WhiteCoat

Stretching the Definition of “Quality”

Attribution: http://www.everystockphoto.com/photographer.php?photographer_id=12419

It seems as if some members of the American Academy of Pediatrics have determined that emergency physicians aren’t performing enough useless tests on teenage girls.

According to a study presented at this year’s AAP conference, only 19 percent of the 77 million girls between ages 14 and 21 that were examined in emergency departments between 2000 and 2009 received pregnancy testing. Subgroup analysis showed that only 42% of those patients complaining of abdominal pain received pregnancy testing and only 28% of those patients exposed to radiation that could cause birth defects received a pregnancy test.

In an American Academy of Pediatrics news release, the researchers, including study author Dr. Monika Goyal, found it “particularly concerning” that rates of pregnancy testing were low even among females with potential reproductive health complaints or with exposure to “potentially teratogenic” radiation (i.e. may cause birth defects) such as chest x-rays or CT scans.

First, note how Dr. Goyal and company mention nothing about the indications for performing pregnancy testing in the emergency department. In other words, they’re bashing emergency physicians for failing to order a test when they haven’t even described when the test should be ordered.

I’m sure that Dr. Goyal’s study didn’t look at repeat visits or specific patient histories. If they’re reporting on 77 million patient visits, they must be using aggregate data. Another problem with the study is that aggregate data doesn’t take into account all of the instances in which a pregnancy test may not be indicated in a female teen.
If the patient had a negative pregnancy test in the doctor’s office two days prior to her current visit, another pregnancy test probably isn’t warranted. If a patient who is known to be pregnant is also having abdominal pain, a pregnancy test is probably a waste of time and money. Patients with epigastric pain or “heartburn” probably don’t need pregnancy tests. Should we get pregnancy tests on every patient with classic UTI symptoms? I don’t.

It’s pretty ridiculous to call the lack of pregnancy testing “particularly concerning” without saying what the “concerns” were. Did the researchers find any cases of bad outcomes due to lack of pregnancy testing in the 77 million cases that they reviewed? I’m sure there will be a few, but then a thorough researcher would then analyze thoses cases to determine whether the bad outcome was due to a failure to perform a pregnancy test. This study did nothing of the sort.

I was also surprised by the specious logic that patients having a chest x-ray performed should have pregnancy testing performed because chest x-rays “may” cause birth defects. The amount of radiation in a chest x-ray is equivalent to the amount of background atmospheric radiation people experience just being alive for 10 days. It is also equivalent to the amount of radiation that someone taking two roundtrip flights from Washington DC to Los Angeles would absorb. Read more about it on Wikipedia if you’re interested. Calling for more pregnancy testing because chest x-rays may be “teratogenic” is like demanding that all women take a pregnancy test before taking a cross-country flight on an airplane because the increased radiation exposure in the upper atmosphere may be just as “teratogenic.”
Whoops. Better not give the TSA any ideas.

Another practical issue is that lead aprons block x-ray radiation. Often x-ray techs and/or doctors will just have a potentially pregnant patient put a lead apron over her abdomen while an x-ray is performed rather than obtaining a pregnancy test. Of course such actions would be “particularly concerning” because no pregnancy test was done.

In the AAP news release, the study’s author stated that the “findings underscore the need to develop quality improvement interventions to increase pregnancy testing in adolescent girls in the emergency department, especially among those with higher risk of pregnancy complications.”

Translation: This study has failed to show that failure to perform pregnancy testing has any relation to quality of care or that it has caused harm to even one patient.

Conclusion: This study was a complete waste of time and money.

3 Responses to “Stretching the Definition of “Quality””

  1. Jamie says:

    What’s funny reading this is that when I was in my late teens/early twenties, when I went to the doctor for ANYTHING, they gave me a pregnancy test eventhough it was not a possibility.

  2. Mary says:

    FYI, a lead shield blocks/absorbs some of the radiation, but internal scatter, which is actually the primary source, cannot be shielded for. In fact, the only time a lead shield is actually effective is if the area being shielded is either within the primary beam, or within 4 – 5 cm of the primary beam. Otherwise, shielding is pretty much done because that is what the patient expects. Here are the sources: NCRP report 147 and http://www.hps.org/publicinformation/ate/q4362.html

  3. Steve says:

    So this study has a MAJOR weakness- it is based on the NHAMCS database. This is a database of outpatient visits that has been data dredged a lot on the theory that it provides very large numbers to look at trends.

    Recent findings- pending publication in Annals of EM- 24% of all patients intubated in the ED go to non-critical care floors…WHHAAATTT??? Yeah, there may be the occasional overflow into a low nurse:patient ratio step down or a discharge after an intubation for an opoid overdose, GHB, etc. But 24%? There is NO WAY this percentage is correct but that is what the database data says.

    Same database, different study- only 56% of patients with ectopic pregnancy had a pregnancy test ordered. 56%???? Anything below 99% is just not reflecting reality. We just know that this cannot be the case.

    Interestingly enough, they listed all the “false positives” that the database listed had an intubation- patients with a chief complaint of “ankle sprain” “scabies” “hypertension” and “rash” to name a few. These were patients with benign diagnoses that were listed as being intubated by the database but we know that there is no way this can be true.

    So from now on, anytime I see a study based on the NHAMSC database I am skipping over it.

Leave a Reply


− six = 2

Popular Authors

  • Greg Henry
  • Rick Bukata
  • Mark Plaster
  • Kevin Klauer
  • Jesse Pines
  • David Newman
  • Rich Levitan
  • Ghazala Sharieff
  • Nicholas Genes
  • Jeannette Wolfe
  • William Sullivan
  • Michael Silverman

Subscribe to EPM