Understanding variation in disease risk: the elusive concept of frailty

Prior exposure to pathogens augments host heterogeneity in susceptibility and has key epidemiological consequences

Pathogen epidemics are key threats to human and wildlife health. Across systems, host protection from pathogens following initial exposure is often incomplete, resulting in recurrent epidemics through partially-immune hosts. Variation in population-level protection has important consequences for epidemic dynamics, but whether acquired protection influences host heterogeneity in susceptibility and its epidemiological consequences remains unexplored. We experimentally investigated whether prior exposure (none, low-dose, or high-dose) to a bacterial pathogen alters host heterogeneity in susceptibility among songbirds. Hosts with no prior pathogen exposure had little variation in protection, but heterogeneity in susceptibility was significantly augmented by prior pathogen exposure, with the highest variability detected in hosts given high-dose prior exposure. An epidemiological model parameterized with experimental data found that heterogeneity in susceptibility from prior exposure more than halved epidemic sizes compared with a homogeneous population with identical mean protection. However, because infection-induced mortality was also greatly reduced in hosts with prior pathogen exposure, reductions in epidemic size were smaller than expected in hosts with prior exposure. These results highlight the importance of variable protection from prior exposure and/or vaccination in driving host heterogeneity and epidemiological dynamics.

Cellular senescence and frailty: a comprehensive insight into the causal links

Senescent cells may have a prominent role in driving inflammation and frailty. The impact of cellular senescence on frailty varies depending on the assessment tool used, as it is influenced by the criteria or items predominantly affected by senescent cells and the varying weights assigned to these items across different health domains. To address this challenge, we undertook a thorough review of all available studies involving gain- or loss-of-function experiments as well as interventions targeting senescent cells, focusing our attention on those studies that examined outcomes based on the individual frailty phenotype criteria or specific items used to calculate two humans (35 and 70 items) and one mouse (31 items) frailty indexes. Based on the calculation of a simple "evidence score," we found that the burden of senescent cells related to musculoskeletal and cerebral health has the strongest causal link to frailty. We deem that insight into these mechanisms may not only contribute to clarifying the role of cellular senescence in frailty but could additionally provide multiple therapeutic opportunities to help the future development of a desirable personalized therapy in these extremely heterogeneous patients.

Chance, ignorance, and the paradoxes of cancer: Richard Peto on developing preventative strategies under uncertainty

During the early 1980s both cancer biology and epidemiological methods were being transformed. In 1984 the leading cancer epidemiologist Richard Peto - who, in 1981, had published the landmark Causes of Cancer with Richard Doll - wrote a short chapter on "The need for ignorance in cancer research", in which the worlds of epidemiology and speculative Darwinian biology met. His reflections on how evolutionary theory related to cancer have become known as "Peto's paradox", whilst his articulation of "black box epidemiology" provided the logic of subsequent practice in the field. We reprint this sparkling and prescient example of biologically-informed epidemiological theorising at its best in this issue of the European Journal of Epidemiology, together with four commentaries that focus on different aspects of its rich content. Here were provide some contextual background to the 1984 chapter, and our own speculations regarding various paradoxes in cancer epidemiology. We suggest that one reason for the relative lack of progress in indentifying novel modifiable causes of cancer over the last 40 years may reflect such exposures being ubiquitous within environments, and discuss the lessons for epidemiology that would follow from this.

Complete sources of cluster variation on the risk of under-five malaria in Uganda: a multilevel-weighted mixed effects logistic regression model approach

BACKGROUND

Malaria, a major cause of mortality worldwide is linked to a web of determinants ranging from individual to contextual factors. This calls for examining the magnitude of the effect of clustering within malaria data. Regrettably, researchers usually ignore cluster variation on the risk of malaria and also apply final survey weights in multilevel modelling instead of multilevel weights. This most likely produces biased estimates, misleads inference and lowers study power. The objective of this study was to determine the complete sources of cluster variation on the risk of under-five malaria and risk factors associated with under-five malaria in Uganda.

METHODS

This study applied a multilevel-weighted mixed effects logistic regression model to account for both individual and contextual factors.

RESULTS

Every additional year in a child's age was positively associated with malaria infection (AOR = 1.42; 95% CI 1.33-1.52). Children whose mothers had at least a secondary school education were less likely to suffer from malaria infection (AOR = 0.53; 95% CI 0.30-0.95) as well as those who dwelled in households in the two highest wealth quintiles (AOR = 0.42; 95% CI 0.27-0.64). An increase in altitude by 1 m was negatively associated with malaria infection (AOR = 0.98; 95% CI 0.97-0.99). About 77% of the total variation in the positive testing for malaria was attributable to differences between enumeration areas (ICC = 0.77; p < 0.001).

CONCLUSIONS

Interventions towards reducing the burden of under-five malaria should be prioritized to improve individual-level characteristics compared to household-level features. Enumeration area (EA) specific interventions may be more effective compared to household specific interventions.

Individual frailty excess hazard models in cancer epidemiology

Unobserved individual heterogeneity is a common challenge in population cancer survival studies. This heterogeneity is usually associated with the combination of model misspecification and the failure to record truly relevant variables. We investigate the effects of unobserved individual heterogeneity in the context of excess hazard models, one of the main tools in cancer epidemiology. We propose an individual excess hazard frailty model to account for individual heterogeneity. This represents an extension of frailty modeling to the relative survival framework. In order to facilitate the inference on the parameters of the proposed model, we select frailty distributions which produce closed-form expressions of the marginal hazard and survival functions. The resulting model allows for an intuitive interpretation, in which the frailties induce a selection of the healthier individuals among survivors. We model the excess hazard using a flexible parametric model with a general hazard structure which facilitates the inclusion of time-dependent effects. We illustrate the performance of the proposed methodology through a simulation study. We present a real-data example using data from lung cancer patients diagnosed in England, and discuss the impact of not accounting for unobserved heterogeneity on the estimation of net survival. The methodology is implemented in the R package IFNS.

The association between dairy intake and risk of cardiovascular disease and mortality in patients with stable angina pectoris

AIMS

The association of dairy products with cardiovascular disease and mortality risk remains heavily debated. We aimed to investigate the association between intake of total dairy and dairy products and the risk of acute myocardial infarction (AMI), stroke, and cardiovascular and all-cause mortality.

METHODS AND RESULTS

We included 1929 patients (80% men, mean age 62 years) with stable angina pectoris from the Western Norway B-vitamin Intervention Trial. Dietary data were obtained via a 169-item food frequency questionnaire. Risk associations were estimated using Cox proportional hazard regression models adjusted for relevant covariates. Non-linear associations were explored visually. The mean (±SD) dairy intake in the study population was 169 ± 108 g/1000 kcal. Median follow-up times were 5.2, 7.8, and 14.1 years for stroke, AMI, and mortality, respectively. Higher intake of total dairy and milk were positively associated with stroke risk [HR (95% CI): 1.14 (1.02, 1.27) and 1.13 (1.02, 1.27), cardiovascular mortality 1.06 (1.00, 1.12) and 1.07 (1.01, 1.13)] and all-cause mortality [1.07 (1.03, 1.11) and 1.06 (1.03, 1.10)] per 50 g/1000 kcal. Higher cheese intake was inversely associated with AMI risk [0.92 (0.83, 1.02)] per 10 g/1000 kcal. Butter was associated with increased AMI risk [1.10 (0.97, 1.24)] and all-cause mortality [1.10 (1.00, 1.20) per 5 g/1000 kcal.

CONCLUSION

Higher dairy and milk consumption were associated with increased risk of mortality and stroke. Cheese was associated with decreased, and butter with increased, risk of AMI. Dairy is a heterogenous food group with divergent health effects and dairy products should therefore be investigated individually.

Familial autoimmunity in patients with idiopathic inflammatory myopathies

BACKGROUND

Familial associations can be indicators of shared genetic susceptibility between two diseases. Previous data on familial autoimmunity in patients with idiopathic inflammatory myopathies (IIM) are scarce and inconsistent.

OBJECTIVES

To investigate which autoimmune diseases (ADs) may share genetic susceptibility with IIM, we examined the familial associations between IIM and different ADs.

METHODS

In this Swedish population-based family study, we assembled 7615 first-degree relatives (FDRs) of 1620 patients with IIM and 37,309 relatives of 7797 matched individuals without IIM. Via register linkages, we ascertained rheumatoid arthritis, other rheumatic inflammatory diseases (RIDs), multiple sclerosis, inflammatory bowel diseases (IBD), type 1 diabetes mellitus, autoimmune thyroid diseases (AITD), coeliac disease (CeD) and myasthenia gravis among the FDRs. We estimated the familial association between IIM and each AD using conditional logistic regression and performed subgroup analyses by kinship.

RESULTS

Patients with IIM had significantly higher odds of having ≥1 FDR affected by other RIDs (adjusted odds ratio [aOR] = 1.40, 95% confidence interval [CI] 1.11-1.78) and greater odds of having ≥2 FDRs affected by CeD (aOR = 3.57, 95% CI 1.28-9.92) compared to the individuals without IIM. In the analyses of any FDR pairs, we observed familial associations for other RIDs (aOR = 1.34, 95% CI 1.14-1.56), IBD (aOR = 1.20, 95% CI 1.02-1.41), AITD (aOR = 1.10, 95% CI 1.02-1.19) and CeD (aOR = 1.37, 95% CI 1.08-1.74) while associations for other ADs were not statistically significant.

CONCLUSION

The observed familial associations may suggest that IIM shares genetic susceptibility with various ADs, information that may be useful for clinical counselling and guiding future genetic studies of IIM.

A hierarchical process model links behavioral aging and lifespan in C. elegans

Aging involves a transition from youthful vigor to geriatric infirmity and death. Individuals who remain vigorous longer tend to live longer, and within isogenic populations of C. elegans the timing of age-associated vigorous movement cessation (VMC) is highly correlated with lifespan. Yet, many mutations and interventions in aging alter the proportion of lifespan spent moving vigorously, appearing to “uncouple” youthful vigor from lifespan. To clarify the relationship between vigorous movement cessation, death, and the physical declines that determine their timing, we developed a new version of the imaging platform called “The Lifespan Machine”. This technology allows us to compare behavioral aging and lifespan at an unprecedented scale. We find that behavioral aging involves a time-dependent increase in the risk of VMC, reminiscent of the risk of death. Furthermore, we find that VMC times are inversely correlated with remaining lifespan across a wide range of genotypes and environmental conditions. Measuring and modelling a variety of lifespan-altering interventions including a new RNA-polymerase II auxin-inducible degron system, we find that vigorous movement and lifespan are best described as emerging from the interplay between at least two distinct physical declines whose rates co-vary between individuals. In this way, we highlight a crucial limitation of predictors of lifespan like VMC—in organisms experiencing multiple, distinct, age-associated physical declines, correlations between mid-life biomarkers and late-life outcomes can arise from the contextual influence of confounding factors rather than a reporting by the biomarker of a robustly predictive biological age.

Aging produces a variety of outcomes—declines in various measures of health and eventually death. By studying the relationship between two outcomes of aging in the same individual, we can learn about the underlying aging processes that cause them. Here, we consider the relationship between death and an outcome often used to quantify health in C. elegans—vigorous movement cessation which describes the age-associated loss of an individuals’ ability to move long distances. We develop an automated imaging platform that allows us to precisely compare this pair of outcomes in each individual across large populations. We find that individuals who remain vigorous longer subsequently have a shorter remaining lifespan—a pattern that holds even after vigorous movement and lifespan timing are both altered by several different mutations and interventions in aging. Modelling our data using a combination of simulation and analytic studies, we demonstrate how the relative timing of vigorous movement cessation and death suggest that these two outcomes are driven by distinct aging processes. Our data and analyses demonstrate how two outcomes of aging can be correlated across individuals with the timing of one predicting the timing of the other, but nevertheless be driven by mostly distinct underlying physical declines.

Heterogeneity in coronary heart disease risk

There is large inter-individual heterogeneity in risk of coronary heart disease (CHD). Risk factors traditionally used in primary risk assessment only partially explain this heterogeneity. Residual, unobserved heterogeneity leads to age-related attenuation of hazard rates and underestimation of hazard ratios. Its magnitude is unknown. Therefore, we aimed to estimate a lower and an approximate upper bound. Heterogeneity was parametrized by a log-normal distribution with shape parameter σ. Analysis was based on published data. From concordance indices of studies including traditional risk factors and additional diagnostic imaging data, we calculated the part of heterogeneity explained by imaging data. For traditional risk assessment, this part typically remains unexplained, thus constituting a lower bound on unobserved heterogeneity. Next, the potential impact of heterogeneity on CHD hazard rates in several large countries was investigated. CHD rates increase with age but the increase attenuates with age. Presuming this attenuation to be largely caused by heterogeneity, an approximate upper bound on σ was derived. Taking together both bounds, unobserved heterogeneity in studies without imaging information can be described by a shape parameter in the range σ = 1-2. It substantially contributes to observed age-dependences of hazard ratios and may lead to underestimation of hazard ratios by a factor of about two. Therefore, analysis of studies for primary CHD risk assessment should account for unobserved heterogeneity.

Individual variation in susceptibility or exposure to SARS-CoV-2 lowers the herd immunity threshold

Individual variation in susceptibility and exposure is subject to selection by natural infection, accelerating the acquisition of immunity, and reducing herd immunity thresholds and epidemic final sizes. This is a manifestation of a wider population phenomenon known as "frailty variation". Despite theoretical understanding, public health policies continue to be guided by mathematical models that leave out considerable variation and as a result inflate projected disease burdens and overestimate the impact of interventions. Here we focus on trajectories of the coronavirus disease (COVID-19) pandemic in England and Scotland until November 2021. We fit models to series of daily deaths and infer relevant epidemiological parameters, including coefficients of variation and effects of non-pharmaceutical interventions which we find in agreement with independent empirical estimates based on contact surveys. Our estimates are robust to whether the analysed data series encompass one or two pandemic waves and enable projections compatible with subsequent dynamics. We conclude that vaccination programmes may have contributed modestly to the acquisition of herd immunity in populations with high levels of pre-existing naturally acquired immunity, while being crucial to protect vulnerable individuals from severe outcomes as the virus becomes endemic.

Polygenic risk for prostate cancer: Decreasing relative risk with age but little impact on absolute risk

Polygenic risk scores (PRSs) for a variety of diseases have recently been shown to have relative risks that depend on age, and genetic relative risks decrease with increasing age. A refined understanding of the age dependency of PRSs for a disease is important for personalized risk predictions and risk stratification. To further evaluate how the PRS relative risk for prostate cancer depends on age, we refined analyses for a validated PRS for prostate cancer by using 64,274 prostate cancer cases and 46,432 controls of diverse ancestry (82.8% European, 9.8% African American, 3.8% Latino, 2.8% Asian, and 0.8% Ghanaian). Our strategy applied a novel weighted proportional hazards model to case-control data to fully utilize age to refine how the relative risk decreased with age. We found significantly greater relative risks for younger men (age 30-55 years) compared with older men (70-88 years) for both relative risk per standard deviation of the PRS and dichotomized according to the upper 90^th percentile of the PRS distribution. For the largest European ancestral group that could provide reliable resolution, the log-relative risk decreased approximately linearly from age 50 to age 75. Despite strong evidence of age-dependent genetic relative risk, our results suggest that absolute risk predictions differed little from predictions that assumed a constant relative risk over ages, from short-term to long-term predictions, simplifying implementation of risk discussions into clinical practice.

Association of timing of birth with mortality among preterm infants born in Canada

OBJECTIVE

To assess the association between time of birth and mortality among preterm infants.

STUDY DESIGN

Population-based study of infants born 22-36 weeks gestation (GA) in Canada from 2010 to 2015 (n = 173 789). Multivariable logistic regression models assessed associations between timing of birth and mortality.

RESULT

Among infants 22-27 weeks GA, evening birth was associated with higher mortality than daytime birth (adjusted odds ratio [AOR] 1.14, 95% CI 1.01-1.29). Among infants 28-32 weeks GA and 33-36 weeks GA, night birth was associated with lower mortality than daytime birth (AOR 0.75, 95% CI 0.59-0.95; AOR 0.78, 95% CI 0.62-0.99, respectively). Sensitivity analysis excluding infants with major congenital anomaly revealed that associations between hour of birth and mortality among infants born 28-32 and 33-36 weeks GA decreased or were not statistically significant.

CONCLUSION

Higher mortality among extremely preterm infants during off-peak hours may suggest variations in available resources based on time of day.

Landfall After the Perfect Storm: Cohort Differences in the Relationship Between Debt and Risk of Heart Attack

Analyses of the Health and Retirement Study (HRS) between 1992 and 2014 compare the relationship between different levels and forms of debt and heart attack risk trajectories across four cohorts. Although all cohorts experienced growing household debt, including the increase of both secured and unsecured debt, they nevertheless encountered different economic opportunity structures and crises at sensitive times in their life courses, with implications for heart attack risk trajectories. Results from frailty hazards models reveal that unsecured debt is associated with increased risk of heart attack across all cohorts. Higher levels of housing debt, however, predict higher rates of heart attack among only the earlier cohorts. Heart attack risk trajectories for Baby Boomers with high levels of housing debt are lower than those of same-aged peers with no housing debt. Thus, the relationship between debt and heart attack varies by level and form of debt across cohorts but distinguishes Baby Boomer cohorts based on their diverse exposures to volatile housing market conditions over the sensitive household formation period of the life course.

Nonproportional hazards model with a frailty term for modeling subgroups with evidence of long-term survivors: Application to a lung cancer dataset

With advancements in medical treatments for cancer, an increase in the life expectancy of patients undergoing new treatments is expected. Consequently, the field of statistics has evolved to present increasingly flexible models to explain such results better. In this paper, we present a lung cancer dataset with some covariates that exhibit nonproportional hazards (NPHs). Besides, the presence of long-term survivors is observed in subgroups. The proposed modeling is based on the generalized time-dependent logistic model with each subgroup's effect time and a random term effect (frailty). In practice, essential covariates are not observed for several reasons. In this context, frailty models are useful in modeling to quantify the amount of unobservable heterogeneity. The frailty distribution adopted was the weighted Lindley distribution, which has several interesting properties, such as the Laplace transform function on closed form, flexibility in the probability density function, among others. The proposed model allows for NPHs and long-term survivors in subgroups. Parameter estimation was performed using the maximum likelihood method, and Monte Carlo simulation studies were conducted to evaluate the estimators' performance. We exemplify this model's use by applying data of patients diagnosed with lung cancer in the state of São Paulo, Brazil.

Predictability in Contemporary Medicine

Medical practice is increasingly coming under the guidance of statistical-mathematical models that are, undoubtedly, valuable tools but are also only a partial representation of reality. Indeed, given that statistics may be more or less adequate, a model is still a subjective interpretation of the researcher and is also influenced by the historical context in which it operates. From this opinion, I will provide a short historical excursus that retraces the advent of probabilistic medicine as a long process that has a beginning that should be sought in the discovery of the complexity of disease. By supporting the belonging of this evolution to the scientific domain it is also acknowledged that the underlying model can be imperfect or fallible and, therefore, confutable as any product of science. Indeed, it seems non-trivial here to recover these concepts, especially today where clinical decisions are entrusted to practical guidelines, which are a hybrid product resulting from the aggregation of multiple perspectives, including the probabilistic approach, to disease. Finally, before the advent of precision medicine, by limiting the use of guidelines to the original consultative context, an aged approach is supported, namely, a relationship with the individual patient.

Systemic alterations play a dominant role in epigenetic predisposition to breast cancer in offspring of obese fathers and is transmitted to a second generation

We previously showed that environmentally-induced epigenetic inheritance of cancer occurs in rodent models. For instance, we reported that paternal consumption of an obesity-inducing diet (OID) increased breast cancer susceptibility in the offspring (F1). Nevertheless, it is still unclear whether programming of breast cancer in daughters is due to systemic alterations or mammary epithelium-specific factors and whether the breast cancer predisposition in F1 progeny can be transmitted to subsequent generations. In this study, we show that mammary glands from F1 control (CO) female offspring exhibit enhanced growth when transplanted into OID females compared to CO mammary glands transplanted into CO females. Similarly, carcinogen-induced mammary tumors from F1 CO female offspring transplanted into OID females has a higher proliferation/apoptosis rate. Further, we show that granddaughters (F2) from the OID grand-paternal germline have accelerated tumor growth compared to CO granddaughters. This between-generation transmission of cancer predisposition is associated with changes in sperm tRNA fragments in OID males. Our findings indicate that systemic and mammary stromal alterations are significant contributors to programming of mammary development and likely cancer predisposition in OID daughters. Our data also show that breast cancer predisposition is transmitted to subsequent generations and may explain some familial cancers, if confirmed in humans.

Treatment Intensification for Elevated Blood Pressure and Risk of Recurrent Stroke

Background

It remains unclear whether physicians' attitudes toward timely management of elevated blood pressure affect the risk of stroke recurrence.

Methods and Results

From a multicenter stroke registry database, we identified 2933 patients with acute ischemic stroke who were admitted to participating centers in 2011, survived at the 1‐year follow‐up period, and returned to outpatient clinics ≥2 times after discharge. As a surrogate measure of physicians' attitude, individual treatment intensification (TI) scores were calculated by dividing the difference between the frequencies of observed and expected medication changes by the frequency of clinic visits and categorizing them into 5 groups. The association between TI groups and the recurrence of stroke within 1 year was analyzed using hierarchical frailty models, with adjustment for clustering within each hospital and relevant covariates. Mean±SD of the TI score was −0.13±0.28. The TI score groups were significantly associated with increased risk of recurrent stroke compared with Group 3 (TI score range, −0.25 to 0); Group 1 (range, −1 to −0.5), adjusted hazard ratio (HR) 13.43 (95% CI, 5.95–30.35); Group 2 (range, −0.5 to −0.25), adjusted HR 4.59 (95% CI, 2.01–10.46); and Group 4 (TI score 0), adjusted HR 6.60 (95% CI, 3.02–14.45); but not with Group 5 (range, 0–1), adjusted HR 1.68 (95% CI, 0.62–4.56). This elevated risk in the lowest TI score groups persisted when confining analysis to those with hypertension, history of blood pressure‐lowering medication, no atrial fibrillation, and regular clinic visits and stratifying the subjects by functional capacity at discharge.

Conclusions

A low TI score, which implies physicians' therapeutic inertia in blood pressure management, was associated with a higher risk of recurrent stroke. The TI score may be a useful performance indicator in the outpatient clinic setting to prevent recurrent stroke.

Potential limitations in COVID-19 machine learning due to data source variability: A case study in the nCov2019 dataset

OBJECTIVE

The lack of representative coronavirus disease 2019 (COVID-19) data is a bottleneck for reliable and generalizable machine learning. Data sharing is insufficient without data quality, in which source variability plays an important role. We showcase and discuss potential biases from data source variability for COVID-19 machine learning.

MATERIALS AND METHODS

We used the publicly available nCov2019 dataset, including patient-level data from several countries. We aimed to the discovery and classification of severity subgroups using symptoms and comorbidities.

RESULTS

Cases from the 2 countries with the highest prevalence were divided into separate subgroups with distinct severity manifestations. This variability can reduce the representativeness of training data with respect the model target populations and increase model complexity at risk of overfitting.

CONCLUSIONS

Data source variability is a potential contributor to bias in distributed research networks. We call for systematic assessment and reporting of data source variability and data quality in COVID-19 data sharing, as key information for reliable and generalizable machine learning.

Impact of Treating Physician on Radiation Therapy Related Severe Toxicities in Men with Prostate Cancer

PURPOSE

The impact of treating physician on radiation therapy (RT) related toxicity is unclear. We carried out a secondary analysis of a randomized controlled study to determine whether the risk of RT-related late toxicities in patients with prostate cancer varies depending on the treating radiation oncologist.

METHODS AND MATERIALS

This is a secondary analysis of a phase 3 randomized controlled study in which patients with prostate cancer with Gleason score ≤7, clinical stage T1b-T3a, and prostate-specific antigen <30 ng/mL were randomized to receive androgen suppression for 6 months, starting either 4 months before or concurrently with definitive prostate radiation therapy. Incidence of late RT-related toxicity was estimated using Kaplan-Meier methods. We applied multivariable semiparametric shared frailty models with gamma distribution to determine the between-physician variation in the hazard of late RT-related grade ≥3 gastrointestinal, genitourinary, or overall toxicity. Patient level covariables included age, risk group, year of enrollment, and treatment regimen. Frailty variance, a measure of unexplained heterogeneity, was estimated with 95% confidence intervals (CIs). Statistical significance was suggested when the lower limit of the 95% CI for the frailty variance was >0. The Commenges-Andersen test was used for P value estimation.

RESULTS

Overall, 426 patients were treated by 9 radiation oncologists. On log-rank test, there was a significant difference in the cumulative incidence of overall grade ≥3 toxicities (P = .001) and grade ≥3 gastrointestinal toxicity (P = .01) among the physician-based clusters. The frailty variance for overall late grade ≥3 toxicity was 0.31 (95% CI, 0.02-1.39; P = .01). The frailty variance for the grade ≥3 gastrointestinal and genitourinary toxicity was 0.84 (95% CI, 0.00-4.20; P = .11) and 0.11 (95% CI, 0.00-1.13; P = .31), respectively.

CONCLUSIONS

In our study, the hazard of overall RT-related late grade ≥3 toxicity varied significantly depending on treating radiation oncologist. Further studies are required to explore the underlying processes that lead to such variations in clinical trials involving radiation therapy in prostate cancer.

Variability in clinicians' understanding and reported methods of identifying high-risk surgical patients: a qualitative study

Background

High-risk patients presenting for surgery require complex decision-making and perioperative management. However, given there is no gold standard for identifying high-risk patients, doing so may be challenging for clinicians in practice. Before a gold standard can be established, the state of current practice must be determined. This study aimed to understand how working clinicians define and identify high-risk surgical patients.

Methods

Clinicians involved in the care of high-risk surgical patients at a public hospital in regional Australia were interviewed as part of an ongoing study evaluating a new shared decision-making process for high-risk patients. The new process, Patient-Centred Advanced Care Planning (PC-ACP) engages patients, families, and clinicians from all relevant specialties in shared decision-making in line with the patient’s goals and values. The semi-structured interviews were conducted before the implementation of the new process and were coded using a modified form of the ‘constant comparative method’ to reveal key themes. Themes concerning patient risk, clinician’s understanding of high risk, and methods for identifying high-risk surgical patients were extricated for close examination.

Results

Thirteen staff involved in high-risk surgery at the hospital at which PC-ACP was to be implemented were interviewed. Analysis revealed six sub-themes within the major theme of factors related to patient risk: (1) increase in high-risk patients, (2) recognising frailty, (3) risk-benefit balance, (4) suitability and readiness for surgery, (5) avoiding negative outcomes, and (6) methods in use for identifying high-risk patients. There was considerable variability in clinicians’ methods of identifying high-risk patients and regarding their definition of high risk. This variability occurred even among clinicians within the same disciplines and specialties.

Conclusions

Although clinicians were confident in their own ability to identify high-risk patients, they acknowledged limitations in recognising frail, high-risk patients and predicting and articulating possible outcomes when consenting these patients. Importantly, little consistency in clinicians’ reported methods for identifying high-risk patients was found. Consensus regarding the definition of high-risk surgical patients is necessary to ensure rigorous decision-making.

Predictive Accuracy of Quick Sequential Organ Failure Assessment for Hospital Mortality Decreases With Increasing Comorbidity Burden Among Patients Admitted for Suspected Infection

OBJECTIVES

Evaluate the accuracy of the quick Sequential Organ Failure Assessment tool to predict mortality across increasing levels of comorbidity burden.

DESIGN

Retrospective observational cohort study.

SETTING

Twelve acute care hospitals in the Southeastern United States.

PATIENTS

A total of 52,187 patients with suspected infection presenting to the Emergency Department between January 2014 and September 2017.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The primary outcome was hospital mortality. We used electronic health record data to calculate quick Sequential Organ Failure Assessment risk scores from vital signs and laboratory values documented during the first 24 hours. We calculated Charlson Comorbidity Index scores to quantify comorbidity burden. We constructed logistic regression models to evaluate differences in the performance of quick Sequential Organ Failure Assessment greater than or equal to 2 to predict hospital mortality in patients with no documented (Charlson Comorbidity Index = 0), low (Charlson Comorbidity Index = 1-2), moderate (Charlson Comorbidity Index = 3-4), or high (Charlson Comorbidity Index ≥ 5) comorbidity burden. Among the cohort, 2,030 patients died in the hospital (4%). No comorbidities were documented for 5,038 patients (10%), 9,235 patients (18%) had low comorbidity burden, 12,649 patients (24%) had moderate comorbidity burden, and 25,265 patients (48%) had high comorbidity burden. Overall model discrimination for quick Sequential Organ Failure Assessment greater than or equal to 2 was the area under the receiver operating characteristic curve of 0.71 (95% CI, 0.69-0.72). A model including both quick Sequential Organ Failure Assessment and Charlson Comorbidity Index had improved discrimination compared with Charlson Comorbidity Index alone (area under the receiver operating characteristic curve, 0.77; 95% CI, 0.76-0.78 vs area under the curve, 0.61; 95% CI, 0.59-0.62). Discrimination was highest among patients with no documented comorbidities (quick Sequential Organ Failure Assessment area under the receiver operating characteristic curve, 0.84; 95% CI; 0.79-0.89) and lowest among high comorbidity patients (quick Sequential Organ Failure Assessment area under the receiver operating characteristic curve, 0.67; 95% CI, 0.65-0.68). The strength of association between quick Sequential Organ Failure Assessment and mortality ranged from 30.5-fold increased likelihood in patients with no comorbidities to 4.7-fold increased likelihood in patients with high comorbidity.

CONCLUSIONS

The accuracy of quick Sequential Organ Failure Assessment to predict hospital mortality diminishes with increasing comorbidity burden. Patients with comorbidities may have baseline abnormalities in quick Sequential Organ Failure Assessment variables that reduce predictive accuracy. Additional research is needed to better understand quick Sequential Organ Failure Assessment performance across different comorbid conditions with modification that incorporates the context of changes to baseline variables.

The effect of treatment delay on time-to-recovery in the presence of unobserved heterogeneity

We study the effect of delaying treatment in the presence of (unobserved) heterogeneity. In a homogeneous population and assuming a proportional treatment effect, a treatment delay period will result in notably lower cumulative recovery percentages. We show in theoretical scenarios using frailty models that if the population is heterogeneous, the effect of a delay period is much smaller. This can be explained by the selection process that is induced by the frailty. Patient groups that start treatment later have already undergone more selection. The marginal hazard ratio for the treatment will act differently in such a more homogeneous patient group. We further discuss modeling approaches for estimating the effect of treatment delay in the presence of heterogeneity, and compare their performance in a simulation study. The conventional Cox model that fails to account for heterogeneity overestimates the effect of treatment delay. Including interaction terms between treatment and starting time of treatment or between treatment and follow up time gave no improvement. Estimating a frailty term can improve the estimation, but is sensitive to misspecification of the frailty distribution. Therefore, multiple frailty distributions should be used and the results should be compared using the Akaike Information Criterion. Non‐parametric estimation of the cumulative recovery percentages can be considered if the dataset contains sufficient long term follow up for each of the delay strategies. The methods are demonstrated on a motivating application evaluating the effect of delaying the start of treatment with assisted reproductive techniques on time‐to‐pregnancy in couples with unexplained subfertility.

Prevalent cohort studies and unobserved heterogeneity

Consider lifetimes originating at a series of calendar times [Formula: see text]. At a certain time [Formula: see text] a cross-sectional sample is taken, generating a sample of current durations (backward recurrence times) of survivors until [Formula: see text] and a prevalent cohort study consisting of survival times left-truncated at the current durations. A Lexis diagram is helpful in visualizing this situation. Survival analysis based on current durations and prevalent cohort studies is now well-established as long as all covariates are observed. The general problems with unobserved covariates have been well understood for ordinary prospective follow-up studies, with the good help of hazard rate models incorporating frailties: as for ordinary regression models, the added noise generates attenuation in the regression parameter estimates. For prevalent cohort studies this attenuation remains, but in addition one needs to take account of the differential selection of the survivors from initiation [Formula: see text] to cross-sectional sampling at [Formula: see text]. This paper intends to survey the recent development of these matters and the consequences for routine use of hazard rate models or accelerated failure time models in the many cases where unobserved heterogeneity may be an issue. The study was inspired by concrete problems in the study of time-to-pregnancy, and we present various simulation results inspired by this particular application.

A dual frailty model for lifetime analysis in maritime transportation

We consider changes in ownership of commercial shipping vessels from an event history perspective. Each change in ownership can be influenced by the properties of the vessel itself, its age and history to date, the characteristics of both the seller and the buyer, and time-varying market conditions. Similar factors can affect the process of deciding when to scrap the vessel as no longer being economically viable. We consider a multi-state approach in which states are defined by the owning companies, a sale marks a transition, and scrapping of the vessel corresponds to moving to an absorbing state. We propose a dual frailty model that attempts to capture unexplained heterogeneity in the data, with one frailty term for the seller and one for the buyer. We describe a Monte Carlo Markov chain estimation procedure and verify its accuracy through simulations. We investigate the consequences of mistakenly ignoring frailty in these circumstances. We compare results with and without the inclusion of frailty.

Trajectory Groups of 24-Hour Systolic Blood Pressure After Acute Ischemic Stroke and Recurrent Vascular Events

Background and Purpose- Blood pressure dynamics in patients with acute ischemic stroke may serve as an important modifiable and prognostic factor. Methods- A total of 8376 patients with acute ischemic stroke were studied from a prospective multicenter registry. Patients were eligible if they had been admitted within 24 hours of symptom onset and had ≥5 systolic blood pressure (SBP) measurements during the first 24 hours of hospitalization. SBP trajectory groups in the first 24 hours were identified using the TRAJ procedure in SAS software with delta-Bayesian Information Criterion and prespecified modeling parameters. Vascular events, including recurrent stroke, myocardial infarction, and death, were prospectively collected. The risk of having vascular events was calculated using the frailty model to adjust for clustering by hospital. Results- The group-based trajectory model classified patients with acute ischemic stroke into 5 SBP trajectory groups: low (22.3%), moderate (40.8%), rapidly stabilized (11.9%), acutely elevated (18.5%), and persistently high (6.4%) SBP. The risk of having vascular events was increased in the acutely elevated (hazard ratio, 1.28 [95% confidence interval, 1.12-1.47]) and the persistently high SBP groups (hazard ratio, 1.67 [95% confidence interval, 1.37-2.04]) but not in the rapidly stabilized group (hazard ratio, 1.13 [95% confidence interval, 0.95-1.34]), when compared with the moderate SBP group. Conclusions- SBP during the first 24 hours after acute ischemic stroke may be categorized into distinct trajectory groups, which differ in relation to stroke characteristics and frequency of subsequent recurrent vascular event risks. The findings may help to recognize potential candidates for future blood pressure control trials.

On prognostic estimates of radiation risk in medicine and radiation protection

The problem of expressing cumulative detrimental effect of radiation exposure is revisited. All conventionally used and computationally complex lifetime or time-integrated risks are based on current population and health statistical data, with unknown future secular trends, that are projected far into the future. It is shown that application of conventionally used lifetime or time-integrated attributable risks (LAR, AR) should be limited to exposures under 1 Gy. More general quantities, such as excess lifetime risk (ELR) and, to a lesser extent, risk of exposure-induced death (REID), are free of dose constraints, but are even more computationally complex than LAR and AR and rely on the unknown total radiation effect on demographic and health statistical data. Appropriate assessment of time-integrated risk of a specific outcome following high-dose (more than 1 Gy) exposure requires consideration of competing risks for other radiation-attributed outcomes and the resulting ELR estimate has an essentially non-linear dose response. Limitations caused by basing conventionally applied time-integrated risks on current population and health statistical data are that they are: (a) not well suited for risk estimates for atypical groups of exposed persons not readily represented by the general population; and (b) not optimal for risk projections decades into the future due to large uncertainties in developments of the future secular trends in the population-specific disease rates. Alternative disease-specific quantities, baseline and attributable survival fractions, based on reduction of survival chances are considered here and are shown to be very useful in circumventing most aspects of these limitations. Another main quantity, named as radiation-attributed decrease of survival (RADS), is recommended here to represent cumulative radiation risk conditional on survival until a certain age. RADS, historically known in statistical literature as "cumulative risk", is only based on the radiation-attributed hazard and is insensitive to competing risks. Therefore, RADS is eminently suitable for risk projections in emergency situations and for estimating radiation risks for persons exposed after therapeutic or interventional medical applications of radiation or in other highly atypical groups of exposed persons, such as astronauts.

Introducing risk inequality metrics in tuberculosis policy development

Global stakeholders including the World Health Organization rely on predictive models for developing strategies and setting targets for tuberculosis care and control programs. Failure to account for variation in individual risk leads to substantial biases that impair data interpretation and policy decisions. Anticipated impediments to estimating heterogeneity for each parameter are discouraging despite considerable technical progress in recent years. Here we identify acquisition of infection as the single process where heterogeneity most fundamentally impacts model outputs, due to selection imposed by dynamic forces of infection. We introduce concrete metrics of risk inequality, demonstrate their utility in mathematical models, and pack the information into a risk inequality coefficient (RIC) which can be calculated and reported by national tuberculosis programs for use in policy development and modeling.

Failure to account for heterogeneity in TB risk can mislead model-based evaluation of proposed interventions. Here, the authors introduce a metric to estimate the distribution of risk in populations from routinely collected data and find that variation in infection acquisition is the most impactful.

Comparing Observed and Unobserved Components of Childhood: Evidence From Finnish Register Data on Midlife Mortality From Siblings and Their Parents

In this study, we argue that the long arm of childhood that determines adult mortality should be thought of as comprising an observed part and its unobserved counterpart, reflecting the observed socioeconomic position of individuals and their parents and unobserved factors shared within a family. Our estimates of the observed and unobserved parts of the long arm of childhood are based on family-level variance in a survival analytic regression model, using siblings nested within families as the units of analysis. The study uses a sample of Finnish siblings born between 1936 and 1950 obtained from Finnish census data. Individuals are followed from ages 35 to 72. To explain familial influence on mortality, we use demographic background factors, the socioeconomic position of the parents, and the individuals' own socioeconomic position at age 35 as predictors of all-cause and cause-specific mortality. The observed part-demographic and socioeconomic factors, including region; number of siblings; native language; parents' education and occupation; and individuals' income, occupation, tenancy status, and education-accounts for between 10 % and 25 % of the total familial influence on mortality. The larger part of the influence of the family on mortality is not explained by observed individual and parental socioeconomic position or demographic background and thus remains an unobserved component of the arm of childhood. This component highlights the need to investigate the influence of childhood circumstances on adult mortality in a comprehensive framework, including demographic, social, behavioral, and genetic information from the family of origin.

Heterogeneity in risk of prostate cancer: A Swedish population-based cohort study of competing risks and Type 2 diabetes mellitus

Most previous studies of prostate cancer have not taken into account that men in the studied populations are also at risk of competing event, and that these men may have different susceptibility to prostate cancer risk. The aim of our study was to investigate heterogeneity in risk of prostate cancer, using a recently developed latent class regression method for competing risks. We further aimed to elucidate the association between Type 2 diabetes mellitus (T2DM) and prostate cancer risk, and to compare the results with conventional methods for survival analysis. We analysed the risk of prostate cancer in 126,482 men from the comparison cohort of the Prostate Cancer Data base Sweden (PCBaSe) 3.0. During a mean follow-up of 6 years 6,036 men were diagnosed with prostate cancer and 22,393 men died. We detected heterogeneity in risk of prostate cancer with two distinct latent classes in the study population. The smaller class included 9% of the study population in which men had a higher risk of prostate cancer and the risk was stronger associated with class membership than any of the covariates included in the study. Moreover, we found no association between T2DM and risk of prostate cancer after removal of the effect of informative censoring due to competing risks. The recently developed latent class for competing risks method could be used to provide new insights in precision medicine with the target to classify individuals regarding different susceptibility to a particular disease, reaction to a risk factor or response to treatment.

Anthropometric measurements and mortality in frail older adults

BACKGROUND

As the number of older adults increases, so does the number of frail older adults. Although anthropometry has been widely used as a way to stratify the overall mortality risk of a person, the significance of these measurements becomes blurred in the case of frail older adults who have changes in body composition. Therefore, the aim of this study is to determine the association of anthropometric measurements (body mass index, knee-adjusted height body mass index, waist-to-hip ratio and calf circumference) with mortality risk in a group of older Mexican adults.

METHODS

This is a longitudinal analysis of the Mexican Health and Aging sub-sample (with biomarkers, n = 2573) from the first wave in 2001, followed-up to the last available wave in 2015. Only frail 50-year or older adults (Frailty Index with a cut-off value of 0.21 or higher, was used) were considered for this analysis (n = 1298). A survival analysis was performed with Kaplan-Meier curves and Cox regression models (unadjusted and adjusted for confounding). Socio-demographic, health risks, physical activity and comorbidities were variables used for adjusting the multivariate models.

RESULTS

From the total sample of 1298 older adults, 32.5% (n = 422) died during follow-up. The highest hazard ratio in the adjusted model was for calf circumference 1.31 (95% confidence interval 1.02-1.69, p = 0.034). Other measurements were not significant.

CONCLUSIONS

Anthropometric measurements have different significance in frail older adults, and these differences could have implications on adverse outcomes. Calf circumference has a potential value in predicting negative health outcomes.

Exploring Selection Bias by Causal Frailty Models: The Magnitude Matters

Counter-intuitive associations appear frequently in epidemiology, and these results are often debated. In particular, several scenarios are characterized by a general risk factor that appears protective in particular subpopulations, for example, individuals suffering from a specific disease. However, the associations are not necessarily representing causal effects. Selection bias due to conditioning on a collider may often be involved, and causal graphs are widely used to highlight such biases. These graphs, however, are qualitative, and they do not provide information on the real life relevance of a spurious association. Quantitative estimates of such associations can be obtained from simple statistical models. In this study, we present several paradoxical associations that occur in epidemiology, and we explore these associations in a causal, frailty framework. By using frailty models, we are able to put numbers on spurious effects that often are neglected in epidemiology. We discuss several counter-intuitive findings that have been reported in real life analyses, and we present calculations that may expand the understanding of these associations. In particular, we derive novel expressions to explain the magnitude of bias in index-event studies.

The surprising implications of familial association in disease risk

BACKGROUND

A wide range of diseases show some degree of clustering in families; family history is therefore an important aspect for clinicians when making risk predictions. Familial aggregation is often quantified in terms of a familial relative risk (FRR), and although at first glance this measure may seem simple and intuitive as an average risk prediction, its implications are not straightforward.

METHODS

We use two statistical models for the distribution of disease risk in a population: a dichotomous risk model that gives an intuitive understanding of the implication of a given FRR, and a continuous risk model that facilitates a more detailed computation of the inequalities in disease risk. Published estimates of FRRs are used to produce Lorenz curves and Gini indices that quantifies the inequalities in risk for a range of diseases.

RESULTS

We demonstrate that even a moderate familial association in disease risk implies a very large difference in risk between individuals in the population. We give examples of diseases for which this is likely to be true, and we further demonstrate the relationship between the point estimates of FRRs and the distribution of risk in the population.

CONCLUSIONS

The variation in risk for several severe diseases may be larger than the variation in income in many countries. The implications of familial risk estimates should be recognized by epidemiologists and clinicians.

Vaccine Effects on Heterogeneity in Susceptibility and Implications for Population Health Management

Heterogeneity in host susceptibility is a key determinant of infectious disease dynamics but is rarely accounted for in assessment of disease control measures. Understanding how susceptibility is distributed in populations, and how control measures change this distribution, is integral to predicting the course of epidemics with and without interventions. Using multiple experimental and modeling approaches, we show that rainbow trout have relatively homogeneous susceptibility to infection with infectious hematopoietic necrosis virus and that vaccination increases heterogeneity in susceptibility in a nearly all-or-nothing fashion. In a simple transmission model with an R₀ of 2, the highly heterogeneous vaccine protection would cause a 35 percentage-point reduction in outbreak size over an intervention inducing homogenous protection at the same mean level. More broadly, these findings provide validation of methodology that can help to reduce biases in predictions of vaccine impact in natural settings and provide insight into how vaccination shapes population susceptibility.

Differences among individuals influence transmission and spread of infectious diseases as well as the effectiveness of control measures. Control measures, such as vaccines, may provide leaky protection, protecting all hosts to an identical degree, or all-or-nothing protection, protecting some hosts completely while leaving others completely unprotected. This distinction can have a dramatic influence on disease dynamics, yet this distribution of protection is frequently unaccounted for in epidemiological models and estimates of vaccine efficacy. Here, we apply new methodology to experimentally examine host heterogeneity in susceptibility and mode of vaccine action as distinct components influencing disease outcome. Through multiple experiments and new modeling approaches, we show that the distribution of vaccine effects can be robustly estimated. These results offer new experimental and inferential methodology that can improve predictions of vaccine effectiveness and have broad applicability to human, wildlife, and ecosystem health.

How serious is bias in effect estimation in randomised trials with survival data given risk heterogeneity and informative censoring?

It is often assumed that randomisation will prevent bias in estimation of treatment effects from clinical trials, but this is not true of the semiparametric Proportional Hazards model for survival data when there is underlying risk heterogeneity. Here, a new formula is proposed for estimation of this bias, improving on a previous formula through ease of use and clarity regarding the role of the mid-study cumulative hazard rate, shown to be an important factor for the bias magnitude. Informative censoring (IC) is recognised as a source of bias. Here, work on selection effects among survivors due to risk heterogeneity is extended to include IC. A new formula shows that bias in causal effect estimation under IC has two sources: one consequent on heterogeneity and one from the additional impact of IC. The formula provides new insights not previously shown: there may less bias under IC than when there is no IC and even, in principle, zero bias. When tested against simulated data, the new formulae are shown to be very accurate for prediction of bias in Proportional Hazards and accelerated failure time analyses which ignore heterogeneity. These data are also used to investigate the performance of accelerated failure time models which explicitly model risk heterogeneity ('frailty models') and G estimation. These methods remove bias when there is simple censoring but not with informative censoring when they may lead to overestimation of treatment effects. The new formulae may be used to help researchers judge the possible extent of bias in past studies. Copyright © 2017 John Wiley & Sons, Ltd.

Statistical methods for unidirectional switch designs: Past, present, and future

Clinical trials may apply or use a sequential introduction of a new treatment to determine its efficacy or effectiveness with respect to a control treatment. The reasons for choosing a particular switch design have different origins. For instance, they may be implemented for ethical or logistic reasons or for studying disease-modifying effects. Large-scale pragmatic trials with complex interventions often use stepped wedge designs (SWDs), where all participants start at the control group, and during the trial, the control treatment is switched to the new intervention at different moments. They typically use cross-sectional data and cluster randomization. On the other hand, new drugs for inhibition of cognitive decline in Alzheimer's or Parkinson's disease typically use delayed start designs (DSDs). Here, participants start in a parallel group design and at a certain moment in the trial, (part of) the control group switches to the new treatment. The studies are longitudinal in nature, and individuals are being randomized. Statistical methods for these unidirectional switch designs (USD) are quite complex and incomparable, and they have been developed by various authors under different terminologies, model specifications, and assumptions. This imposes unnecessary barriers for researchers to compare results or choose the most appropriate method for their own needs. This paper provides an overview of past and current statistical developments for the USDs (SWD and DSD). All designs are formulated in a unified framework of treatment patterns to make comparisons between switch designs easier. The focus is primarily on statistical models, methods of estimation, sample size calculation, and optimal designs for estimation of the treatment effect. Other relevant open issues are being discussed as well to provide suggestions for future research in USDs.

The impact of individual-level heterogeneity on estimated infectious disease burden: a simulation study

Background

Disease burden is not evenly distributed within a population; this uneven distribution can be due to individual heterogeneity in progression rates between disease stages. Composite measures of disease burden that are based on disease progression models, such as the disability-adjusted life year (DALY), are widely used to quantify the current and future burden of infectious diseases. Our goal was to investigate to what extent ignoring the presence of heterogeneity could bias DALY computation.

Methods

Simulations using individual-based models for hypothetical infectious diseases with short and long natural histories were run assuming either “population-averaged” progression probabilities between disease stages, or progression probabilities that were influenced by an a priori defined individual-level frailty (i.e., heterogeneity in disease risk) distribution, and DALYs were calculated.

Results

Under the assumption of heterogeneity in transition rates and increasing frailty with age, the short natural history disease model predicted 14% fewer DALYs compared with the homogenous population assumption. Simulations of a long natural history disease indicated that assuming homogeneity in transition rates when heterogeneity was present could overestimate total DALYs, in the present case by 4% (95% quantile interval: 1–8%).

Conclusions

The consequences of ignoring population heterogeneity should be considered when defining transition parameters for natural history models and when interpreting the resulting disease burden estimates.

Electronic supplementary material

The online version of this article (doi:10.1186/s12963-016-0116-y) contains supplementary material, which is available to authorized users.

Understanding Long-Term Outcomes Following Sepsis: Implications and Challenges

Sepsis is life-threating organ dysfunction due to infection. Incidence of sepsis is increasing and the short-term mortality is improving, generating more sepsis survivors. These sepsis survivors suffer from additional morbidities such as higher risk of readmissions, cardiovascular disease, cognitive impairment and of death, for years following index sepsis episode. In the first year following index sepsis episode, approximately 60 % of sepsis survivors have at least one rehospitalisation episode, which is most often due to infection and one in six sepsis survivors die. Sepsis survivors also have a higher risk of cognitive impairment and cardiovascular disease contributing to the reduced life expectancy seen in this population, when assessed with life table comparisons. For optimal design of interventional trials to reduce these bad outcomes in sepsis survivors, in-depth understanding of major risk factors for these morbid events, their modifiability and a causal relationship to the pathobiology of sepsis is essential. This review highlights the recent advances, clinical and methodological challenges in our understanding of these morbid events in sepsis survivors.

Prostate-specific antigen testing for prostate cancer: Depleting a limited pool of susceptible individuals?

After the introduction of the prostate specific antigen (PSA) test in the 1980s, a sharp increase in the incidence rate of prostate cancer was seen in the United States. The age-specific incidence patterns exhibited remarkable shifts to younger ages, and declining rates were observed at old ages. Similar trends were seen in Norway. We investigate whether these features could, in combination with PSA testing, be explained by a varying degree of susceptibility to prostate cancer in the populations. We analyzed incidence data from the United States' Surveillance, Epidemiology, and End Results program for 1973-2010, comprising 511,027 prostate cancers in men ≥40 years old, and Norwegian national incidence data for 1953-2011, comprising 113,837 prostate cancers in men ≥50 years old. We developed a frailty model where only a proportion of the population could develop prostate cancer, and where the increased risk of diagnosis due to the massive use of PSA testing was modelled by encompassing this heterogeneity in risk. The frailty model fits the observed data well, and captures the changing age-specific incidence patterns across birth cohorts. The susceptible proportion of men is [Formula: see text] in the United States and [Formula: see text] in Norway. Cumulative incidence rates at old age are unchanged across birth cohort exposed to PSA testing at younger and younger ages. The peaking cohort-specific age-incidence curves of prostate cancer may be explained by the underlying heterogeneity in prostate cancer risk. The introduction of the PSA test has led to a larger number of diagnosed men. However, no more cases are being diagnosed in total in birth cohorts exposed to the PSA era at younger and younger ages, even though they are diagnosed at younger ages. Together with the earlier peak in the age-incidence curves for younger cohorts, and the strong familial association of the cancer, this constitutes convincing evidence that the PSA test has led to a higher proportion, and an earlier timing, of diagnoses in a limited pool of susceptible individuals.

How the effects of aging and stresses of life are integrated in mortality rates: insights for genetic studies of human health and longevity

Increasing proportions of elderly individuals in developed countries combined with substantial increases in related medical expenditures make the improvement of the health of the elderly a high priority today. If the process of aging by individuals is a major cause of age related health declines then postponing aging could be an efficient strategy for improving the health of the elderly. Implementing this strategy requires a better understanding of genetic and non-genetic connections among aging, health, and longevity. We review progress and problems in research areas whose development may contribute to analyses of such connections. These include genetic studies of human aging and longevity, the heterogeneity of populations with respect to their susceptibility to disease and death, forces that shape age patterns of human mortality, secular trends in mortality decline, and integrative mortality modeling using longitudinal data. The dynamic involvement of genetic factors in (i) morbidity/mortality risks, (ii) responses to stresses of life, (iii) multi-morbidities of many elderly individuals, (iv) trade-offs for diseases, (v) genetic heterogeneity, and (vi) other relevant aging-related health declines, underscores the need for a comprehensive, integrated approach to analyze the genetic connections for all of the above aspects of aging-related changes. The dynamic relationships among aging, health, and longevity traits would be better understood if one linked several research fields within one conceptual framework that allowed for efficient analyses of available longitudinal data using the wealth of available knowledge about aging, health, and longevity already accumulated in the research field.