This study was conducted in rural Pakistan to assess the dose-response relationship between skin lesions and arsenic exposure and their variation by demographic characteristics. The study included 398 participants (66 participants with skin lesions and 332 without) residing in six previously unstudied villages exposed to ground water arsenic in the range of < 1 to 3090 μg L− 1. The skin lesions identification process involved interview and physical examinations of participants followed by confirmation by a physician according to UNICEF criteria. Urinary inorganic arsenic (iAs), total arsenic (tAs), monomethylarsonic acid (MMA), and dimethylarsinic acid (DMA) were analysed to determine methylation capacity, methylation efficiency and the dose-response relationship with skin lesions. Study participants with skin lesions were found to be exposed to arsenic > 10 μg L− 1 with a daily arsenic intake of 3.23 ± 3.75 mg day− 1 from household ground water sources for an exposure duration of 10–20 years. The participants with skin lesions compared to those without skin lesions showed higher levels of urinary iAs (133.40 ± 242.48 vs. 44.24 ± 86.48 μg g− 1 Cr), MMA (106.38 ± 135.04 vs. 35.43 ± 39.97 μg g− 1 Cr), MMA% (15.26 ± 6.31 vs.12.11 ± 4.68) and lower levels of DMA% (66.99 ± 13.59 vs. 73.39 ± 10.44) and secondary methylation index (SMI) (0.81 ± 0.11 vs. 0.86 ± 0.07). Study participants carrying a lower methylation capacity characterized by higher MMA% (OR 5.06, 95% CI: 2.09–12.27), lower DMA% (OR 0.64, 95% CI: 0.33–1.26), primary methylation index (PMI) (OR 0.56, 95% CI: 0.28–1.12) and SMI (OR 0.43, 95% CI: 0.21–0.88) had a significantly higher risk of skin lesions compared to their corresponding references after adjusting for occupation categories. The findings confirmed that inefficient arsenic methylation capacity was significantly associated with increased skin lesion risks and the effect might be modified by labour intensive occupations.