With the guidance of the ASPE Project Officer, CHSRP staff developed a provisional list of databases for inclusion in the inventory. CHSRP staff then consulted with staff in the DHHS, SSA, VA to identify the key databases that either focus specifically on HIV positive populations or that address issues related to HIV. Federal staff or contractors responsible for development and maintenance of those databases were initially consulted during either face-to-face meetings or telephone or electronic discussions. During the initial discussions, we identified any additional databases that should be considered for inclusion in the inventory, as well as addressed the characteristics of the databases previously identified.
After developing a provisional list of databases, we developed a set of inventory items that were used to review and describe the databases. These items include:
- Purpose of the database;
- Study design used to collect the data;
- Nature of the data collected;
- Unit(s) of analysis;
- Data collection methods;
- General attributes (e.g., time frame covered by the database, size of the database and its various components, medium used to store the database);
- Major data constructs and key data elements within those constructs;
- Strengths and weaknesses of the study design and database, including methods used to validate data collection;
- Gaps in the data collected and factors leading to the gaps; and
- Feasibility of linking databases or using multiple databases to address questions, as well as specific restrictions on allowable uses of individual databases.
CHSRP staff consulted with key federal staff and their contractors during face-to-face or telephone meetings in which each of these elements were examined. Federal staff and their contractors also provided written materials related to the databases that further described these elements. To supplement these consultations, CHSRP staff retrieved written materials from Internet web sites related to the databases. The materials obtained by the CHSRP included: source materials stating the purpose of the databases and planned applications, data collection instruments, database documentation (e.g., file layouts, data dictionaries, etc.), evaluations of the data collection methods used to create the databases, and reports and other published or unpublished materials demonstrating the uses of these databases.
CHSRP staff conducted bibliographic searches for each database to identify journal articles, meeting abstracts, or other published materials that document how the databases have been or are being used to address policy, clinical, administrative, or other topics. The citations included for most of the databases are illustrative rather than an exhaustive list of published articles and reports, due to the large volume of materials published about the databases in this inventory.
CHSRP staff convened an expert panel of HIV researchers and representatives of federal agencies, philanthropies, and industry in July 1999 to discuss potential areas of multi-agency collaboration and new research strategies. Results of the expert panel meeting were used to identify additional databases for inclusion in the inventory, gaps in knowledge that might be addressed through new studies or linkage of exiting databases, and new research topics.
Using the materials gathered in the consultations and bibliographic search, CHSRP staff prepared a summary of each database. The summaries included in this report should be considered synopses of each database rather than a comprehensive review of the database and its attributes. Many of the databases included in the inventory have complex study designs and extensive data sets that cannot be fully captured in this report. The reports and websites that are cited offer more detailed information.
Following completion of the consultations and the materials review, CHSRP staff excluded some databases from the inventory. Some of these databases were surveys that had been conducted on a specific policy topic or undertaken in the mid-1990s, and for which ongoing data collection was not planned. Several other databases were analysis files derived from the databases already included in this inventory. Federal staff then reviewed the inventory for completeness, accuracy, and timeliness.