2008-9 Residency Data 279 fired 28 deaths

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

exPCM

Membership Revoked
Removed
10+ Year Member
15+ Year Member
Joined
Apr 12, 2006
Messages
919
Reaction score
8
2008-9 Residency Data
279 fired
28 deaths
1065 withdrawn (I would bet many of these were pressured into withdrawing)
My comment: I can't help but think that there is a lot of pain behind these numbers
ScreenHunter_01May092341.gif

http://www.acgme.org/acWebsite/annRep/an_2008-09AnnRep.pdf

Members don't see this ad.
 
Last edited:
How many residents were there total?
 
how many residents were there total?

109,482 (over all years).
24071 PGY1.
So about 24K new residents per year and about 3K residents per year leaving/resigning/transferring programs.
So overall this would seem to yield a roughly 12-13% attrition rate.
Any chance that the ACGME would ever publish attrition data for each individual program? I would not hold my breath waiting for that publication.
 
Last edited:
These numbers are truly appalling. Hopefully, we all realize by now that M.D without a residency essentially stands for "MASSIVE DEBT". Something is seriously wrong with the system when you have 3000 newly minted doctors being driven out of hellish residency programs each and every year. That is the equivalent of 20 medical school classes being cast off! And 28 deaths a year? I wonder how many of these are from suicides and residents getting killed in car wrecks driving home post call?

Just for comparison's sake, I recently read that the NYC school system lost a mere 3 out of 55,000 teachers due to incompetence over a recent 2 year period. That is a firing rate of 1.5/55000 per year. In other words, your odds of leaving a residency program are about 1000 times greater than getting booted from the NYC School System. This is the difference between having no power whatsoever at the workplace versus belonging to an iron clad union. Must be a nice feeling knowing that you don't have to worry about getting axed from your job each and everyday.

http://www.nytimes.com/2010/02/24/education/24teachers.html


"The Bloomberg administration has made getting rid of inadequate teachers a linchpin of its efforts to improve city schools. But in the two years since the Education Department began an intensive effort to root out such teachers from the more than 55,000 who have tenure, officials have managed to fire only three for incompetence."
 
109,482 (over all years).
24071 PGY1.
So about 24K new residents per year and about 3K residents per year leaving/resigning/transferring programs.
So overall this would seem to yield a roughly 12-13% attrition rate.
Any chance that the ACGME would ever publish attrition data for each individual program? I would not hold my breath waiting for that publication.

I'm not sure transferred should be counted. I mean if you decided early on that you didn't like urology so you transferred seamlessly to a local radiology program, or if you relocated to a California program because you got engaged to someone who lives there, should we really be calling that a negative? Even "withdrew" is suspect because there are lots of positive reasons one might quit residency for greener pastures. We really should only be looking at the number dismissed (or contracts not renewed) to get useful data as to "attrition".
 
I'm not sure transferred should be counted. I mean if you decided early on that you didn't like urology so you transferred seamlessly to a local radiology program, or if you relocated to a California program because you got engaged to someone who lives there, should we really be calling that a negative? Even "withdrew" is suspect because there are lots of positive reasons one might quit residency for greener pastures. We really should only be looking at the number dismissed (or contracts not renewed) to get useful data as to "attrition".

Your attempt to redefine attrition is not in keeping with the medical literature on this subject.

Example:
Has the 80-hour work week had an impact on voluntary attrition in general surgery residency programs?
Leibrandt TJ, Pezzi CM, Fassler SA, Reilly EF, Morris JB.

Department of Surgery, Abington Memorial Hospital, Abington, PA 19001, USA. [email protected]

Abstract
BACKGROUND: This article attempts to assess the effect of the duty-hour limitations implemented in 2003 on voluntary withdrawal of general surgery residents. STUDY DESIGN: A questionnaire asked the program directors how many categorical general surgery residents left voluntarily in 2003 to 2004, their training levels, why they left, and where they went. Results were compared with an identical study of 2000 to 2001 and analyzed statistically using chi-square analysis. RESULTS: A total of 215 programs (85%) responded, compared with 206 programs (81%) in the previous study. One hundred two programs (48%) reported voluntary attrition of 148 residents, compared with 110 programs (53%) and 167 residents previously. An average of 1.5 residents per program left in programs that reported attrition and 0.7 residents per program in all responders, compared with 1.5 and 0.8 residents in the previous study. In both studies, most programs with attrition lost one (66% [2000 to 2001] and 65% [2003 to 2004]) or two residents (21% [2000 to 2001] and 27% [2003 to 2004]). Most attrition occurred at PGY1 (47%) and PGY2 (28%) levels; a total of 75% of all attrition occurred at these levels, compared with a total of 76% in the previous study. One hundred eleven residents (75%) entered other medical specialties, and 23 (16%) transferred to other general surgery programs, compared with 105 residents (63%) and 40 residents (24%) in the previous study. In both studies, personal issues and work hours/lifestyle were cited as the most common reasons for leaving. In each study, the net loss to general surgery (the number of residents who left voluntarily divided by the total resident population at risk) was 3% for that academic year. Analysis showed no statistically significant difference. CONCLUSIONS: Rates and patterns of attrition seem to have been unaffected by Accreditation Council for Graduate Medical Education work-hours limitations.
 
Horrible.

When I was an intern, I was so unhappy. The best part of my day was the moment I got home and over the evening would feel progressively more miserable as I started to dread the next day. As I walked to work, I sometimes wished a car would hit me so I wouldn't have to be an intern anymore. But more importantly, at the time, I truly believed that nobody was as unhappy as I was and I felt very alone.
 
109,482 (over all years).
24071 PGY1.
So about 24K new residents per year and about 3K residents per year leaving/resigning/transferring programs.
So overall this would seem to yield a roughly 12-13% attrition rate.
Any chance that the ACGME would ever publish attrition data for each individual program? I would not hold my breath waiting for that publication.

.
 
109,482 (over all years).
24071 PGY1.
So about 24K new residents per year and about 3K residents per year leaving/resigning/transferring programs.
So overall this would seem to yield a roughly 12-13% attrition rate.

Your point may be valid, but the numbers don't work that way. You have to take the attrition out of the whole working pool, not just the first years.

Plus, it's not clear at all if the 3000 residents leaving is a 2009 number or a 2003-2009 number. Roughly 100,000 residents active during any given academic year, 3,000 leave. That's an attrition rate of either 500 per year (if it's a 6-year figure) or 3000 per year (if it's a 1-year figure) so .5% (in the first instance) and 3% (in the second).
 
Your point may be valid, but the numbers don't work that way. You have to take the attrition out of the whole working pool, not just the first years.

Plus, it's not clear at all if the 3000 residents leaving is a 2009 number or a 2003-2009 number.
It is perfectly clear to anyone who reads the ACGME annnual report I linked to and compares it to previous annual reports. These are one year reports not some running tally.
Roughly 100,000 residents active during any given academic year, 3,000 leave. That's an attrition rate of either 500 per year (if it's a 6-year figure) or 3000 per year (if it's a 1-year figure) so .5% (in the first instance) and 3% (in the second).

I am sorry to say it seems you don't have a clue about basic statistics. Are you really a PhD or are you just messing with me?
If I have a navy with 100 ships and I build 20 ships per year then it takes 5 years to build those 100 ships. If 2 ships are sunk per year then my attrition rate is 10% (2 ships/20 per year or 10/100 ships per 5 years). Is this a simple enough example for you?

When I posted this I had a feeling I would start seeing posts like this from the statistically challenged.
 
Last edited:
Your point may be valid, but the numbers don't work that way. You have to take the attrition out of the whole working pool, not just the first years.
Then you must add attrition for all the resident years as well. If you want to calculate this accurately without estimates, you take the pgy-1s of a year, and count the dropouts from this group over all their years working as residents. That is your attrition rate for their resident period.

The annual snapshot would be how exPCM did it, he just says that the dropout rates of an average resident year is the same as the dropout rate of the other residents, when they were in this resident year.
 
I am sorry to say it seems you don't have a clue about basic statistics. Are you really a PhD or are you just messing with me?
If I have a navy with 100 ships and I build 20 ships per year then it takes 5 years to build those 100 ships. If 2 ships are sunk per year then my attrition rate is 10% (2 ships/20 per year or 10/100 ships per 5 years). Is this a simple enough example for you?

When I posted this I had a feeling I would start seeing posts like this from the statistically challenged.

I'm not sure I follow this analogy.

So, let's make these two scenarios entirely parallel. And just to clarify a few things.

1. 3k is the annual dropout rate over ALL years, not just PGY1. (unless I'm wrong about this, then disregard anything after this)

2. Attrition rate = members exiting / total members

So, in the ship scenario, you took a longitudinal approach, where you acted like you started with 0 then after 5 years, ended up with 100. And influx is 20 ships a year, while you lose 2.

Now, if we take this logical construct and use it to modify the residency report, we'll have a scenario like this. You start with 0 residents and add 25k annually. However, if my #1 assumption is correct, then we don't really have an annual efflux rate at this point. 3k drop out is over 4 years, and if you assume equal dropout over a 4 year span, then you should end up with 750 for the first year.

I have to conclude that given my two assumptions are correct, I have to agree with dotclash. You have to take the 3k/100k as the attrition rate.
 
Last edited:
Then you must add attrition for all the resident years as well. If you want to calculate this accurately without estimates, you take the pgy-1s of a year, and count the dropouts from this group over all their years working as residents. That is your attrition rate for their resident period.

The annual snapshot would be how exPCM did it, he just says that the dropout rates of an average resident year is the same as the dropout rate of the other residents, when they were in this resident year.

We have someone here who is on the ball and who is likely an excellent physician.

I'm not sure I follow this analogy.

So, let's make these two scenarios entirely parallel. And just to clarify a few things.

1. 3k is the annual dropout rate over ALL years, not just PGY1. (unless I'm wrong about this, then disregard anything after this)

2. Attrition rate = members exiting / total members

So, in the ship scenario, you took a longitudinal approach, where you acted like you started with 0 then after 5 years, ended up with 100. And influx is 20 ships a year, while you lose 2.

Now, if we take this logical construct and use it to modify the residency report, we'll have a scenario like this. You start with 0 residents and add 25k annually. However, if my #1 assumption is correct, then we don't really have an annual efflux rate at this point. 3k drop out is over 4 years, and if you assume equal dropout over a 4 year span, then you should end up with 750 for the first year.

I have to conclude that given my two assumptions are correct, I have to agree with dotclash. You have to take the 3k/100k as the attrition rate.

Another failed statistician here. 3k is for a single year not over 4 years. 12K would be the number over 4 years, with the rate of 3K/year being relatively constant (4 x 3K).

I am getting scared to see the lack of comprehension.
 
Last edited:
I am sorry to say it seems you don't have a clue about basic statistics. Are you really a PhD or are you just messing with me?
If I have a navy with 100 ships and I build 20 ships per year then it takes 5 years to build those 100 ships. If 2 ships are sunk per year then my attrition rate is 10% (2 ships/20 per year or 10/100 ships per 5 years). Is this a simple enough example for you?

When I posted this I had a feeling I would start seeing posts like this from the statistically challenged.

Awfully condescending post. Must be great to be a big bad attending.

At the end of the day, attrition, like survival (for they're the same thing) needs to take into account time until death/attrition. The correct unit is probability/person/unit time.

In the example you give above with the ships, there is not enough detail to calculate an attrition rate. Consider the following possibilities:

1) 10 ships are built this year. 2 of them get sunk this year. Attrition is therefore 2/10 "ship-years"= 20% per ship per year.

2) 10 ships are built this year. 2 of them get sunk *5 years later*. Attrition is now 2/50 ship years=4%/ship/year.

3) 10 ships are built this year. 2 of them get sunk *1000 years later*. Clearly, attrition is basically 0.

So the fact that "2 ships per year out of 10 sink" gives us no indication of attrition unless we know how long those ships took to sink. It's just like survival--trivially, 100 out of 100 people will die, just like 10 out of those 10 ships will either sink or be decomissioned. But we measure survival/attrition by noting how long it takes to reach the endpoint.

Back to the residency question, 24k enter each year. We know 3k leave, but we don't know the length of time--some of those who left were pgy-1 (so took one year), so were pgy-X (where X>1). So we can't calculate an attrition rate.

Bottom line: attrition is just like survival analysis--you need to take into account the time to failure in determining the attrition rate.
 
Awfully condescending post. Must be great to be a big bad attending.

At the end of the day, attrition, like survival (for they're the same thing) needs to take into account time until death/attrition. The correct unit is probability/person/unit time.

In the example you give above with the ships, there is not enough detail to calculate an attrition rate. Consider the following possibilities:

1) 10 ships are built this year. 2 of them get sunk this year. Attrition is therefore 2/10 "ship-years"= 20% per ship per year.

2) 10 ships are built this year. 2 of them get sunk *5 years later*. Attrition is now 2/50 ship years=4%/ship/year.

3) 10 ships are built this year. 2 of them get sunk *1000 years later*. Clearly, attrition is basically 0.

So the fact that "2 ships per year out of 10 sink" gives us no indication of attrition unless we know how long those ships took to sink.
Repeat They sank in one year! -
It's just like survival--trivially, 100 out of 100 people will die, just like 10 out of those 10 ships will either sink or be decomissioned. But we measure survival/attrition by noting how long it takes to reach the endpoint.

Back to the residency question, 24k enter each year. We know 3k leave, but we don't know the length of time--some of those who left were pgy-1 (so took one year), so were pgy-X (where X>1). So we can't calculate an attrition rate.
Yes, we do know that 3k left in one year.
Bottom line: attrition is just like survival analysis--you need to take into account the time to failure in determining the attrition rate.
Must be great to be a smart guy medical student.
Bottom line - this appears to be too hard for you to understand.
I hate to say this but in my medical school we were required to take a biostatistics course and the top half of the class generally had no problems understanding these sorts of statistical scenarios while the bottom half of the class struggled
Maybe the posts are condescending but I am responding to critical and incorrect posts by people who cannot understand a relatively simple analysis.
Reread the post by integrity for comprehension.
Again 24k come in each year and 3k wash out each year.
Ballpark Attrition = 3k/24k = 12.5%. This is a valid estimate.
 
Last edited:
Another failed statistician here. 3k is for a single year not over 4 years. 12K would be the number over 4 years, with the rate of 3K/year being relatively constant (4 x 3K).

I am getting scared to see the lack of comprehension.

Jesus Christ, you're a tool. 3k over residents of PGY 1- 4 - and NOT 3k over just PGY1.



Either way, the point of contention here seems to be solving for annual attrition rate of total residents vs 4 year attrition rate for 25k residents.

The two statements that can be made are:
"3% of total residents wash out every year."
OR

"12.5% of residents that start a 4 year residency wash out over 4 years."
 
Last edited:
Must be great to be a smart guy medical student.
Bottom line - this appears to be too hard for you to understand.
I hate to say this but in my medical school we were required to take a biostatistics course and the top half of the class generally had no problems understanding these sorts of statistical scenarios while the bottom half of the class struggled
Maybe the posts are condescending but I am responding to critical and incorrect posts by people who cannot understand a relatively simple analysis.
Reread the post by integrity for comprehension.
Again 24k come in each year and 3k wash out each year.
Ballpark Attrition = 3k/24k = 12.5%. This is a valid estimate.

This seems to be far too hard for *you* to understand. 3k people left last year--but we don't know how long they were around. Unless they were all PGY-1's, some took longer than a year to leave.

The attrition rate based on this is 12% over the life of the residency. This does not necessarily imply 12% of residents per year are leaving--the latter statement is an upper bound (of sorts). If 12% of residents were leaving per year, we should see 12k leave this year. In fact, we see 3k leaving, implying an average residency length of four years--probably about right. So the attrition rate is 3% per year=12% over the life of the residency. The fact that you can't parse out the difference in these two statements makes me wonder about *your* biostatistics training

I also find it funny you use the term "wash out." As others have pointed out, much of your attrition rate consists of transfers--we don't know the degree to which that's "washing out"/being fired versus people making legitimate decisions to switch fields.
 
This seems to be far too hard for *you* to understand. 3k people left last year--but we don't know how long they were around. Unless they were all PGY-1's, some took longer than a year to leave.

The attrition rate based on this is 12% over the life of the residency. This does not necessarily imply 12% of residents per year are leaving--the latter statement is an upper bound (of sorts). If 12% of residents were leaving per year, we should see 12k leave this year. In fact, we see 3k leaving, implying an average residency length of four years--probably about right. So the attrition rate is 3% per year=12% over the life of the residency. The fact that you can't parse out the difference in these two statements makes me wonder about *your* biostatistics training

I also find it funny you use the term "wash out." As others have pointed out, much of your attrition rate consists of transfers--we don't know the degree to which that's "washing out"/being fired versus people making legitimate decisions to switch fields.
To determine the attrition rate for the Northwestern Medical School Class of 2010 how would you do it?
How about taking the number who entered in 2006 (as the class of 2010) and then seeing how many graduated with their class in 2010. This would help you determine the attrition rate.
The same principles apply to residency although this is beyond your comprehension apparently.
However, this is also a problem area for many others:
http://www.minnesotamedicine.com/PastIssues/November2007/PulseBamboozled/tabid/2339/Default.aspx
 
To determine the attrition rate for the Northwestern Medical School Class of 2010 how would you do it?
How about taking the number who enter in a given year 2006 and then seeing how many graduate with their class in 2010. This would help you determine the attrition rate.
The same principle applies here although it is beyond your comprehension apparently.
However, this is a problem area for many see:
http://www.minnesotamedicine.com/PastIssues/November2007/PulseBamboozled/tabid/2339/Default.aspx

Yeah, I got what you're saying. Your attrition makes more sense in a practical way, though it's not incorrect to say annual attrition is 3%.
 
Yeah, I got what you're saying. Your ship analogy was just unclear, since you said "every year 2 ships sink," which can be interpreted as 2 ships sink total independent of total ship count.


Every year 2 ships sink. Seems crystal clear to me.
 
To determine the attrition rate for the Northwestern Medical School Class of 2010 how would you do it?
How about taking the number who entered in 2006 (as the class of 2010) and then seeing how many graduated with their class in 2010. This would help you determine the attrition rate.
The same principles apply to residency although this is beyond your comprehension apparently.
However, this is also a problem area for many others:
http://www.minnesotamedicine.com/PastIssues/November2007/PulseBamboozled/tabid/2339/Default.aspx



As a general rule then, attrition per year is a much more relevant figure. This is why we call survival analysis "time-to-failure." This is also why survival statistics (attrition from life), are given per year, not over the lifetime. As long as a person gets use out of each year they do not attrition out, attrition per year is the relevant measure. But that's okay, let's do things your way. The CDC/National Center for Health Statistics should start reporting that attrition is 100% over the lifetime, so hey, we haven't made any progress on health since the start of the human race! Similarly, in your ships analogy, as I said above (and which you conveniently ignored, but hey, easier to sling insults than to make substantive points), it's not enough to know that 2 ships out of 10 sink. If it takes 5 years to sink, that's different from one year. I'm not sure why *you* have a problem understanding that, but I guess it's easier just acting like a tool.

The only reason why attrition over lifetime can be useful in the case of med school and residency is that you do not get use of partial completion. Furthermore, in the case of med school, lifetime is fixed at 4 years, so there's clearly a 1:1 mapping from annual attrition to lifetime attrition. But in residency, lifetime varies across specialties. Moreover, while partial completion is largely worthless, there are limited losses from early transfers. Since a large part of your "attrition" is transfers, this needs to be taken into account.

In any case, we clearly agree that it's 3-4% per year and 12% over the lifetime. If your argument is over presentation, you're acting quite the tool.
 
Every year 2 ships sink. Seems crystal clear to me.

As I said above, how old are those ships? If they're 1 day old, that's a high attrition rate. If they're 1,000 years old, that's different. How is this not hard to understand? In two countries of equal population, 10k people die every year. In country A, the 10k are 2 years old. In country B, they're 90 years old. Which country is better off?

And you claim others have no knowledge of statistics?
 
Wow. These numbers are enormous. 1344 are likely getting canned. Multiply that by $225k average annual income x 30 years = $9.1 billion per year in expectation damages.

That's the problem with residency. They need the cheap labor. However, there is not really a strong interest in seeing that the residents finish training because it would increase competition and may decrease salaries for established physicians. The only check is the ACGME disaccrediting programs. However the ACGME allows for a "reasonable" attrition rate.

Lawyers need to now about this figure. Then they can start bringing actions to recoup some of these damages.

The lawyers need to put a bullet in the chamber and let programs play russian roulette with their promotions decisions.

But first one of them has to figure a way to overturn key cases that allow "academic" dismissal for traditionally non-academic shortcomings, like smelling bad, looking unkempt because you been on duty for hours, or being unlikable.

I remain hopeful that some AMG's who did not match attend law school. They have the brains for it. Then they can use their skills to force change.

While they fight for change, they can survive by reviewing charts for malpractice suits. A report that takes two-three hours can pay anywhere from $300-1000. Or they can sue physicians and keep the fire on attendings who are benefiting from this $9.1 billion largess!

Medicine is a ruthless profession. It needs to be responded to in kind.
 
In any case, we clearly agree that it's 3-4% per year and 12% over the lifetime. If your argument is over presentation, you're acting quite the tool.

"lifetime" attrition is the relevant statistic because completing a partial residency is worth NOTHING. It only matters if you finish residency. I think the best snapshot is attrites/graduates per year as this gives credit back for transfers that actually finish residency.
 
"lifetime" attrition is the relevant statistic because completing a partial residency is worth NOTHING. It only matters if you finish residency. I think the best snapshot is attrites/graduates per year as this gives credit back for transfers that actually finish residency.

As I mentioned above, what makes lifetime less relevant is that fact that transfers get some credit (pgy-1), so counting "transfers" as attrition may not be correct.

But I agree with you, you'd want to just see how many people come in and finish--regardless of transfers. But that's not the data we have--but it's also why this 12% attrition rate is a large upper bound.
 
Wow. These numbers are enormous. 1344 are likely getting canned. Multiply that by $225k average annual income x 30 years = $9.1 billion per year in expectation damages.

I really don't know how enormous these numbers are--what's the attrition rate for other professions? Realistically, 3% per year/12% over 4-5 years doesn't seem that high compared to other professions--I hear that law is 20% per year, for example.

As for 225k average annual income, I suppose you need to take into account two factors:

(a) would the guys getting canned really be earning that? Not to be harsh, but presumably some (perhaps even a decent number) are being fired for a reason.

(b) even if they would earn 225k, presumably they're finding some other type of work, so the damage is net the salary in the job they do get (but then tacking on non-financial benefits of being a physician, such as actually enjoying the job).
 
I have some concerns with the assumptions in generating these numbers.

First, I think it would be fairest to focus on only core residencies. The numbers are presented for both core and subs. If someone "washes out" of a sub it's not the end of the universe. They are still a board certfied physician who can practice -- perhaps not in their field of choice, but that's life.

Second, I have a problem including transfers in this total. First, as mentioned above, many transfers are probably by design. Second, even if not, these people got another spot. This happens in all fields -- someone starts in one company, it doesn't work out, and they move somewhere else. Regardless, their career is still on track.

Third, let's exclude those that died. This has nothing to do with being "washed out".

Last, let's assume that all of the withdrawns are really fired. This is certainly an overestimation, as some will be by choice.

This gives us 258+863 = 1121 residents who were "let go" by core programs. Compare this to the denominator of 25346 = 4.4%. (4.4% over 3-4 years, 1.5% per year)

Now, we can argue whether 4.4% is too high, too low, or just right, but I think it's a much more accurate number than 12%.

And, if we assume that of the withdrawn, 50% did so voluntarily (I just made that number up), the actual failure rate is 2.7%.

And, this includes all comers -- US grads, US IMG's, FMG's, DO's etc. It's possible that the "failure" rate is very different between these groups.
 
I have some concerns with the assumptions in generating these numbers.

First, I think it would be fairest to focus on only core residencies. The numbers are presented for both core and subs. If someone "washes out" of a sub it's not the end of the universe. They are still a board certfied physician who can practice -- perhaps not in their field of choice, but that's life.

Second, I have a problem including transfers in this total. First, as mentioned above, many transfers are probably by design. Second, even if not, these people got another spot. This happens in all fields -- someone starts in one company, it doesn't work out, and they move somewhere else. Regardless, their career is still on track.

Third, let's exclude those that died. This has nothing to do with being "washed out".

Last, let's assume that all of the withdrawns are really fired. This is certainly an overestimation, as some will be by choice.

This gives us 258+863 = 1121 residents who were "let go" by core programs. Compare this to the denominator of 25346 = 4.4%. (4.4% over 3-4 years, 1.5% per year)

Now, we can argue whether 4.4% is too high, too low, or just right, but I think it's a much more accurate number than 12%.

And, if we assume that of the withdrawn, 50% did so voluntarily (I just made that number up), the actual failure rate is 2.7%.

And, this includes all comers -- US grads, US IMG's, FMG's, DO's etc. It's possible that the "failure" rate is very different between these groups.
You are again trying to redefine attrition rate in a manner not consistent with the literature on this topic.
In the literature as in the surgery study I posted above the attrition rate included those who left the surgery residency for ANY reason
I never used the term attrition rate as being equivalent to failure rate.
Failure rate is not the same in this situation.
Attrition is not completing the program you started.
If you started at Northwestern med school and flunked out or quit you may still end graduating from another school.
If you drop out of general surgery residency you still may end up completing an anesthesiology residency or EM residency, etc.
The attrition rate for Northwestern med school refers to not completing med school with your entering classmates.
The 12% attrition refers to not completing the initial residency program in which you entered.
A substantial number (likely > 50%) of this 12% will complete residency in another program or another field. However the process still involves a lot of stress even for those who are fortunate enough to secure a residency spot elsewhere.
 
Last edited:
The 12% attrition refers to not completing the initial residency program in which you entered.

Which is a meaningless statistic taken out of context as you have done.

Also, some people just need to be fired. Period. Even residents, fellows and attendings. Just like lawyers, baristas, investment bankers and mechanics sometimes need to be fired. The difference is that it's much harder to fire an incompetent resident than an incompetent barista.
 
Which is a meaningless statistic taken out of context as you have done.

Agreed. Although yes, perhaps it doesn't meet the definition of attrition in the provided link, do we really care how many leave without knowing the circumstances? I know I don't. It is a meaningless statistic without context - most of us don't care if someone leaves by transfer, if they died, changed fields, or willingly left medicine altogether. What we really want to know is how many left against their will, and under what conditions. Too bad we'll probably never get this figure.

Also, some people just need to be fired. Period. Even residents, fellows and attendings. Just like lawyers, baristas, investment bankers and mechanics sometimes need to be fired. The difference is that it's much harder to fire an incompetent resident than an incompetent barista.

Yep. I fail to see an atrocity in firing a resident if they deserved it. Not every "attrition" or firing is some misguided, power hungry PD taking it out on a poor innocent resident.
 
Which is a meaningless statistic taken out of context as you have done.

Also, some people just need to be fired. Period. Even residents, fellows and attendings. Just like lawyers, baristas, investment bankers and mechanics sometimes need to be fired. The difference is that it's much harder to fire an incompetent resident than an incompetent barista.

You are entitled to your opinion. The statistic is not meaningless IMO.
What is meaningless IMO is your unsubstantiated claim that it is much harder to fire a resident than a barista.
I have never heard of a national union getting a resident reinstated
However baristas have been reinstated through their union at Starbucks:
http://www.starbucksunion.org/node/1765
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=102x2863293
 
You are entitled to your opinion. The statistic is not meaningless IMO.
What is meaningless IMO is your unsubstantiated claim that it is much harder to fire a resident than a barista.
I have never heard of a national union getting a resident reinstated
However baristas have been reinstated through their union at Starbucks:
http://www.starbucksunion.org/node/1765
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=102x2863293

Meaningless is a harsh word. But without any context, how is one to interpret the statistic?

As others have pointed out, a large percentage of the attrition is due to transfers to other programs/specialties--not necessarily a bad thing. The relevant factor really is the degree to which people being fired are being appropriately/inappropriate fired--the answer to that question drives a differing set of policies depending on the answer.

Moreover, if the point you're trying to make is that the attrition rate is "too high"," we have to ask what the attrition rate is more generally. My understanding is that attrition is 20% per year for law associates. By that standard, 12% over three years is pretty tame.
 
Meaningless is a harsh word. But without any context, how is one to interpret the statistic?

As others have pointed out, a large percentage of the attrition is due to transfers to other programs/specialties--not necessarily a bad thing. The relevant factor really is the degree to which people being fired are being appropriately/inappropriate fired--the answer to that question drives a differing set of policies depending on the answer.

Moreover, if the point you're trying to make is that the attrition rate is "too high"," we have to ask what the attrition rate is more generally. My understanding is that attrition is 20% per year for law associates. By that standard, 12% over three years is pretty tame.

Interesting stat for the law associates.
I have seen residents fired inappropriately.
Whether the aggregate attrition of residents is "too high" or even "too low" is a matter of opinion.
I believe that some programs gloss over their problems during interview season and when residents see the reality they are disappointed. This can lead to transfers.
 
You are again trying to redefine attrition rate in a manner not consistent with the literature on this topic.
In the literature as in the surgery study I posted above the attrition rate included those who left the surgery residency for ANY reason...

I think you need to realize that the word is being used differently in that article though. If you want to examine attrition from surgery based on hour changes, you need to include transfers. That's really the focus. Folks who don't like surgery or like it better after hour changes are the crux of the article so they absolutely have to include transfers in that definition. However if you want to look at statistics and say how terrible it is that so many folks don't make it through "residency" (as opposed to a specific specialty within the term residency), then you have to exclude transfers because in this case they do, in fact, make it through a residency, just perhaps not in the field they started. So since the article you cited was concerned with attrition from a particular specialty, rather than from completing residency as a whole, it simply makes sense, for totally different reasons to include transfers. I don't think you can look at that and say that the "literature" has defined attrition to include transfers. Instead you have to actually look at the point being made. Here the point is that 80 hours has or doesn't have an affect on folks staying in surgery. The original point in this thread was OP's comment as to whether "there's a lot of pain behind these numbers" in which case transfers really shouldn't be counted. A lot of folks transfer for positive or benign reasons. This is not the same as washing out, getting thrown out, or side-lining your career. This could be a step up for a lot of people. You can't be myopic in how you read articles and proclaim that an article which is very definition specific, as the one you cited effectively defines a term for all purposes. It doesn't.
 
Wow. Go away for a day, come back bashed to heck.

Not that this should really be the focus of the discussion BUT this is the HR definition of attrition:

Attrition = Employees who left during period / ( Employees at beginning of period + Number of employees joining during the period)

If we take 3000 as the yearly out-of-residency rate (just as a number we all understand, even though I think we should take aPD's points as valid),

And we take 75,000 as the number of residents at the beginning of any one year

And we take 25,000 as the number of residents that enter at the beginning of the year,

then the yearly attrition rate (using these numbers) is 3%.

Notice that the rate is calculated over a particular period. Since your attrition number (3000) is a one-year figure, it makes the most sense to calculate a yearly attrition rate.

I think the general point of the post is more interesting than the squabbling of the numbers. It's just numbers, people, no reason to sling insults.
 
do these numbers take into account residents that take time off for research? Technically speaking, my future program will have a horrible attrition rate if it only looks at individuals who complete the program in 5 years. Of the current 8 chiefs, 5 took off 1-2 years for research, so only 3 graduated their original class (or, another way of putting it, of the original class 2010 residents, only 3 have graduated thus far. 3 More are currently PGY3's, another 1 is entering back next year as a PGY4, and 1 more is entering back next year as a PGY3 after 3 years of research...) The current group of PGY2's is going to be even worse. Of the 8 of them, 6 will be entering the lab, 4 of them with 2 year commitments and 2 of them with 1 year commitments. With all this flux, the program also gets years with only graduating 5 (last year) and has a year coming up with 9 chiefs (2 years from now). Not sure how they work it all out, but as a resident it seems good to know the program will work around your wants and not force you to enter the lab or leave the lab if you don't want to.

Using transfers, as others mentioned, depends on which questions you are asking. for what the original post meant to ask, I think they should not be counted... Also, to make the numbers actually make sense, we need a breakdown of PGY year to extrapolate... like, overall attrition could be 12% a year, but if its a 1% attrition rate for PGY2, a 0.5 for PGY3, a 0.1 for PGY4, and only 0.05% rate of attrition for PGY5, then once you get past that intern year hurdle you have tremendous rates of retention... if its 4% PGY1, 4% PGY2, 3% PGY3, 0.6%PGY4, and 0.4%PGY5, then that is a different story
 
It's exPCM - context is just another entity to be manipulated in order to "prove" some conclusion he's already reached.

Coming from you that is a compliment.

You have over 1900 posts but have only started 13 threads in more than 4 years.
This IMO is the epitome of someone who contributes little to this forum except lobbing criticism at the posts of others.
Do you have anything constructive to add here?
 
Coming from you that is a compliment.

You have over 1900 posts but have only started 13 threads in more than 4 years.

I'm sorry, I didn't realize there was a quota. I'll have to start padding my "thread start" count now that I know someone's keeping track. :eek:

This IMO is the epitome of someone who contributes little to this forum except lobbing criticism at the posts of others.

You know who else has a high "thread start" : "total post" ratio?

Hitl--

Er, trolls, that's who.

Do you have anything constructive to add here?

My comment regarding context was simply drawing (more) attention to how your conclusions don't follow the data you actually posted.

aProgDirector said it well. And more politely than me.

Finally, mocking people who take themselves too seriously while misunderestimastanding the data they post is in its own way, constructive. Perhaps the fact that I was mocking you has blinded you to the constructivity of my post?
 
I have seen residents fired inappropriately.

I am curious how you define this. I have seen residents who think they were fired inappropriately, and if you listen to them yammer on you would tend to agree with them. But if you talk to others you would find that the firing was not inappropriate at all. Some such residents have styled themselves as "iconoclasts" or otherwise smarter than everyone else, when in fact they were douchebags who didn't work well with anyone and were dangerous.

I may be misunderstanding you, but are you saying that the majority/most fired residents were unjustifiably let go? I do realize that many programs do not work hard enough to help out otherwise qualified residents who have some difficulties that can be corrected, but at some point you have to realize that many people just aren't cut out to be physicians.

That being said, I have seen residents who were basically forced from programs for being incompatible with the desired style. While not "fired," it reflects poorly on this program - but for statistical purposes it may not look as bad. However, I have also seen programs who are way too loose with their resident education and discipline, and end up graduating poor residents. Is this program a "bad" program because they take their job of producing quality physicians seriously?
 
this happens every year...same old statistics of people leaving, getting fired, etc....it's kind of sad how hard it is to get back into another program again after leaving a prior one for one reason or another. are there any statistics on that?
 
I am curious how you define this. I have seen residents who think they were fired inappropriately, and if you listen to them yammer on you would tend to agree with them. But if you talk to others you would find that the firing was not inappropriate at all. Some such residents have styled themselves as "iconoclasts" or otherwise smarter than everyone else, when in fact they were douchebags who didn't work well with anyone and were dangerous.

I may be misunderstanding you, but are you saying that the majority/most fired residents were unjustifiably let go? I do realize that many programs do not work hard enough to help out otherwise qualified residents who have some difficulties that can be corrected, but at some point you have to realize that many people just aren't cut out to be physicians.

That being said, I have seen residents who were basically forced from programs for being incompatible with the desired style. While not "fired," it reflects poorly on this program - but for statistical purposes it may not look as bad. However, I have also seen programs who are way too loose with their resident education and discipline, and end up graduating poor residents. Is this program a "bad" program because they take their job of producing quality physicians seriously?

I would define an inappropriate firing as one in which the resident was not significantly deficient or, if they did have deficiencies, was not given the opportunity to remediate.
I think too many docs talk about residents being "dangerous" while totally ignoring all the PAs and NPs being churned out to see patients with much less clinical training than residents. Many of these PAs and NPs end up being hired by physicians.
 
Last edited:
I would define an inappropriate firing as one in which the resident was not significantly deficient or, if they did have deficiencies, was not given the opportunity to remediate.
I think too many docs talk about residents being "dangerous" while totally ignoring all the PAs and NPs being churned out to see patients with much less clinical training than residents. Many of these PAs and NPs end up being hired by physicians.

I agree that many programs do not attempt to work with residents who can be salvaged. Sometimes residents just have trouble adjusting to the lifestyle - because residency is basically their first real job and all they know is school. Others have life changes which make them depressed or stressed. Others just need extra help or mentorship in learning how to be doctors.

In regards to your last point, that is not really relevant to whether residents can be dangerous or not. I would wager that people who consider some residents dangerous do so without any consideration of whether PAs or NPs are equally dangerous. It is not relevant. Individual PAs and NPs can also be dangerous. Whether one category is more likely to be dangerous is another argument which also is not very productive.

All the clinical training in the world does not make some physicians less dangerous. All of those who claim to be fighting for physicians against inroads by mid-levels really really need to acknowledge that there are dangerous physicians out there (they are not common but are out there). Don't blame the midlevels for their existence.
 
Top