The danger of ignoring your instincts


Category : Medical Rants

Recently we had a patient admitted for a diagnosis that did not really fit his problem representation. The diagnosis was a convenient one, and easily treated. He initially responded to treatment and we discharged him. The diagnosis assumption nagged at me, but I did not push forward with a test that my mind wanted.

A week later he returned (the dreaded readmission), with the same symptoms. The admitting resident expanded the treatment for the same diagnosis.

The next morning on seeing the patient we were even more uncomfortable than on the first admission.

As often happens, this is a story of community acquired pneumonia that was not CAP. We ordered a CT scan that clarified the abnormal Xrays. We reviewed the CXRs and CT with the radiologist. His symptoms never fit CAP. His CXR could have been CAP. Only the CT scan pointed us in the right direction.

I preach expanding the diagnostic evaluation when the problem representation and the illness script do no match. Yet, doing so is often difficult. Our patient’s diagnosis was delayed a week, with continued discomfort for that entire week.

So I am challenging myself. I “knew” that we did not have the right diagnosis, but “I did not pull the trigger”. I am not unusual. I suspect we all suffer from this error.

The second time I had no hesitation. How do I convince myself to honor my instincts in such patients?

I suspect you all have experienced similar situations. This story (and I have withheld some details for patient confidentiality) likely seems rather common to others.

I hope to do better the next time. Part of not doing better is refusing to rationalize what happened, but rather learn from the experience. The patient improved dramatically when we treated the right process.

My tweetorial collection


Category : Medical Rants

You can probably tell that I tweet much more than I write blog posts these days. I have started doing tweetorials – linked collections of tweets that tell a story. Most tweetorials represent teaching or rather online chalk talks.

They are now being collected (along with other tweetorials that I particularly like and want to use. Here is the link: Medrants Tweetorials

If you do not follow me on twitter or use twitter, you can still use this link to read through these teaching sessions.

The man in 558


Category : Medical Rants

Coming in to meet the students, housestaff and patients for the first day on service always excites me. This Monday was no exception. What awaited me? How many patients would I need to see? What lessons could I impart?

When I arrived we had 11 patients, 2 new and 9 had arrived previously. Going through the list, while routine, always stimulated questions and teaching opportunities. Sometimes the team had questions for me. Sometimes they had a mischievous sense of putting me on the spot. I always love that interplay.

When we got to the man in 558, they told a sad story of an angry man with terminal cancer. He had accepted this fate, but was angry that we could not control his pain. He was refusing opiates because the constipation pain was worse than his cancer pain. The entire team was avoiding seeing this man unless they absolutely had to because he would yell at them. Empathy becomes difficult when confronted with anger.

As we walked around the hospital, my mind never went to the man in 558. We had a variety of medically interesting problems. We had patients to examine and demonstrate physical findings. We had discharge planning.

But inevitably we got to 558. I went into the room to talk with him. The team had painted an accurate picture. He recounted the problem. He was obviously miserable. I told him that we would discuss a plan and involve palliative care. I explained that they were the experts at treating pain and side effects.

Outside the room I did something I rarely have to do, I had the team discuss constipation management. Usually I let the team figure this out without my interference, but in this situation I needed to understand what had been done. As they recited their heroic attempts, I wondered how I could help.

I asked the team if they had tried an opiate antagonist. Having rarely used it, I could not remember the name. So we looked it up – methylnaltrexone. We ordered it and explained the situation to the patient. We hoped it would work. We still called palliative care.

The next day, the team reported that he felt much better and that the new medication had allowed them to restart his opiates. We go to enter the room, and a woman is sitting by his bedside playing the guitar. They are singing a song from Creedence Clearwater Revival. The music therapist is smiling, as is our patient. We learn that he had been a musician – a keyboard player – and that CCR was his favorite group.

The man in 558 became fascinating to everyone. He had great stories to tell. He become the musician rather than the man in 558.

He still had a terminal disease, but he had made his peace. We were able to make his days better.

Palliative care did a great job. Make each day the best it can be. And that went for the team also.

Reflections on risk prediction


Category : Medical Rants

This Monday evening our next podcast becomes available. Dr. Gustavo Heudebert and I discuss another article about risk prediction. This topic has become a recurring theme on the Annals On Call podcast.

Why is risk prediction so important? In 2019 we make many decisions about prevention and testing based on risk prediction. In addition we also estimate harms and benefits. All these predictive models have advantages and flaws. In making a decision for statin use (another upcoming episode), we have to estimate the risk of cardiovascular events, how much taking a statin will decrease that risk, and the probability and type of side effects from taking a statin.

These predictions all come from mathematical modelling. Mathematical modelling is fraught with many hazards. One can apply the same modelling techniques to different databases and develop significantly different models. The episode with Dr. Rod Hayward – Improving Estimation of Cardiovascular Risk – gave a great example of how including different databases and even different time periods changes our risk prediction for cardiovascular events. In that episode, he also explained the problem of temporal trends. Cardiovascular risk has decreased over the past 30+ years. Thus, our predictions should change.

He also pointed out that we never have data on all the risk factors that one would want in a model. The cardiovascular risk prediction models do not take into consideration family history, renal disease or fitness – yet we know that all of these factors modify CV risk.

The same problems exist for benefit and harm prediction. The promise of big data assumes that we have complete data. Yet in medicine we never have complete data.

These predictive models can help our clinical decision making. We should always use them carefully and thoughtfully. Understanding that the numbers are estimates with confidence intervals can help us. They can help us start a conversation with our patients about risks and benefits. The numbers are not magic. We cannot enter them into a computer program to make clinical decisions. Resorting to algorithmic decision making will quickly become dangerous.

Many decisions occur in a “grey zone”. We should take into consideration that patient, his or her concerns, and the totality of their medical situation. Such decisions are complex and understanding these issues makes one wonder about many guidelines and performance measures. We should always remember HL Mencken’s quote, “there is always a well-known solution to every human problem—neat, plausible, and wrong”.

Reflections on my social media Grand Rounds


Category : Medical Rants

Over the past 2 days, after participating in Grand Rounds, I did what I always do after giving a talk. I replayed the process in my head, and tried to understand what I had done well and not as well. But this time, given the new topic, my reflections shed some light on my use of social media and the importance of the “big experiment” of Skyping with Dr. Matt Watto to discuss podcasting.

My first understanding was that blogging has mostly served me. It has given me a platform to become a better writer. But more important it has allowed me to put my ideas down on paper. This process allows me to test out ideas, concepts, and methods for making my thoughts more clear. It allows me to give the ideas a test. The ideas do not all work, but many do. Many become refined as I try to express my thinking.

Facebook on the other hand is simply (for me) a method to keep in touch. I do learn how former students and residents are faring. But this is my least important social media presence.

Twitter has become a major part of my personal continuing education. Because of my twitter habit, I have set a standard of tweeting at least one important medical concept each day. Most often these refer to an article in the literature or a podcast.

Podcasting is rapidly becoming a major method in medical education. I have a colleague who has adopted podcasts as her main method for getting CME and MOC credits. She has young children, works part-time, and would have difficulty going to conferences and classes. She finds podcasts convenient and an important contribution to her work-life balance.

Many residents listen to podcasts either when working out, or when commuting. They love medicine, and enjoy learning during those times.

These ideas seem more clear than prior to developing and giving the Grand Rounds, but none of these are as important as my “big reveal”.

The conversation with Matt clearly was the highlight of our Grand Rounds (it lasted around 20-25 minutes). Both colleagues and residents told me how much they enjoyed the conversation. Both The Curbsiders and Annals On Call are conversations.

When I listen to podcasts, I find the most interesting are those that involve conversations. So, as I was pondering this phenomenon, I wondered if this should be part of more Grand Rounds.

Residents in our program rate morning reports and Clinical Problem Solving conferences as their best learning experiences. These are conversations.

Having 2 or more people discuss a topic, clarifying thoughts and posing follow up questions, helps me absorb a topic better. The classic lecture is usually boring as hell. I have given many Grand Round lectures. After years of honing my skills I generally get good reviews. No matter how well I do, I would greatly prefer to have conversations, discuss patients and try to apply and expand knowledge to helping patients.

I think this is a personal epiphany. We need to figure out how to incorporate conversation into the major of our teaching. I hope this makes sense to you the reader. Typing this is helping me play with these thoughts.

Penicillin allergy, probably not


Category : Medical Rants

The current issue of JAMA has a wonderful review of penicillin allergy. This conclusion is important:

Many patients report they are allergic to penicillin but few have clinically significant reactions. Evaluation of penicillin allergy before deciding not to use penicillin or other ?-lactam antibiotics is an important tool for antimicrobial stewardship.

This concept has great importance. Almost every time I give a pharyngitis talk, someone asks me about second-line antibiotics for patients with “penicillin allergy”. Since penicillin (or amoxicillin) work well against group A strep, group C/G strep and Fusobacterium necrophorum, we want to use a penicillin as our first-line antibiotic.

The review and the wonderful podcast on the topic, make me now want to urge primary care physicians and student health centers should test patients who believe that they are pen allergic prior to needing penicillin.

Given the usefulness of penicillins for significant inpatient infections, we should be testing all “pen allergic” patients in advance. Penicillins remain very important antibiotics and we have inadvertently labelled too many patients as allergic.

2018 #reflection – another great year


Category : Medical Rants

In 3 weeks I will present a Grand Rounds talk on social media and internal medicine. If you had asked me if this would be a possibility when I started blogging in 2002, I would have laughed.

My social media resolution this year was to post on twitter every day with the #5goodminutes . I succeeded. What have I learned?

Having finished my residency 40 years ago, I still love internal medicine. I learned much new this year. Putting some pressure on myself to find something worthwhile to post each day forced me to read more. As usual, our learners continue to push me to become a better physician and educator.

So #medtwitter exploded (at least in my opinion). I have so many great follows that point me to exciting new information. Twitter has become a great way for me to stay connected and informed.

This was a huge year for medical podcasts. I have written about this recently. The excitement of hosting a podcast is difficult to describe. Colleagues and learners tell me how much they like these podcasts. Recording them is fun.

Here is some background on the process. I peruse the Annals of Internal Medicine regularly looking for potential podcast topics. Once I find one, I start looking for guests. Most of the guest are already colleagues, but I have found some wonderful guests previously unknown to me.

Once I find the article and guest, I carefully read and reread the article, highlighting key points. Then I send the guest an outline of the question areas and ask them if they have any suggestions for additional topics or modifications.

I hope my questions represent the questions that listeners want answered. The answers obviously sometimes influence questions that I did not plan. Often I will recapitulate the answer to make certain that I understand and I hope that helps the listeners also.

Once recording has finished, I listen to the “rough draft”, pulling out a few interesting quotes for the introduction. On that listen I often learn more than I did during the interview. I develop “Bob’s Pearls”, and record the introduction and outro.

I then send the podcast to our excellent audio experts at the Annals. They add the music and cut out the dead spaces to make the podcast smoother for listening. They then send me a link and I listen to be certain that everything sounds good.

On the Tuesday morning after the Monday afternoon release I listen once again. Each time I listen to the podcast my retention improves.

Other than medicine and social media, I continue to focus on maintaining my fitness and weight loss. From May 2013 to May 2014 I lost 40 pounds, going from a BMI of 29+ to 24+. I have now maintain that weight for 4.5 years. My fitness addiction – Orange Theory Fitness – that started in September 2015, continues full speed ahead. As of today I have taken 660 classes, making it likely that I will reach 700 classes before my 70th birthday on the Ides of March. Maintaining my weight and fitness improve my quality of life dramatically.

Finally, we just celebrated our 45th wedding anniversary yesterday. We have 2 wonderful children, a great son–in-law and a wonderful girl friend for our son. We have 2 delightful grandsons, one of whom did a marvelous job at his Bar Mitzvah this October.

I continue to work part-time at the VA. For me, part-time is 4.5 months of ward attending. I (and my colleagues and learners) do not believe that I will (or should) stop ward attending anytime soon. Each day on rounds is a chance to help our wonderful learners and deserving patients.

Thanks for reading, listening and following. I hope that I sometimes help others feel my joy and love of our field.

If I were writing sore throat guidelines


Category : Medical Rants

Several tweets asked me to answer this question. How would I rewrite sore throat guidelines? Obviously I am biased. So this is my opinion and I am sticking to it!

  1. I would not change anything about pre-adolescents. Group A strep is the most important bacterial infection and using rapid tests with backup cultures makes sense.
  2. I would change the guidelines for adolescents and young adults. I would treat patients having Centor scores of 3 or 4 with either penicillin or amoxicillin (augmentin would be fine). I would probably treat some 2s if they looked very ill. I would never use macrolides. If the patient is truly penicillin allergic and looked sick then I would use clindamycin.
  3. I would have a printed sheet for all adolescent/young adults. I will do a mockup of that sheet in the next paragraph.
  4. Patients over 30 rarely have pharyngitis. I would treat those patients similarly to adolescent/young adults if they looked sick.

Instructions for adolescents and young adults:

Sore throats should improve over 3-5 days with or without antibiotics. If your sore throat worsens, please return for repeat evaluation.

Beware of the following signs of serious infection:

  1. Rigors – shaking chills
  2. Drenching night sweats
  3. Worsening sore throat
  4. Unilateral neck swelling
  5. Shortness of breath

If you have any of those signs you will likely need intravenous antibiotics, so you should go to an emergency department as soon as possible.


Most Fusobacterium necrophorum do respond to penicillins. They rarely respond to macrolides. If the patient worsens I would either use clindamycin or add metronidazole. If the patient needs IV antibiotics I would choose either piperacillin/tazobactam, penicillin + metronidazole or clindamycin. If you have any suspicion for peritonsillar abscess or Lemierre Syndrome a limited CT of the neck should provide excellent diagnostic information. If you suspect the Lemierre Syndrome, bedside ultrasound of the internal jugular vein should show the clot.

More evidence on suppurative complications from Fusobacterium necrophorum tonsillitis


Category : Medical Rants

Authors from the University of Michigan Department of Otolaryngology have recently published a fascinating article titled Implication of Fusobacterium necrophorum in recurrence of peritonsillar abscess. This article adds to the emerging story of the dangers of incompletely treated Fusobacterium necrophorum pharyngitis.

This article has a wonderful discussion section (of course they cite our work favorably). The gist of the article is included in the abstract :

One hundred fifty-six of the 990 patients in our study developed recurrence of their abscess (16%). The age ranges most susceptible to recurrence included adolescent (22.9%) and young adult groups (17.1%) … The presence of FN was significantly more prevalent in the recurrent group (P?<?0.0001).

This article adds to our understanding of who develops peritonsillar abscess (PTA). 26% were in the 13-18 age group and 44% in the 19-30 age group. The 13-18 age group disproportionately developed recurrent PTA.

The article supports the Danish data that shows Fusobacterium necrophorum as the most common cultured bacteria in PTA (31%). The striking finding of 65% of recurrent PTA had FN in their original culture.

This study adds to our clinical understanding of the devastating potential of this gram negative anaerobic bacteria. We have previously found that FN pharyngitis has the same clinical presentation as strep pharyngitis. Our microbiome study showed that patients with more severe clinical pharyngitis (defined as Centor scores of 3 or 4) differed between group A strep and Fusobacterium necrophorum. The tonsils with FN infection had a less bacterial diversity – this means that FN overwhelms the microbiome in many such patients.

These findings suggest that FN more likely causes suppurative complications. We know from the Cochrane analysis that antibiotics decrease the risk of PTA independent of group A strep testing. This article adds to our growing concern about how to diagnosis and treat Fusobacterium necrophorum pharyngitis.

This article does not address the Lemierre Syndrome. We do know that this syndrome most often follows FN pharyngitis. We cannot prove that appropriate antibiotics would prevent the syndrome, but neither can you provide any evidence that antibiotics would not decrease the syndrome.

As the article documents, the epidemiology of FN pharyngitis, PTA and Lemierre Syndrome overlap almost perfectly. These infections occur primarily in adolescents and young adults. While different articles have differing specific age ranges, one can easily generalize to around 15-30 year old patients. This large age group deserves a different approach to sore throats. One can only hope that the IDSA and CDC will study the growing evidence and develop new guidelines for diagnosing and treating sore throats in these patients.

The neverending story – unintended consequences


Category : Medical Rants

As we look back at 2018 we see a worsening atmosphere for practicing physicians. We have an epidemic of burnout. Many physicians are retiring somewhat prematurely. Some performance measures negatively impact patient care. Patients complain (appropriately) that too many physicians spend their time looking at the computer screen rather than the patient. Increasing patients commit suicide when their opiates are arbitrarily stopped.

All of these observations come about secondary to unintended consequences.

The road to hell is paved with good intentions is a proverb or aphorism. An alternative form is “Hell is full of good meanings, but heaven is full of good works


Health care is complex. Being a physician is complex. We must balance evidence and patient preferences. We treat diseases and patients and sometimes the best treatment for a disease is not the best treatment for the patient.

Everytime we see a new law or regulation, we should wonder what unintended consequences could result. Mandating electronic medical records without understanding how they would impact patient care, physician and nurse satisfaction and medical care costs is a prototypical example. EMRs seem wonderful in theory. They certainly have advantages when seeing patients whose exist in the EMR. But entering the data has become a horrible curse for many health care professionals. EMRs were not developed with enough practicing physicians. Dr. Bob Wachter wrote about this wonderfully in his book The Digital Doctor.

CMS looked at their data and decided that we have too many readmissions within 30 days. So they decided to penalize hospitals that have excessive readmission rates. This article points out the unintended consequence of this regulations – LOWER READMISSIONS LINKED TO HIGHER RISK OF DEATH

Regulators and insurance companies are impacting patient care through their well-intended rules. They do not seem think deeply about their rules. If they would use premortem examinations, perhaps we would have less of these problems. I wrote about this 2 years ago – Incentives without forethought

The continued problem of unintended consequences is one that many medical societies have addressed. I am most familiar with ACP. Their current Patients before Paperwork initiative makes many of these points consistently. Their Performance Measurement Committee (of which I am a member) consistently critiques proposed measures. Unfortunately, we continue to have difficulty influencing the decision makers.

We must all fight this battle every day. The lack of progress is frustrating, but we must not accept bad rules meekly. When the Emperor has no clothes, we should not pretend otherwise. Unfortunately, this is the overwhelming story in health care for 2018.