Scientific Highlights—March 2014

Group Health researchers do it faster

Natural language processing can reduce—by 90 percent—how many charts need to be painstakingly “abstracted” by hand to identify confirmed recurrences of breast cancer, according to David Carrell, PhD. This computed targeting of mentions of certain words in unstructured electronic health record charts can streamline the research process. Dr. Carrell studied nearly 1,500 Group Health patients with Diana Buist, PhD; Jessica Chubak, PhD; Scott Halgrim, MA; Diem-Thy Tran; colleagues at Harvard Medical School, Boston Children's Hospital, and the University of California San Diego. They e-published January 30 in the American Journal of Epidemiology. See abstract.

Here’s how to apply Lean 101 to mental health settings

Group Health’s Behavioral Health Services used Lean process improvement to implement evidence-based practices. E-publishing January 25 in the Journal of Behavioral Health Services & Research, Bradley Steinfeld, PhD, described Lean and how they used it. His coauthors were Gavin Vilander MS; Larry Marx MD; and colleagues at Seattle Children’s Hospital, Molina Healthcare, and the Evidence-Based Practice Institute. Calling for more research to document Lean’s effectiveness in this context, they advised other mental health organizations to start with a relatively simple process, like provider schedules. See abstract.

Too often, seniors fall hard

When seniors have ground-level falls requiring hospitalization, they’re likely to incur severe injury, be readmitted, and die within a year. That’s what Christopher D. Mack, MS—and colleagues at the UW, Harborview Injury Prevention and Research Center, and New York University—found in a study of more than 1,300 consecutive admissions at Harborview Medical Center. In the February Journal of Trauma and Acute Care Surgery, they urged examination of whether improving the quality of post-hospital care can raise longevity and functional outcomes. See abstract.

Is your chair killing you?

Sitting, reclining, and lying down: what do these (in)activities have in common? The more time spent doing them, the higher the risk of death, in 12 years of follow-up of more than 90,000 women in the Women’s Health Initiative. Andrea LaCroix, PhD, published in the February issue of the American Journal of Preventive Medicine with colleagues at the Fred Hutchinson Cancer Research Center, the Universities of California San Diego, Florida, Illinois, and Pittsburgh, and Cornell, Duke, Harvard, Stanford, Stony Brook Universities. Their calculations controlled for multiple potential confounders. See abstract.

How can women kick the hormone therapy habit?

Women often struggle to quit taking hormone therapy—even if tapering dosage—because of menopause symptoms recurring. In 800 women at Group Health and Harvard Pilgrim Health Care, those with greatest difficulty had disturbed sleep and mood—and/or hadn’t received a doctor’s advice. E-publishing January 20 in the Journal of Women’s Health, Katherine Newton, PhD, concluded doctors should advise about this—particularly on managing insomnia or depression, as appropriate. Her co-authors were Susan Reed, MD, MPH; Louis Grothaus, MS; Evette Ludman, PhD; Kelly Ehrlich, MS; Andrea LaCroix, PhD; and a Harvard Medical School colleague. See abstract.

Suicide prevention researchers take cues from patients

In an example of patient-centered research, Ursula Whiteside, PhD, involved “target users” to help develop messages and strategies for engaging people in online suicide-prevention interventions. The target users—Group Health Behavioral Health Services clients currently thinking of suicide—took an anonymous survey to share their intervention preferences. In the February Journal of Medical Internet Research, she described this work with Julie Richards, MPH; Gregory Simon, MD, MPH; Evette Ludman, PhD; and UW colleagues. See abstract.

Patients with worse pain and insomnia benefit more from CBT

When Group Health patients participated in the primary care-based Lifestyles trial, it was those with more severe arthritis pain and insomnia who benefited most—and for the longest time—from cognitive behavioral therapy (CBT) for the two conditions. And including CBT for insomnia led to more improvement than did CBT for pain alone. The authors—Sue McCurry, PhD; Susan Shortreed, PhD; Michael Von Korff, ScD; Ben Balderson, PhD; and colleagues at the University of Washington (UW), Wake Forest University, and Virginia Commonwealth University—published in the February issue of Sleep. See abstract.