Learning to Do No Harm - Annals of Internal Medicine: Fresh Look Blog


Wednesday, September 15, 2021

Learning to Do No Harm

The admission pager pierced the quiet of the workroom and read, “Mr. L admission for heart failure, frequently leaves against medical advice.” Combing through his chart, I saw documentation from several providers: “Patient reported having to pay bills, left against medical advice on day 3,” “Patient reports family emergency, left against medical advice on day 4,” and “Patient left on day 3 without clear etiology though I suspect opioid withdrawal.” As highlighted in an Annals of Internal Medicine article (1), frontline providers have failed to learn how to care for those with opioid use disorder. With this and prior admissions in my mind, I went to the emergency department to meet him.

Mr. L was an older gentleman. You could hear him before you saw him; short, loud gasps filtered through the ajar emergency department door, signaling the labored breathing on the other side. While introducing myself to Mr. L, I noticed that he seemed very uncomfortable. He was in constant motion: He would try to sit up to improve his breathing, only for his legs to swell slightly, tightening his already taut skin, so he would lay back down hoping to find relief for his legs. I asked Mr. L why he came to the hospital and how he was feeling. He shared that since his last hospitalization, he had begun to get progressively short of breath. He used to be able to get to his mailbox, but today he had to ask his elderly father to lift him from the bed. He was ashamed that he had to ask his nearly 80-year-old father for help, and so he thought that he should come in to get some help.

After I understood why Mr. L came in, I asked him about his relationship with drugs. Timidly, he shared that he started buying prescription pills after being injured, and when they became too expensive, he started to use heroin. He had heard of naloxone, but never had any at home, and he often injected with his dad’s insulin needles. He had tried several times to stop but always found it difficult. Slowly, piecing his story together, I asked, “How many days after stopping heroin do you withdraw?” He responded, “Three days, doc.”

His answer unlocked why he left the hospital. With this lens, I examined his chart again. During his past 4 admissions, he had been offered buprenorphine–naloxone twice; however, the medication treatment was started on days 1 and 2 of admission, worsening his withdrawal symptoms. He described that his anxiety and diarrhea always made him want to use heroin again because he didn’t know any other way to make his symptoms stop. After seeing this in his chart, I went to Mr. L's room and asked his permission to start treatment with medications that would help his symptoms. I reassured him that going through withdrawal is difficult, and I asked him 1 favor: If he felt like he had to leave the hospital, he would talk to me before doing so.

Over the next few days I asked Mr. L how his breathing was and if he was diuresing with the medications we were giving him. In addition, I discussed harm reduction practices and the community resources that were available to him for clean syringes, bottled water, and naloxone kits. These skills were not taught to me in medical school or residency but were gained via waiver seminars pursued on my own time. I supplemented this training by asking to rotate on a hospitalist service designed to care for patients with opioid use disorder and facilitate their transition to the community (1).

On day 3, Mr. L asked his nurse for loperamide and then clonidine to help control his symptoms—medications he had never received during his previous admissions. I was anxious about day 4, fearful that when I came in that morning I’d find out that Mr. L had eloped overnight. However, when I printed my list, I was thankful he was still here. Entering his room, Mr. L exclaimed, “Doc, I’m still here! I think I’m ready to go. I just wanted to wait till you got back.”

It was hard to argue with him; he had lost 10 pounds and was now 24 hours off oxygen. We had explored starting treatment with buprenorphine–naloxone; however, given his previous attempts precipitating withdrawal, he declined. He asked me to write the names of his “withdrawal pills” down because he was going to ask his doctors if he can have them the next time he comes to the hospital.

On paper, this admission seems remarkably similar to prior admissions. Mr. L arrived with decompensated heart failure, and after receiving diuretic medications, asked to leave the hospital on day 4. However, my hope is that this hospitalization was different. I hope that Mr. L had less anxiety and fear while withdrawing from opioids, I hope that I gave him the tools to ask for symptom control on his next admission, and I hope that harm reduction teachings will help him to be safer while in the community. The skills I developed to help care for Mr. L are skills that should be taught in undergraduate medical education and reinforced in graduate medical education, instead of requiring individual training. As Weimer and colleagues (1) highlight in their article, physicians frequently do not understand the pathophysiology of opioid use disorder, the stigma associated with the disease, and that the training needed to gain these skills is not always supported with time or funding. With the pace of opioid-related deaths increasing, the opioid pandemic is in its infancy and will require education and policy to change to help our patients (2).  


  1. Weimer MB, Tetrault JM, Fiellin DA. Patients with opioid use disorder deserve trained providers. Ann Intern Med. 2019;171:931-932. [PMID: 31766053] doi:10.7326/M19-2303
  2. Centers for Disease Control and Prevention. Drug overdose deaths. Accessed at www.cdc.gov/drugoverdose/deaths/index.html on 23 July 2021.

No comments:

Post a Comment

By commenting on this site, you agree to the Terms & Conditions of Use.