Piecewise Exponential Model Research Articles

Cerebral microbleed incidence, relationship to amyloid burden: The Mayo Clinic Study of Aging.

To determine the incidence of cerebral microbleeds (CMBs) and the association of amyloid PET burden with incident CMBs. A total of 651 participants, age ≥50 years (55% male), underwent 3T MRI scans with ≥2 separate T2*-weighted gradient recalled echo sequences from October 2011 to August 2017. Eighty-seven percent underwent 11C Pittsburgh compound B (PiB) PET scans. Age-specific CMB incidence rates were calculated by using the piecewise exponential model. Using structural equation models (SEMs), we assessed the effect of amyloid load and baseline CMBs on future CMBs after considering the direct and indirect age, sex, vascular risk factors, and APOE effects. Participants' mean age (SD) was 69.8 (10.0) years at baseline MRI, and 111 participants (17%) had ≥1 baseline CMB. The mean (SD) of the time interval between scans was 2.7 (1.0) years. The overall population incidence rate for CMBs was 3.6/100 person-years and increased with age: from 1.5/100 new CMBs at age 50 to 11.6/100 person-years at age 90. Using the piecewise exponential model regression, the incidence rates increased with age and the presence of baseline CMBs. The SEMs showed that (1) increasing age at MRI or carrying an APOE4 allele was associated with more amyloid at baseline, and higher amyloid, particularly occipital amyloid load, in turn increased the risk of a new lobar CMB; and (2) the presence of CMBs at baseline increased the risk of a lobar CMB and had a larger effect size than amyloid load. Age and APOE4 carrier status act through amyloid load to increase the risk of subsequent lobar CMBs, but the presence of baseline CMBs is the most important risk factor for future CMBs.

Investigation of the effect of the inspection intervals on the track geometry condition

In order to evaluate the railway track geometry condition and plan maintenance activities, track inspection cars run over the track at specific times to monitor it and record geometry measurements. Applying an adequate inspection interval is vital to ensure the availability, safety and quality of the railway track, at the lowest possible cost. The aim of this study has been to investigate the effect of different inspection intervals on the track geometry condition. To achieve this, an integrated statistical model was developed to predict the track geometry condition given different inspection intervals. In order to model the evolution of the track geometry condition, a piecewise exponential model was used which considers break points at the maintenance times. Ordinal logistic regression was applied to model the probability of the occurrence of severe isolated defects. The Monte Carlo technique was used to simulate the track geometry behaviour given different inspection intervals. The results of the proposed model support the decision-making process regarding the selection of the most adequate inspection interval. The applicability of the model was tested in a case study on the Main Western Line in Sweden.

All-cause mortality after antiretroviral therapy initiation in HIV-positive women from Europe, Sub-Saharan Africa and the Americas.

Women account for over half of persons living with HIV/AIDS globally. We examined geographic variation in all-cause mortality after antiretroviral therapy (ART) for women living with HIV (WLWH) worldwide. We pooled data from WLWH at least 18 years initiating ART 2000-2014 within COHERE (Europe) and IeDEA regions (East Africa, West Africa, South Africa, North America, Latin America/Caribbean). Mortality rates were calculated at 0-3, 3-6, 6-12, 12-24 and 24-48 months after ART, and mortality rate ratios were compared with European rates with piecewise exponential parametric survival models based on Poisson regression. One hundred ninety thousand, one hundred and seventy-five WLWH (16% Europe, 47% East Africa, 13% West Africa, 19% South Africa, 1% South America, 3% North America and 2% Central America/Caribbean) were included. The highest death rates occurred 0-3 months after ART [1.51 (95% CI 1.25-1.82) per 100 person-years in Europe, 12.45 (11.30-13.73), 14.03 (13.12-15.02) and 9.44 (8.80-10.11) in East, West and South Africa, and 1.53 (0.97-2.43), 7.83 (5.44-11.27) and 17.02 (14.62-19.81) in North, South America and Central America/Caribbean, respectively] and declined thereafter. Mortality in Europe was the lowest, with regional differences greatest in the first 3 months and smaller at longer ART durations [adjusted rate ratios 24-48 months after ART: 3.63 (95% CI 3.04-4.33), 5.61 (4.84-6.51) and 3.47 (2.97-4.06) for East, West and South Africa; 2.86 (2.26-3.62), 2.42 (1.65-3.55) and 2.50 (1.92-3.26) for North, South America and Central America/Caribbean, respectively]. Global variations in short-term and long-term mortality among WLWH initiating ART may inform context-specific interventions.

PNS300 HOW TO INTERPRET AND ACCOUNT FOR OVER/UNDER-DISPERSION WHEN MODELLING INCIDENCE RATES WITH POISSON MODELS

Our aim was to investigate the impact of over/under-dispersion when modelling incidence rates (e.g. mortality rate) with Poisson models. Events were simulated using parametric time-to-event models with varying rates, hazard functions (decreasing, constant or increasing hazards), levels of censoring, and observation truncation periods (to emulate a fixed duration study). The simulated data were analysed using Poisson, quasi-Poisson and negative binomial regression models to estimate both bias (estimated rate - true rate) and deviations from equidispersion (mean = variance, a critical Poisson assumption). When constant hazards were assumed, mild over-dispersion was observed. Higher over-dispersion was observed with decreasing hazards, while increasing hazards resulted in under-dispersion. Individual censoring and observation time truncation interacted with the hazard function shape and rate in determining over/under-dispersion. Analysis with Poisson and quasi-Poisson models provided unbiased estimates in cases of no censoring or time-truncation. When high over-dispersion was observed, negative binomial model provided heavily biased estimates of the rates. When Poisson regression is employed to model incidence rates, traditional approaches to account for deviations from equidispersion may not be helpful to understand its extent and the reasons for it. Exploratory analysis of the complementary time-to-event outcome data can provide more insight. If the objectives of the study are estimation, Poisson and quasi-Poisson models are robust to mild deviations from equidispersion. For prediction and extrapolation purposes, the use of parametric time-to-event models is recommended. Alternatively, piecewise exponential models can be used to mimic increasing/decreasing hazards over time. Use of negative binomial models in modelling incidence rates is not recommended.

Leydig Cell Function in Male Survivors of Childhood Cancer: A Report From the St Jude Lifetime Cohort Study.

Direct assessment of Leydig cell function in childhood cancer survivors has been limited. The objectives of this study were to describe the prevalence of and risk factors for Leydig cell failure (LCF), Leydig cell dysfunction (LCD), and associated adverse health outcomes. In this retrospective study with cross-sectional health outcomes analysis, we evaluated 1,516 participants (median age, 30.8 years) at a median of 22.0 years after cancer diagnosis. LCF was defined as serum total testosterone less than 250 ng/dL (or 8.67 nmol/L) and luteinizing hormone greater than 9.85 IU/L, and LCD by testosterone as 250 ng/dL or greater and luteinizing hormone greater than 9.85 IU/L. Polytomous logistic regression evaluated associations with demographic and treatment-related risk factors. Log-binomial regression evaluated associations with adverse physical and psychosocial outcomes. Piecewise exponential models assessed the association with all-cause mortality. The prevalence of LCF and LCD was 6.9% and 14.7%, respectively. Independent risk factors for LCF included an age of 26 years or older at assessment, testicular radiotherapy at any dose, and alkylating agents at cyclophosphamide equivalent doses of 4,000 mg/m2 or greater. The risk increased with older age, higher doses of testicular radiotherapy, and cyclophosphamide equivalent doses. LCF was significantly associated with abdominal obesity, diabetes mellitus, erectile dysfunction, muscle weakness, and all-cause mortality. LCD was associated with unilateral orchiectomy and the same risk factors as LCF; no significant associations were found with adverse physical or psychosocial outcomes. Older age, testicular radiotherapy, and exposure to alkylating agents were associated with LCF, which was associated with adverse physical and psychosexual outcomes. LCD, although having similar risk factors, was not associated with adverse health outcomes. Additional studies are needed to investigate the role of sex hormone replacement in mitigating the burden from adverse outcomes in survivors.

Oseltamivir Plus Usual Care versus Usual Primary Care for Influenza-Like-Illness: An Open-Label, Pragmatic, Randomized Controlled Trial

Background: Antivirals are infrequently prescribed in primary care for influenza-like-illness (ILI), mostly because of perceived ineffectiveness in real world primary care, and as individuals who will especially benefit have not been identified in independent trials. We aimed to determine whether adding antiviral treatment to usual primary care for patients with ILI reduces time to recovery overall and in key subgroups. Methods: We conducted an open-label, pragmatic, randomized controlled trial of adding oseltamivir to usual care in patients aged one year and older consulting with ILI in primary care. The primary endpoint was time to recovery (return to usual activities, with fever, head- and muscle-ache minor/absent), following a Bayesian piece-wise exponential model. Baseline nasopharyngeal swabs were analyzed after study completion. Findings: We ascertained the primary outcome for 3059 of the 3266 participants recruited in 15 European countries during three seasonal influenza seasons (2015-2018); 52.0% (1590/3059) had PCR-confirmed influenza infection. Time to recovery was shorter in those given oseltamivir by 1.02 days (95% Bayesian CI: 0.73-1.32). Hazard ratios of proportional benefit were similar across pre-specified subgroups; absolute benefit in time to recovery varied, with an estimated benefit of >2 days in older patients, more severe illness, comorbidity, or longer prior illness duration. The effect was independent of influenza status. Fewer antibiotics were prescribed in the oseltamivir arm (9.3% vs 13.2%, mean difference 4.0; 95% CI: 1.7-6.3), while nausea and/or vomiting was shorter in usual care patients (HR 0.92; 95% CI: 0.85-1.00). Interpretation: Primary care patients with ILI treated with oseltamivir recovered sooner than those managed by usual care alone, irrespective of influenza virus test results. Older, sicker, patients with comorbidities and longer prior illness duration showed greater absolute benefit. Trial Registration: The trial is registered with the ISRCTN Registry, number ISRCTN27908921. Funding Statement: CB reports receiving advisory board fees from Roche Molecular Systems and grant support from Roche Molecular Diagnostics; CB was supported by funding from an NIHR Protection Research Unit on Health Care Associated Infections and Antimicrobial Resistance, by the NIHR MedTech and In Vitro Diagnostics Co-Operative at Oxford NHS Foundation Trust, and by an NIHR Senior Investigator Award. TV is PI in a project that is funded by Innovative Medicines Initiative (IMI) of the European Union, in which Janssen Pharmaceutica is a partner and researcher in another IMI project in which Biocartis, Janssen, Biomerieux and Berry Consultants are partners. In addition he is co PI of a NIHR funded RCT, and PI in several studies funded by the Netherlands Organization of Health Research and Development. CL has received advisory boards fees from MSD and grant support from GILEAD, Global Bridges, Servier, Galenica-Olvos. Declaration of Interests: CB reports receiving advisory board fees from Roche Molecular Systems and grant support from Roche Molecular Diagnostics; CB was supported by funding from an NIHR Protection Research Unit on Health Care Associated Infections and Antimicrobial Resistance, by the NIHR MedTech and In Vitro Diagnostics Co-Operative at Oxford NHS Foundation Trust, and by an NIHR Senior Investigator Award. TV is PI in a project that is funded by Innovative Medicines Initiative (IMI) of the European Union, in which Janssen Pharmaceutica is a partner and researcher in another IMI project in which Biocartis, Janssen, Biomerieux and Berry Consultants are partners. In addition he is co PI of a NIHR funded RCT, and PI in several studies funded by the Netherlands Organization of Health Research and Development. CL has received advisory boards fees from MSD and grant support from GILEAD, Global Bridges, Servier, Galenica-Olvos. All other authors declare no conflict of interest. Ethics Approval Statement: All participating countries gained national research ethics committees and CTA approval as required. Ethics Ref: 15/SC/0138.

Late-onset anorectal disease and psychosocial impact in survivors of childhood cancer: A report from the Childhood Cancer Survivor Study.

The prevalence and associated psychosocial morbidity of late-onset anorectal disease after surgery and radiotherapy for the treatment of childhood cancer are not known. A total of 25,530 survivors diagnosed between 1970 and 1999 (median age at cancer diagnosis, 6.1years; age at survey, 30.2years) and 5036 siblings were evaluated for late-onset anorectal disease, which was defined as a self-reported fistula-in-ano, self-reported anorectal stricture, or pathology- or medical record-confirmed anorectal subsequent malignant neoplasm (SMN) 5 or more years after the primary cancer diagnosis. Piecewise exponential models compared the survivors and siblings and examined associations between cancer treatments and late-onset anorectal disease. Multiple logistic regression with generalized estimating equations was used to evaluate associations between late-onset anorectal disease and emotional distress, as defined by the Brief Symptom Inventory 18 (BSI-18), and health-related quality of life, as defined by the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36). By 45years after the diagnosis, 394 survivors (fistula, n=291; stricture, n=116; anorectal SMN, n=26) and 84 siblings (fistula, n=73; stricture, n=23; anorectal neoplasm, n=1) had developed late-onset anorectal disease (adjusted rate ratio [RR] for survivors vs siblings, 1.2; 95% confidence interval [CI], 1.0-1.5). Among survivors, pelvic radiotherapy with ≥30Gy within 5years of the cancer diagnosis was associated with late-onset anorectal disease (adjusted RR for 30-49.9Gy vs none, 1.6; 95% CI, 1.1-2.3; adjusted RR for ≥50Gy vs none, 5.4; 95% CI, 3.1-9.2). Late-onset anorectal disease was associated with psychosocial impairment in all BSI-18 and SF-36 domains. Late-onset anorectal disease was more common among childhood cancer survivors who received higher doses of pelvic radiotherapy and was associated with substantial psychosocial morbidity.

Estimation of accelerated failure time due to exposure risk using piecewise-exponential model in cohort follow-up study

Piecewise-exponential excess-relative-risk (ERR) model is used for analysis of large cohort follow-up grouped data of exposure risk. Estimation method for the expected-acceleration-time (EAT) to failure by exposure is proposed using the ERR. Method is applied to all solid cancer mortality atomic-bomb survivor grouped data, and gives EAT to death, 9.9 y (95%CI: 8.4, 11.2 y), at 1 Gy during 50 y follow-up (1950–2000). Longevity (all causes of death) is shorter than the reported before. The exposed younger gives greater EAT. The EAT under carcinogenesis mechanistic model with single-point/chronic exposure is mentioned. Competing causes of cancer death may imply the proneness of double cancer.

An extended state space method for three-dimensional elasticity analysis of arbitrary graded sandwich plates

A novel procedure, within the framework of the state space method in linear elasticity, is proposed for three-dimensional analysis of simply supported, multi-layered plates including any number of arbitrary graded layers at arbitrary locations along the transverse coordinate. First, the conventional state space method used for single layer exponentially graded plates is extended to multi-layered plates including any number of exponentially graded layers. This is achieved by incorporating additional matrices for implementing continuity of the state variables at interfaces involving at least one exponentially graded layer. Further, a piece-wise exponential model is used to extend the procedure to plates that include any number of layers whose material properties vary as arbitrary smooth functions along the transverse coordinate. Comparisons with existing elasticity and approximate solutions show that the method is capable of providing numerical results with better accuracy for all the cases studied. New results are obtained for three-dimensional deformation analysis of sandwich plates with graded face sheets and graded soft core, including a detailed parametric study.

Sexual behaviour and incidence of HIV and sexually transmitted infections among men who have sex with men using daily and event-driven pre-exposure prophylaxis in AMPrEP: 2 year results from a demonstration study

Pre-exposure prophylaxis (PrEP) for HIV might induce risk compensation, defined as increased sexual risk behaviour leading to increased incidence of bacterial sexually transmitted infections (STIs). We examined the incidence of HIV, STIs, and sexual behaviour after initiating daily and event-driven PrEP. AMPrEP is a demonstration study that assessed the incidence of HIV and bacterial STIs and sexual behaviour among PrEP users at the STI clinic of the Public Health Service of Amsterdam (Amsterdam, the Netherlands). Eligible adults (aged ≥18 years) were HIV-negative men who have sex with men (MSM) and transgender people who had reported one or more of the following in the previous 6 months: condomless anal sex with casual partners, at least one bacterial STI, use of post-exposure prophylaxis after a sexual risk incident, or an HIV-positive sexual partner with detectable viral load. Participants were given a choice of daily PrEP or event-driven PrEP regimens. Participants were seen every 3 months for HIV and STI testing. Participants also completed self-administered questionnaires about sexual behaviour at each visit. Over the first 2 years, we assessed the number and incidence rate of HIV and chlamydia, gonorrhoea, and syphilis, including STIs diagnosed in between study visits and assessed changes in incidence over time using piecewise exponential survival models. The number of sexual acts and sexual partners were also analysed and changes over time were assessed using negative binomial regression models. This study is ongoing; the data cutoff for this analysis was June 30, 2018. Between Aug 3, 2015, and May 31, 2016, 376 participants were recruited, of whom nine participants were lost to follow-up, thus 367 participants were included in the analysis. Overall, 365 MSM and two transgender women were included, contributing 681·7 person-years of follow-up. At enrolment, 269 (73%) of 367 participants chose daily PrEP and 98 (27%) participants chose event-driven PrEP. Two individuals in the daily PrEP group became HIV-positive; overall HIV incidence rate was 0·30 per 100 person-years (95% CI 0·07-1·19). 253 participants were diagnosed with one or more STI during the first 2 years (incidence rate 90·4 per 100 person-years). STI incidence did not change over time (adjusted incidence rate ratio [aIRR] 1·00 per 3 month follow-up, 95% CI 0·93-1·07). More than a quarter of all STIs were diagnosed from tests done in between study visits. STI incidence was lower in the event-driven PrEP group than the daily PrEP group (aIRR 0·59, 95% CI 0·46-0·75). The number of condomless anal sex acts with casual partners per 3 months increased, but the number of partners and sex acts remained stable. Although the overall incidence of STIs did not change during 2 years of PrEP use, the incidence of STIs was higher among participants using daily PrEP than those using event-driven PrEP, which is likely a result of differences in sexual behaviour. This finding suggests the need to tailor STI prevention interventions according to behavioural profiles. ZonMw, H-TEAM, Internal GGD research funds, Aidsfonds, Stichting AmsterdamDiner Foundation, Gilead Sciences, Janssen Pharmaceutica, M A C AIDS Fund, and ViiV Healthcare.

Polygenic risk of subsequent thyroid cancer after childhood cancer: A report from St. Jude lifetime cohort (SJLIFE) and Childhood Cancer Survivor Study (CCSS).

10060 Background: Subsequent thyroid cancer (STC) is among the most common malignancies in childhood cancer survivors, especially those with thyroid exposure to radiotherapy (RT). Identification of genetic risk factors may inform screening practices. Methods: Twelve SNPs were previously identified as thyroid cancer risk loci in the general population of European ancestry. A polygenic risk score (PRS) was calculated as a sum of risk alleles carried by a survivor, weighted by the natural logarithm of the published per-allele odds ratios (range: 1.2-1.8). With piecewise exponential models, associations of STC rates with PRS were assessed, both overall and stratified by neck RT exposure. Models were adjusted for sex, age at primary diagnosis, attained age, neck RT dose, epipodophyllotoxin therapy, and eigenvectors within survivors of European ancestry from SJLIFE with whole-genome sequencing data and CCSS with SNP data imputed to Haplotype Reference Consortium. Results: Among 2,324 SJLIFE survivors, 61 (43 with, 18 without neck RT) developed STC. The rate of STC was increased by 5.3-fold (95% confidence interval (CI), 2.2-12.6) and 3.1-fold (CI, 1.3-7.7) for survivors in the third and second PRS tertiles, respectively, compared to those in the first tertile, with corresponding cumulative incidence at age of 40 years of 5.3% (CI, 3.3-7.3%), 2.5% (CI, 1.1-3.9%), and 1.0% (CI, 0.005-2.0%), respectively. Stratified by neck RT, the corresponding rate increases were 7.6 (CI, 2.3-25.3) and 3.8 (CI, 1.1-13.4), respectively, among survivors exposed to neck RT; however, no association was observed among survivors without neck RT (only 18 STC cases). Replication was performed among 4,302 CCSS survivors, 100 (61 with, 39 without neck RT) developed STC. The rates of STC were increased by 2.3-fold (CI, 1.4-3.9) and 1.7-fold (CI, 1.0-2.9) for survivors in the third and the second PRS tertiles, compared to those in the first tertile. The similar significant associations were observed in survivors with and without neck RT ( Ptrend = 0.04 and 0.02, respectively). Conclusions: High PRS conferring STC risk can inform screening practices and help personalize and improve survivorship care.

Chronic health conditions (CHC) and late mortality in survivors of acute lymphoblastic leukemia (ALL) in the Childhood Cancer Survivor Study.

10016 Background: The impact of evolving risk-stratified therapy on long-term morbidity and mortality in survivors of childhood ALL remains largely unknown. Methods: All-cause and health-related late mortality (HRM; captures death from late-effects occurring > 5 yrs from diagnosis), subsequent (malignant) neoplasm [S(M)N], CTCAE graded CHC and neurocognitive outcomes were assessed in 5-yr survivors of ALL diagnosed < 21 yrs of age from 1970-99. Therapy combinations defined 6 groups: 1970s-like ( 70s), standard and high risk 1980s- and 1990s-like ( 80sSR, 80sHR, 90sSR, 90sHR), relapse/transplant ( R/BMT). Cumulative incidence and standardized mortality ratios (SMR) were calculated. Piecewise exponential and log-binomial models estimated rate ratios (RR) with 95% confidence intervals (CI). Results: Among 6148 survivors (median age 31.5 yrs), 15-yr cumulative incidence of all-cause mortality was 5.8% (CI 5.3-6.2) and HRM was 1.5% (1.2-1.7). Compared to 70s, HRM was lower for 90sSR and 90sHR (RR 0.1, CI 0.0-0.3; 0.2, 0.1-0.7), similar to that in the US population (SMR; CI: 90sSR 1.1; 0.6-1.9, 90sHR 1.9; 0.8-3.7). 20-yr cumulative incidence of SN was 3.5% (CI 3.1-3.9). Compared to 70s, 90sSR had lower risk of benign meningioma (RR 0.1, CI 0.0-0.3) and SMN (0.3, 0.1-0.6) with no absolute excess risk compared to the US population. 90sSR was associated with a lower risk of CHCs (Table). Conclusions: More recent risk-stratified therapy has succeeded in reducing risk of late mortality and CHCs among long-term survivors of ALL. [Table: see text]

Late cholecystectomy in survivors of childhood cancer: A report from the Childhood Cancer Survivor Study.

e21525 Background: Cholecystectomy (CCY) is among the most common operations performed in the developed world and is offered as a cure for symptomatic gallbladder disease. Whether survivors of childhood cancer undergo CCY at a higher rate than the general population is unknown. Methods: We identified 5-year survivors diagnosed between 1970 and 1999 who self-reported late (> 5 years after cancer diagnosis) CCY. Rates of CCY were determined among the entire cohort and in association with various risk factors and treatment exposures. Adjusted rate ratios (ARR) were estimated with multivariable piecewise exponential models. Results: Among 24,248 survivors (median follow-up 22.3, interquartile range [IQR] 16.2-30.1 years) and 5,038 siblings (median follow-up 26.4, IQR 19.3-33.7 years), the unadjusted cumulative incidence of CCY at age 50 was 7.2% (n = 757) in survivors and 6.5% (n = 168) in siblings. After adjusting for age, sex and race/ethnicity, survivors underwent CCY at higher rates compared to siblings (ARR = 1.3, 95% CI = 1.1-1.5). Relative to siblings, acute lymphoblastic leukemia survivors underwent CCY at a higher rate (ARR = 1.6, 95% CI = 1.3-2.0), all other diagnoses were not independently associated with higher rates of CCY. Among survivors, risk factors for late CCY included female sex, increasing body mass index (BMI) class, exposure to platinum agents and total body irradiation (TBI) (Table). Conclusions: CCY is performed more commonly among childhood cancer survivors relative to siblings. In addition to known risk factors for gallbladder disease, cancer treatment exposures may further enhance risk for CCY. Awareness and education regarding this observation may ensure timely diagnosis and treatment of symptomatic disease. [Table: see text]

Incorporating individual historical controls and aggregate treatment effect estimates into a Bayesian survival trial: a simulation study

BackgroundPerforming well-powered randomised controlled trials (RCTs) of new treatments for rare diseases is often infeasible. However, with the increasing availability of historical data, incorporating existing information into trials with small sample sizes is appealing in order to increase the power. Bayesian approaches enable one to incorporate historical data into a trial’s analysis through a prior distribution.MethodsMotivated by a RCT intended to evaluate the impact on event-free survival of mifamurtide in patients with osteosarcoma, we performed a simulation study to evaluate the impact on trial operating characteristics of incorporating historical individual control data and aggregate treatment effect estimates. We used power priors derived from historical individual control data for baseline parameters of Weibull and piecewise exponential models, while we used a mixture prior to summarise aggregate information obtained on the relative treatment effect. The impact of prior-data conflicts, both with respect to the parameters and survival models, was evaluated for a set of pre-specified weights assigned to the historical information in the prior distributions.ResultsThe operating characteristics varied according to the weights assigned to each source of historical information, the variance of the informative and vague component of the mixture prior and the level of commensurability between the historical and new data. When historical and new controls follow different survival distributions, we did not observe any advantage of choosing a piecewise exponential model compared to a Weibull model for the new trial analysis. However, we think that it remains appealing given the uncertainty that will often surround the shape of the survival distribution of the new data.ConclusionIn the setting of Sarcome-13 trial, and other similar studies in rare diseases, the gains in power and accuracy made possible by incorporating different types of historical information commensurate with the new trial data have to be balanced against the risk of biased estimates and a possible loss in power if data are not commensurate. The weights allocated to the historical data have to be carefully chosen based on this trade-off. Further simulation studies investigating methods for incorporating historical data are required to generalise the findings.

Risk factors associated with revision for prosthetic joint infection following knee replacement: an observational cohort study from England and Wales

SummaryBackgroundProsthetic joint infection is a devastating complication of knee replacement. The risk of developing a prosthetic joint infection is affected by patient, surgical, and health-care system factors. Existing evidence is limited by heterogeneity in populations studied, short follow-up, inadequate power, and does not differentiate early prosthetic joint infection, most likely related to the intervention, from late infection, more likely to occur due to haematogenous bacterial spread. We aimed to assess the overall and time-specific associations of these factors with the risk of revision due to prosthetic joint infection following primary knee replacement.MethodsIn this cohort study, we analysed primary knee replacements done between 2003 and 2013 in England and Wales and the procedures subsequently revised for prosthetic joint infection between 2003 and 2014. Data were obtained from the National Joint Registry linked to the Hospital Episode Statistics data in England and the Patient Episode Database for Wales. Each primary replacement was followed for a minimum of 12 months until the end of the observation period (Dec 31, 2014) or until the date of revision for prosthetic joint infection, revision for another indication, or death (whichever occurred first). We analysed the data using Poisson and piecewise exponential multilevel models to assess the associations between patient, surgical, and health-care system factors and risk of revision for prosthetic joint infection.FindingsOf 679 010 primary knee replacements done between 2003 and 2013 in England and Wales, 3659 were subsequently revised for an indication of prosthetic joint infection between 2003 and 2014, after a median follow-up of 4·6 years (IQR 2·6–6·9). Male sex (rate ratio [RR] for male vs female patients 1·8 [95% CI 1·7–2·0]), younger age (RR for age ≥80 years vs <60 years 0·5 [0·4–0·6]), higher American Society of Anaesthesiologists [ASA] grade (RR for ASA grade 3–5 vs 1, 1·8 [1·6–2·1]), elevated body-mass index (BMI; RR for BMI ≥30 kg/m2vs <25 kg/m2 1·5 [1·3–1·6]), chronic pulmonary disease (RR 1·2 [1·1–1·3]), diabetes (RR 1·4 [1·2–1·5]), liver disease (RR 2·2 [1·6–2·9]), connective tissue and rheumatic diseases (RR 1·5 [1·3–1·7]), peripheral vascular disease (RR 1·4 [1·1–1·7]), surgery for trauma (RR 1·9 [1·4–2·6]), previous septic arthritis (RR 4·9 [2·7–7·6]) or inflammatory arthropathy (RR 1·4 [1·2–1·7]), operation under general anaesthesia (RR 1·1 [1·0–1·2]), requirement for tibial bone graft (RR 2·0 [1·3–2·7]), use of posterior stabilised fixed bearing prostheses (RR for posterior stabilised fixed bearing prostheses vs unconstrained fixed bearing prostheses 1·4 [1·3–1·5]) or constrained condylar prostheses (3·5 [2·5–4·7]) were associated with a higher risk of revision for prosthetic joint infection. However, uncemented total, patellofemoral, or unicondylar knee replacement (RR for uncemented vs cemented total knee replacement 0·7 [95% CI 0·6–0·8], RR for patellofemoral vs cemented total knee replacement 0·3 [0·2–0·5], and RR for unicondylar vs cemented total knee replacement 0·5 [0·5–0·6]) were associated with lower risk of revision for prosthetic joint infection. Most of these factors had time-specific effects, depending on the time period post-surgery.InterpretationWe have identified several risk factors for revision for prosthetic joint infection following knee replacement. Some of these factors are modifiable, and the use of targeted interventions or strategies could lead to a reduced risk of revision for prosthetic joint infection. Non-modifiable factors and the time-specific nature of the effects we have observed will allow clinicians to appropriately counsel patients preoperatively and tailor follow-up regimens.FundingNational Institute for Health Research.

OR18-1 Leydig Cell Function in Adult Survivors of Childhood Cancer

Context: Direct assessment of Leydig cell function in adult survivors of childhood cancer has been limited. Objectives: To describe the prevalence of, and risk factors for, Leydig cell failure (LCF) and associated adverse health outcomes. Design: Retrospective cohort with cross-sectional health outcomes analysis. Patients: 1534 participants (median age 30.8 years; range 18.1-63.8) evaluated at a median of 22.0 years (range 7.5-49.8) after cancer diagnosis. Survivors with LH/FSH deficiency were excluded. Main Outcome Measure: LCF was defined as serum total testosterone < 250 ng/dL (8.67 nmol/L) combined with LH >8.6 IU/L; compensated LCF by testosterone ≥ 250 ng/dL and LH >8.6 IU/L. Polytomous logistic regression evaluated associations between demographic and treatment-related risk factors and LCF or compensated LCF. Log-binomial regression examined associations between these diagnoses and risk of diabetes mellitus, dyslipidemia, abdominal obesity, frailty, erectile dysfunction, and psychological distress. Piecewise exponential models analyzed the association between mortality and LCF/compensated LCF. Results: The prevalence of LCF and compensated LCF was 8.0% (95% confidence interval (CI) 6.7%-9.4%) and 22.8% (95% CI 20.7%-25.0%), respectively. Individuals aged 36-45.9 years (odds ratio (OR) 3.8, 95% CI 1.9-7.9) or ≥46 years (OR 4.9, 95% CI 2.2-11.1) at the time of study had a significantly higher risk of LCF than those 18-25 years old. Participants treated with direct testicular radiotherapy at doses >0-19.9 (OR 81.6, 95% CI 24.9-266.7) or ≥ 20 Gy (OR >999, 95% CI 109.9->999) had a higher risk of LCF than those not exposed to this treatment. Individuals treated with alkylating agents at cyclophosphamide equivalent doses 4000-7999 mg/m2 (OR 3.6, 95% CI 1.9-6.9) or 8000-11999 mg/m2 (OR 3.7, 95% CI 1.9-7.3) or ≥ 12000 mg/m2 (OR 8.7, 95% CI 4.7-16.4) had a higher risk of LCF than those not exposed to these agents. LCF was independently associated with abdominal obesity (prevalence ratio (PR) 1.6, 95% CI 1.2-2.1), diabetes mellitus (PR 2.9, 95% CI 1.8-4.6), erectile dysfunction (PR 1.8, 95% CI 1.4-2.5), frailty (PR 2.5, 95% CI 1.2-5.3) and mortality (PR 4.8, 95% CI 2.3-10.1). For compensated LCF, the risk factor associations were similar to those found for LCF; however, no significant associations were found with adverse physical or psychosocial outcomes. Conclusion: In adult survivors of childhood cancer, older age, direct testicular radiotherapy and high dose alkylating agents were associated with LCF, which was associated with poor general health outcomes. Further studies are needed to investigate the role of sex hormone replacement in mitigating such outcomes.

Therapy-Related Cardiac Risk in Childhood Cancer Survivors: An Analysis of the Childhood Cancer Survivor Study.

The impacts of radiotherapy dose and exposed cardiac volume, select chemotherapeutic agents, and age at exposure on risk for late-onset cardiac disease in survivors of childhood cancer remain unresolved. We determined the rates of severe to fatal cardiac disease in 24,214 5-year survivors in the Childhood Cancer Survivor Study diagnosed between 1970 and 1999 at a median age of 7.0 years (range, 0 to 20.9 years), with a median attained age of 27.5 years (range, 5.6 to 58.9 years). Using piecewise exponential models, we evaluated the association between cardiac disease rates and demographic and treatment characteristics. The cumulative incidence of cardiac disease 30 years from diagnosis was 4.8% (95% CI, 4.3 to 5.2). Low to moderate radiotherapy doses (5.0 to 19.9 Gy) to large cardiac volumes (≥ 50% of heart) were associated with an increased rate of cardiac disease (relative rate, 1.6; 95% CI, 1.1 to 2.3) compared with survivors without cardiac radiotherapy exposure. Similarly, high doses (≥ 20 Gy) to small cardiac volumes (0.1% to 29.9%) were associated with an elevated rate (relative rate, 2.4; 95% CI, 1.4 to 4.2). A dose-response relationship was observed between anthracycline chemotherapy and heart failure with younger children (age ≤ 13 years) at the greatest risk for heart failure after comparable dosing. These observations support advances in radiation field design and delivery technology to reduce cardiac dose/volume and should guide future treatment protocols. They also inform clinical practice guidelines for post-therapy surveillance and risk-reducing strategies.

Global Contraceptive Failure Rates: Who Is Most at Risk?

Contraceptive failure is a major contributor to unintended pregnancy worldwide. DHS retrospective calendars, which are the most widely used data source for estimating contraceptive failure in low‐income countries, vary in quality across countries and surveys. We identified surveys with the most reliable calendar data and analyzed 105,322 episodes of contraceptive use from 15 DHSs conducted between 1992 and 2014. We estimate contraceptive method‐specific 12‐month failure rates. We also examined how failure rates vary by age, education, socioeconomic status, contraceptive intention, residence, and marital status using multilevel piecewise exponential hazard models. Our failure rate estimates are significantly lower than results from the United States and slightly higher than previous studies that included more DHS surveys, including some with lower‐quality data. We estimate age‐specific global contraceptive failure rates and find strong, consistent age patterns with the youngest users experiencing failure rates up to ten times higher than older women for certain methods. Failure also varies by socioeconomic status, with the poorest, and youngest, women at highest risk of experiencing unintended pregnancy due to failure.

Good times, bad times: innovation and survival over the business cycle

High-potential new ventures are a source of economic growth, which policy makers call upon in times of crisis when entrepreneurship is seen as a remedy to economic downturn. Yet at these times new ventures face intensified selection, and survival hinges on heterogeneous capabilities. We examine how the innovative capabilities of new firms created in the Netherlands in 2001-2006, affected their survival likelihood before, during and after the 2007-2008 global financial crisis. We estimate a piecewise exponential model linking survival times, observed in the time period from 2001 to 2015, to longitudinal innovation data from the CIS. Our results show that new ventures innovating within two years from founding benefit of a long-term adaptive survival premium during and after the crisis. This premium and its duration over the stages of the crisis are contingent to the form of innovation: technological innovations entail a more effective and enduring premium, as compared to managerial innovations, which can be even detrimental for survival. Our study has implications for entrepreneurial management, by highlighting how the development of innovative capabilities at founding, lays the foundations for organisational adaptation and resilience in the longer term. Furthermore, our results can inform a policy approach that aims at sheltering from the storm of a financial crisis, those new ventures that do possess the specific and necessary adaptive capabilities, but that are also vulnerable because of the liabilities of newness and smallness. Such an approach could help to maintain alive the process of entrepreneurial experimentation during the crisis, and to boost economic recovery, without dispersing precious resources.

Late Morbidity and Mortality Among Medulloblastoma Survivors Diagnosed Across Three Decades: A Report From the Childhood Cancer Survivor Study.

Treatment of medulloblastoma has evolved from surgery and radiotherapy to contemporary multimodal regimens. However, the impact on long-term health outcomes remains unknown. Cumulative incidence of late mortality (5 or more years from diagnosis), subsequent neoplasms (SNs), and chronic health conditions were evaluated in the Childhood Cancer Survivor Study among 5-year survivors of medulloblastoma diagnosed between 1970 and 1999. Outcomes were evaluated by treatment exposure, including historical therapy (craniospinal irradiation [CSI] ≥ 30 Gy, no chemotherapy), high risk (CSI ≥ 30 Gy + chemotherapy), standard risk (CSI < 30 Gy + chemotherapy), and by treatment decade (1970s, 1980s, 1990s). Rate ratios (RRs) and 95% CIs estimated long-term outcomes using multivariable piecewise exponential models. Among 1,311 eligible survivors (median age, 29 years [range, 6 to 60 years]; median time from diagnosis, 21 years [range, 5 to 44 years]), the 15-year cumulative incidence rate of all-cause late mortality was 23.2% (diagnosed 1970s) versus 12.8% (1990s; P = .002), with a recurrence-related mortality rate of 17.7% versus 9.6% ( P = .008). Lower late mortality rates as a result of other health-related causes were not observed. Among 997 survivors who completed a baseline survey, the 15-year cumulative incidence of SNs was higher among survivors with multimodal therapy (standard risk, 9.5%; historical, 2.8%; P = .03). Survivors treated in the 1990s had a higher cumulative incidence of severe, disabling, life-threatening, and fatal chronic health conditions (56.5% in 1990s v 39.9% in 1970s; P < .001) and were more likely to develop multiple conditions (RR, 2.89; 95% CI, 1.31 to 6.38). However, survivors of standard-risk therapy were less likely to use special education services than high-risk therapy survivors (RR, 0.84; 95% CI, 0.75 to 0.93). Historical changes in medulloblastoma therapy that improved 5-year survival have increased the risk for SNs and debilitating health conditions for survivors yet reduced the need for special education services.