Risks of using medical record and administrative data for prognostic models

Knowledge and insights from a maturing international clinical quality registry

Since 2017, the TrueNTH Global Registry (TNGR) has aimed to drive improvement in patient outcomes for individuals with localized prostate cancer by collating data from healthcare institutions across 13 countries. As TNGR matures, a systematic evaluation of existing processes and documents is necessary to evaluate whether the registry is operating as intended. The main supporting documents: protocol and data dictionary, were comprehensively reviewed in a series of meetings over a 10-month period by an international working group. In parallel, individual consultations with local institutions regarding a benchmarking quality-of-care report were conducted. Four consensus areas for improvement emerged: updating operational definitions, appraisal of the recruitment process, refinement of data elements, and improvement of data quality and reporting. Recommendations presented were drawn from our collective experience and accumulated knowledge in operating an international registry. These can be readily generalized to other health-related reporting programs beyond clinical registries.

The Upper Gastrointestinal Cancer Registry (UGICR): a clinical quality registry to monitor and improve care in upper gastrointestinal cancers

PURPOSE

The Upper Gastrointestinal Cancer Registry (UGICR) was developed to monitor and improve the quality of care provided to patients with upper gastrointestinal cancers in Australia.

PARTICIPANTS

It supports four cancer modules: pancreatic, oesophagogastric, biliary and primary liver cancer. The pancreatic cancer (PC) module was the first module to be implemented, with others being established in a staged approach. Individuals are recruited to the registry if they are aged 18 years or older, have received care for their cancer at a participating public/private hospital or private clinic in Australia and do not opt out of participation.

FINDINGS TO DATE

The UGICR is governed by a multidisciplinary steering committee that provides clinical governance and oversees clinical working parties. The role of the working parties is to develop quality indicators based on best practice for each registry module, develop the minimum datasets and provide guidance in analysing and reporting of results. Data are captured from existing data sources (population-based cancer incidence registries, pathology databases and hospital-coded data) and manually from clinical records. Data collectors directly enter information into a secure web-based Research Electronic Data Capture (REDCap) data collection platform. The PC module began with a pilot phase, and subsequently, we used a formal modified Delphi consensus process to establish a core set of quality indicators for PC. The second module developed was the oesophagogastric cancer (OGC) module. Results of the 1 year pilot phases for PC and OGC modules are included in this cohort profile.

FUTURE PLANS

The UGICR will provide regular reports of risk-adjusted, benchmarked performance on a range of quality indicators that will highlight variations in care and clinical outcomes at a health service level. The registry has also been developed with the view to collect patient-reported outcomes (PROs), which will further add to our understanding of the care of patients with these cancers.

Monitoring quality of care for patients with pancreatic cancer: a modified Delphi consensus

BACKGROUND

Best practise care optimises survival and quality of life in patients with pancreatic cancer (PC), but there is evidence of variability in management and suboptimal care for some patients. Monitoring practise is necessary to underpin improvement initiatives. We aimed to develop a core set of quality indicators that measure quality of care across the disease trajectory.

METHODS

A modified, three-round Delphi survey was performed among experts with wide experience in PC care across three states in Australia. A total of 107 potential quality indicators were identified from the literature and divided into five areas: diagnosis and staging, surgery, other treatment, patient management and outcomes. A further six indicators were added by the panel, increasing potential quality indicators to 113. Rated on a scale of 1-9, indicators with high median importance and feasibility (score 7-9) and low disagreement (<1) were considered in the candidate set.

RESULTS

From 113 potential quality indicators, 34 indicators met the inclusion criteria and 27 (7 diagnosis and staging, 5 surgical, 4 other treatment, 5 patient management, 6 outcome) were included in the final set.

CONCLUSIONS

The developed indicator set can be applied as a tool for internal quality improvement, comparative quality reporting, public reporting and research in PC care.

Lung cancer prognostic index: a risk score to predict overall survival after the diagnosis of non-small-cell lung cancer

INTRODUCTION

Non-small-cell lung cancer outcomes are poor but heterogeneous, even within stage groups. To improve prognostic precision we aimed to develop and validate a simple prognostic model using patient and disease variables.

METHODS

Prospective registry and study data were analysed using Cox proportional hazards regression to derive a prognostic model (hospital 1, n=695), which was subsequently tested (Harrell's c-statistic for discrimination and Cox-Snell residuals for calibration) in two independent validation cohorts (hospital 2, n=479 and hospital 3, n=284).

RESULTS

The derived Lung Cancer Prognostic Index (LCPI) included stage, histology, mutation status, performance status, weight loss, smoking history, respiratory comorbidity, sex, and age. Two-year overall survival rates according to LCPI in the derivation and two validation cohorts, respectively, were 84, 77, and 68% (LCPI 1: score⩽9); 61, 61, and 42% (LCPI 2: score 10-13); 33, 32, and 14% (LCPI 3: score 14-16); 7, 16, and 5% (LCPI 4: score ⩾15). Discrimination (c-statistic) was 0.74 for the derivation cohort, 0.72 and 0.71 for the two validation cohorts.

CONCLUSIONS

The LCPI contributes additional prognostic information, which may be used to counsel patients, guide trial eligibility or design, or standardise mortality risk for epidemiological analyses.