The primary objectives are to (1) estimate national prevalence, phenotypic variations, and risk factors of PCOS in the community; (2) estimate the burden of comorbidities such as metabolic syndrome, dyslipidemia, cardiovascular complications, NAFLD, psychiatric abnormalities, sleep apnea, AGT including type 2 diabetes, and impairment of quality of life (QOL) among known cases of PCOS; and (3) compare the variation in efficacy of standard therapeutic modalities for metabolic dysfunction in women with PCOS (2 sites, All India Institute of Medical Sciences [AIIMS] and Sher-i-Kashmir Institute of Medical Sciences [SKIMS]).

The secondary objective is to assess knowledge and management practices among service providers in the public health sector on PCOS to develop a standard screening and training modules to facilitate management of PCOS at different levels of the public health care system.

Women aged 18-40 years who are not pregnant or lactating and have been residing in the area for 1 year or more will be included in the study. Women who are known to have PCOS will be recorded for the purposes of PCOS prevalence and will only be evaluated in detail for substudy 2 if they are drug-naïve. Women taking oral contraceptives, steroids, antiepileptics, or other drugs known to interfere in glucose or lipid metabolism will not be evaluated in detail and will be excluded from the study.

The study design is a two-stage, cross-sectional, community-based study.

Women aged 18-40 years will be recruited using a multistage sampling technique involving urban and rural areas. The polling booths will be selected randomly at the coordinating center by a random selection process. This study will be conducted in 6 zones of the country involving 10 centers that are likely to represent a majority of ethnic groups.

The following are the criteria for inclusion: women aged 18-40 years who are permanent residents of the area (>1 year), willing to participate in the study and sign informed consent, known to have PCOS, and not on treatment.

Pregnant or lactating women and those with cognitive limitations, physical limitations, or both that prevented them from answering the questionnaire will be excluded from the study.

Women with a history of drug intake such as steroids, androgens, oral contraceptives, antiepileptics, or drugs known to interfere in glucose or lipid metabolism will also be excluded.

A multistage cluster sampling technique is being used in this study. The first step in the sampling procedure will be to have representative data from the 6 zones (N, S, E, W, NE, and Central) of the country (Figure 1). The participating institutions from 6 zones were selected in view of location, logistics, feasibility, and the investigator profile. One urban and one rural Vidhan Sabha constituency near each institution will be selected. The choice of selection of districts and the constituencies will be given to the principal investigators at each site in order to make the study execution simpler. A rural district would mean 70% of subjects pursuing agriculture activities, while urban would mean that less than 30% have agriculture as the source of income.

After the constituencies are selected, the entire list of polling booths will be obtained from the Regional Election Commission offices. Polling booths from rural or urban constituencies will be selected to fulfill the required sample size using a random number table so that each polling booth has an equal chance to get selected. This procedure will be carried out at the coordinating center (SKIMS, Srinagar) by a third party not directly involved in the study. Next, the sites will procure the electoral roll, and voter IDs will be searched for women (aged 18-40 years) to be the potentially eligible subjects. The resulting list of women will be screened using a checklist for eligibility. If a woman is not eligible or refuses to participate in the study, the offer will go to the next woman in the list. If there is more than one eligible woman in a family, the sampling frame will also shift to the next in the voter ID list. The procedure will continue until a total of 1000 women (500 rural and 500 urban) are screened per center (Figure 2).

Sample size was calculated, primarily to estimate the prevalence of PCOS in the country. Previously reported prevalence rates of PCOS have varied from 3.7% to 22.3% [32,33]. For the purpose of calculation of sample size, we presume the prevalence of PCOS in India to be 10%. With an assumed prevalence of 10% and an absolute error margin of 1% in a 95% 2-sided confidence interval and making allowance for about 20% refusals or nonresponse, we need to target a total of 4500 women. Further, being a cluster design and taking a design effect of 2, a total of 9000 women aged 18-40 years will be studied from 10 centers (ie, 1000 per center, except for the 2 centers in Hyderabad, which have been designated with 500 each), half from a rural community and half from an urban community at each center.

Based on a pilot study previously conducted by the chief coordinator using a similar questionnaire [34], 30% (2700/9000) of the women are expected to screen positive (likely to have PCOS) on the screening questionnaire and therefore are required to be investigated in detail. Additionally, of the 70% (6300/9000) of women expected to screen negative on the screening questionnaire, 20% (1260/6300) will require examination and investigation in detail. This will make the sample size of women requiring detailed evaluation and blood testing a total of 3960 (N=2700+1260). Keeping in view the number of centers (N=10), each participant center will screen approximately 1000 women, and a total of 440 women who will be invited for detailed examinations: 20% (140/700) of screen-negative women and all 30% (300/1000) who screen positive (Figure 2).

The subjects will be invited to nearby institutes for detailed clinical, biochemical, hormonal, and sonography assessments. These women will be assessed for clinical detail about menstruation (regularity, duration, flow, amenorrhea), weight gain, unwanted male pattern hair growth, hair fall, infertility, etc. The details about any illness will be recorded. Psychiatric morbidity will be assessed using the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) criteria with the Mini-International Neuropsychiatric Interview, version 4.0. The details of family pedigree will also be recorded in the predesigned questionnaire.

Anthropometric assessment including measurement of height, weight, waist circumference, skinfold thickness on dorsum of hand and triceps, and blood pressure and detailed systemic examination will be undertaken. Height and weight will be measured with subjects in light clothes and without shoes, using standard apparatus. Weight will be measured to the nearest 0.1 kg on a calibrated digital scale. Height will be measured by stadiometer by standard methodology to the nearest 0.1 cm. Waist and hip circumference will be measured to the nearest 0.5 cm with a measuring tape. The waist circumference will be defined as the point midway between the iliac crest and the costal margins (lower ribs), while hip circumference will be defined as the widest circumference over the buttocks and below the iliac crest [35]. Body mass index (BMI) will be derived as weight in kg divided by the square of height in meters. Overweight will be defined using the Asian BMI range of 23.0-24.9 kg/m², while obesity will be defined as having a BMI equal to or greater than 25 kg/m² [36]. Quantitation of hirsutism by a single observer by counting 9 specific body areas for the Ferriman-Gallwey score and assessments of acne vulgaris, androgenic alopecia, and acanthosis nigricans will be performed in all subjects [37].

Ultrasonography of the abdomen will be performed to evaluate the polycystic ovary morphology and grading of fatty liver (NAFLD) and nonalcoholic pancreatic steatosis. The US will be performed at all centers, and MRI of the pelvis (n=100) will be performed primarily at SKIMS, Srinagar, to compare US and standardization of US technique by 2 independent radiologists. MRI will be done as per a randomized selection proportional to the subjects at the site.

The OGTT will be performed after an overnight fast of 10-12 hours. An oral glucose load with 75 g of anhydrous glucose dissolved in 250-350 mL of water will be administered over 3-5 minutes immediately after the basal sample. The blood samples for LH, FSH, testosterone, AMH, SHBG, and DHEAS will be collected on days 3-7 (early follicular phase) of spontaneous or medroxyprogesterone-induced (in amenorrhea patients) menstrual cycle to confirm PCOS. Hematological parameters (hemoglobin, red blood cells, white blood cells, etc), biochemical tests like lipid profile, liver function, kidney function, blood glucose (OGTT), and hormonal assays like T₄, thyrotropin, LH, FSH, PRL, E2, P4, cortisol, testosterone, AMH, DHEAS, 17OHP, and insulin at 0 hours, 30 minutes, and 2 hours (optional) will be carried out at the respective centers, and special hormones like testosterone and SHBG will be done in the central facility to be created at the coordinating center (SKIMS, Srinagar). On the day of testing, women will be called after 10-12 hours of fast, and blood samples (15 ml) will be drawn. Blood will be also collected for glucose and insulin estimation after 30 minutes and 120 minutes of 75 g oral glucose challenge. Insulin samples will be immediately put in a refrigerator, separated in cold centrifuge, and stored at −20 °C until the assay. Plasma will be separated in a refrigerated centrifuge at 1560g for 10 minutes, and sera will be stored at −80 °C until tested.

Plasma glucose will be measured using the glucose oxidase–peroxidase method, and hormonal tests (LH, FSH, T₄, thyrotropin, PRL, E2, cortisol, testosterone, AMH, hsCRP, cytokines, and 25OHD) will use ECLIA on Cobas e411 (Roche Diagnostics USA).

On the basis of one study, metformin could improve the Homeostatic Model Assessment of IR from 2.5 (SD 0.39) to 2.79 (SD 0.39) [38]. Assuming that the baseline was similar, at an α error of 0.05 with power of 90%, and with a projected attrition rate of 20%, we need to recruit 153 women per group. To round off, we will recruit and randomize 160 women with PCOS per group (ie, a total of 320 women). A total of 160 subjects will be recruited and randomized in each of the 2 groups, with 80 subjects in each arm.

Eligible consenting subjects will be randomized in a 1:1 ratio into 2 arms through a blinding procedure.

Arm A compares lifestyle (32 kcal/kg diet per day with 55% carbohydrates, 20% protein, 20% fat, and 20 g of fiber with 30 minutes of brisk walking daily) and metformin (1 g /day) versus lifestyle alone (n=160; 80 per center).

Arm B compares lifestyle (32 kcal/kg diet/day with 55% carbohydrates, 20% protein, 20% fat, and 20 g of fiber with 30 minutes of brisk walking daily) and OCP (30 µg ethinyl estradiol plus 2 mg drospirenone) versus lifestyle alone (n=160; 80 per center).

The subjects will be recruited from the community pool of substudy 1 and the institutional pool. The eligibility based on inclusion and the investigators will assess exclusion criteria. The institutional pool, which comprises the women attending endocrinology, gynecology, or other clinics for various complaints suggestive of PCOS, will be taken for purposes of recruitment in the intervention part of phase 1 or for further assessment in forming a long-term cohort. Potentially eligible patients will be recruited from the outpatient department services of the participating centers. If the subjects qualify on the inclusion and exclusion criteria, they will be informed about the details of the study and hence will be approached for written informed consent. All potentially eligible patients will undergo clinical, biochemical, and imaging (US) assessment to confirm PCOS. A logbook will be maintained for all patients undergo screening, randomized, eligible but not randomized along with reasons for non-randomization. For the intervention study (substudy 3), the subjects will be recruited only from AIIMS, New Delhi, and SKIMS, Srinagar. The sleep study and carotid Doppler at SKIMS and AIIMS, New Delhi, will be conducted for 100 subjects at each site.

The drug-naïve women who qualify on the 1990 NIH criteria (from both the institutional pool and the community pool) will be invited for the study. This will be an open-label randomized intervention trial and will follow a common protocol at each participating center. Women who do not desire fertility and who furnish written informed consent (second time) will undergo a detailed clinical assessment including details of menstrual, drug, and other relevant history followed by clinical examination involving anthropometry (height, weight, waist circumference); Ferriman-Gallwey scoring for hirsutism; grading of acanthosis nigricans, androgenic alopecia, and acne vulgaris (grades 1-3); blood pressure measurement; and a brief relevant systemic examination. Biochemical and hormone estimation will be done at baseline from the second day to the seventh day of the menstruation cycle and at 3 and 6 months. The subjects will be followed up at 6 months, and all these investigations will be repeated at the follow-up. A total of 320 women for the intervention phase (ie, 80 subjects per arm and per center) will be recruited. The study will be done only in 2 participating centers (SKIMS, Srinagar, and AIIMS, New Delhi). The intervention phase will be initially done for a period of 24 weeks.

Standardized study tools will be used to record patient name, contact details, history, family history, personal information (age of menarche, menstrual regularity information, acne severity), food habits, lifestyle habits, previous disease information, drug intake history information, and clinical details including anthropometry (height, skinfold, waist circumference) by using anthropometric tools (SECA 213 height scale, SECA 813 weight scale, SECA 203 nonstretchable tape, and calipers measuring skinfold thickness), blood pressure (using Omron blood pressure instruments), Ferriman-Gallwey score, androgenic alopecia, acanthosis nigricans, acne vulgaris, and systemic examination. Measurement of biochemical parameters (blood glucose, OGTT, lipid profile, liver function, kidney function), protein:creatinine ratio, and 25OHD (in selected cases); hormonal assays (like T₄, thyrotropin, LH, FSH, PRL, E2, P4—21st day optional, cortisol, testosterone, AMH, DHEAS, 17OHP, SHBG, and insulin); and radiological assessment will be done at baseline and after 6 months of treatment.

Patients will be evaluated for any adverse events and efficacy at 90 days (−7 days to +7 days) and 180 days (−7 days to +7 days) through follow-up. All clinical and biochemical investigation will be performed in a similar fashion as done at baseline. If the patient is not available for assessment at 180 days (−7 days to +7 days) follow-up, this will be considered a protocol deviation. The safety of different interventions will be assessed by recording the development of immediate or delayed complications recorded in a form with a prespecified list of safety parameters. Immediate reactions will include allergic reactions, diarrhea, vomiting, deep vein thrombosis, etc in a prespecified checklist.

Field activities will be conducted for the purposes of screening. It will include rapport building among locals and the selection of the study subjects (1 woman per household). It will be the duty of the research staff to approach the selected individuals and to make them understand their role in the study. Consenting individuals will be screened for PCOS phenotype using the predesigned structured questionnaire. The screening questionnaire will be used to obtain information on personal identification details, demographics, socioeconomic status, menstrual history and clinical information, history of any systemic disease, and QOL. The women who have hyperandrogenism or oligomenorrhea with or without obesity and acanthosis nigricans (IR) will be taken as probable cases of PCOS. Screen-positive and selected screen-negative (regular menstrual cycles, no features of hyperandrogenism or insulin resistance) individuals will be informed to visit the hospital for further investigations.