Dianne McWilliam

Nursing home care presents a significant fiscal challenge in the U.S., with an aging population increasing demand and an industry plagued by decades of poor quality of care. With the need for systemic change spotlighted by the Covid-19 pandemic, one proposed solution is more private rooms. Despite growing interest, policymakers lack critical evidence required to determine whether policy should incentivize nursing homes to provide more private rooms: do private rooms actually improve residents' health? This study provides the first causal estimates of the effect of receiving a private room on post-acute nursing home residents' health outcomes. First, using standardized clinical assessments that list residents' room number, I implement a novel method to determine whether residents' rooms are private or shared - data that is otherwise unavailable. Second, to address concerns of selection into room types within nursing homes, I use a new instrumental variable for the room type residents receive: the availability of private rooms in the nursing home on the day before the resident enters. Among ∼380,000 post-acute care residents across the U.S. in 2017-2019, I do not find evidence that receiving a private room improves health outcomes, including mortality, hospital readmission, discharge to home, length of stay, and resident mood. If anything, estimates are suggestive of private rooms potentially being worse for post-acute resident health, with private room receipt leading to lower probability of discharge directly to home, within 30, 60, and 90 days after nursing home entry. These results imply that policymakers seeking to improve post-acute resident health should consider prioritizing other proven nursing home quality improvement interventions. 

Elder maltreatment is prevalent, costly, and understudied. The most common perpetrators of elder maltreatment are family caregivers. With rising labor market participation by family caregivers and an aging population, diverging supply and demand for informal care places greater demands on caregivers increasingly balancing employment and caregiving responsibilities, and caregivers may be more likely to be directly affected by changes in the economy. However, little is known about how macroeconomic conditions affect elder maltreatment. Changes in economic conditions may affect both the amount and nature (e.g., stress, financial strain) of eldercare provided – two factors that could affect elder maltreatment rates. This study will assess trends in domestic elder maltreatment in the U.S. and examine the causal effect of macroeconomic conditions on elder maltreatment, both overall and by maltreatment type. I exploit within-state quarterly variation in macroeconomic conditions (e.g., unemployment rates) and leverage new administrative data capturing rich case-level information on elder maltreatment investigations by state Adult Protective Services programs, from 2015Q4-2019Q4. 

This research is supported by National Institute of Aging of the National Institutes of Health under award number: P30AG066582 (CAPRA Pilot Program Award)

This paper examines the effect of access to paid family leave (PFL) in New York, Rhode Island, New Jersey, and California, on the amount of time individuals spend caring for older adults. State-time variation in the implementation of PFL policies allows for comparison of the differences in individuals’ caregiving before and after PFL access, to that of individuals in states with no change in PFL access, using daily time use data. Event study estimates reveal that PFL implementation in New York and Rhode Island increased time spent on eldercare, driven by a rise in intensive margin caregiving. Individuals in New York and Rhode Island spent an average of 7.4 additional minutes per day providing eldercare, a 114% increase over the pre-PFL mean. Results for California and New Jersey are less conclusive. Policymakers should account for the effects of increased time spent on eldercare on both caregivers’ and care recipients’ wellbeing, including on growing elder abuse.

Objectives: To determine whether the risk of nursing home-to-nursing home transfer is higher among long-term nursing home residents with intellectual and developmental disabilities (IDD), or serious mental illness (SMI), and/or Alzheimer’s disease and related dementias (ADRD), relative to residents without these clinical diagnoses, and to assess the factors associated with transfer overall and for residents with these diagnoses. 

Design: Cross-sectional cohort study of nursing home residents in 2019. 

Setting and Participants: Michigan long-term nursing home residents identified from the Minimum Data Set. 

Methods: Residents who had a nursing home-to-nursing home transfer were allocated into 5 groups: IDD, SMI and ADRD, ADRD, SMI, and all others. We examined transfer rates for each group. We assessed the odds of transfer for each group, adjusting for resident and nursing home characteristics, and behavioral issue and clinical indicators. We used stratified logistic regression to determine factors associated with transfers within each group.

Results: Among 37,638 long-term nursing home residents, 2.3% had a nursing home-to-nursing home transfer. Transfers varied across diagnosis groups: 1.9% in IDD, 2.8% in SMI and ADRD, 1.9% in ADRD, 2.6% in SMI, and 2.5% in all others. After adjustment, residents in the SMI and ADRD group were 39% more likely to transfer than those in the all others group (adjusted odds ratio, 1.39; 95% CI, 1.14-1.68). Aged >=75 years, moderate to severe cognitive impairment, and falls were associated with a lower odds of transfer. Being married and wandering were associated with a higher odds of transfer. Factors associated with transfers varied within each group. 

Conclusions and Implications: Among Michigan long-term nursing home residents, residents with both SMI and ADRD are at the highest risk for transfer to another nursing home when compared with residents with SMI or ADRD alone and with IDD and none of these diagnoses. Understanding the drivers of transfer of this vulnerable group warrants further investigation. 

Within the fundamental determinants of cross-country income inequality, political institutions represent a hallmark factor that societies can influence. Focusing on the portion of inequality explainable by differences in political institutions, we decompose annual cross-country Gini coefficients from 1960–2012. Although inequality has marginally decreased since 1988, the portion that cannot be explained by political institutions has increased markedly, with the explanatory power of institutions falling rapidly from the late 1980s to the early 1990s. This result prevails when using alternative variables, weightings, samples, and controls, and appears to be unlikely to be driven by contemporary regional events alone. However, we find that the link between institutions and income levels has become increasingly nonlinear as countries with the most inclusive political institutions enjoy even higher incomes than before. Our results imply that, if we hold societies responsible for their political institutions, cross-country inequality has become notably less fair since the late 1980s.