Ketosis is associated with many transition cow diseases and the subclinical form has been found to be a common condition in high-producing dairy cows. The objectives of this field study in the Netherlands were (1) to determine risk factors for subclinical ketosis [SCK; 1.2-2.9mmol of β-hydroxybutyrate (BHBA)/L of serum] and clinical ketosis (CK: ≥3.0mmol of BHBA/L of serum) at 7 to 14 d in milk and (2) to assess the association of SCK and CK with production parameters at the first dairy herd improvement (DHI) testing. Twenty-three dairies were enrolled by a local veterinary practice from 2009 to 2010, and 1,715 cows were screened for ketosis by measuring serum BHBA concentrations at 7 to 14 d in milk. Overall, 47.2% of cows had SCK and 11.6% had CK. Mixed generalized logit models with a random effect of herd were used to evaluate cow level factors associated with SCK and CK. The associations of SCK and CK with milk production parameters were tested using mixed linear models with a random effect of herd. Cows at a moderate (3.25-3.75) or fat (≥4) body condition score before calving were more likely to develop SCK and CK than thin (body condition score≤3.0) cows. The risk for developing SCK was higher in parity 2 and older cows compared with heifers, whereas for CK only, parity ≥3 cows had a higher risk. The quarter of the year in which a cow calved was associated with the risk for SCK and CK. For SCK quarter 1 (January-March) and quarter 2 (April-June), and for CK quarter 1, quarter 2, and quarter 3 (July-September) all increased the risk of development of the condition compared with quarter 4 (October-December). An increased yield of colostrum at first milking was associated with increasing risk for SCK and CK. Prolonged previous lactation length and dry period length were both associated with increased odds for SCK and CK. Subclinical ketosis and CK were associated with a higher milk yield, a higher milk fat percentage, and a lower milk protein percentage at first DHI test day. Overall the study reinforces previous findings that the major risk factors for both SCK and CK are increasing parity, overconditioning of animals prepartum, season of calving, and dry period length. In addition, previous lactation length and liters of colostrum have been identified as additional risk factors for the development of ketosis.