Apgar Wars ~ 4th in series: Questions about study’s data — the positive as well as the negative data

by faithgibson on September 22, 2013

in Contemporary Childbirth Politics

Last week Medscape’s posted research by Dr. Amos Grunebaum, Chief of L&D New York-Presbyterian and associate professor at Weill medical College of Cornel University, NYC and his co-author, Dr. Frank Chervenak, director of Maternal-fetal medicine @New York-Presbyterian, who also was an associate professor at the Weill medical College of Cornel University.

These two heavy-hitters used data from the CDC’s national birth certificate databased on live births to compared ZERO five-minutes Apgar scores of newborns between:

  • hospital births attended by MDs and nurse midwives
  • planned home births attended by ‘other’ midwives

Using a newly created stand-along data point for  “zero 5-min Apgar” that has never been scientifically determined to be either accurately or consistently reported, or relevant when removed from the 1- and 10-minute Apgar, these obstetrician-researchers claim to have statistical proven that PHB with “other’ midwives have an 10 to 19 times increase in zero Apgar at 5 minutes by newborns when compared to physician and nurse-midwife attended hospital births.

A ten-fold increase is equal to one baby with a zero Apgar out of every 615 births attended in an OOH setting by a non-nurse midwife.

If true, this describes a life-threatening situation in which a normally-formed, apparently healthy baby who is born with a beating heart suffers a total cardiac and respiratory arrest within less than 5 minutes of its birth. The reason for this is usually unexplained in the moment, but generally presumed by the medical profession to be the result of extreme oxygen deprivation during labor that went on recognized either by L&D nurses or the birth attendant.

This is the ultimate childbirth-related crisis, as under these circumstances such a baby will very likely die or suffer serious neurological damage. This kind of adverse outcome is assumed (until and unless proven otherwise) that whoever was providing care during the labor must have missed crucial indicators that the unborn baby was in trouble. In regard to women planning a home birth and thus laboring OOH, it will be assumed to be the midwife’s fault unless another explanation become obvious after the baby is examined, such as cardiac anomaly or other birth defects.

To have this situation occur with an orders-of-magnitude greater frequency (that is, a 10- to 19-fold increase) in only one setting, or in relation to only one type of birth attendant would indeed be a call for action, both to determine why AND to figure out how to eliminate the problem.

So why aren’t we with moving Heaven and Earth to “determine why AND to figure out how to eliminate the problem?

Questions about both kinds of date — the positive as well as the negative data ~

The two big practical questions is whether:

  • hospital-based birth attendants have figured out how to virtually eliminate critically-low Apgars scores in babies that ultimately die of other causes (since the NNM rate is 2 per 1,000)
  • and/or
  • how such an egregious NNM rate for OOH midwifery could have gone unnoticed by midwives and been missed  scientists until now?

What we do know

The statistician who is in charge of the CDC’s collection of birth records has publicly stated that Apgar reports are in general notoriously unreliable, both in terms of  being consistently included on all birth registration records, and in relation to the accuracy of any one individual recorded number.

Unfortunately for those of us trying to make sense of the study’s claims,  its obstetrician-researchers didn’t include the other 2 critical pieces of Apgar data — the scores for 1-minute and 10-minutes.

The study also did not include any information about other complications, such as a cord accident, shoulder dystocia, or if its mother experienced a placental abruption or other unpredictable emergency. It also does not tell us whether the labor and birth were unattended,  or midwives arrived as the baby as the baby was being born or very shortly afterward.

As a result,  no one can tell what the conditions of care were or the condition of the babies was at the time of the birth (i min Apgar), whether they were successfully resuscitated or not (10-minute Apgar), or (the biggest “if” of all) whether the data itself represents an ‘artifact’ — i.e. something that looks like a fact, but isn’t actually a true.

In these cases, artifacts arrise accidentally from a form of statistical data collection in which the collectors of the data (CDC) have no direct control over the data being collected.

For example, the staff at the CDC  cannot tell whether the information about Agars scoring is being provided direct by the birth attendant him or herself, or filled in by nurses in the hours after the birth or typed in correctly (or left blank) by the hospital clerk who is in charge of filling out all the paperwork to register all births that occur at that hospital some days after the actual birth.

The CDC merely reports the data sent to it,  good, bad and indifferent, a fact the CDC is itself the first to point out.

So it is very possible that mistakes are frequently made the reporting of zero Apgars @ 5-minutes. This may be the result of  different criteria or different circumstances between the two major groups (hospital and OOH). It also may represent under-reporting of the critical 5-minute piece of data by the hospital group, or some unrecognized situation that accounts for OOH midwives mistakenly reporting this one particular critical data point.

One unanswered question is about the resuscitation that would be going on for such a baby in a hospital environment. If CPR is being performed and the NICU resuscitation team is proving chest compression of 60 bpm, would they report the 5-minute Apgar to be zero or would they give the baby a 1 for heart rate under 100, which is typically what would be the case for a baby being resuscitated?

Since I’ve already posted a lengthy critique of what the media and medical publication call ‘alarming’ data,  I won’t  include any additional  information on this study by Grunebaum and Chervenak.

Instead I want to address the cumulative effect of  these frequent and repetitive publishing of inflammatory journal papers and articles in popular magazines and on-line websites such as SLATE.com. It seems that if you say it enough times — PHB and non-nurse midwifery are both orders-of-magnitudes more dangerous — it will have the desired negative effect.

Please go on to part 2:

Negative reports about community midwives and PHB repeatedly promoted in the media via press conferences, press-releases and other aspects of a PR campaign

{ 0 comments… add one now }

Leave a Comment

Previous post:

Next post: