Aug 27 2010

Doctor’s Faith and End of Life Decisions

A new study published in the Journal of Medical Ethics reports on a survey of various characteristics of physicians – specialty, ethnicity, and religious faith – and the way they deal with end of life care. According to these results, ethnicity had little to no effect. The specialty of the practitioner has the greatest effect, with hospital-based doctors discussing and using methods that may hasten death in the terminally ill at 10 times the rate of palliative care specialists. But the most widely reported result is that doctors who are agnostic or atheist were twice as likely to use such methods as deeply religious doctors.

As with most such studies, the two types of questions to ask are – are the results reliable, and if they are reliable what do they mean. To the first point, this is a very weak study. First, it is a self-reporting survey. This is the weakest kind of survey, because it is not strictly scientific – there is a huge effect from self-selection bias. The surveys, in other words, may mostly reflect who is willing to answer the survey, which can overwhelm other factors.

In this survey 8,000 doctors were contacted, and less than four thousand responded. That right there is a massive self-selection factor that renders any results of this survey preliminary at best. Such a survey can be used to generate hypotheses to be confirmed, but not something upon which specific recommendations should be based. Regardless, the authors do just that, recommending:

Greater acknowledgement of the relationship of doctors’ values with clinical decision-making is advocated.

Rather the authors should have concluded: Further testing of the possible relationship between specialty, religious belief, and decision-making is warranted.

But let us take as a hypothetical that these results are truly reflective of doctor practice – what would that mean? To me the most interesting result is that specialty had a ten-fold influence on decision-making. This does reflect my anecdotal experience – that doctors who routinely treat the terminally ill in an in-patient setting are more comfortable and practiced in raising treatment questions that could influence the duration of life. Palliative care specialists, on the other hand, are focused on palliation and may not be as comfortable suggesting withdrawal of care or measures that might hasten death.

This reflects a further underlying  truth that there remains many subcultures within medicine that affect practice. Our stated goal is to have a single evidence-based standard of care, that is optimal for both patient outcomes and cost-effectiveness. But in reality there is much individualization of care and this varies from specialty to specialty and even region to region and hospital to hospital. This suggests that we need better systems for establishing, teaching, and enforcing the standard of care where one exists.

That non-religious doctors entertain treatment decisions that may hasten death at twice the rate as religious doctors is a much smaller effect, and much more suspect given the survey design. But again – assuming this is a real effect, what does it mean? The question is – are non-believers suggesting and making decision that are inappropriate, are religious doctors avoiding appropriate questions and decisions because of their personal faith, or are both groups biased in opposite directions away from a more objective stance somewhere in the middle? This survey provides no information to assess this question definitively.

However, it is interesting that religious doctors do not even discuss these decisions with their patients and their families. This suggests they are avoiding treatment decisions that make them feel personally uncomfortable because of their religious faith. If this is true, then that is something worth exploring further.

I admit my personal bias. I am a non-believer, and a neurologist (a specialty named in the survey as more likely to ask and make decisions that might hasten death). To me it seems that the failure to even ask the question of the patient or family is allowing one’s personal faith to influence medical practice. Whereas I don’t see how my lack of faith would have a similar influence. I frequently deal with terminally ill patients, some very acutely ill, and have had many discussions with their families over prognosis and care. Sometimes I am the treating physician, sometimes I am just a consultant. Either way my primary job is just to provide objective information to the family and let them know the full range of options before them, and to understand the implications of possible decisions. In the end my goal is to make the family feel they are comfortable with the decisions they have made, because they were fully informed. My job is to be non-judgmental and supportive, as long as we are within the fairly wide range of ethically acceptable decisions.

I cannot see how withholding discussion of certain options can better serve the patient or the family. It seems to me this would mostly, or only, serve to prolong the inevitable, prolong patient suffering, and would result in needless futile care. By contrast I never withhold from patients options for being more aggressive in treating a seriously ill patient. The survey did not really address this question, and it would have been a good addition. In fact I directly tell patients and families – if we want to be maximally aggressive and do everything we can, these are our options.

Therefore the reporting of this survey is backwards. It does not indicate that “atheist doctors” are more likely to make decisions that hasten death, but rather than deeply religious doctors are more likely to withhold options from patients and families that may reduce suffering and futile care.


As I stated above, this is not a controlled study but a self-reporting survey, and so the results are highly unreliable. At best it indicates that follow up research is warranted. But taken at face value, if anything this survey shows that culture in medicine still plays a large role in determining practice. And further it suggests that some doctors allow their religious faith to interfere with their decision-making when it comes to end-of-life care.

This will become an increasingly important issue as we face tough societal questions about futile care in a health care system we cannot afford.

12 responses so far