Clinical Outcomes of Patients with Combined Idiopathic Pulmonary Fibrosis and Emphysema in the IPF-PRO Registry

Purpose To assess the impact of concomitant emphysema on outcomes in patients with idiopathic pulmonary fibrosis (IPF). Methods The IPF-PRO Registry is a US registry of patients with IPF. The presence of combined pulmonary fibrosis and emphysema (CPFE) at enrollment was determined by investigators’ review of an HRCT scan. Associations between emphysema and clinical outcomes were analyzed using Cox proportional hazards models. Results Of 934 patients, 119 (12.7%) had CPFE. Compared with patients with IPF alone, patients with CPFE were older (median 72 vs 70 years); higher proportions were current/former smokers (88.2% vs 63.7%), used oxygen with activity (49.6% vs 31.9%) or at rest (30.8% vs 18.4%), had congestive heart failure (13.6% vs 4.8%) and had prior respiratory hospitalization (25.0% vs 16.7%); they had higher FVC (median 71.8 vs 69.4% predicted) and lower DLco (median 35.3 vs 43.6% predicted). In patients with CPFE and IPF alone, respectively, at 1 year, rates of death or lung transplant were 17.5% (95% CI: 11.7, 25.8) and 11.2% (9.2, 13.6) and rates of hospitalization were 21.6% (14.6, 29.6) and 20.6% (17.9, 23.5). There were no significant associations between emphysema and any outcome after adjustment for baseline variables. No baseline variable predicted outcomes better in IPF alone than in CPFE. Conclusion Approximately 13% of patients in the IPF-PRO Registry had CPFE. Physiologic characteristics and comorbidities of patients with CPFE differed from those of patients with IPF alone, but the presence of emphysema did not drive outcomes after adjustment for baseline covariates. Trial registration ClinicalTrials.gov, NCT01915511; registered August 5, 2013.

Multilevel factors predict medication adherence in rheumatoid arthritis: a 6-month cohort study

Non-adherence challenges efficacy and costs of healthcare. Knowledge of the underlying factors is essential to design effective intervention strategies.ObjectivesTo estimate the prevalence of treatment adherence in rheumatoid arthritis (RA) and to evaluate its predictors.MethodsA 6-month prospective cohort study of patients with RA selected by systematic stratified sampling (33% on first disease-modifying rheumatic drug (DMARD), 33% on second-line DMARD and 33% on biologics). The outcome measure was treatment adherence, defined by a score greater than 80% both in the Compliance Questionnaire in Rheumatology and the Reported Adherence to Medication scale, and was estimated with 95% CIs. Predictive factors included sociodemographic, psychological, clinical, drug-related, patient–doctor relationship related and logistic. Their effect on 6-month adherence was examined by multilevel logistic models adjusted for baseline covariates.Results180 patients were recruited (77% women, mean age 60.8). The prevalence of adherence was 59.1% (95% CI 48.1% to 71.8%). Patients on biologics showed higher adherence and perceived a higher medication need than the others; patients on second-line DMARDs had experienced more adverse events than the others. The variables explaining adherence in the final multivariate model were the type of treatment prescribed (second-line DMARDs OR=5.22, and biologics OR=3.76), agreement on treatment (OR=4.57), having received information on treatment adaptation (OR=1.42) and the physician perception of patient trust (OR=1.58). These effects were independent of disease activity.ConclusionTreatment adherence in RA is far from complete. Psychological, communicational and logistic factors influence treatment adherence in RA to a greater extent than sociodemographic or clinical factors.

What difference does multiple imputation make in longitudinal modeling of EQ-5D-5L data? Empirical analyses of simulated and observed missing data patterns

Purpose Although multiple imputation is the state-of-the-art method for managing missing data, mixed models without multiple imputation may be equally valid for longitudinal data. Additionally, it is not clear whether missing values in multi-item instruments should be imputed at item or score-level. We therefore explored the differences in analyzing the scores of a health-related quality of life questionnaire (EQ-5D-5L) using four approaches in two empirical datasets. Methods We used simulated (GR dataset) and observed missingness patterns (ABCD dataset) in EQ-5D-5L scores to investigate the following approaches: approach-1) mixed models using respondents with complete cases, approach-2) mixed models using all available data, approach-3) mixed models after multiple imputation of the EQ-5D-5L scores, and approach-4) mixed models after multiple imputation of EQ-5D 5L items. Results Approach-1 yielded the highest estimates of all approaches (ABCD, GR), increasingly overestimating the EQ-5D-5L score with higher percentages of missing data (GR). Approach-4 produced the lowest scores at follow-up evaluations (ABCD, GR). Standard errors (0.006–0.008) and mean squared errors (0.032–0.035) increased with increasing percentages of simulated missing GR data. Approaches 2 and 3 showed similar results (both datasets). Conclusion Complete cases analyses overestimated the scores and mixed models after multiple imputation by items yielded the lowest scores. As there was no loss of accuracy, mixed models without multiple imputation, when baseline covariates are complete, might be the most parsimonious choice to deal with missing data. However, multiple imputation may be needed when baseline covariates are missing and/or more than two timepoints are considered.

Intermediate Renal Outcomes, Kidney Failure, and Mortality in Obese Kidney Donors

BackgroundObesity is associated with the two archetypal kidney disease risk factors: hypertension and diabetes. Concerns that the effects of diabetes and hypertension in obese kidney donors might be magnified in their remaining kidney have led to the exclusion of many obese candidates from kidney donation.MethodsWe compared mortality, diabetes, hypertension, proteinuria, reduced eGFR and its trajectory, and the development of kidney failure in 8583 kidney donors, according to body mass index (BMI). The study included 6822 individuals with a BMI of <30 kg/m2, 1338 with a BMI of 30–34.9 kg/m2, and 423 with a BMI of ≥35 kg/m2. We used Cox regression models, adjusting for baseline covariates only, and models adjusting for postdonation diabetes, hypertension, and kidney failure as time-varying covariates.ResultsObese donors were more likely than nonobese donors to develop diabetes, hypertension, and proteinuria. The increase in eGFR in obese versus nonobese donors was significantly higher in the first 10 years (3.5 ml/min per 1.73m2 per year versus 2.4 ml/min per 1.73m2 per year; P<0.001), but comparable thereafter. At a mean±SD follow-up of 19.3±10.3 years after donation, 31 (0.5%) nonobese and 12 (0.7%) obese donors developed ESKD. Of the 12 patients with ESKD in obese donors, 10 occurred in 1445 White donors who were related to the recipient (0.9%). Risk of death in obese donors was not significantly increased compared with nonobese donors.ConclusionsObesity in kidney donors, as in nondonors, is associated with increased risk of developing diabetes and hypertension. The absolute risk of ESKD is small and the risk of death is comparable to that of nonobese donors.

Complement activation fragments in cervicovaginal fluid are associated with intra-amniotic infection/inflammation and spontaneous preterm birth in women with preterm premature rupture of membranes

Objective: We sought to determine whether the levels of complement and other inflammatory and angiogenic mediators in cervicovaginal fluid (CVF) are independently associated with intra-amniotic infection and/or inflammation (IAI) and imminent spontaneous preterm birth (SPTB, ≤48 hours of sampling) in women with preterm premature rupture of membranes (PPROM). Study design: This was a retrospective study consisting of 85 singleton pregnant women with PPROM at 20+0 to 33+6 weeks. Amniotic fluid (AF) obtained via amniocentesis was cultured and assayed for interleukin-6. CVF samples collected at the time of amniocentesis were assayed for complement C3a, C4a, and C5a, HSP70, M-CSF, M-CSF-R, S100 A8, S100 A9, thrombospondin-2, VEGF, and VEGFR-1 by ELISA. Results: Multivariate logistic regression analyses revealed that elevated CVF concentrations of complement C3a, 4a and 5a were significantly associated with an increased risk of IAI and imminent SPTB, whereas those of M-CSF were associated with IAI, but not imminent SPTB (P=0.063), after adjustment for baseline covariates (e.g., gestational age at sampling). However, univariate and multivariate analyses showed that the CVF concentrations of angiogenic (thrombospondin-2, VEGF, and VEGFR-1) and inflammatory (HSP70, M-CSF-R, S100 A8, and S100 A9) proteins were not associated with either IAI or imminent SPTB. Conclusions: In women with PPROM, elevated CVF concentrations of complement C3a, C4a, and C5a are independently related to an increased risk of IAI and imminent SPTB. These findings suggest that complement activation in CVF is significantly involved in mechanisms underlying preterm birth and in the host response to IAI in the context of PPROM.

742Analyzing the statistical pitfalls of treating mediators as confounders

Background Numerous studies indicated that infants born small-for-gestational-age (SGA) are at higher risk of overweight. However, the association between SGA and overweight may be due to overcontrolling for body size. This study aimed to analyze the effect of controlling for child’s weight and height in the association between SGA and overweight in children born preterm. Methods Data were obtained from the Preterm Infant Multicenter Growth Study (n = 1089). The association between SGA and overweight at 36 months corrected age (CA) was analyzed using logistic regression models: 1) crude, 2) adjusted for baseline covariates, 3) adjusted for baselines covariates with additional adjustments separately for child’s weight and height at 21 months CA. Marginal structural models (MSM) with stabilized inverse probability weights were used to estimate the direct effect of SGA on overweight. Results The crude and adjusted models yielded a null association (OR, 95% CI: 0.88, 0.26-2.96; 0.95, 0.28-3.29). Adjusting for later height reversed the effect (OR, 95% CI: 2.31, 0.52-10.26), and adjusting for later weight reversed the effect and provided a significant association (OR, 95% CI: 6.60, 1.10-37.14). The MSMs with height and weight considered as mediators indicated no direct effect of SGA on overweight (OR, 95% CI: 0.83, 0.14-5.01; 0.71, 0.18-2.81). Conclusions Overcontrolling for body size can falsely induce an association between SGA and overweight. Key messages Mediators should not be treated as confounders.

Adjustment for Baseline Covariates to Increase Efficiency in RCTs with Binary Endpoint: A Comparison of Bayesian and Frequentist Approaches

Background: In a randomized controlled trial (RCT) with binary outcome the estimate of the marginal treatment effect can be biased by prognostic baseline covariates adjustment. Methods that target the marginal odds ratio, allowing for improved precision and power, have been developed. Methods: The performance of different estimators for the treatment effect in the frequentist (targeted maximum likelihood estimator, inverse-probability-of-treatment weighting, parametric G-computation, and the semiparametric locally efficient estimator) and Bayesian (model averaging), adjustment for confounding, and generalized Bayesian causal effect estimation frameworks are assessed and compared in a simulation study under different scenarios. The use of these estimators is illustrated on an RCT in type II diabetes. Results: Model mis-specification does not increase the bias. The approaches that are not doubly robust have increased standard error (SE) under the scenario of mis-specification of the treatment model. The Bayesian estimators showed a higher type II error than frequentist estimators if noisy covariates are included in the treatment model. Conclusions: Adjusting for prognostic baseline covariates in the analysis of RCTs can have more power than intention-to-treat based tests. However, for some classes of model, when the regression model is mis-specified, inflated type I error and potential bias on treatment effect estimate may arise.

Propensity Score Analysis with Partially Observed Baseline Covariates: A Practical Comparison of Methods for Handling Missing Data

(1) Background: Propensity score methods gained popularity in non-interventional clinical studies. As it may often occur in observational datasets, some values in baseline covariates are missing for some patients. The present study aims to compare the performances of popular statistical methods to deal with missing data in propensity score analysis. (2) Methods: Methods that account for missing data during the estimation process and methods based on the imputation of missing values, such as multiple imputations, were considered. The methods were applied on the dataset of an ongoing prospective registry for the treatment of unprotected left main coronary artery disease. The performances were assessed in terms of the overall balance of baseline covariates. (3) Results: Methods that explicitly deal with missing data were superior to classical complete case analysis. The best balance was observed when propensity scores were estimated with a method that accounts for missing data using a stochastic approximation of the expectation-maximization algorithm. (4) Conclusions: If missing at random mechanism is plausible, methods that use missing data to estimate propensity score or impute them should be preferred. Sensitivity analyses are encouraged to evaluate the implications methods used to handle missing data and estimate propensity score.

Effect of paracentesis on the survival of patients with terminal cancer and ascites: A propensity score-weighted analysis of the EASED study.

e24063 Background: Malignant ascites (MA) is associated with progressive deterioration in quality of life and poor prognosis of patients with cancer. Paracentesis is among the most widely utilised treatments for MA. However, reports on the usefulness of paracentesis in patients with MA are limited. Thus, this study aimed to investigate whether paracentesis affects the duration of survival in such patients. Methods: We performed a post - hoc analysis of a prospective multicentre observational study that investigated the dying process and end-of-life care in patients with terminal cancer, who were admitted to 23 palliative care units in Japan. Survival duration was compared between patients who did (paracentesis group) and did not undergo paracentesis (non-paracentesis group). We used the inverse probability of treatment weighting (IPTW) method to control for baseline covariates between the two groups. Furthermore, subgroup analyses were performed to investigate the IPTW-adjusted hazard ratio (HR) of the paracentesis and non-paracentesis groups according to some of the baseline covariates, including age, sex, Karnofsky Performance Status, primary tumor site, liver dysfunction, and renal dysfunction. Results: Among the 1,896 patients, who were initially enrolled, 568 with ascites were included in the study cohort. Eighty-five (15.0%) patients underwent paracentesis. The most common primary tumour site was the gastrointestinal tract (51.9%, n = 295), followed by the pancreas (22.7%, n = 129). The non-adjusted median durations of survival were 22 days (95% confidence interval [CI]: 16–25) and 12 days (95% CI: 11–13) in the paracentesis and non-paracentesis groups, respectively (HR: 0.69, [95% CI: 0.54–0.88]; p = 0.003). The IPTW-adjusted median durations of survival were 22 days (95% CI: 16–25) and 16 days (95% CI: 12–22) in the paracentesis and non-paracentesis groups, respectively (HR: 0.89, 95% CI: 0.64–1.24; p = 0.492). There were no serious adverse events in the paracentesis group. According to the baseline covariates, a weighted subgroup analysis was performed to compare the survival of patients in the non-paracentesis and paracentesis groups. There was no significant heterogeneity across all subgroups. Paracentesis was not associated with a significant survival risk in any of the subgroups. Conclusions: Paracentesis does not negatively affect the survival of patients with cancer and MA; hence, it can be used as a standard treatment in palliative care settings.