[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.146.176.35. Please contact the publisher to request reinstatement.
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Medical News and Perspectives
October 3, 2012

Clinicians Remain Reluctant to Allow Negative Findings to Influence Practice

JAMA. 2012;308(13):1305-1306. doi:10.1001/2012.jama.12314

Results from a recent review of randomized controlled trials suggest that medically treating mild hypertension in individuals with no previous cardiovascular events or cardiovascular disease does not reduce mortality or morbidity. But, as so often happens with negative findings, it remains to be seen whether this evidence will change clinical practice.

The review, by the Cochrane Collaboration, is at odds with current guidelines in the United States and Europe that call for medical therapy if lifestyle changes cannot control an individual's blood pressure. The Cochrane conclusion follows recent actions by the US Preventive Services Task Force (USPSTF), which has questioned whether screening men for prostate cancer, screening women younger than 50 years for breast cancer, and using electrocardiography to predict coronary heart disease events in low-risk asymptomatic adults provide sufficient potential benefits to offset potential harms. In turn, the Cochrane findings and the USPSTF recommendations have met resistance from the professional societies that have published guidelines promoting these practices and the scientists whose research underlies such guidance.

There are many reasons that clinicians have difficulty incorporating new research findings into practice, including information overload and skepticism about new evidence.

Sharon Straus, MD, MSc, who researches knowledge translation, is not surprised by such resistance. “We do have to be aware that sometimes evidence changes over time, and we need to be open to that change. But that is hard to do,” she said.

Straus, of the University of Toronto, Toronto, Canada, said it is particularly difficult to introduce negative study findings into prevention and screening protocols because it is hard to visualize such interventions as having risk components. “People think that if they get screened for something, it will be a good thing, and we don't think that bad things will happen,” Straus said. “And this is what these [negative findings] highlight—the potential for good and harms.”

HYPERTENSION DOUBTS

The Cochrane researchers, performing a meta-analysis of 4 randomized controlled trials with 8912 individuals, found that treatment for 4 to 5 years with antihypertensive drugs as compared with placebo did not reduce total mortality in asymptomatic patients with mild hypertension, defined as systolic blood pressure of 140 to 159 mm Hg, diastolic blood pressure of 90 to 99 mm Hg, or both. Further, based on a meta-analysis of 3 of the 4 randomized controlled trials involving 7080 asymptomatic individuals, the researchers found antihypertensive drugs as compared with placebo did not reduce incidence of coronary heart disease or stroke or total cardiovascular events. In addition, 9% of patients enrolled in these trials discontinued medical treatment because of adverse effects (Diao D et al. Cochrane Database Syst Rev. 2012;8:CD006742).

James M. Wright, MD, PhD, of the University of British Columbia in Vancouver, Canada, and coauthor of the Cochrane review, said his group has been criticized for basing its conclusions on fairly scant evidence—4 randomized controlled trials, fewer than 9000 individuals, and short follow-up time frames. “It is unfortunate, but it's the only data we have,” said Wright, who is also the coordinating editor of the Cochrane Hypertension Review Group. “The next step would be to have some trials going forward to add to the evidence.”

But such trials will probably never be conducted, because they would require large numbers of relatively healthy enrollees being followed up for many years to determine efficacy of treatments, as assessed by hard end points like myocardial infarction and death, said American Heart Association spokesman Ernesto L. Schiffrin, MD, PhD, of McGill University in Montreal, Canada.

Schiffrin says he is comfortable with current US and European guidelines for treating mild hypertension medically. “A population at low risk followed for 5 years may not allow a demonstration of benefit,” said Schiffrin. “And for that reason, people have gone on to use epidemiological evidence. And you come to a moment when you consider the risk of hypertension is increased to a degree that, even though you cannot prove it in a clinical trial, a medical intervention will produce a benefit.”

Wright is not as trusting of epidemiological evidence. “I have a high blood pressure clinic, and I’ve already changed my practice,” Wright said. “When I see someone with mild hypertension, I inform them of what the evidence shows, and some will say, ‘That's fine, then I don't want to take something that's not proven.’ Others I see may already be on the drugs and doing well, and they continue on them.”

BARRIERS TO CHANGE

Ethan Basch, MD, MSc, a member of the methodology committee of the Patient-Centered Outcomes Research Institute, said incorporating new findings into clinical practice is always difficult. He noted several barriers, including the tendency for physicians to become entrenched in a way of practicing medicine and skeptical of new evidence; improper dissemination of new findings, which fail to reach the decision tools physicians use; confusion that emerges over ambiguous study results, which prompts physicians to fall back on past practices; and information overload, which makes it difficult to keep up with nuanced changes.

Even with these barriers, Basch believes new information can ultimately be melded into practice. “I think we as scientists become invested in the direction of our own research and become blinded to the evidence that contradicts what we are doing,” said Basch, who is also director of cancer outcomes research at the University of North Carolina in Chapel Hill. “But I also think when the science becomes compelling enough, scientists will be swayed.”

×