The associations between occurrence of diseases, milk yield, and body condition score on conception risk after first artificial insemination were analysed in an observational study of 43 farms participating in a herd health program. Data were taken from 9369 lactations, from 4382 cows inseminated between 20 and 180 days in milk in the period 1990 to 1996. Two logistic regression models, one containing data from all lactations and a subset containing data from 1762 lactations with body condition scoring were used to determine pregnancy risk at first AI. A 1-point loss in body condition score during the 100 days after calving resulted in a 20 percent reduction in the risk of conception. Relative within herd kg milk yield had a significant and linear negative association with fertility. Increases and decreases in the percentage milk fat (or milk fat to protein ratios) from the monthly test days near to AI date are significantly (P < 0.01) predictive of AI outcome. Metritis, cystic ovarian disease, lameness, and mastitis gave odds ratios of for pregnancy risk ranging from 0.35 to 1.15, largely dependent on the interval in days from final disease occurrence to first AI. The advantages of coding diseases for linear or non-linear time-dependency were examined. Other diseases with significant effects associations to pregnancy risk were displaced abomasum, milk fever and retained foetal membranes. These diseases showed little relationship between fertility and the number of days since last occurrence. Results of this study confirm the negative effects of milk yield, body score condition loss, and disease on dairy cow fertility. The effects of some diseases on first service conception are strongly dependent on the interval since last disease occurrence. This is especially valid for clinical mastitis which has an extremely weak effect on conception if occurring prior to AI and is associated with > 50 percent reduction in pregnancy risk if occurring in the 3 wk period directly after AI.