In previous research wide variations have been found in both healthcare spending and in health outcomes, with little correlation between the two. These studies were limited to cross-sectional analysis, and tell little about the dynamic process by which these patterns arise. One hypothesis is that variation across regions in rates of technology diffusion, whether for highly effective treatments (with a large impact on health outcomes) or for expensive treatments with unknown value (with a large impact on expenditures), can explain the observed cross-sectional patterns of spending and outcomes. In this proposal, Aim 1 seeks to better understand the diffusion of highly effective healthcare such as hemoglobin A1C (HbA1c) tests for blood glucose control among diabetic patients. Using the national Doximity database on every physician in the U.S., along with information about physician-hospital networks (PHN) and physician social networks, the research team will test why HbA1c diffused so rapidly (and among all racial and ethnic groups) in some areas but not others. They will also test whether more rapid diffusion of HbA1c reduced rates of neuropathy, retinopathy, and amputation.
Aim 2 focuses on the diffusion of generally beneficial treatments but where the treatment can actually harm specific types of patients. Two examples are considered: the rapid growth in implantable cardioverter defibrillators (ICDs), and the growth in new and expensive anticoagulants - dabigatran, apixaban and rivaroxaban.
Aim 3 studies the opposite of diffusion - """"""""exnovation"""""""" or a retreat from use - to ask how physician-hospital networks and regions scaled back on treatments newly found to have poor value for subgroups of patients. The proposal considers two specific treatments: the sharp reduction in carotid endarterectomy (both surgery and stents), and the decline in the use of Rosiglitazone (Avandia), an anti-diabetic drug, following a 2007 publication demonstrating serious cardiovascular risks. In these cases, the most effective exnovation patterns should experience the largest drop in use for the less appropriate patients.
Aim 4 examines the diffusion of treatments with unknown or even adverse consequences, such as the rapid growth in some regions (but not others) in ICU bed capacity. The research team will study the network and diffusion patterns for """"""""extramedical"""""""" treatments - illegal behavior motivated by profit and with no benefit for patients, with one example being the rise and fall of immunoglobulin infusions in 2002-2005. Finally, the research group will use results from these four aims to return to the central hypothesis: can observed differences in treatment-specific diffusion explain observed patterns in regional variations in health outcomes and spending?

Public Health Relevance

Despite the high and variable cost of healthcare in the U.S., there remain persistent and large shortfalls in the quality of care. This study seeks to explain why these variations exist by focusing on patterns of diffusion of new healthcare technologies. Using a national database on physician characteristics, measures of social and practice networks, and clinical registry data linked to Medicare claims data, we seek to understand why some healthcare providers are better able to adopt high-quality care quickly - and use it effectively - and why some healthcare providers instead adopt the most costly and least effective treatments.

Agency
National Institute of Health (NIH)
Institute
National Institute on Aging (NIA)
Type
Research Project--Cooperative Agreements (U01)
Project #
5U01AG046830-02
Application #
8738583
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Haaga, John G
Project Start
2013-09-30
Project End
2018-06-30
Budget Start
2014-08-01
Budget End
2015-06-30
Support Year
2
Fiscal Year
2014
Total Cost
Indirect Cost
Name
Dartmouth College
Department
Miscellaneous
Type
Schools of Arts and Sciences
DUNS #
City
Hanover
State
NH
Country
United States
Zip Code
03755
Moen, Erika L; Bynum, Julie P; Austin, Andrea M et al. (2018) Assessing Variation in Implantable Cardioverter Defibrillator Therapy Guideline Adherence With Physician and Hospital Patient-sharing Networks. Med Care 56:350-357
An, Chuankai; O'Malley, A James; Rockmore, Daniel N et al. (2018) Analysis of the U.S. patient referral network. Stat Med 37:847-866
Bekelis, Kimon; Missios, Symeon; MacKenzie, Todd A (2018) Outcomes of Elective Cerebral Aneurysm Treatment Performed by Attending Neurosurgeons after Night Work. Neurosurgery 82:329-334
Bekelis, Kimon; Missios, Symeon; MacKenzie, Todd A (2018) Correlation of hospital magnet status with the quality of physicians performing neurosurgical procedures in New York State. Br J Neurosurg 32:13-17
Chang, Chiang-Hua; P W Bynum, Julie; Lurie, Jon D (2018) Geographic Expansion of Federally Qualified Health Centers 2007-2014. J Rural Health :
Bekelis, Kimon; Chang, Chiang-Hua; Malenka, David et al. (2018) Direct oral anticoagulant and antiplatelet combination therapy: Hemorrhagic events in coronary artery stent recipients. J Clin Neurosci 50:24-29
Bekelis, Kimon; Missios, Symeon; Shu, Joel et al. (2018) Surgical outcomes for patients diagnosed with dementia: A coarsened exact matching study. J Clin Neurosci 53:160-164
Aschbrenner, Kelly A; Bobak, Carly; Schneider, Emily J et al. (2018) Egocentric social networks and smoking among adults with serious mental illness. Transl Behav Med 8:531-539
Bekelis, Kimon; Missios, Symeon; Coy, Shannon et al. (2018) Association of Hospital Teaching Status with Neurosurgical Outcomes: An Instrumental Variable Analysis. World Neurosurg 110:e689-e698
Missios, Symeon; Bekelis, Kimon (2018) Association of Hospitalization for Neurosurgical Operations in Magnet Hospitals With Mortality and Length of Stay. Neurosurgery 82:372-377

Showing the most recent 10 out of 46 publications