Confessions of a Med School App Reader

Pre-Med & Medical School Pre-Med Topics
1

Now that my and the residency programs’ rank lists are submitted, I’m essentially untouchable (although I could still be identified and I’d like to not tank my career before it starts). As a 4th year med student, I had the opportunity to work in admissions for straight MD (I had previously done interviews and sat in on ad com meetings for the MD/PhD program) and chose to be a primary reader for apps because as I mentioned above, I never saw how applicants were evaluated pre-interview. I saw a good chunk of apps and wanted to present a little insight into what I saw of the process from the other side as the primary reader a la LizzyM. Like LizzyM, I am at a USNWR top 20 medical school and I represent just one of many readers at one of many medical schools.

The Process
Apps would get sent to me in bunches via our electronic system for me to read and evaluate. I had a set of questions I needed to answer about the apps ranging from GPA/MCAT to questions about extracurriculars, personal attributes, LORs etc. Different answers had different point values assigned to them. Additionally, there was a free response section for each question/group of questions for us to elaborate our thoughts and we were encouraged to be as detailed as possible. The final question was asking us to triage the applicant into one of 4 categories (Interview Immediately, Interview 1, Interview 2, and Hold). It appears the 2nd readers had at least 6 categories (Interview Immediately, Interview 1 High Priority, Interview 1 Medium Priority, Interview 1 Low Priority, Interview 2, and Hold) and provide an overall summary of the applicant and our views of them.

We were told at the start of the process that while Interview 2 should mean they are good enough to warrant an interview, they will probably not get interviewed given the level of competition. While there is a numerical score, the final selection by the reader is all that matters, the score is more used to check that readers have some degree of rationality/consistency/objectivity in how they rate applicants. The automated screening based on GPA/MCAT is pretty liberal, but the applications are triaged such that there was a very obvious decline in the GPA/MCAT of applications that I saw over the course of the year and so a weak applicant (GPA/MCAT wise) who submitted early would be read only after the better applicants even if the 4.0/520 submitted several weeks after them. Every application that gets human eyes is read by 2 people pre-interview. 2nd readers are a smaller group of more experienced people. 1st readers are students and the faculty with less experience.

Reminder to the less informed: Unlike undergrad, no one in this process is a “full-time admissions employee.”

Results
I am able to see the final recommendation of the 2nd reader and the final decision of the admissions committee but I cannot see any of their “work.” In other words, I have no idea what the 2nd reader said in free response or how they scored post-interview.

The concurrence between the 2nd reader and me was pretty strong. With them having 6+ categories and me having only 4, I can’t do a direct apples to apples but here is what percentage of my people fell in each of the 2nd readers’ buckets.
Consensus with 2nd reader
My Holds: 89% Hold, 11% Interview 2
My Interview 2s: 13% Hold, 63% Interview 2, 25% Interview 1 Low Priority
My Interview 1s: 33% Interview 2, 50% Interview 1 Low Priority, 17% Interview 1 Medium Priority
My Interview Immediatelys: 14% Interview 1 Low Priority, 36% Interview 1 Medium Priority, 43% Interview 1 High Priority, 7% Interview Immediately

As I said above, there is a final numerical score that they said is used to gauge us as readers. I had a statistically significant correlation (according to ANOVA) between the questionnaire scores and my gestalt recommendation (p=8.7E-10).

Average cGPA (SD)
Hold: 3.37 (0.28)
Interview 2: 3.66 (0.16)
Interview 1: 3.75 (0.18)
Interview Immediately: 3.72 (0.15)
Statistically significant by ANOVA (p=3.4E-4)

Average MCAT (SD)
Hold: 508 (5)
Interview 2: 510 (6)
Interview 1: 512 (4)
Interview Immediately: 512 (4)
NOT statistically significant by ANOVA (p=0.13)

Final Admissions Outcome
Hold: 100% Rejected
Interview 2: 100% Rejected
Interview 1: 100% Rejected
Interview Immediately: 88% Rejected, 7% Waitlist, 5% Pending decision

Discussion
I was surprised that none of the people I saw got in (yet at least). There were a lot of kids that I really liked and thought had really strong apps. I have no idea how they did on interview day so I have no sense as to how much of it is a testament to the competitiveness of the process vs. people not being as impressive in person as they are on paper. I was also surprised that GPA, but not MCAT score correlated with the final recommendation. Even though there is a correlation with GPA, the strength of that correlation is dwarfed by the final score correlation (the odds that the distribution of final scores is random is a million times less than the odds the GPA distribution is random) which takes into account the entire application - that doesn’t surprise me at all. I wish I could see how these kids fared post interview in terms of their interview scores and such. Based on my experiences with MD/PhD, there are a not insignificant percentage of kids who are fantastic on paper and absolute duds on interview day. I was pleased to know that someone with 0 app reading experience was able to evaluate kids in a way that lined up with two senior admissions people - including our dean of admissions. I also assumed we were much closer to pure rolling so it was very surprising to see that apps are triaged pre-reading and that we will circle back to people who submitted early but had weak stats. There were some Interview 1s and even Interview Immediatelys in the lower stats group, but their low stats severely limited their interview slots and chance of getting in compared to how easily they would have gotten an interview earlier if they had a good GPA/MCAT. Keep those scores up!

1 Like
2

So in your experience as a reader, what GPA/MCAT would typicallly lead to an interview invite? 3.7/512? And which GPA is more important, cGPA or sGPA?

3

Based on how my school does it, I think the better question is what GPA/MCAT would prevent you from getting read as soon as you can. I only could statistically analyze the data I could extract in bulk (sorry everyone, not going to go app by app manually extracting data and I don’t have the programming skills to create some sort of program to do it for me) and so doing a correlation with submission date and the date I read the app and scores is too difficult so just kind of going off what it felt like, I would say 3.6/510 was the cutoff for getting read according to your completion date with 3.0/500 being the cutoff for being read at all. And to be clear, I mean if either your GPA or your MCAT are below those, you’re getting “put at the back of the pile” or “getting tossed in the trash immediately.”

Going off our scoring system, they are equivalent (with aoGPA not even being a factor) but I personally put a little more weight in the sGPA and I also would comment on whether a less than stellar GPA was due to a really high proportion of B range grades or if there were any C range grades or lower. I also definitely put more weight into the bio grades vs chem, physics, and math (unless of course, chem, physics, or math was your major). We received no explicit instructions on that front and are ultimately being asked to comment on their “academics” which to my surprise, did include commenting on GPA trends.

4

Final Admissions Outcome
** Hold: 100% Rejected
Interview 2: 100% Rejected
Interview 1: 100% Rejected **
Interview Immediately: ** 88% Rejected, **7% Waitlist, 5% Pending decision

This kind of blows me away. I wonder if these stats are this sobering at much less competitive Med Schools.

5

Our overall acceptance rate is ~2% and it looks like we read ~85% of the apps we receive so I would guess they would look better elsewhere.

6

What made the difference between Interview 1 and Interview Immediately, GPA*MCAT score only or did other factors come in?

7

Other factors only. Look at the GPA/MCAT averages and standard deviations of the interview 1/interview immediately groups.

8

Thanks.
And did you care what the student’s major was?
Say, two applicants have 7 advanced science classes, one is a biology major, the other is a philosophy major, did that come into play at all or not?

9

Probably a slight unconscious bump if a philosophy major is also acing their bio courses. But it’s about how it speaks to academic strength, not at all a desire to have diverse majors.

10

How about the perceived quality of the undergrad school? Say one application with similar GPA/MCAT from a higher ranked college vs lower? Is that quantified in any way for you in the application (using the point values you mentioned) or do you just consider it based on what you know about specific colleges?

There is a perception that med schools will be more forgiving of a slightly lower GPA if it’s coming from an expensive higher ranked private than from a lower ranked public or LAC. Wondering if that is more subliminal per the readers experiences or if it’s truly quantified using a scoring process.

11

Not quantified at all. It certainly didn’t seem like the pre reader triage did at all. It did seem like on the lower end of the GPA spectrum there were more kids from top schools on my list but that might be self selection.

In terms of the final reader decision obviously a reader can choose to take it into account or not. I personally did when it came to the lower GPA kids but not on a this school grade deflates or not, more on a “top tier/upper level state school or not.” The only place where I found myself putting any significant weight into school name was when people did post-Bacs. I definitely considered where the post bacc was done.

Where I probably factored school name in more was actually in evaluating how good an LOR was. Being (or not being) the top x% a Harvard etc. professor has ever seen is very different standard than at other places. I definitely both had instances where I questioned if it was “big fish in a small pond” and on the flip side tried to play up an LOR score (ie in the free response section make it clear that I thought the scoring didn’t do the person justice) because I felt a letter (or
More) was stronger than it might otherwise sound given who wrote it.

Remember, I only get 4 categories so if the academics are there, I’m not really parsing much there. I’m using the other parts of the app a lot more.

12

Looks like the second readers were less impressed with the strong applicants than you were. Mappings of your scores to the median second reader score:

H → H
I2 → I2
I1 → I1L
II → I1M/I1H borderline

13

True, but the other way to look at this is, with so much top chefs available Michelin star restaurants (MD schools), would like to use the finest sieve in the second round to get the finest products / students.

14

On the other hand, if I had the option to designate high, medium, and low priority within interview 1 the distribution of my interview immediately and interview 1s might look quite different.

15

What about paid health work v. Volunteer hours?
Some posters believe that working as an EMT or a CNA is “dinged” compared to volunteer hours. It’s always struck me as strange since it seems to be another way to “ding” working class/middle class kids who need to work in college. If that’s the case, what is the reasoning?

16

@MYOS1634 My impression has always been that wasn’t the issue of paid vs volunteer clinical experience that was the issue. It was paid clinical hours without other volunteering and community service with vulnerable populations (the elderly, rural & urban poor, homeless, mentally ill, newly immigrated/non-English speakers, etc) to demonstrate altruism and a dedication to service.

17

What WOWmom said, clinical exposure (since that’s the better term for it) and community service are two separate things that may or may not overlap.

18

How about multiple retakes of MCAT? Is it clear when the applicant has taken it more than once and does that hurt the applicant? Does the school take the highest sitting or average them?

19

When a student applies to med school, all MCAT scores are sent to the admission committee-even old, expired MCAT scores. There is no score choice or score suppression for the MCAT.

Each school has its own method for handling multiple scores, but the current AAMC recommendation (and most common method) is to average all sittings and use that as the representative score for the applicant. A handful of schools say they will only consider the most recent MCAT score, but they are in the minority. Any student with multiple MCATs should be prepared to answer questions about why s/he has multiple test scores.

Adcomms prefer to see a single strong MCAT score because a student’s MCAT score is predictive of a student’s USMLE performance. The USMLE does not allow retakes if a student passes the exam. (And failing a USMLE has very serious negative consequences for a med student. Even a subsequent great USMLE score will not overcome the stigma of the initial failure.)

20

@iwannabe_Brown & @WayOutWestMom If clinical exposure/shadowing and community service overlap, are they supposed to list in both category or only in one or need to split the hours for both. Any right way to report or any unwritten rule on how to report so that it does not backfire the student.

Context: This semester my D went 5 Sat.days to a free clinic and she loved it. She was doing pretty much every thing, starting from filling the intake questionnaire, taking vitals, sitting with Physicians (shadowing) and understanding how and what they do to diagnose, explaining to the patient the prescription details and when and what they need to take and the variety of cases came to the clinic.

(compared to her prior volunteer hours at a Hospital - where initially allowed only gift shop / front desk, birth center where only deliver blankets and new mom gift and even in emergency room all that she was asked to fold the linens and not much interaction by the clinicians.).

So how should she report her volunteer hours now? Thanks in advance.