About

Andrew Cohen, MD, DPhil, is an Associate Professor of Medicine (Geriatrics) at Yale School of Medicine. He attended medical school at the University of Pennsylvania and earned a doctorate in English literature at Oxford as a Marshall Scholar. Following residency training in internal medicine and a fellowship in geriatrics, he joined the Yale faculty in 2015. His long-term research goal is to improve the care of vulnerable older patients with dementia, with a particular focus on those lacking traditional supports and unable to rely on family or friends as they develop increased cognitive and functional impairments. Dr. Cohen's work has been recognized with honors such as the Paul B. Beeson Emerging Leaders in Aging award from the National Institute on Aging and the Outstanding Junior Investigator of the Year by the American Geriatrics Society. Clinically, he serves as a primary care physician for the Home Based Primary Care program at VA Connecticut.

Research topics

• Medicine
• Psychiatry
• Internal medicine
• Gerontology
• Nursing
• Demography
• Law
• Medical emergency
• Pharmacology
• Pediatrics
• Cardiology
• Family medicine
• Intensive care medicine

Selected publications

• Symptom Trajectories After COVID Hospitalization and Risk Factors for Symptom Burden in Older Persons: a Longitudinal Cohort Study

  The Journals of Gerontology Series A · 2025-06-13

  article

  BACKGROUND: Little is known about how psychosocial factors and geriatric conditions contribute to persistent post-COVID symptoms among older adults. We evaluated symptom burden following COVID-19 hospitalization and identified risk factors for persistent symptoms among -community-dwelling older adults. METHODS: This prospective study recruited 281 older persons (mean age 70.6 years) hospitalized for SARS-CoV-2 infection between June 2020 and June 2021 from Yale-New Haven Health System. Post-COVID symptoms were assessed using a modified Edmonton Symptom Assessment System during hospitalization, and at 1, 3, and 6 months post-discharge. Trajectory analysis identified three symptom trajectories. Multinomial logistic regression evaluated associations between characteristics (sociodemographic, clinical, psychosocial factors, and geriatric conditions) obtained during hospitalization and trajectory membership. RESULTS: Three symptom burden trajectory groups were identified: low (n = 70; 24.9%; reference); moderate (n = 149; 53.0%); and high (62; 22.1%). Female sex (adjusted odds ratio (adjOR)_moderate = 3.10 [95% CI = 1.68-5.72]; adjOR_high = 5.76 [2.70-12.27]), higher depression/anxiety (adjOR_moderate = 1.47 [1.24-1.74]; adjOR_high = 1.72 [1.43-2.07]), and less social support (adjOR_moderate = 0.91 [0.83, 0.99]; adjOR_high = 0.86 [0.78-0.95]) were associated with moderate and high symptom burden. Geriatric conditions, including delirium (adjOR_high = 7.74 [1.56-38.26]), frailty (adjOR_high = 5.26 [1.77-15.68]), impairment of physical function (adjOR_high = 1.18 [1.00-1.40]), and vision impairment (adjOR_high = 4.63 [1.33-16.11]), were associated with high symptom burden. CONCLUSIONS: In older persons hospitalized with COVID-19, female sex, psychosocial factors, and geriatric conditions were associated with higher symptom burden over six months. Future work should investigate the biopsychosocial mechanisms through which psychosocial factors and geriatric conditions contribute to post-COVID symptom burden.

  PublisherDOI

• Comparing Disaster‐Related Surge Staffing Practices Across Rural and Non‐Rural Nursing Homes

  Journal of the American Geriatrics Society · 2025-09-12

  article

  BACKGROUND: Clarifying whether rural and non-rural nursing homes implement different disaster-related staffing approaches is vital to developing targeted preparedness strategies. METHODS: In this retrospective cohort study of 361 nursing homes exposed to Hurricane Michael (2018), we used a comparative interrupted time series design to assess changes in staffing intensity and composition from 30 days before landfall to 7 days after. We examined three outcomes using the Centers for Medicare & Medicaid Services (CMS) Payroll Based Journal: total staffing intensity (hours per resident per day), licensure composition (percentage of registered nurses (RN) and licensed practical nurses (LPN)), and employment composition (percentage of routine staff who are not independent contractors). We evaluated staffing intensity with linear mixed-effects models and licensure and employment composition using linear probability models. We report point estimates and 95% confidence intervals (CI). RESULTS: Of 361 nursing homes, 39.9% were rural. At baseline, rural facilities had 8% (0.92; 95% CI 0.87-0.96) lower total staffing intensity compared to non-rural facilities. In the post-event period, rural facilities increased total staffing by an additional 4% (95% CI 1.02 to 1.06) in the post-event period compared to non-rural facilities. Rural facilities increased their routine staff representation (0.46% points; 95% CI 0.14 to 0.78) but decreased licensed staff representation (-1.00% points; 95% CI -1.39 to 0.61) compared to the post-disaster staffing changes in non-rural facilities. CONCLUSIONS: Distinct disaster-related staffing approaches may heighten baseline staffing differences between rural and non-rural nursing homes. Disaster-related staffing intensity increased in rural facilities, with rural facilities starting from a lower baseline. Although rural facilities may have benefited from greater reliance on routine staff, they were unable to augment licensed nurse staffing. These findings offer foundational evidence regarding the influence of baseline staffing practices and local healthcare workforce conditions on disaster-related staffing strategies.

  PublisherDOI

• Federal inspection timing, not compliance, associated with nursing home post-disaster outcomes

  Health Affairs Scholar · 2025-12-18

  articleOpen access
  PublisherDOI

• Enhancing Dementia Care Via GUIDE: Focus on Older Adults Living Alone With Limited Supports

  Innovation in Aging · 2025-12-01

  articleOpen access

  Abstract Background Approximately one-fourth of older adults with dementia live alone in the United States (about 1.7 million people), often with limited or no caregiver support, making them a particularly challenging population to care for healthcare providers. These individuals often have limited support from caregivers, and it is unclear whether they will benefit from the Guiding an Improved Dementia Experience (GUIDE) national demonstration model, launched by the Centers for Medicare and Medicaid Services in July 2024, which emphasizes support for patient/caregiver dyads. To address this knowledge gap, we elicited the perspectives of national dementia experts on the GUIDE model’s ability to meet the needs of older adults living alone with limited caregiver support. Methods From October-November 2024, teleconference interviews were conducted with 10 national dementia policy experts working in research, consulting, and advocacy institutions. Interview transcripts were analyzed using a combined inductive and deductive content analysis approach. Results All experts concurred that obstacles may prevent older adults living alone with dementia from benefitting fully from GUIDE, for 5 reasons: : 1) no benefit exists equivalent to what is provided for caregivers’ respite ($2,500/year); 2) monthly reimbursements of $215 may be insufficient to support this high-needs population; 3) GUIDE-related medication plans may be impractical; 4) navigators are mostly available via telephone rather than in-home; 5) the GUIDE evaluation ignores living arrangements and includes only interviews with caregiver/patient dyads. Conclusions Findings underscore the importance of developing comprehensive care interventions to effectively support older adults with dementia who live alone with limited caregiver support.

  PublisherOA PDFDOI

• Reply to: Reevaluating Functional Recovery Post‐ <scp>COVID</scp> ‐19: A Call for Broader Considerations

  Journal of the American Geriatrics Society · 2025-05-20

  letterOpen access

  We appreciate the opportunity to respond to the commentary by Sathian and colleagues [1] regarding our recent publication [2]. We welcome the discussion that advances our understanding of post-COVID-19 functional trajectories in older adults. Our study aimed to delineate distinct functional trajectories in the 6 months following hospitalization. A second objective was to determine whether six factors—age, comorbidities, in-hospital delirium, frailty, pre-admission disability, and severity of illness—were associated with trajectory membership. We selected these factors a priori because they are well-established determinants of outcomes in geriatric populations. In analyses of this kind, it is essential that the explanatory variables (in our study, these six factors) temporally precede the outcome (the repeated functional measures used to fit the functional trajectories). To maintain temporal separation, it was therefore not possible to include post-discharge factors such as rehabilitation practices, psychological support, and community care resources as explanatory variables in the analysis. While these post-discharge factors are important, their role in recovery after a COVID-19 hospitalization should be the focus of a future study with a different design. We fully agree with the comment that social determinants of health (SDOH), such as socioeconomic status and healthcare access, play a role in recovery from COVID-19. Our cohort was diverse, with substantial representation of individuals from varied socioeconomic backgrounds, which strengthens the generalizability of our findings. While a full evaluation of SDOH was outside the scope of this study, additional analyses of the VALIANT cohort are underway to further elucidate the contribution of SDOH to other patient-centered outcomes after COVID-19 hospitalization. We hope that this forthcoming work, and future research from other investigators, will help inform the care of disadvantaged older adults who have survived a COVID-19 hospitalization. Preparation of the manuscript: Jim Q. Ho, Andrew B. Cohen, and Lauren E. Ferrante. Review and revision of the manuscript for important intellectual content: all authors. This work was supported by the Yale Claude D. Pepper Older Americans Independence Center (P30AG021342 and P30AG021342-18S1). Dr. Cohen and Dr. Ferrante were supported by Paul B. Beeson Emerging Leaders in Aging awards (K76AG059987 and K76AG057023, respectively) from the National Institute on Aging. Dr. Murphy was supported by the National Center for Advancing Translational Sciences at the National Institutes of Health (UL1TR002014). The funders of this work were not involved in the preparation of the paper. The authors declare no conflicts of interest. This publication is linked to a related article by Sathian et al. To view this article, visit https://doi.org/10.1111/jgs.19533.

  PublisherOA PDFDOI

• Functional Trajectories After <scp>COVID</scp> ‐19 Hospitalization Among Older Adults

  Journal of the American Geriatrics Society · 2025-03-17 · 2 citations

  articleOpen access

  BACKGROUND: Little is known about functional trajectories among older adults who survive hospitalization for coronavirus disease 2019 (COVID-19). We characterized these trajectories over 6 months following discharge and evaluated the associations of potential risk factors with trajectory membership. METHODS: Participants were community-dwelling adults ≥ 60 years of age hospitalized for COVID-19 from June 2020 to June 2021. Interviews completed at 1, 3, and 6 months after discharge included assessments for disability in 15 functional activities. Functional trajectories were identified using latent class analysis. Factors associated with trajectory membership were evaluated using multinomial regression. RESULTS: 311 participants (mean age 71.3 years) were included. Four different functional trajectories were identified: no (43%), mild (16%), moderate (23%), and severe (18%) disability. The pre-admission count of disabilities was independently associated with membership in each non-reference trajectory. Additional factors independently associated with the moderate trajectory included in-hospital delirium (OR 4.12 [95% CI 1.11-15.4]), frailty (OR 1.67 [95% CI 1.12-2.50]) and number of comorbidities (OR 1.41 [95% CI 1.12-1.79]) and with the severe trajectory included in-hospital delirium (OR 12.4 [95% CI 1.93-79.4]), frailty (OR 2.01 [95% CI 1.11-3.62]), number of comorbidities (OR 1.59 [95% 1.11-2.28]), severity of illness (OR 1.46 [95% CI 1.09-1.95]), and age (OR 1.10 [95% CI 1.02-1.18]). CONCLUSIONS: Older survivors of COVID-19 hospitalization experience distinct functional trajectories. Our findings may help inform shared medical decision-making during and after hospitalization and stimulate further research into modifiable risk factors.

  PublisherOA PDFDOI

• Wage Labor History in Central and Southern Africa

  Oxford Research Encyclopedia of African History · 2025-11-18

  reference-entryOpen access1st authorCorresponding

  Abstract Labor in southern Africa changed rapidly over the 19th and 20th centuries as the region became firmly locked into the global capitalist economy. As wage labor developed in southern Africa, it adapted to the rapid changes brought about by the mineral revolution and the spread of commercial agriculture. This was also a period of rapid urbanization, creating both opportunities and challenges for workers. Labor migration and the mobility of both people and ideas tied the region together, often creating a shared experience of working life in both the formal and informal economies. Southern Africa’s political, social, and economic development was fundamentally altered by processes of colonization and the new forms of economic activity that arose from that intrusion. The large-scale exploitation of mineral resources in the late 19th century and the entrenchment of European colonialism reshaped African societies and changed the nature of how, and why, people worked. Broadly speaking, the push of taxes combined with the pull of wages in new mines, farms, and factories slowly coaxed Africans into the capitalist economy. Growing urban centers, and the new occupations they generated, provided alternatives to mine and farm work, as well as providing opportunities for livelihoods based in the “informal” economy, catering for the needs and desires of the growing number of urban residents.

  PublisherOA PDFDOI

• Barriers and Facilitators to Long-Term Services &amp; Supports for Older Adults Living Alone With Cognitive Impairment

  Innovation in Aging · 2025-12-01

  articleOpen accessSenior author

  Abstract Background More than 4 million older Americans with cognitive impairment (CI) live alone. These individuals often face unique challenges due to limited or non-existent support from caregivers. Identifying the barriers and facilitators to accessing long-term services and supports (LTSS) is critical to preventing the adverse outcomes to which this population is particularly vulnerable, including self-neglect, untreated medical conditions, medication mismanagement, falls, and fires. Methods This qualitative study, conducted between May 2016-February 2024, included adults aged 55+ from diverse racial and ethnic backgrounds who were living alone with CI in California, Louisiana, and Michigan, as well as members of their social circle. Factors influencing LTSS access were identified via in-person, semi-structured interviews conducted in English, Spanish, Cantonese, or Mandarin. Transcripts were analyzed with combined inductive and deductive content analysis drawing from the micro-meso-macro framework. Results A total of 119 older adults living alone with cognitive impairment (88 [71.5%] women; median age, 78 years [range, 57-103]) and 56 members of their social circle (45 [80.3%] women; median age, 45 years [range, 39-89]) were interviewed multiple times for a total of 558 interviews. Barriers to accessing LTSS included: healthcare providers being unhelpful, distrustful, dismissive, and unaffordable, with rigid protocols and long waiting periods; cognitive impairment itself in the context of living alone; limited resources. Facilitators included having chronic medical conditions; “hunting” for services; information from attentive healthcare providers; support with transportation. Conclusions Findings underscore the importance of ensuring that critical LTSS are tailored around the needs of older adults living alone with CI.

  PublisherOA PDFDOI

• Development and Feasibility of <scp>PATH</scp> : Preparation for Appropriate Transitions From Home

  Journal of the American Geriatrics Society · 2025-12-31

  articleSenior author

  BACKGROUND: Despite efforts to promote aging in place, millions of caregivers face decisions about whether a person living with dementia (PLwD) should move from home to another residential setting, such as a nursing home or assisted living facility. These decisions are frequently made during times of crisis, and caregivers report many unmet needs regarding decision making. Tools to support this decision making are lacking. METHODS: Focus groups of caregivers and an expert panel provided the content for a decision support tool. The tool was then iteratively reviewed with a new cohort of 53 caregivers, who participated in cognitive interviews and rated the tool's clarity, trustworthiness, and whether it made them more comfortable thinking about care transitions. RESULTS: In focus groups, caregivers highlighted the importance of understanding decision making in the context of the entire relationship between the caregiver and PLwD, and they repeatedly called attention to the role of guilt. The expert panel grappled with the ethical standing of the caregiver's well-being in decision making on behalf of the PLwD. The tool consists of two booklets. The first addresses the cognitive and emotional aspects of decision making and the second provides education. Over 95% of caregivers gave a rating of "Good" or "Very Good" for the booklets' clarity, ease of understanding, and trustworthiness. At least 80% agreed or strongly agreed that they were comfortable thinking about the issues and wanted to learn more. CONCLUSIONS: A decision support tool for caregivers of PLwD that provides education about transitions in care site, along with support for the emotional aspects of decision making, is highly acceptable and supports caregivers' ability to engage with what can be a challenging topic.

  PublisherDOI

• Bridging the Hearing Divide: Policy Solutions for Aging Americans

  American Journal of Public Health · 2025-11-20

  articleOpen accessSenior author

  Hearing loss affects approximately two thirds of adults in the United States aged 70 years or older and frequently remains untreated despite its well-documented harms, including accelerated cognitive decline, increased caregiver burden, and higher health care expenditures. We examine the major barriers to accessing high-quality hearing care, with particular attention to the complex and fragmented landscape of insurance coverage across Medicare, Medicaid, the US Department of Veterans Affairs, private plans, and over-the-counter (OTC) products. We review key legislative and regulatory developments over the past decade, most notably the 2022 establishment of OTC hearing aids, and summarize early opportunities and remaining gaps. We then propose targeted reforms to improve access and affordability, including more consistent Medicaid benefits, selective Medicare expansion, integration of teleaudiology, and strengthened oversight and consumer protections for OTC devices. Finally, we advance a technology-driven policy framework that integrates artificial intelligence–supported risk prediction, teleaudiology, real-time insurance verification, and a transparent device marketplace to modernize delivery and evaluation. Together, these strategies can catalyze a fundamental rethinking of how hearing health is prioritized and managed within the broader United States health care ecosystem. ( Am J Public Health. Published online ahead of print November 20, 2025:e1–e10. https://doi.org/10.2105/AJPH.2025.308298 )

  PublisherOA PDFDOI

Recent grants

• Guardianship and Medical Decision Making for the Unbefriended Elderly

  NIH · $251k · 2016–2019

• Dementia and Decision-Making for Older Adults without Surrogates

  NIH · $1.2M · 2018–2024

Frequent coauthors

• Terri R. Fried

  Yale University

  47 shared

• Gregory Y.H. Lip

  Liverpool Heart and Chest Hospital

  26 shared

• John R. O’Leary

  24 shared

• Andrew R. Zullo

  Brown University

  12 shared

• Eelko Ronner

  Reinier de Graaf Hospital

  12 shared

• Zhanquan Li

  Chinese Academy of Medical Sciences & Peking Union Medical College

  12 shared

• Rupak Datta

  Yale University

  12 shared

• Weihua Li

  Chongqing Three Gorges University

  12 shared

Labs

• GeriatricsPI

Education

• Other, English literature

  Oxford

• M.D.

  University of Pennsylvania

Awards & honors

• Paul B. Beeson Emerging Leaders in Aging award from the Nati…
• Recognition as the Outstanding Junior Investigator of the Ye…