Primary Care Access and Quality (2003-2019)

<h2>Overview</h2>
<p>The Dartmouth Institute for Health Policy and Clinical Practice (TDI) has created a publicly available source of data that provides researchers, payers, regulators, and innovators with metrics that quantify temporal and regional patterns of health care spending and utilization in the United States. Using CMS Medicare claims data (mostly for age >64 enrollees), Atlas researchers built cohorts (“denominators”) and numerous measures or events (“numerators”) which were then used to calculate rates either by geography or for patients assigned to specific hospitals. These rates, which are calculated consistently across time and place, provide researchers with opportunities to evaluate spatial and temporal variation/trends. </p>

<p>This entry contains Dartmouth Atlas rates for measures related to primary care visits, quality of care for diabetic patients, mammography rates, leg amputation rates, and hospitalization rates for ambulatory care sensitive conditions. The denominators and the adjustment methods (e.g., for age, sex, and race) differ across measures, as described in the methods sections of the related publications. Rates are provided at the state, county, hospital referral region (HRR), and hospital service area (HSA) levels. Both overall rates and rates stratified by race (black and non-black) are provided for each geographic region and year. </p>

<p> Users downloading data should review the methods sections of the related publications for context as well as for information about any temporal changes in methods. All reports in the Dartmouth Atlas of Health Care series are available from the National Library of Medicine <a href="https://www.ncbi.nlm.nih.gov/books/NBK584737/">https://www.ncbi.nlm.nih.gov/books/NBK584737/</a> <p>

<p> Note that for the general Dartmouth Atlas rate datasets, which span multiple decades, the author list includes all Dartmouth staff (programmers, analysts, supervisors, etc.) involved in generating all types of Atlas rates across all years. We do not attempt to assign individuals to specific datasets or years. <p>