Deepfake detection with and without content warnings.

The rapid advancement of 'deepfake' video technology-which uses deep learning artificial intelligence algorithms to create fake videos that look real-has given urgency to the question of how policymakers and technology companies should moderate inauthentic content. We conduct an experiment to measure people's alertness to and ability to detect a high-quality deepfake among a set of videos. First, we find that in a natural setting with no content warnings, individuals who are exposed to a deepfake video of neutral content are no more likely to detect anything out of the ordinary (32.9%) compared to a control group who viewed only authentic videos (34.1%). Second, we find that when individuals are given a warning that at least one video in a set of five is a deepfake, only 21.6% of respondents correctly identify the deepfake as the only inauthentic video, while the remainder erroneously select at least one genuine video as a deepfake.

Financial incentives for COVID-19 vaccines in a rural low-resource setting: a cluster-randomized trial.

We implemented a clustered randomized controlled trial with 6,963 residents in six rural Ghana districts to estimate the causal impact of financial incentives on coronavirus disease 2019 (COVID-19) vaccination uptake. Villages randomly received one of four video treatment arms: a placebo, a standard health message, a high cash incentive (60 Ghana cedis) and a low cash incentive (20 Ghana cedis). For the first co-primary outcome-COVID-19 vaccination intentions-non-vaccinated participants assigned to the cash incentive treatments had an average rate of 81% (1,733 of 2,168) compared to 71% (1,895 of 2,669) for those in the placebo treatment arm. For the other co-primary outcome of self-reported vaccinations 2 months after the initial intervention, the average rate for participants in the cash treatment was 3.5% higher than for participants in the placebo treatment (95% confidence interval (CI): 0.001, 6.9; P = 0.045): 40% (602 of 1,486) versus 36.3% (672 of 1,850). We also verified vaccination status of participants: in the cash treatment arm, 36.6% (355 of 1,058) of verified participants had at least one dose of the COVID-19 vaccine compared to 30.3% (439 of 1,544) for those in the placebo-a difference of 6.3% (95% CI: 2.4, 10.2; P = 0.001). For the intention and the vaccination outcomes, the low cash incentive (20 Ghana cedis) had a larger positive effect on COVID-19 vaccine uptake than the high cash incentive (60 Ghana cedis). Trial identifier: AEARCTR-0008775 .

Prevalence, predictors and reasons for COVID-19 vaccine hesitancy: Results of a global online survey.

Vaccine hesitancy has the potential to cripple efforts to end the COVID-19 pandemic. Policy makers need to be informed about the scale, nature and drivers of this problem, both domestically and globally, so that effective interventions can be designed. To this end, we conducted a statistical analysis of data from the CANDOUR survey (n = 15,536), which was carried out in 13 countries representing approximately half of the global population. Both pooled and country-level ordered regression models were estimated to identify predictors of vaccine hesitancy and reasons for not getting vaccinated. We found high levels of hesitancy, particularly in high-income countries. Factors driving moderate hesitancy differed from those driving extreme hesitancy. A lack of trust in health care providers was consistently the underlying driver of more extreme hesitancy. Predictors of moderate hesitancy varied across countries, though being younger and female was typically associated with greater hesitancy. While political ideology played a role in vaccine hesitancy in some countries, this effect was often moderated by income level, particularly in the US. Overall, the results suggest that different interventions such as mass-media campaigns and monetary incentives may be needed to target the moderately versus extremely hesitant. The lack of trust in health care professionals that drives extreme hesitancy may reflect deep societal mistrust in science and institutions and be challenging to overcome.

The COVID-19 pandemic and health-related quality of life across 13 high- and low-middle-income countries: A cross-sectional analysis.

BACKGROUND: Most research on the Coronavirus Disease 2019 (COVID-19) health burden has focused on confirmed cases and deaths, rather than consequences for the general population's health-related quality of life (HRQoL). It is also important to consider HRQoL to better understand the potential multifaceted implications of the COVID-19 pandemic in various international contexts. This study aimed to assess the association between the COVID-19 pandemic and changes in HRQoL in 13 diverse countries. METHODS AND FINDINGS: Adults (18+ years) were surveyed online (24 November to 17 December 2020) in 13 countries spanning 6 continents. Our cross-sectional study used descriptive and regression-based analyses (age adjusted and stratified by gender) to assess the association between the pandemic and changes in the general population's HRQoL, measured by the EQ-5D-5L instrument and its domains (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression), and how overall health deterioration was associated with individual-level (socioeconomic, clinical, and experiences of COVID-19) and national-level (pandemic severity, government responsiveness, and effectiveness) factors. We also produced country-level quality-adjusted life years (QALYs) associated to COVID-19 pandemic-related morbidity. We found that overall health deteriorated, on average across countries, for more than one-third of the 15,480 participants, mostly in the anxiety/depression health domain, especially for younger people (<35 years old) and females/other gender. This translated overall into a 0.066 mean "loss" (95% CI: -0.075, -0.057; p-value < 0.001) in the EQ-5D-5L index, representing a reduction of 8% in overall HRQoL. QALYs lost associated with morbidity were 5 to 11 times greater than QALYs lost based on COVID-19 premature mortality. A limitation of the study is that participants were asked to complete the prepandemic health questionnaire retrospectively, meaning responses may be subject to recall bias. CONCLUSIONS: In this study, we observed that the COVID-19 pandemic was associated with a reduction in perceived HRQoL globally, especially with respect to the anxiety/depression health domain and among younger people. The COVID-19 health burden would therefore be substantially underestimated if based only on mortality. HRQoL measures are important to fully capture morbidity from the pandemic in the general population.

Is There Broad-Based Support in High-Income Countries for COVID-19 Vaccine Donation? Evidence from Seven Countries.

BACKGROUND: Many high-income countries (HICs) have now vaccinated a substantial proportion of their population against COVID-19. Many low-income countries (LICs) may need to wait until at least 2022 before even the most vulnerable 20% of their populations are vaccinated. Beyond ethical considerations, some redistribution of doses would reduce the risk of the emergence and spread of new variants and benefit the economy, both globally and in donor countries. However, the willingness of HIC governments to donate vaccine doses is likely to depend on public support. While previous work has indicated strong average levels of public support in HIC for donation, little is known about how broad-based this support is. OBJECTIVE: To investigate the extent to which support for donation holds across both pre-specified and exploratory subgroups. METHODS: From 24 November-28 December 2020 we conducted an online survey of 8209 members of the general public in seven HIC (Australia, Canada, France, Italy, Spain, UK and USA). We conducted tests of proportions and used Bayesian ordinal logistic regression models to assess the extent of support for donation across population subgroups. RESULTS: We found broad-based support for donations in terms of age, gender, socio-economic status and political ideology. We found no strong evidence that support for donations was higher among those with greater income or a university education. Support for donation among those on the political right and centre was lower than on the left, but 51% (95% confidence interval 48-53%) of respondents who identified with the right supported some level of donation. Those in the more altruistic half of the sample (as captured by willingness to donate money to a good cause) were more likely to support donation than those who were not, but around half of the less altruistic group supported some level of donation. CONCLUSION: There is broad-based support for policymakers in HICs to donate some of their countries' COVID-19 vaccine doses for distribution to LICs.

Gender differences in perceived risk of COVID-19.

OBJECTIVE: We examine gender-based differences in perceived risks related to COVID-19. METHODS: We analyze published findings from COVID-related research on beliefs and attitudes about the health risks posed by the pandemic. We also design and administer a pair of online survey experiments (n = 502) to test if and how responsive men's attitudes are to information about male-specific risks. RESULTS: Across 16 studies, men consistently express lower perceived risk of contracting COVID-19 and less concern about the potential health consequences if they were to catch it. Our experimental results are mixed: Results for one information treatment indicate that men report greater relative risk of adverse outcomes. Men in one of the risk information treatments express less concern for their health if they were to contract the disease. Risk perceptions are positively correlated with self-reported propensity toward protective behaviors. CONCLUSION: Our review of recent studies finds a small-but-consistent gap in men and women's beliefs about their health risks related to the present pandemic. These risk beliefs are crucial determinants of whether individuals take protective measures. Our experimental results suggest that informing men of male-specific risks associated with COVID-19 can reduce their risk perceptions and perceptions of risk and widen the gap between men and women's perceptions.

Polling India via regression and post-stratification of non-probability online samples.

Recent technological advances have facilitated the collection of large-scale administrative data and the online surveying of the Indian population. Building on these we propose a strategy for more robust, frequent and transparent projections of the Indian vote during the campaign. We execute a modified MrP model of Indian vote preferences that proposes innovations to each of its three core components: stratification frame, training data, and a learner. For the post-stratification frame we propose a novel Data Integration approach that allows the simultaneous estimation of counts from multiple complementary sources, such as census tables and auxiliary surveys. For the training data we assemble panels of respondents from two unorthodox online populations: Amazon Mechanical Turks workers and Facebook users. And as a modeling tool, we replace the Bayesian multilevel regression learner with Random Forests. Our 2019 pre-election forecasts for the two largest Lok Sahba coalitions were very close to actual outcomes: we predicted 41.8% for the NDA, against an observed value of 45.0% and 30.8% for the UPA against an observed vote share of just under 31.3%. Our uniform-swing seat projection outperforms other pollsters-we had the lowest absolute error of 89 seats (along with a poll from 'Jan Ki Baat'); the lowest error on the NDA-UPA lead (a mere 8 seats), and we are the only pollster that can capture real-time preference shifts due to salient campaign events.

Citizens from 13 countries share similar preferences for COVID-19 vaccine allocation priorities.

How does the public want a COVID-19 vaccine to be allocated? We conducted a conjoint experiment asking 15,536 adults in 13 countries to evaluate 248,576 profiles of potential vaccine recipients who varied randomly on five attributes. Our sample includes diverse countries from all continents. The results suggest that in addition to giving priority to health workers and to those at high risk, the public favors giving priority to a broad range of key workers and to those with lower income. These preferences are similar across respondents of different education levels, incomes, and political ideologies, as well as across most surveyed countries. The public favored COVID-19 vaccines being allocated solely via government programs but were highly polarized in some developed countries on whether taking a vaccine should be mandatory. There is a consensus among the public on many aspects of COVID-19 vaccination, which needs to be taken into account when developing and communicating rollout strategies.

Lessons from the pandemic on the value of research infrastructure.

The COVID-19 pandemic has shed a spotlight on the resilience of healthcare systems, and their ability to cope efficiently and effectively with unexpected crises. If we are to learn one economic lesson from the pandemic, arguably it is the perils of an overfocus on short-term allocative efficiency at the price of lack of capacity to deal with uncertain future challenges. In normal times, building spare capacity with 'option value' into health systems may seem inefficient, the costs potentially exceeding the benefits. Yet the fatal weakness of not doing so is that this can leave health systems highly constrained when dealing with unexpected, but ultimately inevitable, shocks-such as the COVID-19 pandemic. In this article, we argue that the pandemic has highlighted the potentially enormous option value of biomedical research infrastructure. We illustrate this with reference to COVID-19 response work supported by the United Kingdom National Institute for Health Research Oxford Biomedical Research Centre. As the world deals with the fallout from the most serious economic crisis since the Great Depression, pressure will soon come to review government expenditure, including research funding. Developing a framework to fully account for option value, and understanding the public appetite to pay for it, should allow us to be better prepared for the next emerging problem.