Table 1 Quality Domain assessments

Domain objective

Assess the capture of participants to the registry

Assess the capture of data within specified cohorts

Assess the accuracy of placement of patients into correct cohorts; identify issues with data capture and entry (e.g. transcription errors)

Assess the accuracy of patient-specific data records as a true reflection of individual clinical data and reported outcomes

Assess the reliability of aggregated cohort data against benchmarks determined from evidence based literature

Method of assessment

Ratio of treatment records in the registry to number of patients eligible for participation in the registry.

Calculated by checking archived consult lists containing patients assigned to a cohort against treatment records stored in the electronic database.

Ratio of data captured for patients’ treatment records compared to the total number of variables within the CDS for each cohort.

Calculated by dividing the number of patients at time (x) with data (i) available, by the number of patients eligible for collection of (x)(i).

Assessed for all treatment records entered into the registry.

All treatment records were retrieved and diagnosis was checked against cohort inclusion and exclusion criteria.

Any cohort assignment that did not match the diagnosis was flagged and the contributing surgeon notified.

Outlier analysis utilising quartiles method was performed on current age, age at surgery, height and weight.

Assessed for all treatment records that had a diagnosis entered into the registry.

Validate individual patient data records to original data / patient submitted forms.

Determined by comparing source data and data transcribed to the registry software.

Data validation performed by a registry custodian member who was independant to data entry.

Assessed for all treatment records entered into the registry.

The highest quality evidence of appropriate patient outcomes were used to benchmark aggregated cohort PROMs data.

Assessed for all treatment records with PROMS data captured to the registry.

Audits performed during pilot period (July 2017 - Aug 2018)

12

6

6

4

4

Benchmark

90% [9, 13]

90% [9, 13]

95% set internally by registry custodian team

90% [9, 13]

Varied depending on PROM