Exploring the triggers of premature and early menopause in India: a comprehensive analysis based on National Family Health Survey, 2019-2021.

The study utilizes the latest round of National Family Health Survey, conducted during 2019–2021. The first (1993–1993), second (1998–1999), third (2005–2006), fourth (2015–2016) and fifth (2019–2021) round of National Family Health Survey (NFHS) for analysing the overall trend and the recent round for the rest of the study. The National family health survey (NFHS) is a large-scale, multi-round survey conducted in representative sample of household throughout India. The NFHS is a collaborative project of the International Institute for Population Sciences (IIPS), Mumbai, India; ORC Macro, Calverton, Maryland, USA and the East–West center, Honolulu, Hawaii, USA. The Ministry of Health and Family Welfare (MoHFW), Government of India, designated IIPS as nodal agency, responsible for providing coordination and technical guidance for the NFHS. The fifth round of NFHS was conducted in 2019–2021, with the interruption of Covid-19 pandemic in between. The survey was carried out for 707 districts, as of 31st March 2017. Due to the pandemic lockdown, the fieldwork was in two phases- Phase I covered 17 states and union territories from 17th June 2019 to 30th January 2020; and Phase II covered 11 states and union territories from 2nd January 2020 to 30th April 2021. There were 17 field agencies who collected information from 636,699 households, 724,115 women and 101,839 men 28 . The pregnant women and lactating mothers during the time of survey are excluded from the analysis. Women who had undergone hysterectomy, that is, having surgical menopause are also excluded. Two separate datasets are created for the analysis and the sample sizes for analysis of premature menopause is 429,446 (ages 15–39), and early menopause is 79,643 (ages 40–44). The women who had their last menstrual cycle one or more year ago is considered to be menopausal women, and those who had their last menstrual cycle beyond 12 months below the age of 40, is identified as prematurely menopausal women , while menopause between 40 and less than 45 years as early menopause . Both are dichotomized as ‘1’ and ‘0’. The socio-economic and demographic factors considered in the study to observe the patterns of premature and early menopause across the population subgroups. The educational level of women is categorized as no education, primary, secondary and higher education. Caste is grouped into four categories: scheduled caste (SC), scheduled tribe (ST), other backward class (OBC), and others. Religion is categorized as Hindu, Muslims and Others (including Sikh, Buddhist/Neo-Buddhist, Jain, Jewish, Parsi/Zoroastrian, no religion, and other). Place of residence is given as rural and urban in the survey. In the survey, a household wealth index was calculated by combining household amenities, assets, and durables and categorizing households in a range from poorest to richest, corresponding to wealth quintiles from lowest to highest. The working status of the woman are taken as employed and not employed. Marital status is categorized as- never married, currently married, and widowed/divorced/separated. The states and union territories are recoded according to regions as- North (Chandigarh, Delhi, Haryana, Himachal Pradesh, Jammu and Kashmir, Punjab, Rajasthan); North East (Assam, Arunachal Pradesh, Manipur, Meghalaya, Mizoram, Nagaland and Tripura); Central (Chhattisgarh, Madhya Pradesh, Uttarakhand Uttar Pradesh); East (Bihar, Jharkhand, Odisha, West Bengal); West (Dadra and Nagar Haveli and Daman and Diu, Goa, Gujarat, Maharashtra); South (Andhra Pradesh, Karnataka, Kerala, Puducherry, Tamil Nadu, Telangana, Andaman and Nicobar Islands, Lakshadweep) 29 . The lifestyle behavior were considered for observing the effect of it on the occurrence of premature and early menopause. Consumption of tobacco is considered from the survey as smoking categorized as no and yes. Regular drinking of alcohol is also dichotomized as no and yes. Unhealthy diet pattern is indicated by regular consumption of fried foods and aerated drinks, again grouped as no and yes. The biological or reproductive factors were considered to observe the plausible determining factors of premature and early menopause. The age at menarche is the age of onset of menstrual cycle, and is grouped as ‘12 or less’, ‘13–15’, and ‘more than 15’. Women with no children ever born, that is, with zero parity were considered as nulliparous and the variable is dichotomized. Age at first birth provided details about the respondents' ages at the time of the birth of their first child. These categories were ‘below 18 years’, ‘18–24 years’, and ‘25 + years’ for this variable. The usage of contraceptive methods to delay or prevent pregnancy was a question that respondents were asked. Injectables, pills, and emergency contraception were all regarded in this study as hormonal contraceptives. Menstrual hygiene was categorized as—hygienic methods (sanitary napkins, locally prepared napkins, tampons, menstrual cup), unhygienic methods (cloths, nothing, others) and both. Termination of pregnancy was dichotomized as yes or no. The respondent's measured mass (weight) and height at the time of the interview were used to calculate body mass index (BMI). The SECA 874 U digital scale was used to measure weight, while the SECA 213 stadiometer was used to measure height. Underweight (18.5 kg/m2), normal (18.5–24.9 kg/m2), and obese/overweight (25 kg/m2) were the outcomes of dividing body weight by square of body height, or BMI. During the survey, blood samples from the respondents were taken for anaemia testing. In this study, haemoglobin levels below 12 g/dL were deemed anaemic, and those above 12 g/dL were deemed not to be anaemic. A finger-stick blood sample was used to measure random blood glucose using an Accu-Chek Performa glucometer and glucose test strips. Diabetics were defined as those with a random blood glucose level of 200 mg/dl or higher. The study variables are described using descriptive statistics at first. Bivariate analysis including cross-tabulation and chi-square tests are done to observe significant associations between premature/early menopause and the correlates. Estimates of the hazard ratio (HR) for the explanatory variables that are expected to influence menopausal state are obtained using Cox proportional hazard regression models. In the survival analysis, the median age (median survival time) was the age at which 50% of the women were anticipated to still be in their reproductive years. A survival technique is chosen over alternative models, such as logit analysis, because it appropriately handles censored observations, which are a feature of this data set, and takes into consideration the duration from menarche to the event of interest, that is, menopause. The assumption of proportional hazards is tested to eliminate bias in the estimates of HRs by looking at plots of the Schoenfeld residuals and by using time-dependent variables. When computing the Cox partial likelihood using the Efron technique 30 , modifications are performed to account for linked observations (several women reaching menopause at the same age). Cox-proportionate hazard model will be as follows: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h\left( {t|X} \right) = h_{0} \left( t \right){\text{exp}}\left( {\beta \cdot X} \right)$$\end{document} h t | X = h 0 t exp β · X where t i  represents the survival time (premature or early menopause age) for the ith individual. h(t)  is the hazard function determined by a set of covariates. X i is the vector of covariates for the i th individual. β is the vector of model coefficients that measure the impact of covariates. The term  h(0)  is called the baseline hazard, that corresponds to the value of the hazard if all the x i ’s are equal to zero. The likelihood function of the model is given by \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L\left( \beta \right) = \mathop \prod \limits_{i = 1}^{n} \left[ {\frac{{e^{{\left( {\beta_{i} \cdot X_{i} } \right)}} }}{{\mathop \sum \nolimits_{{j \in R\left( {t_{i} } \right)}} e^{{\left( {\beta_{j} \cdot X_{j} } \right)}} }}} \right]^{{C_{i} }}$$\end{document} L β = ∏ i = 1 n e β i · X i ∑ j ∈ R t i e β j · X j C i where n is the sampled women. R(t i ) is set of women who are at risk of premature or early menopause at time t i , for the i th individual. C i indicates censored observations for the i th individual. With respect to the parameter vector β , the likelihood function aims to maximise the probability of observing the data. The model coefficients are usually estimated by partial likelihood estimation techniques, which take into account just those who witness the event or are censored at the time t i . The exponential of the associated coefficient represents the hazard ratio (HR) for a predictor variable: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$HR = e^{{\beta_{j} }}$$\end{document} H R = e β j Hazard ratios are interpretated as- HR = 1: No effect. HR  1: Increase in Hazard. A total of four models were fitted with the covariates taken in the following manner: Model 1 Socio-economic and demographic factors. Model 2 Socio-economic and demographic factors + Lifestyle behaviour. Model 3 Socio-economic and demographic factors + Lifestyle behaviour + Reproductive factors. Model 4 Socio-economic and demographic factors + Lifestyle behaviour + Reproductive factors + Anthropometric and Bio-chemical factors. The analysis is based on secondary data available in public domain for research; thus, no approval was required from any institutional review board (IRB). The survey agencies had conducted the field work with prior consent from the respondents. The NFHS survey was conducted in accordance with the relevant ethical guidelines and regulations.

The sample of age 15–39 consists of 2.23% of premature menopausal women. Around 34% of the women resided in urban areas and 66% in rural areas. About 15.15% of the sample women did not have schooling/education. Majority of the sample belonged to Hindu religion (81.17%) and 26% belonged to scheduled caste or scheduled tribe. Around 37% women belong to poor wealth index. About 57.7% of the samples were currently married, 39.5% never married and 2.78% were widowed/divorced/separated. The age at menarche is 13 to 15 years for 79% of these women. Around 48% of the sample is nulliparous. The age at first birth is 18 to 24 years for 66% of the women. There are 8.8% and 10.3% women who use hormonal contraceptives and had terminated pregnancies, respectively. According to the body mass index of the sample, 21.7% are underweight, 15.3% are overweight and 5.2% are obese. There are 53.2% women in the age-group 15–39 who are anemic and 0.8% diabetic (Table 1 ). Table 1 Percentage distribution of the sample by the study variables for the sample. Aged 15–39 Aged 40–44 n % n % Premature Menopause  No 4,19,871 97.77  Yes 9575 2.23 Early Menopause  No 66,740 83.8  Yes 12,903 16.2 Current age  15–19 1,14,462 26.65  20–24 73,360 17.08  25–29 70,847 16.5  30–34 80,145 18.66  35–39 90,632 21.1  40 24,095 30.25  41 12,216 15.34  42 16,783 21.07  43 13,754 17.27  44 12,796 16.07 Residence  Urban 1,45,467 33.87 28,013 35.17  Rural 2,83,979 66.13 51,630 64.83 Education  No education 65,045 15.15 31,880 40.03  Primary 43,866 10.21 12,092 15.18  Secondary 2,40,817 56.08 28,842 36.21  Higher 79,718 18.56 6829 8.58 Employment  Not employed 48,484 75.1 7594 64.13  Employed 16,074 24.9 4249 35.87 Wealth index  Poorest 74,532 17.36 13,166 16.53  Poorer 84,830 19.75 15,216 19.11  Middle 89,070 20.74 16,249 20.4  Richer 91,896 21.4 17,006 21.35  Richest 89,118 20.75 18,006 22.61 Caste  SC 93,646 21.81 16,061 20.17  ST 39,408 9.18 6990 8.78  OBC 1,84,841 43.04 34,587 43.43  Others 1,11,551 25.98 22,005 27.63 Religion  Hindu 3,48,561 81.17 65,636 82.41  Muslim 58,433 13.61 9402 11.81  Others 22,452 5.23 4605 5.78 Marital status  Never married 1,69,588 39.49 1010 1.27  Currently married 247,919 57.73 71,663 89.98  Widowed/divorced/separated 11,939 2.78 6970 8.75 Regions  Northern 77,900 18.14 15,917 19.99  Western 144,495 33.65 22,952 28.82  Southern 72,503 16.88 13,962 17.53  Eastern 4215 0.98 793 1.00  Central 40,717 9.48 7791 9.78  North-Eastern 89,616 20.87 18,229 22.89 Smoking  No 4,26,223 99.25 78,190 98.18  Yes 3223 0.75 1453 1.82 Drinking  No 4,26,841 99.39 78,682 98.79  Yes 2605 0.61 961 1.21 Regular consumption of fried foods and aerated drinks  No 2,23,609 52.07 44,009 55.26  Yes 2,05,837 47.93 35,634 44.74 Age at Menarche  12 or less 32,497 17.17  13–15 1,49,363 78.93  More than 15 7371 3.9 Nulliparity  No 2,24,481 52.27 76,578 96.15  Yes 2,04,965 47.73 3065 3.85 Menstrual hygiene  Hygienic methods 99,120 52.45  Unhygienic methods 37,902 20.06  Both 51,948 27.49 Age at first birth  Below 18 49,290 22.71 16,651 21.76  18–24 1,43,887 66.3 47,735 62.39  25+ 23,846 10.99 12,120 15.84 Hormonal contraceptive use  No 3,91,326 91.12 69,570 87.35  Yes 38,120 8.88 10,073 12.65 Had terminated pregnancy  No 3,85,104 89.67 66,734 83.79  Yes 44,342 10.33 12,909 16.21 BMI category  Underweight 89,902 21.77 7146 9.25  Normal weight 2,38,138 57.66 41,509 53.75  Overweight 63,371 15.34 20,130 26.06  Obese 21,616 5.23 8446 10.94 Anemic  No 2,01,075 46.82 37,021 46.48  Yes 2,28,371 53.18 42,622 53.52 Glucose level  Non-diabetic 4,12,942 99.22 74,465 96.69  Diabetic 3238 0.78 2549 3.31 N 4,16,180 77,014 Percentage distribution of the sample by the study variables for the sample. Coming to the sample of 40 to 44 aged women, 16.2% of them attained early menopause. The socio-economic and demographic variable distribution are almost similar to the former sample, however here 40% women had no schooling/education. About 89.9% of the samples were currently married, 1.27% never married and 8.75% were widowed/divorced/separated. Around 3.8% of the sample is nulliparous. The age at first birth is 18 to 24 years for 62.4% of the women. There are 12.6% and 16.2% women who use hormonal contraceptives and had terminated pregnancies, respectively. According to the body mass index of the sample, 9.3% are underweight, 26.1% are overweight and 10.9% are obese. There are 3.3% women in the age-group 15–39 who are diabetic (Table 1 ). The prevalence of premature and early menopause has decreased gradually over time. Premature menopause was highest during 1998–1999, 3.4% and reduced to 2.76% in 2005–2006 and remained almost constant in the next five years also, and further reduced to 2.23% in recent years. Early menopause was around 21% during the 90 s, increased to 24% in 2005–06, then reduced to 17.5% in the next five years, to 16% in recent years (Fig.  1 ). Figure 1 Trends in prevalence of premature and early menopause from 1992 to 2021. Trends in prevalence of premature and early menopause from 1992 to 2021. The bivariate analysis results for premature menopause sample aged 15–39 are displayed in Table 2 . Premature menopause is highest in the age-group 35–39, with 6.1%. Women residing in rural areas (2.6%), has no education (6.1%), employed (2.7%), belonging to poor wealth quintile (2.8%), and other backward class (2.5%), has significantly higher percentage of experiencing premature menopause. Marital status showed and significant association with premature menopause, with around 4.1% prevalence among those who were widowed/divorced/separated. Regional prevalences showed that Northern (2.65%) and Western (2.63%) regions had relatively higher proportions of premature menopause. Table 2 Cross-tabulations and chi-square tests of premature menopause with the possible correlates (age group 15–39). Premature Menopause (%) χ 2 value DF p -value Age-group  15–19 0.42 9.00E+03 4  < 0.001  20–24 0.55  25–29 1.07  30–34 2.96  35–39 6.13 Residence  Urban 1.5 244.05 1  < 0.001  Rural 2.6 Education  No education 6.12 5.40E+03 3  < 0.001  Primary 3.45  Secondary 1.5  Higher 0.6 Employment  Not employed 1.91 36.25 1  < 0.001  Employed 2.7 Wealth index  Poorest 2.86 413.08 4  < 0.001  Poorer 2.87  Middle 2.45  Richer 1.99  Richest 1.12 Caste  SC 2.24 235.75 3  < 0.001  ST 1.99  OBC 2.46  Others 1.93 Religion  Hindu 2.31 170.82 2  < 0.001  Muslim 2  Others 1.63 Marital status  Never married 0.43 4.20E+03 2  < 0.001  Married 3.37  Widowed/divorced/Separated 4.09 Region  Northern 2.65 304.10 5  < 0.001  Western 2.63  Southern 2.19  Eastern 1.27  Central 1.36  North-Eastern 1.69 Smoking  No 2.22 24.54 1  < 0.001  Yes 4.05 Drinking  No 2.22 40.37 1  < 0.001  Yes 4.1 Regular consumption of fried foods and aerated drinks  No 2.12 36.04 1  < 0.001  Yes 2.27 Age at Menarche  12 or less 1.06 88.31 2  < 0.001  13–15 0.44  More than 15 0.5 Nulliparity  No 3.76 4.90E+03 1  < 0.001  Yes 0.56 Menstrual hygiene  Hygienic methods 0.4 65.65 2  < 0.001  Unhygienic methods 0.76  Both 0.4 Age at first birth  Below 18 6.25 1.20E+03 2  < 0.001  18–24 3.26  25+ 1.57 Hormonal contraceptive use  No 2.26 8.57 1  < 0.001  Yes 1.95 Had terminated pregnancy  No 2.08 416.78 1  < 0.001  Yes 3.53 BMI category  Underweight 1.64 563.24 3  < 0.001  Normal weight 2.11  Overweight 3.21  Obese 3.52 Anaemic  No 2.5 119.67 1  < 0.001  Yes 1.99 Glucose level  Non-diabetic 2.23 148.33 1  < 0.001  Diabetic 5.09 Cross-tabulations and chi-square tests of premature menopause with the possible correlates (age group 15–39). Tobacco (4%), alcohol (4.1%) and regular fried food (2.3%) consumption are observed to be significantly associated with premature menopause. The women whose age at menarche is more than 15 years, have higher percentage of premature menopause (1.1%). Nulliparity was not observed to be a significant correlate. Premature menopause was comparatively higher among those who used unhygienic methods of menstruation (0.76%). Those whose age at first birth was below 18 years of age, premature menopause was higher among them (6.3%). There is also an observed significant association of premature menopause with hormonal contraceptive use (1.9%) and termination of pregnancy (3.5%). Overweight (3.2%) and obese (3.5%) women are observed to experience premature menopause in comparison to lower BMI categories. Premature menopause is significantly higher among diabetic women (5.1%) (Table 2 ). For the sampled age-group 40–44 years for measuring early menopause, the bivariate analysis is displayed in Table 3 . Menopause evidently increases with increasing age, and majority, around 23% attained at the age 44. The patterns are quite similar as premature menopause, where women residing in rural areas (18.1%), has no education (21%), belonging to poor wealth quintile (19%), and scheduled caste (17.4%), has significantly higher percentage of experiencing early menopause. Marital status also showed and significant association with early menopause, with around 20% prevalence among those who were widowed/divorced/separated. Western regions showed significantly higher proportion of early menopausal women (18.89%). Table 3 Cross-tabulations and chi-square tests of early menopause with the possible correlates (age group 40–44). Early Menopause (%) χ 2 value DF p -value Age  40 12.65 819.38 4  < 0.001  41 12.83  42 16.57  43 18.69  44 22.94 Residence  Urban 12.76 144.89 1  < 0.001  Rural 18.07 Education  No education 20.91 977.24 3   0.001  Employed 16.49 Wealth index  Poorest 18.89 298.02 4  < 0.001  Poorer 19.04  Middle 17.98  Richer 15.24  Richest 11.14 Caste  SC 17.42 170.73 3  < 0.001  ST 15.93  OBC 16.81  Others 14.44 Religion  Hindu 16.36 185.85 2  < 0.001  Muslim 16.52  Others 13.34 Marital status  Never married 12.39 113.04 2  < 0.001  Married 15.87  Widowed/divorced/Separated 20.13 Regions  Northern 15.56 261.38 5  < 0.001  Western 18.89  Southern 16.01  Eastern 13.25  Central 14.65  North-Eastern 14.31 Smoking  No 16.1 5.16 1  < 0.05  Yes 21.71 Drinking  No 16.15 3.94 1  < 0.05  Yes 20.06 Regular consumption of fried foods and aerated drinks  No 15.79 14.85 1   0.001  Yes 17.44 Age at first birth  Below 18 24.16 1.30E + 03 2  < 0.001  18–24 15.36  25+ 8.27 Hormonal contraceptive use  No 16.72 118.4 1  < 0.001  Yes 12.64 Had terminated pregnancy  No 16.06 9.84 1  < 0.001  Yes 16.91 BMI category  Underweight 20.44 125.93 3  < 0.001  Normal weight 15.89  Overweight 15.92  Obese 15.87 Anemic  No 18.29 124.42 1  < 0.001  Yes 14.38 Glucose level  Non-diabetic 16.05 31.94 1  < 0.001  Diabetic 21.31 Cross-tabulations and chi-square tests of early menopause with the possible correlates (age group 40–44). Consumption of tobacco (21.7%), alcohol (20.1%) and regular fried food (16.4%) are found to be significantly associated with early menopause. Early menopause is higher among those whose age at first birth was below 18 years of age (24.1%). There is also an observed significant association of early menopause with hormonal contraceptive use (12.6%) and termination of pregnancy (16.9%). On the contrary to premature menopause, early menopause is higher among underweight women (20.4%). Early menopause is significantly higher among diabetic women (21.3%) (Table 3 ). Results from Survival models: The multiple cox-proportional hazard models taking premature and early menopause as the event is displayed in Table 4 and Table S1 . The model 3 and 4, is observed to be predicting premature menopause as an event the best. The results are reported for the model 4, where all the possible explanatory variables are adjusted. Table 4 Adjusted hazard ratios and its confidence intervals, for observing the effect of covariates on the occurrence of premature menopause. Model 1 Model 2 Model 3 Model 4 Residence  Urban ®  Rural 1.016 (0.875–1.179) 1.023 (0.881–1.188) 0.613 (0.398–1.042) 0.672 (0.427–1.060) No education ®  Primary 2.673 (0.689–10.37) 1.785 (0.425–7.494) 0.734*** (0.621–0.868) 0.733*** (0.620–0.867)  Secondary 3.103* (0.960–10.03) 2.673 (0.822–8.690) 0.676*** (0.589–0.775) 0.672*** (0.586–0.770)  Higher 1.580 (0.445–5.618) 1.408 (0.391–5.073) 0.413*** (0.313–0.545) 0.410*** (0.310–0.541) Employment  Not employed ®  Employed 0.879 (0.780–1.190) 0.878 (0.779–1.189) 1.188** (1.073–1.371) 1.136*** (1.032–1.217) Wealth index  Poorest ®  Poorer 1.134 (0.964–1.335) 1.134 (0.964–1.335) 0.832 (0.476–1.454) 0.917 (0.513–1.641)  Middle 1.100 (0.926–1.305) 1.102 (0.928–1.309) 0.985 (0.559–1.736) 1.093 (0.602–1.984)  Richer 0.914 (0.750–1.113) 0.919 (0.753–1.121) 0.844 (0.450–1.583) 0.865 (0.441–1.694)  Richest 0.579 (0.268–1.253) 0.716 (0.322–1.592) 0.623*** (0.486–0.798) 0.632*** (0.492–0.812) Caste  SC ®  ST 0.997 (0.825–1.205) 0.986 (0.814–1.193) 0.748 (0.418–1.341) 0.681 (0.371–1.247)  OBC 0.847 (0.685–1.046) 0.839 (0.677–1.040) 0.671 (0.350–1.288) 0.615 (0.313–1.209)  Others 0.873 (0.546–1.395) 0.837 (0.514–1.365) 0.805 (0.520–1.521) 0.802 (1.117–1.517) Religion  Hindu ®  Muslim 0.934 (0.781–1.117) 0.911 (0.758–1.096) 1.233 (0.737–2.063) 1.188 (0.683–2.068)  Others 0.841 (0.678–1.043) 0.838 (0.676–1.040) 1.465 (0.818–2.626) 1.719* (0.955–3.095) Marital status  Never married ®  Married 0.999 (0.873–1.185) 0.986 (0.814–1.193) 1.012 (0.978–1.334) 1.347* (1.126–1.762)  Widowed/divorced/separated 0.893 (0.686–1.248) 0.839 (0.677–1.040) 2.134** (1.965–2.335) 2.671** (1.890–2.793) Regions  North ®  West 1.183*** (1.077–1.317) 1.238*** (1.091–1.455) 1.259*** (1.087–1.459) 1.257*** (1.083–1.459)  South 1.221*** (1.078–1.389) 1.258*** (1.097–1.487) 1.261*** (1.086–1.464) 1.262*** (1.086–1.466)  East 0.729 (0.478–1.306) 0.893 (0.589–1.158) 0.822 (0.584–1.256) 0.820 (0.583–1.154)  Central 1.287** (1.103–1.692) 1.322** (1.084–1.712) 1.326** (1.034–1.700) 1.327** (1.034–1.701)  North-east 1.159 (0.967–1.273) 1.152 (0.964–1.226) 1.145 (0.988–1.325) 1.142 (0.985–1.323) Smoking  No ®  Yes 0.795 (0.509–1.241) 1.340** (1.117–1.529) 1.208** (1.123–1.490) Drinking  No ®  Yes 1.015 (0.716–1.439) 1.894 (0.452–7.933) 1.004 (0.136–7.394) Regular consumption of fried foods and aerated drinks  No ®  Yes 1.072** (1.037–1.227) 1.435** (1.086–2.323) 1.443* (1.180–2.367) Age at Menarche  More than 15s ® 12 or less  3.190*** (1.696–6.000)  2.630*** (1.365–5.071)  13–15  0.974 (0.610–1.555)  0.859 (0.535–1.380) Nulliparity  No ®  Yes 0.914 (0.441–1.892) 0.875 (0.404–1.894) Menstrual hygiene  Hygienic methods ®  Unhygienic methods 1.660** (1.025–2.687) 1.579* (1.056–2.608)  Both 1.399 (0.914–2.143) 1.264 (0.809–1.974) Age at first birth  Below 18 ®  18–24 0.381 (0.136–1.062) 0.318* (0.106–0.954)  25+ 0.887 (0.457–1.722) 0.887 (0.439–1.792) Hormonal contraceptive use  No ®  Yes 1.472* (1.142–1.573) 1.535** (1.159–1.795) Had terminated pregnancy  No ®  Yes 0.896 (0.313–2.561) 1.028* (1.013–1.941) BMI category  Underweight ®  Normal weight 0.545*** (0.368–0.805)  Overweight 0.303** (0.107–0.854)  Obese 0.866 (0.258–2.905) Anemic  No ®  Yes 0.927 (0.640–1.343) Glucose level  Non-diabetic ®  Diabetic 3.579** (3.473–4.07) ®Reference category. *** p  < 0.001; ** p  < 0.01; * p  < 0.05. Adjusted hazard ratios and its confidence intervals, for observing the effect of covariates on the occurrence of premature menopause. ®Reference category. *** p  < 0.001; ** p  < 0.01; * p  < 0.05. The survival analysis demonstrated that women with higher levels of education have lower chances of having premature menopause, in comparison with those who have no education. Women who are employed have a 13.6% higher probability of experiencing premature menopause in comparison to unemployed women (HR 1.136; p  < 0.001). The women in higher wealth quintile, that is, richest have 36.8% lower chances of having premature menopause than poorest women (HR 0.632; p  < 0.001). Widowed/divorced/separated women are at a much higher risk of premature menopause in comparison to never married (HR 2.671; p  < 0.01). Region as a control showed that in the central and southern region of the country the risk of experiencing premature menopause is 32.7% (HR 1.327; p  < 0.05) and 26.2% (HR 1.262; p  < 0.01) higher respectively, in comparison to Northern regions. Smoking is observed to be a significant predictor of premature menopause, with 20.8% higher risk (HR 1.208; p  < 0.01). Regular consumption of fried food increases the risk off experiencing premature menopause by 44.3% (HR 1.443; p  < 0.05). There is increased risk of premature menopause for the one whose age at menarche is 12 or less years compare to 15 or more age at menarche (HR 2.63; p  < 0.001). Practicing unhygienic menstrual methods increases the risk by 57.9% in comparison to those who practice hygienic methods (HR 1.579; p  < 0.05). There is lower risk of premature menopause for those the women whose age at first birth is 18–24 in comparison to if the age at first birth is below 18 years (HR 0.318; p  < 0.05). The risk was also higher for women who had ever used any of the hormonal contraceptives (HR 1.535; p  < 0.05). The women who ever had any terminated pregnancy have higher changes of attaining menopause prematurely (HR 1.028; p  < 0.05). The diabetic women have 3.5 times higher chances of experiencing premature menopause (HR 3.579; p  < 0.01) (Table 4 ). The result for early menopause followed a similar pattern with smoking (HR 1.134; p  < 0.001) and termination of pregnancy (HR 1.116; p  < 0.001) being two of the significant predictors (Table S1 ).

The study has several strengths owing to its national representativeness and methodological robustness. Since the analysis is based on a national-level population survey that covered large-scale data on women’s health that helped to assess the maximum factors affecting early age at menopause. It was also possible to exclude the women who had undergone hysterectomy as with the surgery done these women have lower estrogen levels than other women who have attained premature or early menopause naturally 56 . This analysis controlled for most of the factors that could confound the results. The study has some limitations owing to the cross-sectional nature. There are chances of recall bias of the date of last menstrual cycle since it is self-reported. The fact that the current study employed data from women between the ages of 15 and 49 was a clear restriction. In order to get over this restriction, we employed survival models in our research. These models work best with data that is 'time to event data' and that contains censored cases. Since the study has analysed secondary data, more detailed micro-studies would help in better understanding of the premature or early menopausal cases.

The high number of women attaining menopause at younger ages is a matter of concern. This study is an important contribution to literature that provides robust prevalence estimate for premature and early menopause, and holistically analyses possible explanatory factors using a large-scale national data. It is crucial that underprivileged women have access to the proper diet and healthcare measures because early menopause is associated with osteoporosis and other health problems. To further understand the relationships between general undernutrition brought on by poverty and specific micronutrient deficiencies that may have an impact on ovarian reserve, such as vitamin D deficiency, more research is required. In India, both the proportion and the absolute number of post-menopausal women are growing, therefore it's critical to revamp public reproductive healthcare facilities to include the right kinds of treatment for them. The current health care system faces a difficulty in providing adequate care for the many women who experience early menopause. The demands of women going through early menopause should be taken into account by the government programmes already in place that are designed to meet the needs of childbearing women. Women who are approaching menopause want assistance in managing the symptoms brought on by hormonal shift and the menopausal transition. Women may endure vasomotor symptoms, urinogenital issues, and psychological issues during this time. These women may be able to cope with the discomfort they experience with the support of the therapy and counselling that health care professionals can offer. It is possible to teach healthcare professionals to offer the required counselling and advice to deal with the issues faced by menopausal women; these actions are low-cost and simple to incorporate into current programmes. Another strategy to encourage women going through premature menopause to seek out the appropriate medical care is to raise public knowledge of the detrimental effects of premature menopause on health and the significance of doing so. It is suggested to ensure health care access to underprivileged women has been supported by the data in wealth index, social and demographic characteristics, and lifestyle habits, and there is dire need to expand health care services and thus the budgetary allocations.

The study estimated the prevalence of premature and early menopause, and further analyzed its determining factors, utilizing a large-scale national population survey. Previous studies based on NFHS or other national surveys mostly included women who underwent surgical menopause as well 31 – 33 , however this study excluded it in order to avoid overestimation of the prevalence rates for premature and early menopause, with the confounding effect of large number of hysterectomies. The estimated prevalence of premature menopause is 2.2% and early menopause is 16.2%. A study by 34 , had estimated premature menopause to be 1.5% using the DLHS 2007–2008 data in India. Another study based on the US and Korean population too computed the prevalence of premature menopause, as 1.7% and 2.8%, respectively, and early menopause as 3.4% and 7.2% in US and Korea, respectively 35 . However, none of the studies had excluded the pregnant and lactating mothers from the sample, which was considered in the present study, thus making the current estimation more robust. Overall, the study's socioeconomic, family planning, and demographic characteristics were substantially linked to premature and early menopause. Women in rural settings are more likely than those in urban areas to have premature menopause because they are less likely to have access to health care services 36 . The degree of education has been demonstrated to be a significant explanatory factor in premature and early menopause. Premature and early menopause rates increased among women with lower levels of education, This result is in line with other studies as well 16 , 37 . According to the present study, women from poorer households were more likely to go through a premature and early menopause, which is consistent with earlier studies in India or other regions 15 , 18 . There is a possible poverty and nutrition link in this regard. This phenomenon could be explained in the way that women in rural areas in poorer households, and also having less or no education have lack of awareness coupled with inaccessibility of healthcare services and poor nutritional diet. The intersectionality of residential, economic and educational vulnerability has compounded effect that may lead to early onset of menopause. Lifestyle choices such as tobacco, alcohol and junk food consumption were observed to be contributing factors towards premature and early menopause. Tobacco smoking during reproductive cycle, or in general smoking had been found to be a significant factor causing premature menopause 32 . The risk was found to be comparatively higher among female cigarette smokers than the women who did quit smoking. By lowering the flow of oestrogen, tobacco's anti-estrogen actions speed up the start of menopause 38 . Overall, this compromises the typical hormonal balance in women, which has an impact on the entire system. When smoking combined with an early menopause, the hormonal imbalances result in sleep problems, depressive symptoms, and eventually impaired cognitive performance 20 , 39 , 40 . Meat and alcohol consumption and physical inactivity, were independently found to have significant association with premature menopause in previous studies 41 , 42 . Dietary pattern and nutritional status of a woman had been also found to be contributing factors to premature menopause. This risk is higher among women with low body mass index (BMI) or malnourished which is also observed from the study results 43 , 44 . Fat tissue is where oestrogen is stored, and extremely thin women have lower reserves of oestrogen, which are more easily exhausted. Women with higher BMIs have higher amounts of estrone (E1) and estradiol (E2) in their bodies, which can delay menopause. BMI is a key factor in determining endogenous oestrogen levels 45 . Though, they study showed premature menopause to be higher among overweight/obese women, whereas early menopause was higher among underweight women. Early age at menarche if found to have a significant association with premature and early menopause. A study by Mishra et al. (2017), reported that women who had menarche before age of 13, had double the risk of experiencing premature menopause and 31% higher risk of early menopause 46 . Early menarche has been linked to poor reproductive functioning, including irregular periods 47 , 48 , PCOS 49 , and a slightly higher risk of endometriosis 50 , according to earlier studies. In contrary to previous studies 46 , 48 that reported the nulliparous women had higher chances of experiencing early onset of menopause, the present study did not show any significant association. The age at first birth is another attributable factor, which shows that premature and early menopause is higher among the women who age at first birth is lower than 18 years 33 . Women who do not become pregnant typically have an earlier menopause than those who have children. Additionally, it is true that common factors—ranging from genetics to childhood environmental factors like obesity, psychological stress, and social environment—may explain the association and have an impact on menopause, the date of the first period, and fertility 46 . Interestingly the use of hormonal contraceptives such as, injectables, pills, and emergency contraception also emerged as an explanatory variable for premature and early menopause. Though there has been studies earlier that stated those who used oral contraception had experienced premature menopause less 51 – 53 . History of termination of pregnancy showed association with premature and early menopause, however not established much in earlier studies. In addition to harming the uterus, abortion can result in ovarian dysfunction, which can lead to the failure of the ovaries. Able to produce eggs of average, everyday quality. Hormone production volume can also be impacted. Additionally, ovarian infection or blocked fallopian tubes may result from unsafe or ongoing abortion, full ovary removal, which harms and ages the ovaries 54 . Another interesting result is that of the association of higher glucose level with premature and early menopause. Women who have either type 1 diabetes (before age 30) or type 2 diabetes (between 30 and 39 years) are more likely than identical women without diabetes to experience menopause earlier in life. There might be a connection between diabetes and changes in the body, reproductive system, and ageing and functioning of the ovaries 55 . Future studies are required to determine the potential causes of the association between diabetes and premature or early menopause. The study provides a baseline for carrying out future research in India by considering more factors and conducting causal inferences.

The event of ovarian failure is termed as menopause. Majority women normally attains menopause in the age bracket of 45 and 55 1 . According to the World Health Organization (WHO), natural menopause is defined as “permanent cessation of menstruation resulting from the loss of ovarian follicular activity”, which is normally recognized after a year-long consecutive amenorrhea 2 . This is an essential physical and hormonal event in a healthy woman’s reproductive cycle. With increasing age, the ovarian function depletes and decreases its production of oestrogen and progesterone hormones, and thus the gradual decline in fecundity 3 , 4 . So basically, menopause is the transition of a woman’s life from reproductive phase to a non-reproductive phase, which has biological, emotional, sociocultural significance 5 . The age distribution of menopause is like a Gaussian curve ranging from age 40 to 54, but the general clustering is around 45–55 6 . Some women, due to ovarian insufficiency attains menopause at an early age due to lifestyle factors and hormonal imbalances. Menopause occurring before the age of 40 is premature and between 40 and 44 years age is early, since the natural age of menopause lies between 45 and 50. The cessation of menses is marked by amenorrhea, rise in gonadotrophin levels and oestrogen deficiency 7 , 8 . According to a PAN India study by Ahuja (2016), there is a strong association between early onset of menopause and various factors such as illiteracy, poor socio-economic background, underweight, parity, and age at pregnancy 9 . Studies have shown that age at menarche, breastfeeding of previous child, age at first pregnancy, plays a very important role in determining the onset of menopause 10 . There are also effects of nulliparity, usage of oral contraceptive pills, having a live birth or not, on the onset of natural menopause 11 , 12 . Menopausal age is associated with a number of factors with includes smoking 13 , 14 , level of education 15 , working status 16 , abortion 17 , body mass index 16 and food habits 18 . Smoking specifically have shown adverse impacts on reproductive health and heavy smokers were observed to reach menopause earlier 19 , 20 , thus making it important to study its association. Tobacco consumption has anti-estrogenic impacts on the female body which can lead to estrogen function resistance 20 . There has not been any consistent association between the onset of menopause and the above stated factors. In the coming decades, both the proportion and total number of Indian women aged 45 and beyond are expected to rise sharply. In India, there were around 96 million women who were 45 years of age or older as of the 2011 census, and this figure is projected to rise to 401 million by 2026 21 . Women in India could, on average, spend 30 years in the postmenopausal stage of life because the average life expectancy at age 45 is 30 years. The post-menopausal population may provide significant problems to the provision of public healthcare in the future due to the health concerns associated with these years, including hypertension, heart disease, osteoporosis, and a deterioration in overall quality of life 22 , 23 . Increasing urbanization and changing lifestyle pose greater challenges to public health. Identifying factors associated with early menopause are necessary. The age at menopause is associated with the risk of several chronic diseases such as cardiovascular diseases, breast and endometrial cancers, and osteoporosis 24 – 27 . Although this issue is ignored in India, premature or early menopause has numerous short- and long-term negative health consequences. Health problems related to early onset menopause are not well known in India because the determinants and prevalence of premature menopause are not well documented, and thus comes our major research question regarding the same. The hypothesis is, to test. There is no significant effect of socio-economic, demographic, medical and lifestyle behaviour on premature and early onset of menopause against There is a significant effect of socio-economic, demographic, medical and lifestyle behaviour on premature and early onset of menopause. Thus, the study aims to estimate the prevalence of both premature and early menopause, and examine the potential associated factors in India.

Supplementary Information. Supplementary Information.