ATTACHMENT K - Procedures for Accessing Case Records and Abstracting Data

ATTACHMENT K:

PROCEDURES FOR ACCESSING CASE RECORDS AND ABSTRACTING DATA

PROCEDURES FOR LOCATING CASE RECORDS AND ABSTRACTING DATA

Contact with Tribes/ITOs. As part of the Outreach effort, NORC staff will contact the ITO by email and telephone to explain the data needs for the study. We will hold a teleconference with each ITO to discuss the approval process for obtaining the participant list and access to the case record data. We will delineate in detail the specific data elements that are necessary to develop the sample frame for the participant profile and survey. In addition, we will ask about site-specific quality assurance procedures to gain a better understanding of the quality of the information recorded in the case records, how information is updated, and where the records are stored.

Negotiate a Data Sharing Agreement. Once the requirements for gaining access have been established, we will negotiate a Data Sharing Agreement with each ITO in order to obtain case record data through a secure means. The Data Sharing Agreement will also address tribal ownership of the de-identified data once the study is completed.¹ We will ensure that the Data Sharing Agreement incorporates language on the future use of the datasets to create a Public Use File. Upon executing the Data Sharing Agreement, NORC will contact the ITO and make arrangements for conducting the case record review and onsite visit.

Obtain list of FDPIR Participating Households. NORC will contact each ITO to obtain a count of participating households for the reference month. We will also request a list of participating households to develop the sampling frame (as described below).

Implement site specific sampling strategy. For each ITO we will develop a site-specific sampling frame based on a list of current participating households provided by the ITO for the reference month. The study team will provide detailed instructions and technical assistance for any ITO that wishes to draw the sample. For tribes with electronic records, sample selection may occur prior to the site visit. For tribes lacking electronic records, drawing the sample will need to occur as part of preparation for the site visit by FDPIR staff or onsite by the study team. Detailed procedures will be developed for both electronic and paper case records and provided to each ITO.

Obtain case record data. Our working assumption is that the majority of the ITOs do not have automated case record files and we plan to conduct onsite visits with those Tribes to abstract the data. Three scenarios for conducting the case record review are detailed below. These scenarios will be refined as we learn more about record-keeping for each ITO in the sample during our outreach and discuss their preferred procedures for the case record review.

(1) For most sites, case record review will involve abstracting information onsite from eligibility forms that are completed by applicants to FDPIR. At the USDA-Tribal Leader consultations with the Tribes in the study sample, we learned that very few Tribes maintain electronic records that would be amenable to electronic data transfer. Tribes/ITOs will draw the sample in advance (based on specifications provided by the study team). Research staff trained in data abstraction procedures will conduct the case record review. They will conduct onsite visits to abstract the data in FDPIR offices, using the case record review template provided in Attachment B1a. These data will be entered into a secure database for aggregation and analysis.

(2) For those sites that maintain electronic data files (to-date, we are aware of only one such site in the sample), research staff will consult with the Tribe to establish a data format and a secure method of data transfer, which may include shipping encrypted data on an external hard drive or uploading it to a secure file transfer program (FTP) link. NORC will be able to accept the data files in the most common formats—ASCII, SAS, STATA, SPSS, and MS Excel. After receiving the data files, NORC will examine the data for quality assurance using routine data cleaning procedures and conducting test data runs. Any data quality issues that might be encountered during this process will be reported back to the ITO and resolved immediately.

(3) For very small sites, the research team will offer FDPIR programs the option of copying and transmitting redacted copies of case records through a secure and traceable carrier. Research staff will then abstract the data from these records to obtain the necessary data elements. Per NORC data security procedures, all paper case records will be locked in secure files and destroyed upon completion of the abstraction process. Based on the consultations, we expect that few Tribes will select this option.

Create case record review file. Data abstracted from the ITO-specific case records will be compiled into an Excel workbook using a common template.

1 We will inquire as to whether ITOs would like to be directly involved in any stage of the sampling process (e.g., drawing the site-specific sample). If they express interest, we will organize training and other skill-building activities to actively engage them and to ensure that Tribal capacity-building efforts are met through this data collection.

1