The goal of this study is to improve the use of preventive health services by implementing the AWV Practice Redesign Toolkit to stimulate AWV visit use in diverse practices across the United States. This is a stepped wedge cluster randomized controlled trial. The intervention will be implemented in a total of 16 small-to-mid-size primary care practices (at the practice level) with clinicians and staff. Eight practices will participate in the main clinical trial, and 8 in a replication study. The main clinical trial (with community-based practices) and replication study (with Federally Qualified Health Center practices) will be conducted simultaneously. Electronic health record (EHR) data extractions will be used to collect outcomes in a population cohort of patients. Semi-structured interviews will be conducted with clinicians/staff and patients to assess intervention implementation. The investigators hypothesize that the toolkit implementation will increase AWV visit use and consequently, use of preventive health services.

This is a multi-site, stepped wedge cluster randomized controlled trial to evaluate the effect of the AWV Practice Redesign Toolkit on AWV and preventive health services use. The study will take place in 16 primary care practices around the United States. Eight practices will participate in the main clinical trial, and 8 in a replication study. The main clinical trial (with community-based practices) and replication study (with Federally Qualified Health Center practices) will be conducted simultaneously. Recruitment occurs at the level of the practice. The investigators hypothesize that the intervention will increase uptake of AWVs and use of preventive health services.

The stepped wedge cluster design is a pragmatic study design that includes an initial control period in which no clusters receive the intervention. Randomized clusters cross to the intervention condition at regular intervals, and there is a period at the end of the study where all clusters have been exposed to the intervention. Data are collected throughout the study so that each cluster contributes to both the control and intervention periods. The intervention will be implemented sequentially over 16 months, with 4 clusters of 4 practices each switching from control to intervention every 4 months between months 4-20. By month 20 of the study all practices will have completed intervention implementation. Each cluster will have a 4 month transition period during intervention implementation.

In addition to the procedures described above, the investigators will collect qualitative semi-structured interview data on a subset of patients (up to n=10), and clinicians/staff (up to n=10) from each practice to assess fidelity to the intervention.

Outcomes will be assessed through EHR data extractions, which will be led by the DARTNet Institute (co-investigator Wilson Pace). Data will be sent from the DARTNet Institute to the American Academy of Family Physicians National Research Network (AAFP NRN) for cleaning, de-identification and creation of flags. The AAFP NRN has established contracts and agreements with the DARTNet Institute that enables them to access DARTNet's data. The AAFP NRN will deliver a limited dataset to investigators at the University of California, Los Angeles for analyses.

There are no inclusion or exclusion criteria based on sex/gender, race, and/or ethnicity.

Inclusion Criteria:

• have Medicare coverage
• aged 50 or older
• has at least one visit to the practice in the past 12 months
• not deceased

Practice eligibility criteria:

• care for patients with Medicare insurance

The research team will engage the University of California Curation Center (UC3) to serve as the data repository for all components of the project. The research team will work with UC3 to curate and deliver to the curation center definition of data elements extracted from EHR and source data in a standardized, de-identified format. We also will include statistical codes for analyses in the form of R, SAS or STATA files.

The DRYAD system within the UC3 provides a platform to publish the metadata concerning the project that researchers globally can access. This system will serve as the data repository in which we will house the data management plan for the source data as well as metadata that describes the source data and code books of the underlying data elements, with record counts such that outside researchers can have a full understanding of what is housed in the system.

Within 6 months of publication of the main effects paper. Data will be permanently archived and available.

Where possible data will be shared in an open file format, so proprietary software is not required to view or use the files. All datasets will be indexed by the Thomson-Reuters Data Citation Index, Scopus, and Google Dataset Search. Each dataset is given a unique Digital Object Identifier or DOI. Entering the DOI URL in any browser will take the user to the dataset's landing page. Dryad also provides a faceted search and browse capability for direct discovery.

Dryad has implemented the Make Data Count project recommendations. This means that that views and downloads on each dataset landing page are standardized against the COUNTER Code of Practice for Research Data. Within this framework, Dryad also exposes all related citations to a dataset on the landing page. These are updated each time a new citation from an article or other source has been published.