Biostats PhD for not a huge bio fan.

[trynagetby]

Hi all, so I've been trying to think of ways to broaden my application chances. One of my rec writers recommended bio-statistics programs. To be honest, while I don't hate biology I've also never been particularly interested in it. I enjoyed a biostatistics REU but more for the statistics side part than the bio. I don't mind and think I would actually enjoy the more applications-focus of biostatistics programs. Would it make sense to apply to biostatistics programs for me or would I end up being miserable?

Edited October 14, 2020 by trynagetby

[StatsG0d]

They don't really teach any biology in biostatistics programs. It's only necessary to learn/re-learn bio if you're interested in statistical genetics.

[Geococcyx]

You can definitely find the sorts of statistics you like in biostat departments.  Just to drive the point home, it looks like you did some research on Bayesian classification -- a reasonable example of how that may work in biostat may be Briana Stephenson, who worked on Bayesian clustering methods at UNC biostat (and now is a prof at Harvard biostat).  Out of your other possible interests from your profile: plenty of biostat folks do causal inference (e.g. Hudgens, also at UNC), and for your more machine-learning sorts of approaches, one example may be the sort of precision medicine work that Kosorok at UNC biostat and Laber at NC State (which is combined stat/biostat) have done.  These examples are obviously north carolina-slanted, but this isn't a unique situation (e.g. Witten does biostat ML at UW).  To StatsGOd's point, most of these applications don't require that you be an expert in biology (and correspondingly, there are plenty of geostatisticians who aren't earth scientists and whatnot).

I think bayessays has some nuanced opinions about whether biostat departments are the right choice for you in terms of being miserable or not; I'll let them weigh in if they desire and eschew addressing it myself.

[bayessays]

1 hour ago, Geococcyx said:

 

I think bayessays has some nuanced opinions about whether biostat departments are the right choice for you in terms of being miserable or not;

I think people who are not interested in applied statistics at all and want to do more theoretical work should go to statistics programs.  Also some biostatistics programs do not have opportunities to teach since the departments don't have undergraduate students.  OP sounds like they would enjoy biostat.

[trynagetby]

Thank you everybody for your super thoughtful and helpful advice. I'll definitely add some biostatistics schools to my list.

7 hours ago, StatsG0d said:

They don't really teach any biology in biostatistics programs. It's only necessary to learn/re-learn bio if you're interested in statistical genetics.

I was more worried that the research involved would require some in depth biological knowledge even if self-studied/gleaned. Is it possible to succeed in bio-statistics with only a superficial understanding of biology?

[Casorati]

1 hour ago, trynagetby said:

Thank you everybody for your super thoughtful and helpful advice. I'll definitely add some biostatistics schools to my list.

I was more worried that the research involved would require some in depth biological knowledge even if self-studied/gleaned. Is it possible to succeed in bio-statistics with only a superficial understanding of biology?

Biostatistics is still statistics with more emphasis on medical data. The bio prefix is misleading since biostatistics has little to do with biology. I have not taken a single course in biology and I still did and probably will do research in biostatistics.

Edited October 15, 2020 by Casorati

[cyberwulf]

1 hour ago, trynagetby said:

Is it possible to succeed in bio-statistics with only a superficial understanding of biology?

Yes. 

You'll probably find as you get engaged in research you'll learn (and hopefully become interested in) some of the underlying biology related to the problem(s) you're studying. A lack of interest in biological applications would make someone a bad fit for a biostat program, but a lack of experience and knowledge isn't an issue.

[bayessays]

10 hours ago, cyberwulf said:

A lack of interest in biological applications would make someone a bad fit for a biostat program

I think it's probably important to point out to someone unfamiliar with biostatistics programs that "biological applications" is pretty loosely defined.  A few people do genetics/bioinformatics stuff, but is analyzing clinical trial, epidemiological data, or health records really biological in any meaningful way where somebody not liking "biology" would be an impediment?

[StatsG0d]

4 hours ago, bayessays said:

I think it's probably important to point out to someone unfamiliar with biostatistics programs that "biological applications" is pretty loosely defined.  A few people do genetics/bioinformatics stuff, but is analyzing clinical trial, epidemiological data, or health records really biological in any meaningful way where somebody not liking "biology" would be an impediment?

I agree and I was going to comment something similar. I would say that you at least should have some interest in public health. If you're just in it for the math, you'd probably be happier at a statistics department.

[cyberwulf]

12 hours ago, bayessays said:

I think it's probably important to point out to someone unfamiliar with biostatistics programs that "biological applications" is pretty loosely defined.  A few people do genetics/bioinformatics stuff, but is analyzing clinical trial, epidemiological data, or health records really biological in any meaningful way where somebody not liking "biology" would be an impediment?

I see your point, but being in a department is about more than your dissertation work. Faculty and most students in biostat departments are excited about working on biomedical problems, and so while someone without much interest in biology might be able to get through the Ph.D., they would likely feel quite isolated from their peers and mentors that they don't share a passion with.

[bayessays]

9 minutes ago, cyberwulf said:

they would likely feel quite isolated from their peers and mentors that they don't share a passion with.

Absolutely, I think this is an important point that people overlook.  I absolutely agree with@StatsG0d that you need some interest in public health.

[StatsG0d]

9 hours ago, cyberwulf said:

I see your point, but being in a department is about more than your dissertation work. Faculty and most students in biostat departments are excited about working on biomedical problems, and so while someone without much interest in biology might be able to get through the Ph.D., they would likely feel quite isolated from their peers and mentors that they don't share a passion with.

I think it's naive to think that all (or even the vast majority of) biostats faculty members are fully engaged / committed to public health and/or biomedical problems. There are many professors who are in biostatistics simply because applications of their theoretical research area naturally fits in the realm of biostatistics (e.g., imaging, dimension reduction, historical data borrowing). This is probably mostly true for faculty members who were trained in stats departments rather than biostats departments. 

That said, I wholeheartedly agree with you regarding students. 

[trynagetby]

Thank you everyone for your thoughts. I reviewed the top bio-statistics departments and definitely think I would as happier, if not happier in them as in a pure statistics program. I'll be applying to a mix of programs now.

[cyberwulf]

On 10/16/2020 at 9:19 AM, StatsG0d said:

There are many professors who are in biostatistics simply because applications of their theoretical research area naturally fits in the realm of biostatistics (e.g., imaging, dimension reduction, historical data borrowing).

Possibly, but such faculty are a dying breed. The trend in biostatistics is towards being more deeply embedded within the biomedical domains they specialize in.