AHRQ Quality Indicators banner

[Federal Register: May 16, 2006 (Volume 71, Number 94)]
[Page 28345-28346]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]

Agency for Healthcare Research and Quality

AHRQ Quality indicators Workgroup on Risk Adjustment Approaches to Administrative Data

AGENCY: Agency for Healthcare Research and Quality (AHRQ), HHS.

ACTION: Notice of request for nominations.

SUMMARY: The Agency for Healthcare Research and Quality (AHRQ) is seeking nominations for members of an AHRQ-convened Workgroup on risk adjustment specifically aimed at the AHRQ Quality Indicators (QIS). This Workgroup is being formed as part of a structured approach for evaluating risk-adjustment and the appropriateness of hierarchical modeling methodology for the AHRQ Quality Indicators at the area and/or provider levels. The Workgroup will evaluate appropriate technical and methodological approaches currently available, and will also discuss and suggest strategies as to what risk adjustment approach(s), if any, would best fit AHRQ QI user needs. As part of this effort and using the AHRQ QIs, the Workgroup member will be addressing several key issues for the development of a risk adjustment methodology, including but not limited to:

  • Statistical and methodological issues related to the development and validation of risk adjusted models that predict patient outcomes using administrative data, and are suitable for assessing quality at different levels (individual hospital, State, region).
  • Methods for comparing the performance of hierarchical methods with previously employed methods based on administrative data to improve predictive and discriminant ability, and overall fit.
  • Appropriate use of sub-sampling techniques for model validation.
  • Computation of confidence intervals for assessing provider-specific and State-level performance in comparison to national summary statistics (means or percentiles).

For additional information about the AHRQ QIs, please visit the AHRQ Web site at http://qualityindicators.ahrq.gov.

Specifically, the AHRQ QIs Risk Adjustment Workgroup will consist of up to 9 individuals who are familiar with different risk adjustment methodologies including hierarchical modeling approaches. The Workgroup will have a series of conference calls to discuss the technical and policy issues surrounding risk adjustment for the AHRQ QIs and will then assist AHRQ in developing a report that will aim to summarize the discussions and suggestions of the workgroup, which will be made available for public comment.

DATES: Please submit nominations on or before June 15, 2006. Self- nominations are welcome. Third-party nominations must indicate that the individual has been contacted and is willing to serve on the workgroup. Notification of selected candidates will be contacted by AHRQ no later than June 29, 2006.

ADDRESSES: Nominations can be sent in the form of a letter or e-mail, preferably as an electronic file with an e-mail attachment and should specifically address the submission criteria as noted below. electronic submissions are strongly encouraged. Responses should be submitted to:

ATTN: Project Officer, AHRQ Quality Indicators Project, Agency for Healthcare Research and Quality, Center for Delivery, Organization and Markets, 540 Gaither Road, Room 5121, Rockville, MD 20850, E-mail: projectofficer@qualityindicators. ahrq.gov.

Submission Criteria

To be considered for membership on the AHRQ QI Workgroup, please send the following information for each nominee:
  1. A brief nomination letter highlighting experience/knowledge relevant in the development and use of risk adjustment methodology including hierarchical modeling approaches and familiarity with the AHRQ QIs and health care administrative data. (See selection criteria below.) Please include full contact information of nominee: name, title, organization, mailing address, telephone and fax numbers, and e- mail address.
  2. Curriculum vita (with citations to any pertinent publications).

Nominee Selection Criteria

Nominees should have technical expertise in health care quality measurement development, and a familiarity with statistical methods in the area of risk adjustment as well as hierarchical modeling. More specifically, each candidate will be evaluated using the following criteria:

  • Knowledge of recent risk-adjustment and hierarchical modeling approaches published in the literature;
  • Peer-reviewed publications relevant to the development and use of risk-adjustment, hierarchical modeling; performance measures and reporting;
  • Expertise in statistical methods relevant to the evaluation of alternative approaches to risk-adjustment and hierarchical modeling;
  • Experience with development of measures based on administrative data and its uses;
  • Expertise in hospital quality improvement and patient safety;
  • Familiarity with the AHRQ Quality Indicators and their application; and,
  • Availability to participate in conference calls and provide written comments starting from late June through September 2006.

Time Commitment

In an effort to provide for expert input and for recommendations on how to improve on the existing risk adjustment approach to administrative data, we are initiating a review process that will require participation in approximately

[[Page 28346]]

four to five conference calls with some pre and post evaluation time (approximately 13 hours). Results from this process will influence the development of risk-adjustment and hierarchical modeling approaches for the AHRQ Quality Indicators. Beginning in late June through September, selected nominees will be asked to participate in the following activities:

Workgroup Activities

  1. Provide evaluative comments on current methodology for risk- adjustment and hierarchical modeling (2.0 hours) and participate in subsequent Workgroup call (1.0 hour);
  2. Participate in second Workgroup conference call to discuss suggested changes to the current modeling methodology, including the adoption of hierarchical methods (1.5 hour);
  3. Provide evaluative comments on AHRQ's new draft or revised methodology (1.5 hour);
  4. Participate in third Workgroup call to respond to each others' comments and questions or provide additional clarifications regarding draft methodology (1.5 hours);
  5. Review draft summary document (1.5 hour);
  6. Participate in fourth Workgroup call. Provide suggestions for summary document for public comment (2.0 hours); and,
  7. Participate in final Workgroup call. Discuss and respond to public comments (2.0 hours).

Please note that should additional conference calls be necessary, Workgroup members are expected to make every effort to participate. The Workgroup will conduct business by telephone, e-mail, or other electronic means as needed.


Mamatha Pancholi, Center for Delivery, Organization, and Markets,
Agency for Healthcare Research and Quality,
540 Gaither Road, Rockville, MD 20850;
Phone: (301) 427-1470; Fax: (301) 427-1430;
E-mail: mamatha.pancholi@ahrq.hhs.gov

Marybeth Farquhar, Center for Delivery, Organization, and Markets,
Agency for Healthcare Research and Quality,
540 Gaither Road, Rockville, MD 20850;
Phone: (301) 427-1317; Fax: (301) 427-1430;
E-mail: marybeth.farquhar@ahrq.hhs.gov.



The AHRQ Quality Indicators (AHRQ QIs) are a unique set of measures of health care quality that make use of readily available hospital inpatient administrative data. The QIs have been used for various purposes. Some of these include tracking, hospital self-assessment, reporting of hospital-specific quality or pay for performance. The AHRQ QIs are provider- and area-level quality indicators and currently consist of four modules: the Prevention Quality Indicators (PQI), the Inpatient Quality Indicators, the Patient Safety Indicators (PSI), and the Pediatric Quality Indicators (PedQIs). In response to feedback from the AHRQ QI user community, AHRQ is committed to developing risk adjustment approaches in an effort to provide an overall view of quality that is complete, useful and easily understandable to consumers and others with the health care field.

Dated: May 8, 2006.
Carolyn M. Clancy,

[FR Doc. 06-4574 Filed 5-15-06; 8:45am]