The rate at which clinically relevant genomic tests are being developed is accelerating for chronic adult-onset in areas such as cardiology, psychiatry and oncology. These test results can affect medical management and long-term health. Consistent with this trend, VA is increasing its delivery of genomic services for diagnostic and predictive purposes. Comparative effectiveness research of genomic service delivery models is critical to ensure that VA develops, implements and disseminates genomic services that are both effective and veteran-centered.
The objective of this pilot study was to adapt and augment existing measures of genomics care process and patient outcomes to address the unique characteristics and needs of Veteran receiving genomic medical services. Specifically, the proposed study had four aims: (1) Identify the key process steps and patient outcomes of VA genomic services, (2) Develop two data collection instruments to measure those (a) process characteristics and (b) outcome dimensions, (3) Rigorously test those instruments for comprehension and ease of administration, and (4) Finalize the instruments for use in a future comparative effectiveness study of genomics services.
This study used classic measurement development methods and procedures, including in-depth in-person provider interviews and patient focus groups as well as phone interviews to identify clinically relevant domains of genomics service processes and patient outcomes. Cognitive interviewing was used to assess potential respondents' understanding of the resulting instruments.
Two convenience samples were used: 1) VA-based providers, primarily physicians, who ordered genomic tests for VA patients and had done so at least once within the past 12 months. 2) Veterans who received genomic services from the VABHS-based geneticist. Patients who elected to forego genomic testing, and patients who received only pharmacogenomic testing, were excluded.
We also obtained input from our Genomics Expert Advisory Panel via round-table discussions about the processes and outcomes of genomic services that should be measured in VA. The panel included three VA-based genomics experts and one geneticist from a private healthcare system highly respected for its work in quality assessment and improvement.
Qualitative data from the focus groups and interviews were transcribed and coded by two individuals both trained in clinical genetics and qualitative analysis. The Analytic Induction approach to analysis of the data was used. This iterative approach involves a deductive analysis followed by an inductive analysis to identify new domains. We initially defined a preliminary set of codes based on a literature review and expert panel input. Additional codes describing new issues were identified and integrated as the analysis proceeded. The coders analyzed each transcript independently and then reviewed the results together, coming to consensus on the themes and subsequent domains identified as important by the study participants.
Based on these analyses, one preliminary measurement tool for patients was developed. Given the complexity of the care processes involved, separate versions of the provider survey were developed for genomics and non-genomics providers. Cognitive interviews to assess the comprehension of potential respondents were conducted for all three tools. Findings from the cognitive interviews were compiled and reviewed by the research team, and revised versions were created. This process was repeated for the patient outcome measurement tool.
The test-retest reliability of the genomics expert version of the provider survey was assessed by administering that instrument to 2 VA-based genomics providers twice at least two weeks apart. A minimum correlation coefficient of 0.7 between the two sets of responses was regarded as indicative of instrument's stability.
A total of 19 patients (12 women) participated in the initial focus groups to inform the measurement tool's development. Seven were clinically symptomatic; 10 were referred for predictive testing and two for carrier testing. Several themes emerged. For example, principal reasons for pursuing genomics testing were to diagnose illness, reduce anxiety about disease risk, and concern for children future health. Providing choices regarding medical management and lifestyle behaviors was considered valuable, as was the time allocated for the counseling session. An additional seven patients participated in cognitive interviews conducted to evaluate the draft instruments.
Eight providers representing five different disciplines were interviewed. All respondents used clinical evaluations and personal/family histories to identify patients for referral. The extent of genetic counseling, as well as the test ordering and approval process, varied widely. Coordination between the provider, the pathologist and the send-out lab was identified as an important facilitator of this process. There was confusion as to what services could be provided to non-veteran family members.
The development of instruments to measure process and patient outcomes is a necessary first step to conducting studies to describing the variability of current practice, assessing the quality of VA genomics services, and identify best practices. For example, using these tools future studies could characterize and compare genomic services delivery models and potentially develop a method of triaging patients to the best mode based on patient characteristics.
The qualitative data collected during this pilot suggest that the delivery and accessibility of VA-based genomic services could benefit from having more genomic experts on staff, standardizing the genomic testing approval and ordering process, and clarification regarding what services can be provided to non-veteran family members.
The option to pursue genetic testing and counseling, when medically indicated, appears to contribute to the patient's quality of life by providing a sense of greater empowerment and control over their health and lifestyle choices. Veteran satisfaction with the steps in the genetic testing process, with the support provided to non-veteran family members, and with outcomes such as patient empowerment should be considered when assessing the value of and the quality of genomics services.
External Links for this Project
Grant Number: I01HX000424-01
None at this time.