. As shown in table table2,2, after univariate and multivariate regression analysis, the most important risk factor associated with risk of breast cancer was age at marriage. The mean age at marriage was significantly higher in cases than in controls, in both univariate and multivariate analysis. The mean age at menarche was not significantly different between cases and controls (p = 0.593) (table (table3).3). Results indicate that the mean age at menopause was less in the case group than in the control group (p = 0.04), but in multivariate analysis no significant relationship was seen. Mean age at first pregnancy was not significantly different between cases and controls (table (table3)3) (p = 0.77). No difference was seen between the 2 groups with regard to live births, live children, history of still births, and death of children aged < 2 years (p > 0.05). No significant correlation was seen between the risk of breast cancer and marriage status (p = 0.702), educational level less than diploma (p = 0.594), history of breast cancer in second-degree relatives (p = 0.37), and breast feeding. ORs were not calculated for smoking and other diseases due to the low amount of positive cases.
Results show that the mean age of patients was 47.15 ± 10.36 years, which is very similar to other studies in Iran and other countries [8, 9, 10, 11, 12, 13]. In our study, there was no significant correlation between age at marriage and risk of breast cancer. In another study carried out in Shiraz, a significant rise in the risk of breast cancer was seen when marriage occurred after 25 years of age . Maybe sexual activity and its effect on sex hormone secretion play a role here. In the present study, significant correlation was seen between breast cancer and menarche occurring before 13.37 years of age and menopause before 46.6 years of age [8, 14]. However, another study showed an OR = 0.72 for menarche after 17 years of age , and others even showed a protective role of late-age menarche . It has been said that a lower age at menarche and a higher age at menopause can influence the risk of developing breast cancer [17, 18]. Maybe the fact that hormone replacement therapy is not used in our area plays a role in the present results [2, 19]. The results also show an increase in the risk of breast cancer in women who deliver their first child at age less than 20.51 years (not significant). Numerous studies indicated that age at first delivery can be a risk factor [8, 14, 16, 17, 20]. Also, nulliparity and the total number of children have been reported as risk factors [17, 21, 22] but we did not find this in the present study. Others showed a protective effect of parity and breast feeding with a 7% decrease in breast cancer risk seen for every 12 months of breast feeding . In the present study, breastfeeding for a total of > 64.35 months had an OR = 0.73. A protective effect of breastfeeding was reported before [22, 23, 24, 25, 15, 18], although others reported no protective effect . In our area, no significant correlation was seen between risk of breast cancer and marriage. This was similar to other studies [20, 14, 27], but some authors have reported a higher risk for married women . Some interfering variables like oral contraceptives can be included here. No significant relationship was seen with regard to lower educational levels and family history of breast cancer, but the latter was reported to be a risk factor in other reports [8, 9, 18]. Here, we only saw a significant correlation between risk of breast cancer and lower age at menopause (which was not confirmed in multivariate analysis).
Breast cancer development is a multifactorial process which includes a complex interaction between physical activity, diet, energy intake, and body weight. This suggests that continual updating of knowledge on primary prevention and identification of behavioral changes will reduce the risk . Thus, prevention strategies in our area could include avoidance of modifiable risk factors such as late first pregnancy, absence or short duration of breast feeding, and use of hormone replacement therapy in postmenopausal women which is encouraged in some studies. Our findings have highlighted the heterogeneity of breast cancer etiology across populations and indicated the need for further studies. The most interesting result of this study was the significant correlation between age at marriage and risk of breast cancer, which could be a modifiable risk factor, especially in our area. Also, treatment of infertility, planning for earlier detection of cancer in those with a positive history of breast cancer in first-degree relatives or in persons with a medical history of other cancers or benign lesions of the breast could be modifiable risk factors applicable for this area and potentially helpful in the quest to change the pattern of breast cancer.
There was no conflict of interest.
The authors thank all colleagues in the health care facilities of Golestan Province who helped gather data, and the Deputy of Research at Golestan University of Medical Sciences who funded the project.