Are you a good doctor? Is your complication rate higher or lower than others? Are you competent to do procedures? Will someone ever tell you that you should not be doing something that you have been trained to do? Will patients compare you to others in your community? Are the data that are now available online for all to see going to affect your practice? Will your reimbursement be based on some measurement of the quality of care? Is the measurement tool valid? Are the data valid?

All of these are questions that plague us every day. As a residency program director I was required to sign a piece of paper that verified that graduates from my training program were competent to be “independent practitioners.” I wonder if what I was signing really was truthful. As a former department chairman, I was required to evaluate new hires with a focused professional practice evaluation and evaluate my faculty with an ongoing professional practice evaluation. These tools were put into place to identify the competency and cost-effectiveness of my faculty and to help determine that they were providing safe and compassionate patient care. The point of this is that we are being measured every day we practice . More and more, our ability to provide care is being scrutinized by hospital administrators, the government, and our patients. There is no getting away from this: we are now undergoing a period of close scrutiny and being compared to others far more intensely than when I began in medicine.

The article entitled “Impact of surgeon annual volume on short-term maternal outcome in cesarean delivery” by Drukker et al of Israel in this month’s American Journal of Obstetrics and Gynecology is another example of quantifying medicine and evaluating competency. This manuscript indicates that obstetrical providers who perform a low volume of cesarean deliveries (lowest quartile of the number of cesarean deliveries) have the highest complication rates. Many may interpret these data and think that the competency of these “low-volume” providers is inadequate and they should not be providing cesarean deliveries as part of their obstetrical care. If this information were taken at face value, many providers would not be allowed to perform cesarean deliveries. This could lead to many hospitals not being able to provide obstetrical care and create an access problem to intrapartum care in many regions of the country. Providers in larger hospitals who have limited practices or who are slowing down but have years of experience may also be told that they are not qualified to perform procedures that they were doing for many years. These providers may be denied privileges at the hospital or insurers may not reimburse them for the care they provided.

The article concludes that the attending physicians who perform <20 cesarean deliveries a year have the highest composite complication rates. However, the conclusion begs several questions: who really was doing the surgery, the resident or the attending, and is the small increase in severe complications really clinically significant?

We all know that statistically significant findings may not be clinically significant. Four of the 7 outcomes variables used in this study had no statistically significantly difference between the 2 provider groups. This makes the reader wonder why the composite adverse maternal outcome is really significant. The most significant difference identified between the low-volume and high-volume group was urinary and gastrointestinal tract injuries. This occurred in only 4 more cases of 1000 cesarean deliveries (0.6-0.2%). Is this enough to tell someone he or she should not be doing cesarean deliveries anymore? The other statistically significant finding, a drop in hemoglobin of >3 g/dL, did not result in an increase in blood transfusions. Do these statistically significant differences represent clinically significant findings, findings that might lead to suspension of clinical privileges or denial of payment?

The finding of prolonged maternal hospitalization of ≥10 days seems more clinically relevant but the reasons for delayed discharge are never discussed, such as genitourinary or gastrointestinal injuries, and low hemoglobin. Obviously the secondary outcomes of statistically significant skin-to-delivery and skin-to-skin times may be statistically significant, but they are not clinically significant and should not in any way be used to judge quality of care of any individual provider. This is a great example of statistics that do not have any bearing on clinical care being used to justify a conclusion.

There are many reasons to question whether this article should be used to guide care.

• 1.
  

  Body mass index was not included in the demographic information evaluated. Many papers have been published that indicate that body mass index is a significant contributory factor to surgical morbidity and mortality.

  
  
• 2.
  

  A baseline cesarean delivery rate of 11% is remarkable and makes one wonder if the data are applicable to current practice in the United States.

  
  
• 3.
  

  There is no clear understanding of who the actual surgeon was since either a resident or an attending may have been the primary surgeon.

Despite concerns about the applicability of the findings, this article has the potential of having significant impact on care, both positively and negatively. It could result in improving the quality of care by channeling patients to providers who perform cesarean deliveries more frequently with less complications or it could create a barrier to access to care and prevent patients from receiving care close to their homes or by physicians with whom the patient has a long-standing relationship. For good or bad, this type of information will be utilized in the future. Data will be used to determine how care will be provided and who will provide that care in all aspects of medicine. This is not new.

We have seen this with the advent of hospitalists, laborists, minimally invasive surgeons, and urogynecologists. The concept that those who do things routinely get better results is now well accepted throughout medicine. Maybe “laborists” will not become “cesarean sectionists.” If data from a study like this were to be used to guide who performs cesarean deliveries, the most common operation in the United States, we must be sure that the statistical significance of what is found has real clinical value. I do not believe this article can be said to do that but I do think it is a harbinger of the future. Stay tuned!