Identifying and evaluating COVID-19 effects on short-term statistics

The economic downturn due to lockdown measures at the beginning of the COVID-19 crisis raised the question whether any adaptations to the short-term statistics (STS) were needed to ensure accurate and relevant output. We limit ourselves to STS on turnover and related variables like volume of production. We looked into the different stages of the production process - from data collection to output - and anticipated a number of potential lockdown effects. With respect to output relevance, there was an increased interest in faster and specific output. With respect to the output accuracy, we took measures to check whether the anticipated effects really occurred and measures to mitigate the consequences. Examples of such measures are the calculation of an additional editing score function, alternative imputations and extensions of the regular analysis step. In this paper we give an overview of the anticipated effects, the subsequent measures that we took, we evaluate to what extent the anticipated effects occurred in practice and we mention some unforeseen effects. We end this paper by discussing to what extent the developed measures are also useful to keep after the economy has recovered. © 2023 - IOS Press. All rights reserved.

The importance of cultural tailoring of communicators and media outlets in an influenza vaccination awareness campaign: a digital randomized trial.

The COVID-19 pandemic has exposed the vulnerability of ethnic minorities again. Health inequity within ethnic minorities has been explained by factors such as higher prevalence of underlying disease, restricted access to care, and lower vaccination rates. In this study, we investigated the effect of cultural tailoring of communicators and media outlets, respectively, on vaccine willingness in an influenza vaccination campaign in the Netherlands. A total of 1226 participants were recruited from two culturally non-tailored media outlets (Dutch newspaper and Facebook), and one media outlet tailored to a large community in the Netherlands with Indian ancestry. The participants from all three media outlets were randomly exposed to a vaccination awareness video delivered by a physician with an Indian or Dutch background, followed by an online survey. Cultural tailoring compared to cultural non-tailoring of communicators showed no difference in improvement of vaccine willingness (13.9% vs. 20.7% increment, respectively, p = 0.083). However, the media outlet tailored to the community with Indian ancestry, resulted in a higher improvement of vaccine willingness compared to non-tailored media outlets (46.7% vs. 14.7% increment, respectively, p < 0.001, unadjusted OR = 5.096). These results suggest that cultural tailoring of media outlets may be critical to effectively reach out to ethnic minorities to help optimize vaccination rates and improve general health.

Primary results of the cTRAK TN trial: A clinical trial utilising ctDNA mutation tracking to detect minimal residual disease and trigger intervention in patients with moderate and high risk early stage triple negative breast cancer

Background Detection of circulating tumour DNA (ctDNA) in patients (pts) who have completed treatment for early-stage triple negative breast cancer (TNBC) is associated with a very high risk of future relapse. Identifiying those at high risk of subsequent relapse may allow tailoring of further therapy to delay or prevent recurrence. The c-TRAK TN trial assessed the utility of prospective ctDNA surveillance in pts treated for TNBC and the activity of pembrolizumab (P) in pts with ctDNA detected. Methods c-TRAK TN, a multi-centre phase II trial with integrated prospective screening component, enrolled pts with early-stage TNBC and either residual disease following neoadjuvant chemotherapy, or tumour size >20mm and/or axillary lymph node involvement if adjuvant chemotherapy was given. Tumour tissue was sequenced to identify somatic mutations suitable for tracking using personalised digital PCR ctDNA assays (BioRad QX200). Pts had "active" ctDNA surveillance via blood sample testing every 3 months to 12 months (potential up to 18 months if S samples missed due to COVID) during which time if ctDNA was detected (ctDNA+) pts could be randomised 2:1 to P (200mg i.v. q 3 weeks for 1 year) or observation (Obs). Pts and clinicians were blinded to ctDNA+ results unless they were allocated P, when staging scans were done and those free of clinical recurrence started treatment. Following advice from the Independent Data Monitoring Committee, the Obs arm closed on 16/06/2020 with all subsequent ctDNA+ pts allocated P. Following the completion of active ctDNA surveillance, 3-monthly visits continued to 24 months to be analysed retrospectively. The aim was to recruit 150 pts to ctDNA surveillance, assuming 30% would be ctDNA+ within 12 months, allowing ctDNA+ rate to be estimated with a 2-sided 95%CI of +/-7.3%. Co-primary endpoints are i) rates of ctDNA detection by 12 and 24 months from start of ctDNA surveillance;ii) rates of sustained ctDNA clearance on P defined as absence of detectable ctDNA, or disease recurrence 6 months after starting P. Results 208 pts were registered between 30/01/18 and 06/12/19, 185 had tumour sequenced, 171 (92.4%) had trackable mutations, and 161 entered ctDNA surveillance. The rate of ctDNA detection by 12 months after start of surveillance was 27.3% (44/161, 95% CI 20.6-34.9). ctDNA+ rates from baseline, 3, 6, 9 and 12 month ctDNA samples were 23/161 (14.3%), 6/115 (5.2%), 6/99 (5.1%), 7/84 (8.3%), and 2/84 (2.4%) respectively. An additional 2 pts were ctDNA+ on COVID extended active surveillance at 15 (1/51, 2%) or 18 months (1/11, 9%). 7 pts relapsed without prior ctDNA detection. 45 pts entered the therapeutic component of the trial (initially 31 to P and 14 to Obs). 1 Obs pt was re-allocated to P. Of pts allocated to P, 72% (23/32) had metastatic disease at time of ctDNA detection on staging scans (75% (12/16) who were ctDNA+ at baseline and 69% (11/16) at other timepoints). 4 pts declined to start P, largely due to COVID concerns. Of the 5 pts who commenced P, at the time of analysis none achieved sustained ctDNA clearance and 4 had recurred. In pts allocated to Obs, median time to recurrence was 4.1 months (95% CI: 3.2-not-defined). Conclusion The c-TRAK TN trial is to our knowledge the first study to assess the proof-of-principle of whether ctDNA assays have clinical utility in guiding further therapy in TNBC. Relatively few pts commenced P treatment precluding assessment of potential activity. At enrollment, patients had a relatively high of rate of undiagnosed metastatic disease when imaged. Our findings have implications for future trial design, emphasizing the importance of early start of ctDNA testing, and more sensitive and/or more frequent ctDNA testing regimes.

Electrocardiogram-based mortality prediction in patients with COVID-19 using machine learning.

BACKGROUND AND PURPOSE: The electrocardiogram (ECG) is frequently obtained in the work-up of COVID-19 patients. So far, no study has evaluated whether ECG-based machine learning models have added value to predict in-hospital mortality specifically in COVID-19 patients. METHODS: Using data from the CAPACITY-COVID registry, we studied 882 patients admitted with COVID-19 across seven hospitals in the Netherlands. Raw format 12-lead ECGs recorded within 72 h of admission were studied. With data from five hospitals (n = 634), three models were developed: (a) a logistic regression baseline model using age and sex, (b) a least absolute shrinkage and selection operator (LASSO) model using age, sex and human annotated ECG features, and (c) a pre-trained deep neural network (DNN) using age, sex and the raw ECG waveforms. Data from two hospitals (n = 248) was used for external validation. RESULTS: Performances for models a, b and c were comparable with an area under the receiver operating curve of 0.73 (95% confidence interval [CI] 0.65-0.79), 0.76 (95% CI 0.68-0.82) and 0.77 (95% CI 0.70-0.83) respectively. Predictors of mortality in the LASSO model were age, low QRS voltage, ST depression, premature atrial complexes, sex, increased ventricular rate, and right bundle branch block. CONCLUSION: This study shows that the ECG-based prediction models could be helpful for the initial risk stratification of patients diagnosed with COVID-19, and that several ECG abnormalities are associated with in-hospital all-cause mortality of COVID-19 patients. Moreover, this proof-of-principle study shows that the use of pre-trained DNNs for ECG analysis does not underperform compared with time-consuming manual annotation of ECG features.

PO58 CURRENT AND FUTURE MANAGEMENT OF PATIENTS WITH ADVANCED HR+ HER2- ABC WITH PIK3CA MUTATIONS IN THE UK: THE PIK PATHWAY UK PROJECT

Background: This survey aimed to expand understanding of the current and future UK management of patients with HR+ HER2-advanced/metastatic BC with PIK3CA mutations. Methods: Surveys were completed at one-to-one meetings with UK oncologists involved in HR+ HER2- aBC management, either in person or by video conference between February and October 2020. Questions (multi-option and open questions and Likert scales) were designed to understand the unmet need for HR+ HER2- aBC patients, current and future PIK3CA testing and treatment sequencing (pre/post-introduction of PI3Ki) and perceptions around managing adverse events. Responses were entered into a digital survey platform. Results: 264 oncologists were contacted to participate in the survey. Unprecedented pressures on NHS clinicians due to the COVID-19 pandemic limited availability to participate in the survey. As such, 36 oncologists working at 31 geographically-dispersed UK hospitals completed interviews. For HR+ HER2- aBC, 92% oncologists agreed/strongly agreed that the PIK3CA mutation was prognostically significant and 92% agreed/strongly agreed that a PIK3CA mutation was predictive of a response with PI3Ki. 58% oncologists had previously tested for PIK3CA mutations and 33% had clinical experience with PI3Ki. 53% oncologists are currently testing for PIK3CA, 79% of which are testing in the research setting only. Of those currently testing for PIK3CA mutations, 21% test at diagnosis of metastatic disease and the remainder at first- or second-line disease progression. The median (IQR) percentage of patients estimated to have available archival tissue for PIK3CA testing was 80% (33%-95%) including metastatic tissue in 50% (28%-73%). OS followed by PFS and then HRQoL were the priority primary treatment goals. Efficacy was the most important factor driving a decision to use a particular treatment. In the setting of PI3Ki becoming routinely available, the majority of clinicians would use a PI3Ki + fulvestrant as next line therapy following progression on CDK4/6i + AI in patients with a PIK3CA mutation (89% for patients progressing with primary endocrine resistance and 97% for patients progressing with secondary endocrine resistance (ABC guidelines). The PI3Ki-related Grade 1/2 toxicities of most concern to oncologists were diarrhoea and hyperglycaemia (40% and 35% concerned/strongly concerned, respectively). The AE of most concern to patients was thought to be diarrhoea (49% rated diarrhoea as the most concerning AE, 92% placed diarrhoea in the top 3 most concerning AEs). Opinion was that introduction of PI3Ki into UK clinical practice would result in increased utilisation of endocrinology support, increased consultation, clinic and clinician admin time, and more clinical assessments. Conclusions: The survey provides contemporary insights into UK aBC treatment pathways, suggesting how PI3Ki may be integrated following progression on first line CDK4/6i, and the potential impact of the general introduction of PI3Ki.

Using an app to educate patients on the pros and cons of extended anticoagulation after venous thromboembolism: A randomized controlled trial

Background : Adequate patient education is essential to enable patients to engage in shared decision-making (SDM) when deciding to stop or continue anticoagulation after 3 months of anticoagulation for venous thromboembolism (VTE). Aims : To evaluate the effect of an interactive, educational app on patients' level of satisfaction with information, perceived level of knowledge, decisional conflict and SDM when deciding on treatment duration after VTE. Methods : This randomized controlled trial in 1 academic and 3 general hospitals in The Netherlands included adult patients with VTE without malignancy or other indication for anticoagulation. Patients were randomized in a 1:1 ratio to receive the app (intervention group) in addition to the standard of care. The app contains information on VTE and anticoagulation on an interactive timeline, created for this study. In the week preceding the consultation when treatment duration is decided, patients were provided with daily videos using push notifications. Outcomes were assessed through online, self-reported questionnaires at baseline, 1-2 days before and 1 day after the consultation. Data were analyzed using t-tests and linear mixed models for repeated measurements. Results : The trial was terminated early as the inclusion rates dropped due to the COVID-19 pandemic. Data of 56 patients were analyzed (mean age 57±13;27% female). Satisfaction with received information was very heterogeneous and generally higher after the consultation (Figure 1). On a numeric rating scale from 0 to 10, patients who received the app were 0.86 points (95% CI 0.04 to 1.68;p 0.04) more satisfied with the provided information (Table 2). Patients who received the app experienced significantly less decisional conflict. No significant differences in satisfaction with knowledge, perceived knowledge and physician-reported SDM were observed. FIGURE 1 Change of satisfaction with received information over time stratified by intervention group. Measured at baseline, 1-2 days prior to consultation (“pre”) and 1 day after consultation (“post”). Colored lines indicate individual patients;blue thick line and confidence band indicate mean and 95% confidence interval Conclusions : An educational app about VTE and anticoagulation increases patients'satisfaction and reduces decisional conflict when deciding to stop or continue anticoagulation after initial treatment for VTE.

Electrocardiogram-based mortality prediction in patients with COVID-19 using machine learning

Background The electrocardiogram (ECG) is an easy to assess, widely available and inexpensive tool that is frequently used during the work-up of hospitalized COVID-19 patients. So far, no study has been conducted to evaluate if ECG-based machine learning models are able to predict allcause in-hospital mortality in COVID-19 patients. Purpose With this study, we aim to evaluate the value of using the ECG to predict in-hospital all-cause mortality of COVID-19 patients by analyzing the ECG at hospital admission, comparing a logistic regression based approach and a DNN based approach. Secondly, we aim to identify specific ECG features associated with mortality in patients diagnosed with COVID-19. Methods and results We studied 882 patients admitted with COVID-19 across seven hospitals in the Netherlands. Raw-format 12-lead ECGs recorded after admission (<72 hours) were collected, manually assessed, and annotated using pre-defined ECG features. Using data from five out of seven centers (n = 634), two mortality prediction models were developed: (a) a logistic regression model using manually annotated ECG features, and (b) a pre-trained deep neural network (DNN) using the raw ECG waveforms. Data from two other centers (n = 248) were used for external validation. Performance of both prediction models was similar, with a mean area under the receiver operating curve of 0.69 [95%CI 0.55- 0.82] for the logistic regression model and 0.71 [95%CI 0.59-0.81] for the DNN in the external validation cohort. After adjustment for age and sex, ventricular rate (OR 1.13 [95% CI 1.01-1.27] per 10 ms increase), right bundle branch block (3.26 [95% CI 1.15-9.50]), ST-depression (2.78 [95% CI 1.03-7.70]) and low QRS voltages (3.09 [95% CI 1.02-9.38]) remained as significant predictors for mortality. Conclusion: This study shows that ECG-based prediction models at admission may be a valuable addition to the initial risk stratification in admitted COVID-19 patients. The DNN model showed similar performance to the logistic regression that needs time-consuming manual annotation. Several ECG features associated with mortality were identified.

An interrupted time series analysis of the lockdown policies in India: a national-level analysis of COVID-19 incidence.

India implemented a national mandatory lockdown policy (Lockdown 1.0) on 24 March 2020 in response to Coronavirus Disease 2019 (COVID-19). The policy was revised in three subsequent stages (Lockdown 2.0-4.0 between 15 April to 18 May 2020), and restrictions were lifted (Unlockdown 1.0) on 1 June 2020. This study evaluated the effect of lockdown policy on the COVID-19 incidence rate at the national level to inform policy response for this and future pandemics. We conducted an interrupted time series analysis with a segmented regression model using publicly available data on daily reported new COVID-19 cases between 2 March 2020 and 1 September 2020. National-level data from Google Community Mobility Reports during this timeframe were also used in model development and robustness checks. Results showed an 8% [95% confidence interval (CI) = 6-9%] reduction in the change in incidence rate per day after Lockdown 1.0 compared to prior to the Lockdown order, with an additional reduction of 3% (95% CI = 2-3%) after Lockdown 4.0, suggesting an 11% (95% CI = 9-12%) reduction in the change in COVID-19 incidence after Lockdown 4.0 compared to the period before Lockdown 1.0. Uptake of the lockdown policy is indicated by decreased mobility and attenuation of the increasing incidence of COVID-19. The increasing rate of incident case reports in India was attenuated after the lockdown policy was implemented compared to before, and this reduction was maintained after the restrictions were eased, suggesting that the policy helped to 'flatten the curve' and buy additional time for pandemic preparedness, response and recovery.