International CR Registry

The user sub-committee has a formal site on-boarding process. Participating sites will be encouraged to enter data on all patients meeting inclusion / exclusion criteria to reduce bias. Prospective / concurrent data collection will be encouraged, to support quality improvement activities.

The company Dendrite will host the registry and ensure data security. Privacy protections will be in place, and no identifying information such as names will be entered into the registry. Data entered from sites into the registry is secured through encryption added to secure HTTP, done using Transport Layer Security. Stored data is also encrypted using InterSystems Cache database software.

There is an ICRR program survey assessing structural CR indicators. The registry variable list for patient-level data was developed based on recommendations of the Core Outcome Set-STAndards for Development (COS-STAD), through a delphi process. The final variable list and data dictionary can be found here: https://globalcardiacrehab.com/ICRR-Governance.

There are 10 program-reported variables and 16 patient-reported variables (some variables are assessed at multiple time points), assessed pre-program, post-program (dependent upon duration of each program) and each year from initial assessment (until patients expire or are unable to complete the assessments, which is ascertained by the program, or patient opt-out/withdrawal). Sites can arrange electronic data capture for the program-reported variables to contribute data to the registry, rather than manual electronic entry.

Descriptive statistics (i.e. means and standard deviations for continuous variables and frequencies with percentages for categorical variables) will be used to describe the quality of care (outcomes below). For testing change in patient outcomes pre to post-program, paired t-tests will also be used for continuous variables and chi-square test for categorical variables.

Univariate and multivariate approaches will be used to identify factors associated with outcomes, including the use of guideline-indicated therapies. Associations between patient characteristics and outcomes tested via logistic regression will be reported with odds ratios and 95% confidence intervals. Associations will be considered statistically significant when 2-sided α is <0.05. Appropriate statistical corrections for repeated measures will be performed. To limit the influence of confounding, multivariate adjustment will be used, including propensity weighting, or other techniques. Risk adjustment (e.g., socioeconomic status, indication, comorbidities, age) will be undertaken.

The registry has a data quality dashboard which displays by site completeness of data entry at each assessment point, to help reduce bias. The registry has in-built minimum and maximum values for continuous variables to reduce data entry errors. The data dictionary clearly defines each variable, and new sites are trained prior to start. The registry is being usability tested prior to launch, and piloted in Qatar. The research sub-committee will oversee a data quality process; Data audits may be supported by a registry database analyst / professional statistician.

Patient perspectives were considered in the development of the variable list and outcomes. A registry page for patients is in development (https://globalcardiacrehab.com/ICRR-for-Patients). Patients have a voice on ICRR's user sub-committee. Patients providing data pre and post-program may receive a lay summary of their progress. Lay summaries of annual registry reports may be shared with interested patients.

Inclusion criteria

SITES:

• CR program (defined in intervention section)
• in low-resource setting (i.e., low or middle-income country according to World Bank, or in region within a higher-income country where there is under-development of CR related to financial resources, lack of healthcare system resources, lack of patient and provider awareness, and/or patient disadvantage [e.g., limited social resources, geographic barriers])

Exclusion criteria

SITES:

-program cannot enter data in English

PATIENTS:

• not capable to understand the registry information letter for cognitive or mental reasons (although they would also likely be excluded from CR).
• not proficient in the language of the local ethics-approved registry information letter /consent form.