Childbirth, risk and the illusion of control

Garrison Keillor famously explained that all the children in Lake Woebegone are above average. Maybe homebirth and natural childbirth advocates come from Lake Woebegone. They appear to believe that when in comes to childbirth complications, all of them are at below average risk.

Natural childbirth and homebirth advocates like to tell themselves and each other to “trust birth.” They consider it a practical strategy, and appear not to realize that it is a common cognitive error known as “optimistic bias.” Simply put, individuals tend to underestimate their risk of various bad outcomes, regardless of known statistics and often in the face of known risk factors. For example, many smokers rationalize their smoking by estimating that their personal risk of getting lung cancer is much lower than it really is.

The paper Unrealistic Optimism About Susceptibility to Health Problems: Conclusions from a Community-Wide Sample, written by Neil Weinstein, was published in 1986. It has been cited by no fewer than 962 other scientific papers because its descriptions and insights are so important. The article explores different reasons for optimistic bias. Although the paper is not specifically about childbirth, its descriptions and conclusions seem particularly apt in considering the current emphasis among childbirth advocates on “trusting birth.”

There are two sources of optimistic bias that seemly particularly apt in describing the outlook of natural childbirth and homebirth advocates. The first is a cognitive error that is often made in connection with uncommon events:

… Any factor that makes us think our own risk is low could lead us to claim that we are below average in risk if we fail to recognize that the same factor may apply to others as well. From this point of view, hazards rated low in frequency could lead to optimistic biases in comparative risk judgments because we forget that the hazards are just as unlikely to strike our peers. Similarly, lack of experience with a problem may make it difficult to imagine how it might affect us and lead us to claim that our own risk is below average….

When considering the chance of being affected by an uncommon event, NCB and homebirth advocates believe that however uncommon the even is, it is even less likely to happen to them than the average person. They fail to realize that they are average persons.


… [P]eople use their past experience to predict their future vulnerability. For many hazards, people seem to hold the mistaken belief that if they have not yet experienced the problem, they are exempt from future risk (absent/exempt)… [P]eople may believe that vulnerability is a constitutional matter, so if the problem has not appeared, their bodies must be resistant. Furthermore, some problems may seem to be caused by one’s behavior or personality (e.g., obesity and drug addition), and people may conclude that the absence of a problem at their age means that they do not have the weakness of character that allows it to develop.

This strategy is also favored by NCB and homebirth advocates. They never had a health problem before, so they assume they will not develop a problem during childbirth. They had a successful homebirth in the past, so they assume all future homebirths will be successful. They “eat right” and stay “informed” so they will not be prey to childbirth complications.

Weinstein conducted interviews with almost 300 individuals, evenly divided between men and women. The participants were asked to evaluate their risk of developing each of 32 different conditions. The most remarkable finding is the most basic. Participants judged themselves to be less likely than average to develop virtually every single condition or disease (31 out of 32).

Although respondents believed themselves to be at less risk than average for virtually every event ranging from developing asthma, to breaking a bone, to being mugged, there were difference in how much less they thought they faced.

Consistent with predictions and with previous studies, optimistic biases increase with perceived preventability, with perceived embarrassment, and most strongly, with the belief that one is exempt from risk if the problem has not yet appeared…


Optimistic biases decrease with experience, perceived frequency, and the perceived extent of others’ worry. Optimism was unrelated to perceived seriousness.

In other words, respondents judge their risk of experiencing a bad outcome as markedly lower than average if they thought that they could prevent the outcome, if the outcome was embarrassing, or if they had no personal experience of the outcome. That sounds like an excellent description of NCB and homebirth advocates’ approach to the inherent risks of childbirth. They falsely believe that they have the power to prevent bad outcomes; they will be embarrassed in front of their peers in the NCB or homebirth movement if they have a bad outcome; and they erroneously believe that if they have no personal experience of the bad outcome it isn’t likely to happen to them.

Weinstein found that, amazingly, actual risk factors don’t seem to moderate individuals’ assessment of personal risk:

… Cigarette smoking is a powerful risk factor for many illnesses. It is strongly correlated with subjects’ judgments of lung cancer risk, has a small correlation with risk judgments for heart attack and cancer in general, and has no significant correlation with the perceived risk of stroke. Furthermore, except for frequency of drinking, none of the other behavioral risk factors is correlated with perceived risk. There is no association between auto safety belt use and perceived risk of injury in an auto accident, between excercise or dietary cholesterol and heart disease risk, or between flossing of one’s teeth and vulnerability to gum disease. Even for drinking, the relationship between the behavior and the perceived risk is optimistically skewed. What one finds is that 77% of the people who never drink place themselves in the lowest comparative risk category, but of those who drink alcoholic beverages four or more times a week, only 18% place themselves in any of the above average risk categories. In other words people give themselves credit when they do not drink but do not acknowledge that frequent drinking may place them in a higher than average risk category.

NCB and homebirth advocates practice the same strategy of denial. Twins, breech and VBAC are well known risk factors for bad outcome yet NCB and homebirth advocates (including homebirth midwives) insist that these are not “high risk” situations.

Weinstein concludes:

… The principal characteristics determining the amount of bias elicited by the 69 different hazards that have been considered were (1) the belief that if the problem has not yet appeared, one is exempt from future risk; (2) the perception that the problem is preventable by individual action; (3) the perception that the hazard is infrequent; and (4) lack of experience with the hazard. If a hazard had such characteristics, people had a strong tendency to conclude that their own risk was less than the risk faced by their peers…

Not only do these factors account for individuals’ underestimation of their personal risk, I suspect that the same factors account in large part for the difference in outlook between obstetricians on the one hand and NCB and homebirth advocates on the other. Doctors KNOW that just because a problem has not developed in the past does not mean it can’t happen in the future; doctors KNOW that most pregnancy complications have nothing to do with individual action; doctors KNOW that infrequent events do happen; and, perhaps most importantly, doctors have direct personal experience of these complications occurring in low risk women.

NCB and homebirth advocates like to pretend that doctors “pathologize” birth. The reality is exactly the opposite. NCB and homebirth advocates are in denial about the risk of developing childbirth complications. They trumpet their philosophy of “trusting birth” without having any idea that it marks them as hopelessly ignorant of actual risks, in deep denial about their personal risk, and so naive that they believe that if they haven’t heard of it, it can’t happen to them.

  • when we can control the weather, we will be able to control childbirth. so far neither of those things has happened.