February 6, 2017

OB Coding: Delivering Accurate Coding Remains a Challenge: Part II

EDITOR’S NOTE: This is the second and final installment in a two-part series on OB coding. Part I appeared in the Jan. 31, 2017 edition of ICD10monitor news.

It is easy to identify an obstetrics inpatient who has delivered a child from the codes on her abstract. The primary diagnosis (PD) is always an “O” (for obstetrics) code. If she came in for an “unrelated” condition, there should be an O code – as a physician, I cannot recall a single patient who went on to deliver during an admission whose PD was not a complication of (or complicating) the pregnancy.

As an example, a patient in the third trimester who was involved in a motor vehicle collision and brought in for observation who went on to deliver would warrant the “O9A.22, Injury, poisoning, and certain other consequences of external causes complicating childbirth” code. You then would add the codes that told the remainder of the story: what was injured, the circumstances of the incident, the outcome of the delivery, how many weeks pregnant she was, how she delivered, whether there any other complications, etc.

There are O codes indicating that a condition in any other body system is impacting the pregnancy. If the rest of Chapter 15 doesn’t have a specific code, numerous “obstetric conditions not elsewhere classified which are complicating pregnancy, childbirth, and the puerperium” can be found in O94-O9A. Viral hepatitis, malnutrition, obesity, acute asthma attack (“diseases of the respiratory system”), acute appendicitis (“diseases of the digestive system”), cancer (“malignant neoplasm”) – everything is covered. When I reviewed charts and found a solitary code for anemia, for example D50.9, Iron deficiency anemia, unspecified, I knew a code was missing (O99.02, Anemia complicating childbirth).

OB cases are unusual in that a PD may be present on admission indicator-no (POA-N). A patient may come into the hospital full-term due to spontaneous rupture of membranes with spontaneous onset of labor. Under general coding rules, this would establish the principal diagnosis, because it is the reason that occasioned the admission. If the delivery is uneventful, it gets codified as O80, Encounter for full-term uncomplicated delivery. But if there is a complication, there is no available principal diagnosis code for “full-term SROM with onset of spontaneous labor.” Therefore, the guidelines mandate selection of the complication as the PD, albeit POA-N. For vaginal deliveries, the PD corresponds to the main circumstances or complication of the delivery. For cesarean sections, you choose either the condition that resulted in the performance of the cesarean or the reason the patient was admitted, even if it was unrelated to the condition resulting in the cesarean.

Cesareans warrant a deeper dive in general. If there is no indication (or more precisely, no medical indication –“OB going on vacation next week” probably isn’t really a legitimate indication), “O82, Encounter for cesarean delivery without indication” is the code. This should be a very rare occurrence, as it is inappropriate to subject a patient to a needless, risky surgery. If a patient goes into spontaneous labor prior to the date of a scheduled cesarean, “O75.82, Onset (spontaneous) of labor after 37 completed weeks of gestation but before 39 completed weeks gestation, with delivery by (planned) cesarean section” is utilized, with the reason for the planned cesarean section (such as cephalopelvic disproportion or maternal care for scar from previous cesarean) to be listed as the PD. Examples of other reasons for planned cesareans include O34.1_, Maternal care for benign tumor of corpus uteri (which requires an additional code, like D25.1, Intramural leiomyoma of uterus) and placenta previa.

In November 2013, the American College of Gynecology (ACOG) Committee on Obstetric Practice Society for Maternal-Fetal Medicine redefined “full-term” to reflect improved neonatal outcomes. The current definitions are:

  • Pre-term: < 37 weeks
  • Early term: 37 weeks to 38 6/7 weeks
  • Full-term: 39 weeks to 40 6/7 weeks
  • Late-term: 41 weeks to 41 6/7 weeks
  • Post-term: > 42 weeks
This set up a coding-clinical disconnect. Pre-term is in the tabular list as “before 37 completed weeks of gestation,” and we may infer “post-term” because it has a parenthetical clause of 40-42 weeks in the index. When ICD-10-CM guidelines are revised to reflect this, it will likely cause a DRG shift because “O48.0, Post-term pregnancy” is a valid reason for admission for IOL, and additional diagnoses often risk-adjust the DRG. If the patient is instead considered late-term, the additional diagnosis/complication will be moved to PD and there may be no comorbid condition to change the severity of illness (SOI) or add a comorbid complication (CC).

The provider needs to be crystal clear as to the reason that caused the patient to be admitted. If she is there for an IOL, what is the impetus? Common diagnoses for induction are premature rupture of membranes (due to matters such as duration of time between rupture and onset of labor), placental insufficiency, the fetus being large for its gestational age, oligohydramnios, pre-eclampsia, and term pregnancy with preexisting conditions that may jeopardize the fetus or the mother (such as diabetes or hypertension). If the history and physical has a laundry list of “pregnancy c/b (complicated by),” it may pose a problem for the coder to select the likely principal diagnosis. It would not be unreasonable to get the OB provider to offer the reason/s he or she thinks was the precipitant for IOL, by template.

There is no limit on having complications or preexisting conditions, so a patient may have more than one of the conditions listed as IOL reasons. There are also other conditions the OB provider considers in his or her risk assessment, and these often risk-adjust when documented correctly and picked up by the coder. Examples of these are malnutrition (specify severity), poor fetal growth, maternal care for cervical incompetence or recurrent pregnancy loss, obesity, or drug use complicating pregnancy. One in particular that clinical documentation improvement (CDI) personnel could be on the lookout for is depression, with all of its specificity (recurrent major depressive disorder or single episode with severity) documented.

Once the patient begins labor, complicating conditions may arise, and they may serve as the reason for the decision to deliver operatively by cesarean. These need to be documented in a codable format. Two illustrative examples I ran across are:

  • “Arrest of labor:” This does not index to “transverse arrest,” which codes as “O64.0, Obstructed labor due to incomplete rotation of fetal head,” nor to O63.0, Prolonged first stage (of labor) or O63.1, Prolonged second stage (of labor).
  • “Chorio:” 2017 edits of the ICD-10-CM codes gave us trimester-specific chorioamnionitis codes. Providers need to either spell it out once or create an acronym expansion so the coder can pick it up in compliant fashion.
Other conditions that precipitate a cesarean that I would like to point out are:

  • O66.4_, Failed trial of labor subcategory, including attempted vaginal birth after previous cesarean section, and attempted application of vacuum extractor and forceps
  • O75.2, Pyrexia (fever) during labor, not elsewhere classified. This is the code used if chorioamnionitis is ruled out. The consumption of resources is exactly the same, and they risk-adjust the same as well.
The reason for admission, induction, and decision to perform a cesarean should be consistent throughout the chart. It is confusing and problematic to have inconsistencies and contradictory information in the H&P, progress notes, delivery note, cesarean operative note, and discharge summary. Additionally, providers should be aware that a delivery note by a nurse is not amenable to coding by the inpatient coder. Sometimes I would find information in that note exclusively, and we couldn’t use it.

There are certain conditions that are found as “maternal care for” codes as well as signifying variants that obstruct labor. You don’t use both; if the condition was present prior to labor/childbirth, but didn’t impact on the labor or delivery, you use the “maternal care for” version. If it exists and then impairs the labor or delivery, you use the “obstructed labor” version. A prime example of this is:

  • O32.1XX0, Maternal care for breech presentation, not applicable or unspecified (fetus) versus O64.1XX0, Obstructed labor due to breech presentation, not applicable or unspecified (fetus).
If a patient has a multiple gestation, including a PD of complication specific to multiple gestations, a code from O30, specifying number of fetuses, placentas, and amniotic sacs is indicated. This is used in addition to the appropriate Z37 code, which adds the detail as to whether the babies were stillborn or live. If the provider refers to “Fetus A,” you are permitted to infer this as fetus 1, etc.

Did complications arise during the delivery or in the postpartum period (defined as from delivery to six weeks post-term)? “Puerperium” means the period between childbirth and the return of the uterus to its normal size. Be sure to select the correct code that designates trimester in childbirth (during labor), or in puerperium (postpartum).

The most commonly missed risk-adjusting factor is acute blood loss anemia (ABLA). When I evaluated an OB service line and determined the reason the case mix index (CMI) was so low compared to peers, the underlying cause was that they never documented ABLA. The chair told me, “all our patients have ABLA!” I emphatically agreed. My recommendation was to use a combined threshold estimated blood loss (EBL) and drop in hematocrit, as well as to document the treatment or monitoring of the ABLA. If they were not going to transfuse, I suggested documentation of iron therapy, and repeat hemoglobin/hematocrit levels were planned. Coupling this with education noting that sustaining ABLA is not a patient safety indicator was sufficient to change behavior and improve CMI.

The last pieces of information that should be on every record with a delivery are an outcome of delivery code from Z37 and a listing of weeks of gestation designation from Z3A. The exception to this is when a patient delivers prior to admission to the hospital. In that case, you only use Z39.0, Encounter for care and examination of mother immediately after delivery, since she didn’t actually deliver during this admission and wasn’t technically pregnant during this admission.

If the provider paints a picture of the backstory to the pregnancy, the circumstances of the admission, labor, and delivery; any complications that arise; and what became of the pregnancy, you should be able to code it, thereby telling the story in codes.

The key is to think about how the situation unfolded and to be able to compliantly code it. This may take querying of a group of providers who we really never asked anything of before, and who will need some training as to how to respond.

I have confidence you are up for the challenge; and I hope this article helped.
Erica E. Remer, MD, FACEP, CCDS

Erica Remer, MD, FACEP, CCDS has a unique perspective as a practicing emergency physician for 25 years, with extensive coding, CDI, and ICD-10 expertise. As Physician Advisor for University Hospitals Health System in Cleveland, Ohio for four years, she trained 2,700 providers in ICD-10, closed hundreds of queries, fought numerous DRG clinical determination and medical necessity denials, and educated CDI specialists and healthcare providers with engaging, case-based presentations. She transitioned to independent consulting in July, 2016. Dr. Remer is a member of the ICD10monitor editorial board and co-hosts Talk Ten Tuesdays.