Passive smoking a significant predictor of Polycystic Ovarian Syndrome: Logistic Regression based Evidence from Karachi, Pakistan.

Abstract

Objective: To determine the predictors of occurrence of Polycystic Ovarian Syndrome in females of Karachi using Logistic Regression model.

Methods: A descriptive cross-sectional study was conducted from (January-June 2022) after the approval of the Institutional Bioethical Committee. Following the pilot study, validated structured questionnaire was disseminated in person and online. Data obtained from 599 individuals was kept confidential and analyzed using SPSS version 23 for descriptive statistics including chi-square based cross tabulations to assess the relationship of predictors with existence and vulnerability to the syndrome. Subsequently, inferential analysis using logistic regression was used to predict the odds of having or being vulnerable to the syndrome.

Results: The existence of the syndrome was significantly associated with symptoms (p˂0.01), level of education (p˂0.05) , marital status (p˂0.001), house hold work frequency (p˂0.05), being tested for diagnosis (p˂0.001), father/brother smoking (p˂0.001), source of information (p˂0.05) and awareness of the syndrome in sex education (p˂0.05). Regression model revealed that irregularity of menses (p˂0.001, OR=3.25), hirsutism (p˂0.05, OR=2.08) and acne (p˂0.05, OR=1.74) significantly increased the likelihood of suffering from the syndrome. Females subjected to passive smoking were 2.46 times more likely to suffer from the syndrome (p˂0.05). Thus, exposure to passive smoking increases the vulnerability to having Polycystic Ovarian Syndrome.

Conclusion: The identification of variables involved in the occurrence of Polycystic Ovarian Syndrome will help to ascertain the etiology of newly diagnosed females. Passive smoking could be regarded as one of the factors in the etiology of Polycystic Ovarian Syndrome.