Tag Archives: Logistic Models

Statistical models which describe the relationship between a qualitative dependent variable (that is, one which can take only certain discrete values, such as the presence or absence of a disease) and an independent variable. A common application is in epidemiology for estimating an individual’s risk (probability of a disease) as a function of a given risk factor.

Episiotomy and obstetric outcomes among women living with type 3 female genital mutilation: a secondary analysis

Reprod Health. 2016 Oct 10;13(1):131.FREE

Episiotomy and obstetric outcomes among women living with type 3 female genital mutilation: a secondary analysis.

Rodriguez MI, Seuc A, Say L, Hindin MJ

BACKGROUND: To investigate the association between type of episiotomy and obstetric outcomes among 6,187 women with type 3 Female Genital Mutilation (FGM).

METHODS: We conducted a secondary analysis of women presenting in labor to 28 obstetric centres in Burkina Faso, Ghana, Kenya, Nigeria, Senegal and Sudan between November 2001 and March 2003. Data were analysed using cross tabulations and multivariable logistic regression to determine if type of episiotomy by FGM classification had a significant impact on key maternal outcomes. Our main outcome measures were anal sphincter tears, intrapartum blood loss requiring an intervention, and postpartum haemorrhage.

RESULTS: Type of episiotomy performed varied significantly by FGM status. Among women without FGM, the most common type of episiotomy performed was posterior lateral (25.4 %). The prevalence of the most extensive type of episiotomy, anterior and posterior lateral episiotomy increased with type of FGM. Among women without FGM, 0.4 % had this type of episiotomy. This increased to 0.6 % for women with FGM Types 1, 2 or 4 and to 54.6 % of all women delivering vaginally with FGM Type 3. After adjustment, women with an anterior episiotomy, (AOR = 0.15 95 %; CI 0.06-0.40); posterior lateral episiotomy (AOR = 0.68 95 %; CI 0.50-0.94) or both anterior and posterior lateral episiotomies performed concurrently (AOR = 0.21 95 % CI 0.12-0.36) were all significantly less likely to have anal sphincter tears compared to women without episiotomies. Women with anterior episiotomy (AOR = 0.08; 95%CI 0.02-0.24), posterior lateral episiotomy (AOR = 0.17 95 %; CI 0.05-0.52) and the combination of the two (AOR = 0.04 95 % CI 0.01-0.11) were significantly less likely to have postpartum haemorrhage compared with women who had no episiotomy.

CONCLUSIONS: Among women living with FGM Type 3, episiotomies were protective against anal sphincter tears and postpartum haemorrhage. Further clinical and research is needed to guide clinical practice of when episiotomies should be performed.

This article can be accessed in this LINK

Determinants of Elongation of the Labia Minora in Tete Province, Central Mozambique Findings of a Household Survey

African Journal of Reproductive Health. 2016; 20(2): 111-121.LMEFREE

Determinants of elongation of the labia minora in Tete Province, Central Mozambique: Findings of a household survey

Martínez Pérez G, Bagnol B, Chersich M, Mariano E, Mbofana F, Hull T, Martin Hilber A


A WHO-supported provincial-level population-based survey was conducted in 2007 to understand the determinants and implications forhealth of vaginal practices. A total of 919 women aged 18-60 were selected randomly for enrolment. This is the first population-based study of females in Tete Province, Mozambique. At some time over their lives, 98.8% of women had practiced elongation of their labia minora and a quarter (24.0%) had done so in the past month. Currently practicing women were more likely to have engaged in sex recently, and used contraceptives and condoms at last sex than women who had stopped labial elongation. Younger age, residence in rural areas and having two or more male partners were also determinants of current practice. Women commonly reported they practiced for no specific reason (62.8%). Discomforting itchiness and lower abdominal pain were more frequent in women who had stopped labial elongation than in women who were currently practicing. Although women may not report current vaginal ill health, it is possible that prospective cohort studies could uncover alterations in genital vaginal flora or other indicators of impact on women’s health. The findings of this study do not suggest that labial elongation is linked with high-risk behaviors for HIV transmission. .

This article can be accessed in this LINK

Women’s empowerment and the intention to continue the practice of female genital cutting in Egypt.

Arch Iran Med. 2009 Mar;12(2):154-60. FREE

Women’s empowerment and the intention to continue the practice of female genital cutting in Egypt.

Afifi M.

Department of Primary Health Care, MOH (HQ), P.O. Box 1853, Dubai, UAE. afifidr@yahoo.co.uk

BACKGROUND: The study aimed to (dis)prove the association of the level of women’s empowerment with their future intention to perpetuate female genital cutting for their daughters.

METHODS: In a national representative community-based sample of 14,393 currently-married women in Egypt, the level of empowerment, intention to continue the practice, and other socio- demographic variables were collected in the 2000 Egypt Demographic and Health Survey. Secondary in-depth analysis was conducted on data downloaded from MEASURE Demographic Health Surveys (MEASURE DHS) website.

RESULTS: About 14% of the women intended to discontinue the practice. Twenty-six percent of the women were empowered in all household decisions. Levels of women’s empowerment adjusted for age, residence, education, interaction between empowerment and education, work status, and female genital cutting status of currently-married women were entered in six logistic regression models in a sequential way.

CONCLUSION: In the last model, those of high levels of empowerment and education were 8.06 times more likely not intending to perpetuate female genital cutting for their daughters than low- empowered low-educated women.

This article can be accessed for free in this LINK

Female genital cutting and mother’s age at birth are associated with the sex of offspring in Africa.

Biodemography Soc Biol. 2008 Fall;54(2):141-51.

Female genital cutting and mother’s age at birth are associated with the sex of offspring in Africa.

Martina JF, Williams RC.

School of Human Evolution and Social Change, Arizona State University, Tempe, AZ 85287-2402, USA. john.martin@asu.edu


Female genital cutting (FGC) is widespread in Africa. Its short and long term medical consequences are widely reported. Here we report that FGC is associated with and may contribute to increases in the sex ratio at birth (Odds Ratio = 1.019; 95% C.I. = 1.007, 1.032) while mother’s age at birth is inversely associated with probability of male birth (Odds Ratio = 0.998; 95% C.I. = 0.997, 0.999) in a generalized linear, logistic model with the probability of a male birth the dependent variable and eight potential explanatory variables applied to 413,384 births in 22 African countries. While these two associations are statistically significant, their odds ratios are close to 1.0 and they only slightly decrease the null deviance of the model. Therefore sex determination in these data is still a mostly stochastic process determined by the random variation in X or Y sperm production, transport, and conception.

This article can be purchased in this LINK

The effects of female genital mutilation on the onset of sexual activity and marriage in Guinea.

Arch Sex Behav. 2009 Apr;38(2):178-85. Epub 2007 Oct 18.

The effects of female genital mutilation on the onset of sexual activity and marriage in Guinea.

Van Rossem R, Gage AJ. Vakgroep Sociologie, Universiteit Gent, Korte Meer 3-5, 9000, Ghent, Belgium. ronan.vanrossem@ugent.be


Female genital mutilation (FGM) is almost universal in Guinea and practiced by all ethnic and religious groups and social classes, although the prevalence of the various types of FGM varies by socioeconomic group. A common explanation for FGM practices is that they contribute to the social control over female sexuality and enhance the marriageability of women. These claims were tested using the 1999 Guinea Demographic and Health Survey (DHS) (N = 6753). Event history techniques were used to examine the effect of type of FGM on the age at first sex and the age at first marriage and logistic regression for the effect of FGM on premarital sex. The results showed that the type of FGM had a significant zero-order effect on the age at first marriage and the prevalence of premarital sex, but not on the age at first sex. However, these effects became non-significant once controls for age, religion, ethnicity, education, residence, and wealth were added to the model. Variations in sexual behavior, therefore, were unrelated to type of FGM, but reflected differences in the social characteristics of the participants.

This article can be viewed online in this LINK