The aim of this study was to find predictors of neuropathy and reactions, determine the most sensitive methods for detecting peripheral neuropathy, study the pathogenesis of neuropathy and reactions and create a bank of specimen, backed up by detailed clinical documentation. A multi-centre cohort study of 303 multibacillary leprosy patients in Northern India was followed for 2 years. All newly registered MB patients requiring a full course of MDT, who were smear positive and/or had six or more skin lesions and/or had two or more nerve trunks involved, were eligible. A detailed history was taken and physical and neurological examinations were performed. Nerve function was assessed at each visit with nerve conduction testing, warm and cold detection thresholds, vibrometry, dynamometry, monofilaments and voluntary muscle testing. Because the latter two are widely used in leprosy clinics, they were used as 'gold standard' for sensory and motor impairment. Other outcome events were type 1 and 2 reactions and neuritis. All subjects had a skin biopsy at registration, repeated at the time of an outcome event, along with a nerve biopsy. These were examined using a variety of immunohistological techniques. Blood sampling for serological testing was done at every 4-weekly clinic visit. At diagnosis, 115 patients had an outcome event of recent onset. Many people had skin lesions overlying a major nerve trunk, which were shown to be significantly associated with an increased of sensory or motor impairment. The most important adjusted odds ratios for motor impairment were, facial 4.5 (1.3-16) and ulnar 3.5 (1.0-8.5); for sensory impairment they were, ulnar 2.9 (1.3-6.5), median 3.6 (1.1-12) and posterior tibial 4.0 (1.8-8.7). Nerve enlargement was found in 94% of patients, while only 24% and 3% had paraesthesia and nerve tenderness on palpation, respectively. These increased the risk of reactions only marginally. Seven subjects had abnormal tendon reflexes and seven abnormal joint position sense. In all but one case, these impairments were accompanied by abnormalities in two or more other nerve function tests and thus seemed to indicate more severe neuropathy. At diagnosis, 38% of a cohort of newly diagnosed MB leprosy patients had recent or new reactions or nerve damage at the time of intake into the study. The main risk factor for neuropathy found in this baseline analysis was the presence of skin lesions overlying nerve trunks. They increased the risk of sensory or motor impairment in the concerned nerve by 3-4 times. For some nerves, reactional signs in the lesions further increased this risk to 6-8 times the risk of those without such lesions. Patients with skin lesions overlying peripheral nerve trunks should be carefully monitored for development of sensory or motor impairment.