Low-value service utilization is common among all older adults, and utilization of some high-value services decreases after the onset of cognitive decline.
ABSTRACT
Objectives: Cognitive impairment and dementia have rising prevalence and impact the health care utilization and lives of older adults. Receipt of low-value (LV) care and underutilization of high-value (HV) care by individuals with these cognitive disorders may have negative consequences for patient health, health system efficiency, and societal welfare. Evidence on health care value among cognitively impaired individuals is limited; we thus ascertained receipt of LV and HV health care in older adults with normal cognition, cognitive impairment without dementia (CIND), and dementia.
Study Design: Retrospective cohort study of Health and Retirement Study data linked to Medicare claims (1996-2018).
Methods: We examined the association between cognitive decline and the receipt of 5 LV and 7 HV services vs individuals with no change in cognition.
Results: Receipt of LV care ranged from 4% to 13% regardless of cognitive status. Cognitive decline (from unimpaired to either CIND or dementia) was associated with decreased probability of receipt of 1 LV service (colorectal cancer screening at 85 years and older [5-percentage-point reduction; P = .047]) and 3 HV services (glucose-lowering drugs [7-percentage-point reduction; P = .029], statins [32-percentage-point reduction; P = .045], and antiresorptive therapy [61-percentage-point reduction; P = .019]).
Conclusions: LV service receipt is wasteful and may be harmful, but it was not consistently associated with cognitive status. Lack of HV care for those with cognitive impairment could be a missed opportunity to improve well-being or reduce preventable adverse events. Our results suggest opportunities for improving the quality of care received by all older adults, including those with cognitive impairment.
Am J Manag Care. 2024;30(7):316-323. https://doi.org/10.37765/ajmc.2024.89580
Takeaway Points
Low-value care utilization is common among older adults, including those with cognitive impairment without dementia and those with dementia, and utilization of some high-value services decreases as individuals experience cognitive decline.
CMS defines low-value (LV) health care as services from which the majority of consumers would not derive a clinical benefit and high-value (HV) health care as services that most individuals would benefit from and have a strong clinical evidence base demonstrating appropriate care.1 LV care can cause iatrogenic harm and unnecessarily raise costs, whereas HV services have demonstrated benefit for patients and society.2 Each year, approximately 25% to 42% of Medicare beneficiaries receive some form of LV care, suggesting that wasteful use of unnecessary services is pervasive among older adults.3 Importantly, older adults also have a higher prevalence of cognitive impairment without dementia (CIND) and of dementia, herein referred to collectively as cognitive impairment. Together, these conditions affect an estimated 20% to 35% of individuals 65 years and older in the US,4,5 and they have a high and growing burden that disproportionately affects older adults and care partners of minoritized racial and ethnic groups.6,7 With the increased prevalence of cognitive impairment at older ages coinciding with more health care utilization in general,6 even small amounts of LV care, or deficits in HV care, could have considerable consequences for patient welfare, equity, and health system waste.2,8,9
Cognitive impairment could affect both the demand and supply of medical treatments and interventions in several ways. Perception of benefit may be altered by cognitive impairment, both among affected individuals and their care partners at home, and it may also be altered among clinicians because of insufficient information about symptoms, communication difficulties, less applicable treatment guidelines, different expectations of patient adherence or life expectancy, or different perceptions of patient and physician autonomy. It is also possible that transitions in care that coincide with cognitive decline could alter the use of interventions. There is partial support for some of these mechanisms from existing evidence, including lower rates of guideline-concordant care for specific diseases among individuals with cognitive impairment.10-12 This may reflect changes in the way clinicians make recommendations about care once cognitive impairment is present. In addition, responsibility for care decisions that are typically informed by patient preference will eventually shift toward care partners, and new experimental evidence shows that surrogate decision makers are less likely to choose life-extending care for individuals with dementia compared with those without dementia.13 Despite the potential significance of understanding the use of LV and HV care in this important and vulnerable subpopulation, no published studies to our knowledge have specifically examined patterns in relation to cognitive impairment.
In this study, we examined the relationship between the use of 5 LV and 7 HV services in individuals with incident CIND and dementia vs those with no change in cognition. These 12 services were selected in accordance with task force recommendations and peer-reviewed literature because of their unambiguous lack of clinical benefit (LV) or unambiguous evidence of clinical benefit (HV) for older adults regardless of cognitive status. We hypothesized that the difficulties in providing care for individuals with cognitive decline (as described in the previous paragraph) would lead to more LV and less HV care while also acknowledging the possibility that differing care regimens, outside normal patterns of care, could potentially insulate them from common LV care practices that occur for older adults without cognitive impairment.
METHODS
Data and Sample
We conducted a retrospective cohort study of Health and Retirement Study (HRS) respondents whose data were linked to Medicare claims data in the years 1996 to 2018 (waves 3-14). The HRS, conducted by the University of Michigan (grant No. NIA U01AG009740), includes standardized cognitive assessment every 2 years.14 After in-person baseline interviews, most follow-up interviews since 1996 have been conducted remotely via telephone and featured the Telephone Interview for Cognitive Status (TICS). Among HRS respondents, 80% have consented to have their data linked with Medicare claims, which allows observation of all diagnoses and health care utilization. Participation in HRS does not include disclosure of cognitive or other research assessments to the patient’s treating clinicians, although participants are free to do so if they choose.
Our sample includes person-waves of community-dwelling HRS respondents, 65 years and older, with linkable Medicare claims data and continuous enrollment in traditional (fee-for-service) Medicare for 12 months after the date of cognitive assessment. The characteristics of these individuals are shown in Table 1. Individuals were censored upon death, discontinuation of HRS interviews, or disenrollment from Medicare. In longitudinal analyses in which we examined changes in cognitive status across time, we required individuals to have at least 4 consecutive waves (representing at least 6 years of follow-up) of observation in HRS, with censoring after a change in cognitive category. In some analyses, we imposed further restrictions based on sex, age, and diagnoses based on relevance to the definition of each LV and HV service (details described in the following section).
LV and HV Services
The outcomes of this study were binary measures of the utilization of services that are widely believed to be LV and HV and are listed in Table 2. To select these services, we examined lists from the Choosing Wisely initiative, the US Preventive Services Task Force, and peer-reviewed literature.3,8,9,15,16 Through an iterative process of discussion with practicing clinicians with geriatric expertise (E.A.P., S.T., S.B.) and experts in health care value (D.B., A.M.F., A.B.), we chose services that are generally accepted as either LV or HV regardless of the cognitive status of the patient, meaning that the clinical benefit and overall utility of the service do not depend on the patient’s degree of cognitive impairment. A detailed description of this process can be found in eAppendix A (eAppendices available at ajmc.com). The key considerations for each service were the time horizon to benefit, quantifiable effects on mortality or quality of life, burden associated with the service, and possible effects on care partners. Accordingly, the LV services will have a longer time horizon to benefit, moderate to high burden of care, or lack of relative net benefit, even for adults with dementia. Conversely, the HV services have a short to intermediate time horizon to benefit, low to moderate burden of care, and relative net benefit even for adults with dementia. The benefits of the HV services occur regardless of age. We also required that services be commonly used in current clinical practice with older adults and observable in claims data.
Our analytic sample varied depending on which service was the outcome of interest. In all analyses (adjusted and unadjusted), we restricted to person-waves with the age, sex, and diagnoses that would make the person eligible to receive the designated LV or HV service. For example, in our analyses of prostate-specific antigen–based screening for prostate cancer (LV for men 75 years and older during the data collection period), we omitted women and everyone younger than 75 years. Claims for prescription drugs are only observable after the introduction of Part D in 2006; analyses for prescription drug outcomes were thus restricted to the years 2006 to 2018. For services where value depends on past diagnoses, we required continuous Medicare enrollment for 12 months prior to the date of cognitive assessment. For details of sample selection for each service, please see Table 3 [part A and part B]. This table displays the set of restrictions applied overall (sample A) as well as service-specific restrictions for different analyses (samples B and C). Note that some service definitions required inclusion and exclusion diagnoses, which are listed in eAppendix B; specific National Drug Codes are listed in eAppendix C.
Cognitive Impairment Measure
The primary exposure was a decline in cognitive status, as measured by the modified TICS at the respondent’s most recent HRS interview. A 27-point scale was calculated based on immediate and delayed 10-noun free recall, serial 7 subtraction, and a backward count from 20. Proxy responses were not used. Cognitive status was defined as normal (12-27), CIND (7-11), and dementia (0-6). It should be noted that CIND is similar, but not identical, to mild cognitive impairment.17 These cut points have been validated against similar definitions from the Aging, Demographics, and Memory Study, during which neuropsychological and clinical assessments were used to obtain gold-standard diagnoses of CIND and dementia.17,18 Notably, these cut points do not equate to a clinical diagnosis of CIND or dementia made in a health care setting, where a diagnosis may affect how patients seek care and how physicians make decisions about care provision; indeed, other research has shown that utilization may change at the time of diagnosis, with some studies showing an increase and others showing a decrease.19,20 This study, however, used survey-based measures of cognitive status that providers were not necessarily aware of; accordingly, our analyses focused on the specific role of cognition (and not diagnoses) in decisions for the provision of HV and LV services.
Statistical Analysis
We calculated the unadjusted probabilities of use for each service, stratified by cognitive status. In our primary analyses, we identified the associations of each service with changing to CIND and changing to dementia compared with remaining cognitively normal. We used linear probability models with individual fixed effects, with separate regressions examining binary dependent variables for each of the 12 services of interest.
Utilizationit+1 = β0 + β1CIND_switchit + β2ADRD_switchit + β3Xi + β4Wt + εit
Each dependent variable was equal to 1 if the individual used that service in the 12 months after the cognitive assessment from that wave. The exposure variables were binary indicators of whether the individual had changed cognitive status (to either CIND or dementia) after being measured as cognitively normal in all previous waves. Individuals were censored after a change in cognitive status, meaning that any subsequent change in status did not contribute to the analysis. In addition to individual fixed effects (Xi), we also included wave fixed effects (Wt) in all analyses; these adjustments control for unobserved confounding across individuals and time. SEs were clustered at the individual level. The coefficients measured in these analyses can be interpreted as follows: β1 is the change in probability of using each service associated with changing from normal cognition to CIND, and β2 is the change in probability of using each service associated with changing from normal cognition to dementia. Both are measured in comparison to consistently normal cognition and are adjusted for individual and time-varying confounding (individual and wave fixed effects). Relationships with P values less than .05 were considered significant.
RESULTS
Table 1 shows the sociodemographic characteristics of the set of HRS respondents with linkable Medicare claims and sufficient cognitive assessment data for our statistical analyses. This sample was 64% women, with a mean age of 78.7 years at the time of cognitive assessment. Reported race was 87% White, 11% Black, and 3% other; 6% reported Hispanic ethnicity.
Table 2 shows the percentages using each service, stratified by cognitive status, among the person-waves where the individual had the sex, age, and diagnoses making them eligible to receive the service of interest. Use of the LV services ranged from 4% to 13%. All 5 LV services had lower (unadjusted) utilization probabilities for individuals with dementia than for individuals with normal cognition. Among respondents with CIND, utilization probabilities generally fell between the probabilities for individuals without cognitive impairment and those for individuals with dementia. All HV services had lower utilization probabilities (unadjusted) among individuals with dementia compared with those with normal cognition, except for antiresorptive therapy for osteoporosis, which was slightly more common among individuals with dementia.
Table 4 shows the results of adjusted longitudinal analyses of the association between the use of each service and changes in cognition from normal to CIND and normal to dementia. The use of several services was significantly associated with these transitions in cognitive status. For example, among those with diabetes, incident CIND was associated with a 7-percentage-point reduction in the probability of a glucose-lowering drug prescription (coefficient = –0.07; P = .029), which is considered an HV service. Among those with hyperlipidemia, incident dementia was associated with a lower probability of being prescribed a statin (HV; coefficient = –0.32; P = .045). Among those with osteoporosis, incident dementia was associated with a lower probability of use of antiresorptive therapy (HV; coefficient = –0.61; P = .019). Incident CIND was also associated with a lower likelihood of colorectal cancer screening (LV at 85 years and older; coefficient = –0.05; P = .047).
DISCUSSION
This study examined the receipt of clinical services considered to be LV and HV and the relationship between service receipt and cognitive status, as measured by a validated cognitive assessment instrument.17,18 Value was defined as an unambiguous lack of or evidence of clinical benefit, without reference to the cognitive status of an individual. We found that LV care receipt is relatively common, with unadjusted probabilities ranging from 4% to 13% regardless of cognitive status (Table 2). With analyses that adjusted for individual-level and time-varying confounding over at least 6 consecutive years of follow-up, we found that the use of several services was lower when individuals experienced cognitive decline but that use of other services did not change.
The relatively common use of LV services over 20-plus years of observation and regardless of cognitive status suggests persistent waste in the health care of older adults. The lower utilization of several LV services in CIND and dementia, relative to no cognitive impairment (Table 2), suggests that decisions about the use of such services may be influenced by cognitive status and that individuals with cognitive impairment are not at special risk for increased exposure to LV care. On the other hand, lower use of 3 HV services after cognitive decline—glucose-lowering drugs, statins, and antiresorptive therapy—is worrisome because these individuals would be expected to benefit from these services. A drop in the use of HV services may represent a missed opportunity because optimal (evidence-based) care could result in improvement or stability of health, quality of life, and well-being.20 These results have implications for care partners, who may be able to advocate for better care for individuals with cognitive impairment, and for providers, who have a responsibility to provide HV services and avoid LV services.
We hypothesized that cognitive decline would be associated with receipt of more LV services and fewer HV services, but our results were mixed (Table 4). The drop in utilization for some HV services could reflect the difficulties of care for individuals with cognitive decline, such as communication difficulties, unclear treatment guidelines, different expectations of patient adherence or life expectancy, transitions of care, or different perceptions of patient and physician autonomy. It is also possible that as cognition worsened, there may have been fewer opportunities for both HV and LV services to be provided; this has been demonstrated in other studies, in which individuals with undiagnosed dementia were found to be more likely to not show up for scheduled appointments.21
The inconsistency of results across services and across the dimension of value suggests that as hypothesized, there were indeed changes in care, but in ways that were not reflective of evidence-based recommendations that define value. These findings are consistent with those of other studies showing persistent use of LV services, emphasizing the importance of policies that disincentivize LV care. Related evidence from Schwartz et al, who examined the receipt of 26 LV services, showed that 25% to 42% of individuals received at least 1 LV service in a given year.3 This relatively common use of LV care is costly.9 At the same time, the 7 HV services we examined appear to be underutilized (3%-78%, depending on the service), representing a missed opportunity for care with proven benefits. Other recent work examined changes in LV and HV care utilization that occurred during the COVID-19 pandemic and found that after utilization dropped off at the start of the pandemic, both HV and LV services rebounded.22 As with our study, this work suggests an unnuanced approach to clinical decision-making, with insufficient attention to evidence-based recommendations.
Strengths and Limitations
The strengths of this study include the linkage of longitudinal, validated cognitive assessment data to claims data for services that are widely regarded as unambiguously LV or HV when applied without regard to cognitive status. One limitation is that despite the proven clinical benefit of these services, unobserved personal factors may affect an individual’s health care preferences for—or actual benefit from—a given service. For example, benefits from antihypertensives, statins, and glucose-lowering drugs would only be realized with good adherence to those therapies. This means that for some individuals, services we designated as either high (or low) value may not be truly high (or low) value. It is possible that utilization of other LV and HV services, not examined here, may have different associations with cognitive decline. We did not consider services unique to the care of individuals with cognitive impairment, as value assessment has not been rigorously applied to such services in that context. Although we used individual fixed effects to control for potential confounding by time-invariant respondent characteristics, we did not control for confounding that varies within an individual across time. For example, we did not control for confounding related to an individual entering hospice; however, this was extremely rare in our sample (only 2 individuals used hospice during the follow-up period). Relatedly, the focus on incident cognitive decline led to a less representative sample because we did not examine those transitioning from CIND to dementia. One limitation of linear probability models is that predictions can be outside the range of possible values for a probability (ie, < 0 or > 1); however, their ability to estimate relationships with individual fixed effects in data where some individuals have full separation of the outcome makes them preferable to alternative estimation methods. There is a possibility that multiple hypothesis testing may have led to spurious significant associations. We did not perform the Bonferroni correction because of its tendency to produce false negatives, which occurs when multiple hypotheses are not independent of each other, as was the case in our study.23 Finally, some of our analyses have sample sizes that may be too small to identify statistically significant associations.
CONCLUSIONS
CMS has explicitly called for programs and policies, such as value-based insurance design, to enhance equity and address “overused, higher-cost services,”1 which includes LV services. These policies cannot, however, be adequately targeted and implemented without a better understanding of which patient subgroups and specific clinical services are most affected. This study confirms that individuals experiencing cognitive decline, like other older adults, may be exposed to persistent overuse of LV care, and it provides new evidence on potential underuse of HV care following a transition in cognitive status. The point of newly detectable cognitive decline could therefore be a focal point for reevaluation of evidence-based recommendations regarding general health care. Future work should continue to characterize the quality of care delivered to cognitively impaired older adults,24 with the goal of developing dementia-specific guidelines to guide providers in the delivery of care that both reduces the risk of harm and promotes optimal health and well-being.
Author Affiliations: The Comparative Health Outcomes, Policy, and Economics (CHOICE) Institute, University of Washington (DB, SJ, AB), Seattle, WA; Division of Gerontology and Geriatric Medicine (EAP) and Department of Psychiatry and Behavioral Sciences (ST, SB), University of Washington School of Medicine, Seattle, WA; Department of Health Systems and Population Health, University of Washington School of Public Health (EAP), Seattle, WA; Department of Family Medicine, Keck School of Medicine, University of Southern California (SB), Los Angeles, CA; Center for Value-Based Insurance Design, University of Michigan (AMF), Ann Arbor, MI.
Source of Funding: This work was supported by a Research Starter Grant from the PhRMA Foundation. Dr Barthold is supported by NIA K01AG071843. Support for data access and analyses for this research came from the University of Washington (UW) Population Health Initiative, UW’s Student Technology Fee program, the UW’s provost’s office, and a Eunice Kennedy Shriver National Institute of Child Health and Human Development research infrastructure grant, P2C HD042828, to the Center for Studies in Demography and Ecology at UW.
Prior Presentation: A summary of this study was presented at the annual meeting of ISPOR—The Professional Society for Health Economics and Outcomes Research in May 2022.
Author Disclosures: Dr Basu received consulting fees from Salutis Consulting LLC outside of this work. Dr Fendrick reports serving as a consultant to AbbVie, CareFirst BlueCross BlueShield, Centivo, Community Oncology Alliance, EmblemHealth, Employee Benefit Research Institute, Exact Sciences, GRAIL, Health at Scale Technologies,* HealthCorum, Hopewell Fund, Hygieia, Johnson & Johnson, Medtronic, MedZed, Merck, Mother Goose Health,* Phathom Pharmaceuticals, Proton Intelligence, RA Capital Management, Sempre Health,* Silver Fern Healthcare,* Teladoc Health, US Department of Defense, Virginia Center for Health Innovation, Washington Health Benefit Exchange, Wellth,* Yale New Haven Health System, and Zansors* (asterisks indicate equity interest); research funding from Arnold Ventures, National Pharmaceutical Council, Patient-Centered Outcomes Research Institute, Pharmaceutical Research and Manufacturers of America, and Robert Wood Johnson Foundation; and outside positions as co–editor in chief of The American Journal of Managed Care®, past member of the Medicare Evidence Development & Coverage Advisory Committee, and partner at VBID Health, LLC. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (DB, ST, AMF); acquisition of data (DB); analysis and interpretation of data (DB, SJ, AB, EAP, ST, SB, AMF); drafting of the manuscript (DB, EAP, ST, SB); critical revision of the manuscript for important intellectual content (DB, SJ, AB, EAP, ST, SB, AMF); statistical analysis (DB, SJ, AB); obtaining funding (DB, AMF); administrative, technical, or logistic support (DB); and supervision (DB, AMF).
Address Correspondence to: Douglas Barthold, PhD, The Comparative Health Outcomes, Policy, and Economics (CHOICE) Institute, University of Washington, 1959 NE Pacific St, Box 357630, Seattle, WA 98195. Email: barthold@uw.edu.
REFERENCES
1. CMS; HHS. Patient Protection and Affordable Care Act; HHS notice of benefit and payment parameters for 2021; notice requirement for non-federal governmental plans. Fed Regist. 2020;85(94):29164-29262.
2. Mafi JN, Parchman M. Low-value care: an intractable global problem with no quick fix. BMJ Qual Saf. 2018;27(5):333-336. doi:10.1136/bmjqs-2017-007477
3. Schwartz AL, Landon BE, Elshaug AG, Chernew ME, McWilliams JM. Measuring low-value care in Medicare. JAMA Intern Med. 2014;174(7):1067-1076. doi:10.1001/jamainternmed.2014.1541
4. Zissimopoulos JM, Tysinger BC, St Clair PA, Crimmins EM. The impact of changes in population health and mortality on future prevalence of Alzheimer’s disease and other dementias in the United States. J Gerontol B Psychol Sci Soc Sci. 2018;73(suppl 1):S38-S47. doi:10.1093/geronb/gbx147
5. Langa KM, Levine DA. The diagnosis and management of mild cognitive impairment: a clinical review. JAMA. 2014;312(23):2551-2561. doi:10.1001/jama.2014.13806
6. 2022 Alzheimer’s disease facts and figures. Alzheimers Dement. 2022;18(4):700-789. doi:10.1002/alz.12638
7. Navaie-Waliser M, Feldman PH, Gould DA, Levine C, Kuerbis AN, Donelan K. The experiences and challenges of informal caregivers: common themes and differences among Whites, Blacks, and Hispanics. Gerontologist. 2001;41(6):733-741. doi:10.1093/geront/41.6.733
8. Cliff BQ, Avanceña ALV, Hirth RA, Lee SD. The impact of Choosing Wisely interventions on low-value medical services: a systematic review. Milbank Q. 2021;99(4):1024-1058. doi:10.1111/1468-0009.12531
9. Oronce CIA, Fendrick AM, Ladapo JA, Sarkisian C, Mafi JN. The utilization and costs of Grade D USPSTF services in Medicare, 2007-2016. J Gen Intern Med. 2021;36(12):3711-3718. doi:10.1007/s11606-021-06784-8
10. Levine DA, Langa KM, Galecki A, et al. Mild cognitive impairment and receipt of treatments for acute myocardial infarction in older adults. J Gen Intern Med. 2020;35(1):28-35. doi:10.1007/s11606-019-05155-8
11. Levine DA, Langa KM, Fagerlin A, et al. Physician decision-making and recommendations for stroke and myocardial infarction treatments in older adults with mild cognitive impairment. PLoS One. 2020;15(3):e0230446. doi:10.1371/journal.pone.0230446
12. Sloan FA, Trogdon JG, Curtis LH, Schulman KA. The effect of dementia on outcomes and process of care for Medicare beneficiaries admitted with acute myocardial infarction. J Am Geriatr Soc. 2004;52(2):173-181. doi:10.1111/j.1532-5415.2004.52052.x
13. Nicholas LH, Langa KM, Halpern SD, Macis M. How do surrogates make treatment decisions for patients with dementia: an experimental survey study. Health Econ. 2024;33(6):1211-1228. doi:10.1002/hec.4810
14. Health and Retirement Study. Accessed January 20, 2021. https://hrs.isr.umich.edu/about
15. Chernew ME, Fendrick AM, Buxbaum J, Budros M. V-BID X: creating a value-based insurance design plan for the exchange market. Health Care Markets and Regulations Lab. June 2019. Accessed January 20, 2021. https://vbidcenter.org/wp-content/uploads/2019/09/VBID-X-White-Paper-92019.pdf
16. Cassel CK, Guest JA. Choosing wisely: helping physicians and patients make smart decisions about their care. JAMA. 2012;307(17):1801-1802. doi:10.1001/jama.2012.476
17. Crimmins EM, Kim JK, Langa KM, Weir DR. Assessment of cognition using surveys and neuropsychological assessment: the Health and Retirement Study and the Aging, Demographics, and Memory Study. J Gerontol B Psychol Sci Soc Sci. 2011;66(suppl 1):i162-i171. doi:10.1093/geronb/gbr048
18. Langa KM, Larson EB, Crimmins EM, et al. A comparison of the prevalence of dementia in the United States in 2000 and 2012. JAMA Intern Med. 2017;177(1):51-58. doi:10.1001/jamainternmed.2016.6807
19. Hoffman GJ, Maust DT, Harris M, Ha J, Davis MA. Medicare spending associated with a dementia diagnosis among older adults. J Am Geriatr Soc. 2022;70(9):2592-2601. doi:10.1111/jgs.17835
20. McCormick WC, Kukull WA, van Belle G, Bowen JD, Teri L, Larson EB. The effect of diagnosing Alzheimer’s disease on frequency of physician visits: a case-control study. J Gen Intern Med. 1995;10(4):187-193. doi:10.1007/BF02600253
21. Lee SJ, Larson EB, Dublin S, Walker R, Marcum Z, Barnes D. A cohort study of healthcare utilization in older adults with undiagnosed dementia. J Gen Intern Med. 2018;33(1):13-15. doi:10.1007/s11606-017-4162-3
22. Shahzad M, Song Z, Chernew ME, Fendrick AM. Changes in use of low-value services during the COVID-19 pandemic. Am J Manag Care. 2022;28(11):600-604. doi:10.37765/ajmc.2022.89031
23. Moran MD. Arguments for rejecting the sequential Bonferroni in ecological studies. Oikos. 2003;100(2):403-405. doi:10.1034/j.1600-0706.2003.12010.x
24. Gotanda H, Nuckols T, Mori K, Tsugawa Y. Comparison of the quality of chronic disease management between adults with and without dementia. JAMA Netw Open. 2021;4(5):e219622. doi:10.1001/jamanetworkopen.2021.9622
Despite Record ACA Enrollment, Report Reveals Underinsured Americans are in Crisis
November 21st 2024Despite significant progress in expanding health insurance coverage since the Affordable Care Act (ACA) was enacted, millions of Americans still face critical gaps in access and affordability to health care.
Read More
Sustaining Compassionate Trauma Care Across Communities
September 30th 2024September is National Recovery Month, and we are bringing you another limited-edition month-long podcast series with our Strategic Alliance Partner, UPMC Health Plan. In our final episode, we speak with Lyndra Bills, MD, and Shari Hutchison, MS.
Listen
Racial Inequities in Guideline-Adherent Breast Cancer Care and Timely Treatment
November 19th 2024Older non-Hispanic Black adults with early-stage breast cancer are less likely to receive timely treatment and guideline-concordant care, increasing their risk of death compared with non-Hispanic White women.
Read More