Monday, October 27, 2008

Anti-Smoking Bias Evident Not Only in Research, But in Policy Analysis As Well

New Study Claims to Find Support for Ban on Smoking in Homes with Children, But Conclusion was Pre-Determined

A new study published online ahead of print in the American Journal of Public Health purports to provide an ethical analysis of policies that ban smoking in homes in which children are present. Based on an analysis of the implications of such a policy for the principles of autonomy and nonmaleficence, the authors conclude that smoke-free home policies are justified in order to protect children from harm (see: Jarvie JA, Malone RE. Children’s secondhand smoke exposure in private homes and cars: an ethical analysis. Am J Public Health, December 2008).

In analyzing smoking bans in homes with children from the perspectives of autonomy and nonmaleficence, the authors conclude: “Restricting smoking by adults in cars or homes in which children are present is the most ethically justifiable position, minimizing paternalism, while respecting autonomy and emphasizing nonmaleficence toward children.”

However, in their analysis, the authors acknowledge the following: “lack of knowledge, lack of resources, and economic or psychosocial burdens may challenge the definition of many adult behaviors, including smoking, as maleficent or nonbeneficent. For example, a single parent addicted to nicotine and in early recovery from heroin addiction who lives in public housing with smoke-free common areas, located in a neighborhood with high rates of violent crime, may decide that smoking inside with the window open is safer for her young children than risking the trek to find a place to smoke outside. This journey might expose her family to neglect or violence, or expose the mother to drugs during her still-tenuous recovery. In some cases, therefore, it may be argued that the immediacy of addressing other injurious factors, including the context within which adult smoking occurs, may take precedence.”

Nevertheless, the authors conclude: “Analogous to the process of regulating public secondhand smoke, policy for private domains may serve most effectively to enhance changes in social norms. Creating a norm of unacceptability for childhood secondhand smoke exposure poses no threat to adult autonomy.”

The Rest of the Story

To the careful reader, this paper is an example of the justification of a pre-determined conclusion regarding a policy via a purportedly objective policy analysis. This story demonstrates that a strong bias on the part of anti-smoking researchers is present not only in research, but in policy analysis as well.

Here, the authors have apparently reached a pre-determined conclusion that smoke-free home policies are justified to protect children and attempted, post-hoc, to supply some sort of ethical analysis that would justify this policy. Their attempt to do so is weakly disguised, however, as it is easily apparent that the authors are not actually allowing their opinions to be informed by the results of their own analysis.

Most notably, the authors established – from their analysis – two key points in the paper:

First, the authors established that banning smoking in homes with children present does not necessarily represent a policy of nonmaleficence, because there are situations in which the smoking parent may and family may be subject to greater risks, such as that from exposure to drugs, neglect, or violence. The authors present precisely such a situation: in this situation, the authors themselves argue that choosing to smoke inside the home with the window open might be an option associated with less risk to the family than observing the law and going outside to smoke.

Second, the authors established that banning smoking in homes with children present does not protect individual autonomy and in fact may interfere with the ability of a parent to make a wise decision that minimizes overall risk for the family. The authors actually present an example which illustrates a situation where a parent would be subjecting the family to greater risk by following a policy of not smoking in the home.

However, in both cases, the authors ignore their own arguments and present a conclusion which is inconsistent with the arguments that they themselves advance. First, they conclude that “Restricting smoking by adults in cars or homes in which children are present is the most ethically justifiable position, minimizing paternalism, while respecting autonomy and emphasizing nonmaleficence toward children.” But as they have themselves shown, such a policy may neither emphasize nonmaleficence nor respect autonomy.

Second, they conclude that banning smoking in homes with children “poses no threat to adult autonomy.” From their own analysis, such a policy clearly does pose a threat to adult autonomy: namely, it removes adult autonomy over decisions about risk management for their children, themselves, and their families.

While I am not here arguing the merits or lack thereof for these home smoking bans, what I am pointing out is the bias apparent in this policy analysis. The conclusion of the piece is inconsistent with the findings of the analysis, suggesting that the authors came to a pre-determined conclusion and tried to create a post-hoc justification, rather than conducting the analysis and then basing their conclusion on the results.

While not as important as my conclusion that there is a severe bias present in the article, it is still worth noting that the piece also fails to address the fundamental question of how one can justify banning smoking in homes with children present, but not also justify policies that would regulate the foods that parents feed their children, the activities that parents allow their children to participate in, and a host of other risks to which parents expose their children. Until and unless the researchers can explain why regulating smoking is justified, but regulating these other risks is not, then they have not succeeded in providing the policy analysis that would be necessary to support the enactment of such policies.

No comments: