Thursday, July 19, 2012

New Study Violates Ethical Principles by Asking Parents to Expose their Children to Secondhand Smoke; Study Should Not Have Been Approved or Published

A new study published in the July issue of Tobacco Control reports the results of a randomized controlled trial designed to study the relationship between secondhand smoke exposure and sleep bruxism in children. Sleep bruxism is a condition in which individuals grind or clench their teeth during sleep, which can lead to jaw muscle fatigue, pain, and tooth wear. The objective of the study was to determine whether secondhand smoke exposure is a cause of sleep bruxism in children.

The study subjects were 498 white children, recruited from schools in Naples, Italy. Children were examined for sleep bruxism and parents were asked to report how often they smoked in the presence of the child. Children who exhibited sleep bruxism and were exposed to secondhand smoke were then randomized into two groups.

According to the study:

"The smoking members of the families in group 1 were asked not to smoke in the presence of the child for a period of 6 months, while those in group 2 were asked not to change their smoking habits."

After 6 months, 38% of the children in group 1 exhibited sleep bruxism, while 90% of the children in group 2 still exhibited sleep bruxism. Based on these results, the paper concludes that secondhand smoke exposure is a cause of sleep bruxism in children.

According to the article, the study was approved by the Ethics Committee of the Second University of Naples.

The Rest of the Story

This study violates basic ethical principles of research conduct because it exposes children to unreasonable and unnecessary risks, intentionally encourages parents to put their children at risk, and fails to incorporate alternative methods that would reduce these risks.

According to the Helsinki declaration: " In medical research involving human subjects, the well-being of the individual research subject must take precedence over all other interests."

Furthermore, the Helsinki declaration states that:

"The benefits, risks, burdens and effectiveness of a new intervention must be tested against those of the best current proven intervention, except in the following circumstances:
  • The use of placebo, or no treatment, is acceptable in studies where no current proven intervention exists; or
  • Where for compelling and scientifically sound methodological reasons the use of placebo is necessary to determine the efficacy or safety of an intervention and the patients who receive placebo or no treatment will not be subject to any risk of serious or irreversible harm. Extreme care must be taken to avoid abuse of this option."
In the present study, children in group 1 were provided with an intervention to reduce their exposure to secondhand smoke: parents were asked not to smoke in the presence of their children. Children in group 2 were not provided with any treatment. According to the Helsinki declaration, this would have been acceptable only if: (1) there was no treatment available; or (2) children receiving no treatment would not be subject to any risk of serious harm. Neither of these conditions are met.

Furthermore, this study goes beyond simply providing no treatment to the children in group 2. Instead of simply observing these children over time, the study protocol called for "asking" the parents not to change their smoking habits.

In other words, the investigators knowingly and intentionally placed the children in group 2 at significant risk of health damage. The paper itself acknowledges that: "Exposure to secondhand smoke (SHS) represents a serious public health threat, remaining a preventable cause of morbidity among children. Adverse health effects have been found in children exposed to SHS, including increased risk of pneumonia, bronchitis, respiratory illness, wheezing, middle ear effusions, otitis media and cardiovascular damage."

Thus, the children in group 2 were intentionally and knowingly put at increased risk of pneumonia, bronchitis, respiratory illness, wheezing, middle ear effusions, otitis media and cardiovascular damage for the purpose of having a group in which secondhand smoke exposure was persistent for six months, to allow the investigators to compare the effects of removing secondhand smoke exposure with the effects of continuing that exposure.

In other words, the children in group 2 were placed at significant risk. This is not acceptable.

In the United States, the federal regulations on the protection of human subjects (section 46.406a) would have allowed this research to be conducted on minors only if: "The risk represents a minor increase over minimal risk." This condition is clearly not met, as the risks of pneumonia, bronchitis, respiratory illness, and cardiovascular damage are a major increase over minimal risk.

Moreover, there was an alternative procedure available that would have provided the same scientific knowledge without putting children at substantially more than minimal risk. The investigators could have encouraged all parents not to smoke around their children and then followed all the children over time as a single group. Then, they could have assessed changes in secondhand smoke exposure and related those changes (or lack of changes) to the persistence or disappearance of sleep bruxism.

It is important for me to emphasize that there is no justification for the investigators asking parents to continue to smoke in the presence of their children. 

Because this research violates the Helsinki declaration and because it also violates the U.S. federal regulations on the protection of human subjects, I do not believe that the journal should have published the paper. The journal, as a policy, will only publish research that conforms with the Helsinki declaration. This policy appears to have been violated in this case.

A commentary published simultaneously in the July issue of Tobacco Control expresses similar concerns about this study's violation of ethical principles, arguing that the study "contravenes the Helsinki declaration."

In the commentary, Barrientos-Gutierrez et al. state that: "Preventing family members from stopping smoking and actively seeking children to be exposed to a dangerous pollutant is reckless, as the benefits of knowing if SHS is a risk factor for bruxism cannot overcome the dangers of exposing children to SHS or the benefits of quitting."

They conclude: "Considering children's high vulnerability and inherent risks from SHS, we cannot think of a single research question that could justify actively seeking exposing children to SHS or preventing parents from stopping smoking to provide a healthier environment for themselves and their families."

In the same issue, the authors of the study respond to this commentary. Rather than acknowledge that they made a mistake, they instead defend the study. But in defending the study, they deliver a definitive knock-out blow to their argument that the study was ethical. They point out that after being randomized to group 2, a number of the families dropped out of the study, refusing to participate because: "aware of the risks of SHS, [they] decided to reduce it and therefore did not participate."

If the subjects themselves realized that they were being put at undue risk by agreeing to participate in the study, then it is quite clear that this research was unethical. In fact, after a number of families refused to participate because of the risks to which they were being asked to be exposed, the IRB should have been informed and the study should have been halted.

The authors go on to justify their research by noting that "all of the parents of group 2 remaining in the trial were those who reported not being able to reduce children's exposure to SHS." This is an unacceptable argument. A feeling that one is not able to take a particular action is no justification for investigators putting the children of those subjects at risk and failing to deliver any intervention to encourage those parents to quit, reduce their smoking, or not smoke in the presence of their children.

Let me provide an analogy to show how unacceptable these arguments are. Suppose researchers wanted to determine whether crack cocaine use during pregnancy causes infant health effects. The investigators recruit a group of pregnant crack cocaine users and randomize half of them to stop using crack cocaine. The other half of the subjects are randomized to a group that is instructed not to stop using crack cocaine.

A concerned physician at the hospital writes to the investigators asking how this study can be allowed to proceed. The investigators respond by arguing, first, that there is no ethical problem because after being randomized to the continued use of crack group, subjects are free to withdraw from the study if they don't want to put their fetuses at risk of harmful health effects.

This argument is ridiculous, because we don't protect human subjects by giving them the option of withdrawing from a study if they don't want to assume health risks that are unreasonable.

Second, the researchers argue that the subjects in the group randomized to continue crack cocaine use were not confident about their ability to stop using. This argument fails because the lack of self-efficacy in a human subject should not be used as an excuse to put that individual at risk. In fact, these subjects are a particularly vulnerable group that needs most to be protected from risk.

While I was disappointed to see that the authors defended their research, the most disappointing part of the story was that the journal stood by its decision to publish the paper. Apparently, the journal felt it was doing a service by publishing commentaries about the ethics of the study and a response from the study authors. But that is not sufficient. If the study was unethical and violated basic ethical principles of ethical conduct, then it should not have been published. Publishing a study in which pregnant women were encouraged to continue smoking crack cocaine would not be justified by the journal publishing commentaries about the ethics of the study in the same issue.

No comments: