Wednesday, February 27, 2019

Improving outcomes that matter for patients with type 2 diabetes

When comparing therapies for type 2 diabetes, physicians, patients, and quality measures often get caught up in the disease-oriented outcome of glycemic control. However, a 2014 editorial in American Family Physician pointed out that of the medications then available to lower blood sugar levels, only metformin reduced mortality and clinically relevant complications. Since that time, studies of some newer diabetes medications have demonstrated mortality benefits in patients with cardiovascular disease. However, the best second-line medication after metformin remains unclear. Although the World Health Organization guideline recommended inexpensive sulfonylureas as second-line therapy in low-resource settings, the American College of Physicians and the American Academy of Family Physicians suggested "the choice of drug [after metformin] should be based on a conversation with the patient about benefits, possible harms, and cost."

In the February 15 issue of AFP, Drs. Joshua Steinberg and Lyndsay Carlson applied the STEPS criteria (safety, tolerability, effectiveness, prince, and simplicity) to each of the 10 categories of diabetes medications (including insulin). Their analysis confirmed that metformin should be first-line pharmacotherapy for most persons with type 2 diabetes. Other key points from this article include:

Safety - Sulfonylureas, insulins, meglitinides, and pramlintide increase risk of hypoglycemia. Metformin and acarbose require monitoring, dose adjustments, or discontinuation in patients with chronic kidney disease.

Tolerability - Side effects across different drug classes range from gastrointestinal effects (metformin, acarbose, meglintinides, pramlintide, GLP-1 receptor agonists, SGLT-2 inhibitors) to weight gain (sulfonylureas), edema (TZDs), severe arthralgias (DPP-4 inhibitors), and genital and urinary tract infections (SGLT-2 inhibitors).

Effectiveness - Recent trials showed improved patient-oriented outcomes from some GLP-1 receptor agonists and SGLT-2 inhibitors in patients at high cardiovascular risk or with known cardiovascular disease. Acarbose also reduces cardiovascular events.

Price - Metformin, acarbose, sulfonylureas, and generic pioglitazone are the most affordable options.

Simplicity - Acarbose and meglitinides are taken three times daily before meals, while insulins, GLP-1 receptor agonists, and pramlintide require subcutaneous injections.


This post first appeared on the AFP Community Blog.

Tuesday, February 19, 2019

Living, teaching, and valuing the pursuit of health equity

My younger son, who turned seven last week, had a minor health scare in December. After a few days of a cough and runny nose, one morning he complained that he didn't want to walk because his legs hurt. We gave him some liquid ibuprofen, wondering if these were just growing pains, but the pain kept coming back. Adding to our concern, this is a child with a high threshold for pain; for example, he barely blinked when having a cavity filled at age three. Over the next week or so, he developed red and purple blotches on his lower legs and feet that looked like this and this.

After consultations with our family doctor and other physician friends and family, he was diagnosed with Henoch-Schonlein purpura, a relatively rare condition that, fortunately, resolves spontaneously in most but can cause kidney disease in up to half of affected patients. Through the first week of January, his legs continued to hurt off and on as the rash slowly faded, but his kidney function remained normal, and the only medical bills we received were from a single clinic visit and some outpatient blood tests covered by our health insurance. We were lucky, not only because my son avoided complications, but because we had the advantage of being well positioned to obtain further care for him had they occurred.

Adjusted U.S. ESRD incidence rates, 2000-2015

Last week, I led a team-based learning exercise for the first-year class at Georgetown University School of Medicine on disparities in kidney (renal) disease. Not only are some racial and ethnic groups more likely to suffer from end-stage renal disease (when kidney function has deteriorated to the point that dialysis or a kidney transplant is often needed), but this unequal burden is unevenly distributed geographically, reflecting disparities in socioeconomic status. The graphic below, showing much higher rates of end-stage renal disease in the majority-African American northeast and southeast quadrants of Washington, DC, coincides with my years of practicing in these areas and noticing clusters of fast-food chains and dialysis centers around primary care clinics.

End-stage Renal Disease in Washington, DC

Similar large disparities in mortality and life expectancy are present nationally. In a 2006 paper, Dr. Christopher Murray (profiled in Jeremy Smith's "Epic Measures") described "Eight Americas," collections of U.S. counties defined by a mixture of race, geography, socioeconomics, and population density that demonstrated striking differences in mortality patterns. In a more recent analysis, Dr. Murray and colleagues concluded that geographic disparities in life expectancy have worsened over the past three decades: "Compared with the national average, counties in Colorado, Alaska, and along both coasts experienced larger increases in life expectancy between 1980 and 2014, while some southern counties in states stretching from Oklahoma to West Virginia saw little, if any, improvement over this same period."

Observing that overall U.S. life expectancy has been falling since 2015 after decades of steady improvement, a recent editorial in the Annals of Internal Medicine called on health research funders to "honestly recognize the interactive roles of biology and the socioeconomic and political environment ... [and] align health research resources toward an integrative model of science that seriously investigates the socioeconomic and political determinants of health alongside the biological ones." Reducing disparities in HIV/AIDS, for example, will require not only more clinical resources and affordable drugs, but more studies of policy interventions to improve social and living conditions that increase the risk of acquiring HIV infections in the first place. This means going beyond meeting individual-level social needs to changing the conditions that make people sick in communities and populations (including the Eight Americas).

As "value-based" health care payment models rapidly expand, based on the argument that we ought to be paying for good health outcomes rather than the volume of health services, we are recognizing that social determinants put some populations at a health disadvantage that effective interventions, such as increasing primary care physician supply, cannot fully overcome. As a result, physicians and health systems who serve underserved populations may be judged as providing "poor quality" care (and paid less) simply because their patients' outcomes were so much worse to begin with.

What can be done about this? Dr. Krisda Chaiyachati and colleagues have argued for a "disparities-sensitive frame shift" in value-based payment:

The health care industry cannot ignore true instances of poor quality, but it also should not worsen health care for at-risk populations. To address this tension, value-based payment models should ... integrate measures of equity into hospitals' financial calculus, incentivizing hospitals to tackle the disparities challenge without losing sight of quality. ... Concurrently, we should start paying hospitals to reduce disparities directly.

Similarly, Dr. Christopher Frank observed that "value-based payments will only work when we decide that health equity is an important value to reward."

Make no mistake: I'm happy that my son got well and has stayed well. I'm not happy that it is inherently more difficult for other children to have the positive outcome that he experienced. Reducing these disadvantages and pursuing health equity informs my clinical practice, teaching in population health, and writing projects.

Monday, February 11, 2019

Does subspecialist-oriented medical care add sufficient value to be worth the added cost?

The latest Graham Center One-Pager in the February 1 issue of American Family Physician contained good news and bad news for primary care. Examining the entry of medical students into residency programs between 2008 and 2018, Dr. Robert Baillieu and colleagues reported that the total number of graduates who entered Family Medicine through the National Residency Matching Program increased by 64% over the past decade. However, the annual proportion of U.S. allopathic (MD) graduates remained static at around 50%, reflecting the continued migration of most students into higher-paying medical subspecialties.

Two of my previous posts reviewed research demonstrating that students entering family medicine are more likely to make patient-centered, cost-conscious clinical decisions and that primary care physicians who trained in low-cost hospital service areas are more likely to provide high-value care in practice. The late health services researcher Barbara Starfield, MD, MPH once argued that a lack of investment in primary care is a major reason that the U.S. health system spends so much but produces poor outcomes:

The thing that is wrong with our current health care system is that it is not designed to produce the best effectiveness, efficiency and equity in health services because it is too focused on things that are unnecessary and of high cost rather than arranging services so that the most needed services are provided when needed and with high quality. [This] is the case because the country has not put sufficient emphasis during the past 50 years on a good infrastructure of primary care. ... We have done a reasonably good job at making subspecialty care available, but a lot of subspecialty care is not necessary if you have good primary care. So we end up with a very expensive system that does things unnecessarily.

In a recent nationally representative study in JAMA Internal Medicine, Dr. David Levine and colleagues examined associations between receipt of outpatient primary care and care value and patient experience. Using Dr. Starfield's definition of primary care as "first-contact care that is comprehensive, continuous, and coordinated," the authors compared the quality and experience of care in more than 70,000 U.S. adults with and without primary care who participated in the Medical Expenditure Panel Survey from 2012 to 2014. 70% of the primary care clinicians identified by patients were family physicians (19% were general internists). After adjustment for potential sources of confounding, respondents with primary care were more likely to receive high-value preventive care and counseling and to report better patient experiences than those without primary care. However, respondents with primary care were also slightly more likely to receive low-value prostate cancer screening and antibiotics for respiratory infections.

In an accompanying editorial that noted the disparity in primary care investment between the U.S. (7% of total health care spending) and the health systems of other industrialized nations (20%), Dr. Allan Goroll asked: "Does primary care add sufficient value to deserve better funding?" Although this formulation recognizes that the American status quo is a subspecialist-oriented health system, it seems to me that the question ought to be, "Does subspecialist-oriented medical care add sufficient value to primary care be worth the added cost?" From this study and previously published evidence, the answer appears to be no.


This post first appeared on the AFP Community Blog.