Algorithmic fairness in pandemic forecasting: lessons from COVID-19.
NPJ digital medicine
2022; 5 (1): 59
Racial and ethnic minorities have borne a particularly acute burden of the COVID-19 pandemic in the United States. There is a growing awareness from both researchers and public health leaders of the critical need to ensure fairness in forecast results. Without careful and deliberate bias mitigation, inequities embedded in data can be transferred to model predictions, perpetuating disparities, and exacerbating the disproportionate harms of the COVID-19 pandemic. These biases in data and forecasts can be viewed through both statistical and sociological lenses, and the challenges of both building hierarchical models with limited data availability and drawing on data that reflects structural inequities must be confronted. We present an outline of key modeling domains in which unfairness may be introduced and draw on our experience building and testing the Google-Harvard COVID-19 Public Forecasting model to illustrate these challenges and offer strategies to address them. While targeted toward pandemic forecasting, these domains of potentially biased modeling and concurrent approaches to pursuing fairness present important considerations for equitable machine-learning innovation.
View details for DOI 10.1038/s41746-022-00602-z
View details for PubMedID 35538215
Marketplace Health Insurance Ratings: Most Potential Enrollees Have Access To Plans Of Medium Or High Quality.
Health affairs (Project Hope)
2022; 41 (3): 390-397
The Affordable Care Act (ACA) Marketplace plays a critical role in providing affordable health insurance for the nongroup market, yet the accessibility of plans from insurers with high quality ratings has not been investigated. Our analysis of recently released insurer quality star ratings for plan year 2020 found substantial variation in access to high rated plans in the federally facilitated ACA Marketplace. In most participating counties (1,390 of 2,265, or 61.4percent), the highest-rated ACA Marketplace insurer had a three-star rating. Fewer than one-third of counties (703, or 31.0percent) had access to four- or five-star-rated insurers. Fewer than 10percent (172, or 7.6percent) had access to only one- or two-star-rated insurers. In plan-based analyses, each one-point increase in star rating was associated with a $28 increase in the average monthly plan premium. Counties with the highest proportion of residents obtaining individual coverage through the ACA Marketplace and counties with more insurers were the most likely to have access to plans from high-rated insurers. We found no systematic racial or ethnic disparities in access to plans from high-rated insurers. Policy makers should continue to monitor the quality of available health plans.
View details for DOI 10.1377/hlthaff.2021.00922
View details for PubMedID 35254939
Early social distancing policies in Europe, changes in mobility & COVID-19 case trajectories: Insights from Spring 2020
2021; 16 (6): e0253071
Social distancing have been widely used to mitigate community spread of SARS-CoV-2. We sought to quantify the impact of COVID-19 social distancing policies across 27 European counties in spring 2020 on population mobility and the subsequent trajectory of disease.We obtained data on national social distancing policies from the Oxford COVID-19 Government Response Tracker and aggregated and anonymized mobility data from Google. We used a pre-post comparison and two linear mixed-effects models to first assess the relationship between implementation of national policies and observed changes in mobility, and then to assess the relationship between changes in mobility and rates of COVID-19 infections in subsequent weeks.Compared to a pre-COVID baseline, Spain saw the largest decrease in aggregate population mobility (~70%), as measured by the time spent away from residence, while Sweden saw the smallest decrease (~20%). The largest declines in mobility were associated with mandatory stay-at-home orders, followed by mandatory workplace closures, school closures, and non-mandatory workplace closures. While mandatory shelter-in-place orders were associated with 16.7% less mobility (95% CI: -23.7% to -9.7%), non-mandatory orders were only associated with an 8.4% decrease (95% CI: -14.9% to -1.8%). Large-gathering bans were associated with the smallest change in mobility compared with other policy types. Changes in mobility were in turn associated with changes in COVID-19 case growth. For example, a 10% decrease in time spent away from places of residence was associated with 11.8% (95% CI: 3.8%, 19.1%) fewer new COVID-19 cases.This comprehensive evaluation across Europe suggests that mandatory stay-at-home orders and workplace closures had the largest impacts on population mobility and subsequent COVID-19 cases at the onset of the pandemic. With a better understanding of policies' relative performance, countries can more effectively invest in, and target, early nonpharmacological interventions.
View details for DOI 10.1371/journal.pone.0253071
View details for Web of Science ID 000671698800008
View details for PubMedID 34191818
View details for PubMedCentralID PMC8244916
Impacts of social distancing policies on mobility and COVID-19 case growth in the US
2021; 12 (1): 3118
Social distancing remains an important strategy to combat the COVID-19 pandemic in the United States. However, the impacts of specific state-level policies on mobility and subsequent COVID-19 case trajectories have not been completely quantified. Using anonymized and aggregated mobility data from opted-in Google users, we found that state-level emergency declarations resulted in a 9.9% reduction in time spent away from places of residence. Implementation of one or more social distancing policies resulted in an additional 24.5% reduction in mobility the following week, and subsequent shelter-in-place mandates yielded an additional 29.0% reduction. Decreases in mobility were associated with substantial reductions in case growth two to four weeks later. For example, a 10% reduction in mobility was associated with a 17.5% reduction in case growth two weeks later. Given the continued reliance on social distancing policies to limit the spread of COVID-19, these results may be helpful to public health officials trying to balance infection control with the economic and social consequences of these policies.
View details for DOI 10.1038/s41467-021-23404-5
View details for Web of Science ID 000658773600013
View details for PubMedID 34035295
View details for PubMedCentralID PMC8149701
Comparison of General Surgical Practice Size and Setting in 2017 vs 2013 in the US
JAMA NETWORK OPEN
2021; 4 (4): e216848
As health care delivery markets have changed and new payment models have emerged, physicians in many specialties have consolidated their practices, but whether this consolidation has occurred in surgical practices is unknown.To examine changes in the size of surgical practices, market-level factors associated with this consolidation, and how place of service for surgical care delivery varies by practice size.A cross-sectional study of Medicare Data on Provider Practice and Specialty from January 1 to December 31, 2013, compared with January 1 to December 31, 2017, was conducted on all general surgeon practices caring for patients enrolled in Medicare in the US. Data analysis was performed from November 4, 2019, to January 9, 2020.Practice sizes in 2013 and 2017 were compared relative to hospital market concentration measured by the Herfindahl-Hirschman Index in the hospital referral region.The primary outcome was the change in size of surgical practices over the study period. Secondary outcomes included change in surgical practice market concentration and the place of service for provision of surgical care stratified by surgical practice size.From 2013 to 2017, the number of surgical practices in the US decreased from 10 432 to 8451. The proportion of surgeons decreased in practices with 1 (from 26.2% to 17.4%), 2 (from 8.3% to 6.6%), and 3 to 5 (from 18.0% to 16.5%) surgeons, and the proportion of surgeons in practices with 6 or more surgeons increased (from 47.6% to 59.5%). Hospital concentration was associated with an increase in the size of the surgical practice. Each 10% increase in the hospital market concentration was associated with an increase of 0.204 surgeons (95% CI, 0.020-0.388 surgeons; P = .03) per practice from 2013 to 2017. Similarly, a 10% increase in the hospital-level HHI was associated with an increase in the surgical practice HHI of 0.023 (95% CI, 0.013-0.033; P < .001). Large surgical practices increased their share of Medicare services provided from 36.5% in 2013 to 45.6% in 2017. Large practices (31.3% inpatient in 2013 to 33.1% in 2017) were much more likely than small practices (19.0% inpatient in 2013 to 17.7% in 2017) to be based in hospital settings and this gap widened over time.Surgeons have increasingly joined larger practices over time, and there has been a significant decrease in solo, small, and midsize surgical practices. The consolidation of surgeons into larger practices appears to be associated with hospital market concentration in the same market. Although overall care appears to be more hospital based for larger practices, the association between the consolidation of surgical practices and patient access and outcomes should be studied.
View details for DOI 10.1001/jamanetworkopen.2021.6848
View details for Web of Science ID 000645614000001
View details for PubMedID 33909056
View details for PubMedCentralID PMC8082320
- Policy Implications of Artificial Intelligence in Surgery Artificial Intelligence In Surgery: Understanding The Role Of AI In Surgical Practice McGraw Hill. 2021
Association of community-level social vulnerability with US acute care hospital intensive care unit capacity during COVID-19.
Healthcare (Amsterdam, Netherlands)
2021; 10 (1): 100611
The COVID-19 pandemic has placed unprecedented stress on US acute care hospitals, leading to overburdened ICUs. It remains unknown if increased COVID-19 ICU occupancy is crowding out non-COVID-related care and whether hospitals in vulnerable communities may be more susceptible to ICUs reaching capacity. Using facility-level hospitalization data, we conducted a retrospective observational cohort study of 1753 US acute care hospitals reporting to the US Department of Health and Human Services Protect database from September 4, 2020 to February 25, 2021. 63% of hospitals reached critical ICU capacity for at least two weeks during the study period, and the surge of COVID-19 cases appeared to be crowding out non-COVID-19-related intensive care needs. Hospitals in the South (OR = 3.31, 95% CI OR 2.31-4.78) and West (OR = 2.28, 95% CI OR 1.51-3.46) were more likely to reach critical capacity than those in the Northeast, and hospitals in areas with the highest social vulnerability were more than twice as likely to reach capacity as those in the least vulnerable areas (OR = 2.15, 95% CI OR 1.41-3.29). The association between social vulnerability and critical ICU capacity highlights underlying structural inequities in health care access and provides an opportunity for policymakers to take action to prevent strained ICU capacity from compounding COVID-19 inequities.
View details for DOI 10.1016/j.hjdsi.2021.100611
View details for PubMedID 34979442