CHS Research News Center News

Group Health Center for Health Studies and Kaiser Permanente Northern California recently began enrolling volunteers in one of the nation’s largest and most comprehensive studies of pancreatic cancer.

Unique in its ability to rapidly identify cases following diagnosis, the PACIFIC ("Pancreatic Cancer Investigation: Finding Causes") study will help scientists better understand environmental and genetic factors that may cause cancer of the pancreas. This, in turn, could lead to better screening and prevention of the disease, which strikes 29,000 Americans annually, is the fifth leading cause of cancer death in the United States, and has a five-year survival rate of less than 5 percent. The study is funded by a five-year $6 million grant from the National Cancer Institute.

Working with colleagues from the Fred Hutchinson Cancer Research Center (FHCRC) and the University of Washington (UW), scientists from Group Health and Kaiser will use their organizations’ computerized clinical information systems to identify 745 cases soon after diagnosis and collect information via interviews and blood specimens. Data from these cases will then be compared to information from 1,041 other study participants who do not have pancreatic cancer. The scientists plan to include extra numbers of African Americans in the control group to help determine why this racial group has higher rates of pancreatic cancer than the general public.

"The risk factors for pancreatic cancer are not well known, in part because cases of pancreatic cancer are very difficult to identify quickly enough to enroll in research studies," explains Meg Mandelson, PhD, an associate investigator at Group Health’s Center for Health Studies, associate member of the FHCRC, and a co-principal investigator of the study. "By the time a patient has developed signs and symptoms of pancreatic cancer, the disease is usually already fairly advanced." About half of all patients with pancreatic cancer die within six months of diagnosis.

Located deep in the abdomen behind the stomach and other organs, the pancreas gland is difficult to reach surgically, so there is no single path to diagnose pancreatic cancer. Unlike diseases such as colon cancer, which is nearly always detected from a surgical biopsy, pancreatic cancer may be diagnosed using a combination of imaging (x-rays), blood tests, and surgical biopsies. Historically, this complexity has made it difficult for researchers to identify and reach patients in time to collect data needed for comprehensive evaluation. But new computerized clinical information systems at Group Health and Kaiser are changing this. The researchers in this study conduct daily surveillance of test results in all three areas (radiology, lab, and biopsies) to track nearly all patients with pancreatic cancer in the system soon after diagnosis.

"Group Health and Kaiser Northern California are two of just a handful of research institutions nationwide that have the capacity to do this ‘ultra-rapid case identification’ the study requires," says Mandelson. Once identified, the patient’s physician can give the researchers permission to contact the patient to arrange a 45-minute interview regarding their medical history, family history, medication use, diet, and other environmental factors, and to collect a blood specimen for study.

The researchers are establishing a specimen repository at UW’s Institute for Public Health Genetics and FHCRC to store blood collected in the study so that it can be used over time in studies of the molecular characteristics of pancreatic cancer.

"The study would not be possible without the support and commitment of many people at Group Health and Kaiser—especially their clinical teams in oncology and gastroenterology," Mandelson says.

"We believe that combining data from the study specimen repository with information from patient interviews will allow us to take major steps toward understanding and preventing pancreatic cancer," she adds.

With the cost of prescription drugs rising faster than any other part of health care, more insurers and health systems have begun adopting restrictive policies for coverage of newer drugs. How can consumers and health care organizations be sure that such policies don’t lean too far in either direction—either restricting necessary treatment or wasting consumers’ premiums on expensive medicines that offer no real advantage?

These issues are addressed by experts from Group Health in an article in the October 2005 issue of the Journal of General Internal Medicine. The piece presents five principles that organizations can use to protect patients when they compare the benefits, risks, and costs of drugs to include in their "formularies," the lists of drugs approved for coverage.

"Manufacturers of new drugs make a range of claims about clinical benefits, differences in cost, and cost effectiveness," said Greg Simon, MD, MPH, a Group Health psychiatrist and researcher who is the lead author of the article. "Decision makers need some principles for sorting through those claims." While other groups have established broad principles for addressing the problem, the Group Health authors offer "more specific rules for identifying the right questions and the most important evidence," Simon explained.

In addition to Simon, co-authors on this paper include Bruce Psaty, MD, PhD, a Group Health Center for Health Studies senior investigator and University of Washington professor of medicine and epidemiology who is also a member of an Institute of Medicine Committee on the Assessment of the U.S. Drug Safety System; Group Health physician Marc Mora, MD, former chair of the committee that establishes formulary policies for Group Health; and Group Health pharmacist Jennifer Berg Hrachovec, PharmD, MPH.

"The money we have available for health care is not unlimited," said Simon. "Health insurers and government agencies that pay for health care have a responsibility to make sure the money we have is used for the greatest benefit. Our society tends to assume that a newer, more expensive treatment must be more effective. That assumption can be bad for our pocketbooks and our health."

Principles for evidence-based drug formulary policy

1. Give more weight to true experiments than to models or simulations, and carefully examine assumptions of such models.

The highest standard for finding true differences between newer and older treatments is the "randomized controlled trial," where study participants are assigned to different treatments by chance. Those "gold standard" comparisons are costly and may take years to complete. Policy makers sometimes rely on simulated comparisons, but those comparisons are often based on expert opinion and subject to bias. "When models or simulations are the best data available, appropriate skepticism is necessary," the authors write.

As an example, they point to Cox-2 inhibitor drugs such as Vioxx and Celebrex, which are more expensive, but no more effective on average, than older anti-inflammatory drugs. Simulated comparisons supported by the pharmaceutical industry showed that Cox-2 inhibitors could save money in the long run by reducing the risk of ulcers and intestinal bleeding. But that claim was not consistently demonstrated in randomized comparisons. The most recent evidence now finds that Cox-2 inhibitors probably do not reduce intestinal side effects and probably do increase risk of heart problems.

"Fortunately, many health care organizations had weighed the true scientific evidence of the potential risks and benefits of these drugs and decided years ago not to include them in their formularies," said Mora. "The principles we present can help organizations make these kinds of sound decisions in the interest of patient safety."

2. Give more weight to real health outcomes than changes in risk factors.

Comparisons should give more credit to medications that are actually proven to result in less illness and fewer deaths, rather than simply changing risk factors. For example, many long-term medications, such as cholesterol or blood pressure drugs, are prescribed to reduce future risk of disease. While newer blood pressure drugs have been shown to reduce blood pressure, older and more affordable drugs are probably more effective for actually preventing heart disease, explained Simon.

3. Look at the full range of alternatives rather than those selected by industry when considering claims for advantages of new treatments.

Studies needed for new drug approval often compare a placebo with a current standard treatment. But sometimes there are many standard treatments available, and the manufacturer of a new drug may not always choose the most informative comparison. Psaty and his colleagues have developed a method called "network meta-analysis" that can be used to make valid comparisons across several treatments that are studied separately. This can be "especially important when the comparisons most interesting to clinicians and policy-makers differ from those that are most interesting or advantageous to research sponsors," the authors write.

4. Understand that variation in effects across individuals or subgroups argues against restrictions on first-line treatment, but only if those differences are predictable.

Drugs approved for the same problem often show equal efficacy on average, but their effects can vary widely between individuals, the authors explain. "This …can be misinterpreted to support either unnecessarily broad or inappropriately restrictive formulary policies." If it’s not possible to predict at the beginning of treatment which drug will have the best effect, then starting with the most affordable alternative makes sense, the authors contend. But a more expensive medication is often justified for patients who respond poorly to first-line treatment.

The authors use the class of drugs for depression known as serotonin reuptake inhibitors (SRIs) as an example. Direct randomized comparisons of SRIs show no difference in average effectiveness. But studies also show that half the patients who fail to respond to one SRI may do well on another. It’s not possible to predict who will do well on which drug at the beginning of treatment. Therefore, it makes sense for an insurance company to only cover the most affordable drug for anyone starting treatment for the first time, the authors write. Patients who do poorly with the first medication should be allowed to use one of the more expensive alternatives.

5. Understand that variation in effects argues against requiring changes in ongoing treatment.

With this principle, the authors argue that formulary changes should not necessarily result in requirements that patients switch to new drugs. "In addition to concern about variability in effect, decisions …must consider added visits and other costs of medication changes, as well as the likelihood that many physicians and patients may be hesitant to disrupt stable treatment," the authors write.

Patients whose type 2 diabetes was accompanied by minor or major depression had higher mortality rates, compared to patients with type 2 diabetes alone according to a recent University of Washington (UW)/Group Health study that appears in the November 2005 edition of Diabetes Care, published by the American Diabetes Association.

The researchers surveyed and followed up 4,154 Group Health patients with type 2 diabetes. The patients filled out written questionnaires. With patients' consent, automated diagnostic, laboratory, and pharmacy data were collected. The researchers also reviewed Washington state mortality data to analyze diabetes complications and deaths.

Depression is common among people who have type 2 diabetes. This high prevalence can have unfortunate repercussions. Both minor and major depression among people with diabetes are strongly linked with increased mortality.

"Depression may be associated with increased mortality in patients with diabetes because of both behavioral and biologic factors," the researchers noted in their article. More work, they added, is needed to untangle the cause-and-effect relationships among depression, behavior, diabetes complications, and mortality.

Wayne Katon, MD, professor and vice chair of the UW Department of Psychiatry and Behavioral Sciences, led the recent study. The research team included Group Health researchers Carolyn Rutter, PhD, Greg Simon, MD, MPH, Elizabeth Lin, MD, MPH, Evette Ludman, PhD, and Michael Von Korff, ScD, as well UW’s Paul Ciechanowski, MD, MPH, and Bessie Young, MD, MPH, and Leslie Kinder, PhD, from the Veterans Affairs Puget Sound Health Care System.

Previous studies by this research group have shown that patients with depression and diabetes are less likely to follow diet and exercise guidelines or to check their blood glucose levels, and to have more lapses in filling their prescriptions for oral hypoglycemic, lipid-lowering, and high blood pressure medications. People with depression and diabetes were also more likely to have three or more heart disease risk factors, such as smoking, obesity, and a sedentary lifestyle, compared to people with diabetes alone.

Patients with both depression and diabetes are also significantly more likely to have cardiovascular and cerebrovascular complications. Depression may increase complications, not only because of poor self-care, but possibly through the brain chemistry and nervous system abnormalities that accompany depression, the researchers noted. They added that people may also become depressed in response to changes in their ability to function or because of physical symptoms, such as chronic pain from nerves damaged by diabetes.

In the UW and Group Health study, patients with diabetes accompanied by minor depression were less educated and were less likely to be Caucasian, in comparison to the diabetes patients without depression. Patients with diabetes and major depression were significantly younger, less likely to be married, and more likely to be female than were diabetes patients without depression. Both those with major and minor depression were more likely to have two or more diabetes complications, and were more likely to have another medical condition in addition to diabetes. They were also more likely to smoke, to be sedentary, to be obese, and to have been treated with insulin. Compared to diabetes patients with minor depression, those with major depression were more likely to be younger, female, and unemployed.

The researchers pointed out that a sedentary lifestyle was an important, independent predictor of mortality from diabetes. Earlier studies have shown that lack of physical activity can predict depression, and, conversely, that depression can predict the development of a sedentary lifestyle. Other studies have shown that improvements in treating depression in diabetes patients can lead to the patients exercising more and to better physical functioning.

Study shows characteristics of women who refuse follow-up tests for breast cancer

In a study of women with breast cancer, researchers from Kaiser Permanente, Group Health, and other health plans examined characteristics of women who refused recommended follow-up testing after a positive breast cancer screening test, or a visit to a medical provider for breast cancer symptoms. Those refusing were more likely to be 75 or older and have six or more children. The study appears in the Nov. 8, 2005, Journal of the National Cancer Institute Monographs.

"What is surprising is that these women who declined the recommended follow-up procedures were getting regular medical care and had a number of clinical visits prior to cancer diagnosis," says lead researcher Sheila Weinmann, PhD, with Kaiser Permanente’s Center for Health Research in Portland, Ore.

"Results of this study showed that refusal was not associated with socioeconomic status or race," says Joyce Gilbert, MPH, a researcher with Kaiser Permanente’s Care Management Institute, adding that "the nature of refusal is not yet fully understood."

Researchers found the most frequently documented reasons women refused follow-up care included denying there was a problem or having a fatalistic view of their medical prognosis. Others expressed fear of further diagnostic tests and surgery and fear of discomfort from a mammogram. Most refusals occurred at a clinical visit but some also took place during mammography appointments or during follow-up telephone calls. Sixty-one percent of the refusers had breast cancer symptoms noted in their medical chart sometime within the three years before breast cancer diagnosis.

The study looked at medical records of 1,347 women age 50 and older who had been diagnosed with late-stage breast cancer and an equal number of women with early-stage breast cancer. The women, who were diagnosed from 1995 to 1999, were from five Kaiser Permanente sites, Group Health Cooperative, and the Henry Ford Health System. These organizations are part of the Cancer Research Network, a consortium of 11 research organizations affiliated with nonprofit integrated health care delivery systems and the National Cancer Institute.

Seven percent of the women refused a health provider's advice for a follow-up procedure sometime during the three years before diagnosis. The women refusing breast cancer follow-up services were almost twice as likely as non-refusers to be in the late-stage group.

Researchers from Group Health Center for Health Studies who contributed to the study were R. Kevin Beverly, MA, Emily Westbrook, and Bill Barlow, PhD.

Group Health Center for Health Studies has named David E. Arterburn, MD, MPH, to develop and lead a new program of research in physical activity, weight management, and dietary change at Group Health.

Arterburn, who will join the CHS faculty in February, is an associate professor with the University of Cincinnati’s Division of General Internal Medicine. His research to date has focused on the efficacy and safety of drug and surgical therapies for obesity, the epidemiology of obesity, and obesity’s impact on quality of life, health services use, and costs of care. He also serves as medical editor for the Foundation for Informed Medical Decision Making, where he is developing a video-based decision support tool for bariatric surgery, as well as other tools for behavior and drug treatment for obesity.

Arterburn is a 1997 graduate of the University of Kentucky College of Medicine. He received his master of public health degree from the University of Washington School of Public Health where he was a fellow in the Northwest Veterans Affairs Health Services Research and Development Field Program.

The Group Health Physical Therapy Department has received the 2005 Birnbaum Award for Supporting Research that Benefits Group Health Members. The award, named for Group Health founders Hilde and Bill Birnbaum, is given by the Center to a Group Health work team that has made extraordinary contributions to successful research at the Cooperative. CHS staff members nominate teams and the winner is selected by CHS faculty.

The Physical Therapy Department was cited for "contributing substantially to the success of a series of randomized controlled trials of interventions for patients with back pain." Participating therapists showed "outstanding motivation and follow-through by contributing to the development of protocols, training therapists, and delivering protocol-based interventions," the CHS researchers wrote.

Service Line Director Brent Proctor accepted the award on behalf of the therapists during the Birnbaum Lecture on October 13.

Research Highlights

The work of several CHS researchers is prominently featured in a special Journal of the National Cancer Institute Monograph that focuses exclusively on articles from the Cancer Research Network (CRN). Published on November 8, 2005, the monograph was edited by Ed Wagner, MD, MPH, director of the CRN and a senior investigator at Group Health Center for Health Studies (CHS), and Thomas M. Vogt, MD, MPH, associate director of the Kaiser Permanente’s Center for Health Research Hawaii. It features 18 articles by scientists conducting cancer research within large, integrated care systems. Of these, 14 are co-authored by CHS researchers and/or CHS affiliate faculty members. Research topics include breast and cervical cancer screening, prophylactic mastectomy, health system smoking policies, racial disparities in cancer survival, and the response of health care organizations to the findings of the Women’s Health Initiative hormone therapy study. The monograph "addresses substantive and methodologic issues of screening, disparities, translation of research into practice, cancer care, risk reduction, and health services, as well as the expanding infrastructure that makes it possible to examine key research questions across multiple health care systems," Wagner and Vogt write.

Having diabetes is a strong risk factor for sudden cardiac death, according to a study of nearly 6,000 Group Health patients that was published in the June 24, 2005, online edition of the European Heart Journal. Researchers from the University of Washington’s (UW’s) Cardiovascular Health Research Unit joined lead author Xavier Jouven, MD, PhD, of the University of Paris to conduct the study, which compared the cases of patients in four groups: 1) no diabetes, 2) borderline diabetes, 3) diabetes without microvascular disease, and 4) diabetes with microvascular disease. They found that when compared to no diabetes, there was a progressively higher risk of sudden cardiac death linked to borderline diabetes, diabetes without microvascular disease, and diabetes with microvascular disease. High glucose levels were also associated with the risk of sudden cardiac death both in the absence and presence of microvascular disease.

Being underweight, being obese, smoking, and having depressive symptoms were linked to the development of "frailty," a geriatric syndrome that predicts several poor outcomes in older women, including death, hospitalization, hip fracture, and disability. These are the findings of a study co-authored by CHS Senior Investigator Andrea LaCroix, PhD, that involved more than 40,000 women aged 65 to 79. The researchers measured several components of frailty, including muscle weakness, impaired walking, exhaustion, and low physical activity. The investigation was led by Nancy Fugate Woods, PhD, dean of the UW School of Nursing, as part of the Women’s Health Initiative (WHI) Observational Study. LaCroix, in addition to her work at CHS, is co-principal Investigator of the WHI Clinical Coordinating Center at the Fred Hutchinson Cancer Research Center. The study appears in the August 2005 issue of the Journal of American Geriatric Society.

Women with moderate-to-severe urinary incontinence are two to three times more likely to be depressed than continent women according to a UW/Group Health study, published in the September 2005 issue of Obstetrics and Gynecology and led by Jennifer Melville, MD, MPH, a UW associate professor of medicine and CHS affiliate investigator. The researchers concluded that women with moderate-to-severe urinary incontinence should be screened for comorbid major depression and offered treatment if depression is present. Study co-authors included CHS’ Kristin Delaney, MPH, and Katherine Newton, PhD, and UW’s Wayne Katon, MD, MPH, also an affiliate investigator with CHS.

Although Canadians are concerned about access to family physicians, the aging of the population may not be the problem, according to a study co-authored by CHS Associate Investigator Rob Reid, MD, PhD, and published in the Spring 2005 issue of the Canadian Journal of Aging. Along with colleagues from the University of British Columbia’s Centre for Health Services and Policy Research, Reid studied the influence of population and workforce aging on supply and use of family physicians between 1991 and 2001. Among the researchers’ conclusions: Family physician use rates increased among older adults and decreased among younger adults. Also, older family physicians provided many more services than their predecessors and younger family physicians provided many fewer. "In terms of impact on future requirements for family physicians, both changes in age-specific rates of use, and changes in age-specific patterns of family physician productivity trump population aging as key drivers," the authors wrote. Reid also serves as associate director of Group Health’s Department of Preventive Care.

Volume 18, Issue 1
Winter 2006

CHS Research News
Vol 18, Issue 1, Winter 2006

Center News

New Findings

Research Highlights

Volume 18, Issue 1 Winter 2006

CHS Research News Vol 18, Issue 1, Winter 2006

Center News

New Findings

Research Highlights

Volume 18, Issue 1
Winter 2006

CHS Research News
Vol 18, Issue 1, Winter 2006