HIM 675 Module 5 Final Project Milestone Two Example

Final Project Milestone Two: From Bedside to Claim. Literature Synthesis and Conceptual Framework for the Cimarron Heights Present-on-Admission Study

[Student Name]

Southern New Hampshire University

HIM 675: Research Methods and Evaluation

Final Project Milestone Two

[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 Two: From Bedside to Claim. Literature Synthesis and Conceptual Framework for the Cimarron Heights Present-on-Admission Study

The Module Two review described individual studies of present-on-admission accuracy. This milestone does something different. It argues from the literature toward the study proposed for Cimarron Heights Medical Center, a composite teaching hospital in Tulsa, and it organizes what is known into a conceptual framework that names the study's variables and the relationships between them. The synthesis is built around three claims, each supported by more than one source, and the framework that follows shows where in the path from bedside to claim the study will look for errors.

Claim One: The Flag Changes What Hospitals Report

The flag's purpose is to tell apart what a patient carried in the door from what developed during the stay, and the evidence shows that this separation materially changes reported performance. When one academic system applied present-on-admission values to its patient safety indicators, rates fell for nearly every indicator and significantly for several, including pressure ulcers (Bahl et al., 2008). Goldman et al. (2011) reached the same conclusion from the opposite direction, warning that flags should be made more accurate before being used in assessments tied to payment. Together these studies establish that the flag is consequential, which justifies the effort of measuring it.

Claim Two: The Flag Is Often Wrong, in Both Directions

The best direct evidence comes from blind re-abstraction. Goldman et al. (2011) found that coded flags matched re-abstraction in roughly three of four California records and that errors split nearly evenly between over-reporting and under-reporting, with hospital type predicting which direction dominated. That pattern matters for the Cimarron Heights study in two ways. Because errors run both ways, a single accuracy figure would hide what is happening, so the study must measure over- and under-reporting separately. And because teaching hospitals tended to under-report, Cimarron Heights' apparent over-reporting of pressure injuries would be unusual for its type and worth explaining.

Claim Three: Errors Begin in Documentation

Coders can assign a flag only from what clinicians record, so the quality of documentation at admission sets the ceiling on the flag's accuracy. Weiskopf and Weng (2013) describe record data quality in terms of completeness, correctness and concordance, among other dimensions, and a missing admission skin finding is a failure of completeness that makes a correct flag impossible. For pressure injuries, the tools for a complete admission record exist. The Braden Scale, developed by Bergstrom et al. (1987), structures risk assessment, and the revised staging system described by Edsberg et al. (2016) gives nurses shared definitions for each stage. When a structured assessment is completed and documented within 24 hours, coders have evidence for the flag. When it is not, they must infer, query or default. The literature does not directly test this link for flags, which is the gap the study's third research question addresses.

The Conceptual Framework

The framework traces each flag through four steps. In the first, a condition is either present or absent when the patient arrives; this is the true state the study tries to reconstruct through re-abstraction. In the second, clinicians document or fail to document that state, which the study captures through the presence and timing of a structured admission assessment, the variable at the heart of RQ3. In the third, the coder reads the available documentation and either assigns a flag, queries a clinician or defaults; the study records whether a query occurred and how much coding experience the coder had. In the fourth, the flag appears on the claim, where it is compared with re-abstraction to produce the outcome measures of agreement, over-report and under-report.

Contextual factors act on every step. Admission source matters because patients transferred from nursing homes or arriving through the emergency department may carry conditions that are harder to see or document quickly. Unit type matters because staffing and workflows differ between intensive care and medical floors. Condition matters because a fall injury is obvious at the moment it happens, while a deep tissue injury may not appear for days. These factors become control variables in the analysis.

How the Framework Guides the Study

The framework does three jobs. It defines the outcome precisely, as agreement between step four and the true state at step one. It identifies the main explanatory variable, documentation at step two, and the conditions under which its effect should be strongest. And it shows what the study cannot see: a condition present at arrival that no one ever documented will look absent to both the coder and the re-abstractor, so the study will underestimate under-reporting for undocumented conditions. That limitation is built into any record-based design and will be stated plainly in the methods.

Rival Explanations

A good framework also names explanations that would compete with the study's main hypothesis. Three are plausible at Cimarron Heights. The first is coder habit: some coders may default to yes for patients transferred from nursing homes regardless of documentation, which would produce errors even when an assessment exists. The second is query practice: if coders rarely query nurses, missing documentation may matter less than whether anyone asks. The third is timing of discovery: deep tissue injuries and some blood clots surface after admission even though they began before it, so disagreement may reflect genuine clinical uncertainty rather than poor practice. The study addresses the first two by recording admission source, coder experience and queries, and the third by analyzing clinically undetermined values separately. If documentation remains associated with disagreement after these factors are considered, the case for the hypothesis is stronger. Stating these rivals in advance also keeps the analysis honest, because it commits the study to testing them rather than explaining them away after the results arrive.

Conclusion

The literature supports three claims: flags change what hospitals report, flags are often wrong in both directions and documentation at admission sets the limit on their accuracy. The framework turns those claims into a sequence of measurable steps and places each study variable within it. Milestone Three will translate the framework into a design, a sample, an instrument and an analysis plan.

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

Bergstrom, N., Braden, B. J., Laguzza, A., & Holman, V. (1987). The Braden Scale for predicting pressure sore risk. Nursing Research, 36(4), 205-210. https://doi.org/10.1097/00006199-198707000-00002

Edsberg, L. E., Black, J. M., Goldberg, M., McNichol, L., Moore, L., & Sieggreen, M. (2016). Revised National Pressure Ulcer Advisory Panel pressure injury staging system. Journal of Wound, Ostomy and Continence Nursing, 43(6), 585-597. https://doi.org/10.1097/WON.0000000000000281

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

Weiskopf, N. G., & Weng, C. (2013). Methods and dimensions of electronic health record data quality assessment: Enabling reuse for clinical research. Journal of the American Medical Informatics Association, 20(1), 144-151. https://doi.org/10.1136/amiajnl-2011-000681