HINF 510 Week 6 New and Advanced Technologies in Clinical Information Systems Example

An AI Scribe, a Blood Pressure Cuff That Reports Home and a Chatbot We Said No To: Evaluating New Technologies After a Health Center Network's Record Goes Live

[Student Name]

University of Phoenix

HINF/510: The Systems Life Cycle

Week 6 Assignment

[Instructor Name]

[Date]

The health center network, its pilots and decisions are composites written for a model paper; research findings and federal reports come from the sources listed.

Within a month of the final clinic group's switch to the new record, three vendors pitched the health centers: an ambient artificial intelligence scribe that drafts notes from recorded visits, a remote blood pressure monitoring program with cellular cuffs and a patient-facing chatbot to answer questions and book appointments. Each vendor promised savings and better care. The chief operating officer asked the clinical informatics committee, which had replaced the implementation steering committee, to evaluate them. This paper describes the framework used and the decisions made.

New Technology and the Life Cycle

New technologies enter the systems life cycle in the maintenance and optimization phase, often as additions to the core record. Each should pass through the same discipline as the record itself: planning, analysis of needs and workflows, design, testing, implementation and evaluation. Adding tools too soon after go-live risks overwhelming staff still learning the record.

The Evaluation Framework

The committee rated each proposal on six questions. Does it fit the network's strategy of integrated, equitable primary care? What does published evidence show? Is the network ready in infrastructure, staffing and workflow? Would any patients or staff be left out? What are the privacy and safety risks? What will it cost, and how will it be paid for?

Proposal One: The Ambient Scribe

An ambient scribe uses a smartphone microphone, with the patient's consent, to capture the conversation, then produces a draft note; nothing enters the record until the clinician edits and signs it. At one large medical group, the tool was used by 3,442 physicians in 303,266 patient encounters in the first 10 weeks after implementation, and early assessments reported less time spent documenting and more attention to patients (Tierney et al., 2024).

Evaluating the Scribe

Fit: strong, since documentation burden was the clinicians' top complaint. Evidence: promising but early, mostly from large systems and English-language visits. Readiness: good, as clinicians now use the new record smoothly. Equity: a concern, because about 40% of the network's visits are in Spanish and some use interpreters; accuracy in these visits must be tested. Privacy and safety: the tool must obtain consent, retain no audio after processing and never finalize a note without clinician review, since drafts can contain errors. Cost: about $150,000 a year for 40 clinicians, to be weighed against time saved.

Proposal Two: Home Blood Pressure Telemonitoring

The monitoring program gives patients with uncontrolled hypertension cellular cuffs that send readings to the record, reviewed by a clinical pharmacist who adjusts medications under a protocol. In a cluster randomized trial, home telemonitoring with pharmacist management brought blood pressure under control and kept it there at the 6- and 12-month checks for 57.2% of patients, against 30.0% of those in usual care, and control remained better six months after the intervention ended (Margolis et al., 2013). The cuff did not lower anyone's blood pressure; the pharmacist who acted on its readings did.

Evaluating Telemonitoring

Fit: strong, since hypertension control is a core quality measure and about 11,000 of the network's patients have hypertension. Evidence: strong for programs that pair monitoring with active management. Readiness: the record can receive device data, and the network employs two clinical pharmacists. Equity: cellular cuffs work without home internet, an advantage for patients without broadband. Privacy and safety: readings must be reviewed within set times, with escalation rules for dangerous values. Cost: devices and pharmacist time, with reimbursement uncertain for health centers.

A Caution From Federal Auditors

Remote monitoring has drawn federal scrutiny. A 2024 federal review found that about 43% of Medicare enrollees who received remote patient monitoring did not receive all three components of the service, such as education and setup, device supply and treatment management, and recommended stronger oversight of how the service is used and billed (Office of Inspector General, 2024). The network's program will document every component and focus on patients who benefit, not on volume.

Proposal Three: The Chatbot

The chatbot vendor offered a tool to answer patient questions, triage symptoms and book appointments by text, in English and Spanish. Fit: moderate, since phone lines are overloaded. Evidence: little published evidence on safety or outcomes for symptom advice in populations like the network's. Readiness: the portal and scheduling templates are only weeks old. Equity: Spanish quality was unverified, and many patients speak Indigenous languages from Mexico and Central America. Safety: symptom triage errors could delay urgent care. The committee declined the chatbot for now, agreeing to revisit it in a year for scheduling only, without symptom advice.

What the Clinicians Said

The committee asked clinicians directly. In a short survey of 88 clinicians, 71% said they would try an ambient scribe, mostly to reduce evening documentation, while several behavioral health clinicians worried that patients would speak less openly if they knew a device was listening. The committee therefore excluded behavioral health visits from the first pilot and required clinicians to ask each patient, in the patient's language, before starting the tool. Clinicians also asked that the scribe produce notes matching the new short templates rather than long narratives, which the vendor agreed to configure.

Paying for What Works

Cost and payment shaped each decision differently. The scribe will be paid from operating funds, justified only if it saves enough clinician time to add visits or reduce burnout-related turnover, which costs the network an estimated $250,000 each time a physician leaves. Telemonitoring can draw partly on grant funding for hypertension control and on payment from managed care plans with quality incentives. The chatbot's vendor offered a free first year, which the committee noted was not a reason to adopt a tool whose safety was unproven.

The Decisions

The committee approved a six-month pilot of the ambient scribe with 12 volunteer clinicians, including four who see many Spanish-speaking patients, and a telemonitoring program for 300 patients with uncontrolled hypertension at four clinics. Both pilots have defined outcomes and stop rules.

Measures

For the scribe: documentation time per visit and after hours, note accuracy reviewed monthly by clinician auditors, patient consent and satisfaction and clinician burnout scores. For telemonitoring: blood pressure control at three and six months, time from a high reading to pharmacist action, device use and cost per patient controlled.

Governance of Artificial Intelligence

Because the scribe uses artificial intelligence, the committee created a standing review for any AI tool: a description of how it was developed and validated, testing on the network's own patients and languages, a named clinical owner and monitoring for errors and unequal performance.

Conclusion

New technologies can extend a clinical information system, but only when evaluated with the same discipline as the system itself. Applying one framework to three proposals led the network to pilot an ambient scribe with attention to language and consent, launch telemonitoring backed by strong trial evidence and careful documentation and decline a chatbot whose risks outweighed the evidence. The life cycle continues, with each new tool entering through planning, testing and evaluation.

References

Margolis, K. L., Asche, S. E., Bergdall, A. R., Dehmer, S. P., Groen, S. E., Kadrmas, H. M., Kerby, T. J., Klotzle, K. J., Maciosek, M. V., Michels, R. D., O'Connor, P. J., Pritchard, R. A., Sekenski, J. L., Sperl-Hillen, J. M., & Trower, N. K. (2013). Effect of home blood pressure telemonitoring and pharmacist management on blood pressure control: A cluster randomized clinical trial. JAMA, 310(1), 46-56. https://doi.org/10.1001/jama.2013.6549

Office of Inspector General. (2024). Additional oversight of remote patient monitoring in Medicare is needed (OEI-02-23-00260). U.S. Department of Health and Human Services. https://oig.hhs.gov/documents/evaluation/10001/OEI-02-23-00260.pdf

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