The Outcome Assessment and Information Set (OASIS) plays an essential role in home healthcare. To date, few researchers have investigated the psychometric properties of the OASIS in actual clinical practice. This study used several methodologies to estimate reliability and validity of OASIS when clinicians were conducting assessments as part of the routine care process. Analyses from a delayed assessment approach (n = 259), a simultaneous assessment approach (n = 105) and an assessment of consistency between OASIS and the Centers for Medicare and Medicaid Services (CMS) 485 (n = 141) provided insights into OASIS reliability and validity. Reliability for many items was considerably lower than prior studies and there were notable inconsistencies between several domains of OASIS and the CMS 485. Based on these findings it was also unclear whether prior reliability findings could be generalized to practice because the previous research did not account for the stresses and unpredictability of the home healthcare setting. Further, the inconsistencies between the OASIS and CMS 485 suggested that the OASIS might not truly reflect the patient's condition. Implications for improving the use of the instrument in practice settings are discussed.