Family Medicine Research That Provides Compelling, Urgent Data to Improve Patient Care

For patients with depression, it is not just remission that matters, but the time to reach remission. In a fascinating and important article, Garrison et al¹ find substantial differences in the amount of time it took to reach remission among patients with depression treated with collaborative care versus usual care. Much of the previous literature emphasizes remission itself as the primary outcome. This article provides dramatic results, that is, more than a year difference in time to remission for collaborative care patients compared with those without collaborative care. All were diagnosed with depression or dysthymia; all were treated with antidepressants. The Robert Graham Center also presents some useful data that are pertinent to the collaborative treatment of depression. While family physicians frequently work with nurse practitioners and/or physicians assistants,² the number of family physicians working directly in the office with behavioral specialists or psychiatrists is much lower at about one fifth of family physicians. To increase collaborative mental health care, these numbers need to increase. The recent Journal of the American Board of Family Medicine supplement on integrated care³ provides related helpful information.

Health literacy is known to be important in health outcomes. Health care providers often work to overcome literacy issues, but often without much guidance or certainty of improved outcomes. The Agency for Health care Research and Quality has a health literacy toolkit that Weiss et al⁴ tested for medication reconciliation in 2 practices. The primary implemented items were not particularly time consuming, and the results were impressive: substantially more drug therapy problems were identified and revisions undertaken. For those not using the toolkit methods, it's time to start!

In another persuasive report, Chang et al⁵ find that patients who initiate a discussion of intrauterine devices (IUDs) keep their IUDs much longer than do patients for whom the physician introduced the idea of an IUD. The difference among this population was more than a year! The physician-initiated IUDs were removed at 6 months, on average. Follow-up research is needed to identify interventions to improve these numbers.

Our readers in the United States are now coding visits using the International Classification of Diseases, 10th Revision (ICD-10). And, by now, they may well know how easy or tough the new codes are to use. Yet, these same readers are unlikely to know how well they are doing, because reimbursement rates have not yet been well tested. Boyd and coauthors⁶ provide a great service by estimating the inherent coding difficulties embedded in the change from International Classification of Diseases, 9th Revision (ICD-9), to ICD-10. About a quarter of the diagnosis code transitions from ICD-9 to ICD-10 were found to be convoluted, and a third of the payments were convoluted—exactly what many physicians fear. A quarter of diagnoses is a large number any day, and could actually mean at least one diagnosis for most patients seen. One third of payments is also a huge figure. The impact on our ability to compare former with current data regarding our quality of care systems or research could easily be as large. The authors provide some estimates of transition costs, as well as the impacts in other countries, but not an answer to the overall value (dollar or otherwise) of the transition to ICD-10.

It is possible for family physicians to assist in improving electronic health records (EHRs). After interviewing EHR vendors, Olayiwola et al⁷ provide 6 strategies for immediate implementation—for the sake of all US physicians! Some of this effort should clearly go toward EHR assistance for ICD-10 coding.

Alcohol dependence is a chronic disorder frequently encountered among family medicine patients. Long-term, repeated, multifaceted efforts often are necessary for people with alcohol dependence to stop or substantially reduce their alcohol intake and, by doing so, their risks for bad outcomes. In the report from Berger et al,⁸ patients did not drink on ∼80% of days for 3 months after treatment. Specifically for those who drank daily, the goal of abstinence was associated with better outcomes over 3 months. Daily drinkers seemed to have an easier time being abstinent than they did cutting the amount they drank in any one day, suggesting a triggering or reinforcing effect of any alcohol intake. Most patient change occurred in the first 2 weeks and then was maintained. A goal of abstinence versus reduction was selected by the patient, without specific direction. This study suggests that we should help daily drinkers choose an abstinence goal. Keeping the long-term goal in mind is necessary to help understand that relapses are a part of the trajectory to long-term health and are to be expected.

In other clinical research, Logue et al⁹ provide evidence that supports the concurrent use of the Charlson comorbidity index with a polypharmacy score to improve the prediction of hospital readmissions. Trimeloni and Spencer¹⁰ present tactics to deal with an under-expected, understudied, and under-discussed problem: breast milk oversupply. An excellent review on preexposure prophylaxis for HIV prevention is authored by Conniff and Evensen.¹¹ Ledford et al¹² investigate the outcome of implementing computed tomography for reducing lung cancer mortality in a community hospital. The multiple articles on patient-centered medical homes^{13⇓⇓⇓⇓⇓⇓–20} are separately discussed by Thomas Rosenthal, who wrote the classic patient-centered medical home article, one of the most cited and read articles from the Journal of the American Board of Family Medicine.²¹