Tag Archives | epidemiology

What, precisely, is the precision medicine initiative?

First of all, it’s a big deal, high on President Obama’s agenda and with its own page on the White House Web Site

Near term goals

One immediate goal of the Precision Medicine Initiative will be to significantly expand efforts in cancer genomics to create prevention and treatment successes for more cancers.

But the long term goals are broader

the Initiative will 1) support a national network of scientists who possess the talent and skills to develop new approaches for answering critical scientific and medical questions and 2) launch a national cohort study of a million or more Americans to propel our understanding of health and disease. The goal is to set the foundation for a new way of doing research that fosters open, responsible data sharing with the highest regard to patient privacy, and that puts engaged participants at the center.

Precision Medicine Initiative logo

You’ll soon recognize the PMI logo

Some specific research questions would be to:
  • Identify genomic variants that affect drug response
  • Assess clinical validity of genomic variants associated with disease
  • Identify biomarkers that are early indicators of disease
  • Understand chronic diseases and best management strategies
  • Understand genes/pathways/factors that protect from disease

In the process, we will learn about EHRs, mhealth, patient engagement and new research methodologies.


Workshops conducted by the PMI Working Group are open to the public if space is available, and can be viewed on webcasts. The recent workshop on Participant Engagement and Health Equity (July 1 and 2, 2015) was phenomenal, and worthwhile catching the video.

Epidemiology and data visualization

We’ve always understood the value of data visualization

John Snow

If you’ve taken a look at any overview about epidemiology, or attended one lecture on the subject, you’ve heard of John Snow, the Victorian anesthesiologist, who tromped around London in his spare time, asking people about their water supply (water was delivered by different companies, and drawn from different pumps, accordingly). He associated one water supplier with a higher incidence of cholera deaths, inferring that something in the water was causing cholera – and this was before scientists had embraced the germ theory, and well before pathogens had been identified as causes of infectious disease.

Data visualization

John Snow’s map showing cases of cholera in 19th century London.

In addition to his excrutiating hand calculations of infections and death rates, he mapped the data. We can see references to his maps on all kinds of data visualization sites. Epidemiologists have always known about his maps; now they are garnering attention from non-epidemiologists.

Florence Nightingale

Another Victorian who didn’t mind using pencil and paper to add up lots and lots of numbers, was Florence Nightingale. Yes, the lady with the lamp, was also an ardent mathematician/statistician. She invented a type of diagram, “coxcombs” to visualize mortality by different causes.

Data visualization

Florence Nightingale called these diagrams, “coxcombs”.

But was she an epidemiologist? The term “epidemiologist” wasn’t in use when she was doing her work, but “she used statistics to measure health, identify causes of mortality, evaluate health services, and reform institutions.” (Stolley and Lasky, Investigating Disease Patterns, 1995).

Variation in Vancomycin Use in Pediatric Hospitalizations in the 2008 Premier Database

How much variation in use is too much?

Vancomycin is indicated for the treatment of serious or severe infections caused by susceptible strains of methicillin-resistant (beta-lactam-resistant) staphylococci.  Because of concerns about the development of drug-resistant bacteria, recommendations to prevent the spread of vancomycin resistance have been in place since 1995 and include guidelines for inpatient pediatric use of vancomycin.  With such guidelines in place, it is of special interest to compare inpatient pediatric vancomycin administration across hospitals.

Our recent publication, “Pediatric Vancomycin Use in 421 Hospitals in the United States 2008” published in PLOS ONE on 8/16/2012 (Lasky T, Greenspan J, Ernst FR, and Gonzalez L), compares vancomycin use in all pediatric hospitalizations (hospitalizations of children under age 18) in 421 hospitals in the Premier database.

Key Findings

  • Vancomycin was administered to children at 374 hospitals in the Premier hospital database.
  • Another 47 hospitals with 17,271 pediatric hospitalizations (13,233 under age 2) reported no vancomycin use during 2008.
  • The number of pediatric hospitalizations with vancomycin use ranged from 0 to 1225 at individual hospitals.
  • Most hospitals (221) had fewer than 10 pediatric hospitalizations with vancomycin use in the study period.
  • 21 hospitals (5.6% of hospitals) each had over 200 hospitalizations with vancomycin use, and together, accounted for more than 50% of the pediatric hospitalizations with vancomycin use.
  • Percentage of hospitalizations with vancomycin use ranged up to 33.3% when hospitals with few pediatric hospitalizations were kept in the sample, the high percenetages being an artifact of the small number of hospitalizations in the denominator. For this reason, percentage, by itself, may not be a useful indicator in small hospitals.
  • In hospitals with more than 100 pediatric hospitalizations with vancomycin use, the percentage with vancomycin use ranged from 1.26 to 12.90, a 10 fold range in the prevalence of vancomycin use.
  • Our stratified analyses and logistic modeling showed variation in vancomycin use by individual hospital that was not explained by hospital or patient characteristics including: bed size, teaching status, region of the country, rural or urban geography, and patient sex, race, APR-DRG risk of mortality and APR-DRG severity of illness.

For Discussion and Further Investigation

Until recently, few studies have compared pediatric antibiotic use across large numbers of hospitals or geography, and it was not possible to assess variation in use across institutions. Hospital variation in care of adults has been studied for several decades, much of it made possible by large Medicare claims databases. With the availability of aggregated data for pediatric hospitalizations we can begin describing and attempting to understand variation in pediatric practice. This first study of hospital variation in pediatric vancomycin use raises questions for further research.

Crowdsourcing to obtain study samples and conduct epidemiologic surveys

It seemed like a fanciful idea, but the title caught my eye, “Crowdsourcing Graphical Perception: Using Mechanical Turk to Assess Visualization Design” (Jeffrey Heer, Michael Bostock ACM Human Factors in Computing Systems (CHI), 203–212, 2010).

OK, google “Mechanical Turk” – it is a service from Amazon that matches “workers or Turkers” with “requestors” for small tasks and asked “workers” to assess different visualization  designs.  If you know this already, you are way ahead of where I was last Tuesday. The authors got 186 respondents to assess some  visualizations for a total study cost of $367.77, and the authors were able to compare their study results to published literature to conclude that the method was viable.  If “Turkers” can assess visualizations, then they presumably they can respond to other types of questionnaires or surveys about a range of issues.

This piqued my curiosity, and I searched for more information on this issue.  One of the first questions of concern to any epidemiologist would be the degree and types of selection biases associated with using crowdsourcing and related approaches to obtain what are essentially survey samples and responses. I was surprised to see a growing body of literature in this arena, the most recent of which is Jennifer Jacquet’s article,on the Scientific American blog (July 7, 2011), “The Pros and Cons of Amazon Mechanical Turk for Scientific Surveys” (aren’t surveys some of what we do?).

I found several other studies characterizing the demographics of samples of respondents on Mechanical Turk:

“The New Demographics of Mechanical Turk”, March 9, 2010 Panos Ipeirotis, NYU School of Business,

Ross, Irani, Silberman, Zaldivar and Tomlinson, “Who are the Crowdworkers? Shifting Demographics in Mechanical Turk” CHI 2010,

Going further, some folks are pursuing the idea of using this methodology for subject recruitment in experimental research: “Using Mechanical Turk as a Subject Recruitment Tool for Experimental Research”, Berinsky, Huber and Lenz, October 7, 2011,

Clearly, some very bright minds are exploring the potential of this mechanism.  I would think that we epidemiologists will have a lot to contribute in this arena, and will see great benefits, as well.

Powered by WordPress. Designed by Woo Themes

michael kors outlet