Randomized Communication Nudges to Increase Primary Care Engagement in Medicaid

Those in the both the treatment arm and control arm receive the standard new member MaineCare enrollment packet. Those in the treatment arm also receive a simplified mailer approximately 7 days later that assigns new members a primary care provider (PCP). Providers are selected from a list of MaineCare-enrolled who are accepting new patients and who are within 30 miles of a new enrollee’s home, when possible. Enrollees in the treatment arm also receive a prepaid mailer to notify MaineCare if they do not want to see their assigned provider because they already have a preferred PCP. The streamlined mailer also encourages members to make an appointment for a check-up with their new PCP and emphasizes the low-cost and value of primary care. Treatment members also receive a reminder in the mail 14 days after the initial enrollment packet which contains the name and contact information of the assigned PCP as well as instructions on how to choose a different PCP, if desired. Those in the control arm only receive the standard enrollment packet without the additional 7-day and 14-day mailers, and without an assigned PCP.

2021-03-16

2021-12-17

Primary care visit within 90 days of enrollment in MaineCare
Primary care within 365 days of enrollment in MaineCare
Number of primary care visits in the first 365 days after enrollment in MaineCare
Emergency department visit within 365 days of enrollment in MaineCare, classified by primary-care treatable and preventable/avoidable
Number of emergency department visits within 365 days of enrollment in MaineCare, classified by primary-care treatable and preventable/avoidable
Inpatient hospitalization within 365 days of enrollment in MaineCare
Number of inpatient hospitalizations within 365 days of enrollment in MaineCare
Ambulatory-care sensitive admissions within 365 days of enrollment in MaineCare
Number of ambulatory-care sensitive admissions within 365 days of enrollment in MaineCare
per year
Annual MaineCare enrollment duration
Any loss of MaineCare coverage per year

Ambulatory-care sensitive admissions will be based on the Billings NYU classification system and Agency for Healthcare Research and Quality (AHRQ) Prevention Quality Indicators.1,2 Classification of emergency department visits will be based on the NYU ED algorithm and the Johnson (2017) “patch” that extends this algorithm to include more ICD codes.3 This approach classifies ED visits into one of the following four categories: (1) nonemergent; (2) emergent, primary care treatable; (3) emergent, ED care needed, but preventable/avoidable; and (4) emergent, ED care needed, not preventable/avoidable.

1. Billings J, Anderson GM, Newman LS. Recent findings on preventable hospitalizations. Health Aff (Millwood). 1996;15(3):239-249. doi:10.1377/hlthaff.15.3.239
2. AHRQ QI: Prevention Quality Indicators Overview. Accessed November 21, 2023. https://qualityindicators.ahrq.gov/measures/pqi_resources
3. Johnston KJ, Allen L, Melanson TA, Pitts SR. A “Patch” to the NYU Emergency Department Visit Algorithm. Health Services Research. 2017;52(4):1264. doi:10.1111/1475-6773.12638

Rate of primary care visits per year
Rate of Ambulatory-care sensitive admissions per year
Rate of emergency department visits per year
Rate of inpatient hospitalization per year
Annual flu shots
Annual A1C screening (excluding anyone with diabetes)
Annual cholesterol screening
Mammogram in women 50-64
Chlamydia testing 19-24
A1c for enrollees with diabetes
Microalbumin for enrollees with diabetes
Receipt of statin prescription for enrollees with a coronary artery disease or diabetes diagnosis

The experimental design is an individual-level randomization among newly-enrolled members of the MaineCare Primary Care Case Management (PCCM) Program. The protocol is as follows:
• Every Tuesday by 10am during the trial period, MaineCare administrators will pull a report of new PCCM members. Those who were enrolled in PCCM less than one year ago, who received standard enrollment packets more than 1 day before the report, and who were under the age of 19 are removed.
• Member IDs are then anonymized and sent to academic partner for randomization in batch form, corresponding to each week of enrollment.
• The academic partners will send the list back to MaineCare by 5pm with randomization status for each new member.
• Before day 7 following the enrollment packet mailing, the active Medicaid provider database will be used to assign those in the treatment group a PCP who is taking new patients that is nearest to their home. If multiple providers qualify across participants, assignment will be rotated across providers to balance new patient load.
• Information is populated in mailers and sent to enrollees in treated arm by business day 7.
• Those who have not yet contacted MaineCare are sent a 14-day reminder card.
• Member responses via phone, email, or mail are tracked for the duration of the intervention period.

Not available

Randomization conducted in office by a computer program.

Individual

No

0

Total N: 4,212

Treatment: 862
Control: 3,350

We have powered this study to detect an increase of 4 percentage points on our outcome of a primary care visit within one year of enrollment, which is of clinical and policy significance in the Medicaid expansion evaluation literature as well as in a meta-analysis of the effect of mail-based interventions to increase primary care attendance.4–6 Using a two-sample test for a different between proportions and a two-sided significance level of 5%, a sample of 862 members in the treatment group and 3,350 members in the control group will be sufficient to detect a difference of four percentage points with 89% power. 4. Baicker K, Taubman SL, Allen HL, et al. The Oregon Experiment — Effects of Medicaid on Clinical Outcomes. New England Journal of Medicine. 2013;368(18):1713-1722. doi:10.1056/NEJMsa1212321 5. Miller S, Wherry LR. Health and Access to Care during the First 2 Years of the ACA Medicaid Expansions. N Engl J Med. 2017;376(10):947-956. doi:10.1056/NEJMsa1612890 6. Stubbs ND, Geraci SA, Stephenson PL, Jones DB, Sanders S. Methods to reduce outpatient non-attendance. Am J Med Sci. 2012;344(3):211-219. doi:10.1097/MAJ.0b013e31824997c6