Accounting for the impact of the Covid-19 pandemic on routine data resources: development and application of novel methodology for the analysis and reporting of population-based cancer statistics

Aims of the research The aims of the project are: O1. To assess the overall impact of the COVID-19 pandemic on: the timing and stage of cancer diagnoses; treatment allocation for cancer patients; and non-cancer death rates in the general population. O2. To ensure that we can fairly compare and monitor cancer survival and incidence over periods of time including years during or following the pandemic. O3. To make sure that research studies that rely on routine data resources can continue to make fair comparisons despite the impact of the pandemic. O4. To communicate cancer survival statistics in a patient-focussed and meaningful manner. Background to the research Monitoring the number of cancer cases and patient survival is vital. This information is used in the planning and delivery of health services, and in the evaluation of diagnostics and treatment at the population level. Comparisons are made between risk factor groups such as age, deprivation, sex, regions/countries and stage. These comparisons help to identify groups to target with the aim of reducing differences in incidence and survival. During the pandemic routine delivery of the health service was severely disrupted, as cancer screening services were paused, the availability of GPs, oncologists and other healthcare providers was reduced, and there were delays to accessing cancer treatments. Similar impacts were seen across many disease areas including cardiovascular disease, kidney disease and diabetes. As a result, there have been large changes to the data due to potentially worse cancer survival and increased non-cancer death rates in the general population. These population death rates, which have been increased due to Covid-19, are needed to estimate common cancer survival measures. Ignoring these changes can result in bias and excluding affected data is wasteful. Hence, it is important to develop statistical methods to ensure we make best use of the data. Design and methods used To address the aims, I will apply the following methods: M1. I will search for previously published research regarding the scale of the impact of the pandemic on cancer-related clinical practice and routine data resources. I will conduct simulation studies to scope the impact of these changes on cancer survival. I will investigate the impact of the reduced cancer diagnosis services. M2. I will evaluate methods to adjust for the decrease in reported cancer cases to fairly report survival metrics. I will develop approaches to alter population-mortality files to enable the estimation of unbiased cancer survival metrics. M3. I will develop methods to quantify the contribution of a) delayed diagnosis b) suboptimal treatment and c) varying competing mortality on cancer survival. M4. I will develop two interactive webtools: one to provide patient-focussed metrics of cancer survival for a range of cancer types and one for researchers to illustrate the sensitivity analysis. I will develop a workshop to disseminate the findings of this project. Patient/service user, carer and public involvement I have identified a National Cancer Research Institute (NCRI) PPI representative to be named PPI advisory group lead. This patient has been involved in the development of this project through review of this plain English summary, and the scientific abstract. I will approach further patients to form the remainder of the advisory group. We will continue discussions regarding patients' experiences of cancer diagnosis and treatment during the Covid-19 pandemic, and any concerns they may have to further shape the project. Dissemination I will disseminate my findings from this project by publishing my work in journals and presenting at international conferences. I will develop a dissemination workshop and two webtools, one patient-focussed and the other researcher-focussed, to improve the communication of reported cancer survival metrics.