Crunch time
Vicki Conn
You are a doctor or nurse with a clinical question. Obviously, a good place to look for the answer is in the many scientific journals related to the topic of your interest. So off to the medical library you go. And the answer is... confusing!
There are hundreds of journals that publish thousands of articles on a huge array of topics. Even if you successfully narrow the field to publications or articles related to your question, you find dozens of studies that offer conflicting results. Which ones do you trust?
That’s another problem because determining which studies are best requires careful review of the methodology, sample and statistical techniques. It would take days to figure out which results are worth heeding.
Thank goodness there’s a meta-analysis on your topic! The researchers who use this technique do the work for you. Meta-analysis involves combing through the tangled web of research and creating a summary and conclusion regarding the current wisdom on any given subject.
If you wonder just how important meta-analysis is, look at Professor Vicki Conn’s study, “Chronic Illness Exercise Interventions: A Meta-Analysis.” The National Institute of Nursing Research thought this meta-analysis important enough to award a five-year, almost $900,000 grant.
Conn is continuing her pursuit of exercise interventions meta-analysis with a $1.7 million-dollar NIH grant, which was awarded in early 2006.
“There’s no other way to accurately summarize results across studies,” Conn says. So Conn and a team of researchers are beginning the long and arduous process of locating and examining every study that’s been published about exercise and the chronically ill. By conducting a meta-analysis, Conn hopes that trends will emerge, making it easier to formulate definitive conclusions about the best types of exercise programs for this population.
Among the chronically ill, exercise may be particularly important for managing symptoms and disease progression. For instance, diabetics can control their glucose levels better if they regularly exercise, and people with heart disease decrease their risk of life threatening episodes by staying in shape.
Despite the staggering amount of existing research, Conn notes that even summaries of studies focus on a single population, such as people with arthritis or diabetics. “There is no summary of the literature across chronic illnesses,” she says. “And there’s no reason to believe that it takes different motivation for these groups to exercise.
“There are so many individual studies out there, and it looks like the findings are contradictory. One study looks at self-monitoring, one adds walking with a friend, another adds a reward for exercising. How do you compare these different interventions? We want to summarize across all the studies more accurately,” she says.
Conn’s meta-analysis of exercise and the elderly was published in a 2002 Annals of Behavioral Medicine journal, found that people age 60 and older are more likely to exercise if they join a gym or commit to classes at a senior center rather than being left to their own devices at home. In doing her initial study, Conn noticed that some of the research in her meta-analysis dealt with elderly people who had chronic illnesses and that interventions were more effective if they were delivered to these very specific patient populations. “I saw a window of opportunity,” she says. “That helped lead to the analysis on this population.”
Conn is fortunate to have help from one of the world’s top methodological experts in meta-analysis, Adam Hafdahl. As one of the country’s foremost expert in the technique, Hafdahl will participate in Conn’s study by performing the complex statistical calculations involved in determining the final outcomes.
The bulk of Conn’s time on the project will be spent in tracking the thousands of studies in medical and nursing literature. With a team of research assistants, she begins by sifting through the computerized databases that have become the first stop for most academics and clinicians who need to conduct a typical literature search.
After checking these high-profile sources, Conn and her team become detectives hunting down what she refers to as “fugitive literature.” At this point, the team will delve into the murky world of unpublished research and minor journals.
It may not be as thrilling and dangerous as a James Bond chase, but for every article scanned by Conn and her team, many more leads are uncovered in the form of citations that must be found and analyzed. They also search through conference presentation abstracts, usually contacting the authors to make more connections.
“We really have to be organized,” Conn admits. “Ninety percent of our time is spent seeking out the studies that need to be included.” This translates into at least four years of dedicated work by Conn and many graduate students who want a taste of this type of research process.
Once Conn and her colleagues have completed the statistical analysis of primary findings and calculated the exact numbers associated with intervention effects, they will have a list of interventions that have been proven the most effective for encouraging exercise. “This will be very powerful in terms of influencing practice,” she says. For instance, public health departments may refer to Conn’s conclusions in determining what types of cost-effective and persuasive programs to offer the chronically ill.
You may think this type of demanding, well-funded study would engulf a researcher. But Conn simultaneously completed another project she had been working on since 1997. “Nursing Interventions to Enhance Older Women’s Exercise” ended its five-year, $652,000 National Institute of Nursing Research grant in 2002. Findings suggest activity professionals should incorporate self-monitoring, intense contact between themselves and participants. Groups of individuals with particular chronic illnesses are most likely to be successful.
Looking back on the study, Conn says with satisfaction, “We enrolled 200 women, and our intervention worked!”
In this study, Conn tested a brief, motivational strategy versus a prompting intervention that she developed. The first strategy involved a brief contact with the subjects, but Conn found that the limited contact was not strong or long enough to result in an increase in exercise. On the other hand, her intervention, which used phone calls from the study team to inquire about how well subjects’ exercise programs were working, proved successful in increasing and maintaining exercise among older women.
Besides requiring subjects, who were an average age of 75, to report on their exercise activity, Conn issued each subject a wearable device that kept track of how many steps the woman had taken each day for a week. By the end of the study period, Conn found that her prompting intervention group had taken more steps than when they began the program and were reporting intentional exercise routines as part of their regular schedule. In fact, the results show that the prompting group increased their exercise activity three times more than the single-contact group.
Conn hopes senior centers, retirement villages and nursing homes will take note. Using her prompting program, older women could be successful in reaping the benefits of regular exercise.
Conn is dedicated to the impressive improvements to quality of life for the population that exercise equipment and gym marketers largely ignore. Through all her research projects, Conn says with a laugh that “my purpose is to get people who don’t look too good in spandex to exercise anyway.”