IntroductionDelays in cancer diagnosis arose from the commencement of non-pharmaceutical interventions (NPI) introduced in the UK in March 2020 in response to the COVID-19 pandemic. Our earlier work predicted this will lead to approximately 3620 avoidable deaths for four major tumour types (breast, bowel, lung, and oesophageal cancer) in the next 5 years. Here, using national population-based modelling, we estimate the health and economic losses resulting from these avoidable cancer deaths. We also compare these with the impact of an equivalent number of COVID-19 deaths to understand the welfare consequences of the different health conditions.MethodsWe estimate health losses using quality-adjusted life years (QALYs) and lost economic productivity using the human capital (HC) approach. The analysis uses linked English National Health Service (NHS) cancer registration and hospital administrative datasets for patients aged 15–84 years, diagnosed with breast, colorectal, and oesophageal cancer between 1st Jan to 31st Dec 2010, with follow-up data until 31st Dec 2014, and diagnosed with lung cancer between 1st Jan to 31st Dec 31 2012, with follow-up data until 31st Dec 2015. Productivity losses are based on the estimation of excess additional deaths due to cancer at 1, 3 and 5 years for the four cancer types, which were derived from a previous analysis using this dataset. A total of 500 random samples drawn from the total number of COVID-19 deaths reported by the Office for National Statistics, stratified by gender, were used to estimate productivity losses for an equivalent number of deaths (n = 3620) due to SARS-CoV-2 infection.ResultsWe collected data for 32,583 patients with breast cancer, 24,975 with colorectal cancer, 6744 with oesophageal cancer, and 29,305 with lung cancer. We estimate that across the four site-specific cancers combined in England alone, additional excess cancer deaths would amount to a loss of 32,700 QALYs (95% CI 31,300-34,100) and productivity losses of £103.8million GBP (73.2–132.2) in the next five years. For breast cancer, we estimate a loss of 4100 QALYS (3900–4400) and productivity losses of £23.2 m (18.2–28.6); for colorectal cancer, 15,000 QALYS (14,100–16,000) lost and productivity losses of £35.7 m (22.4–48.7); for lung cancer 10,900 QALYS (9,900–11,700) lost and productivity losses of £38.3 m (14.0–59.9) for lung cancer; and for oesophageal cancer, 2700 QALYS (2300–3,100) lost and productivity losses of £6.6 m (–6 to –17.6). In comparison, the equivalent number of COVID-19 deaths caused approximately 21,450 QALYs lost, as well as productivity losses amounting to £76.4 m (73.5–79.2).ConclusionPremature cancer deaths resulting from diagnostic delays during the first wave of the COVID-19 pandemic in the UK will result in significant economic losses. On a per-capita basis, this impact is, in fact, greater than that of deaths directly attributable to COVID-19. These results emphasise the importance of robust evaluation of the trade-offs of the wider health, welfare and economic effects of NPI to support both resource allocation and the prioritisation of time-critical health services directly impacted in a pandemic, such as cancer care.
ABSTRACT Background Regulators and oncology healthcare providers are increasingly interested in using observational studies of real‐world data (RWD) to complement clinical evidence from randomized controlled trials for informed decision‐making. To generate valid evidence, RWD studies must be carefully designed to avoid systematic biases. The clone‐censor‐weight (CCW) method has been proposed to address immortal time and other time‐related biases. Methods The objective of this manuscript is to de‐mystify the CCW method for cancer researchers by describing and presenting its core components in an accessible and digestible format, using visualizations and examples from cancer‐relevant studies. The CCW method has been applied in diverse settings, including investigations of the effects of surgery within a certain time after cancer diagnosis, the continuation of annual screening mammography, and chemotherapy duration on survival. Results The method handles complex data wherein the treatment group to which an individual belongs is unknown at the start of follow‐up. The three steps of the CCW method involve cloning or duplicating the patient population and assigning one clone to each treatment strategy, artificially censoring the clones when their observed data are inconsistent with the assigned strategy and weighting the cloned and censored population to address selection bias created by the artificial censoring. Conclusions The CCW method is a powerful tool for designing RWD studies in cancer that are free from time‐related biases and successfully, to the extent possible, emulate features of a randomized clinical trial.
Abstract Background The effectiveness of COVID-19 monoclonal antibody and antiviral therapies against severe COVID-19 outcomes is unclear. Initial benefit was shown in unvaccinated patients and before the Omicron variant emerged. We used the OpenSAFELY platform to emulate target trials to estimate the effectiveness of sotrovimab or molnupiravir, versus no treatment. Methods With the approval of NHS England, we derived population-based cohorts of non-hospitalised high-risk individuals in England testing positive for SARS-CoV-2 during periods of dominance of the BA.1 (16/12/2021-10/02/2022) and BA.2 (11/02/2022-21/05/2022) Omicron sublineages. We used the clone-censor-weight approach to estimate the effect of treatment with sotrovimab or molnupiravir initiated within 5 days after positive test versus no treatment. Hazard ratios (HR) for COVID-19 hospitalisation or death within 28 days were estimated using weighted Cox models. Results Of the 35,856 [BA.1 period] and 39,192 [BA.2 period] patients, 1,830 [BA.1] and 1,242 [BA.2] were treated with molnupiravir and 2,244 [BA.1] and 4,164 [BA.2] with sotrovimab. The estimated HRs for molnupiravir versus untreated were 1.00 (95%CI: 0.81;1.22) [BA.1] and 1.22 (0.96;1.56) [BA.2]; corresponding HRs for sotrovimab versus untreated were 0.76 (0.66;0.89) [BA.1] and 0.92 (0.79;1.06) [BA.2]. Interpretation Compared with no treatment, sotrovimab was associated with reduced risk of adverse outcomes after COVID-19 in the BA.1 period, but there was weaker evidence of benefit in the BA2 period. Molnupiravir was not associated with reduced risk in either period. Funding UKRI, Wellcome Trust, MRC, NIHR and HDRUK.
Cancer mortality has been examined among ethnic South Asian migrants in England and Wales, but not by generation of migration.Using South Asian mortality records, identified by a name-recognition algorithm, and census information, age-standardised rates among South Asians, and South Asian vs non-South Asian rate ratios, were calculated.All-cancer rates in ethnic South Asians were half of those in non-South Asians in first-generation (all-cancer-standardised mortality ratio (SMR) in males 0.51 and in females 0.56) and subsequent-generation South Asians (SMR in males 0.43 and in females 0.36). The higher mortality in first-generation South Asians for liver (both sexes), oral cavity and gallbladder cancer (females), particularly marked among Bangladeshis, was reduced in subsequent generations.
Rationale: Whether patients with coronavirus disease (COVID-19) may benefit from extracorporeal membrane oxygenation (ECMO) compared with conventional invasive mechanical ventilation (IMV) remains unknown. Objectives: To estimate the effect of ECMO on 90-day mortality versus IMV only. Methods: Among 4,244 critically ill adult patients with COVID-19 included in a multicenter cohort study, we emulated a target trial comparing the treatment strategies of initiating ECMO versus no ECMO within 7 days of IMV in patients with severe acute respiratory distress syndrome (PaO2/FiO2 < 80 or PaCO2 ⩾ 60 mm Hg). We controlled for confounding using a multivariable Cox model on the basis of predefined variables. Measurements and Main Results: A total of 1,235 patients met the full eligibility criteria for the emulated trial, among whom 164 patients initiated ECMO. The ECMO strategy had a higher survival probability on Day 7 from the onset of eligibility criteria (87% vs. 83%; risk difference, 4%; 95% confidence interval, 0–9%), which decreased during follow-up (survival on Day 90: 63% vs. 65%; risk difference, −2%; 95% confidence interval, −10 to 5%). However, ECMO was associated with higher survival when performed in high-volume ECMO centers or in regions where a specific ECMO network organization was set up to handle high demand and when initiated within the first 4 days of IMV and in patients who are profoundly hypoxemic. Conclusions: In an emulated trial on the basis of a nationwide COVID-19 cohort, we found differential survival over time of an ECMO compared with a no-ECMO strategy. However, ECMO was consistently associated with better outcomes when performed in high-volume centers and regions with ECMO capacities specifically organized to handle high demand.
Abstract This study aimed to evaluate the association between thyroid dysfunction and breast cancer risk. We included 239,436 females of the UK Biobank cohort. Information on thyroid dysfunction, personal and family medical history, medications, reproductive factors, lifestyle, and socioeconomic characteristics was retrieved from baseline self‐reported data and hospital inpatient databases. Breast cancer diagnoses were identified through population‐based registries. We computed Cox models to estimate hazard ratios (HRs) of breast cancer incidence for thyroid dysfunction diagnosis and treatments, and examined potential confounding and effect modification by comorbidities and breast cancer risk factors. In our study, 3,227 (1.3%) and 20,762 (8.7%) women had hyper‐ and hypothyroidism prior to the baseline. During a median follow‐up of 7.1 years, 5,326 (2.2%) women developed breast cancer. Compared to no thyroid dysfunction, there was no association between hypothyroidism and breast cancer risk overall (HR = 0.93, 95% confidence interval (CI): 0.84–1.02, 442 cases), but we found a decreased risk more than 10 years after hypothyroidism diagnosis (HR=0.85, 95%CI 0.74–0.97, 226 cases). There was no association with hyperthyroidism overall (HR=1.08, 95%CI 0.86–1.35, 79 cases) but breast cancer risk was elevated among women with treated hyperthyroidism (HR=1.38, 95%CI: 1.03–1.86, 44 cases) or aged 60 years or more at hyperthyroidism diagnosis (HR=1.74, 95%CI: 1.01–3.00, 113 cases), and 5–10 years after hyperthyroidism diagnosis (HR=1.58, 95%CI: 1.06–2.33, 25 cases). In conclusion, breast cancer risk was reduced long after hypothyroidism diagnosis, but increased among women with treated hyperthyroidism. Future studies are needed to determine whether the higher breast cancer risk observed among treated hyperthyroidism could be explained by hyperthyroidism severity, type of treatment or aetiology.
In cancer epidemiology using population-based data, regression models for the excess mortality hazard is a useful method to estimate cancer survival and to describe the association between prognosis factors and excess mortality. This method requires expected mortality rates from general population life tables: each cancer patient is assigned an expected (background) mortality rate obtained from the life tables, typically at least according to their age and sex, from the population they belong to. However, those life tables may be insufficiently stratified, as some characteristics such as deprivation, ethnicity, and comorbidities, are not available in the life tables for a number of countries. This may affect the background mortality rate allocated to each patient, and it has been shown that not including relevant information for assigning an expected mortality rate to each patient induces a bias in the estimation of the regression parameters of the excess hazard model. We propose two parametric corrections in excess hazard regression models, including a single-parameter or a random effect (frailty), to account for possible mismatches in the life table and thus misspecification of the background mortality rate. In an extensive simulation study, the good statistical performance of the proposed approach is demonstrated, and we illustrate their use on real population-based data of lung cancer patients. We present conditions and limitations of these methods, and provide some recommendations for their use in practice.
BackgroundThe National Health Service (NHS) cancer plan for England was published in 2000, with the aim of improving the survival of patients with cancer. By contrast, a formal cancer strategy was not implemented in Wales until late 2006. National data on cancer patient survival in England and Wales up to 2007 thus offer the opportunity for a first formal assessment of the cancer plan in England, by comparing survival trends in England with those in Wales before, during, and after the implementation of the plan.MethodsWe analysed population-based survival in 2·2 million adults diagnosed with one of 21 common cancers in England and Wales during 1996–2006 and followed up to Dec 31, 2007. We defined three calendar periods: 1996–2000 (before the cancer plan), 2001–03 (initialisation), and 2004–06 (implementation). We estimated year-on-year trends in 1-year relative survival for patients diagnosed during each period, and changes in those trends between successive periods in England and separately in Wales. Changes between successive periods in mean survival up to 5 years after diagnosis were analysed by country and by government office region of England. Life tables for single year of age, sex, calendar year, deprivation category, and government office region were used to control for background mortality in all analyses.Findings1-year survival in England and Wales improved for most cancers in men and women diagnosed during 1996–2006 and followed until 2007, although not all trends were significant. Annual trends were generally higher in Wales than in England during 1996–2000 and 2001–03, but higher in England than in Wales during 2004–06. 1-year survival for patients diagnosed in 2006 was over 60% for 12 of 17 cancers in men and 13 of 18 cancers in women. Differences in 3-year survival trends between England and Wales were less marked than the differences in 1-year survival. North–South differences in survival trends for the four most common cancers were not striking, but the North West region and Wales showed the smallest improvements during 2001–03 and 2004–06.InterpretationThe findings indicate slightly faster improvement in 1-year survival in England than in Wales during 2004–06, whereas the opposite was true during 2001–03. This reversal of survival trends in 2001–03 and 2004–06 between England and Wales is much less obvious for 3-year survival. These different patterns of survival suggest some beneficial effect of the NHS cancer plan for England, although the data do not so far provide a definitive assessment of the effectiveness of the plan.FundingOffice for National Statistics (contract NT-04/2355A); Cancer Research UK (programme grant C1336/A5735).
Summary In cancer epidemiology using population-based data, regression models for the excess mortality hazard is a useful method to estimate cancer survival and to describe the association between prognosis factors and excess mortality. This method requires expected mortality rates from general population life tables: each cancer patient is assigned an expected (background) mortality rate obtained from the life tables, typically at least according to their age and sex, from the population they belong to. However, those life tables may be insufficiently stratified, as some characteristics such as deprivation, ethnicity, and comorbidities, are not available in the life tables for a number of countries. This may affect the background mortality rate allocated to each patient, and it has been shown that not including relevant information for assigning an expected mortality rate to each patient induces a bias in the estimation of the regression parameters of the excess hazard model. We propose two parametric corrections in excess hazard regression models, including a single-parameter or a random effect (frailty), to account for possible mismatches in the life table and thus misspecification of the background mortality rate. In an extensive simulation study, the good statistical performance of the proposed approach is demonstrated, and we illustrate their use on real population-based data of lung cancer patients. We present conditions and limitations of these methods and provide some recommendations for their use in practice.
Abstract Background Large and complex population-based cancer data are becoming broadly available, thanks to purposeful linkage between cancer registry data and health electronic records. Aiming at understanding the explanatory power of factors on cancer survival, the modelling and selection of variables need to be understood and exploited properly for improving model-based estimates of cancer survival. Method We assess the performances of well-known model selection strategies developed by Royston and Sauerbrei and Wynant and Abrahamowicz that we adapt to the relative survival data setting and to test for interaction terms. Results We apply these to all male patients diagnosed with lung cancer in England in 2012 ( N = 15,688), and followed-up until 31/12/2015. We model the effects of age at diagnosis, tumour stage, deprivation, comorbidity and emergency presentation, as well as interactions between age and all of the above. Given the size of the dataset, all model selection strategies favoured virtually the same model, except for a non-linear effect of age at diagnosis selected by the backward-based selection strategies (versus a linear effect selected otherwise). Conclusion The results from extensive simulations evaluating varying model complexity and sample sizes provide guidelines on a model selection strategy in the context of excess hazard modelling.
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)
