HI501 Unit 6 Adoption Evaluation Example

Paper Scripts After Six O'Clock: A Unit 6 Adoption Evaluation of Electronic Prescribing for Controlled Substances in Urgent Care, Framed by UTAUT

[Student Name]

Purdue University Global

HI501: Health Informatics

Unit 6 Adoption Evaluation

[Instructor Name]

[Date]

Harlow Ridge Health, its clinics, its staff and every figure are composite; the theory, regulations and studies cited are real.

What Adoption Means Here

In this evaluation, adoption means use observed in the prescribing logs: the share of controlled substance prescriptions sent electronically rather than printed. Training completion and a satisfaction survey were available but set aside. Nearly every prescriber completed training, and the survey's average score was favorable, yet neither says what prescribers actually do at 8 p.m. with a patient waiting. In this composite state, paper prescriptions for controlled substances remain legal, so electronic use is voluntary, which makes the logs a fair measure of choice.

Theory

UTAUT, the unified acceptance model that Venkatesh and colleagues assembled from eight earlier theories, proposes four main influences on use: performance expectancy, the belief that the system helps with the job; effort expectancy, how easy it is; social influence, whether important others expect its use; and facilitating conditions, whether the organization and infrastructure support it. It adds four moderators: gender, age, experience and voluntariness (Venkatesh et al., 2003). Reviews of acceptance models in health care warn that they explain intentions better than behavior and can miss the work context (Holden & Karsh, 2010), which is why this evaluation starts from behavior in the logs and uses the theory to organize explanations rather than to predict. The data allow analysis of experience, since employed prescribers and locum prescribers differ sharply, but not of gender or age, which the logs do not record.

How the Evaluation Was Done

Log data came from the prescribing module's audit trail for the six months after go-live, extracted by the clinical informatics analyst with prescriber identities replaced by a role code: employed physician, employed advanced practice provider or locum. Each controlled prescription was classed as electronic or printed, and each signing attempt as successful at first try or retried. Shifts were assigned by the time of signing. Interviews lasted about fifteen minutes, were held in person or by phone between patients, and followed six open questions, one or more for each construct, plus a closing question about anything else that made electronic prescribing harder. Two reviewers coded the interview notes separately and then agreed on a final code for each statement. No patient information was collected.

Results From the Logs

Six months of composite log data across six clinics:

1. Controlled prescriptions per month: about 1,180.

2. Share sent electronically, day shifts, 8 a.m. to 6 p.m.: 84 percent.

3. Share sent electronically, evening shifts, 6 p.m. to 10 p.m.: 57 percent.

4. Median time to sign one controlled prescription electronically, including the second factor: 38 seconds, against 9 seconds for a noncontrolled prescription.

5. Second-factor attempts that failed and were retried: 11 percent by day, 23 percent in the evening.

6. Share of evening controlled prescriptions written by locum prescribers: 46 percent; by day, 12 percent.

The evening gap is not a story about attitude: almost half of evening prescribing is done by people the system was never set up to serve.

Results From Interviews

Short interviews with 8 prescribers, 5 employed and 3 locum, and 4 medical assistants were coded against the four constructs. Performance expectancy was high. Employed prescribers valued fewer pharmacy callbacks about illegible or altered scripts and liked that the prescription and the state monitoring check sat in one workflow. Locums agreed in principle. Effort expectancy was where the system lost ground. Federal rules require two-factor authentication to sign each controlled prescription (21 C.F.R. § 1311.115), and Harlow Ridge implemented the second factor as a push notification to the prescriber's phone. Two clinics have weak mobile signal in their back offices, and the push often arrives late or not at all, which explains much of the evening retry rate when the front office is closed and prescribers chart in the back. One prescriber said, "I can print in five seconds or wait for my phone to find a bar." Social influence was moderate. The medical director uses the system and has asked others to, but evening shifts have no lead physician on site. Facilitating conditions explained the largest share of the gap. Before a prescriber can sign controlled prescriptions electronically, their identity must be verified and their credentials bound to the system (21 C.F.R. § 1311.105), which at Harlow Ridge takes about two weeks. Locum prescribers, who work short assignments and mostly cover evenings, are often not enrolled during their first weeks and sometimes never. They are not choosing paper; they have no other option.

Discussion

Read through the theory, the evening pattern has two causes that look like resistance but are not. The first is a facilitating condition: locums cannot use the system because enrollment is slow. The moderator that matters is experience, in a narrow sense: experience with this organization's enrollment process, which locums lack. The second is effort: a second factor that depends on phone signal fails most often in the rooms where evening prescribers work. Performance expectancy and social influence are not the problem, and an awareness campaign aimed at them would spend effort where the barrier is not. This matters because the usual response to low adoption is more training and more reminders from leadership. Here, both would land on people who already believe in the system and already use it whenever they can. The evening prescriber who prints a script is, in most cases, either not enrolled or standing in a room where the phone cannot receive the push. Neither problem yields to persuasion.

Recommendations

1. Start identity proofing and enrollment for locum prescribers during credentialing, before their first shift, with a target of 90 percent of evening controlled prescriptions written by enrolled prescribers within three months.

2. Offer a hard token or a code-based second factor that works without signal as an alternative to the push notification, for the two clinics with poor coverage.

3. Report the electronic share by shift and by prescriber type monthly, so the evening rate is visible to the medical director.

4. Re-run this evaluation at twelve months, adding the share of printed controlled prescriptions written by enrolled prescribers, which would isolate the effort problem once enrollment is fixed.

Limits

The interviews are few, and locum prescribers were hardest to reach, so their views are thin. The logs show what was sent but not why each printed script was printed. And because use is voluntary, a change in state law requiring electronic prescribing would change the meaning of these figures entirely.

References

21 C.F.R. § 1311.105 (2024). https://www.ecfr.gov/current/title-21/section-1311.105

21 C.F.R. § 1311.115 (2024). https://www.ecfr.gov/current/title-21/section-1311.115

Holden, R. J., & Karsh, B.-T. (2010). The technology acceptance model: Its past and its future in health care. Journal of Biomedical Informatics, 43(1), 159-172. https://doi.org/10.1016/j.jbi.2009.07.002

Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27(3), 425-478. https://doi.org/10.2307/30036540