Upcycling: Bringing new life to existing health data

From the Health Affairs Blog:

By now, most of us are familiar with recycling. Items with reclaimable value are collected; then base materials are salvaged to create new products—often of lesser quality.

Fewer people are familiar with the term upcycling, a form of recycling that involves reconceiving, and sometimes adding to, existing items with the goal of giving them a different purpose and higher value. Though the term may be unfamiliar, the concept is old. A hundred years ago, farm families upcycled feed sacks into dresses, and old doors into furniture. The fleece jackets we now see everywhere, often made from reclaimed plastic bottles, are a modern example of upcycling.

In the world of health care data, too, opportunities exist to upcycle by adding to and repurposing existing information. It makes good sense to leverage investments in data collection, many of which have already been made for other reasons, such as public health surveillance and provider billing: doing so reduces data collection and cost burdens. The California HealthCare Foundation, based in Oakland, is actively pursuing data-upcycling initiatives as part of its mission to make useful information about health care quality publicly available. Two such efforts are highlighted below.

Maternity Care Data
More than 500,000 California women give birth each year; yet little information exists to guide decisions about where to seek high-quality maternity care. Providers themselves often lack information about their own performance. To help fill this gap, the California Maternal Data Center was launched in 2012 (it is cofunded by the California HealthCare Foundation and the Centers for Disease Control and Prevention and operated by the California Maternal Quality Care Collaborative). By repurposing data that hospitals and the state government have long collected, it provides metrics on the quality of maternity care. The California Maternal Data Center links birth certificate data (for example, birth weight, delivering provider) with information included in patient discharge data (for example, diagnostic and procedure codes related to the birth) that hospitals are already required to submit to the state.

With this combination of data, the center is able to produce robust measures, such as rates of cesarean sections, episiotomies, and vaginal birth after C-section, on all California hospitals providing maternity care. With a small amount of additional work, hospitals can voluntarily submit additional data elements from targeted medical chart reviews (that is, reviews of a subset of charts identified by the California Maternal Data Center to be most relevant) to generate other measures, such as elective delivery before thirty-nine weeks. Many participating hospitals are using the center’s data to facilitate quality improvement, and plans are under way to support public reporting at the hospital and physician-practice levels.

To read the complete post, click here.

CSCAR: Statistical consulting @ UM

The Center for Statistical Consultation & Research (CSCAR) provides free consulting in data analysis & statistical methods to all UM researchers.  We are able to assist with the design, analysis, interpretation, & communication of research studies involving data.

CSCAR also presents workshops on statistical methods and software for data management & analysis.  Fall, Winter and Spring workshop offerings include Statistics Review, SAS, SPSS, Stata, SEM & Analysis with R.

Visit the CSCAR web page for current offerings http://cscar.research.umich.edu/ for additional information.

Children’s EHR format available in USHIK

From the Agency for Healthcare Research & Quality (AHQR):

The Agency for Healthcare Research and Quality’s (AHRQ’S) Children’s EHR Format has moved to the United States Health Information Knowledgebase (USHIK). This  helps align the Children’s EHR Format with related knowledgebase information, such as IT artifacts related to meaningful use, health IT standards, & related clinical quality measures, as well as a friendlier user interface

The Children’s EHR Format, developed by AHRQ & the Centers for Medicare and Medicaid Services (CMS), with input from the American Academy of Pediatrics, was released in February 2013.  The Children’s EHR Format is a set of child-specific requirements (& other requirements of special importance for children) that an EHR should meet to perform optimally for the particular health care needs of children.

Find the Children’s EHR Format here under the “Child EHR Format” tab.  For more information, follow this link: healthit.ahrq.gov/childehrformat.

MEPS update

New data files & tables have been added to the  Medical Expenditure Panel Survey web site:

    • MEPS 1996-2011 Pooled Linkage Variance Estimation File (MEPS HC-036)
    • MEPS 1996-2011 Replicates for Variance Estimation File (MEPS HC-036BRR)
    • 2011 Non-Group Health Insurance Tables (No. 6 Tables Series)
    • Updated MEPS Household Component Summary Data Tables Technical Notes
    • 2011 Expenditures by Medical Condition Tables

Form more information, visit the MEPS web site.

Meaningful use updates in USHIK

From the Agency for Healthcare Research & Quality (AHRQ):

Meaningful Use Portal is the “one-stop shop” for Meaningful Use Stage 1 and Stage 2 Clinical Quality Measures, their computation logic, their data elements, & the vocabularies and codes the data elements may take on. This portal has been updated with new functionality:

  • Contains updates to the Clinical Quality Measures & Value Sets including the release package from which they originated. Users can now sort & filter the measures by release package & can download current or historical packages detailing the state of the CQM & Value Sets at the time they were released.
  • The Value Sets contained in the Meaningful Use portal have been updated to include each Code System’s identifier (OID). This information has been added to the screens, the API, & the exports containing Value Set information.
  • The short name for each EH CQM’s is now on the CQM list pages. Users can search or sort by this field.

For more information visit the USIK website.

Project Tycho – Data for health

TychoBraheProject Tycho, from the University of Pittsburgh (with funding from the Bill & Melinda Gates Foundation & the National Institutes of Health), is a database created to advance the availability & use of public health data for science & policy.

The project has completed the digitization of the entire history of weekly Nationally Notifiable Disease Surveillance System (NNDSS) reports for the United States (1888-2013) into a database in computable format (Level 3 data); standardized a major part of these data for online access (Level 2 data); and a subset of the U.S. data was cleaned further and used for a study on the impact of vaccination programs in the United States that was recently published in the NEJM (Level 1 data).

Read more about this fascinating project here.  You can also find it on the Health Statistics research guide on the “Incidence/Prevalence” & “Visualization” pages.

Hat tip to Gillian Mayman!

New database from CMS: Medicare Provider Charge Data

The Department of Health & Human Services has created a database that for the first time gives consumers information on what hospitals charge.  The data, on the charges for services that are provided during the 100 most common Medicare inpatient stays and 30 common outpatient services, show significant variation across the country and within communities.

For example, average inpatient charges for services a hospital may provide in connection with a joint replacement range from a low of $5,300 at a hospital in Ada, Okla., to a high of $223,000 at a hospital in Monterey Park, Calif.  Even within the same geographic area, hospital charges for similar services can vary significantly. For example, average inpatient hospital charges for services that may be provided to treat heart failure range from a low of $21,000 to a high of $46,000 in Denver, Colo., and from a low of $9,000 to a high of $51,000 in Jackson, Miss.

Access the database here and on the Health Statistics research guide.


Quandl: A new tool for datasets


Quandl is a new way to find & use data on the internet.   Quandl has indexed millions of time-series datasets–all open & free– from over 400 sources. Download a Quandl dataset in any format that you want. Visualize, save, share, authenticate, validate, upload, index, merge, & and transform data.

Here’s an example of data on infanct mortality in Haiti, from the World Bank.


New web site for USHIK database

The United States Health Information Knowledgebase (USHIK) is an on-line, publicly accessible registry and repository of healthcare-related data, metadata and standards. USHIK is funded and directed by the Agency for Healthcare Research and Quality (AHRQ) with management support in partnership with the Centers for Medicare & Medicaid Services (CMS).

The web site was recently redesigned for clarity and usability:  it’s easier to search and filter information, to find things on the menu, which now is on top, Draft Measures available for public feedback are in a new portal, and the site now better meets requirements for section 508 accessibility and usability.

See the new web site here.

How Low-Income Americans Experience Health Care: Ranking the States

From the Commonwealth Fund:

The first-ever health system performance scorecard focusing on low-income Americans has one overriding message: where you live matters.

The new Commonwealth Fund report Health Care in the Two Americas, along with the accompanying interactive map,provides the first state-by-state comparison of the health care experiences of the 39 percent of Americans earning less than twice the federal poverty level ($47,000 a year for a family of four or $23,000 for an individual). It finds that access to affordable health care and quality of care vary greatly depending on the state. In comparing the care experiences of low-income populations with higher-income groups, the report also finds striking income-related differences within each state.

Read the report & the interactive map here.