Table 1.
Umbrella review domain and summary of quantitative findings statement (SRa source) | Strength of the evidence according to the GRADE-URb criteriac | |
Patient characteristics | ||
|
Patients with better controlled diabetes are more likely to enroll or use a portal as compared to other patients with diabetes [14,31,35]. | Moderate |
|
Patients with private insurance in the US context are more likely to enroll or use a portal [14,15,31,34,35]. | Moderate |
|
Patients with higher illness(es) burden or need are more likely to enroll or use a portal [14,15,31,34]. | Moderate |
|
White people are more likely to enroll or use a portal [14,15,31,32,34]. | Moderate |
|
Middle-aged people (≤65 years) are more likely to enroll or use a portal [14,17,31,32,34]. | Moderate |
|
People who have a higher income are more likely to enroll or use a portal [14,15,31,35]. | Moderate |
|
Males with diabetes are more likely to enroll or use portal as compared with females with diabetes [14,35]. | Moderate |
|
Patients with higher health literacy are more likely to enroll or use a portal [14,17,30]. | Low |
|
Females are more likely to access online information and use a portal [17,31,34]. | Low |
|
People who have a higher education level are more likely to enroll in and use a patient portal [14,31,35]. | Low |
Patient-related facilitators | ||
|
Patients are more likely to register and use a portal after portal-related education and training [32]. | Moderate |
Patient satisfaction | ||
|
Patients who use patient portals report higher satisfaction with communication, treatment, medications, and care [16,31,34,35]. | Moderate |
Behavioral effects | ||
|
Use of patient portals can increase adherence, mostly medication adherence across different patient populations [16,17,31,33,34]. | Moderate |
|
Use of patient portals can improve screening, vaccinations, examinations, and/or care across different patient populations [31,34,35]. | Moderate |
|
Use of patient portals can improve visit preparation and communication and information sharing between patients and providers [14,16,31,34]. | Low |
Service utilization effects | ||
|
Health care provider’s workload related to contacts and messaging does not change with patient portal adoption [34]. | Moderate |
|
Patients’ access to social support and mental health and testing services does not change with portal use [31,33]. | Moderate |
|
Hospitalization rates do not change with patient portal use [16,31,34]. | Low |
|
Emergency department visits do not change with patient portal use [16,31,34]. | Low |
|
Phone or messaging volume received by health care providers does not change with patient portal use [16,17,31,34]. | Low |
|
Patient portal use results in an increase in office, primary care, specialist, outpatient, or after-hour visits [15-17,31,34,35]. | Low |
|
Patient portal use does not reduce no-show rates [17,34]. | Low |
Clinical outcomes | ||
|
There is improvement in HbA1cd levels for patients with diabetes who use patient portals [15-17,31]. | Moderate |
|
There is improvement in LDLe, HDLf, cholesterol, or lipids for patients with diabetes who use patient portals [15,16,31,35]. | Low |
|
There is no change in systolic and diastolic blood pressure for patients with diabetes or hypertension who use patient portals [16,31,35]. | Low |
|
Psychosocial, cognitive function, BMI, symptom stability, and depression and anxiety status does not change across multiple patient populations who use patient portals [16,17,31]. | Low |
Patient-oriented outcomes | ||
|
Patient empowerment and self-efficacy scores do not change with portal use [16,31]. | Low |
aSR: systematic review.
bGRADE-UR: Grading of Recommendations, Assessment, Development, and Evaluations at the Level of an Umbrella Review.
cIndicates the strength of the evidence and was calculated based on study limitations, directness, consistency, precision, and reporting of bias. The ratings are from high, moderate, and low. Any statements we evaluated as insufficient were moved to the supporting evidence tables in Multimedia Appendix 4.
dHbA1c: hemoglobin A1c.
eLDL: low-density lipoprotein.
fHDL: high-density lipoprotein.