This post is intended for students in a dual degree program, such as an MD/DO/PhD. I give some advice, but take it with a grain of salt — everyone takes their own path. 

Changing views 

When I was finishing graduate school, I wanted to apply for a post-doc rather than a residency.  During clinical rotations, my viewpoint shifted.  I wanted to continue clinical training.  Be prepared for changes.

Versatile skills

Because things can change during training, I would recommend trying to spend some time during graduate training developing versatile skills: skills that apply to many different areas.  I took many courses, but the ones that were most versatile were the most foundational, some of which I audited rather than took, such as undergraduate probability and mathematical statistics.  Following up on this, I benefited from widely applicable courses in statistical inference, linear models, Bayesian statistics, graphical models, and causal inference.  Survival is also essential to clinical trials.  Some of the more advanced courses provide more mathematical maturity, but one can start research without them.

Another fundamental skill is coding (and expressing things using matrix operations). I recommend learning a scripting language (R, Python, etc).  You can sometimes study more complex objects using simulations than you can with closed-form equations (although often you can study a phenomenon by using a simple example).  A final highly versatile skill is optimization. Beneath everything is an optimization problem (it is the essence of statistical estimation, or learning).

Also, going from clinic to research and back and forth, where different languages are used, along with great coaching, strengthened my ability to communicate. Of all the skills, communication is perhaps the most versatile.

Statistics and clinic

In general, I saw my research everywhere in clinical medicine, not just in the literature, but also on the floors. I saw the equations behind most decisions.  It allowed me to think about how to better incorporate evidence into healthcare.  To me, being able to do this was the major deliverable of my training.

When we have a strong understanding of core topics in statistics, it better directs us toward what we should be doing in medical research, such as adjusting lab reference ranges, treatment effects, and test characteristics for covariates, or predicting side effects in addition to primary outcomes like mortality. 

Day-to-day tips

After I finished my PhD, a few days before starting clinical rotations, I went to the hospital, and I tried to form some positive memories.  For example, I went to the cafeteria, and found some food I liked, or I chatted with a friend.  Similarly, a program alum (Benjamin Plog) gave the following advice during a seminar: structure the PhD like a 9-5 job, it will make it easier to transition back to clinical rotations. Even if it is not possible to do this in the early years, try to do it in the later years.  If your sleep schedule gets off track, which can happen during graduate school, I recommend looking into mindfulness/meditation, because there will be times during third year when you must be rested, even if you are having trouble sleeping.

Also, at least for me, treating research like a job, where I showed up every day and recorded my work, led to more steady progress than waiting around for inspiration to strike. Also be sure to document everything, including your partial progress (especially your partial progress, as ultimately everything is partial progress). I had a document for my advisors at each meeting, and I sent a follow up each Friday.  This was sometimes tedious, but it helped us stay on the same page, it helped me show progress, and it helped in terms of getting input (and finally writing manuscripts).  Especially if one is on a tight schedule, like I was, having these documents and the strong communication that resulted between student and mentor helps.

Finally, if you are in stats research (or maybe any research), I would check out J. Michael Steele’s advice on grad school.

Your own path

We live in a competitive society. Note that competition can be fun and useful.  However, it does not need to influence you too much. Even if you are not the best in either medicine or research, know that you are in the intersection, you will be good at that, and that is valuable.  

It helps to cultivate an inner sense of achievement. It might consist of celebrating a problem you are making progress toward solving, or identifying a personal milestone in your work, etc. For me, one of these milestones was a 2 minute presentation I gave during a class at the very end of medical school, where I advocated for sometimes-neglected treatment effect heterogeneity analyses. I knew what I was talking about, and I could tell that the message resonated. Being able to do this — to travel far to a distant academic land, find something there, and then bring it back to others in a way that was digestible in 2 minutes, felt meaningful.

I will say, although you should focus most on your internal progress, take care to still meet external expectations.  If you decide to work too much according to your own goals, it can impact your grades, test scores, etc. These things matter for what comes next.  In particular, your residency application will depend most on your third year of medical school, even though third year is just one year of 8+. Your residency application can then impact the next 3-7 years of your life.

The intersection

You will be at the intersection of your research field and medicine. Being good in an intersection takes discipline, because it’s sometimes tempting during training to go too far toward research or medicine.  I did the former by taking many years in the graduate phase (although my own story was complex, because I transferred and switched departments). 

Remember, you always make certain choices that can help you stay centered. One of these is choosing an advising team. In general, they say choose a mentor, not a topic, which I think is good advice. When choosing mentors and advisors, think, what kind of person do you want to be like? What is their track record with students? What is the average time to degree? One could also say choose a department, not a topic, because it takes a village to train a graduate student. Does the department have a track record of working with dual degree students? Etc. 

Choosing a topic is also important.  Sometimes hot topics can lead to longer duration of degree, so beware. Careful of choosing a research direction that you think will extend your graduate training beyond 4 or 5 years. You can look at other students working in the field, or with the labs, to get a sense of whether this will happen; however, even if you make a good decision, your PhD might take longer for reasons outside of your control (in general, I wish PhDs were fixed length).  I think, getting a 4 year PhD and then doing 2 years of research during residency or after is better on average in terms of life and physician-scientist career than getting a 6 year PhD and then doing no research in residency. A 4 year PhD may feel like you are moving fast, but remember who you are, and your place in the intersection.

Sometimes, I found I wanted to make choices, like spending extra time, taking a new project, or submitting to the best possible journal.  Sometimes this was out of excitement, sometimes it was to fill a hole.  If this happens to you, and it is the former, it may be possible to find excitement in your current project (I love Steele’s advice in this respect or Eliot’s poem on seeing the same thing with new eyes). If you are instead doing these things to fill a hole, try to understand why this might be the case.  Was there trauma? Is it competition? Insecurity?  Try not to let things like this make major choices for you. Mindfulness helped me here.

I do not know a lot about PSTPs, but it seems the PSTP can be nice, because your research then strengthens your residency application (the residencies benefit directly from your skill in research), and the PSTP structure facilitates a physician-scientist career.  As far as I know, most internal medicine PSTPs include a fellowship, so they are 6+ years with 3 of these years being research (so you can re-start topics you might have found interesting during the PhD).  This is ideal if one wants to do a fellowship, but difficult if not. Psychiatry has PSTPs that do not require a fellowship.  

Ultimately, there is a tension between the clinical-research balance I recommend and the 7 years I took to study statistics, and the fact that the extra time I spent studying statistics, and in particular the time I spent studying statistical models for medical decision making, also gave me a new, different perspective on clinical medicine, which has been engaging for me, and I hope will be valuable to society. You might choose otherwise; listen to your gut, and I hope my story helps build your intuition.

Life

Seek balance in terms of work and personal life. Common advice is to start your life during the MD/PhD degree, which can be 8+ years. To do this, I would recommend that you try to seek out predictable environments as much as possible, both in research and clinical specialty. To me, predictability is often as important as hours in terms of being able to also have a life outside of work. Things might sometimes feel too predictable. However, for me, life brought enough excitement in its own right.  In addition, all of the transitions between worlds in a dual degree will also bring excitement/stress.

Balance may be productive in its own right; in the process of taking time to focus on things outside of work, you might discover things about yourself that not only improve life, but also make you better as a healthcare provider and a scientist. 

In general, community is important. As I went through training, I began to prioritize community more, partly due to Covid, which occurred during my graduate studies. I remember loving statistics, and then during the pandemic one day being in my apartment alone, having just done 5 hours of mathematical derivations, and realizing: I like statistics, but I also really liked talking to people about statistics.

Try to hold on to the community, somehow. I still don’t really know how to do this well. Many of your PhD friends will leave when they find jobs. Many of your medical school friends will leave in the match. I wish there were more academic jobs and more residency spots, and therefore it were easier to stay in one place geographically, and to maintain community. 

Community also leads to organization, and working together for a better future. Friends outside of research and medicine can also help in terms of community that will not leave with the job market or the match.

Specialty

Another major decision is specialty. If you don’t know what specialty you would like to apply into, clinical experiences during your PhD years, which were called Longitudinal Clerkship Experiences in my program, can really help.  Try to find a preceptor who might then write a letter of recommendation for your residency application, since it is hard to scramble for letters during third year, especially around application time. Just ask if they would be willing to do it, no need to actually then ask them to write it, if you change your mind in terms of specialty, etc. Also, use the experience to try to find a specialty that you would feel alright with, and one that does not depend too much on future performance in third year, since this is difficult to predict. You can always change your mind later.

In general, the longer you are in research (or the longer it feels), the more LCEs you should do.  This may be tough. The longer you are in the PhD, the more you will feel pressure to finish. However, specialty choice is a major decision. The harder third year is, the more difficult it is to make a major decision, and the longer you are out (or the longer it feels), the harder third year will be.

No matter what, returning to clinical training can be challenging.  In my opinion, this is a major reason people have trouble choosing a specialty: feeling afraid of choosing any specialty. If you notice this, it may be worth stepping back and trying to identify the stressors rather than trying to force a decision.

Like in grad school with choosing a topic, my sense is that it is sometimes worth choosing a mentor (or an environment) instead of a specialty. If you are still having difficulty choosing a specialty, which can be stressful, it might be helpful to note that you are, in the process, learning about yourself, and you can be yourself in multiple different specialties.

As a dual degree student, also, your ability to find a career that suits you is further facilitated by your role as a researcher. For example, you might find that you want to go into radiology, but you like talking to patients. You might realize that there are two parts to this: you like to help people, and you like to talk to people. As a radiologist, you help people (on a large scale), so that requirement is fulfilled. To talk to people—besides making sure to talk with colleagues—as a researcher, you can collaborate or even have a lab, which can lead to many meaningful interactions. 

Teaching hospital

After being in research, it can be difficult to return to clinicals, because it feels like mistakes can cause harm. Nowadays, post-Covid, everyone wears jackets/fleeces. The white coats, and the medical student’s short white coats, are all but gone. I however always used to wear my short white coat, as a symbol that I needed someone to double-check my work.  The feeling that one needs to be double-checked may be more pronounced the longer one is away from clinic (or the longer it feels one is away), and it is a conscientious and caring instinct.  I though encourage you to feel more free to make mistakes than I felt.  To learn, you must feel free to make mistakes. There will be many people above you to prevent mistakes from causing harm: they are trained exactly for this, and they want to do this for you and for your future patients. 

Also, often, it is not mistakes that are bad, but, instead, missing the big picture. To see the big picture, one needs to take risks, and make mistakes; sometimes it helps to present without notes, etc, or take the extra time for yourself, even a minute before a presentation, and summarize the case mentally without looking at notes.  It’s difficult, because seeing the big picture is not necessarily about ignoring details, it is about synthesizing them, and this is a difficult thing to do, at least for my brain, which is a more bottom-up processor, I think.

If given feedback, even if you feel singled out, try to think of it as a gift. A Dean of Research (Dr. Dewhurst) once said something during a research seminar like: although criticism can hurt, there is also a pearl. Practice finding the pearl.

A scaffold

Ultimately, medicine is information overload. It helps to have a scaffold when presenting a case. As a professor at the school, Dr. Fong, told us, a great scaffold is the differential. So, when thinking about a presentation, the first thing to do is develop a differential, and then structure the presentation around it.  Mention symptoms that help you narrow the differential (and order tests and labs to do this too). That is how healthcare providers think, and it is what they expect to hear about.

For dual degree students re-entering clinical training, who might have lost vocabulary and knowledge during the PhD, shelf and step exams can be a major stressor.  I think it is helpful to learn to think about a differential during exam questions as well, as I describe here. Full disclosure, this didn’t help me do too great on exams, but it made studying much more engaging. Also, I almost always studied on a treadmill, so I did not have highlighting or even the physical act of writing to help me stay engaged, and so I needed a new technique.  

Community in school

Connect with others in the dual training program —they will understand you.  They might be in different departments, but they will be talking about the same things: the day-to-day grind, thesis proposals, thesis defenses, transitions, clinical work, etc.  They have so much to teach you. Take advantage of structured, consistent opportunities to do this, such as dinner seminars and social events. It is nice to be on your own path, as any dual degree student will be, but it can be very difficult to be isolated, especially in high-pressure academic environments. Also, although you can obtain similar community in either the clinical world or the research world, and this is important, avoiding dual degree events in favor of departmental or medical school events may also tempt you to go too far in one direction or another (e.g., to take extra time in graduate school or leave graduate school early).

Also, keep in contact with administrators, mentors, and other role models — listen to them. They will help you through graduate training in a way that works for you. When returning to third year, listen to medical school advisors as well. They can help you ease into clinical work and think about specialties.

Improving patient care as motivation

The goal of improving patient care is a guiding light. It is a gift to relieve suffering through one’s work (an option in any career, not just medicine).  In your career, you might do varying degrees of patient care, but you still will have had the training, and the direction. It can feel like a lot, since there is always work to do, but even making a small dent in it can help, sometimes in big ways.  

Being at the intersection of the clinical and research worlds can be a tough road. You are a link between research and clinical medicine. Prepare to sometimes feel misunderstood. If you remember the patient, it will give you strength, and you will understand, and so will I.