Data management affects reproductive performance indicators in Swedish dairy herds

We studied how data management affected reproductive performance indicators at herd level and how herd characteristics affected the agreement of the indicators. Data from 483 herds in the Swedish Official Milk Recording Scheme was used. The herd level performance indicators were: number of inseminations per animal submitted to artificial insemination ( AI), percentage of non-returns at 56 days, number of days from calving to first and last insemination, percent pregnant per insemination and per animals submitted to AI, and fertility index. Agreement was assessed by the change in performance indicators from a routine calculation to an extra calculation where updated data was included. Changes of more than +/-5% were considered indicative of disagreement. The amount of disagreement for the reproductive performance indicators was variable. Percentage of non-returns at 56 days showed no disagreement, while number of inseminations per animal submitted to AI showed largest disagreement. Some herd characteristics, e. g. do-it-yourself-inseminations, percent pregnancy checked and rate of inseminations, affected amount of disagreement.