Providing patient-friendly educational materials alongside laboratory test results reduced patient messaging by only 15 messages per week and did not produce clinically meaningful changes in communication patterns, according to a quality improvement study from Vanderbilt University Medical Center that was recently published in JAMA Network Open.
In the interrupted time series analysis, the researchers evaluated 829,902 test results reviewed by 205,139 patients during 2024. Following implementation of educational formats in August, the proportion of results followed by patient-initiated messages within 24 hours decreased from 17.5% to 17.3%, a statistically nonsignificant change. When stratified by care setting, tests ordered in primary care showed a statistically significant 0.8 percentage-point decrease in messaging. However, microbial, antibody, and polymerase chain reaction (PCR) tests ordered outside primary care demonstrated increased messaging.
The intervention targeted tests with historically high message volumes: basic metabolic panel, comprehensive metabolic panel (CMP) including liver function tests (LFT), complete blood cell count (CBC), thyroid panel or thyrotropin tests, urine microalbumin tests, and selected microbial, antibody, and PCR tests. Health literacy experts collaborated with specialists in internal medicine, laboratory medicine, and clinical informatics to develop content presented through the MyChart-based patient portal. Among patients who reviewed results, 130,284 (64%) were female and 74,852 (37%) were male, with a mean age of 51 years. Most viewed results (531,823; 64%) were flagged as abnormal, and 575,582 results (69%) were reviewed by patients before their physicians. Among reviewed results, 144,623 (17%) prompted patient messages within 24 hours. Conversely, 12,433 unreviewed results (7%) had patient-initiated messages within 24 hours, suggesting messaging occurred for unrelated needs.
Specific test categories showed variable responses. For primary care orders, statistically significant decreases occurred in messaging following review of CBCs, CMP and LFTs, thyroid panels and thyrotropin levels, and urine microalbumin tests. When stratified by result status, abnormal CMP and LFT results showed decreased messaging, as did normal thyroid panel and thyrotropin test results. Normal microbial, antibody, and PCR tests demonstrated increased messaging.
The educational content included test descriptions, basic interpretation with graphic representation of results and reference ranges, and notes indicating follow-up from the clinical team for concerning findings. Implementation followed the 21st Century Cures Act requirement for immediate electronic access to test results, which has been associated with patients viewing more than 45% of test results before ordering physicians and doubling of patient-initiated messages since implementation in 2021.
Multivariable segmented logistic regression models, with standard errors clustered by patient, controlled for age group, sex, ethnicity, race, preferred language, insurance type, time enrolled in portal at result release, and whether patients reviewed results before physicians. Researchers evaluated messaging behavior from January 1 through December 31, 2024, and intervention deployed on August 7.
The Vanderbilt University Medical Center Institutional Review Board approved study procedures and granted informed consent waiver. The analysis followed Standards for Quality Improvement Reporting Excellence and Reporting of Studies Conducted Using Observational Routinely Collected Health Data reporting guidelines.
Disclosures can be found in the study.
Source: JAMA Network Open