Differences in Diabetic Prescription Drug Utilization and Costs Among Patients With Diabetes Enrolled in Colorado Marketplace and Medicaid Plans, 2014-2015.

Importance: Increasing prices of antidiabetic medications in the US have raised substantial concerns about the effects of drug affordability on diabetes care. There has been little rigorous evidence comparing the experiences of patients with diabetes across different types of insurance coverage. Objective: To compare the utilization patterns and costs of prescription drugs to treat diabetes among low-income adults with Medicaid vs those with Marketplace insurance in Colorado during 2014 and 2015. Design, Setting, and Participants: This cross-sectional study included diabetic patients enrolled in Colorado Medicaid and Marketplace plans who were aged 19 to 64 years and had incomes between 75% and 200% of the federal poverty level during 2014 and 2015. Data analysis was conducted from September 2020 to April 2021. Exposures: Health insurance through Colorado Medicaid or Colorado's state-based Marketplace. Main Outcomes and Measures: Primary outcomes were drug utilization (prescription drug fills) and drug costs (total costs and out-of-pocket costs). The secondary outcome was months with an active prescription for noninsulin antidiabetic medications. An all payer claims database was combined with income data, and linear models were used to adjust for clinical and demographic confounders. Results: Of 22 788 diabetic patients included in the study, 20 245 were enrolled in Medicaid and 2543 in a Marketplace plan. Marketplace-eligible individuals were older (mean [SD] age, 52.12 [10.60] vs 47.70 [11.33] years), and Medicaid-eligible individuals were more likely to be female (12 429 [61.4%] vs 1413 [55.6%]). Medicaid-eligible patients were significantly more likely than Marketplace-eligible patients to fill prescriptions for dipeptidyl peptidase 4 inhibitors (adjusted difference, -3.7%; 95% CI, -5.3 to -2.1; P < .001) and sulfonylureas (adjusted difference, -6.6%; 95% CI, -8.9 to -4.3; P < .001). Overall rates of insulin use were similar in the 2 groups (adjusted difference, -2.3%; -5.1 to 0.5; P = .11). Out-of-pocket costs for noninsulin medications were 84.4% to 95.2% lower and total costs were 9.4% to 54.2% lower in Medicaid than in Marketplace plans. Out-of-pocket costs for insulin were 76.7% to 94.7% lower in Medicaid than in Marketplace plans, whereas differences in total insulin costs were mixed. The percentage of months of apparent active medication coverage was similar between the 2 groups for 4 of 5 drug classes examined, with Marketplace-eligible patients having a greater percentage of months than Medicaid-eligible patients for sulfonylureas (adjusted difference, 5.3%; 95% CI, 0.3%-10.4%; P = .04). Conclusions and Relevance: In this cross-sectional study, drug utilization across multiple drug classes was higher and drug costs were significantly lower for adults with diabetes enrolled in Medicaid than for those with subsidized Marketplace plans. Patients with Marketplace coverage had a similar percentage of months with an active prescription as patients with Medicaid coverage.

The general fault in our fault lines.

Pervading global narratives suggest that political polarization is increasing, yet the accuracy of such group meta-perceptions has been drawn into question. A recent US study suggests that these beliefs are inaccurate and drive polarized beliefs about out-groups. However, it also found that informing people of inaccuracies reduces those negative beliefs. In this work, we explore whether these results generalize to other countries. To achieve this, we replicate two of the original experiments with 10,207 participants across 26 countries. We focus on local group divisions, which we refer to as fault lines. We find broad generalizability for both inaccurate meta-perceptions and reduced negative motive attribution through a simple disclosure intervention. We conclude that inaccurate and negative group meta-perceptions are exhibited in myriad contexts and that informing individuals of their misperceptions can yield positive benefits for intergroup relations. Such generalizability highlights a robust phenomenon with implications for political discourse worldwide.

COVID-19-Related Insurance Coverage Changes and Disparities in Access to Care Among Low-Income US Adults in 4 Southern States.

Importance: It is unclear how the COVID-19 pandemic and its associated economic downturn have affected insurance coverage and disparities in access to health care among low-income families and people of color in states that have and have not expanded Medicaid. Objective: To determine changes in insurance coverage and disparities in access to health care among low-income families and people of color across 4 Southern states and by Medicaid expansion status. Design Setting and Participants: This random-digit dialing telephone survey study of US citizens ages 19 to 64 years with a family income less than 138% of the federal poverty line in in 4 states (Arkansas, Kentucky, Louisiana, and Texas) was conducted from October to December 2020. Using a difference-in-differences design, we estimated changes in outcomes by Medicaid expansion status overall and by race and ethnicity in 2020 (n = 1804) compared with 2018 to 2019 (n = 5710). We also explored barriers to health care and use of telehealth by race and ethnicity. Data analysis was conducted from January 2021 to March 2021. Exposures: COVID-19 pandemic and prior Medicaid expansion status. Main Outcomes and Measures: Primary outcome was the uninsured rate and secondary outcomes were financial and nonfinancial barriers to health care access. Results: Of 7514 respondents (11% response rate; 3889 White non-Latinx [51.8%], 1881 Black non-Latinx [25.0%], and 1156 Latinx individuals [15.4%]; 4161 women [55.4%]), 5815 (77.4%) were in the states with previous expansion and 1699 (22.6%) were in Texas (nonexpansion state). Respondents in the expansion states were older, more likely White, and less likely to have attended college compared with respondents in Texas. Uninsurance rate in 2020 rose by 7.4 percentage points in Texas (95% CI, 2.2-12.6; P = .01) and 2.5 percentage points in expansion states (95% CI, -1.9 to 7.0; P = .27), with a difference-in-differences estimate for Medicaid expansion of -4.9% (95% CI, -11.3 to 1.6; P = .14). Among Black and Latinx individuals, Medicaid expansion was associated with protection against a rise in the uninsured rate (difference-in-differences, -9.5%; 95% CI, -19.0 to -0.1; P = .048). Measures of access, including having a personal physician and regular care for chronic conditions, worsened significantly in 2020 in all 4 states, with no significant difference by Medicaid expansion status. Conclusions and Relevance: In this survey of US adults, uninsured rates increased among low-income adults in 4 Southern states during the COVID-19 pandemic, but Medicaid expansion states, that association was diminished among Black and Latinx individuals. Nonfinancial barriers to care because of the pandemic were common in all states.

Replicating patterns of prospect theory for decision under risk.

Prospect theory is among the most influential frameworks in behavioural science, specifically in research on decision-making under risk. Kahneman and Tversky's 1979 study tested financial choices under risk, concluding that such judgements deviate significantly from the assumptions of expected utility theory, which had remarkable impacts on science, policy and industry. Though substantial evidence supports prospect theory, many presumed canonical theories have drawn scrutiny for recent replication failures. In response, we directly test the original methods in a multinational study (n = 4,098 participants, 19 countries, 13 languages), adjusting only for current and local currencies while requiring all participants to respond to all items. The results replicated for 94% of items, with some attenuation. Twelve of 13 theoretical contrasts replicated, with 100% replication in some countries. Heterogeneity between countries and intra-individual variation highlight meaningful avenues for future theorizing and applications. We conclude that the empirical foundations for prospect theory replicate beyond any reasonable thresholds.