Supporting Statement B
Uniform Data System, OMB Control No. 0915-0193
1. Respondent Universe and Sampling Methods
The respondent universe for the 2012 Uniform Data System (UDS) includes all health center grantees who receive HRSA funding in CY 2012. The unit of analysis is at the grantee level, and the entire universe will be included in the dataset (1287 grantees). Expected response rate for the data collection effort is 100%. UDS data have been collected annually from all grantees since before 2000, and the actual response rate has been 100% of the universe. Since all grantees submit data, there are no sample size calculations. The universe is included, rather than a sample of health center grantees, because the UDS data reporting activities are part of programmatic requirements; all grantees are required to submit data to BPHC so that the data can be used to track grantees’ performance and inform performance improvement efforts.
2. Procedures for the Collection of Information
No statistical methods are needed for sample selection, because the entire universe of health center grantees is required to submit data to the UDS. Since the universe is reporting, we are able to calculate the true population parameters, rather than estimating statistics based on a sample.
Data are collected and submitted by health center grantees. Data are collected through administrative, financial, and medical records reviews. Data are not based on survey questionnaires, so interviewers are not applicable to the data collection process. Grantees are given advance notice when the UDS is open for data submission in their Electronic Handbook (EHB). They have 3 months to submit and finalize their data after the editor review described below.
BPHC has several strategies for assuring data quality. For instance, BPHC provides very specific written instructions in the UDS Manual that is updated annually on how grantees should report the data to the UDS. BPHC also hires a contractor to provide training and technical assistance during UDS reporting activities. The contractor works with Primary Care Associations in all states to provide annual UDS data reporting training to grantees; each health center typically sends 1 or 2 representatives to the trainings and they report back to the center and assist with the data reporting. The contractor also runs a consultation helpline which any grantee can access if they require assistance in any data collection activities. Once data have been submitted to the UDS, the contractor performs numerous data checks and screenings to identify and resolve any data discrepancies or outliers. The contractor has numerous staff members who work one-on-one with health centers to ensure that all outliers are real or that incorrect data entries are corrected. They also conduct sensitivity analyses to determine the impact of any data edits.
3. Methods to Maximize Response Rates and Deal with Nonresponse
All funded grantees are required to complete the UDS so there are no issues of non-response. Training is provided for grantees and technical assistance is available for all grantees to facilitate reporting for the UDS. A toll free number is also available for assistance and to address questions, and grantees may also receive help by asking questions via email. Since the dataset includes the universe of grantees, the UDS provides accurate and reliable population parameters; there are no issues of reliability since this is not a sample of grantees.
The expected response rate is 100%. This is based on several years experience collecting UDS data, when actual response rates for each year have been 100%. Follow-up procedures for getting all grantees to submit their data include email notifications and telephone calls to health center administrators, with as many contacts as are necessary for them to complete data submission.
4. Tests of Procedures or Methods to be Undertaken
All UDS tables used for data collection and reporting are approved by OMB before they are implemented. (See OMB Control Number 0195-0193, Expiration date 01/31/2014).
BPHC pilot tests major changes in data collection before they are proposed. For example, in 2008 a pilot test was conducted before new clinical performance measures were introduced. In 2011, a pilot test was conducted for new data collection and reporting of staff tenure data. The results of pilot tests were described in each applicable supporting statement for the proposed data collections.
For the clinical performance measures, the data collection methods and requirements have not changed since they were implemented in 2008. For each measure, grantees may report on a sample of 70 patients or for all patients within the specified universe (e.g., for the blood pressure control measure, the universe is all patients age 18 to 85 years with a diagnosis of hypertension, seen at least twice during the reporting year). BPHC encourages health centers to report on all patients that meet the identified criteria for each measure using their Electronic Health Records (EHRs). HRSA considers this method to present a relatively low burden to grantees, especially in view of the value of the data for grantee quality improvement. Grantees have been very positive about reporting the clinical measures. Very few critical comments have been received about the burden of data collection for existing and proposed clinical measures.
5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or
Analyzing Data
Consultation on data collection was obtained from the organization and individual listed below. John Snow, Inc. is a public health research and consulting firm that has provided consultation on the proposed UDS. JSI has worked with HRSA staff in developing and reviewing the instructions and materials for grantees, and has provided valuable technical and analytical expertise.
Arthur Stickgold
John Snow Inc.
44 Farnsworth Street
Boston, MA 02210
Tel: 617-482-0617
Email: [email protected]
| File Type | application/msword |
| Author | Seleda.Perryman |
| Last Modified By | ctac |
| File Modified | 2012-02-09 |
| File Created | 2012-02-09 |