HIM 500 Module 9 Final Project Example

Final Project: From Having a Record to Using It Well, Health IT Recommendations for Bramble Bay Medical Center

[Student Name]

Southern New Hampshire University

HIM 500: Healthcare Informatics

Final Project

[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: From Having a Record to Using It Well, Health IT Recommendations for Bramble Bay Medical Center

Executive Summary

Bramble Bay Medical Center adopted its electronic record more than a decade ago, but it has not yet turned the record into reliably useful information. Clinicians override 91% of medication alerts, most transfer patients arrive without electronic outside records and physicians spend long hours documenting after clinic. This report recommends five actions over eighteen months: join national exchange, rationalize medication alerts, pilot ambient documentation under strict conditions, adopt record rules for AI-drafted content and defer a predictive decision support package until the basics are fixed. The eighteen-month cost is about $540,000, mostly staff time and interfaces. Leadership is asked to approve the roadmap and the governance changes that support it.

Where Bramble Bay Stands

The hospital is typical of its generation. Adler-Milstein and Jha (2017) showed that federal incentives drove a rapid wave of hospital record adoption, and Bramble Bay installed its system in 2012 within that wave, configuring it quickly to meet deadlines. The results are visible today: default alerts that clinicians ignore, exchange limited to two neighboring hospitals and documentation practices built around billing rather than care. None of these problems requires a new record system. Each requires using standards, governance and design better.

How Options Were Judged

Every option went through the seven-step process adopted in Milestone One: define the need in measurable terms, set requirements, check standards and certification, review legal duties, weigh evidence, test usability and estimate five-year cost. Options were scored on weighted criteria agreed before any vendor demonstration. The legal review in Milestone Two added conditions for each technology, and the recommendations below carry those conditions.

Recommendation 1: Join National Exchange

Joining a qualified network under national trusted exchange would let Bramble Bay query records from all six regional hospitals instead of two. Holmgren and Adler-Milstein (2017) found that integrating outside information into clinicians' workflow was the least common exchange capability among US hospitals, so the recommendation includes an admission reconciliation workflow that brings outside medications, problems and results into the chart. Part 2 records must be segmented before go-live. The target is outside records for 80% of transfers within a year.

Recommendation 2: Rationalize Medication Alerts

The Module Six review found that most overridden alerts were overridden for good reason, while a few dangerous overrides, especially for renal dosing, were hidden among them. Retiring low-value alerts, reserving interruptions for high-severity risks and redesigning renal dosing alerts to recommend a dose follow the features that Kawamoto et al. (2005) linked to effective decision support. A decision support committee will govern new alerts. The target is a 60% reduction in interruptive alerts per 100 orders with fewer inappropriate overrides of high-severity alerts in quarterly review.

Recommendation 3: Pilot Ambient Documentation Under Conditions

Documentation burden is real. Sinsky et al. (2016) found that ambulatory physicians spent nearly two hours on the record and desk work for every hour with patients, and the hospital's own logs show outpatient physicians spending well over an hour each evening finishing notes. Tierney et al. (2024) reported that a large health system's ambient scribe rollout was generally well received and reduced perceived burden. A 16-week pilot with 20 clinicians is recommended, conditional on contract terms that limit recording retention to 30 days, prohibit model training on Bramble Bay recordings without written approval and require breach notice within 48 hours. Health information staff will audit note accuracy monthly, and the pilot will stop if serious errors exceed the agreed threshold.

Recommendation 4: Adopt Record Rules for AI-Drafted Content

Before any AI tool writes in the record, the hospital should adopt rules that keep the signing clinician as author, mark notes drafted with ambient tools, define signing as attestation of full review, route corrections through the amendment policy, exclude recordings from the legal record and require coder review of AI-suggested codes. The health information department should own these rules and the audits that enforce them.

Recommendation 5: Defer Predictive Decision Support

The vendor's predictive package should wait twelve months. Adding new alerts before rationalization would deepen alert fatigue, and the models have not been validated in patients like Bramble Bay's. After rationalization, each model should be tested silently against local outcomes before any clinician sees its scores. If a model performs well in the silent trial, it will enter the same evaluation process as any other tool, including a review of the developer's disclosures about how it was built and a check that its alerts fit within the rationalized alert budget.

Roadmap and Costs

Table 1 sequences the recommendations over eighteen months to fit the informatics team's capacity of roughly one major project per quarter, with record rules adopted first because they are policy work.

Table 1. Eighteen-Month Roadmap and Estimated Costs

Note. Estimates by the author with finance and informatics; mostly staff time, interfaces and licenses.

Governance, Measures and Risks

The informatics steering committee will own the roadmap, with a decision support committee for alerts and the health information department for record rules. Progress will be reported quarterly on five measures: outside records available for transfers, interruptive alerts per 100 orders, inappropriate high-severity overrides, after-hours documentation time among pilot clinicians and note accuracy in the ambient audit. Cresswell et al. (2013) found that large health IT efforts depend on leadership, user involvement and ongoing support after launch, so each project will have a clinical champion and a post-launch support period. The main risks are staff capacity, vendor contract terms and clinician trust; each is addressed by sequencing, conditions set before signing and transparent reporting.

Conclusion

Bramble Bay does not need a new record; it needs to use the one it has with better standards, cleaner decision support, stronger exchange and clear rules for new tools. The five recommendations, sequenced over eighteen months and governed with measurable targets, would move the hospital from having a record to using it well.

References

Adler-Milstein, J., & Jha, A. K. (2017). HITECH Act drove large gains in hospital electronic health record adoption. Health Affairs, 36(8), 1416-1422. https://doi.org/10.1377/hlthaff.2016.1651

Cresswell, K. M., Bates, D. W., & Sheikh, A. (2013). Ten key considerations for the successful implementation and adoption of large-scale health information technology. Journal of the American Medical Informatics Association, 20(e1), e9-e13. https://doi.org/10.1136/amiajnl-2013-001684

Holmgren, A. J., & Adler-Milstein, J. (2017). Health information exchange in US hospitals: The current landscape and a path to improved information sharing. Journal of Hospital Medicine, 12(3), 193-198. https://doi.org/10.12788/jhm.2704

Kawamoto, K., Houlihan, C. A., Balas, E. A., & Lobach, D. F. (2005). Improving clinical practice using clinical decision support systems: A systematic review of trials to identify features critical to success. BMJ, 330(7494), Article 765. https://doi.org/10.1136/bmj.38398.500764.8F

Sinsky, C., Colligan, L., Li, L., Prgomet, M., Reynolds, S., Goeders, L., Westbrook, J., Tutty, M., & Blike, G. (2016). Allocation of physician time in ambulatory practice: A time and motion study in 4 specialties. Annals of Internal Medicine, 165(11), 753-760. https://doi.org/10.7326/M16-0961

Tierney, A. A., Gayre, G., Hoberman, B., Mattern, B., Ballesca, M., Kipnis, P., Liu, V., & Lee, K. (2024). Ambient artificial intelligence scribes to alleviate the burden of clinical documentation. NEJM Catalyst, 5(3). https://doi.org/10.1056/CAT.23.0404