Your Amazon Orders Could Be Gold

By  |  November 5, 2018 | 

Suppose I told you one of your patients:

–Didn’t vote in the last five elections

–Bought a 24 pack of Snickers bars and 12 pack of Cheetos on Amazon every three months

–Participates in a  Burger King Whopper loyalty discount program

–Owns a Harley Davidson

What conclusions would you draw as they relate to future illness, adherence, or their state of health?

That is a darn good question. You might infer an awful lot–but would it tell you more than knowing the patient’s zip code and prescription refill list only? Sometimes simple wins.

This is the dicey world of analytics—and if you need proof have a look at what is out there now.

Which gets me to the matter I want to highlight—the results of a JAMA study whose primary conclusion I cite below (given what we are all trying to do at Geisinger I had tremendous interest in the findings).

“This work reaffirms that the social environment is associated with health outcomes. However, these results suggest that information about the environment in which a person lives may not contribute much more to population risk assessment than is already provided by EHR data. Although this result does not mean that integrating social determinants of health into the EHR has no benefit, researchers may be able to use EHR data alone for population risk assessment.”

We have tons and tons of data. Tons. But if you can reliably predict who will have an untoward clinical outcome, in the case of the study more frequent or higher numbers of hospital visits, with what we already had 3, 5, or 7 years ago, why go through the hassle. A “just the facts ma’am approach” might suit us all fine for a lot less sweat and cost.

So what did the investigators do?

Firstly, with rule building, you start with derivation set of patient data and test it in an f/u validation cohort (90 and 122K patients, respectively–a generous sample). The authors looked to see whether the model they assembled using neighborhood characteristics, again, added to what we already had, could predict those folks that might have a shorter time to use of the ED and wards, as well as a shorter time to hospitalizations due to accidents, asthma, flu, MI, and CVA. If you can identify them early, you intervene with preemptive interventions.

And I’ll be darn, contrary to my priors, the extra juice did not improve risk prediction of events beyond what clinicians could already glean from the EHR.  A patient’s age, sex, race/ethnicity, and insurance status plus a few other things might be as good as a data run with a variable stack of employment, education, and home value thrown in. Zip code is everything.

Here it is in the words of the researchers–repeated again:

 “[…] information about the environment in which a person lives may not contribute much more to population risk assessment that is already provided by EHR data. Although this result does not mean that integrating social determinants of health into the EHR has no benefit, researchers may be able to use EHR data alone for population risk assessment.”?

I have seen very few studies of this sort; clinically relevant modeling with social and behavioral variables included in a large dataset purposed for risk prediction. And as always, I discover. Lessons like this over a career humble and make you wise.

Not all that glitters is gold–and venture capitalists and futurists take heed.  Not every vendor or seller has rigorously studied its product, or outcomes of its application have gone unpublished. And in my book, that amounts to nothing more than high priced woo-woo peddling.

One Comment

  1. Avatar
    Jack Chase November 12, 2018 at 1:40 pm - Reply

    Brad- interesting post. To me, the benefit of information about social determinants of health is more useful related to treatment (both of the individual and to resource planning for population health) than it is for risk prediction. In San Francisco’s public health department, we serve clients disproportionately affected by homelessness, financial poverty, food insecurity, social marginalization, mental illness, trauma and domestic violence and substance use disorders. To create successful medical treatment plans for patients and populations, we have to treat psychosocial illness with equal sophistication and intensity as conventional biomedical illness. This seems to me the compelling reason that we should ask and document these factors. Best,
    Jack Chase MD, Medical Director of Care Coordination, Zuckerberg San Francisco General Hospital

Leave A Comment

About the Author: Bradley Flansbaum

Bradley Flansbaum
Bradley Flansbaum, DO, MPH, MHM works for Geisinger Health System in Danville, PA in both the divisions of hospital medicine and population health. He began working as a hospitalist in 1996, at the inception of the hospital medicine movement. He is a founding member of the Society of Hospital Medicine and served as a board member and officer. He speaks nationally in promoting hospital medicine and has presented at many statewide meetings and conferences. He is also actively involved in house staff education. Currently, he serves on the SHM Public Policy Committee and has an interest in payment policy, healthcare market competition, health disparities, cost-effectiveness analysis, and pain and palliative care. He is SHM’s delegate for the AMA House of Delegates. Dr. Flansbaum received his undergraduate degree from Union College in Schenectady, NY and attended medical school at the New York College of Osteopathic Medicine. He completed his residency and chief residency in Internal Medicine at Long Island Jewish Medical Center in New York. He received his M.P.H. in Health Policy and Management at Columbia University. He is a political junky, and loves to cook, stay fit, read non-fiction, listen to many genres of music, and is a resident of Danville, PA.


Related Posts

February 11, 2020 |  1
I am feeling discouraged. Actually, I am pretty frustrated. I truly believed that if we could “choose wisely” and cut out the nearly $1 trillion of waste in health care that we would make a real dent in costs in health care. Not only that, I even thought that decreasing these total costs of care […]
September 24, 2019 |  0
Look up from your stack of journals lately? You will note DNA analysis has gotten awfully popular. Mom, dad, and the rest of the clan want in and the temptation to obtain the blueprint that makes you, you has an almost hypnotic-like draw. You are curious if what grandma told you—half of your “G, C, […]
June 10, 2019 |  0
Public and private payers incentivizing providers (P4P) to coax desired clinical behaviors have failed. Both CMS derived measures and those used by commercial insurers, many of which overlap, fail to pass Good Housekeeping standards and lack the reliable characteristics indicators must possess. Whether those indicators are valid, attributable, or meaningful makes all the difference in […]