Page 1 of 1

SAS programming for biology PhDs

PostPosted: Mon Aug 29, 2016 9:11 am
by D.I.P.
Hi, I am in 3rd year of my PhD in molecular biology/cell biology where I don't have any programming skill and have no experience in handling big data. I may pursue a career in clinical research in future. Should I take a SAS certification?

Re: SAS programming for biology PhDs

PostPosted: Tue Aug 30, 2016 7:07 am
by D.X.

First a comment, you say you "may" pursue a career in clinical Research in the future, what does that mean? It may be beneficial to share what exactly you'd like to be doing.

Second, the SAS certification in the absense of knowing what you want may not be the best course of Action. Not all or not many Jobs in clinical Research do not Need someone who knows SAS.

That being said, those who are SAS trained and use that a part of thier every day Job are staticians. The clinical Trial Manager or trial lead will hardly if ever get their Hands wet on SAS, they go to thier Stat's Person. Is this what you're looking for? In clinical research they Play a key role supporting or own the statitical analyis plan (SAP) for a clinical study (which is an offical part of the clinical study protocol and Report, CTP and CSR respectively) - both studies of an interventional nature i.e RCT and non-interventional nature, i.e. observational Need SAPs and staticial Expertise to guide design and interpretation. An upside to this Job is that you are actively involved in study planning and you are among the first if not the first to see the data coming out of a study. However can be a very operational role spending alot of time QA/QCing data.

The other People who may know a bit of SAS programing though they may use other programs are Epidmiologists this is where you get into big data analytics, i like to think of big data analytics when it Comes to clinical data as a Marketing re-branding of what we call epidemologic Research. Same Statistical and investigative approaches are used (pooling, extraction, Analysis etc.). Is this where you want to go?

I first recommend you have an idea of what you want to do, i.e. is it Epidemologic type Research (generally hypotheisis generating, and providing so called 'real-world' insights into Patient outcomes)? or statistics suppoting a clinical Trial team? Then in that case get a good biostatistics Background and experience. Then if warrented worry about SAS programing later. SAS Training with context or statistics Expertise is well i think useless especially when not linked to a solid career strategy which may be your case still as a 3rd year PhD Student.

Do your career Research first maybe get a Hands on experience then take a decision. To reiterate those who SAS expereince are generally deep dive statistics experts. The spend lots of time in closed rooms, florencent lighting and tend to suffer Vitamin D deficiency and anemia. (just kidding).


Re: SAS programming for biology PhDs

PostPosted: Thu Sep 01, 2016 1:59 pm
by Cory
Just a tidbit on software used for data analysis. I'd recommend that you look at "R" as a way to get started and explore your interest in stats before diving into any certification.

The software is open source, widely used, growing acceptance in corporate and regulatory space and superbly useful even for individual small project use.


Re: SAS programming for biology PhDs

PostPosted: Thu Sep 01, 2016 5:35 pm
by D.I.P.
Thanks for your comments. I basically work on liver inflammation and immunology where I mostly do wet lab experiments. I am not quite happy with my career and I'd like to venture into epidemiology/clinical research.

Re: SAS programming for biology PhDs

PostPosted: Sun Sep 04, 2016 11:46 am
by D.X.
To what extent have you done your research to understand the subject matter. ? Do you know the types of questions asked and are you aware of the statics or typed of statistics appied? The is alot of correlative statistics used. Remember that observational or epidemeological studies are hypothisis generating and do not show statstically causality only linkage.association. ...irrespective of prospective or retrospective. How is your regression analysis staistics? You will work in the world of odds ratios and risk ratios. Etc etc. have you read some good epi studies?

Start to familiarize your self a bit. A bit off theapeutic area for your but i like Nephrology. Look up papers by Kalantar-Zadeh from the academic side..a well accepted and followed epidemiologist in that field plus an practicing HCP. From the industry side look up papers by B Bradbury - also on the neph side but a bit dated he along with K-Z have a few good papers on methodologies applied that can translate easilty to other theapeutic areas.

Then from there you can bridge to the bigger ones.

Good luck.


Re: SAS programming for biology PhDs

PostPosted: Wed Jun 28, 2017 4:30 pm
by S Mehra
I will also love to hear how to transition into a epidemiology or clinical research job.
I have done a doctorate in genetic epidemiology and thus know (if not advance but) statistics for analyzing case control studies-OR, RR, CI, power, Correlation, Logistic regression etc.But have no programming background. Therefore, I am currently educating myself with base SAS & clinical SAS and eventually would like to appear for the base SAS certification. Thereafter, I want to move to clinical research/epidemiology based job. What i am not sure whether my academic and previous work experience will hep me get a postdoc or a clinical research associate or clinical data manager position? Whether it will be wise to pursue more of epidemiology based or clinical trial related jobs. Please guide.

Re: SAS programming for biology PhDs

PostPosted: Sat Jul 01, 2017 1:50 pm
by D.X.
Hi Mehra,

If you are looking for Clinical Res. associate or clinical research jobs in a company, those are usually linked to clinical trials or the conduct of. Staticians are always needed not so much fir cRA jobs but clinical reseach yes.

There are jobs, where you maybe be involved in obervational studies that can fall under clinical research and development - these studies use the epideomlogic statistcs you describe.

programming knowlefge can be nice to have when, say pooling data, but running analysis and making sense of the data is key.

There is need for good epedemilogists and associated statistican knowlegde, one hot area yiu can look at is Health Technlogy Assessment, HTA, where epi knowledge is key to understand and track value for medicine both at industry anf government levels.

Another area and say buzz word to be attentive to is "Real World Evidence" - here observational data sets from a variety od sources such as claims data or registries are if interest - analysis here uses the epi statistics you described - again both government and industry are playing here - becoming more inrensified as health care budgets become more constricted and limited so linking back to HTA. A key issue with RWE is dara quality so maybe programming skills can help to a certain extent but most you cant control so bak to how you analyze and interpret.

Inside a pharma there are epidemiology groups who work criss functionally with clinical development, pricing and market access, policy, and where relevant commercial teams etc.

Best DX