Background and aim - Alternative therapies based on homeopathy can be effective in improving reproductive performance in intensive pig breeding. In this study, the effect of a homeopathic complex on reproductive performance of sows under intensive farming has been investigated.Material and methods - Over period of three years, 186 sows were recruited from a farm where a large proportion of animals were suffering from prolonged weaning-to-oestrus intervals (WEI) and weaning-to-service intervals (WSI). Sows were allocated to two groups; once per month, one group was given a homeopathic complex (Borax 10 mK plus Lycopodium 10 mK), while the other group was given a hydro-alcoholic solution (placebo). The follow-up period started one week before the expected date of oestrus, continued for two pregnancies and ended after the weaning of the second farrowing. To evaluate reproductive performance, during the follow-up we collected data on quantitative parameters such as the average number of stillbirths, newborns, and repeat services per farrowing. Time-related data such as WEI, WSI, length of the two pregnancies and weaning periods were also collected to measure the length of the follow-up of each sow. Differences in quantitative parameters between the two groups were evaluated using parametric and non-parametric statistics. Time-related data were used to plot Kaplan–Meier curves and in Cox regression models to evaluate whether treated sows had a higher probability of experiencing a shorter follow-up in comparison to untreated sows.Results - We did not found significant differences in the number of newborns, while the number of stillbirths was higher in the treatment group, even if the difference was slightly significant (p-value = 0.03). The number of repeat services was lower in the treatment group, and this difference was highly significant (p-value < 0.001). Results from the Cox regression models suggest that the end of the follow-up was reached by sows of the treatment group at about twice the rate of sows of the control group (model 2, Hazard Ratio-treatment = 2.27; 95%CI: 1.56–3.24).