Posted by: notdeaddinosaur | August 20, 2011

Meaningful Use Requires Meaningless Data

Race is a medically meaningless concept.

Spare me the few tired cliches about prostate cancer, diabetes, and sarcoidosis being more common in blacks than whites, or even the slightly increased risk of ACEI cough in patients of Asian descent. We screen Jews of Ashkenazi descent for Tay Sachs without any racial labeling. All that information is readily accessible under the Family History section of the medical history. It is no more than custom which dictates the standard introductory format including age, race, and gender. It turns out I’ve blogged about this before at some length (pretty good post, actually). What is new is the advent of electronic medical records.

Much hullabaloo has been made about federal stimulus funds allocated to doctors as payments for adopting EMRs; “up to $44,000!” Here’s the problem with that figure, though, including how it breaks down (source here):

[M]aintaining [an EMR] costs multiple thousands of dollars a year. Bear in mind that they’re not talking about a lump sum payment of $44,000. It’s $18,000 the first year, $12,000 the second year, $8000 the next, $4000 the next, and then $2000, for a total of $44,000 spread over five years. FOR A SYSTEM EXPECTED TO COST AN AVERAGE OF $10,000 PER DOCTOR PER YEAR, not counting the start up costs, which run in the vicinity of $50,000. $44,000 over five years for something that will cost us $90,000 over the same period? And that’s even if they actually get around to giving out the money in the first place! According to this, in order to qualify for “meaningful use,” EMRs must be used for ePrescribing, for communicating with other EMRs like labs and hospitals, and for transmitting information on performance measures (the paternalistic proxy for “quality”) to the government.

Just because my electronic systems didn’t end up costing me anything, it turns out that even though I bill Medicare less than $25,000, I’m still eligible to apply for some of the stimulus money. So just for shits and giggles, I hooked up with a government funded entity whose stated purpose in life is to help me get that money. Cool.

I’ve had a couple of visits with them so far. It turns out that my freebie EMR has features which I hadn’t bothered using yet, mainly because they didn’t seem particularly useful in the provision of medical care — that’s what I do, remember? — to patients. One of them was a so-called “Demographics” section, right below such vital information as patient name, address, phone numbers, and birth date. This section contains three pieces of information I have to enter, one from a set of radio buttons, and two from pick lists, mechanisms that allow for the collection of what is known as “structured data” instead of just information I type into the EMR “free form”.

The first item is “Ethnicity”. There are three radio button options: Hispanic, Non-hispanic, and Unspecified (the default).

The second item is “Preferred Language”, to be selected from a pick list. I can only enter one option.

The third item is “Race”, again to be selected from a pick list. They include “African or African American”, “Asian or Asian American”, “European or Caucasion American”, plus several other basically meaningless classifications. (For example, what entry do I use for an individual from the Indian subcontinent?) Unlike “Language”, I can enter as many of these options as I wish.

What? The? F?

Aside from the language entry, which could perhaps be useful in a very large, very diverse practice, neither “Ethnicity” (limited to Hispanic or Not) nor “Race” has any possible legitimate bearing on diagnosis, treatment, or any other aspect of medical care. And yet an integral part of Government-defined “meaningful use” consists of completing this section of the medical record.

Interestingly, a stated later requirement is for me to submit information from my EMR to the government, ostensibly for what they’re currently calling “reporting purposes”.

Now, what government function uses demographic data like race and ethnicity (ie, Hispanic or not)? Would that be the tracking of, say, voting patterns? And doesn’t it seem like a handy way to collect that data, neatly sorted by address and birth date, rather than having to use the decidedly old-fashioned, up-to-a-decade-out-of-date but actually legal way of tracking that information through the census?

How Orwellian to require that “Meaningful Use” incorporate the recording of medically meaningless data.


  1. To paraphrase an old magazine article, the government already knows where I live and work, how much I make, what I drive, if I pay my taxes, and when I travel out of the country.

    And, by the way, didn’t they get this race and ethnicity information the first hundred times they collected it from me directly? What, does this info change from year to year? Can I be a non-Hispanic Asian this year and be a non-Hispanic African-American the next? It would be fun, though my mom might object.

  2. The reason for this is that the government is trying to track “disparities in healthcare” based on race. It seems that in particular inner city blacks don’t live as long and have more chronic and acute medical problems. The liberal theory is that this is merely because they are black and that white people get better health care because they are white.

    The conveniently ignore myriad social issues that complicate the picture and which are not amenable to blaming the medical establishment.

    So, now you will be monitored to see if you are discriminating in your treatments based on the patient’s race.


  3. When I was a research assistant, we had to collect this info from participants and report it (in a pooled manner) to the feds in our grant progress reports. It always felt invasive to me to ask about race/ethnicity. I used to let participants fill it out themselves and my instructions included “if you are comfortable indicating this…” as well as information that we were asking because we were federal grant funded and because the agency (usually NIH in our case) wanted to be sure we were collecting data on all humans and not just white men. Most people filled it out; several found it confusing. I was told by my P.I. later that the researchers were supposed to fill this out. I found this insulting to the participants – so I’m just supposed to say “yeah, he looks totally white”?

    One of our research faculty members, who had lived in a particularly ethnically divided part of Europe during some very bad times, flat out refused to deal with the form. I guess if you grow up with entities (e.g. Nazis) who will hunt people down to persecute and/or kill them because of ethnic background, you become a bit extra sensitive to disclosures of that information.

    It’s a shame that the activities and uses you need to engage in to be eligible for credit for using EMR are so foolish. I’m a fan of EMR in concept, but like so many things…in practice (and management) it leaves me pretty cold.

  4. For some reason, being in Texas makes me not want to identify as “White”. When I was applying for a driver’s license, I tried leaving the Race question blank, but they wouldn’t let me. Nor would they let me put “Jewish” as a race. Ultimately, I just put “Other” which is amusing because I have previously claimed to be white.

    Ultimately though, since it is meaningless, it doesn’t really matter what you put. It’s not like the government is going to go run genetic tests on all your patients to make sure they are the exact race you say they are.

  5. The Feds always want this (and a lot of other) info when federal money is being spent, to document for us pesky taxpayers how they’re spending our money. Asking for race info is sometimes a legacy of the not too distant history when public funds can and were distributed and used in racist ways, and sometimes a way to determine whether particular programs or interventions are reaching underserved groups.

    It was a particular pain in the neck when I was writing grant proposals for public health programs and had to incorporate special approaches to reach a particular racial group that constituted an infinitesimal percentage of the population in my state.

  6. There is actually a bigger purpose to marrying the race/ethnicity and other demographic details along with your diagnosis and treatment patterns under these new rules/guidelines. One of the other, rather large, components of this healthcare reform act was to establish organizations that are solely tasked with researching, studying and evaluating the comparative effectiveness of drugs. There are certainly diagnoses and treatments that are more or less effective based upon a patient’s demographic profile and understanding these and their relation to disease prevalence and treatment effectiveness is a main goal of these agencies. While evaluating the cost of these effective treatments is not included in these analysis, it’s likely that will also eventually be part of the decision-making process. But ideally, it’s to ensure that the most effective treatments on the market are available for coverage universally.

    I acknowledge that demographic information about patients is certainly captured during the census, but as our government agencies don’t speak to each other, this info is not married with the EMR data that is being sent; therefore, they ask again.

    DinoDoc, I agree that the specificity with which you are allowed to choose in your forms is ridiculous. Very rarely in the data I use am I given enough detail to make use of this information appropriately. If not mandatory for you to fill out in the EMR, then I doubt we’ll improve usefulness going forward.

  7. […] […]

  8. SusieQ324 – The comparative effectiveness of drugs and the prevalence of diseases is dictated by the specific genes a person has, and attempting to try “guessing” what genes a person has by asking them their race/ethnicity is a failed attempt at practicing medicine. To truly determine the relative effectiveness of different potential drugs one would have to sequence the DNA of the pool of patients, identify what genes indicate increased susceptibility to side-effects and then encourage everyone that may take the drug to have their DNA sequenced to determine if they are susceptible to suffering from the side effects. Simply asking a person their race/ethnicity does not capture any medically meaningful data.

    What happens when you have the South African whose parents were emigrants from Spain in the 1700s stating that they are African-American and not Hispanic.

    Meaningful Use is only Meaningful in that it demonstrates the government is too busy using concepts proven wrong by modern technology while technology continues to leave the true dinosaurs we call bureaucrats further and further behind.

  9. I’m all for research and learning for the greater good, but why would they only want to know if you are HISPANIC OR NON-HISPANIC? Even if you could justify using this information to help mend the “disparities medical care for minorities”, this will never be acheived.

    The major point no one has mentioned is this: you will only be able to track those MEDICARE patients that SEEK TREATMENT. It is not a true study of the population, because this means of collecting information for “meaningful use” is neglecting to account for those that are not covered on medicare (64 years and under) OR those not seeking treatment.

    For the Meaningful Use for Dummies version of requirements see below:

    Click to access Beginners_Guide.pdf

    Another thing no one else has mentioned… the only points/rules/regulations (whatever you want to call them) in the meaningful use requirements that MAY actually help patient care are not monitored in any way. For example, providers are required to provide patients and “electronic copy” of their medical records within three days of a request. Providers have to report when they were requested AND when they were given. 1) You are exempt from reporting this if no one requests their records electronically. 2) It would just be easier to NOT report it to the system in the first place 3) A majority of medicare patients have trouble with email, we have not received a SINGLE request for an electronic copy of their records. 4) One of the “benefits” of the EMR systems is all the “defaults” you can place in the system. You can choose to default “gave patient electronic records” when you select “patient requests records”. So basically, there’s no checks or balances. You can almost report anything you want. (Not that good doctors would, but it’s the bad ones I’m pretty sure they are trying to regulate….unsuccessfully.

    Good doctors will choose not to accept medicare anymore because of all the hassle and extra administrative work they will entail (and soon penalties). The good doctors will have enough patients with “other” insurance companies to get by. The bad doctors will be all that will be left for medicare patients. I wonder what their care will be like then.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: