HIM 675 Module 3 Final Project Milestone One Example

Final Project Milestone One: Present or Acquired? Problem, Purpose and Research Questions for a Study of Present-on-Admission Accuracy at Cimarron Heights Medical Center

[Student Name]

Southern New Hampshire University

HIM 675: Research Methods and Evaluation

Final Project Milestone One

[Instructor Name]

[Date]

The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.

Final Project Milestone One: Present or Acquired? Problem, Purpose and Research Questions for a Study of Present-on-Admission Accuracy at Cimarron Heights Medical Center

Every secondary diagnosis on an inpatient claim carries a small flag that answers one question: was this condition present when the patient arrived? The answer determines whether a wound, an infection or an injury from falling is counted as harm the hospital caused, and for several conditions it also affects payment. At Cimarron Heights Medical Center, a composite 340-bed teaching hospital in Tulsa, Oklahoma, there is reason to doubt the answers. This milestone sets out the background, problem, purpose, research questions and hypotheses, definitions and boundaries of a study to measure how accurate the hospital's flags are and what predicts their errors.

Background

Medicare began requiring present-on-admission reporting for inpatient claims in 2008, and the flag was quickly adopted for quality measurement because it allows claims to separate conditions patients brought with them from complications that developed in the hospital. Pine et al. (2007) showed that adding the flag to administrative data improved risk-adjusted mortality models across 188 Pennsylvania hospitals. The flag's usefulness, however, depends on its accuracy. Goldman et al. (2011) checked California claims against a fresh reading of the chart and found the flag right roughly three times in four, wrong in both directions the rest of the time. Bahl et al. (2008) showed that whether conditions are recognized as present on admission substantially changes patient safety indicator rates. The flag is therefore a small field with large consequences.

Problem Statement

In fiscal 2026, an internal audit at Cimarron Heights found that 41 of 112 pressure injuries of stage 2 or worse, 37%, were coded as present on admission although no skin finding appeared in nursing or physician documentation from the first 24 hours. If those flags are wrong, the hospital is understating its hospital-acquired pressure injury rate, reporting inaccurate data to regulators and possibly receiving payment it is not entitled to. If they are right, the hospital's admission documentation is failing to record wounds that patients already have, which is a care problem of its own. No one has measured the accuracy of the flags for this or other hospital-acquired conditions at Cimarron Heights, and coders and nurses disagree about where the errors arise. The problem this study addresses is that the hospital does not know how accurate its present-on-admission flags are for hospital-acquired conditions, in which direction they err or what documentation practices are associated with errors.

Purpose

The purpose of this quantitative study is to measure the agreement between coded present-on-admission flags and an independent re-abstraction of the medical record for four hospital-acquired conditions among adult inpatients discharged from Cimarron Heights Medical Center in fiscal 2026, and to examine whether disagreement is associated with the absence of a structured skin and risk assessment documented within 24 hours of admission.

Research Questions and Hypotheses

The study asks three questions.

RQ1: Among adult inpatients discharged in fiscal 2026 with one of the four study conditions coded as a secondary diagnosis, what proportion of coded present-on-admission flags agree with independent re-abstraction, overall and for each condition? This question is descriptive and has no hypothesis, but the study expects overall agreement between 70% and 85%, consistent with earlier research.

RQ2: Among records that disagree, are errors more often over-reports, conditions flagged as present on admission that re-abstraction finds were acquired in the hospital, or under-reports? H2: Over-reports will exceed under-reports for pressure injuries, because the audit suggests flags are being set without documentation, while the direction for other conditions is left open.

RQ3: Is the absence of a structured skin and risk assessment documented within 24 hours of admission associated with disagreement between the coded flag and re-abstraction, after accounting for condition, admission source and unit type? H3: Records without such an assessment will have higher odds of disagreement than records with one. The null hypothesis is that the odds are equal.

Definitions

Present-on-admission flag: the indicator assigned to each secondary diagnosis on an inpatient claim, with values meaning yes, no, documentation insufficient to determine, clinically undetermined and exempt from reporting. Agreement: the coded flag matches the re-abstraction flag, with documentation insufficient and clinically undetermined treated as separate categories. Over-report: coded yes, re-abstracted no. Under-report: coded no, re-abstracted yes. Structured admission assessment: a completed skin inspection and fall and venous thromboembolism risk screening in the nursing admission record within 24 hours. Re-abstraction: blind review of the full record by a credentialed coder who has not seen the original flag.

Significance

The results matter to several groups. Quality leaders need to know whether reported hospital-acquired condition rates reflect reality before they set targets. Finance and compliance need to know whether payments for these conditions are correct and whether any refunds are owed. Nursing needs to know whether admission assessments are being completed and documented, since an undocumented wound is both a data error and a missed care opportunity. For the field, the study adds recent, condition-specific evidence on flag accuracy for nursing-sensitive conditions, which earlier research did not focus on.

There is also a practical reason to study the question now. Cimarron Heights is preparing to replace its nursing documentation templates next year. If the study shows that missing admission assessments drive flag errors, the new templates can be designed to require the assessment before the admission note is signed, which would be far cheaper than correcting flags one claim at a time.

Scope, Assumptions and Limitations

The study is limited to one hospital, one fiscal year, adults and four conditions. It assumes that the full record, including nursing flowsheets and scanned outside documents, is available for re-abstraction and that re-abstraction by trained coders, checked for agreement between reviewers, is an adequate gold standard. Its main limitations are that a single site limits generalization, that a gold standard drawn from the same record cannot detect conditions no one documented and that the cross-sectional design can show association but not cause for RQ3. These limits will be addressed as far as possible in the methods milestone.

References

Bahl, V., Thompson, M. A., Kau, T.-Y., Hu, H. M., & Campbell, D. A. (2008). Do the AHRQ patient safety indicators flag conditions that are present at the time of hospital admission? Medical Care, 46(5), 516-522. https://doi.org/10.1097/MLR.0b013e31815f537f

Goldman, L. E., Chu, P. W., Osmond, D., & Bindman, A. (2011). The accuracy of present-on-admission reporting in administrative data. Health Services Research, 46(6pt1), 1946-1962. https://doi.org/10.1111/j.1475-6773.2011.01300.x

Pine, M., Jordan, H. S., Elixhauser, A., Fry, D. E., Hoaglin, D. C., Jones, B., Meimban, R., Warner, D., & Gonzales, J. (2007). Enhancement of claims data to improve risk adjustment of hospital mortality. JAMA, 297(1), 71-76. https://doi.org/10.1001/jama.297.1.71