OBJECTIVES
Administrative records such as Medicare fee‐for‐service (FFS) claims provide accurate information on services paid for by Medicare. However, the increasing availability of electronic health records means many researchers may be inclined to rely on data coded in hospital information systems rather than claims. The current quality and accuracy of hospital reports on the use of post‐acute care (PAC) services are not known.
DESIGN
This study examined differences in the PAC use between hospital discharge status recorded on Medicare Provider and Analysis Review inpatient hospital records and claims for PAC services.
SETTING
In addition to assessments of the three types of Medicare‐reimbursed PAC (home health agency [HHA], skilled nursing facility [SNF], and inpatient rehabilitation facility [IRF]), the analysis also considered home without PAC services as a default discharge location.
PARTICIPANTS
The analysis was conducted using data for FFS beneficiaries who participated in the Medicare Current Beneficiary Survey and had one or more inpatient hospitalizations from 2006 to 2011.
MEASUREMENTS
This study measured discrepancies between hospital‐reported discharges to PAC and PAC use based on Medicare claims.
RESULTS
The study found that, on average, 27.9% of hospital reports of discharging to Medicare‐covered PAC services were not substantiated by Medicare PAC claims. Among all the discharge pathways, discharging to HHAs had the highest discrepancy rate (29.6%), followed by IRFs (14.7%) and SNFs (13.8%).
CONCLUSION
The study results call for cautions about the extent to which the reported discharge locations on hospital claims may differ from actual PAC services used. Assuming that Medicare FFS claims were complete and accurate, researchers using the discharge status reported on Medicare hospital claims should be aware of possible measurement errors when using hospital‐reported discharge locations.
from Wiley: Journal of the American Geriatrics Society: Table of Contents https://ift.tt/2snGXry
via IFTTT
No comments:
Post a Comment