This trial is taking place in Los Angeles, CA at clinics within the UCLA Health System.

Despite the Advisory Committee on Immunization Practices (ACIP) recommendation in 2010 that all people above 6 months of age should receive an annual flu vaccine, vaccination rates remain low: at 6m-4.9 yrs. (70%), 5-17.9 yrs. (56%), 18-64.9 yrs. (38%), and >65 yrs. (63%). The investigators will assess the effectiveness and cost-effectiveness of 1, 2, 3 MyChart R/R messages as compared to the standard of care control (no messages).

Annual epidemics of influenza cause substantial morbidity in the U.S. with up to 40,000 deaths/year and many hospitalizations, emergency and outpatient visits, and significant costs. Concerns about pandemic influenza elevate the need to prevent flu outbreaks.

Numerous studies, including Cochrane or systematic reviews, and reports by the CDC and the Task Force on Community Preventive Services, highlight 4 evidence-based strategies to raise child and adult influenza vaccination rates: 1) increase patient demand by reminder-recall or education, 2) expand patient access to influenza vaccinations (e.g., flu vaccine clinics), 3) implement provider strategies such as prompts or standing orders, and 4) use societal strategies (e.g., reducing costs).

Reminder/recall (R/R), sent by phone, mail or other modality, can improve child and adult influenza vaccination rates. However, fewer than one-fifth of pediatric or adult primary care practices utilize patient R/R. Barriers are lack of finances, personnel, and algorithms to identify eligible patients.

A technological breakthrough that might overcome these barriers involves patient portals-- secure, web-based communication systems, embedded within electronic health records (EHRs), for patients and providers to communicate with each other via email and the internet. Portals are used by about half of Americans and half of UCLA patients. Portals can theoretically improve upon phone, mail or text R/R by adding information such as web links, videos, or images, allowing patients to schedule their own visits, and linking to the medical chart to customize messages.

For this randomized control trial, the intent is to evaluate the impact of patient portal (MyChart) reminder recalls - either 1, 2, or 3 reminders versus the standard of care control group, specifically in relation to raising influenza vaccination rates among UCLA Health System's primary care patients aged 6 months and older.

The proposed design of this 4-arm RCT:

1. Standard of care control (no messages)
2. Up to 1 portal R/R messages
3. Up to 2 portal R/R messages
4. Up to 3 portal R/R messages

Hypothesis 1: >1 portal R/R will increase vaccination rates vs. no R/R. Hypothesis 2: More R/R messages will raise vaccination rates (3R/R > 2R/R > 1R/R > 0R/R).

For the primary analysis, only the data from the randomly selected index patients (1 index patient per household) will be included.

For relevant study arms, the first R/R message will be sent in October 2018.

Inclusion Criteria:

• A patient within the UCLA Health System identified by the system as a primary care patient per an internal algorithm, with a documented primary care visit within the last 3 years as of 8/1/18.

Exclusion Criteria:

• A patient not identified by the UCLA Health System's internal algorithm as a primary care patient.

Aggregate results will be prepared for publication at the end of the trial.

Patients were eligible for study participation if they were at least 6 months of age at the start of intervention (ie, eligible for influenza vaccination), a primary care patient within the UCLA Health System according to the health system's criteria, and an active patient portal user (≥1 log-in during the past 12 months), other than initial account log-in.

This pragmatic, 4-arm randomized clinical trial was performed from October 1, 2018, to April 2, 2019, across the UCLA (University of California, Los Angeles) health care system. A total of 164 205 patients in 52 primary care practices who had used the patient portal within 12 months were included. Adverse events were not assessed/monitored.

We conducted a trial of whether reminder sent via the portal impacted influenza vaccination rates; we did not collect adverse event or serious event information.