What Do We (Still) Need to Know About Long-term Care Management?


Over the past 40 years there have been many long-term services and supports policies and programs that have used care management to determine eligibility, allocate resources, or coordinate care. However, the evidence base for understanding the efficacy and effectiveness of care management itself remains weak. This article summarizes long-standing questions about care management, surveys recent demonstration projects and policies that incorporate care management, and reviews important relevant developments in defining and measuring quality of home- and community-based services.

Key Words:

care management, long-term services and supports, home- and community-based services, Managed Long-term Services and Supports, quality

The evidence base for care management in long-term services and supports (LTSS) has a mixed history, going back to the Channeling Demonstration (Kane, 1988), which concluded in large part that neither “basic” or “enhanced” case management, in conjunction with an expanded set of home- and community-based services (HCBS) had an effect on the main outcome of preventing or delaying institutionalization over an eighteen to twenty-four month time period.

However, for all of its careful measurement and design, many questions were left unanswered (Applebaum, 2012). Some of these old questions remain open, and new questions have arisen. In 1980, when Channeling was first conceived, only 10 percent of Medicaid LTSS expenditures went to HCBS. Across subsequent decades, care-managed HCBS programs have grown steadily, and now account for 57 percent of Medicaid LTSS spending (Eiken et al., 2018).

Thus, even though there are gaps in our understanding of how care management functions, there is evidence that these programs have been able to shift the locus of care away from nursing homes and toward community settings. And, the recent shift to MLTSS in twenty-three states (AdvancingStates, 2020; Lewis et al., 2018) has increased the importance of understanding how care management functions, as it is now a key tool in private sector management of Medicaid LTSS benefits.

Old Questions That Deserve Answers

Looking back to the findings from the Channeling Demonstration (Applebaum, 2012), it is clear that case managers who implemented the intervention learned a lot about their local communities and how to serve program participants. However, the evaluation was not designed to generate evidence regarding the organization and delivery of case management itself.

For example, agency directors and program managers have little guidance regarding staffing ratios—agencies vary widely in the number of participants per case manager. Usually this is taken as an indicator of “intensity,” however, there is no evidence to support a given level of intensity. Related questions have to do with how specific cases should be assigned to individual case managers: Should “difficult” individuals be assigned to local experts, or distributed across the team so that no case manager has more than a few time-consuming participants?

Variability in how care management is reimbursed continues. Under Medicaid Waiver programs, states can elect to pay for care management as a service or as an administrative function (Centers for Medicare & Medicaid Services, 2019a). This distinction has important state budget consequences. It also has bearing on a separate issue of conflict of interest (more on this later).

However, as an empirical question, there is no evidence regarding whether the payment model has an impact on participant outcomes. Paying for care management as a service creates an incentive for agencies to submit more bills, thus leading states to either impose limits on the number of units of service that can be delivered, or impose a global budget. By contrast, as an administrative function, agencies may have an incentive to minimize the amount of care management per participant, e.g., by increasing caseloads. In addition, there is an open question as to whether being “part of the system” undermines the potential for care managers to play an advocacy role.

In 2012, a survey by National Association of States United for Aging and Disability (NASUAD, 2012) found it was more common for state programs to pay for care management as a service than as an administrative cost. With the shift to MLTSS, treating care management as an administrative function resolves concerns over conflict of interest, thus the dominant payment model may have shifted over time. The consequences of these changes in payment model have not been carefully examined.

The role of care management at the interface between LTSS and medical care remains an important unanswered question. One end of the spectrum is the fully integrated PACE model, which has demonstrated benefits for participants. Despite much growth over time, PACE only covers a small segment of people who use LTSS. For most people who use LTSS, there is little formal connection between their medical doctors and their care managers. But there is emerging evidence around other models of coordinating medical care and LTSS, for example the Independence at Home model in the general community and the Veterans Affairs (Valluru et al., 2019).

Future research should examine whether care management capacity should be based in primary or geriatric care, e.g., by hiring social workers, or if office staff can make referrals to community resources (e.g., Area Agencies on Aging [AAAs]). Current reimbursement models do not incentivize physician practices to develop this capacity, so innovations in risk-sharing may be needed to make it financially viable.

New Questions Begging for Attention

The LTSS environment has been changing dramatically in recent years. As of 2020, twenty-three states have implemented a form of Medicaid “Managed LTSS” in either regional or statewide programs (AdvancingStates, 2020). These programs use managed care organizations to deliver Medicaid-financed HCBS that was traditionally managed through the aging network. While state MLTSS programs vary in terms of the populations covered (older adult, people with physical or intellectual disabilities, those who are dually eligible for Medicare and Medicaid [duals]), they share the common feature of placing a private corporation at risk for the delivery and management of LTSS. To greater or lesser degree, they all disrupt the traditional role of care management.

In many states, care management was formerly provided by AAAs, state units on aging, or independent agencies. Under MLTSS, some or all of these functions are handled by managed care organizations. In the Pennsylvania Community HealthChoices program, AAAs conduct functional assessments to determine program eligibility, but employees or contractors to managed care organizations perform the care management.

Similarly, in the Tennessee CHOICES program, person-centered care coordination is provided by the MLTSS managed care plans (Kasten, Saucier, and Burwell, 2013). But in Massachusetts, under the Senior Care Options program, care coordination is conducted by independent agencies, and in Ohio AAAs provide care management in partnership with managed care plans.

This national experiment has proceeded without a strong evidence base. There are important questions that should be raised about the role of care managers who are employees of a corporation whose profitability depends upon controlling expenditures. State payments to MLTSS plans are structured to incentivize favoring HCBS, rather than nursing homes.

An overview of eight states’ experience with the Financial Alignment Demonstration found that enrollment was far lower than expected.

However, profit growth will depend upon either decreasing spending over time or increasing state payment rates. In such a system, care managers who advocate for individual participants may face resistance to “outlier” care plans that are more expensive than the average. In addition, the pressure to shift more of the burden of LTSS onto family or other funding sources will increase. Nursing home placement, while more expensive, reduces the strain on care management; it is unknown whether individual care managers face performance targets that reward community living. These questions will only grow in importance as these programs mature, and unit costs increase over time.

Incorporating care management into managed care organizations under MLTSS raises the question: Has the focus on conflict-free care management been effective? The concern was organizations providing both care management and direct care had an incentive to self-deal—by referring individuals to their own organization and potentially by inflating the amount of care needed. The implementation of rules regarding conflict-free care management led to divestment of functions into separate agencies; this was a key requirement in the Balancing Incentive Program (Wiener et al., 2015).

However, MLTSS plans are not precluded from having an ownership stake in LTSS providers; instead such integration is seen as a positive feature, because it facilitates data-sharing and multidisciplinary collaboration. On the surface this appears to be taking step back from the conflict-free policy. Aside from the evaluation of the Balanced Incentive Program, which included other major features, neither the move toward conflict-free case management or the potential for new conflicts under MLTSS has been studied.

Medicare Innovations Are Blurring Boundaries

Several innovations in the Medicare program have begun to blur the boundary between the traditional acute-care focus of the program and long-term needs of beneficiaries. These include Special Needs Plans (SNPs), financial alignment demonstrations, Accountable Care Organizations, and the introduction of Special Supplemental Benefits. These innovations explicitly or implicitly involve a care management function, however, there has been little direct investigation into the way care management is conducted distinct from overall program outcomes.

SNPs are Medicare Advantage plans targeted to special populations such as duals, people who are institutionalized, or those who have a “severe or disabling chronic condition” (Centers for Medicare & Medicaid Services [CMS], 2019b). A key feature of these plans is that there needs to be a “model of care” approved by CMS that includes care coordination for all members. Thus, each SNP has its own approach to organizing and delivering care coordination, including the degree to which it interacts with LTSS or community resources. While there has been some research on the impact of SNPs (Chen, Yang, and Gajadhar, 2018; Zhang and Diana, 2018), the effectiveness of care coordination itself has not been examined.

The Affordable Care Act (Grabowski, 2012) included an opportunity for states to improve care coordination for duals through integrated financing using managed care plans (Morley, Walsh, and Wilkin, 2013). The goal of the Financial Alignment Initiative was to determine whether aligning incentives between acute and long-term care would lead to savings. Early evidence from Massachusetts (Barry et al., 2015; Medicare Payment Advisory Commission, 2016) found that the “LTC Coordinator” role was challenging for the health plans to implement. The final model used external personnel (who were not part of the health plans or provider organizations) to reduce potential conflicts of interest.

Notably, that program excluded participants in Medicaid HCBS waiver programs; arguably the most challenging subgroup of people to serve. An overview of eight states’ experience found that enrollment in the program was far lower than expected (Grabowski et al., 2017). Recent evidence suggests that participants’ perceptions of care coordination are very positive (CMS, 2020). However, evidence from California suggests that care coordination was a weak aspect of the implementation in that state (Graham et al., 2018). Given the variation in program design and target populations across participating states, research is clearly needed on the role of case management for LTSS in this type of financing model.

Accountable Care Organizations (ACOs) are provider-based organizations that can “share savings” by controlling spending for a defined population (McWilliams et al., 2020). The growing body of evidence suggests that the primary pathway by which ACOs have reduced spending is by managing post-acute care (McWilliams et al., 2020; Trombley et al., 2019), in part by establishing preferred networks of nursing facilities (Kennedy et al., 2020). This body of research relies mainly on analysis of Medicare claims data, thus there is an important gap in our understanding of the tools being used to manage the hospital discharge process and the interface with long-term care providers. Because many people initiate LTSS after a hospitalization and post-acute care, the role of case management in ACOs is an important boundary that deserves further inquiry.

In 2018, CMS began allowing Medicare Advantage plans to offer “Special Supplemental Benefits” to enrollees with certain chronic illnesses (Coleman, 2019) to help address social determinants of health. These benefits include delivered meals, adult daycare, and non-medical transportation. While benefit availability is limited to a small fraction of Medicare enrollees, there is a need for research on the impact of addressing these needs and on the process by which Medicare Advantage plans use care management to identify members with needs and connect them with potentially beneficial services (Thomas et al., 2019).

There Are No Easy Answers

There are several challenges to generating knowledge about care coordination in the current environment. The policy experiments described above represent attempts to integrate Medicaid and Medicare, physical health services, and LTSS. The relatively high rate of all-cause and potentially avoidable hospitalization among HCBS users suggests there is an opportunity for improvement through better coordination across LTSS and physical health (Walsh et al., 2012; Wysocki, Kane, Dowd, et al., 2014; Wysocki, Kane, Golberstein, et al., 2014). In each of the integrated financing demonstrations, there is some form of care coordination within the domains of LTSS, physical health and behavioral health, as well as coordination across those domains. Even programs attempting a degree of financial integration typically “carve out” behavioral health. In addition, while participation in Medicaid managed care plans is mandatory, participation in Medicare managed care is not. Therefore, participants may opt-out of fully integrated products, or may choose to enroll in a Medicare plan that is from a Managed Care Organization (MCO) that competes with their Medicaid plan. Thus, coordination of care is made more difficult by the need to communicate and share protected participant data between private companies.

Another critical factor is that all of the demonstration projects use some form of risk sharing with privately owned MCOs. Under these arrangements, MCOs have substantial leeway to innovate and develop care coordination models. Thus, there is variation across different programs, across states, and even within states. Research on best practices also may be hindered to the extent that these models are proprietary by the MCOs.

While state and federal evaluations of these demonstration projects are underway, there is no concentrated effort to develop systematic knowledge about the underlying mechanism of care coordination. The fundamental challenge is that various demonstration projects do not systematically vary the key dimensions of care coordination. The projects are designed to test different approaches to financial integration, rather than methods of care coordination.

CMS-sponsored evaluations of financial alignment and MLTSS programs focus on programmatic outcomes for individuals (e.g., reducing preventable hospitalizations; increasing HCBS use), and society (e.g., reducing or slowing growth in costs) (Irvin et al., 2015). To the extent that the evaluations examine care coordination, the effort is descriptive. For example, the evaluation of care coordination in the California financial alignment demonstration was limited to telephone surveys of participants who rated their experience (Graham et al., 2018). Florida’s MLTSS program evaluation captured quantitative measures of care management (e.g., timeliness) and qualitative data on participant experience (e.g., responsiveness to requests) (Beitsch et al., 2016). However, the program’s design did not permit assessment of different care management models. So even though care management (or coordination) is a core component, the results of the various evaluations will not be informative about the best way to deliver it.

A research agenda that focused on the best way to deliver coordinated or integrated benefits should examine several key dimensions. First, should care management be internal to an integrated payor, or should it be conducted by an outside entity? Care coordination that is internal to an MCO managing public benefits may have different incentives than if they are located in an external agency.

External care coordinators may have interests that are more closely aligned with participants, and may be more likely to take an advocacy role. However, they will have relatively less control over care plan authorization.

‘Research on best practices also may be hindered to the extent that these models are proprietary by the MCOs.’

Second, what are the qualifications and training for care managers, especially if the responsibilities include chronic disease management as well as LTSS? Increasing the professionalization of care coordinators seems desirable, but such a move may disqualify a substantial amount of the current workforce. Certification efforts (National Committee for Quality Assurance, 2020) offer the promise of establishing a standardized skill set, but have not been externally validated.

Third, what is the ideal frequency, mode of contact, and caseload? How does that vary with the complexity of participant needs? This is a long-standing question, and one that has been upended by restrictions placed on in-person contact during the COVID-19 pandemic.

Fourth, if care management crosses organizational boundaries, what is the best way to share participant data, and how important is it to have real-time data? Health information exchanges are being used to improve coordination of care between Medicaid and Medicare plans, however this solution remains imperfect because it depends upon provider participation. The prospects for answering these questions will depend upon the level of interest from program sponsors (i.e., CMS and state Medicaid agencies), as high quality data will rely on implementation of a consistent model across a sufficiently large population.

Two recently developed survey instruments, the Consumer Assessment of Health Plans—Home and Community Based Services (CAHPS-HCBS) (Agency for Health Care Policy and Research, 2020b) and the National Core Indicators—Aging and Disabilities (NCI-AD) (National Core Indicators, 2020) are being used by states to measure the performance of LTSS systems. These new surveys are significant because they are potentially sensitive to variation in the quality of care management. The CAHPS-HCBS is available for free, and includes eight questions specifically about case management, and three questions about choice and control. These items can be used to construct three National Quality Forum endorsed measures (See Table 1; National Quality Forum, 2017).

Table 1. National Quality Forum (NQF) Care Management Measures
Measure Item
Global Rating
  • What number would you use to rate the help you get from your case manager? (0 to 10)
  • Would you recommend the case manager who helps you to your family and friends if they needed case management services?
Case Manager is Helpful
  • Able to contact this case manager when needed.
  • Case manager helped when asked for help with getting or fixing equipment.
  • Case manager helped when asked for help with getting other changes to services.
Choosing the Services that Matter to You
  • Person-centered service plan included all of the things that are important.
  • Staff knows what’s on the service plan, including the things that are important.


The Agency for Health Care Quality and Research is creating a benchmarking database that will allow states that adopt the CAHPS-HCBS to compare the performance of their programs to each other (Agency for Health Care Policy and Research, 2020a). The NCI–AD, which requires a licensing fee, has an extensive set of questions about service coordination. These items can be aggregated to generate statistics on program performance (See Table 2; National Core Indicators, 2020), and participating states can receive benchmarking reports. As states continue to innovate with new financing and delivery mechanisms, these two instruments have the potential to track improvement and change with reliable and valid measures.

Table 2. National Core Indicators—Aging and Disability Service Coordination Measure
  • Proportion of people who know whom to contact if they want to make changes to their services
  • Proportion of people who know whom to contact if they need help with services or have a complaint
  • Proportion of people who want help planning for future changes in their service needs
  • Proportion of people whose long-term care services meet all their current needs and goals
  • Proportion of people whose case manager/care coordinator talked to them about services that might help with their unmet needs and goals

Source: National Core Indicators. 2020. National Core Indicators–Aging and Disabilities (NCI–AD).


A key strength of both the NCI–AD and CAHPS–HCBS measures is that they prioritize participant self-reporting. However, these measures do not provide extensive detail on the actual experience of care management itself, or the interaction between care managers and care coordinators (where the latter are involved primarily with medical care rather than LTSS). Program sponsors that seek to evaluate critical details—for example, whether or not LTSS care managers communicate with primary care or specialist physicians—should examine other measures. An extensive review of care coordination measures (McDonald et al., 2014) identified eighty instruments, and mapped each one to key dimensions of the care management experience. Federal demonstration projects and state Medicaid agencies should consider more detailed data collection to address questions regarding the effectiveness of care management systems.

One final and fundamental observation about care management in integrated programs is that it is difficult to observe coordination between care managers and physicians. Approaches that use claims data have been developed to describe the consistency of medical care participants receive (Pollack et al., 2015). However, without examining narrative notes from care coordinators and medical providers, it may be impossible to assess whether effective communication and care planning is actually taking place. Qualitative approaches may be required to establish whether higher levels of financial integration between physical health and LTSS, for example, lead to better management of chronic disease for people who also use HCBS.


In recent years there have been a number of changes in the LTSS policy environment that affect the role of care managers. These changes highlight the importance of long-standing empirical questions about the organization and financing of case management itself, and raise new questions. Considerable energy and investment are being made into programs that attempt to align incentives for providing high quality medical and social services for people with disabilities. These changes have had an impact on the traditional organization of care management as part of the aging network (Polivka and Luo, 2019). As care management is incorporated, and in some cases reinvented, it is critical to continue to ask and answer questions about how care management can help achieve the most important goal of supporting program participants’ independence and self-determination.

Howard B. Degenholtz, PhD, is an associate professor in Health Policy and Management, and faculty at the Center for Bioethics and Health Law, at the University of Pittsburgh Graduate School of Public Health in Pennsylvania.


AdvancingStates. 2020. "MLTSS Map." Retrieved January 28, 2021.

Agency for Health Care Policy and Research. 2020a. "CAHPS Databases." Retrieved February 2, 2021.

Agency for Health Care Policy and Research. 2020b. "CAHPS Surveys and Tools to Advance Patient-Centered Care." Retrieved February 2, 2021.

Applebaum, R. 2012. "Channeling: What We Learned, What We Didn't—and What It All Means Twenty-Five-Years Later." Generations 36(1): 21.

Barry, C., et al. 2015. Early Insights from One Care: Massachusetts’ Demonstration to Integrate Care and Align Financing for Dual Eligible BeneficiariesRetrieved February 2, 2021.

Beitsch, L. M., et al. 2016. Independent Assessment of the Florida Statewide Medicaid Managed Care Long-term Care Program. Retrieved February 2, 2021.

Centers for Medicare & Medicaid Services (CMS). 2019a. Application for a §1915(c) Home and Community-Based Waiver: Instructions, Technical Guide and Review Criteria. Retrieved January 28, 2021.

CMS. 2019b. "Special Needs Plans: What is a Special Needs Plan?" Retrieved January 28, 2021.

CMS. 2020. Care Coordination Snapshot for the Capitated ModelRetrieved February 2, 2021.

Chen, B. K., Yang, Y. T., and Gajadhar, R. 2018. "Early Evidence from South Carolina's Medicare-Medicaid Dual-eligible Financial Alignment Initiative: An Observational Study to Understand Who Enrolled, and Whether the Program Improved Health?" BMC Health Services Research 18(1): 913. doi:10.1186/s12913-018-3721-6.

Coleman, K. 2019. Implementing Supplemental Benefits for Chronically Ill Enrollees. Retrieved February 2, 2021.

Eiken, S., et al. 2018. Medicaid Expenditures for Long-Term Services and Supports in FY 2016. Retrieved January 28, 2021.

Grabowski, D. C. 2012. "Care Coordination for Dually Eligible Medicare-Medicaid Beneficiaries Under the Affordable Care Act." Journal of Aging & Social Policy 24(2): 221–32. doi:10.1080/08959420.2012.659113.

Grabowski, D. C., et al. 2017. "Passive Enrollment of Dual-Eligible Beneficiaries Into Medicare and Medicaid Managed Care Has Not Met Expectations." Health Affairs (Millwood) 36(5): 846–54. doi:10.1377/hlthaff.2016.1082.

Graham, C. L., et al. 2018. "Beneficiaries Respond To California's Program to Integrate Medicare, Medicaid, and Long-Term Services." Health Affairs (Millwood) 37(9): 1432–41. doi:10.1377/hlthaff.2018.0452.

Irvin, C., et al. 2015. Medicaid 1115 Demonstration Evaluation Design Plan. Retrieved February 2, 2021.

Kane, R. A. 1988. "The Noblest Experiment of Them All: Learning from the National Long-Term Care Channeling Demonstration." Health Services Research 22(1): 189–98.

Kasten, J., Saucier, P., and Burwell, B. 2013. How Have Long-Term Services And Supports Providers Fared In The Transition To Medicaid Managed Care? A Study Of Three StatesRetrieved February 2, 2021.

Kennedy, G., et al. 2020. "Accountable Care Organizations and Post-Acute Care: A Focus on Preferred SNF Networks." Medical Care Research and Review 77(4): 312–23. doi:10.1177/1077558718781117.

Lewis, E., et al. 2018. The Growth of Managed Long-Term Services and Supports Programs: 2017 Update. https://www.medicaid.gov/media/3406. Retrieved February 2, 2021.

McDonald, K. M., et al. 2014. Care Coordination Measures Atlas Version 4. (Prepared by Stanford University under subcontract to American Institutes for Research on Contract No. HHSA290-2010-00005I. AHRQ Publication No. 14-0037- EF.). Rockville, MD:

McWilliams, W. J., et al. 2020. "Savings or Selection? Initial Spending Reductions in the Medicare Shared Savings Program and Considerations for Reform." Milbank Quarterly 98(3): 847–907. doi:10.1111/1468-0009.12468.

Medicare Payment Advisory Commission. 2016. Report to the Congress: Medicare and Health Care Delivery System. Retrieved February 2, 2021.

Morley, M. A., Walsh, E. G., and Wilkin, J. C. 2013. "Financing Integrated Care in the Financial Alignment Demonstration." Generations 37(2): 24–9.

National Association of States United for Aging and Disability. 2012. Results of February 2012 Survey on Medicaid Funded Long-Term Services and Supports: Case ManagementRetrieved January 28, 2021.

National Committee for Quality Assurance. 2020. "Long-Term Services and Supports Distinction for Health Plans." Retrieved February 2, 2021.

National Core Indicators. 2020. "National Core Indicators—Aging and Disabilities (NCI-AD)." Retrieved February 2, 2021.

National Quality Forum. 2017. "CAHPS Home- and Community-Based Services Measures." Retrieved March 8, 2020.

Polivka, L., and Luo, B. 2019. "Neoliberal Long-Term Care: From Community to Corporate Control." The Gerontologist 59(2): 222–9. doi:10.1093/geront/gnx139.

Pollack, C. E., et al. 2015. "Patient Sharing and Quality of Care: Measuring Outcomes of Care Coordination Using Claims Data." Medical Care 53(4): 317–23. doi:10.1097/MLR.0000000000000319.

Thomas, K. S., et al. 2019. "Perspectives of Medicare Advantage Plan Representatives on Addressing Social Determinants of Health in Response to the CHRONIC Care Act." JAMA Network Open 2(7): e196923. doi:10.1001/jamanetworkopen.2019.6923.

Trombley, M. J., et al. 2019. "Early Effects of an Accountable Care Organization Model for Underserved Areas." New England Journal of Medicine 381(6): 543–51. doi:10.1056/NEJMsa1816660.

Valluru, G., et al. 2019. "Integrated Home- and Community-Based Services Improve Community Survival Among Independence at Home Medicare Beneficiaries Without Increasing Medicaid Costs." Journal of the American Geriatrics Society 67(7): 1495–501. doi:10.1111/jgs.15968.

Walsh, E. G., et al. 2012. "Potentially Avoidable Hospitalizations of Dually Eligible Medicare and Medicaid Beneficiaries from Nursing Facility and Home- and Community-Based Services Waiver Programs." Journal of the American Geriatrics Society 60(5): 821–9. doi:10.1111/j.1532-5415.2012.03920.

Wiener, J., et al. 2015. Descriptive Overview And Summary Of Balancing Incentive Program Participating States At BaselineRetrieved February 2, 2021.

Wysocki, A., Kane, R. L., Dowd, B., et al. 2014. "Hospitalization of Elderly Medicaid Long-term Care Users Who Transition From Nursing Homes." Journal of the American Geriatrics Society 62(1): 71–8. doi:10.1111/jgs.12614.

Wysocki, A., Kane, R. L., Golberstein, E., et al. 2014. "The Association Between Long-term Care Setting and Potentially Preventable Hospitalizations Among Older Dual Eligibles." Health Services Research 49(3): 778–97. doi:10.1111/1475-6773.12168.

Zhang, Y., and Diana, M. L. 2018. "Effects of Early Dual-Eligible Special Needs Plans on Health Expenditure." Health Services Research 53(4): 2165–84. doi:10.1111/1475-6773.12778.