Mixed Proportional Hazard Model Research Articles

Residential green space and age at menarche in German and Australian adolescent girls: A longitudinal study.

A large multicentre European study reported later onset of menopause among women residing in greener areas. This influence on the timing of a reproductive event like menopause, raises the question whether similar associations can be observed with timing of menarche. We investigated whether exposure to residential green space was related to the age at menarche in German and Australian adolescent girls. The analytic samples comprised of 1706 German and 1474 Australian adolescent girls. Percentage of green space was calculated in 1000m buffers around a residential address or its surrogate at the previous follow-up. Mixed effects Cox proportional hazard models were used to explore the associations. The survival object was the occurrence of menarche at the time of follow-up (15-year follow-up of the German cohorts and the study wave at 14-15 years in the Australian cohort) and number of years since baseline (10-year follow-up in the German cohort and the study wave at 10-11 years in the Australian cohort). Participants who did not reach menarche were included as censored observations. A greener residence was not associated with the age at menarche. Null findings were consistent in the general population and in analyses stratified by socioeconomic status or urbanicity in both countries. Urban residents were more likely to have earlier menarche, and this association was consistent across Germany and Australia. The results of our analysis do not support the hypothesis that residing in places with more green space can influence timing of menarche. However, given the limitations of our study, researchers should not be discouraged to further explore environmental risk factors of early menarche.

Joint modeling of time to diabetic retinopathy and change in fasting blood sugar among type 2 diabetic patients, Northwest Ethiopia

This study aimed to assess changes in fasting blood sugar (FBS) levels, time to diabetic retinopathy (DR) and its predictors among type 2 diabetes patients in Ethiopia. An institution-based retrospective follow-up study was conducted at the University of Gondar Comprehensive Specialized Hospital. The linear mixed effect model and Cox proportional hazard models were fitted separately, and later, the two models were fitted jointly using R software. Variables with a p value < 0.05 were considered significant predictors in the adjusted analysis. The incidence rate of DR was 2 per 100-person year of observation with a median follow-up time of 90.8 months (IQR 63.4). The current value and rate of change in FBS level were significant predictors of time to DR (AHR = 1.35; 95% CI 1.12–1.63) and (AHR = 1.70; 95% CI 1.21–2.39), respectively. Hypertension (AHR = 2.49; 95% CI 1.32–4.66), taking > 1 antidiabetic oral agent (AHR = 4.90; 95% CI 1.07–20.0) and more than 10 years duration (AHR = 0.17, 95% CI 0.06–0.46) were predictors of time to DR. This study revealed that the current value of FBS and the rate of FBS change were significantly associated with the time to DR.

Socioeconomic disadvantage and human papillomavirus (HPV) vaccination uptake

ImportanceDespite availability of safe and effective human papillomavirus (HPV) vaccines, vaccination uptake remains low in the U.S. Research examining the impact of neighborhood socioeconomic status on HPV vaccination may help target interventions. ObjectiveTo examine the association between area deprivation and HPV vaccine initiation and completion. Design, setting, participantsRetrospective cohort study of individuals aged 11–18 years residing in the upper Midwest region. Receipt of HPV vaccination was examined over a three-year follow-up period (01/01/2016–12/31/2018). Main outcomes and measuresOutcomes of interest were initiation and completion of HPV vaccination. Demographic data were collected from the Rochester Epidemiology Project (REP). Area-level socioeconomic disadvantage was measured by calculating an Area Deprivation Index (ADI) score for each person, a measure of socioeconomic disadvantage derived from American Community Survey data. Multivariable mixed effect Cox proportional hazards models were used to examine the association of ADI quartiles (Q1-Q4) with HPV vaccine series initiation and completion, given initiation. ResultsIndividuals residing in census block groups with higher deprivation had significantly lower likelihood of HPV vaccine initiation (Q2: HR = 0.91, 0.84–0.99 Q3: HR = 0.83, 0.76–0.90; Q4: HR = 0.84, 0.74–0.96) relative to those in the least-deprived block groups (Q1). Similarly, those living in block groups with higher deprivation had significantly lower likelihood of completion (Q2: HR = 0.91, 0.86–0.97; Q3: HR = 0.87, 0.81–0.94; Q4: HR = 0.82, 0.74–0.92) compared to individuals in the least-deprived block groups (Q1). Conclusions and relevanceLower probability of both HPV vaccine-series initiation and completion were observed in areas with greater deprivation. Our results can inform allocation of resources to increase HPV vaccination rates in our primary care practice and provide an example of leveraging public data to inform similar efforts across diverse health systems.

Survival of colorectal cancer: the role of contextual, individual and tumour-specific characteristics

Abstract Background Colorectal cancer (CRC) is the second most common cancer in women and the third most common cancer in men in Germany. One in two people suffering from CRC dies within five years after diagnosis. The aim of our ongoing research project is to investigate the interaction of contextual, individual and tumour-specific characteristics and their effect on CRC survival in Bavaria. Methods A total of 124,958 incident cases of CRC (ICD-10 C18-C20) from the Bavarian Cancer Registry are linked with data on area deprivation based on the Bavarian Index of Multiple Deprivation (BIMD). We include all years with a CRC-specific case completeness of 90% or more, i.e. 2004 to 2018. We calculate Kaplan-Meier curves as well as mixed proportional hazards models including non-linear effects. Our modelling strategy starts from a baseline model with random effects at district, municipality, and individual level, which we continuously extend to include regional (deprivation), individual (age at diagnosis, sex) as well as tumour-specific variables (tumour site, grading, cancer stage, histological subtype, diagnosis of multiple primary tumours). Results Preliminary results show significant random effects at district and municipality level as well as a negative impact of deprivation on CRC survival for all sites. Individual as well as tumour-specific characteristics, such as age, grading, and cancer stage, demonstrate the strongest effects on CRC survival and partly explain survival differences according to area level (districts, municipalities), and level of deprivation. The effects of individual as well as tumour-specific characteristics vary between districts and municipalities in Bavaria depending on their level of area deprivation. Conclusions Contextual, individual and tumour-specific characteristics have an impact on CRC survival. Future studies may also include health services characteristics in CRC survival analysis. Key messages CRC survival is inversely associated with area deprivation in Bavaria. The effects of individual and tumour-specific characteristics, e.g. age and cancer stage, increase with the level of deprivation.

Long-term prognosis of teeth with class III furcation involvement.

Evaluation of survival of teeth with class III furcation involvement (FI) ≥5 years after active periodontal treatment (APT) and identification of prognostic factors. All charts of patients who completed APT at the Department of Periodontology of Goethe-University Frankfurt, Germany, beginning October 2004 were screened for teeth with class III FI. APT had to be accomplished for ≥5 years. Charts were analysed for data of class III FI teeth at baseline (T0), at accomplishment of APT (T1), and at the last supportive periodontal care (T2). Baseline radiographic bone loss (RBL) and treatment were assessed. One-hundred and sixty patients (age: 54.4 ± 9.8 years; 82 females; 39 active smokers; 9 diabetics, 85 stage III, 75 stage IV, 59 grade B, 101 grade C) presented 265 teeth with class III FI. Ninety-eight teeth (37%) were lost during 110, 78/137 (median, lower/upper quartile) months. Logistic mixed-model regression and mixed Cox proportional hazard model associated adjunctive systemic antibiotics with fewer tooth loss (26% vs. 42%; p=.019/.004) and RBL (p=.014/.024) and mean probing pocket depth (PPD) at T1 (p < .001) with more tooth loss. Subgingival instrumentation with adjunctive systemic antibiotics favours retention of class III furcation-involved teeth. Baseline RBL and PPD at T1 deteriorate long-term prognosis.

Changes in Serum Thyroid Function Predict Cognitive Decline in the Very Old: Longitudinal Findings from the Newcastle 85+ Study.

Background: Low serum thyrotropin (TSH) has been associated with an increased risk of cognitive impairment in observational studies of older individuals, but the mechanism underlying this is unclear. We investigated the association between changes in thyroid status and cognitive impairment in very old adults, using prospective data from the Newcastle 85+ study. Method: A cohort of 85-year-old individuals was assessed for health status and thyroid function. Complete data from a comprehensive multidimensional measure of health and repeat thyroid function were available for 642 participants with normal free thyroid hormones and TSH levels ranging between 0.1 and 10 mU/L. Cognitive performance, assessed using Mini-Mental State Examination (MMSE) and Cognitive Drug Research battery was examined by using linear mixed, logistic regression, and Cox proportional hazard models in relation to baseline and 3-year changes in serum TSH, free thyroxine (fT4), and free triiodothyronine (fT3). Results: Over 3 years, declining serum TSH was associated with reductions in fT4 and fT3, and an increased risk of incident cognitive impairment by 5 years (odds ratio1.77 [95% confidence interval: 1.19-2.61]; p = 0.004). A greater reduction in MMSE score was associated with larger TSH decline, at 3 (p = 0.001) and 5 years (p < 0.001), respectively. Steady fT4 concentrations were found in participants with rising TSH. Conclusions: In contrast to physiological expectation, in this group of 85-year-olds, a declining serum TSH was associated with reductions in free thyroid hormones over time. A decreasing serum TSH trajectory over time anticipated cognitive decline in later life. Declining TSH concentrations are a biomarker for cognitive impairment in later life.

Longitudinal Investigation of Pubertal Milestones and Hormones as a Function of Body Fat in Girls

Background: Studies comparing the timing and pace of puberty in overweight/obese girls (OW/OB) vs normal weight girls (NW) have produced conflicting results; some suggest earlier activation of the central components of the reproductive axis in OB while others are more consistent with a peripheral source of estrogen (e.g. adipose tissue) driving puberty in OB. Importantly, there have been no longitudinal assessments of both clinical and biochemical pubertal markers in OB vs. NW. Methods: 90 healthy pre-menarchal girls (26 OW/OB, 54 NW) from the community, aged 8.2-14.7 years, completed 2.8 ± 1.7 (mean, SD) study visits over the course of 4 years. Visits included dual-energy x-ray absorptiometry to calculate percent total body fat (TBF), Tanner staging, breast ultrasound for morphological staging (BMORPH; stages A-E), pelvic ultrasound, hand x-ray (bone age, BA), blood tests for reproductive hormones, and urine collection to determine a vaginal maturation index (VMI), an index of estrogen exposure in urogenital epithelial cells. Menarchal status was determined at each visit and via follow-up questionnaires. The effect of TBF on hormones and markers of estrogen action, the pace of breast maturation, and age at menarche were determined using a mixed, multi-state, or Cox proportional hazards model, respectively. Mixed and Cox models controlled for BMORPH at visit 1 (V1) and race. Results: NW girls were older than OW/OB (11.3 vs. 10.2 yrs, p<0.01) at V1, more likely to be non-Hispanic White (66 vs. 40%, p=0.03), and had more advanced breast morphology BMORPH (p<0.01). LH, E2, VMI, BA, and ovarian and uterine volume increased with time with no effect of TBF. There was an interaction between time and TBF for FSH, INHB, E1, Total T, Free T, and A’dione (p<0.05): levels were initially similar in all TBF groups, but after 1 yr, levels increased in girls with higher TBF, plateaued in girls with mid-range TBF, and decreased in girls with lower TBF. Girls with higher TBF progressed through BMORPH stage D (corresponding to growth/arborization of the breast ductal system), more slowly than girls with lower TBF but achieved menarche at a younger age (risk 1.04x higher per 1 unit increase in TBF). Conclusions: Intensive reproductive phenotyping of girls during the pubertal transition reveals that both the neuroendocrine and ovarian components of the axis are generally preserved in girls with higher TBF but that the axis appears to be activated earlier than in girls with lower TBF. In late puberty, however, girls with higher TBF demonstrate subtle differences in standard hormonal (e.g. serum FSH, INHB, and androgen) and clinical (e.g. delayed growth of breast bud) markers of puberty. Investigation of the mechanistic basis for these differences and their potential clinical consequences for girls with higher TBF deserves further study.

Gene-based analysis of bi-variate survival traits via functional regressions with applications to eye diseases.

Genetic studies of two related survival outcomes of a pleiotropic gene are commonly encountered but statistical models to analyze them are rarely developed. To analyze sequencing data, we propose mixed effect Cox proportional hazard models by functional regressions to perform gene-based joint association analysis of two survival traits motivated by our ongoing real studies. These models extend fixed effect Cox models of univariate survival traits by incorporating variations and correlation of multivariate survival traits into the models. The associations between genetic variants and two survival traits are tested by likelihood ratio test statistics. Extensive simulation studies suggest that type I error rates are well controlled and power performances are stable. The proposed models are applied to analyze bivariate survival traits of left and right eyes in the age-related macular degeneration progression.

Longitudinal Investigation of Pubertal Milestones and Hormones as a Function of Body Fat in Girls

Epidemiologic studies demonstrated that overweight/obese girls (OW/OB) undergo thelarche and menarche earlier than normal weight girls (NW). There have been no longitudinal studies to specifically investigate how body weight/fat affects both clinical and biochemical pubertal markers in girls. 90 girls (36 OW/OB, 54 NW), aged 8.2-14.7 years, completed 2.8 ± 1.7 study visits over the course of four years. Visits included dual-energy x-ray absorptiometry to calculate total body fat (TBF), Tanner staging, breast ultrasound for morphological staging (BMORPH; A-E), pelvic ultrasound, hormone tests, and assessment of menarchal status. The effect of TBF on pubertal markers was determined using a mixed, multi-state, or Cox proportional hazards model, controlling for baseline BMORPH. NW were older than OW/OB (11.3 vs. 10.2 yrs, p<0.01) at baseline and had more advanced BMORPH (p<0.01). LH, estradiol, and ovarian and uterine volumes increased with time with no effect of TBF. There was a time x TBF interaction for FSH, inhibin B, estrone, total and free testosterone, and androstenedione: levels were initially similar, but after 1 yr, levels increased in girls with higher TBF, plateaued in girls with mid-range TBF, and decreased in girls with lower TBF. Girls with higher TBF progressed through BMORPH stage D more slowly but achieved menarche earlier than girls with lower TBF. In late puberty, girls with higher TBF demonstrate differences in standard hormonal and clinical markers of puberty. Investigation of the underlying causes and clinical consequences of these differences in girls with higher TBF deserves further study.

Cerebrospinal fluid \u03b1-synuclein predicts neurodegeneration and clinical progression in non-demented elders

BackgroundAccumulating reports have suggested that α-synuclein is involved in the pathogenesis of Alzheimer’s disease (AD). As the cerebrospinal fluid (CSF) α-synuclein has been suggested as a potential biomarker of AD, this study was set out to test whether CSF α-synuclein is associated with other AD biomarkers and could predict neurodegeneration and clinical progression in non-demented elders.MethodsThe associations between CSF α-synuclein and other AD biomarkers were investigated at baseline in non-demented Chinese elders. The predictive values of CSF α-synuclein for longitudinal neuroimaging change and the conversion risk of non-demented elders were assessed using linear mixed effects models and multivariate Cox proportional hazard models, respectively, in the Alzheimer’s disease Neuroimaging Initiative (ADNI) database.ResultsThe CSF α-synuclein levels correlated with AD-specific biomarkers, CSF total tau and phosphorylated tau levels, in 651 Chinese Han participants (training set). These positive correlations were replicated in the ADNI database (validation set). Using a longitudinal cohort from ADNI, the CSF α-synuclein concentrations were found to increase with disease severity. The CSF α-synuclein had high diagnostic accuracy for AD based on the “ATN” (amyloid, tau, neurodegeneration) system (A + T+ versus A − T − control) (area under the receiver operating characteristic curve, 0.84). Moreover, CSF α-synuclein predicted longitudinal hippocampus atrophy and conversion from MCI to AD dementia.ConclusionsCSF α-synuclein is associated with CSF tau levels and could predict neurodegeneration and clinical progression in non-demented elders. This finding indicates that CSF α-synuclein is a potentially useful early biomarker for AD.

Do funding sources influence long-term patient survival in pediatric liver transplantation?

Socioeconomic status has been associated with inferior outcomes after multiple surgical procedures, but has not been well studied with respect to pediatric liver transplantation. This study evaluated the impact of insurance status (as a proxy for socioeconomic status) on patient and allograft survival in pediatric first-time liver transplant recipients. Our retrospective analysis of the UNOS data base from January 2002 through September 2017 revealed 6997 pediatric patients undergoing first-time isolated liver transplantation. A mixed Cox proportional hazards model adjusted for donor, recipient, and program characteristics determined the RR of insurance status on allograft and patient survival. All results were considered significant at P<.05. All statistical results were obtained using R version 3.5.1 and coxme version 2.2-10. Medicaid status had a significant negative impact on long-term survival after controlling for multiple covariates. Pediatric patients undergoing first-time isolated liver transplantation with Medicaid insurance had a RR of 1.42 [confidence interval: 1.18-1.60] of post-transplant death. Pediatric patients undergoing first-time isolated liver transplantation have multiple risk factors that may impact long-term survival. Having Medicaid insurance almost doubles the chances of dying post-liver transplant. This patient population may require more global support post-transplant to improve long-term survival.

Using chitosan microparticles to treat metritis in lactating dairy cows

The main objective of this study was to evaluate the efficacy of intrauterine administration of chitosan microparticles (CM) in curing metritis in dairy cows. A secondary objective was to evaluate the effects of metritis treatments on milk yield, survival, and reproductive performance. Cows with a fetid, watery, red-brownish vaginal discharge were diagnosed with metritis. Holstein cows (n = 826) with metritis from 3 dairies located in northern Florida were blocked by parity (primiparous or multiparous) and, within each block, randomly assigned to one of 3 treatments: CM (n = 276) = intrauterine infusion of 24 g of CM dissolved in 40 mL of sterile distilled water at the time of metritis diagnosis (d 0), 2 (d 2), and 4 (d 4) d later; ceftiofur (CEF; n = 275) = subcutaneous injection of 6.6 mg/kg ceftiofur crystalline-free acid in the base of the ear at d 0 and d 3; Control (CON; n = 275) = no treatment applied at metritis diagnosis. All groups could receive escape therapy if condition worsened. Cure was considered when vaginal discharge became mucoid and not fetid. A group of nonmetritic (NMET; n = 2,436) cows was used for comparison. Data were analyzed by generalized linear mixed and Cox's proportional hazard models. Cows in CM and CON had lesser risk of metritis cure on d 12 than cows in CEF (58.6 ± 5.0 vs. 61.9 ± 4.9% vs. 77.9 ± 3.9, respectively). The proportion of cows culled within 60 days in milk (DIM) was greater for cows in CM than for cows in CEF and CON (21.5 ± 2.7 vs. 9.7 ± 1.9 vs. 11.3 ± 2.0%, respectively). Treatment did not affect rectal temperature or plasma nonesterified fatty acids, β-hydroxybutyrate, and haptoglobin concentrations. Milk yield in the first 60 DIM differed for all treatments, and it was lowest for CM (35.8 ± 0.3 kg/d), followed by CON (36.8 ± 0.3 kg/d) and CEF (37.9 ± 0.3 kg/d). The hazard of pregnancy up to 300 DIM was lesser for CM than CEF (hazard ratio = 0.62; 95% CI: 0.50-0.76), for CM than CON (hazard ratio = 0.77; 95% CI: 0.62-0.95) and for CON than CEF (hazard ratio = 0.80; 95% CI: 0.65-0.99). Culling was greater, and milk yield and fertility were lesser for CEF than NMET. In summary, CM did not improve the cure of metritis, and was detrimental to milk yield, survival, and fertility compared with CON. In contrast, CEF increased the cure of metritis, milk yield, and fertility compared with CM and CON. Finally, the negative effects of metritis on milk yield culling and fertility could not be completely reversed by CEF.

Heap: A command for fitting discrete outcome variable models in the presence of heaping at known points

Self-reported survey data are often plagued by the presence of heaping. Accounting for this measurement error is crucial for the identification and consistent estimation of the underlying model (parameters) from such data. In this article, we introduce two commands. The first command, heapmph, estimates the parameters of a discrete-time mixed proportional hazard model with gammaunobserved heterogeneity, allowing for fixed and individual-specific censoring and different-sized heap points. The second command, heapop, extends the framework to ordered choice outcomes, subject to heaping. We also provide suitable specification tests.

TESTING A PARAMETRIC TRANSFORMATION MODEL VERSUS A NONPARAMETRIC ALTERNATIVE

Despite an abundance of semiparametric estimators of the transformation model, no procedure has been proposed yet to test the hypothesis that the transformation function belongs to a finite-dimensional parametric family against a nonparametric alternative. In this article, we introduce a bootstrap test based on integrated squared distance between a nonparametric estimator and a parametric null. As a special case, our procedure can be used to test the parametric specification of the integrated baseline hazard in a semiparametric mixed proportional hazard model. We investigate the finite sample performance of our test in a Monte Carlo study. Finally, we apply the proposed test to Kennan’s strike durations data.

Adults transplanted as children as retransplant candidates: Analysis of outcomes support optimism in a population mislabeled as high risk.

Adult liver transplant programs have heretofore been hesitant to perform liver retransplantation in adult patients who underwent primary liver transplantation as a child (P_A). Areas of concern include: (a) potential disruption in care when transferring from a pediatric to an adult transplant center; (b) generally inferior outcomes of retransplantation; (c) reputation of young adults for non-adherence to post-transplant regimen; and (d) potential higher work effort for equivalent outcomes. To examine these concerns, we reviewed data on all US liver adult retransplants from 10/01/1987 to 9/30/2017. We propensity matched the P_A patients to patients who received both primary and retransplantation as adults (A_A), with ≥550days between transplants. A mixed Cox proportional hazards model with program size and time period of transplantation as random variables revealed that retransplantation of P_A patients produced no significantly different graft survival or patient survival rates than retransplantation of the matched A_A patients. Therefore, inferior rates of liver retransplantation in these patients and concerns about continuity of care in changing transplant programs are not as believed in the wider liver transplant community. In conclusion, liver transplant centers should be optimistic about retransplanting adults who received their primary transplants as children.

A matched-pair analysis comparing stereotactic radiosurgery with whole-brain radiotherapy for patients with multiple brain metastases

Stereotactic radiosurgery (SRS) is an emerging treatment for patients with multiple brain metastases (BM). The present work compares the SRS of multiple brain metastases with whole-brain radiotherapy (WBRT). We performed a matched-pair analysis for 128 patients with multiple BM treated with either SRS or WBRT over a 5-year period. Patients were matched pairwise for seven potential prognostic factors. A mixed Cox Proportional Hazards model with univariate and multivariate analysis was fitted for overall survival (OS).Distant intracranial progression-free survival (icPFS) and local control were assessed using a Fine and Gray subdistribution hazard model and considering death as competing event. Patients undergoing SRS had a median of 4 BM (range 3-16). 1-year local control of individual BM following SRS was 91.7%. Median OS in the SRS subgroup was 15.7months (IQR 9.7-36.4) versus 8.0months (interquartile range, IQR 3.8-18.0) in the WBRT subgroup (HR 2.25, 95% CI [1.5; 3.5], p<0.001). Median icPFS was 8.6 (IQR 3.4-18.0) versus 22.4 (IQR 5.6-28.6) months, respectively (HR for WBRT 0.41, 95% CI [0.24; 0.71], p=0.001). Following SRS, synchronous BM diagnosis (HR 2.51, 95% CI [1.30; 4.70], p=0.004), higher initial number of BM (HR 1.21, 95% CI [1.10; 1.40], p=0.002) and lung cancer histology (HR 2.05, 95% CI [1.10; 3.80], p=0.024) negatively impacted survival. Excellent clinical performance (KPI 90%) was a positive prognosticator (HR 0.38, 95% CI [0.20; 0.72], p=0.003), as was extracerebral tumor control (HR 0.48, 95% CI [0.24; 0.97], p=0.040). Higher initial (HR 1.19, 95% CI [1.00; 1.40], p<0.013) and total number of BM (HR 1.23, 95% CI [1.10; 1.40], p<0.001) were prognostic for shorter icPFS. This is the first matched-pair analysis to compare SRS alone versus WBRT alone for multiple BM. OS was prolonged in the SRS subgroup and generally favorable in the entire cohort. Our results suggest SRS as a feasible and effective treatment for patients with multiple BM.

Nationwide Study of County Health Rankings and Lung Transplant Outcomes

There is limited information about whether social determinants of health affect lung transplant outcomes. In this study we aimed to examine associations between nationwide County Health Rankings (CHR) and its domain specific sub-ranks and lung transplant graft and patient survival. United Network for Organ Sharing and CHR datasets were linked by using transplant recipient's zip code of residence. Patient- and county-level characteristics were assessed for all adult lung transplant recipients from 2010-2018. CHR were calculated across all 3149 US counties and averaged across the eight study years and counties were assigned to a tertile for each of two overall ranks and six domain-specific sub-ranks. Mixed effects Cox proportional hazards models were used to examine associations between County Health Rankings and graft survival and patient survival post-lung transplant. Out of 15909 Lung Transplant recipients, 7084 resided in best, 6072 in middle and 2753 in worst terile. The Health Outcomes rank, Length of Life sub-rank, Quality of Life sub-rank, Clinical Care sub-rank, Social and Economic Factors sub-rank, and Physical Environment sub-rank were independently associated with both graft and patient survival (between 7 and 15% increased risk among counties in the worst tertiles compared to the best). However, these associations did not remain statistically significant when models were adjusted for all relevant patient characteristics, center lung transplant volume, and random effects for transplant center and county. While county-level differences do affect lung transplant outcomes, these differences are largely accounted for by individual patient and transplant-specific characteristics. Further research is needed to ascertain whether these factors act as mediator factors.

New Evidence Supporting Increased Use of Split Liver Transplantation.

Young children and small-statured candidates are dying on liver candidate waitlists. The Organ Procurement and Transplantation Network and United Network for Organ Sharing have proposed a split liver (SL) variance encouraging transplant programs to split more livers to aid these smaller statured candidates. We evaluated the US experience of splitting donor livers during 2002-2016. The results of our analysis provide new evidence to support this variance. During 2002-2016, SL grafts from 935 donors were transplanted into 1870 recipients. Controlling for recipient factors and using time period and program identification as random variables, a multivariable mixed Cox proportional hazards model for graft failure revealed that donor aged 3-10 years had a relative risk (RR) of 3.94 (2.86-5.44), and donor aged >30 years had a RR of 1.94 (1.59-2.35) for graft failure. Donor-to-recipient body surface area ratio <0.90 had a RR of 1.40 (1.13-1.75). Programs with experience transplanting <23 SLs had a RR of 1.43 (1.21-1.75). The same program transplanting both split segments had a RR of 1.38 (1.20-1.59). Splitting protocols based on these findings could improve graft survival after SL transplantation, which would encourage programs to opt into splitting more livers. The new protocols may consider donor age restrictions, necessary program experience, donor with body surface area appropriate for recipient, and improved logistical factors to share segments between transplant programs. The result would likely be a drastic reduction in liver waitlist deaths for young children and small-statured candidates.

A competing risks model with time‐varying heterogeneity and simultaneous failure

This paper proposes a new bivariate competing risks model in which both durations are the first passage times of dependent Lévy subordinators with exponential thresholds and multiplicative covariates effects. Our specification extends the mixed proportional hazards model, as it allows for the time‐varying heterogeneity represented by the unobservable Lévy processes and it generates the simultaneous termination of both durations with positive probability. We obtain nonparametric identification of all model primitives given competing risks data. A flexible semiparametric estimation procedure is provided and illustrated through the analysis of a real dataset.

Endogenous censoring in the mixed proportional hazard model with an application to optimal unemployment insurance

SummaryWe examine the sensitivity of estimates of the MPH model with respect to assumptions on the censoring mechanism in the context of an economic model of optimal unemployment insurance. We assume a parametric model for the duration of interest and leave the distribution of censoring unrestricted, allowing it to be correlated with observed and unobserved characteristics. We provide a practical characterization of the identified set with moment inequalities and suggest methods for estimating this set. We apply this approach to estimate the elasticity of unemployment exit rate with respect to unemployment benefit. Finally, we investigate welfare consequences of our estimates.