A second wave of crisis in the NHS after the COVID-19 pandemic
What is the aim of the study and why is it important?
The COVID-19 pandemic could affect the population in several dimensions. The virus will have direct effects on people who catch it, but also indirect effects on the rest of the population due to the disruption it is causing to life and access to healthcare. In this study, we will estimate specifically the indirect effects of COVID-19 on “day-to-day” healthcare service utilization. Using a unique dataset (QResearch) that links primary care data, secondary care data (Hospital Episode Statistics) and cancer registry data and mortality data, we will investigate how the pandemic has disrupted the utilisation of healthcare services for routine and emergency visits, and the consequences of this disruption for patients. Our main outcome measures will relate to how the disruption in NHS hospitals affected the health of patients measured across several dimensions, namely: the number of visits to A&E or specialists; hospital admissions; drug prescriptions; deterioration of comorbidities; and mortality. Using HES data we will identify the group of patients whose appointments were rescheduled or cancelled. We will also construct a control group looking at the people who were being treated before the pandemic so that we can estimate the likely pattern of patients whose treatment may have been disrupted if the pandemic had not occurred. In addition, we will specifically look at patients with a cancer diagnosis to see if they suffered serious effects from the disruption to hospital care and access.
How is the research being done?
We will initially identify all patients whose appointments were delayed/cancelled and those diagnosed with cancer (registered with QResearch general practices during the period) between 1st January 2015 and the December 2020 (or the latest date for which data is available).
Considering first a group of patients with all conditions who had appointments delayed/cancelled, we will first produce descriptive statistics for the variables included in our dataset. We will cross-tabulate summaries of the use of health care services for individuals categorised by age, gender, region, ethnicity, and local area deprivation after the start of the pandemic.
Additionally, we will produce statistical models that can be used to estimate, the additional time that patients were forced to wait for new appointments; increases in mortality, and the causes of death; and the number of drug prescriptions and tests, plus number of visits to GPs, A&E, and hospital admissions.
We will compare the modelled durations for people whose appointments were cancelled/postponed before and during the pandemic (October 2019-July 2020) versus patients whose appointments were cancelled/delayed between 2015 and 2018. The idea is to measure how a disruption in secondary care will affect patients and the use of healthcare. Furthermore, this analysis will help us to identify if there is some discrimination in terms of age, gender, or deprivation areas during the pandemic.
After this broader analysis we will focus on patients with cancer. We will estimate their outcomes survival curves, median survival times, and hazard ratios of the diagnosis, stage of cancer, mortality, referral time, and waiting time. Again, given our interest in inequality, we will consider the differences across groups with different demographic characteristics to discover any inequality in care and health outcomes. The impact of the pandemic on cancer patients is an important issue because cancer patients have high levels of secondary hospital use and the government should adopt policies that protect this particularly vulnerable group of people.
Lead Applicant Organisation Name
University of Oxford
Location of research
University of Oxford
Date on which research approved
Project reference ID
Generic ethics approval reference
Are all data accessed are in anonymised form?
Brief summary of the dataset to be released (including any sensitive data)
We will include patients aged 0 to 100 years registered with practices contributing to the QResearch database. Our cases will be selected using HES outpatient data as those individuals whose appointments were not attended between January 2015 and July 2020. We will select all the patients whose appointments where not attended, distinguishing between appointments cancelled by the patients themselves and those cancelled by the hospitals. We will then follow these patients to measure their use of health care services in the months following that unattended appointment. We will also place a special focus on patients with a diagnosis of cancer to see if they were particularly affected by the interruptions to treatment.
We are using data covering primary care from EMIS, secondary care from HES, cancer registry and demographic data from the ONS.
COVID-19 Research Response Fund, University of Oxford
Public Benefit Statement
Dr Catia Nicodemo, University of Oxford
Professor Julia Hippisley-Cox, University of Oxford
Dr Jakub Lonsky, University of Liverpool
Dr Stuart Redding, University of Oxford
Professor Stavros Petrou, University of Oxford
Dr Cesar Garriga, University of Oxford
Trusted Research Environment (TRE)