> For the complete documentation index, see [llms.txt](https://wiki.medicaiddatalearningnetwork.org/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://wiki.medicaiddatalearningnetwork.org/advanced/identifying-claims-from-acute-care-hospitals.md). # Identifying Claims from Acute Care Hospitals {% hint style="info" icon="pen-to-square" %} Contributed by Shashi Kapadia () and Kenneth Karan () {% endhint %} ## Motivation For a research study focused on hospitalized patients, we needed to select claims from acute care hospitals and aggregate them into contiguous stays or episodes. Not all claims in the inpatient file are in fact from acute care hospitals, as claims in this file may be from other types of inpatient facilities, and some appear to be claims for outpatient services. We needed to develop criteria to ensure that constructed episodes were only comprised of inpatient claims. ## Approach We adapted two distinct approaches that were suggested in T-MSIS TAF Methodology briefs #5011 and #5241. Both documents employed several common criteria and branched off into individualized criteria, with the two documents using different variables to select claims from acute care hospitals. Rather than select one of the two approaches, we cast a wider net and implemented a combination by counting a claim as being from an acute care hospital one of several criteria were met. ## Criteria required of by all approaches *All* of the criteria below must be satisfied * Claim in the TAF Inpatient File. * Have a service delivery claim type code: 1, 3, W, A, C, or U * Have an institutional revenue center code that is not missing and not zeroes and only has alphanumeric values. * Claim start date and claim end dates are valid. Valid dates are defined as: * If a date is before 10/1 of the year before the year on the file name, then it is set to missing. This accounts for scalar dates of zeroes or other erroneous values. For example, if the admission date was 1/1/1960 or 1/1/2015, and file being read was TAF\_INPATIENT\_HEADER for 2016, then the admission date would be set to missing. * If a claim end date is after a claim start date, then the two dates are exchanged. * Inpatient claim start dates are the earliest non-missing value of admission date (admsn\_dt) or service begin date (srvc\_bgn\_dt). * Inpatient claim end dates are the latest non-missing value of discharge date (dschrg\_dt) or service begin date (srvc\_bgn\_dt). In addition, *any* one of the criteria below must be satisfied. * A bill type code (bill\_type\_cd) with the prefix of 011, 012, 041, or 042. * A taxonomy code (blg\_prvder\_txnmy\_cd or blg\_prvdr\_nppes\_txnmy\_cd) of 282N00000X, 286500000X, 282J00000X, or 284300000X. * A type of service (tos\_cd) code of 001, 060, 084, 086, 090, 091, 092, or 093. ## Key Considerations and Limitations * Appendix B of the TAF DQ Methodology Brief #5241 refers to hospital taxonomy codes other than the ones listed above. These were omitted because they refer to specific hospital subtypes that were not counted in the brief’s definition of Inpatient Hospital claims, as shown in Table A.1, row 5. Nonetheless these additional taxonomies may be of interest for specific studies. * Since our method was developed, the Lown Institute published a blog entry documenting the presence of inpatient claims in the TAF Other Service File. We have not implemented that method and so did not include it here, but it stands to reason that our method may be extended to select claims from the Other Services file that meet all other criteria. * This approach is not benchmarked against an external data source. We attempted to reconcile what seemed to be two largely divergent approaches in two TAF technical briefs but were unable to validate either approach against others. ## Sources * TAF DQ Methodology Brief #5011 p. 3-4. * TAF DQ Methodology Brief #5241 p. 3-4, 8. * Lown Institute Substack “Hospital inpatient admissions recorded in the TAF other services file” 8/8/25. Accessed at --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://wiki.medicaiddatalearningnetwork.org/advanced/identifying-claims-from-acute-care-hospitals.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.