Category: Statistics

Biometrics. 2021 Dec 24. doi: 10.1111/biom.13615. Online ahead of print.

ABSTRACT

While most statistical methods for the analysis of longitudinal data have focused on retrospective models of association, new advances in mobile health data have presented opportunities for predicting future health status by leveraging an individual’s behavioral history alongside data from similar patients. Methods that incorporate both individual-level and sample-level effects are critical to using this data to its full predictive capacity. Neural networks are powerful tools for prediction, but many assume input observations are independent even when they are clustered or correlated in some way, such as in longitudinal data. Generalized linear mixed models (GLMM) provide a flexible framework for modeling longitudinal data but have poor predictive power particularly when the data is highly nonlinear. We propose a generalized neural network mixed model (GNMM) that replaces the linear fixed effect in a GLMM with the output of a feed-forward neural network. The model simultaneously accounts for the correlation structure and complex nonlinear relationship between input variables and outcomes, and it utilizes the predictive power of neural networks. We apply this approach to predict depression and anxiety levels of schizophrenic patients using longitudinal data collected from passive smartphone sensor data.

PMID:34951484 | DOI:10.1111/biom.13615

Artif Organs. 2021 Dec 24. doi: 10.1111/aor.14141. Online ahead of print.

ABSTRACT

BACKGROUND: M101 is an extracellular hemoglobin isolated from a marine lugworm and is present in the medical device HEMO₂ life®. The clinical investigation OXYOP was a paired kidney analysis (n = 60) designed to evaluate the safety and performance of HEMO₂ life® used as an additive to preservation solution in renal transplantation. The secondary efficacy endpoints showed less delayed graft function (DGF) and better renal function in the HEMO₂ life® group but due to the study design cold ischemia time (CIT) was longer in the contralateral kidneys.

METHODS: An additional analysis was conducted including OXYOP patients and patients from the ASTRE database (n = 6584) to verify that the decrease in DGF rates observed in the HEMO₂ life® group may not be due solely to the shorter CIT but also to HEMO₂ life® performance. Kaplan-Meier estimate curves of cumulative probability of achieving a creatinine level below 250 µmol/L were generated and compared in both groups. A Cox model was used to test the effect of the explanatory variables (use of HEMO₂ life® and CIT). Finally, a bootstrap strategy was used to randomly select smaller samples of patients and test them for statistical comparison in the ASTRE database.

RESULTS: Kaplan-Meier estimate curves confirmed the existence of a relation between DGF and CIT and Cox analysis showed a benefit in the HEMO₂ life® group regardless of the associated CIT. Boostrap analysis confirmed these results.

CONCLUSIONS: The present study suggested that the better recovery of renal function observed among kidneys preserved with HEMO₂ life® in the OXYOP study is a therapeutic benefit of this breakthrough innovative medical device.

PMID:34951495 | DOI:10.1111/aor.14141

J Biomech Eng. 2021 Dec 24. doi: 10.1115/1.4053362. Online ahead of print.

ABSTRACT

BACKGROUND: Multiple biomechanical shoulder simulators have been described in the literature, with a trend toward increasing complexity to better simulate clinical scenarios. Our objective was to develop an advanced, novel shoulder joint simulator and compare outcomes at two separate institutions, for a typical shoulder joint motion simulation.

METHODS: Identical shoulder simulators were developed & deployed at both institutions. Eight cadaveric upper extremities were tested by simulating actively controlled, arm elevation in the plane of the scapula for two sequential test conditions (intact and non-destructive simulated cuff-tear), each repeated for a total of five trials. Muscle forces and joint translations were recorded for both conditions. The intact condition was repeated following simulated cuff-tear to assess effect of testing order. Statistical analyses were aimed at assessing repeatability and reproducibility of results within specimens, between specimens, and between institutions.

RESULTS: The highest average forces were observed for the middle deltoid (233N or 32.5% body weight), followed by infraspinatus (99.0N), and posterior deltoid (93.7N) muscles. Differentiation between test conditions was unhindered by variability between repeated trials. Data from testing repeated over time, and between the two institutions were not significantly different.

CONCLUSIONS: The novel shoulder simulator produced repeatable results with low trial-to-trial variation and outcomes were comparable between the two institutions. The results demonstrated a consistent response in muscle forces and humeral translation for the simulated rotator cuff tear condition. Such advanced shoulder simulators could thus be used for evaluating and optimizing surgical interventions and implant strategies.

PMID:34951458 | DOI:10.1115/1.4053362

Occup Med (Lond). 2021 Dec 24:kqab153. doi: 10.1093/occmed/kqab153. Online ahead of print.

ABSTRACT

BACKGROUND: Research has shown that of the myriad services available to veterans in the UK, very few have been independently evaluated. This report presents the results of a randomized controlled trial assessing the impact of Time Line Therapy™ delivered by the Warrior Programme (a third-sector organization).

AIMS: This study was aimed to determine if the intervention is effective in reducing emotional and functional difficulties in ex-service personnel.

METHODS: A mixed-design analysis of variance model was used to investigate whether the Warrior Programme had a statistically significant impact on self-reported scores. The intervention and control group provided data on measures prior to and immediately after the intervention, and at 3-month follow-up.

RESULTS: Those in the intervention group (n = 23) reported statistically significant improvements in self-reported scores immediately following intervention: the Clinical Outcomes in Routine Evaluation (CORE) scores (CORE global distress mean difference [MD] = 45.0, 95% confidence interval [CI] 31-60) (CORE subjective well-being MD = 5.9, 95% CI 3.5-8.3) (CORE functioning MD = 16.7, 95% CI 11.4-21.9) (CORE problems/symptoms MD = 19.4, 95% CI 13.1-25.7), general self-efficacy (MD = -9.8, 95% CI -13.6 to -6.8), anxiety (MD = 8.6, 95% CI 5.2-12.1), depression (MD = 10, 95% CI 6.6-13.5), post-traumatic stress disorder (MD = 26.3, 95% CI 17-25) and functional impairment (MD = 11.1, 95% CI 5.3-16.8) over time, compared to the control group (n = 29). However, score improvement was not sustained over time or statistically significant at follow-up.

CONCLUSIONS: The Warrior Programme was effective in reducing emotional and functional difficulties in ex-service personnel immediately after the intervention, but the effect was not sustained at 3-month follow-up.

PMID:34951474 | DOI:10.1093/occmed/kqab153

Curr Alzheimer Res. 2021 Dec 22. doi: 10.2174/1567205019666211222151851. Online ahead of print.

ABSTRACT

OBJECTIVE: The objective of this study is to examine the association of olfactory function and genetic predisposition of Alzheimer’s disease (AD) with cognitive performance in adults.

METHODS: A total of 2049 Chinese adults from Rugao Longevity and Ageing Study (RuLAS, n=1460, mean age 78 years) and Central China Cohort (CCC, n=589, mean age 48 years) were included in this study. A standard interview-based survey, clinical information, and blood samples were collected in both cohorts. Olfactory function in terms of olfactory identification was measured by the brief version of the Chinese Smell Identification Test consisted of 18 full points. Cognitive performance was measured by the Chinese version of the Mini-mental State Examination. A genetic risk score (GRS) was calculated from 5 single nucleotide polymorphisms, which were robustly related to Alzheimer’s disease in Caucasians and cognitive performance in our Chinese population.

RESULTS: In the pooled analyses, participants at the lowest quartile of olfactory function had significantly higher odds of cognitive impairment (adjusted odds ratio [95% CI] =1.45 [1.00 to 2.09], Ptrend =0.005), and such association was stronger among participants with a stronger genetic predisposition of Alzheimer’s disease (β coefficient±SE, -0.06±0.03 in participants with a lower GRS vs. -0.19±0.05 in those with a higher GRS, respectively, Pinteraction=0.01). Similar associations were observed in RuLAS (P-trend=0.06) and in CCC (P-trend<0.001).

CONCLUSION: In this study, a decreased olfactory function was associated with worse cognitive performance in adults, especially among participants with a higher genetic risk of Alzheimer’s disease. Further studies are warranted to evaluate the causal relationship between olfaction and cognitive performance.

PMID:34951364 | DOI:10.2174/1567205019666211222151851

Exp Clin Transplant. 2021 Dec;19(12):1263-1270. doi: 10.6002/ect.2021.0438.

ABSTRACT

OBJECTIVES: Comparisons of COVID-19 incidence between kidney transplant recipients and patients who did not receive kidney transplant are underexplored in various geographic regions.

MATERIALS AND METHODS: This Indian, single-center, retrospective study analyzed COVID-19 data of patients hospitalized between May 12, 2020, and January 11, 2021. A propensity matching score was used to compare outcomes between the 2 groups. We also used multivariable Cox proportional hazard analyses to assess association of kidney transplantation with mortality.

RESULTS: Of the 1627 COVID-19 cases, 179 were kidney transplant recipients and 1448 were not kidney transplant patients (control group). Ofthe 436 reported in-hospital deaths, 20 (11.1%) were in the kidney transplant group and 416 (28.7%) were in the control group. Propensity matching identified 98 kidney transplantrecipients and167 controlpatients. InKaplanMeier survival plots for these patients, there was no statistical difference in mortality (log-rank, Mantel Cox test; P = .07) or severity (log-rank, Mantel Cox test; P = .07) with regard to COVID-19. In Cox analysis, age groups from 61 to 70 years (hazard ratio = 1.5; 95% CI, 1.0-2.2; P = .04), 71 to 80 years (hazard ratio = 1.64; 95% CI, 1.0-2.5; P = .02), and >80 years (hazard ratio = 1.91; 95% CI, 1.1-3.1; P = .01)were associatedwith statistically significant greater mortality.Having a kidney transplant (hazard ratio = 0.43; 95% CI, 0.3-0.7; P = 0.001) was not associated with mortality.

CONCLUSIONS: In our analysis, age was the most important predictor of mortality. Kidney transplant status was not found to have an independent association with mortality and severity.

PMID:34951346 | DOI:10.6002/ect.2021.0438

Exp Clin Transplant. 2021 Dec;19(12):1334-1340. doi: 10.6002/ect.2021.0402.

ABSTRACT

OBJECTIVES: Vaccination against SARS-CoV-2 may reduce COVID-19 mortality and complications in solidorgan transplant recipients, and we evaluated the associated antibody responses and adverse effects in this high-risk population.

MATERIALS AND METHODS: This prospective observational study (April-June 2021) included 10 liver and 38 kidney transplant recipients who received 2 vaccine doses (Sinovac, n = 31; or BioNTech, n = 17) and 56 healthy adults (Sinovac), all of whom provided 3 blood samples (prevaccination, 4 weeks after first dose, and 4-6 weeks after second dose) for quantitative tests (Abbott Quant assay forimmunoglobulin G antibodies against SARS-CoV-2 spike protein). Type I error was α = .05 in all statistical analyses (SPSS, version 25).

RESULTS: We analyzed demographic data, antibody responses, and adverse events after 2 doses of SARSCoV-2 vaccine, comparedimmune responses from solidorgan transplant recipients (median age, 36.5 years) versus healthy patients (median age, 37.5 years), and observed significantly higher seropositivity in healthy versus transplant patients after Sinovac vaccination (100% vs 67.5%; P = .001). However, we observed no significant seropositive differences for Sinovac versus BioNTech second doses in transplantrecipients. Median SARS-CoV-2 immunoglobulin G level after second dose was significantly higher in BioNTech (1388.6 AU/mL) versus Sinovac patients (136.6 AU/mL) (P = .012). The seropositivity difference between the 2 vaccines was significant in participants 24 to 44 years old (P = .040). The rate of at least 1 side effect was 82.4% (n = 14) for BioNTech vaccine and 32.3% (n = 10) for Sinovac vaccine, and the difference was statistically significant.The most common side effect was arm pain (significantly higher in BioNTech group).

CONCLUSIONS: Solid-organ transplant recipients demonstrated inadequate vaccine responses (higher risk of complications and mortality) versus healthy patients. Furthermore, immune responses may differ between vaccines. Therefore, additional vaccine doses and strict control measures remain crucial.

PMID:34951350 | DOI:10.6002/ect.2021.0402

Clin Exp Hypertens. 2021 Dec 24:1-8. doi: 10.1080/10641963.2021.2018599. Online ahead of print.

ABSTRACT

INTRODUCTION: Left ventricular (LV) diastolic dysfunction and structural abnormalities are common cardiac changes in hypertension (HTN), and several factors other than high blood pressure (BP) may play a role in these changes. The aim of this study was to reveal the relationship between triglyceride glucose (TyG) index, a novel parameter for insulin resistance (IR), with LV diastolic function and structure in hypertensive patients.

MATERIAL AND METHOD: A total of 119 newly diagnosed, untrated hypertensive patients free of diabetes and/or cardiovascular complications were included in this study. IR was estimated with the TyG index calculated from ln [fasting TG (mg/dL) × fasting blood glucose (mg/dL)/2]. Two-dimensional and Doppler echocardiographic examinations were performed to assess LV diastolic functions and structure.

RESULTS: Based on median TyG index, 51 patients was assigned as group I (<8.7) and 68 patients as group II (>8.7). In patients with high TyG index, left atrial volume index (LAVi) (p < .001) LV mass index (LVMI) (p = .016), E/e‘ ratio (p < .001) increased, and e’ velocity (p < .001) and E/A ratio (p = .028) decreased. There was a statistically significant correlation between TyG index and these parameters (all p > .05). Stepwise multiple regression analysis demonstrated that the relationship of TyG index with LV diastolic function and structure was independent of potential confounders (all p < .001).

CONCLUSION: This study suggest that a high TyG index is related to LV diastolic functional impairment and structure abnormality in newly diagnosed hypertensive patients in the absence of diabetes or CVD.

PMID:34951339 | DOI:10.1080/10641963.2021.2018599

Stat Methods Med Res. 2021 Dec 24:9622802211064996. doi: 10.1177/09622802211064996. Online ahead of print.

ABSTRACT

Many clinical trials incorporate stopping rules to terminate early if the clinical question under study can be answered with a high degree of confidence. While common in later-stage trials, these rules are rarely implemented in dose escalation studies, due in part to the relatively smaller sample size of these designs. However, even with a small sample size, this paper shows that easily implementable stopping rules can terminate dose-escalation early with minimal loss to the accuracy of maximum tolerated dose estimation. These stopping rules are developed when the goal is to identify one or two dose levels, as the maximum tolerated dose and co-maximum tolerated dose. In oncology, this latter goal is frequently considered when the study includes dose-expansion cohorts, which are used to further estimate and compare the safety and efficacy of one or two dose levels. As study protocols do not typically halt accrual between escalation and expansion, early termination is of clinical importance as it either allows for additional patients to be treated as part of the dose expansion cohort to obtain more precise estimates of the study endpoints or allows for an overall reduction in the total sample size.

PMID:34951338 | DOI:10.1177/09622802211064996

Ann Clin Biochem. 2021 Dec 24:45632211063142. doi: 10.1177/00045632211063142. Online ahead of print.

ABSTRACT

OBJECTIVE: A series of physiological changes in thyroid function occur during pregnancy and differ from those non-pregnant women. This study aimed to establish the pregnancy-specific reference intervals of TSH and FT4 using an indirect method based on the healthy pregnant women from southwest China population.

METHODS: Thyroid function test results which available on the Laboratory Information System (LIS) were collected from the pregnancies who visited the Obstetric Clinic or the Department of Gynecology between 1 January 2015, and 30 December 2020. We grouped the data by trimesters to establish the reference intervals (RIs) based on the clinical consensus of different levels of TSH and FT4 at different weeks of gestation. All arrangements were referenced to the document CLSI EP28-A3C.

RESULTS: A total of 33,040 thyroid function test results of pregnant women, aged 31 (28,33) years were statistical analyzed. Estimated RIs for TSH and FT4 in the first, second and third trimesters corresponding to the 2.5^th and 97.5^th percentiles in TPOAb negative were 0.02-5.23, 0.03-5.24, 0.37-5.68 mIU/L, 11.66-20.69, 10.1-18.59, 9.85-16.86pmol/L, respectively.

CONCLUSION: This study provides trimester-specific RIs for TSH and FT4 among healthy pregnant women in southwest China which guides clinicians to diagnosis and screen for thyroid disorders in this region.

PMID:34951326 | DOI:10.1177/00045632211063142