Jill has had a great series of posts on The Unnecesarean about a large malpractice payment based on jury findings of a lack of appropriate informed consent prior to a shoulder dystocia that resulted in a permanently affected child, and a follow up post on how to properly counsel a patient with a baby with a large estimated fetal weight.
As you may have expected, both of these posts had lively comments sections. I only got involved in the second post. Yes, I haven’t just been MIA here, I have been MIA in the interwebs in general. However, I tend to jump in when I actually get a chance to read something, and then see someone creating straw men arguments or grossly misinflating aspects of the conversation in order to make a point, which happened a few times in the second comment thread.
My first comment was in regards to informed consent. A few commenters acted as if there is an exact formula for informed consent, and it includes presenting every worst case scenario, even if the risks of that scenario are diminishingly rare. Also, some commenters were treating the one verdict and award in this very specific case as the totality of case law on informed consent.
ACOG has a Committee Opinion on Informed Consent which discusses the complicated and amorphous subject of malpractice case law and informed consent. This opinion, which I recommend that you read, like I recommend you read all of the links I am including (yeah, I know, you haven’t got all day, but still), states that first of all, informed consent is more of an ethical issue than merely a legal issue. Secondly, the adequacy of disclosure, which is the issue that the huge malpractice payout in the original Unnecesarean post hinged on, has been judged by different criteria in different cases. In recent history, “common practice of the profession” was the most common trend for judgments. That could be troublesome, because standard of care in different areas can be quite variable, and not necessarily evidence based or best for the patient. Now, the trend seems to be moving towards the “reasonable person” criterion, which can also be troublesome. Especially if the all the commenters on these threads are “reasonable people.” Ahem.
Physicians are notoriously poor at presenting risk (pdf) in a way a reasonable person can understand. Many practitioners will very selectively and erratically present risk, sometimes exaggerating, downplaying or completely omitting risks or benefits in order to lead the patient in a certain direction. Ignoring that, statistics are still highly complicated even with the best of intentions. This article recommends using “natural frequencies”, such as saying three to five people out of ten taking Prozac will report some sexual dysfunction, as opposed to saying there is a 30 to 50% probability of sexual dysfunction. Many people will assume the latter will mean that every time they have a sexual encounter, there will be a 30 to 50% chance of there being a problem. Percentages or other comparative methods (__ times more likely) can be tricky.
For example, in Liu et al’s Maternal mortality and severe morbidity associated with low-risk planned cesarean delivery versus planned vaginal delivery at term, one of the outcomes measured was any hysterectomy. I picked this specific outcome because it is a good example of how to discuss the numbers, but also because one of the commenters grossly misrepresented this particular risk, stating it was ONLY a risk of vaginal birth, and not at all associated with cesarean section.
In this retrospective study, there were 27 hysterectomies in 46,766 cesarean deliveries, and 376 hysterectomies in 2,292,420 vaginal deliveries. That is the same as 0.6 per thousand cesareans, and 0.2 per thousand vaginal deliveries. The adjusted odds ratio of any hysterectomy is 3.2 higher odds for cesarean than vaginal delivery. So, three times higher, or 320% higher. Sounds huge, right? But, the absolute risk difference is 0.4 per 1,000. Or, four hysterectomies per 10,000 cesareans. Does increased risk of hysterectomy need to be part of the informed consent for cesarean section? Does it need to be part of the informed consent for vaginal delivery? How frequent does an adverse event need to be for it to deserve a mention? Does an adverse outcome such as nerve injury resulting in foot drop, usually due to epidural or spinal anesthesia, which only appears in isolated case reports, not even in large studies such as this, need to be mentioned?
I still haven’t touched on the topic of how to counsel a patient who is near term and has high estimated fetal weight. This is a complicated topic, and I don’t think I am going to cover it in this post. Based on the evidence, including the ACOG position statements on the topic and UpToDate’s review of the literature, shoulder dystocia is unpredictable and unpreventable. Prophylactic cesarean section does not prevent nerve injuries or neonatal death. Induction of labor (which is disappointingly common in these cases) actually increases neonatal poor outcomes. Instrumental vaginal delivery (use of vacuum extraction or forceps) increases the risk of shoulder dystocia. Estimated fetal weight is a tool with poor accuracy, given a rating of I for insufficient evidence to support its use by the United States Preventative Services Task Force. This list does not even take into account the maternal history and characteristics.
I think informed consent for any pregnant person should include the chance of a shoulder dystocia. I think as the risk factors increase (estimated fetal weight greater than 4500 g, gestational diabetes, prior macrosomic baby, prior shoulder dystocia, male fetal gender, small maternal pelvic size), that increased risk should be presented. If a practitioner is acting out of fear of a lawsuit in the extremely rare case that there is a very poor outcome, the practitioner should mention this fear.