A Bayesian hierarchical approach to account for evidence and uncertainty in the modeling of infectious diseases: An application to COVID-19.

Infectious disease models can serve as critical tools to predict the development of cases and associated healthcare demand and to determine the set of nonpharmaceutical interventions (NPIs) that is most effective in slowing the spread of an infectious agent. Current approaches to estimate NPI effects typically focus on relatively short time periods and either on the number of reported cases, deaths, intensive care occupancy, or hospital occupancy as a single indicator of disease transmission. In this work, we propose a Bayesian hierarchical model that integrates multiple outcomes and complementary sources of information in the estimation of the true and unknown number of infections while accounting for time-varying underreporting and weekday-specific delays in reported cases and deaths, allowing us to estimate the number of infections on a daily basis rather than having to smooth the data. To address dynamic changes occurring over long periods of time, we account for the spread of new variants, seasonality, and time-varying differences in host susceptibility. We implement a Markov chain Monte Carlo algorithm to conduct Bayesian inference and illustrate the proposed approach with data on COVID-19 from 20 European countries. The approach shows good performance on simulated data and produces posterior predictions that show a good fit to reported cases, deaths, hospital, and intensive care occupancy.

Long-term benefits and harms of early colorectal cancer screening in German individuals with familial cancer risk.

Individuals with a family history of colorectal cancer (CRC) may benefit from early screening with colonoscopy or immunologic fecal occult blood testing (iFOBT). We systematically evaluated the benefit-harm trade-offs of various screening strategies differing by screening test (colonoscopy or iFOBT), interval (iFOBT: annual/biennial; colonoscopy: 10-yearly) and age at start (30, 35, 40, 45, 50 and 55 years) and end of screening (65, 70 and 75 years) offered to individuals identified with familial CRC risk in Germany. A Markov-state-transition model was developed and used to estimate health benefits (CRC-related deaths avoided, life-years gained [LYG]), potential harms (eg, associated with additional colonoscopies) and incremental harm-benefit ratios (IHBR) for each strategy. Both benefits and harms increased with earlier start and shorter intervals of screening. When screening started before age 50, 32-36 CRC-related deaths per 1000 persons were avoided with colonoscopy and 29-34 with iFOBT screening, compared to 29-31 (colonoscopy) and 28-30 (iFOBT) CRC-related deaths per 1000 persons when starting age 50 or older, respectively. For iFOBT screening, the IHBRs expressed as additional colonoscopies per LYG were one (biennial, age 45-65 vs no screening), four (biennial, age 35-65), six (biennial, age 30-70) and 34 (annual, age 30-54; biennial, age 55-75). Corresponding IHBRs for 10-yearly colonoscopy were four (age 55-65), 10 (age 45-65), 15 (age 35-65) and 29 (age 30-70). Offering screening with colonoscopy or iFOBT to individuals with familial CRC risk before age 50 is expected to be beneficial. Depending on the accepted IHBR threshold, 10-yearly colonoscopy or alternatively biennial iFOBT from age 30 to 70 should be recommended for this target group.

Using a Bayesian hierarchical approach to study the association between non-pharmaceutical interventions and the spread of Covid-19 in Germany.

Non-Pharmaceutical Interventions (NPIs) are community mitigation strategies, aimed at reducing the spread of illnesses like the coronavirus pandemic, without relying on pharmaceutical drug treatments. This study aims to evaluate the effectiveness of different NPIs across sixteen states of Germany, for a time period of 21 months of the pandemic. We used a Bayesian hierarchical approach that combines different sub-models and merges information from complementary sources, to estimate the true and unknown number of infections. In this framework, we used data on reported cases, hospitalizations, intensive care unit occupancy, and deaths to estimate the effect of NPIs. The list of NPIs includes: "contact restriction (up to 5 people)", "strict contact restriction", "curfew", "events permitted up to 100 people", "mask requirement in shopping malls", "restaurant closure", "restaurants permitted only with test", "school closure" and "general behavioral changes". We found a considerable reduction in the instantaneous reproduction number by "general behavioral changes", "strict contact restriction", "restaurants permitted only with test", "contact restriction (up to 5 people)", "restaurant closure" and "curfew". No association with school closures could be found. This study suggests that some public health measures, including general behavioral changes, strict contact restrictions, and restaurants permitted only with tests are associated with containing the Covid-19 pandemic. Future research is needed to better understand the effectiveness of NPIs in the context of Covid-19 vaccination.

Colorectal Cancer Screening for Persons With a Positive Family History­Evaluation of the FARKOR Program for the Secondary Prevention of Colorectal Cancer in Persons Aged 25 to 50.

BACKGROUND: Persons with a positive family history of colorectal cancer (CRC) are more likely than others to develop CRC and are also younger at the onset of the disease. Nonetheless, the German Federal Joint Committee (G-BA, Gemeinsamer Bundes - ausschuss) recommends screening all persons aged 50 and above regardless of their family history. FARKOR was a project supported by the Innovation Fund of the G-BA to study the feasibility, efficacy, and safety of a risk-adapted early detection program for CRC among persons aged 25 to 50 without any specific past medical history. METHODS: Physicians in private practice in Bavaria documented their activities relating to FARKOR online. The FARKOR process comprised a declaration of consent, a simplified family history for CRC, an optional, more comprehensive family history, a counseling session for participatory decision-making on further measures, and various modalities of screening (an immunological fecal occult blood test [iFOBT], colonoscopy, or no screening). Related physician activities outside the FARKOR process were assessed by record linkage between study data and data of the patients' health insurance carriers. RESULTS: The simplified family history was documented in 25 847 persons and positive for CRC in 5769 (22.3%). 3232 persons had a more comprehensive family history, among whom 2054 (63.6%) participated in screening measures. 1595 underwent colonoscopy; 278 persons who had already undergone colonoscopy in the preceding five years were excluded from the analysis. Colonoscopy revealed adenoma in 232 persons (17,6 %), advanced adenoma in 78 (5.9%) and carcinoma in 4 (0.3%). There were no serious complications. CONCLUSION: The detection rates in this study corresponded to those of persons aged 55 to 59 in the current early detection program. Despite numerous problems in the performance of the study (inconsistencies in documentation, external performance of screening measures on program participants), the results support the feasibility of a risk-adapted early detection program in the young target population with a family history of CRC.

Diagnostic Utility of Temporal Muscle Thickness as a Monitoring Tool for Muscle Wasting in Neurocritical Care.

Temporalis muscle (TM) atrophy has emerged as a potential biomarker for muscle wasting. However, its diagnostic utility as a monitoring tool in intensive care remains uncertain. Hence, the objective of this study was to evaluate the diagnostic value of sequential ultrasound- and computed tomography (CT)-based measurements of TM thickness (TMT). With a prospective observational design, we included 40 patients without preexisting sarcopenia admitted to a neurointensive care unit. TMT measurements, performed upon admission and serially every 3−4 days, were correlated with rectus femoris muscle thickness (RFT) ultrasound measurements. Interrater reliability was assessed by Bland Altmann plots and intraclass correlation coefficient (ICC). Analysis of variance was performed in subgroups to evaluate differences in the standard error of measurement (SEM). RFT decline was paralleled by ultrasound- as well as CT-based TMT measurements (TMT to RFT: r = 0.746, p < 0.001; CT-based TMT to ultrasound-based RFT: r = 0.609, p < 0.001). ICC was 0.80 [95% CI 0.74, 0.84] for ultrasound-based assessment and 0.90 [95% CI 0.88, 0.92] for CT-based TMT measurements. Analysis of variance for BMI, Heckmatt score, fluid balance, and agitation showed no evidence of measurement errors in these subgroups. This study demonstrates the clinical feasibility and utility of ultrasound- and CT-based TMT measurements for the assessment of muscle wasting.

Survival after left ventricular assist device implantation correlates with a novel device-based measure of heart rate variability: the heart rate score.

OBJECTIVES: The heart rate score (HRS) serves as a device-based measure of impaired heart rate variability and is an independent predictor of death in patients with heart failure and a cardiac implantable electrical device. However, no data are available for predicting death from the HRS in patients with end stage heart failure and a left ventricular assist device. METHODS: From November 2011 to July 2018, a total of 56 patients with a pre-existing cardiac implantable electrical device underwent left ventricular assist device implantation at our 2 study sites. The ventricular HRS was calculated retrospectively during the first cardiac implantable electrical device follow-up examination following the index hospitalization. Survival during follow-up was correlated with initial HRS. RESULTS: During the follow-up period, 46.4% of the patients (n = 26) died. The median follow-up period was 33.2 months. The median HRS after the index hospitalization was 41.1 ± 21.8%. More patients with an HRS >65% died compared to patients with an HRS <30% (76.9% vs 14.4%; P = 0.007). CONCLUSIONS: In our multicentre experience, survival of patients after an left ventricular assist device implant correlates with the HRS. After confirmation of our findings in a larger cohort, the effect of rate-responsive pacing will be within the scope of further investigation.