Table 2 Examples of Hybrid 1, 2, and 3 studies with recommendations for additional categories (shaded)

Name

Systems of Support to Increase Colorectal Cancer Screening and Follow-up (SOS)

Strategies and Opportunities to STOP Colon Cancer in Priority Populations (STOP)

Benefit for Increasing Colorectal Cancer Screening in Priority Populations (BeneFIT)

Research aims

To compare the effectiveness of an EHR-linked automated mailed and stepped intensity CRC screening program to usual care

To compare the Adoption, Reach, Implementation, and Maintenance of a clinic-based EHR-embedded mailed program to usual care

To evaluate implementation of mailed FIT programs by two health insurance plans

Research questions

Primary question: Will an automated mailed CRC screening program increase screening uptake?

Implementation questions: What is the incremental cost-effectiveness of the stepped intensity program?

Is the program acceptable to patients and how could it be improved?

Primary question: Compared to usual care in safety net clinics can a mailed fecal testing program be adapted and implemented in safety net clinics and will it increase CRC screening uptake in vulnerable eligible adults and compared to usual care in safety net?

Co-primary question: What contextual factors influence adoption, reach, implementation, and maintenance of the program?

Primary question: Can different implementation strategies be used to increase the adoption, reach, effectiveness, and maintenance of a mailed FIT program? What do the programs cost? What contextual factors (inner and outer setting) influence program implementation?

Secondary question: What is the FIT return rate among eligible health plan enrollees?

Unit of randomization (or comparison)

Patient (patient-level randomized trial)

Clinic (clinic-level randomized trial)

Naturalistic experiment (no randomization) with comparative outcomes

Comparison condition

Usual care

Usual care

Two different implementation models, no usual care comparator

Sampling frames

Patients age-eligible not current for CRC screening, limited number of exclusions

Clinics with sufficient numbers of age-eligible patients not current for CRC screening, almost no exclusion

Two Medicaid/Medicare health plans and their enrollees within two states - age eligible and not current for CRC, almost no exclusions

Program design

Research team designed

Designed through collaboration between the EHR vendor, clinics, and research team

Health plan designed, with the researchers serving as consultants

Program delivery

Research team delivered the mailings. The research team closely supervised clinical staff and the study paid for their time. Program fidelity was monitored; variation minimized.

Delivered by clinics with training assistance and implementation facilitation (plan, do, see, act cycle) by the research team. Program delivery fidelity varied

Delivered by the health plans with the research team providing consultative assistance. The health plans managed program components and program fidelity

Evaluation method

Quantitative and summative

Mixed methods: quantitative and qualitative, formative and summative

Mixed methods: quantitative and qualitative, formative and summative

Measures

Comparative colorectal cancer screening rates and cost-effectiveness

FIT screening rates overall and variation by clinic implementation and patient characteristics, RE-AIM measures, variation in costs by clinic, qualitative assessments of factors related to clinic success and challenges

RE-AIM measures comparing the two implementation models. Costs of implementation. Comparisons of FIT test return rates across the two implementation models

Potential design challenges

Verbal consent required, leading to decreased Reach and external validity

Variation in implementation fidelity. IRB was not a challenge: single institutional review board, waiver of consent

Quasi-experimental design might lead to bias. Implementation and effectiveness outcomes were not directly comparable between the two health plans (no randomization, no control comparison)