MICLIST 3/18/20

- Ask not what your country can do for you, ask what you can do to increase ventilator availability. I typically avoid HuffPost, but this article offers excellent research into the problem.

- This one is definitely a downer - implies that young people have a significant burden of serious disease requiring hospitalization and ICU. It could be that young people are disproportionately infected, or that we're just not seeing the hospitalized older people due to delays in testing. All the same, in a period of uncertainty, I interpret any signal like this as an argument for playing it safe up front.

- A 50% solution is a lot better than nothing at all. I'm surprised we don't see more of this - employees making makeshift PPE. As supplies run out, do we think Mike Pence is going to save the day? By next week?

- The Brits switch tack. This has me bummed since I was rooting for their approach. Nonetheless, I'm still impressed - it takes guts to change approach when the best understanding changes.

- Lipsitch answers back (to Ioannidis piece here.) Seems like Ioannidis could be the defender of the Brits' approach, and Lipsitch the rest of the world. Given recent events, it seems Lipsitch's view stands taller.

MICLIST: Most Interesting COVID Links I Saw Today 3/17/20

In no particular order:

- The top, top thing on my mind is ventilators. This is maybe the most fixable near term way to save lives. Tom Inglesby at the Center for Health Security gives a great status update here. In WWII, we retrofitted car manufacturing to produce A LOT of tanks, pronto. I think we gotta do the same here AS WELL AS generate the workforce to operate these things.

- John Ioannidis, famed researcher and medical skeptic, threw down some more skepticism. This one really made me stop and think. With information so spotty, perhaps the social interventions will be needed for many months, and once we resume normal life, COVID will come rushing back?

- This preprint claims that the virus is 1,000 times less infectious after 48 hrs on its favorite surfaces, plastic and steel. Cardboard can support it for about an hour tops. When it gets aerosolized (lighter than air, so it can persists with air currents and travel all over), it still weakens, becoming ten times less infectious in 3 hours. This is good news.

- This thread from Defender of Seattle Trevor Bedford shows the simplest explanation I know of for the nationwide spread leading up to this past weekend. Short version - maybe 20,000 people COVID positive in the US. We are flying blind.

- The Brits are impressive. Gutsy. Their (ironic) plan is to wait until they see the whites of their eyes. By not closing schools (kids will spend less time with grandparents) and not shutting down the country (economy will function, which matters, particularly for people living paycheck to paycheck), they plan to absorb the shock by relying on sick people to stay home. At a time where everyone else is ramping up extreme measures, the Brits seem to be gambling. Except they are using good science. Which reminds me that we're all gambling. I hope it goes well for them. It'll be fascinating to see how it works out.

- This paper from Imperial College London seems to be the talk of Twitter today and, it's a downer. See a good breakdown here. Bottom line - the least bad options still look bad.

- Tyler Cowen posts some great COVID material, from well outside the usual sources. If you want perspective, I'd check out Marginal Revolution daily (I'd do that anyway, it's the best website.)

- Give it a rest with handwringing about ibuprofen. Until there's a reason that this largely safe anti-inflammatory is a bad idea, it's not a bad idea.

- MGH telecast their grand rounds. Dynamite. Maybe we'll learn how to be better knowledge workers in medicine when this is all over - if it's worth saying, it's worth recording.

Comparing COVID to the flu

To gauge the seriousness of COVID, the flu is a good comparison because we are familiar with it, it appears to have approximate severity and transmissibility. The case fatality rate for the seasonal flu is commonly listed at 0.1%. However, this number is likely falsely high.

On average, roughly 15% of the US population gets infected with flu virus every year, though this figure varies from 5% to 20% on any given year. However, one quarter of those infected essentially have minimal or no symptoms.

Since there are so many asymptomatic/mild cases of flu, it is difficult to determine the total burden of disease, and hence the true fatality rate. A high quality review found that on average, and across regions, the flu kills 5.9 people out of every 100,000 of the general population, whether infected or not.

We can extrapolate from this to get a rough estimate of the fatality rate of the flu. If 5.9 out of 100,000 people in a given population die of the flu, and the flu infects roughly 15% of those 100,000, this means that the rate of flu death is around 0.04%. This is less than half of the commonly reported flu fatality rate of 0.1%.

The best case we have right now for lowest fatality rate of COVID is 0.6% coming out of South Korea. From our analysis above, it seems COVID is around ten times as deadly as the seasonal flu.

Why I'm taking novel coronavirus seriously

The TL;DR can be found in this document (also pasted below). If found this document after I completed this post and it lines up with almost everything I'm saying. Also note, this post represents my best understanding as of 3/5/2020 and I hope to make addenda as better information becomes available. Also please note that most of these outcomes are in reference to the US. The situation is potentially far worse for the developing world and I don't mean to minimize those consequences.

There is a lot to talk about the novel coronavirus emerging in China, but how do we decide how and when to make significant changes in our lives as opposed to watchful waiting?

To get a bearing on this, two questions are on my mind - how dangerous is the infection, and how many people will get it. Both factors are required for good estimates of harm.

Strong answers are challenging because most of the data comes from China and other regions which are in the midst of outbreaks. The numbers aren't gathered in controlled settings, and the areas of outbreak don't necessarily compare well with the US. However, we know much more than nothing.

Question 1) How severe is it?
One way of characterizing severity is by the number of infected people who die, or the case fatality rate. One of the first studies emerging from China was published in JAMA Feb 24, and it suggested a case fatality rate of 2.3%, or 23 deaths out of every thousand infected with the virus.(1) In a follow up study in the New England Journal the death rate was 1.4%, or 14 in every thousand. Finally, at the time of this writing, South Korea has reported 6,284 cases and only 40 deaths, for a fatality rate of 0.6%, or 6 in every thousand.

To give this some quick context, the case fatality rate of the seasonal flu is commonly stated to be 0.1%, or one death per thousand, though a recent review puts it closer to 0.05%.

These numbers are sobering, as it appears the novel coronavirus is 5 to 50 times as deadly as the flu. These numbers approach the deadliness of the catastrophic 1918 flu pandemic. Fortunately, most researchers think the 2.3% number is an overestimate. That's because in most cases, coronavirus infections are expected to be asymptomatic or mild. And patients with mild symptoms rarely get tested. How often do you go to the doctor if you're only a little sick? Some researchers therefore think the fatality rate is significantly lower than 1%. This makes intuitive sense, and the South Korean data to support this.(2) However, an attempt to quantify the numbers of asymptomatic or mild infections seemed to corroborate the worst - asymptomatic or mild cases are relatively rare.

So what to make of this?

We have to wait and see as the US ramps up testing. My guess is that the numbers in the US will be substantially lower than China or the developing world, if for no other reason than that the US has a stronger social and healthcare infrastructure. It will likely be several weeks before we have a reliable estimate. Until we know better, I think it's reasonable to treat this as a fatality rate between 0.5 to 1%.

A few caveats that you've probably heard - the fatality is skewed towards those who are older or have serious health issues. Also, there seems to be very little danger for children - the JAMA study reported 416 cases among children less than 10 years old and there were no hospitalizations and no fatalities. For people under 50 with no medical issues, the fatality appears the same or even less than the seasonal flu.

This graphic summarizing these points can be found here.

Question 2) How transmissible is it?
Some infections are containable, meaning that dramatic efforts to track down and quarantine early cases can stop the spread of a disease throughout the wider population. This has been successful with SARS and Ebola. Other diseases are unable to be contained, such as the flu. Given the nature of how the flu is spread, when a new strain emerges, it is considered impossible to prevent from disseminating around the globe. At the time of this writing, it is widely believed that this virus is not containable, although the WHO continues to state that containment is a possibility.

If this virus behaves like the flu and spreads throughout the globe, how many people will be infected? At first, it seems like it will reach everyone, given enough time. However, depending on the nature of the virus itself and its specific modes of transmission, far less than 100% of people would get infected. In a typical year, the seasonal flu infects roughly 12% of the US population, though this can be as high as 20%.

Marc Lipsitch, epidemiologist at Harvard specializing in pandemics, has said in a piece in the Atlantic that he envisions 40 to 70% of the adult US population contracting the virus, based on unpublished models. Fortunately, he recently revised down this prediction to 20 - 60% based on new data.

I haven't seen anyone else modeling transmission, so this 20-60% range seems like the best guess if we assume it cannot be contained.

Conclusions from these two questions
If we use conservative figures from above, the adult US population is roughly 250,000,000. If 20% of them get infected, that's 50,000,000. A 0.5% mortality rate equals 250,000 deaths. (I leave it to you to run the numbers for a worst case scenario in the US.) For comparison, last year's flu caused 34,000 deaths and 500,000 hospitalizations.

Putting it all together
Many hospitals are currently at capacity due to this year's unusually active flu season. Speaking of the situation in Springfield, nearly every bed in the building is occupied, and there are people lined up in the halls in the emergency room. Even a modest surge of patients would be overwhelming, forcing the hospital to board patients in the lobby and convert operating rooms and procedure rooms into makeshift intensive care units. Compound this with the need to prevent spread of coronavirus to non-infected patients, and you can see the logistical nightmare this poses to the healthcare system. For numbers, the JAMA study noted hospitalization rates of 20%. If the true number is half that, we are still looking at more than doubling the current number of patients with flu.

The best way to ease the burden of a surge like this is to delay the spread of the disease, as shown in the graphic below which I found here. If everyone gets sick all at once, this will exceed capacity and therefore increase the mortality as patients go without intensive care, and those with non-infectious conditions can't access treatment. Imagine dealing with a burst appendix or a car accident or a pregnancy when the hospital is already overflowing.

I often hear the argument that "It's inevitable - we're all going to get it so what's the point?" The point is that, by flattening out the above curve and spreading out the time over which people get infected (what's called mitigation), we can potentially save lives. The benefits of this behavior always go unseen - you never know which 70-year-old did not pick up the disease because you washed your hands that one time - but these actions are matters of life and death.

Lastly, preliminary research from the Seattle outbreak indicates there has been cryptic spreading in that area for over 6 weeks. There is reason to believe the same is occurring in other cities in the US, but this is masked by the fact that the US has had delayed testing.

Therefore, until better information is available, I am behaving as if the virus is actively spreading in my area. I am doing what I can to flatten the curve.

What I'm doing:

- Called my parents, both over 65, and told them to act as if the virus is already in their area.

- Wash my hands or use alcohol gel before during and after entering a public place, before and after eating, after using a commonly touched surface (like any time I use my credit card).

- Carry hand sanitizer in my pocket.

- Regularly wipe down my work station and cell phone with alcohol or Clorox wipes.

- Don't shake hands.

- Keep 6 feet from others when I can and avoid ill-appearing people.

- Cough and sneeze into my armpit and dispose of tissues immediately.

- Try to avoid touching my face.

If I get sick: 

- Stay home.

- Eat and sleep separately from my family.

- Wear a surgical mask.

- Call ahead before seeing my doctor and ask about getting tested for coronavirus.

I have also stocked up on a 60 day supply of essential medications and non-perishable food in line with these suggestions from the CDC.

Final Thoughts:

There are reasons for optimism - the final numbers on severity and transmissibility could be much lower than what we're currently seeing. Effective mitigation strategies might emerge (such as apps that enable you to avoid proximity to infections). Warm weather could slow its spread by getting people out of doors. Treatments might emerge faster than expected. And I would be the happiest person if that occurred and all of you reading this concluded that I'm an alarmist. 

Nonetheless, the most effective time to start safe behavior is at the beginning of a crisis. The best time to avoid infection is at the earliest point of exposure. And the cheapest time to buy preparatory materials is before they are out of stock. 

“Everything we do before a pandemic will seem alarmist. Everything we do after will seem inadequate.”
Michael Leavitt, Director of the Dept of Health and Human Services 2007

................................................................................................
As mentioned at the start, the below summary is from Ethan. I have some quibbles, but mostly agree.

Updated Mar. 4, 2020

Coronavirus in brief 

Bottom Line.

• Coronavirus is significantly worse than the flu, but not the zombie apocalypse. No need to panic, but it probably makes sense to prepare.
• It is going to affect day-to-day-life in western countries, including the U.S.
• You and your family will probably face personal risk of illness by the end of the year.
• You can prepare by

• Stocking at least 1 month of nonperishable food, pet food, and other necessities, and 3 months of prescription medications.
• Relocating away from dense cities and/or shifting to working from home, if possible.
• Learning how to properly wash your hands, and practicing not touching your face.
• Avoiding travel after March of this year, and/or planning with cancellation option.
• Making plans to care for and protect the elderly from exposure to the virus.
• Carrying hand sanitizer, and using it frequently (every 30 min outside your home, before you eat or touch your face).
• Buying enough hand sanitizer and hand soap for at least 1 month.
• Wiping commonly contacted items (phone, keyboard, headphones etc) down with disinfectant regularly.
• Avoiding crowded places (e.g. concerts, subways, theatres, buses, airports etc) without protection.
• For essential travel, buying N95 respirators, if you can, and learning how to use them, including shaving facial hair. These are probably less effective than hand hygiene, so lower priority.

~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*

What does the virus do?

• The virus causes coughing, sneezing, fever, pneumonia, and in severe cases kidney failure and death.
• 80% of cases are relatively mild. The rest look like moderate to severe pneumonia.
• Approximately 1% of people who catch the virus die.
• After symptoms show, it takes 3 weeks - 1 month for severe cases to resolve.
• Risk is much higher for people over 40.
• Children appear to be relatively unaffected.
• Men may be twice as susceptible as women, although it is too early to tell with confidence.
• Immunity may not last long, and no-one has it to start with.

Where is the virus now (Mar 4)?

• 94,000+ confirmed cases worldwide, most in China. 3,200+ deaths. Likely more unreported.
• 36 countries have more than 10 confirmed cases outside of China.
• Japan, Iran, Italy, and South Korea all had an exponential growth of cases from 10s to 100s in less than a week.
• 138 cases in the U.S. 1 case, in Northern California, is likely the first spread without link to China, suggesting the virus is spreading undetected in the United States.
• New cases in washington appear to be the result of hidden transmission for weeks, suggesting 50-1500 cases may be under the radar there.

What do we know about the virus?

• It likely arose from a crossover, or “zoonosis” from animals in China, sometime in late November early december of 2019.
• It is most closely related to a virus called SARS which caused a small epidemic in 2002. It is also related to viruses that cause the common cold.

How does the virus spread?

• Probably similarly to the flu. Being within 6 feet of a cough or sneeze, touching a surface that has been coughed on and then touching your face, or eating food that has been coughed on are all ways to spread the virus.
• Relatively quickly. Approximately doubling the number of infected people every week.
• The virus can probably survive on many types of surfaces for 2-7 days.
• Some people who are infectious and can spread the virus do not show visible symptoms.
• It takes ~5 days (range of 2-14) for symptoms to develop.

Can we treat it?

• Not right now. No vaccine or approved medical countermeasure.
• Supportive care like mechanical ventilation can significantly decrease risk of death if ICU rooms are available.
• An antiviral, called remdesivir, is in clinical trials and shows some signs of efficacy.
• Historical timelines for new drug / vaccine rollout suggest mass availability in 2021.

Shouldn’t I be more worried about the flu?

• No. This is worse.
• The seasonal flu kills <0.1% of infected people. This kills ~1%. That is 10X worse.
• The coronavirus spreads a little faster than the flu.
• You have some natural immunity to flu even though each season the strain is different. You probably have no immunity against this coronavirus.
• We have a reliable vaccine against seasonal flu. We will not have a vaccine or effective treatment for coronavirus for some time.
• Seasonal flu is very well characterized and understood. This virus is still under intensive study, and all the numbers I give have uncertainty, which means that it may be worse than our best guess. Long-term effects of catching the virus are unknown.

How can I think about my and my family’s risk?

• Look at these charts for risk of death by age group.
• Consider risk factors (source) which make the disease more dangerous, such as cardiovascular disease, diabetes, lung conditions + smoking, high blood pressure, and cancer.
• Think about the population density of places you go to regularly. Ask yourself: “How many people have been here in the last week?”. Avoid places where that number is large, and, take extra precautions.
• Think about exponential spread. In the early stages, it will be doubling every week approximately. Really think about that- it means the odds of infection are about 2X higher each week this continues. So it’s twice as safe to travel April 1 than April 7. And twice as safe on April 7 as the 14th. I find that extremely counterintuitive, and chances are you will too.

This all sounds crazy.

• Yeah, it does. The info I’ve presented above makes this look like probably the worst pandemic since the 1918 Flu. 
• However, what’s presented above is an aggregation of facts and high-quality estimates from the scientific literature and expert recommendations, as best as I could find them.
• The outlook presented here is largely shared by experts at: the CDC, the World Health Organization, the Harvard School of Public Health, the Johns Hopkins Center for Health Security, MIT, a biosecurity summit I recently attended, and by most of my colleagues in the biosecurity space.
• Unfortunately, I think this is the world we live in.

Other resources

• Dashboard for tracking the spread.
• Article on prep.

Recommended:

This podcast with Marc Lipsitch. It's 27 minutes and it's pretty scary. It comes before Lipsitch's revised-down numbers.

This talk on youtube with Amesh Adalja is excellent, and a bit more optimistic than Lipsitch.

You're Likely to get the Coronavirus in the Atlantic.

STATNews has excellent coronavirus coverage, brief informative accessible articles. Their coronavirus topic section is here.

Twitter is an excellent way to stay up to date. I have a list of coronavirus epidemiologist and writers, including journalists from STATNews, that you can access here. I don't recommend it, but my Twitter handle is @astupple.

If you haven't watched it, the movie Contagion is a must see - the drama comes from its verisimilitude, not heroic characters or plot devices.

This podcast from early February interestingly walks through the thinking on the data almost as it is emerging.

For a deeper look at pandemics and human extinction, this 80,000 Hours podcast is informative, and this one is more specific to coronavirus.

......................................................................................................

1- Though it acknowledged a wide range, from 2.9% in Hubei vs. 0.4% outside Hubei.
2- Important caveat - it takes one to three weeks to get sick enough from coronavirus to die. If South Korea's massive testing campaign is catching people early in their disease course, then they may be testing people before they get sick enough to die.

 

Family Meetings: My Approach

Preparation:
- What information do we need from the patient/family? What are the key decisions we are helping them make?
- What key information do we need to give them? (Usually this info supports item 1.)
- What is the basic prognosis? Days to weeks, weeks to months, months to years, more? What information is essential to clarify this? Can we get the consultants to help clarify?
- Confirm who is the decider. Find the paperwork that confirms the healthcare proxy, which is crucial if the patient lacks capacity.
- Make a brief list of all major organ systems that are impaired.
- Schedule a time and place, try to set aside an hour of your time. Who needs to be there? I try to only include consultants who have known the patient a long time and who the patient identifies as their doctor. Otherwise, lots of providers can make for a confusing and long session.

Get the room ready:
- Get chairs for everyone.
- See if there is a separate room where everyone can meet in and try to reserve it.

Before going in:
- Decide who will be the meeting leader. There should be one point-person that the discussion goes through so that the important items get addressed. Others are there for supporting comments and details.
- Review with the care team what the main decision points are.

During the Meeting:
- Close the door and turn off the TV.
- Introduce everyone by relationship to the patient and by role in the care team, and shake their hand.
- Identify the healthcare proxy and verbalize this.
- Orient yourself initially toward the patient.
- Update the patient about the state of things. Review the major organ systems that are impaired, and the basic treatments in very general terms. "You have an infection in your blood and we are giving you antibiotics. Your heart is weak which makes it hard to breathe, and we are giving you medicine to dry out your lungs. Unfortunately, this medicine uses your kidneys, but they are weak too, and so we are having a hard time. And you are also unfortunately bleeding into your bowel so we are giving you blood transfusions."
- Ask if there are any questions. Answer them in the most general terms possible. Avoid getting too specific about the medical details.
- Get straight to the key question you have. Be straightforward in asking it. If the patient does not have capacity, direct the question at the healthcare proxy. "We need to talk about something we call code status. Do you recall anyone discussing this with you before?"
- Deliver the information that you need to deliver. "We have been giving you a blood thinner to protect from stroke, but this medicine makes you bleed. Since we haven't been able to fix the bleeding, we stopped it."

Closing the Meeting:
- Reassure the patient that we are doing as much as you can and want to make sure that everything we are doing lines up with their preferences and wishes.
- Thank everyone for coming and shake their hand.
- Write a "Goals of Care" progress note that includes the names of everyone present, a brief summary of the discussion, and any major decisions that were made.

Tips:
- Focus on the decider, either the patient if lucid or the healthcare proxy. A family member or friend may be the most engaging. Answer their questions briefly, but bring your attention back to the decider.
- Never argue. Give your impression and understanding and recommendation, but don't argue about details. Allow patients and family's to contradict you. Correct only when they are being disrespectful of the staff, and do so gently.
- Try to let the patient/family do most of the talking. Try to let them finish all sentences before jumping in. Let them set the pace of the conversation.
- Try not to be overly cheery. We instinctively greet people upbeat and happy, but it often doesn't match the context. Smile and be pleasant, but not joyful.

Advance Directive Discussion Tips:
I envision a ladder: Full code -> DNR/DNI -> Do Not Escalate Care -> CMO

I often use language like this:
"For most patients, at some point the burden of treatments reaches a point where those treatments are not worth it. This happens on a spectrum. For an otherwise healthy 24 year old, if they get sick we put them through a lot (machines, lines and tubes, surgeries, etc), because their organs are all healthy and they have so much potential life ahead of them. For an 80 year old with terminal cancer, we don't do that stuff. Most patients fall somewhere in between. Lots of times, patients say 'do what you can using hospitals and intravenous medicines and fluids, but don't hook me up to lines and tubes, don't put me in the intensive care unit.' We call this Do Not Escalate Care. If antibiotics and fluids don't seem to be working and the patient starts declining, we focus on their comfort and let them pass away in peace."

"We think Mr. Smith is at the end of his life. Some people in this situation want as many days as possible. They are willing to go through anything: 'Keep me in the hospital, poke me with needles for labs and imaging, do procedures, hook me up to lines and tubes and machines. Whatever it takes, even if it means I'm in pain and uncomfortable at the end. Other people would prefer a period of calm and some clarity at the end. They want to focus on maximizing the quality of the days remaining, rather increasing the number of days. They say 'Stop checking labs and vitals, stop with the treatments. If something happens, just let it take it's natural course and carry me off.' We call that Comfort Measures Only. The team that makes this happen is called hospice."

We often focus on "making the person DNR" or otherwise getting them to change their code status. This is the wrong approach. It's most important to give the proper information and get a clear decision that the whole family acknowledges.

Uncomplicated Gram Neg Bacteremia - I can do oral antibiotics for how long?

Someone finally studied it! There's decent evidence that you can switch to oral antibiotics with good bioavailability (in this study they primarily used Bactrim or quinolones for UTI bugs) when patients with gram negative bacteremia become clinically stable and get the same outcomes as with IV therapy. It's not an RCT, but it's almost 5,000 patients at several centers in a cohort that is closely matched with controls. And it's for hospitalized patients.

 

Association of 30 Day Mortality with Oral Step-Down vs Continued Intravenous Therapy in Patients Hospitalized with Enterobacteriaceae Bacteremia

 

Even better, an actual RCT of 600 patients with gram negative bacteremia, uncomplicated (ie- good source control, no mechanical valves or anything like that), which compared 7 to 14 days of antibiotics and showed no difference.

 

Seven versus fourteen Days of Antibiotic Therapy for uncomplicated Gram-negative Bacteremia: a Non-inferiority Randomized Controlled Trial.

 

It's a good case for doing what feels reasonable - start with IV antibiotics, once patient has been afebrile ~48 hrs and feeling better (no renal failure or stuff like that) and you have sensitivities, then switch to levofloxacin or Bactrim to complete 7 days. No PICC. BAM!

 

Ward Expectations

Disclaimer: These are my personal expectations and may not overlap with other attendings. I offer them more as a learning tool to see what an experienced clinician considers good performance.

Work hard:
- Show up on time, don't leave until your work is done or handed off.
- Do what you say you are going to do, update your team when you make changes to the plan.

Take Ownership:
- Act like you are the main doctor taking care of this patient, and that all patient problems are your problems.

Medications:
We are internal medicine, we own the med list, and we are responsible for an accurate list for any of our patients. This responsibility is easy to dodge by simply importing the existing list, and since we often care for patients for whom we did not do the initial med rec, that means assuming the list is wrong until we have checked it ourselves.
- Double check the following home meds against pharmacy records or patient/family report.
     - Anticoagulation
     - Antiplatelet (clopidogrel)
     - Diuretic
     - Immunosuppressives
     - Any other drug that is critical

Interacting with patients:
- Introduce yourself in terms of your role in the hospital in terms the patient can understand.
"Hello, I'm Dr. Stupple. I know you have seen a lot of doctors. I'm taking over from Dr. _____. I'm your main doctor right now. I'm on a team that includes Dr. _____. You have gastroenterologists and surgeons who are seeing you as well, so there's three teams. My job is to coordinate everyone."
- Be respectful of everyone. Patients that trigger anger and resentment are often the most vulnerable and suffering the most.
- Give them time to feel heard. Letting someone speak uninterrupted until they get out everything they want to say is maybe the most powerful way to communicate compassion.
- Explain to them what is happening and update them when things change. Ex- if you order a CT scan, tell them why it's happening rather than letting the nurse or transportation tech do it.
- Discuss medical issues at a 6th grade level of language and understanding.
- Speak with the healthcare proxy or close family once every day or two.

Nursing and other staff:
- Be respectful. Among other things, control the urge to be dismissive- when nurses are concerned, patient mortality reliably increases.
- Be approachable.
- Respond to pages quickly, regardless who they are from. It's easy to say you respect everyone, but actions speak louder. Responding quickly to pages tells the recipient that you value their concerns. Think how much you appreciate it when someone "important" gets back to you immediately. It's gold.

Workflow:
- Top priority is generally to call consults, then submit orders, talk to patients and families, and finally get notes done.
- Be skeptical of all incoming information. Double check important vitals, confirm history with the patient or close family.
- Don't dismiss incoming information. Treat emergencies and urgencies as such until you know they are resolved.
- Don't delay placing diet orders.

Presenting:
- Start with chief complaint or one-liner to orient the team. (This is an excellent habit because it teaches you to prioritize and summarize.)
- Save the discussion for the assessment/plan, don't editorialize the data.
- On rounds, take a guess at a diagnosis, list alternatives, and look for tests you could do to make things clearer. Don't obsess about trying to get the diagnosis right, just tell us what you're thinking, including "I don't know."
- Master the problem list. Have a problem for each medication the patient is on at home, including meds that are on hold, and each med in the hospital.


Master Level:
- Focus on solving problems for patients, rather than simply getting your job done. Ex- A patient admitted with CHF can be rapidly improved with IV lasix and tempting to discharge. But focusing on their problem of volume overload and frequent admissions, we can diurese them to their dry weight.
- Prioritize items that will impact care, such as delaying diagnosis and treatment or prolonging hospital stay. This requires good medical knowledge as well as....
- Maintain situational Awareness: Outmaneuver the bottlenecks, such as PICC lines and IV access, procedures, consents.
- Extend your plan all the way until discharge. Clarify what conditions need to be met before you will discharge.
- Speak definitively when you are definite about what you are telling them. Avoid using qualifiers like "we might have to do an LP" or "it looks like you might have bacteria in your blood stream."

Personal Extras:
- I avoid maintenance fluids. If a patient needs fluid, I make an estimate of how much and write to give that amount over a period of a few hours. Then I reassess and consider more fluids. Maintenance fluids are hard to know how much they received and can easily be forgotten about over a period of days.
- I never order colace since several RCT's show it doesn't work.
- With exceptions, I rarely alter patient's diet from home because their treatment should be tailored to what they usually eat, not a special in hospital diet.