September 23, 2019

Pregnancy Test for Female Abdominal Pain Patients (ACEP-24)

At-a-glance details of the measure

Formal Title
Percentage of emergency department visits for female patients aged 14 through 50 years old who present to the ED with a chief complaint of abdominal pain who have had a pregnancy test (urine or serum) ordered.

Patient Safety (as distinct from Patient Experience, Efficiency/Cost Reduction, Effective Care, and Population Health).

Process (as opposed to Outcome, or Efficiency measure)

Welcome to the first edition of an ongoing series, where the quality measures supported by CEDR are explained and discussed. The purpose of this series is to peek under the hood of these quality measures, to help educate emergency physicians about their rationale, history, and what goes into each measure’s calculation. I hope this will enable both practicing physicians in the ED and ED medical directors to choose appropriate quality measures for their practice, and helps them better document the critical data elements to optimize scoring of

The first measure I chose to examine is ACEP-24 – pregnancy testing for female abdominal pain pages. This is among the simplest measures to calculate, and also among the least controversial – nothing changes the
differential diagnosis of abdominal pain so much as a positive pregnancy test. No one wants to miss an ectopic or diagnose a pregnancy via CT scan. It’s no surprise ACEP-24 has roots that go back a long way – it’s based on an NQF submission from 2008, and includes citations from a 2000 ACEP policy and a 1989 Annals paper on the unreliability of patient history for determining pregnancy.

The challenge, however, is capturing patient histories and presentations, and physician behavior, in a way that CEDR can recognize - to facilitate an accurate calculation of the quality measure. You could write a detailed note about a patient’s recent hysterectomy, or you could document that you dipped the patient’s urine and checked the U-preg test yourself – but your EHR needs to capture this data discretely, and transmit that data to CEDR in a way that the registry can use to calculate the measure. When I say “discrete” data in this case, I mean an entry in the Past Medical History field of a chart, or the Orders area. Remember CEDR can’t infer meaning from notes – the registry has to look in pre-defined areas of a chart to see if a case meets criteria and quality standards were met.

Let’s take a look at ACEP-24’s numerator, denominator and exclusions, and how that figures in to the Merit-based Incentive Payment System (MIPS) calculation.

The numerator is Emergency department visits for patients who have had a pregnancy test (urine or serum) ordered. That’s fair, and it ought to be easy enough to capture orders from an EHR. But be wary of the point-of-care pregnancy test – like strep tests and guaiac tests, some POC pregnancy and UAs can be easy to perform without an electronic order. If this is a habit at your shop, this will hurt your numerator and thus, your overall performance.

The denominator for ACEP-24 is all emergency department visits for female patients aged 14 through 50 years old who present to the ED with a chief complaint of abdominal pain. This also seems fair, and straightforward.
But make sure you’re not artificially shrinking the denominator by eliminating a lot of related complaints – does your triage nurse pick a chief complaint from a narrow, well-defined drop-down list in your EHR, or are free-text entries allowed? Check your daily census to see if “abd. pain” and “distended belly” and “n/v with abdominal pain” are listed chief complaints – if so, it’s likely the denominator CEDR would calculate for this measure is smaller that it should be – which could serve to artificially magnify the effect of the numerator.

The exclusions for this measure make sense – An EP shouldn’t be expected to check a pregnancy test on a patient who has had a hysterectomy, or is post-menopause, or is obviously pregnant. But for these exclusions,
do a chart review: where are you capturing information on hysterectomy, menopause, and pregnancy status?

This information should exist in the patient’s past medical and surgical history, or problem list, or some other discrete field (for instance, some EHRs can capture visits that occur during “episodes of pregnancy” that persist
for a period of time, after a pregnancy is diagnosed). If it’s not, you’re missing key exclusions, inflating your denominator and likely making your practice look less safe than it is.

Performing well on these quality measures is important, when it comes to calculating MIPS scores. The Quality component of MIPS is a big fraction of the overall MIPS score – your practice is rewarded points based on
performance over a historical benchmark. Also your ED could stand to gain significant bonus payments if the physicians are performing above a certain threshold in their quality measures – and could get penalized if they
fall in the bottom quartile.

Fortunately, practices can choose which quality measures to report – so they should choose measures where they know they’re accurately capturing the data that will lead to a favorable calculation.

When our CEDR study group looked at data elements from a random sampling of EDs, we found considerable variability in a few of the elements that went into the calculation of ACEP-24. Some sites just rarely seemed to
capture a diagnosis of hysterectomy or menopause. Even the diagnosis of pregnancy varied – a few EDs seemed to have a rate less than 1%, while others reported 3-4% of their patients were pregnant. That variability
may simply be explained by variations in the ED patient population – but if I was at an ED that didn’t seem to report a lot of visits from pregnant patients, I’d be worried my EHR wasn’t transmitting this data to CEDR in a
discrete manner that CEDR could recognize. Maybe the ED docs know the patient is pregnant, and write about it in their notes, but unless pregnancy status is captured in a place like History or Problem List or Chief Complaint or Diagnosis, it’s going to be missed by the registry.

So even though ACEP-24 seems straightforward – a reasonable and relatively simple quality measure – be sure you’re capturing the right data elements in pre-defined places in your EHR. Particularly, make sure the point-of-care U-preg tests are documented discretely as orders, rather than a note. And make sure your exclusions like hysterectomy or current pregnancy