Determinants of Tubal Ligation in Puebla, Mexico

Abstract

Tubal ligation provides an effective and reliable method by which women can choose to limit the number of children they will bear. However, because of the irreversibility of the procedure and other potential disadvantages, it is important to understand factors associated with women's choice of this method of birth control. Between May 1999 and August 2000, data were collected from 755 women aged 40 to 60 years from a cross-section of neighborhoods of varying socio-economic make-up in Puebla, Mexico, finding a tubal ligation rate of 42.2%. Multiple logistic regression models were utilized to examine demographic, socio-economic, and reproductive history characteristics in relation to women's choice of tubal ligation. Regression analyses were repeated with participants grouped by age to determine how the timing of availability of tubal ligation related to the decision to undergo the procedure. The results of this study suggest that younger age, more education, use of some forms of birth control, and increased parity were associated with women's decisions to undergo tubal ligation. The statistically significant difference of greater tubal ligation and lower hysterectomy rates across age groups reflect increased access to tubal ligation in Mexico from the early 1970s, supporting the idea that women's choice of tubal ligation was related to access.

KEYWORDS:

• contraception
• reproductive health
• risk factors
• Mexico

Notes

This study was funded by NSF Grant # 9805299.

Notes. ** = p < .01;

* = p < .05 in two-tailed t-test for continuous variables, and in chi-square test for categorical variables.

Notes. ** = p < .01;

* p < .05. Excluded categories in parentheses. ¹Multivariate logistic regression, adjusting for other factors shown in the model.

Notes. ** = p < .01;

* p < .05. Excluded categories in parentheses. ¹Multivariate logistic regression, adjusting for other factors shown in the model.

1. All statistical analyses were performed in Stata 11.1.

2. Age at first birth was not included in the model; missing values reduced the number of cases with a bias against younger women, and the effect was not significant.