Mixture and Non-mixture Cure Models for Health Technology Assessment: What You Need to Know.

There is increasing interest in the use of cure modelling to inform health technology assessment (HTA) due to the development of new treatments that appear to offer the potential for cure in some patients. However, cure models are often not included in evidence dossiers submitted to HTA agencies, and they are relatively rarely relied upon to inform decision-making. This is likely due to a lack of understanding of how cure models work, what they assume, and how reliable they are. In this tutorial we explain why and when cure models may be useful for HTA, describe the key characteristics of mixture and non-mixture cure models, and demonstrate their use in a range of scenarios, providing Stata code. We highlight key issues that must be taken into account by analysts when fitting these models and by reviewers and decision-makers when interpreting their predictions. In particular, we note that flexible parametric non-mixture cure models have not been used in HTA, but they offer advantages that make them well suited to an HTA context when a cure assumption is valid but follow-up is limited.

Evaluation of Flexible Parametric Relative Survival Approaches for Enforcing Long-Term Constraints When Extrapolating All-Cause Survival.

OBJECTIVES: Parametric models are used to estimate the lifetime benefit of an intervention beyond the range of trial follow-up. Recent recommendations have suggested more flexible survival approaches and the use of external data when extrapolating. Both of these can be realized by using flexible parametric relative survival modeling. The overall aim of this article is to introduce and contrast various approaches for applying constraints on the long-term disease-related (excess) mortality including cure models and evaluate the consequent implications for extrapolation. METHODS: We describe flexible parametric relative survival modeling approaches. We then introduce various options for constraining the long-term excess mortality and compare the performance of each method in simulated data. These methods include fitting a standard flexible parametric relative survival model, enforcing statistical cure, and forcing the long-term excess mortality to converge to a constant. We simulate various scenarios, including where statistical cure is reasonable and where the long-term excess mortality persists. RESULTS: The compared approaches showed similar survival fits within the follow-up period. However, when extrapolating the all-cause survival beyond trial follow-up, there is variation depending on the assumption made about the long-term excess mortality. Altering the time point from which the excess mortality is constrained enables further flexibility. CONCLUSIONS: The various constraints can lead to applying explicit assumptions when extrapolating, which could lead to more plausible survival extrapolations. The inclusion of general population mortality directly into the model-building process, which is possible for all considered approaches, should be adopted more widely in survival extrapolation in health technology assessment.

Modeling the multi-state natural history of rare diseases with heterogeneous individual patient data: A simulation study.

Multi-state survival models are used to represent the natural history of a disease, forming the basis of a health technology assessment comparing a novel treatment to current practice. Constructing such models for rare diseases is problematic, since evidence sources are typically much sparser and more heterogeneous. This simulation study investigated different one-stage and two-stage approaches to meta-analyzing individual patient data (IPD) in a multi-state survival setting when the number and size of studies being meta-analyzed are small. The objective was to assess methods of different complexity to see when they are accurate, when they are inaccurate and when they struggle to converge due to the sparsity of data. Biologically plausible multi-state IPD were simulated from study- and transition-specific hazard functions. One-stage frailty and two-stage stratified models were estimated, and compared to a base case model that did not account for study heterogeneity. Convergence and the bias/coverage of population-level transition probabilities to, and lengths of stay in, each state were used to assess model performance. A real-world application to Duchenne Muscular Dystrophy, a neuromuscular rare disease, was conducted, and a software demonstration is provided. Models not accounting for study heterogeneity were consistently out-performed by two-stage models. Frailty models struggled to converge, particularly in scenarios of low heterogeneity, and predictions from models that did converge were also subject to bias. Stratified models may be better suited to meta-analyzing disparate sources of IPD in rare disease natural history/economic modeling, as they converge more consistently and produce less biased predictions of lengths of stay.

Perils of Randomized Controlled Trial Survival Extrapolation Assuming Treatment Effect Waning: Why the Distinction Between Marginal and Conditional Estimates Matters.

OBJECTIVES: A long-term, constant, protective treatment effect is a strong assumption when extrapolating survival beyond clinical trial follow-up; hence, sensitivity to treatment effect waning is commonly assessed for economic evaluations. Forcing a hazard ratio (HR) to 1 does not necessarily estimate loss of individual-level treatment effect accurately because of HR selection bias. A simulation study was designed to explore the behavior of marginal HRs under a waning conditional (individual-level) treatment effect and demonstrate bias in forcing a marginal HR to 1 when the estimand is "survival difference with individual-level waning". METHODS: Data were simulated under 4 parameter combinations (varying prognostic strength of heterogeneity and treatment effect). Time-varying marginal HRs were estimated in scenarios where the true conditional HR attenuated to 1. Restricted mean survival time differences, estimated having constrained the marginal HR to 1, were compared with true values to assess bias induced by marginal constraints. RESULTS: Under loss of conditional treatment effect, the marginal HR took a value >1 because of covariate imbalances. Constraining this value to 1 lead to restricted mean survival time difference bias of up to 0.8 years (57% increase). Inflation of effect size estimates also increased with the magnitude of initial protective treatment effect. CONCLUSIONS: Important differences exist between survival extrapolations assuming marginal versus conditional treatment effect waning. When a marginal HR is constrained to 1 to assess efficacy under individual-level treatment effect waning, the survival benefits associated with the new treatment will be overestimated, and incremental cost-effectiveness ratios will be underestimated.

Flexible parametric methods for calculating life expectancy in small populations.

BACKGROUND: Life expectancy is a simple measure of assessing health differences between two or more populations but current life expectancy calculations are not reliable for small populations. A potential solution to this is to borrow strength from larger populations from the same source, but this has not formally been investigated. METHODS: Using data on 451,222 individuals from the Clinical Practice Research Datalink on the presence/absence of intellectual disability and type 2 diabetes mellitus, we compared stratified and combined flexible parametric models, and Chiang's methods, for calculating life expectancy. Confidence intervals were calculated using the Delta method, Chiang's adjusted life table approach and bootstrapping. RESULTS: The flexible parametric models allowed calculation of life expectancy by exact age and beyond traditional life expectancy age thresholds. The combined model that fit age interaction effects as a spline term provided less bias and greater statistical precision for small covariate subgroups by borrowing strength from the larger subgroups. However, careful consideration of the distribution of events in the smallest group was needed. CONCLUSIONS: Life expectancy is a simple measure to compare health differences between populations. The use of combined flexible parametric methods to calculate life expectancy in small samples has shown promising results by allowing life expectancy to be modelled by exact age, greater statistical precision, less bias and prediction of different covariate patterns without stratification. We recommend further investigation of their application for both policymakers and researchers.

Using temporal recalibration to improve the calibration of risk prediction models in competing risk settings when there are trends in survival over time.

We have previously proposed temporal recalibration to account for trends in survival over time to improve the calibration of predictions from prognostic models for new patients. This involves first estimating the predictor effects using data from all individuals (full dataset) and then re-estimating the baseline using a subset of the most recent data whilst constraining the predictor effects to remain the same. In this article, we demonstrate how temporal recalibration can be applied in competing risk settings by recalibrating each cause-specific (or subdistribution) hazard model separately. We illustrate this using an example of colon cancer survival with data from the Surveillance Epidemiology and End Results (SEER) program. Data from patients diagnosed in 1995-2004 were used to fit two models for deaths due to colon cancer and other causes respectively. We discuss considerations that need to be made in order to apply temporal recalibration such as the choice of data used in the recalibration step. We also demonstrate how to assess the calibration of these models in new data for patients diagnosed subsequently in 2005. Comparison was made to a standard analysis (when improvements over time are not taken into account) and a period analysis which is similar to temporal recalibration but differs in the data used to estimate the predictor effects. The 10-year calibration plots demonstrated that using the standard approach over-estimated the risk of death due to colon cancer and the total risk of death and that calibration was improved using temporal recalibration or period analysis.

Fair comparisons of cause-specific and relative survival by accounting for the systematic removal of patients from risk-sets.

BACKGROUND: In population-based cancer studies it is common to try to isolate the impact of cancer by estimating net survival. Net survival is defined as the probability of surviving cancer in the absence of any other-causes of death. Net survival can be estimated either in the cause-specific or relative survival framework. Cause-specific survival considers deaths from the cancer as the event of interest. Relative survival incorporates general population expected mortality rates to represent the other-cause mortality rate. Estimation approaches in both frameworks are impacted by the systematic removal of patients from the risk-set, commonly referred to as informative censoring in the cause-specific framework. In the relative survival framework, the Pohar Perme estimator combats the effect of this systematic removal of patients through weighting. When the two frameworks have been compared, informative censoring is rarely accounted for in the cause-specific framework. METHODS: We investigate the use of weighted cause-specific Kaplan-Meier estimates to overcome the impact of informative censoring and compared approaches to defining weights. Individuals remaining in the risk-set are upweighted using their predicted other-cause survival obtained through various model-based approaches. We also compare weights derived from expected mortality rates. We applied the approaches to US cancer registry data and conducted a simulation study. RESULTS: Using weighted cause-specific estimates provides a better estimate of marginal net survival. The unweighted Kaplan-Meier estimates have a similar bias to the Ederer II method for relative survival. Weighted Kaplan-Meier estimates are unbiased and similar to the Pohar Perme estimator. There was little variation between the several weighting approaches. CONCLUSION: In comparisons of cause-specific and relative survival, it is important to compare "like-with-like", therefore, a weighted approach should be considered for both frameworks. If researchers are interested in obtaining net measures in a cause-specific framework, then weighting is needed to account for informative censoring.

Survival Extrapolation Incorporating General Population Mortality Using Excess Hazard and Cure Models: A Tutorial.

BACKGROUND: Different parametric survival models can lead to widely discordant extrapolations and decision uncertainty in cost-effectiveness analyses. The use of excess hazard (EH) methods, which incorporate general population mortality data, has the potential to reduce model uncertainty. This review highlights key practical considerations of EH methods for estimating long-term survival. METHODS: Demonstration of methods used a case study of 686 patients from the German Breast Cancer Study Group, followed for a maximum of 7.3 y and divided into low (1/2) and high (3) grade cancers. Seven standard parametric survival models were fit to each group separately. The same 7 distributions were then used in an EH framework, which incorporated general population mortality rates, and fitted both with and without a cure parameter. Survival extrapolations, restricted mean survival time (RMST), and difference in RMST between high and low grades were compared up to 30 years along with Akaike information criterion goodness-of-fit and cure fraction estimates. The sensitivity of the EH models to lifetable misspecification was investigated. RESULTS: In our case study, variability in survival extrapolations was extensive across the standard models, with 30-y RMST ranging from 7.5 to 14.3 y. Incorporation of general population mortality rates using EH cure methods substantially reduced model uncertainty, whereas EH models without cure had less of an effect. Long-term treatment effects approached the null for most models but at varying rates. Lifetable misspecification had minimal effect on RMST differences. CONCLUSIONS: EH methods may be useful for survival extrapolation, and in cancer, EHs may decrease over time and be easier to extrapolate than all-cause hazards. EH cure models may be helpful when cure is plausible and likely to result in less extrapolation variability. HIGHLIGHTS: In health economic modeling, to help anchor long-term survival extrapolation, it has been recommended that survival models incorporate background mortality rates using excess hazard (EH) methods.We present a thorough description of EH methods with and without the assumption of cure and demonstrate user-friendly software to aid researchers wishing to use these methods.EH models are applied to a case study, and we demonstrate that EHs are easier to extrapolate and that the use of the EH cure model, when cure is plausible, can reduce extrapolation variability.EH methods are relatively robust to lifetable misspecification.

Impact of a Prior Cancer Diagnosis on Quality of Care and Survival Following Acute Myocardial Infarction: Retrospective Population-Based Cohort Study in England.

BACKGROUND: An increasing proportion of patients with cancer experience acute myocardial infarction (AMI). We investigated differences in quality of AMI care and survival between patients with and without previous cancer diagnoses. METHODS: A retrospective cohort study using Virtual Cardio-Oncology Research Initiative data. Patients aged 40+ years hospitalized in England with AMI between January 2010 and March 2018 were assessed, ascertaining previous cancers diagnosed within 15 years. Multivariable regression was used to assess effects of cancer diagnosis, time, stage, and site on international quality indicators and mortality. RESULTS: Of 512 388 patients with AMI (mean age, 69.3 years; 33.5% women), 42 187 (8.2%) had previous cancers. Patients with cancer had significantly lower use of ACE (angiotensin-converting enzyme) inhibitors/angiotensin receptor blockers (mean percentage point decrease [mppd], 2.6% [95% CI, 1.8-3.4]) and lower overall composite care (mppd, 1.2% [95% CI, 0.9-1.6]). Poorer quality indicator attainment was observed in patients with cancer diagnosed in the last year (mppd, 1.4% [95% CI, 1.8-1.0]), with later stage disease (mppd, 2.5% [95% CI, 3.3-1.4]), and with lung cancer (mppd, 2.2% [95% CI, 3.0-1.3]). Twelve-month all-cause survival was 90.5% in noncancer controls and 86.3% in adjusted counterfactual controls. Differences in post-AMI survival were driven by cancer-related deaths. Modeling improving quality indicator attainment to noncancer patient levels showed modest 12-month survival benefits (lung cancer, 0.6%; other cancers, 0.3%). CONCLUSIONS: Measures of quality of AMI care are poorer in patients with cancer, with lower use of secondary prevention medications. Findings are primarily driven by differences in age and comorbidities between cancer and noncancer populations and attenuated after adjustment. The largest impact was observed in recent cancer diagnoses (<1 year) and lung cancer. Further investigation will determine whether differences reflect appropriate management according to cancer prognosis or whether opportunities to improve AMI outcomes in patients with cancer exist.

Acute heart failure presentation, management, and outcomes in cancer patients: a national longitudinal study.

AIMS: Currently, little evidence exists on survival and quality of care in cancer patients presenting with acute heart failure (HF). The aim of the study is to investigate the presentation and outcomes of hospital admission with acute HF in a national cohort of patients with prior cancer. METHODS AND RESULTS: This retrospective, population-based cohort study identified 221 953 patients admitted to a hospital in England for HF during 2012-2018 (12 867 with a breast, prostate, colorectal, or lung cancer diagnosis in the previous 10 years). We examined the impact of cancer on (i) HF presentation and in-hospital mortality, (ii) place of care, (iii) HF medication prescribing, and (iv) post-discharge survival, using propensity score weighting and model-based adjustment. Heart failure presentation was similar between cancer and non-cancer patients. A lower percentage of patients with prior cancer were cared for in a cardiology ward [-2.4% age point difference (ppd) (95% CI -3.3, -1.6)] or were prescribed angiotensin-converting enzyme inhibitors or angiotensin receptor antagonists (ACEi/ARB) for heart failure with reduced ejection fraction [-2.1 ppd (-3.3, -0.9)] than non-cancer patients. Survival after HF discharge was poor with median survival of 1.6 years in prior cancer and 2.6 years in non-cancer patients. Mortality in prior cancer patients was driven primarily by non-cancer causes (68% of post-discharge deaths). CONCLUSION: Survival in prior cancer patients presenting with acute HF was poor, with a significant proportion due to non-cancer causes of death. Despite this, cardiologists were less likely to manage cancer patients with HF. Cancer patients who develop HF were less likely to be prescribed guideline-based HF medications compared with non-cancer patients. This was particularly driven by patients with a poorer cancer prognosis.

Comparing the number and length of primary care consultations in people with and without intellectual disabilities and health needs: observational cohort study using electronic health records.

BACKGROUND: In the United Kingdom, 15-min appointments with the general practitioner (GP) are recommended for people with complex health conditions, including intellectual disabilities and health needs, but we do not know whether this happens. AIMS: We compared number and length of primary care consultations (GP, nurse, other allied health, other) for people with and without intellectual disabilities and health needs. METHODS: Linked primary care data from the Clinical Practice Research Datalink (CPRD) in England were used to investigate face-to-face and telephone primary care consultations in 2017-2019. Health needs investigated were: epilepsy; incontinence; severe visual/hearing impairments; severe mobility difficulties; cerebral palsy; and percutaneous endoscopic gastrostomy feeding. Age and gender-standardized consultation rates per year (Poisson), duration of consultations, and the proportion of "long consultations" (≥15 min) were reported. RESULTS: People with intellectual disabilities (n = 7,794) had 1.9 times as many GP consultations per year as those without (n = 176,807; consultation rate ratio = 1.87 [95% confidence interval 1.86-1.89]). Consultation rates with nurses and allied healthcare professionals were also twice as high. Mean GP consultation time was 9-10 min regardless of intellectual disability/health need status. Long GP consultations were less common in people with intellectual disabilities (18.2% [17.8-18.7] vs. 20.9% [20.8-21.0]). Long consultations with practice nurses were more common in people with health needs, particularly severe visual loss. CONCLUSIONS: People with intellectual disabilities and/or health needs tend to have more, rather than longer, GP consultations compared with the rest of the population. We recommend further investigation into the role of practice nurses to support people with intellectual disabilities and health needs.

Estimating causal effects in the presence of competing events using regression standardisation with the Stata command standsurv.

BACKGROUND: When interested in a time-to-event outcome, competing events that prevent the occurrence of the event of interest may be present. In the presence of competing events, various estimands have been suggested for defining the causal effect of treatment on the event of interest. Depending on the estimand, the competing events are either accommodated or eliminated, resulting in causal effects with different interpretations. The former approach captures the total effect of treatment on the event of interest while the latter approach captures the direct effect of treatment on the event of interest that is not mediated by the competing event. Separable effects have also been defined for settings where the treatment can be partitioned into two components that affect the event of interest and the competing event through different causal pathways. METHODS: We outline various causal effects that may be of interest in the presence of competing events, including total, direct and separable effects, and describe how to obtain estimates using regression standardisation with the Stata command standsurv. Regression standardisation is applied by obtaining the average of individual estimates across all individuals in a study population after fitting a survival model. RESULTS: With standsurv several contrasts of interest can be calculated including differences, ratios and other user-defined functions. Confidence intervals can also be obtained using the delta method. Throughout we use an example analysing a publicly available dataset on prostate cancer to allow the reader to replicate the analysis and further explore the different effects of interest. CONCLUSIONS: Several causal effects can be defined in the presence of competing events and, under assumptions, estimates of those can be obtained using regression standardisation with the Stata command standsurv. The choice of which causal effect to define should be given careful consideration based on the research question and the audience to which the findings will be communicated.

Reference-Adjusted Loss in Life Expectancy for Population-Based Cancer Patient Survival Comparisons-with an Application to Colon Cancer in Sweden.

BACKGROUND: The loss in life expectancy, LLE, is defined as the difference in life expectancy between patients with cancer and that of the general population. It is a useful measure for summarizing the impact of a cancer diagnosis on an individual's life expectancy. However, it is less useful for making comparisons of cancer survival across groups or over time, because the LLE is influenced by both mortality due to cancer and other causes and the life expectancy in the general population. METHODS: We present an approach for making LLE estimates comparable across groups and over time by using reference expected mortality rates with flexible parametric relative survival models. The approach is illustrated by estimating temporal trends in LLE of patients with colon cancer in Sweden. RESULTS: The life expectancy of Swedish patients with colon cancer has improved, but the LLE has not decreased to the same extent because the life expectancy in the general population has also increased. When using a fixed population and other-cause mortality, that is, a reference-adjusted approach, the LLE decreases over time. For example, using 2010 mortality rates as the reference, the LLE for females diagnosed at age 65 decreased from 11.3 if diagnosed in 1976 to 7.2 if diagnosed in 2010, and from 3.9 to 1.9 years for women 85 years old at diagnosis. CONCLUSIONS: The reference-adjusted LLE is useful for making comparisons across calendar time, or groups, because differences in other-cause mortality are removed. IMPACT: The reference-adjusted approach enhances the use of LLE as a comparative measure.

Generating high-fidelity synthetic time-to-event datasets to improve data transparency and accessibility.

BACKGROUND: A lack of available data and statistical code being published alongside journal articles provides a significant barrier to open scientific discourse, and reproducibility of research. Information governance restrictions inhibit the active dissemination of individual level data to accompany published manuscripts. Realistic, high-fidelity time-to-event synthetic data can aid in the acceleration of methodological developments in survival analysis and beyond by enabling researchers to access and test published methods using data similar to that which they were developed on. METHODS: We present methods to accurately emulate the covariate patterns and survival times found in real-world datasets using synthetic data techniques, without compromising patient privacy. We model the joint covariate distribution of the original data using covariate specific sequential conditional regression models, then fit a complex flexible parametric survival model from which to generate survival times conditional on individual covariate patterns. We recreate the administrative censoring mechanism using the last observed follow-up date information from the initial dataset. Metrics for evaluating the accuracy of the synthetic data, and the non-identifiability of individuals from the original dataset, are presented. RESULTS: We successfully create a synthetic version of an example colon cancer dataset consisting of 9064 patients which aims to show good similarity to both covariate distributions and survival times from the original data, without containing any exact information from the original data, therefore allowing them to be published openly alongside research. CONCLUSIONS: We evaluate the effectiveness of the methods for constructing synthetic data, as well as providing evidence that there is minimal risk that a given patient from the original data could be identified from their individual unique patient information. Synthetic datasets using this methodology could be made available alongside published research without breaching data privacy protocols, and allow for data and code to be made available alongside methodological or applied manuscripts to greatly improve the transparency and accessibility of medical research.

Health Needs and Their Relationship with Life Expectancy in People with and without Intellectual Disabilities in England.

Health needs are common in people living with intellectual disabilities, but we do not know how they contribute to life expectancy. We used the Clinical Practice Research Datalink (CPRD) linked with hospital/mortality data in England (2017-2019) to explore life expectancy among people with or without intellectual disabilities, indicated by the presence or absence, respectively, of: epilepsy; incontinence; severe visual loss; severe visual impairment; severe mobility difficulties; cerebral palsy and PEG feeding. Life expectancy and 95% confidence intervals were compared using flexible parametric methods. At baseline, 46.4% (total n = 7794) of individuals with intellectual disabilities compared with 9.7% (total n = 176,807) in the comparison group had ≥1 health need. Epilepsy was the most common health need (18.7% vs. 1.1%). All health needs except hearing impairment were associated with shorter life expectancy: PEG feeding and mobility difficulties were associated with the greatest loss in life years (65-68% and 41-44%, respectively). Differential life expectancy attenuated but remained (≈12% life years lost) even after restricting the population to those without health needs (additional years expected to live at 10 years: 65.5 [60.3, 71.1] vs. 74.3 [73.8, 74.7]). We conclude that health needs play a significant role but do not explain all of the differential life expectancy experienced by people with intellectual disabilities.

Obtaining long-term stage-specific relative survival estimates in the presence of incomplete historical stage information.

BACKGROUND: Completeness of recording for cancer stage at diagnosis is often historically poor in cancer registries, making it challenging to provide long-term stage-specific survival estimates. Stage-specific survival differences are driven by differences in short-term prognosis, meaning estimated survival metrics using period analysis are unlikely to be sensitive to imputed historical stage data. METHODS: We used data from the Surveillance, Epidemiology, and End Results (SEER) Program for lung, colon and breast cancer. To represent missing data patterns in less complete registry data, we artificially inflated the proportion of missing stage information conditional on stage at diagnosis and calendar year of diagnosis. Period analysis was applied and missing stage at diagnosis information was imputed under four different conditions to emulate extreme imputed stage distributions. RESULTS: We fit a flexible parametric model for each cancer stage on the excess hazard scale and the differences in stage-specific marginal relative survival were assessed. Estimates were also obtained from non-parametric approaches for validation. There was little difference between the 10-year stage-specific marginal relative survival estimates, regardless of the assumed historical stage distribution. CONCLUSIONS: When conducting a period analysis, multiple imputation can be used to obtain stage-specific long-term estimates of relative survival, even when the historical stage information is largely incomplete.

Immortal time bias for life-long conditions in retrospective observational studies using electronic health records.

BACKGROUND: Immortal time bias is common in observational studies but is typically described for pharmacoepidemiology studies where there is a delay between cohort entry and treatment initiation. METHODS: This study used the Clinical Practice Research Datalink (CPRD) and linked national mortality data in England from 2000 to 2019 to investigate immortal time bias for a specific life-long condition, intellectual disability. Life expectancy (Chiang's abridged life table approach) was compared for 33,867 exposed and 980,586 unexposed individuals aged 10+ years using five methods: (1) treating immortal time as observation time; (2) excluding time before date of first exposure diagnosis; (3) matching cohort entry to first exposure diagnosis; (4) excluding time before proxy date of inputting first exposure diagnosis (by the physician); and (5) treating exposure as a time-dependent measure. RESULTS: When not considered in the design or analysis (Method 1), immortal time bias led to disproportionately high life expectancy for the exposed population during the first calendar period (additional years expected to live: 2000-2004: 65.6 [95% CI: 63.6,67.6]) compared to the later calendar periods (2005-2009: 59.9 [58.8,60.9]; 2010-2014: 58.0 [57.1,58.9]; 2015-2019: 58.2 [56.8,59.7]). Date of entry of diagnosis (Method 4) was unreliable in this CPRD cohort. The final methods (Method 2, 3 and 5) appeared to solve the main theoretical problem but residual bias may have remained. CONCLUSIONS: We conclude that immortal time bias is a significant issue for studies of life-long conditions that use electronic health record data and requires careful consideration of how clinical diagnoses are entered onto electronic health record systems.

Five ways to improve international comparisons of cancer survival: lessons learned from ICBP SURVMARK-2.

BACKGROUND: Comparisons of population-based cancer survival between countries are important to benchmark the overall effectiveness of cancer management. The International Cancer Benchmarking Partnership (ICBP) Survmark-2 study aims to compare survival in seven high-income countries across eight cancer sites and explore reasons for the observed differences. A critical aspect in ensuring comparability in the reported survival estimates are similarities in practice across cancer registries. While ICBP Survmark-2 has shown these differences are unlikely to explain the observed differences in cancer-specific survival between countries, it is important to keep in mind potential biases linked to registry practice and understand their likely impact. METHODS: Based on experiences gained within ICBP Survmark-2, we have developed a set of recommendations that seek to optimally harmonise cancer registry datasets to improve future benchmarking exercises. RESULTS: Our recommendations stem from considering the impact on cancer survival estimates in five key areas: (1) the completeness of the registry and the availability of registration sources; (2) the inclusion of death certification as a source of identifying cases; (3) the specification of the date of incidence; (4) the approach to handling multiple primary tumours and (5) the quality of linkage of cases to the deaths register. CONCLUSION: These recommendations seek to improve comparability whilst maintaining the opportunity to understand and act upon international variations in outcomes among cancer patients.

Non-parametric estimation of reference adjusted, standardised probabilities of all-cause death and death due to cancer for population group comparisons.

BACKGROUND: Ensuring fair comparisons of cancer survival statistics across population groups requires careful consideration of differential competing mortality due to other causes, and adjusting for imbalances over groups in other prognostic covariates (e.g. age). This has typically been achieved using comparisons of age-standardised net survival, with age standardisation addressing covariate imbalance, and the net estimates removing differences in competing mortality from other causes. However, these estimates lack ease of interpretability. In this paper, we motivate an alternative non-parametric approach that uses a common rate of other cause mortality across groups to give reference-adjusted estimates of the all-cause and cause-specific crude probability of death in contrast to solely reporting net survival estimates. METHODS: We develop the methodology for a non-parametric equivalent of standardised and reference adjusted crude probabilities of death, building on the estimation of non-parametric crude probabilities of death. We illustrate the approach using regional comparisons of survival following a diagnosis of rectal cancer for men in England. We standardise to the covariate distribution and other cause mortality of England as a whole to offer comparability, but with close approximation to the observed all-cause region-specific mortality. RESULTS: The approach gives comparable estimates to observed crude probabilities of death, but allows direct comparison across population groups with different covariate profiles and competing mortality patterns. In our illustrative example, we show that regional variations in survival following a diagnosis of rectal cancer persist even after accounting for the variation in deprivation, age at diagnosis and other cause mortality. CONCLUSIONS: The methodological approach of using standardised and reference adjusted metrics offers an appealing approach for future cancer survival comparison studies and routinely published cancer statistics. Our non-parametric estimation approach through the use of weighting offers the ability to estimate comparable survival estimates without the need for statistical modelling.