Islamic American doctors and premature babies.

Two daunting methodological difficulties arise when we try to study the influence of doctors' religious beliefs on their clinical practices. First, religious beliefs are a private matter. We don't know what goes on in people's hearts. Surveys can tell us what people self-report about whether they follow certain rituals, observe holidays, pray, go to church or believe in God. But we cannot know whether people are telling the truth. Nor can we know what these practices mean to those who do them. They likely mean very different things to different people. The second methodological difficulty relates not to people's reported religious beliefs but, instead, to their reported clinical practices. Doctors' responses to survey questions about hypothetical situations may or may not reflect what they actually do in the experiential reality of complex clinical situations. This is especially true if the hypothetical questions are about clinical situations that they rarely experience in their own professional life. Nevertheless, there are plausible correlations between doctors' self-reported clinical practices and their self-reported levels of religiosity. Curlin and colleagues have shown that doctors who consider themselves to be very religious are less likely than less religious doctors to report that they would refer pregnant women for abortions, prescribe birth control for unmarried women or provide terminal sedation for dying patients 1. In this issue, Arzuaga et al. report the results of a survey of a group of Muslim physicians in the United States regarding their attitudes about resuscitation of ‘periviable’ infants 2. The group of Muslim physicians was identified through their membership in a religious organisation. Thus, the respondents are likely to be more active and engaged in the American Muslim community than are other Muslim physicians in the United States. Thus, the study must be interpreted with a third caveat, in addition to those noted above, regarding the generalisability of the results. The study addresses the controversial issue of whether these physicians' religious beliefs affect their clinical practice. The results are not straightforward. They showed that higher degrees of self-reported religiosity were associated with a higher likelihood of ‘endorsing resuscitation’ for babies born at 22 and 23 weeks’ gestation. Overall, however, the reported rates of ‘endorsing resuscitation’ are similar to the actual practices of doctors at many leading academic medical centres 3. So are these doctors' clinical practice shaped by their religious beliefs? Or by American culture? Muslim society in the United States (as well as in the rest of the world) is extremely diverse 4. Muslims come from different countries and cultures, belong to different branches of the religion and different sects within those branches and may or may not have studied Muslim theology. In this sense, they are no different than Christians, Jews or Hindus. The labels describe complex and diverse groups of people. Muslim Americans are not uniformly religious and devout. Even among the more religiously devout, there are some sharp distinctions between being a good Muslim and being an Islamic extremist. As with people of other religions or nationalities, many Muslim Americans emigrated from war-torn or autocratic countries in the Middle East in order to practice – or not practice – their religion and politics more freely in the United States. The authors suggest that Muslims are more likely than doctors from other religious groups to resuscitate babies over parental objections. That may be true. However, it may only be that both groups – Muslims and doctors from other religious groups – give answers on surveys that are consistent with their deeply held philosophical, political or religious beliefs. The study did not clearly explain why an American Muslim Physician would choose to resuscitate a non-viable infant based solely on their belief. In principle, the Qur'an, which is the ultimate reference for Muslims, condemns killing of a living soul and does encourage preservation of life: ‘And do not kill any one whom Allah has forbidden, except for a just cause’ (The Qur'an, Al-Isra’.17:33). However, there is no clear definition of viability in Qur'an and there are multiple references to foetal development that one can interrupt in varying ways: ‘And they ask you about the soul. Say: The soul is one of the commands of my Lord, and you are not given aught of knowledge but a little’ (Al Qur'an, Al-Isra’. 17:85). These varying views reflect the flexibility of Islam. There is no single right answer to complex dilemmas. Religious beliefs and practices evolve over time. Of note, the study did not ask physicians the reasons for their choices or refer to Qur'an. Religion is clearly an important factor to analyse in understanding the choices that doctors and patients make. Because it is so important, studies must be detailed, rigorous and replicable. This study is a first step. Overall, the authors are to be applauded for exploring the complex relationships between religious beliefs and clinical practices. We hope that this study is just a first step and that future research will examine (i) the relationship between hypothetical responses to a survey and actual practices; (ii) the basis for decisions regarding resuscitation and non-resuscitation; (iii) more precise comparisons between people of different religions who have varying self-reported levels of religiosity; and (iv) the impact of acculturation on the practices of doctors who have emigrated from their homelands to another country. Authors have no financial disclosures. There is no conflict of interest of the authors to disclose.