Ultra-low dose hormonal pills have become the favoured treatment option over low dose pills in PCOS. Existing literature shows marked heterogeneity in metabolic effects of pills with different compositions. This might have been due to improper choice of outcome variable for comparison. Total cholesterol - high density lipoprotein ratio (TC/HDL) discordance with Low Density Lipoprotein cholesterol (LDL-C) was a better marker of future risk of atherosclerotic cardio-vascular disease and stable to variations. Present study was a randomised controlled trial conducted to compare prevalence of TC/HDL and LDL discordance among non-obese PCOS women treated with hormonal pills of different compositions. Role of baseline participant features and pill composition on discordance after treatment was determined. Women were randomised into three arms, two arms receiving ultra-low dose pills (either Ethinyl Estradiol 20µg with drosperinone 3mg or Ethinyl Estradiol 15µg with gestodene 60µg) and one arm receiving low dose pill (Ethinyl Estradiol 30µg with desogestrel 150 µg). Discordance was observed in more than 1/4th participants prior to intervention. After one year of treatment less than 1/5th of participants were discordant. Ultra-low dose pill users had lower discordance, LDL and TC compared to low dose pill users after one year of treatment. Random Forest a non-linear classifier showed the highest accuracy in predicting discordance. Baseline Parameters having maximal impact on occurrence of discordance were triglyceride (TG), HOMA-IR, BMI and HDL.Ultra-low dose hormonal pills have become the favoured treatment option over low dose pills in PCOS. Existing literature shows marked heterogeneity in metabolic effects of pills with different compositions. This might have been due to improper choice of outcome variable for comparison. Total cholesterol - high density lipoprotein ratio (TC/HDL) discordance with Low Density Lipoprotein cholesterol (LDL-C) was a better marker of future risk of atherosclerotic cardio-vascular disease and stable to variations. Present study was a randomised controlled trial conducted to compare prevalence of TC/HDL and LDL discordance among non-obese PCOS women treated with hormonal pills of different compositions. Role of baseline participant features and pill composition on discordance after treatment was determined. Women were randomised into three arms, two arms receiving ultra-low dose pills (either Ethinyl Estradiol 20µg with drosperinone 3mg or Ethinyl Estradiol 15µg with gestodene 60µg) and one arm receiving low dose pill (Ethinyl Estradiol 30µg with desogestrel 150 µg). Discordance was observed in more than 1/4th participants prior to intervention. After one year of treatment less than 1/5th of participants were discordant. Ultra-low dose pill users had lower discordance, LDL and TC compared to low dose pill users after one year of treatment. Random Forest a non-linear classifier showed the highest accuracy in predicting discordance. Baseline Parameters having maximal impact on occurrence of discordance were triglyceride (TG), HOMA-IR, BMI and HDL.