He Called Himself Architect of My 47-Page Opioid Safety Protocol — Then the Joint Commission Asked Him to Explain the 41% Threshold
He listed himself as Protocol Architect on my opioid safety work — then the Joint Commission asked him to explain the threshold data.
Wren Alcott had been working on version 14 of the protocol since 6:40 AM.
Her desk at St. Dunstan Medical Center pharmacy was in the corner by the window — a position she had chosen when she moved to the medication safety role because the window faced east and the morning light was useful.
The pharmacy protocol was open on her screen: a 47-page decision tree governing when a clinical pharmacist at St. Dunstan could authorize an override of a physician’s opioid prescription.
She had written all 47 pages.
She had written all 14 versions of those pages, beginning three years ago with a 12-page draft and ending here, with version 14, which had reduced over-dispensing events at St. Dunstan by 41% across three years of implementation.
She had validated every step with clinical data she had pulled herself.
She had run the analysis herself.
Nobody else at St. Dunstan did this work at this level of specificity.
Terri Owens — her junior colleague, 29, two years out of pharmacy school — was at the desk across the room.
She said: “Section 4. The threshold parameters. Can you walk me through the logic again?”
Wren said: “The threshold is 40 milligrams of morphine equivalent per day for outpatient settings. That’s the cutoff below which override authorization is standard.”
She said: “Where does that number come from?”
Wren said: “It comes from 18 months of our own incident reports. Not from national benchmarks — from St. Dunstan’s data. I ran the analysis in 2022 and 2023. The national benchmark at the time was 50 milligrams. Our internal data showed 40 was more appropriate for our patient population and our prescribing patterns.”
She said: “And the decision tree below the threshold?”
Wren said: “Three criteria, in order. Patient history, indication, and prior authorization status. If all three are present and within parameters, the override is standard. If any one is missing, escalation is required. The logic is in section 4B, lines 12 through 34.”
Terri said: “And if a pharmacist disagrees with the physician’s indication?”
Wren said: “That’s section 6. Documented disagreement, clinical rationale on file, escalation to clinical pharmacist supervisor. The decision tree doesn’t eliminate judgment. It structures when judgment is required.”
Terri looked at the screen.
She said: “This is the most complicated document I’ve read in two years.”
Wren said: “It’s the most complicated document I’ve written in sixteen years.”
She said it without pride and without apology.
The glasses on her forehead had been there for three hours.
She did not know this.
The Joint Commission accreditation schedule arrived by email at 11:53 AM.
It was from the hospital’s accreditation coordinator, forwarded from TJC.
She opened it.
Interview schedule: Pharmacy Protocol — Opioid Safety Initiative.
She scrolled.
Interview Panel — Primary Contact: Dr. Elliott Marsh, Chief of Pharmacy. Role: Protocol Architect, Opioid Safety Initiative.
She read that line.
She scrolled further.
Support Staff: Wren Alcott, MTM-BC. Role: Clinical Support Staff.
She read both lines.
She opened the protocol on her second screen.
Version 14.
Revision history: she scrolled to the bottom.
Version 1 — Revised by: W. Alcott, 2023-01.
Version 2 — Revised by: W. Alcott, 2023-03.
Version 3 — Revised by: W. Alcott, 2023-06.
She scrolled through 14 entries.
All 14: Revised by: W. Alcott.
She closed the revision history.
She folded the accreditation email printout — she had printed it to check the date — and put it in her pocket.
She put her glasses on her face, which meant she had taken them down from her forehead, which meant she had finally noticed they were there.
She went back to the protocol.
She had a version 14 review to complete before tomorrow.
There had been a time — three years ago, the week she finished version 1 — when Elliott Marsh had read the first draft and said: “This is exactly what we need.”
He had said it like he had been waiting for someone to write it.
He had been sitting at his desk with the printed draft in front of him, and she had been standing in the doorway.
She had said: “I want to test the threshold with three months of incident data before we roll it out.”
He had said: “Take what you need.”
He had looked at her over his reading glasses.
He had said: “This is good work, Wren.”
He had meant it.
She had believed, then, that he understood what the work was and that it was hers.
(Drop “PROTOCOL” in the comments if you want to read what happened in the accreditation interview — and why Elliott couldn’t answer the question.) 👇
Hospital COO Diane Farrow passed Wren in the hallway outside the pharmacy on a Tuesday morning, two days before the accreditation team’s scheduled arrival.
She was smiling.
She said: “Wren — the TJC team is so impressed with the pharmacy work. Elliott’s protocol write-up is excellent.”
Wren said: “The protocol is mine. I wrote all 14 versions.”
Diane stopped.
She was carrying a clipboard and a coffee cup.
She held the clipboard slightly lower.
She said: “I see.”
She said it carefully.
She was reconsidering something.
She said: “I’ll need to speak with Elliott.”
She was already adjusting her route.
She did not call Wren later.
She called Elliott.
There had been a time — year 1 of the protocol, the month before the first implementation — when Elliott had introduced Wren at an administration meeting.
He had said: “Wren Alcott, our clinical pharmacist, has been implementing our new medication safety initiative.”
Diane had been in the room.
She had nodded.
She had looked at Wren.
Wren had said: “The protocol is now active on all inpatient units. We’ll have three-month outcome data in October.”
She had said it because it was true and it was what the room needed to know.
Diane had written something down.
She had not asked Wren’s last name.
She had looked at Elliott for the next question.
Wren had believed, then, that her work would speak for itself.
She had been correct that it would speak.
She had underestimated how long that would take.
The morning before the TJC interview, Wren was in the hallway outside the accreditation conference room.
She was listed as Clinical Support Staff — available if needed.
She had brought her tablet.
She had the incident data files on the tablet because she always had the incident data files with her during accreditation periods.
She had the protocol — all 14 versions — in the file app.
She had read the interview question list, which the accreditation coordinator had distributed to all support staff as standard procedure.
She had read question 4: Section 4 threshold parameters — can you walk us through the clinical rationale for the 40mg morphine equivalent cutoff?
She had read it standing in the hallway.
She had opened the incident data file.
She had sat in the chair provided for support staff.
She had read through the 2022 analysis — the one she had run when she was determining whether 40 or 50 milligrams was the appropriate threshold for St. Dunstan specifically.
She had read through the 2023 update.
She knew this data better than she knew most things.
She had been sitting in the chair for forty minutes when the conference room door opened and the TJC assistant appeared.
She was already standing.
Elliott was in the conference room with Dr. Greta Nolan, the TJC lead surveyor, and two colleagues.
He had been describing the protocol outcomes.
He was accurate about the outcomes.
He was precise about the 41% reduction in over-dispensing events.
He was confident.
Then Greta said: “Dr. Marsh, can you walk us through the clinical decision logic in Section 4 — specifically, the threshold parameters for override authorization? Forty milligrams morphine equivalent seems specific. Can you explain the internal data that informed that number?”
Elliott said: “The threshold was derived from current national benchmarks and best practices in opioid safety management.”
Greta said: “The national benchmark at the time of protocol development was 50 milligrams, based on the CDC guidelines. Your protocol uses 40. What internal data supported the lower threshold?”
Elliott said: “The protocol accounts for our patient population.”
Greta said: “Do you have the internal data analysis that informed the threshold parameters?”
There was a pause.
Elliott said: “The internal analysis is in the supporting documentation.”
Greta said: “We’d like to review it. And speak with whoever ran the analysis.”
She turned to her assistant.
She said: “Is the clinical pharmacist available?”
The assistant opened the door.
Wren was already standing.
The morning before the TJC interview, Wren counted the revision history entries one more time.
She was at her desk at 7:15 AM, before Terri arrived.
She opened version 14.
She scrolled to the revision history.
14 entries.
All 14: Revised by: W. Alcott.
She looked at the pre-accreditation interview schedule, which she had read three times since receiving it.
Protocol Architect: Dr. Elliott Marsh.
Clinical Support Staff: Wren Alcott.
She looked at the revision history.
She looked at the interview schedule.
She put the interview schedule in her desk drawer.
She closed the revision history.
She opened the incident data file.
She had work to do before the interview.
She opened the 2022 analysis — the one she had run when she was determining the 40-milligram threshold.
She read through it.
She read through the 2023 validation.
She read through section 4B, the decision tree logic.
She had read these documents many times.
She read them again.
When the assistant came to the hallway at 9:45 AM to ask whether the clinical pharmacist was available, she was standing.
There had been a moment — version 6, two years into the protocol development — when Diane Farrow had stopped Elliott in the elevator and said: ‘I hear the pharmacy numbers are improving.’
Elliott had said: ‘The protocol is working.’
He had not mentioned Wren.
Diane had said: ‘Good work.’
The elevator had opened.
They had both gotten off.
Wren had been at her desk at that moment, reading version 7 data.
She had not been in the elevator.
She had not known the conversation happened.
She learned about it later, from a colleague who had been in the elevator.
The colleague had said: ‘Elliott takes credit for everything, doesn’t he.’
She had said: ‘He runs the department.’
She had believed that then.
She was less certain what she believed now, reading the TJC schedule with ‘Clinical Support Staff’ beside her name.
She put the schedule in the drawer.
She opened the incident data.
She had work to do.
Elliott Marsh had read all 14 versions of the protocol.
He was sitting in the accreditation conference room with Dr. Greta Nolan’s question in front of him and he was understanding, for the first time, that reading 14 versions and knowing the logic of 14 versions were different things.
He had read them.
He had approved them.
He had signed each approval page.
He had understood the outcomes.
He had understood the methodology well enough to describe it to administration.
He had not understood the 40-milligram threshold well enough to explain why it was 40 and not 50 — because he had approved it on the basis that Wren had analyzed the data and recommended 40, and that had been sufficient for his approval.
He had never asked for the analysis.
It had not occurred to him to need it.
He had been, he was realizing now, describing a building from the outside.
He had known the address.
He had known the floor plan in general terms.
He had never been inside.
The specific moment this had been decided was a Tuesday morning in October, three years before the accreditation, when Wren had handed him the version 4 draft with the 40-milligram threshold recommendation.
She had attached a 12-page analysis.
She had said: “The national benchmark is 50. I’m recommending 40. The analysis is in the appendix.”
He had said: “If the data supports it, I trust your recommendation.”
He had initialed the approval on the cover page.
He had put the 12-page appendix in the file.
He had not read the appendix.
He had thought: she’s the medication safety specialist.
He had thought: that’s what I hired her for.
He had thought: my role is to ensure this goes through the institutional process correctly.
He had not examined whether approving an analysis and understanding an analysis were the same thing.
He was examining it now, in the conference room, with Greta Nolan waiting.
He could not explain the threshold.
He had not understood it.
He had approved it.
He had put his name on the pre-accreditation materials as Protocol Architect and had not understood the threshold that was the center of the protocol.
He looked at the door.
He said: “Let me get the clinical pharmacist.”
In the hallway, Wren had been sitting in the support staff chair with the incident data on her tablet.
She had been sitting there for forty minutes.
She had read through the 2022 analysis.
She had read through the 2023 update.
She had read section 4 of the protocol from memory, without needing the screen.
She knew why the threshold was 40.
She knew it was not a national benchmark.
She knew it was 18 months of St. Dunstan’s own incident data, her own analysis, her own recommendation, validated twice.
She had been sitting with the glasses in her hand — she had taken them down from her forehead when the assistant appeared, for no particular reason except that she wanted something to hold.
The paper on the tablet in front of her.
The glasses in her hand.
She was not nervous.
She had the data.
She had always had the data.
When the assistant opened the door, she was already standing.
Elliott had also read the pre-accreditation materials before submitting them to TJC.
He had read his own role: Protocol Architect.
He had read her role: Clinical Support Staff.
He had thought: I am the department head. The department head is the protocol architect. That is what that means.
He had not looked up whether TJC had a definition for Protocol Architect that differed from his definition.
He had not asked whether his definition and the licensing board’s definition were the same.
He had submitted the materials.
He had been confident in them.
He was in the conference room now understanding that confidence and knowledge were different things.
He was in the conference room now understanding that approving all 14 versions of a document and knowing what was inside all 14 versions of a document were different things.
He had approved.
He had not known.
He had put his name on the pre-accreditation materials as Protocol Architect.
He had been wrong in the specific way that people are wrong when they do not examine their own frameworks.
He had not examined his framework.
He was sitting with the consequence of not examining it.
The door was closed.
Wren was in the hallway with the data.
He had to let her in.
Wren came into the conference room and sat down.
She set her tablet on the table.
She put her glasses back on her forehead — she had been holding them since the assistant came to the door.
She set the incident data file open on the tablet.
Greta Nolan said: “Thank you for joining us. I’m Dr. Nolan, TJC lead surveyor for the pharmacy review. Can you walk us through the threshold parameters in Section 4? Specifically, why 40 milligrams morphine equivalent rather than the national benchmark of 50?”
Wren said: “Because the national benchmark doesn’t account for our patient population and prescribing patterns. In 2022, I ran an analysis of 18 months of St. Dunstan’s incident data — over-dispensing events, outcomes, contributing factors. The analysis showed that our over-dispensing rate was highest in the 40-to-50 milligram range. Setting the threshold at 50 would have missed the majority of our problematic prescribing. Setting it at 40 required override review for a larger class of prescriptions. The 41% reduction in over-dispensing events since implementation is the outcome.”
Greta said: “Do you have that internal data analysis?”
Wren opened the file on the tablet.
She said: “This is the 2022 analysis. And this is the 2023 validation update.”
She turned the tablet toward Greta.
Greta looked at it.
She looked at it for a full minute.
She said: “This is the most rigorous protocol documentation I’ve seen in a community hospital setting.”
She said it to Wren.
She was not looking at Elliott.
She said: “Can you walk me through the override decision logic in 4B?”
Wren said: “Three criteria, in order. Patient history, indication, and prior authorization status. If all three are present and within parameters, the override is standard. If any one is missing, escalation is required.”
Greta said: “And the escalation pathway?”
Wren said: “Clinical pharmacist supervisor. Documented disagreement on file. The logic doesn’t eliminate judgment. It structures when judgment is required.”
Greta said: “Who designed this structure?”
Wren said: “I did.”
Greta wrote in her notes.
She wrote for thirty seconds.
She said: “We have what we need for this section. I’d like to return to the implementation data in the afternoon session.”
She looked at Wren.
She said: “Thank you. We’ll want you in the afternoon session as well.”
She said it directly.
She was not asking Elliott.
Elliott did not speak for the remaining four minutes of the morning session.
Terri Owens was at the pharmacy counter when the morning session ended.
She had been there for the last two hours, covering the counter while Wren was in accreditation support.
She had told the TJC assistant at 9:15 AM — when the assistant came to the counter to ask where the clinical pharmacist was — “Wren Alcott. She’s in the hallway.”
She had said it without hesitation.
When the conference room door opened and the morning session ended, Terri saw Wren come out first.
She saw Elliott come out second, behind her.
She saw that Greta Nolan was still talking to Wren at the doorway of the conference room.
She did not say anything.
She nodded once when Wren passed the counter.
She went back to the prescription she was reviewing.
Diane Farrow received the TJC preliminary finding by email at 4:47 PM.
She read it.
She read the attribution line: Protocol Design — Wren Alcott, MTM-BC, Clinical Pharmacist, St. Dunstan Medical Center.
She forwarded it to Elliott.
She did not add a message.
Elliott found Wren at her desk at 5:15 PM.
She was reviewing the afternoon data she would present the following day.
He stood in the doorway.
He said: “I didn’t know the threshold had internal data behind it.”
She said: “I sent you the incident analysis in October 2022. It was the appendix to version 4.”
He said: “I approved the protocol. I didn’t read the appendix.”
She said: “I know.”
She went back to the tablet.
The conversation was two minutes.
He left.
The following morning, at 6:54 AM, before Terri arrived at the pharmacy, Elliott filed the correction with TJC.
He typed it himself.
Protocol Architect: Wren Alcott, MTM-BC.
Department approval and administrative lead: Dr. Elliott Marsh.
He looked at what he had typed.
He submitted it.
He went to the kitchen.
He came back to his desk.
He opened a new document.
He wrote: Re: State Pharmacy Board — Medication Safety Award Nomination.
He wrote: The St. Dunstan opioid protocol was designed by Wren Alcott, MTM-BC, Clinical Pharmacist.
He wrote: The protocol is entirely her intellectual and clinical work.
He signed it.
He did not list himself as co-nominee.
He submitted it to the state pharmacy board nominations portal at 7:09 AM.
He closed the document.
He had an 8 AM patient safety committee meeting.
He prepared for it without looking at the nomination again.
The afternoon session with TJC lasted 38 minutes.
Greta Nolan asked about the implementation methodology.
Wren answered for 22 minutes.
She described the staggered rollout across 7 inpatient units.
She described the pharmacist training program she had designed to accompany the protocol.
She described the feedback loop built into the first three months: weekly outcome reviews, two adjustments in month two, the decision to hold the threshold at 40 milligrams when three weeks of data suggested moving higher.
She held the threshold at 40.
Month six data confirmed it.
Elliott was in the room for 38 minutes.
He did not speak.
When the session ended, Greta shook Wren’s hand.
She said: ‘We will be in touch about the final report.’
She was looking at Wren.
She did not look at Elliott.
He held the door for the TJC team.
He signed the sign-out form.
He walked back to his office.
He sat at his desk without opening any documents.
Forty minutes later, he opened a new document.
He typed: Re: TJC Pre-Accreditation Materials Correction.
He did not finish it that evening.
He finished it at 6:54 AM the following morning.
The nomination he submitted for the state pharmacy board was 400 words long.
He described the protocol design process, the 18 months of incident data analysis, the implementation methodology, the outcome data.
He described all of it as Wren’s work.
He described his own role in one sentence: Department head approval and administrative oversight.
He did not elaborate on his role.
He signed the nomination.
He submitted it.
He went to the kitchen.
He came back to his desk.
He had a patient safety committee agenda to prepare.
He opened it.
He worked on it for an hour.
He did not think about the nomination again that morning.
He thought about it at 11:30 AM, when Terri Owens passed his office on the way to the pharmacy counter and he saw the pharmacy door and thought about Wren at her desk with the incident data on her screen and the glasses on her forehead.
He had seen her like that for three years.
He had thought: she is the one who does the work.
He had thought: and I am the one who leads the department.
He had believed these were compatible descriptions.
He was sitting with a more careful understanding of what compatible means.
The TJC accreditation report was published six weeks after the interview.
It arrived in Wren’s institutional email as a PDF attachment.
She opened it at her desk — 8:23 AM, a Thursday — and scrolled to the pharmacy section.
Opioid Safety Protocol: St. Dunstan Medical Center.
Protocol Design: Wren Alcott, MTM-BC, Clinical Pharmacist.
She read it.
She put the report in the tray on the corner of her desk.
She did not file it yet.
She had a version 1 of the medication reconciliation protocol to start.
She was at the same desk she had always been at.
The screen was showing a different document — 4 months of medication reconciliation incident data, 324 events, which she was reading to find the pattern that would tell her where the protocol failures were.
The reading glasses were on her forehead.
She had pushed them up when she moved from the printed incident summary to the screen — the screen was close enough to read without them, and she had been in this position for two hours.
She knew this because she had looked at the clock when she started.
She did not know the glasses were on her forehead.
This was the same as it had always been.
She was reading the incident data the way she always read incident data: by looking for the frequencies first, then the outliers, then the conditions that made outliers possible.
The pattern in the medication reconciliation data was beginning to appear.
It was in the discharge protocols.
She could see it developing across three event types, in the third and fourth months of the audit period.
She was not ready to write the protocol yet.
She needed two more months of data.
She needed to understand whether the pattern she was seeing was structural or situational.
She had been here before — at the beginning of a problem that required years to solve correctly.
The TJC report was in the tray.
She had not filed it.
She would file it when she finished the new read-through.
Her name was in it.
She knew what it said.
The pre-accreditation materials submitted to TJC — the ones listing Elliott as Protocol Architect — were in the hospital’s regulatory file.
They existed.
The correction was also in the file.
The published TJC report superseded them both.
The hospital’s three years of administration meeting minutes said: “Dr. Marsh’s medication safety initiative.”
Those minutes were in the permanent record.
She was not asking for them to be corrected.
She knew what the TJC report said.
That was the document that would matter to anyone who needed to verify her work.
The administration meeting minutes were a record of what Elliott said when he had the option of saying something else.
She was not going to explain that to anyone.
Elliott appeared in her doorway at 3:40 PM.
He said: “The board meeting is next month. I’d like you to present the protocol outcomes. Not me. You.”
She said: “I’ll need the incident data slides from the 2022 and 2023 analyses.”
He said: “They’re yours. They’ve always been yours.”
She said: “I know.”
She looked back at the screen.
He looked at the glasses on her forehead for a moment.
He did not mention them.
Nobody ever mentioned them.
He said: “I’ll let Diane know you’re presenting.”
She said: “Good.”
He left.
She found the pattern on page 6.
The glasses were warm.
The state pharmacy board letter arrived eight weeks later.
Subject: Medication Safety Award — Nomination Confirmation.
She was listed as the nominee.
The nomination letter described the opioid safety protocol as her intellectual and clinical work.
She read it.
She put it in the tray beside the TJC report.
She did not file either document immediately.
She had a read-through to finish.
The pre-accreditation materials from TJC — the ones listing Elliott as Protocol Architect — were in the hospital’s regulatory file.
The correction was in the regulatory file.
The TJC published report was the authoritative record.
Three years of administration meeting minutes said: Dr. Marsh’s medication safety initiative.
Those minutes were in the permanent record.
She was not asking for them to be corrected.
The meeting minutes were a record of what Elliott said when he had the option of saying something else.
The TJC report was what the institution said after it saw the evidence.
She knew which document would be relevant when anyone needed to verify her work.
She was not going to explain that to anyone.
Terri passed the desk at 4:17 PM.
She did not mention the glasses.
She never mentioned the glasses.
Nobody ever mentioned the glasses.
Wren did not know they were there.
She found the pattern on page 6.
The glasses were warm.
