Managing Chronic Illness

Abstract

Models of care coordination can significantly improve health outcomes for older adults with chronic illnesses if they can engage participants. The purpose of this study was to examine the impact of nursing contact on the rate of participants’ voluntary disenrollment from a care coordination program. In this retrospective cohort study using administrative data for 1,524 participants in the Health Quality Partners Medicare Care Coordination Demonstration Program, the rate of voluntary disenrollment was approximately 11%. A lower risk of voluntary disenrollment was associated with a greater proportion of in-person (vs. telephonic) nursing contact (Hazard Ratio [HR] 0.137, confidence interval [CI] [0.050, 0.376]). A higher risk of voluntary disenrollment was associated with lower continuity of nurses who provided care (HR 1.964, CI [1.724, 2.238]). Findings suggest that in-person nursing contact and care continuity may enhance enrollment of chronically ill older adults and, ultimately, the overall health and well-being of this population

Keywords

care coordination participant enrollment nursing contact

Background

Nearly 50% of Americans live with chronic illnesses such as hypertension, disorders of lipid metabolism, non-traumatic joint diseases, and upper respiratory diseases (Anderson, 2010). Among adults above the age of 65, more than 70% live with two or more chronic illnesses; moreover, older adults with chronic illnesses account for a substantial portion in all Medicare expenditures (Anderson, 2010; Thorpe & Howard, 2006). Between 1987 and 2002, for example, 16% of total Medicare spending was attributed to the increased costs of treating three significant health problems—hypertension, diabetes, and hyperlipidemia (Thorpe & Howard, 2006).

Care coordination, is a benefit for some fee-for-service Medicare beneficiaries (Medicare Program; Revisions to Payment Policies under the Physician Fee Schedule; Clinical Laboratory Fee Schedule & Other Revisions to Part B for Current Year 2014 [Final Rule], 2013); it is also a promising strategy to help older adults manage chronic illnesses and improve their health outcomes over time (Hong, Siegel, & Ferris, 2014). Defined as “the deliberate organization of patient care activities between two or more participants . . . involved in a patient’s care to facilitate the appropriate delivery of health care services” (Agency for Healthcare Research and Quality, 2007), care coordination is associated with improved patient and organizational outcomes, including the rate of mortality and hospital readmission, the quality of patient care and satisfaction, and health care costs (Brown, Peikes, Peterson, Schore, & Razafindrakoto, 2012; Centers for Medicare and Medicaid Services, 2001; Coburn, Marcantonio, Lazansky, Keller, & Davis, 2012; Marek et al., 2014; Peikes, Peterson, Brown, Graff, & Lynch, 2012; Rantz et al., 2014; Schore, Peikes, Peterson, Gerolamo, & Brown, 2011; Verhaegh et al., 2014).

Care coordination is frequently provided by nurses (as well as social workers and other health professionals) who interact with program participants over time in a range of settings, such as primary care practices, insurance companies, health care systems, and regional care management settings (Hong et al., 2014). Available evidence indicates that nursing contact, particularly in-person versus telephonic nursing contact, is associated with more effective delivery of care coordination services (Brown et al., 2012; Coburn et al., 2012; Counsell et al., 2007; Marek et al., 2014; Peikes, Chen, Schore, & Brown, 2009; Peikes et al., 2012); however, studies have not described many other characteristics of nursing contact that may affect the delivery of effective care coordination (e.g., the continuity of nurses who provide care, the frequency of nursing contacts, and modes of nursing contact such as individual vs. group contacts).

Although care coordination models vary greatly in both program design and effectiveness, a widely held belief is that sustained participant engagement is a key prerequisite to program effectiveness and overall health and well-being of its participants (Brown et al., 2012; Peikes et al., 2009; Peikes et al., 2012; Rantz et al., 2014). Health Quality Partners (HQP), the longest running clinical partner in the Centers for Medicare and Medicaid Services (CMS; 2001) Medicare Coordinated Care Demonstration (MCCD), is a care coordination program with a strong emphasis on engaging older adults and their family caregivers as participants in care of chronic conditions (Brown et al., 2012; Hong et al., 2014); thus, new data that describe HQP strategies for coordinating care with older adults are a potentially rich source for evidence to guide practice and research. HQP is a non-profit quality improvement organization that has provided “advanced preventive care,” a form of community-based care coordination services, to approximately 2,380 Medicare beneficiaries with chronic illness. In HQP, experienced and extensively trained registered nurses assess participants’ physical, functional, cognitive, behavioral, and social needs, and provide a range of interventions including (a) person-centered problem solving to support medication management and treatment adherence, (b) home and group exercise programs for fall prevention, (c) collaboration with primary and specialty care providers to optimize treatment for common chronic conditions, and (d) structured group interventions for weight loss, and curriculum-based disease-specific education to promote self-management. To provide these services, HQP nurses interact with program participants throughout the intervention via a series of in-person and telephonic, individual and group encounters (Coburn et al., 2012; Schore et al., 2011). Studies reveal that older adults who participated in the HQP program, compared with usual care participants, experienced a reduction in all-cause mortality (Coburn et al., 2012; Schore et al., 2011); moreover, among HQP participants, those with the greatest risk of poor health outcomes also experienced fewer 30-day hospital readmissions and lower Medicare costs (Brown et al., 2012; for additional information about the HQP Program, see http://www.hqp.org).

Although recent findings suggest the value of care coordination for older adults, such as participants in HQP, prior studies have not determined how nursing contact influences older adults’ decisions to continue participating or to voluntarily disenroll from care coordination programs. Significantly, prior studies have not examined the relationship between nursing contact and patient enrollment/disenrollment in care coordination programs. Understanding the predictors of disenrollment from a care coordination program is critical to identify strategies that care coordinators may use to increase the number of participants who are exposed to the protective benefits of effective services. Thus, the primary aim of this study was to identify characteristics of nursing contact in a care coordination program associated with program disenrollment. Prior studies of voluntary disenrollment from primary care indicate that patients often disenroll due to dissatisfaction with a health-related encounter (Nagraj et al., 2013) or a poor relationship with a health care provider (Safran, Montgomery, Chang, Murphy, & Rogers, 2001). Because evidence also suggests that nursing contact during the first 6 months of enrollment contributes to the quality of health care services (O’Connor, Bowles, et al., 2014; O’Connor, Hanlon, & Bowles, 2014; Rogers, Perlic, & Madigan, 2007), a second aim was to define the characteristics of nursing contact, within the first 6 months of enrollment, which predict patients’ disenrollment from a care coordination program.

Method

This retrospective cohort study used administrative data from HQP to measure the influence of nursing contact on participant disenrollment in a CMS care coordination program. The study was approved by the Institutional Review Board at the University of Pennsylvania.

Data

Data were collected by HQP care coordination nurses as a part of the CMS MCCD. To facilitate the analysis, investigators at HQP (KC and SM) created a file of data for all Medicare fee-for-service beneficiaries who were enrolled in the HQP Program between April 1, 2002, and October 23, 2013. An identification number, created for each HQP participant, was used to link participant data of three types: participant enrollment status, characteristics of nursing contact, and HQP participant characteristics (the data are described below). Data were stored at HQP and were accessed for statistical analysis by research team members.

Sample

The sample of HQP participants consisted of Medicare beneficiaries (N = 1,524), who resided in Eastern Pennsylvania, were aged 65 years or older, were treated by a primary care physician who agreed to participate in the CMS MCCD study, and were diagnosed with one of six conditions—heart failure, coronary artery disease, asthma, diabetes, hypertension, or hyperlipidemia (Coburn et al., 2012). Potential HQP participants were excluded if they were diagnosed with dementia, end-stage renal disease, schizophrenia, or active cancer; had a life expectancy less than 6 months; or were current or anticipated nursing home residents (Coburn et al., 2012).

Variables and Measures

The study relied on a time to event analysis; thus, it is necessary to define the event of interest as well as the time to that event. The event of interest (e.g., dependent variable) was voluntary disenrollment, defined as a member’s decision to formally terminate his or her enrollment in HQP services. Voluntary disenrollment, as opposed to involuntary disenrollment, is an active choice made by the HQP participant. If a participant died, left the program because of geographic relocation, or moved to a nursing home, it was not considered voluntary disenrollment. Time to the event (e.g., time to voluntary disenrollment) was computed from the date of program enrollment and is operationally defined below.

The independent variable of interest was nursing contact among HQP nurse coordinators and participants between April 1, 2002, and October 23, 2013. Nursing contact was operationalized using seven metrics including (a) the count of all nursing contacts, (b) contact type (e.g., the proportion of in-person contact vs. telephone contact), (c) contact setting (e.g., the proportion of individual contact vs. group contact), (d) contact initiation (e.g., the proportion initiated by participant, family, or nurse), and (e) count of nurses per participant (e.g., the number of different individual nurses patients interacted with over the course of the study). To account for variability in the length of time that participants were enrolled in the HQP Program, nursing contact was also operationalized as contact intensity (e.g., the average number of nursing contacts per participant per year) and nurse continuity (e.g., average number of nurses per participant per year). Finally, for descriptive purposes only, nursing contact was also classified in four categories (e.g., the proportion of nursing activities in participant contacts that included assessing, monitoring, educating, and referring).

Other independent variables were participants’ age, gender, and clinical and functional characteristics, including (a) primary diagnosis (e.g., heart failure, coronary artery disease, asthma, diabetes, hypertension, or hyperlipidemia), (b) health rating score (e.g., participants’ self-report of perceived health as excellent, good, fair, or poor), (c) documentation of a fall in the year before enrollment, (d) a yes/no item to indicate any difficulty with basic activities of daily living (BADL), (e) a yes/no item to indicate any difficulty with instrumental activities of daily living (IADL), and (f) a yes/no item to indicate whether any equipment was used to support mobility (e.g., canes, walkers, and wheelchairs).

Analysis

Data were analyzed using SAS Version 9.2 (SAS Institute, Inc., Cary, North Carolina). For all participants enrolled in the HQP program during this study period, we described voluntary disenrollment, nursing contact, and other independent variables using frequencies and percentages for categorical measures and means/medians along with measures of dispersion (SD, Interquartile range [IQR]) for continuous measures. Descriptions were aggregated to the participant level and stratified where necessary.

For the primary aim, we estimated and visually described disenrollment rates using Kaplan–Meier methodology and measured time as number of months enrolled from date of program enrollment to either date of voluntary disenrollment, last disenrollment date (e.g., date of death), or 10/23/2013 for participants still active (date of data collection), whichever occurred first. We generated multivariable Cox proportional hazards regression models to examine the relationships between participant characteristics and attributes of nursing contact with voluntary disenrollment from HQP. First, simple Cox proportional regression models were used to examine the bivariate relationship between individual nursing contact variables and participant characteristics with voluntary disenrollment. Subsequently, a stepwise approach was used for the multivariate model using any predictor variables found significant in the bivariate models at the 0.20 level. Because of collinearity between phone versus in-person contact and group versus individual contact, phone versus in-person contact was chosen for consideration in final modeling, and therefore, contact type instead of contact setting was used. Because participants were enrolled for varying lengths of time in the HQP program, we described (a) intensity of nursing contact by calculating the average number of nursing contacts per participant per year and (b) continuity of nursing contact by calculating the average number of nurses per participant per year. The derived final multivariate model included the proportion of in-person contacts, contact intensity, and nurse continuity. The proportional hazards model assumption was tested visually using log–log survival plots and inferentially using scaled Schoenfeld residuals. There was no indication of model assumption violation, and thus, the semi-parametric Cox proportional hazards model was used for multivariable analysis.

For our secondary aim, we examined a subset of all participants having greater than 6 months (181 days) enrollment in HQP. Using this subgroup, we again estimated and visually described voluntary disenrollment rates using Kaplan–Meier and Cox Proportional Hazards methodology and measured time to voluntary disenrollment in the same manner as the primary aim but starting at the 181st day of enrollment. The same predictors applied to achieve the primary aim were used but calculated for just the first 6 months of enrollment.

Results

The study sample included 1,524 participants who were enrolled in the HQP program between April 1, 2002, and October 23, 2013. The final study sample size was 1,283 participants after excluding 241 participants (15.8%) with incomplete data (e.g., participants without a baseline assessment, without accounts for fee-for-service Medicare, and without an enrollment date). Participants’ demographic, clinical, functional, and disenrollment characteristics are summarized in Table 1. On average, participants were 75 years of age and predominantly female (59%). Participants were most commonly diagnosed with hypertension (43%), coronary artery disease (18%), and diabetes mellitus (17%). Approximately 77% reported their general health as excellent or good, whereas 20% reported it as fair and 3% as poor. Among the 1,283 participants with complete data, 138 participants (approximately 11%) voluntarily disenrolled from the HQP program. Figure 1 represents the Kaplan–Meier survival curve for voluntary disenrollment time.

Table 1.

Participant Characteristics (N = 1,283 Participants).

Variable	Category	M	Median	SD	IQR
Age at enrollment	Years	75.27	75	7	(69.00, 80.00)
		n		%
Voluntary disenrollment	Yes	138		10.76
Gender	Female	756		58.92
Primary diagnosis	Diabetes mellitus	185		17.02
	Hyperlipidemia	130		11.96
	Hypertension	466		42.87
	Coronary artery disease	199		18.31
	Heart failure	59		5.43
	Asthma	48		4.42
	Null	196		^a
Health rating score	Excellent	223		17.38
	Good	762		59.39
	Fair	260		20.27
	Poor	38		2.96
Fall in the past year	Yes	366		28.53
Difficulty with BADLs	Yes	314		24.47
Difficulty with IADLs	Yes	508		39.59
Mobility equipment	Yes	439		34.22

Note. BADL = basic activities of daily living; IADL = instrumental activities of daily living; IQR = interquartile range

Nulls were not used in sample number to calculate primary diagnosis percentages.

Figure 1.

Aim 1: Kaplan–Meier curve for time to voluntary disenrollment.

In Table 2, the characteristics of nursing contact in the HQP are described. Over the duration of participants’ enrollment in HQP (M = 60.2 months), nurses contacted participants an average of 94 times (M = approximately 21 contacts per year). The majority of nursing contacts were in-person (53%), one to one encounters (85%), and initiated by nurses (93%). On average, 1.58 different HQP nurses contacted participants in each year of their enrollment.

Table 2.

Nursing Contact (N = 1,283 Participants).

Variable	Category	M	Median	SD	IQR
Contacts	Count of all contacts	94.44	75	82.96	(36.00, 129.00)
Contact type	Proportion of in-person contact	0.53	0.52	0.23	(0.35, 0.71)
	Proportion of telephone contact	0.47	0.48	0.23	(0.29, 0.65)
Contact setting	Proportion of individual contact	0.85	0.97	0.23	(0.79, 1.00)
	Proportion of group contact	0.15	0.03	0.23	(0.00, 0.21)
Contact initiation	Proportion initiated by participant	0.06	0.03	0.08	(0.00, 0.09)
	Proportion initiated by family	0.01	0	0.04	(0.00, 0.01)
	Proportion initiated by nurse	0.93	0.95	0.1	(0.89, 0.99)
Count of nurses/participant	Number of nurses	3.78	3	2.54	(2.00, 5.00)
Contact intensity (Count of contacts/ days in program) × 365	Average number of contacts per year	20.97	16.74	13.98	(12.43, 24.98)
Nurse continuity (Count of nurses/days in program) × 365	Average number of nurses per year	1.58	0.8	4.78	(0.51, 1.29)
Identify proportion	Proportion of nursing time focused on identification of referrals	0.01	0	0.03	(0.00, 0.01)
Education proportion	Proportion of nursing time focused on education about treatments and wellness approaches	0.83	0.86	0.16	(0.78, 0.92)
Monitor proportion	Proportion of nursing time focused on monitoring changes in health and support	0.86	0.89	0.15	(0.83, 0.94)
Assessment proportion	Proportion of nursing time focused on assessment of participant needs	0.56	0.58	0.26	(0.39, 0.77)

As noted earlier, the primary aim of the study was to identify the characteristics of nursing contact and participant characteristics that predict voluntary disenrollment from the care coordination program. After bivariate Cox Proportional Hazard modeling, tests of multicollinearity, and stepwise multivariate approaches, three variables remained for modeling in the final multivariate analysis, including the proportion of in-person contacts, the average number of contacts per year, and the continuity of nursing contact (Table 3). Findings from the multivariate analysis indicated (a) a higher proportion of in-person (vs. telephonic) nursing contacts was associated with lower odds of voluntary disenrollment (HR = 0.137, confidence interval [CI] [0.050, 0.376]), and (b) a higher count of nursing contacts per year was associated with lower odds of voluntary disenrollment (HR = 0.936, CI [0.907, 0.966]). In contrast, lower nursing continuity (e.g., a higher number of different nurses providing care coordination over time) was associated with greater odds of voluntary disenrollment (HR = 1.964, CI [1.724, 2.238]).

Table 3.

Predictors of Voluntary Disenrollment From the HQP Program (N = 1,283 Participants).

Variable	Hazard ratio	95% Hazard ratio confidence limits		p value
In-person contact proportion	0.137	0.05	0.376	.0001
Average number of contacts per year	0.936	0.907	0.966	<.0001
Average number of nurses per year	1.964	1.724	2.238	<.0001

Note. HQP = Health Quality Partners.

The secondary aim was to identify characteristics of nursing contact within the first 6 months of enrollment and participant characteristics that predict voluntary disenrollment from the care coordination program. The participants’ demographic, clinical, functional, and disenrollment characteristics for Aim 2 are similar to the characteristics for the primary aim and, thus, are not shown. After bivariate Cox Proportional Hazard modeling and stepwise multivariate approaches, two variables remained for modeling in the multivariate analysis: the average number of nursing contacts per year and participant age at the time of enrollment in the program (Table 4). Findings from the multivariate analysis indicated that a higher count of nursing contacts in the first 6 months of participants’ enrollment in the program was associated with lower hazard of voluntary disenrollment (HR = 0.973, CI [0.956, 0.991]). In addition, a higher age at enrollment was associated with lower hazard of voluntary disenrollment (HR = 0.960, CI [0.930, 0.990]). The Kaplan–Meier curve for Aim 2 is similar to the curve for the primary aim and, thus, is not shown.

Table 4.

Predictors of Voluntary Disenrollment in First 6 Months of Enrollment (N = 1,283 Participants).

Variable	Hazard ratio	95% Hazard ratio confidence limits		p value
Average number of contacts per year	0.973	0.956	0.991	.004
Age on enrollment	0.96	0.93	0.99	.0104

Discussion

Prior studies indicate that care coordination provided by nurses is an effective strategy for reducing costs and improving participant outcomes for older adults with chronic illnesses vulnerable for poor outcomes (Brown et al., 2012; Hong et al., 2014). In this study, we examined nursing activities in one care coordination program—HQP—and learned that nursing contact with participants also predicted patterns in their enrollment in the program over time. This finding is significant because it suggests that nursing contact with older adults is a potentially modifiable attribute of care coordination programs and may be useful for increasing patient engagement in care coordination programs, ultimately improving health outcomes among chronically ill older adults living in the community.

In this study of 1,283 HQP participants, we found that 138 participants (approximately 11%) voluntarily disenrolled from the program. To our knowledge, prior studies have not described the rate of voluntary disenrollment from care coordination programs; thus, benchmarks were not available for interpreting the significance of the rate of disenrollment. Moreover, our findings do not describe the reasons that participants chose to voluntarily disenroll from HQP; however, our finding that approximately 11% of participants voluntarily disenrolled suggests an opportunity for research and quality improvement interventions designed to explain and prevent avoidable disenrollment from the care coordination program.

To this end, we identified two characteristics of nursing contact that predicted voluntary disenrollment and that suggest a starting point for research and quality improvement interventions. Our finding that a greater hazard of voluntary disenrollment was associated with fewer in-person (vs. telephonic) nursing contact indicates that in-person nursing contact may be an important strategy for sustaining enrollment in a care coordination program; this finding warrants confirmation in studies designed to evaluate care in multiple coordination programs. Prior studies also indicate that supportive relationships between older adults and community-based nurse case managers were valued components of care management interventions and were associated with improved outcomes such as participant satisfaction and the rate of hospital readmissions (Bradway et al., 2012; Naylor, 2002; Naylor, Aiken, Kurtzman, Olds, & Hirschman, 2011). As noted by Parry, Kramer, and Coleman (2006),

Competence alone may be insufficient to engage participants in the self-management aspects of the intervention. Rather, it is possible that the combination of competence and the perception of a caring relationship may have been responsible for eliciting the engagement of patients.

In this study, we found that (a) a greater hazard of voluntary disenrollment was associated with fewer nursing contacts with participants in HQP per year, and (b) a higher hazard of voluntary disenrollment was also associated with a lower continuity of nurses that provided care coordination with participants. These findings suggest that the type, frequency, and continuity of nursing contact predict sustained enrollment and, thereby, increase the potential benefits of effective programs.

Limited prior evidence suggests that “frontloading” or intensive contact with older adults early in their enrollment in health care services increases the quality of participant outcomes (O’Connor, Hanlon, & Bowles, 2014; Rogers et al., 2007). We found a modest relationship between more frequent nursing contact with participants in the first 6 months of their enrollment in HQP and a lower hazard of voluntary disenrollment. This finding suggests that frontloading of nursing contact during the early months of participant enrollment may also be a useful strategy for optimizing participant outcomes in a care coordination program. Finally, a lower age on enrollment also predicted a marginally greater hazard of voluntary disenrollment. This finding suggested that younger participants were somewhat more likely to voluntarily disenroll within the first 6 months compared with older care coordination participants. This finding warrants additional study as prior research has not explored the acceptance of care coordination services according to age among older adults.

Due to limitations, study findings must be interpreted with caution. First, the study was conducted using secondary data. Available data did not include potentially useful predictors of disenrollment, such as (a) the reasons that participants chose to disenroll, (b) strategies that the HQP nurses may have used to minimize disenrollment, and (c) participant-level data such as comorbidity, level of education, race, ethnicity, participation in Medicaid, and satisfaction with the care coordination program. Second, participant-level data included only the description of HQP participants at baseline. Thus, time-varying data, such as BADL, diagnosis, and support at home, were not available to describe the potential impact of changes in function, support, and health on decisions to disenroll from the program. Third, the study was conducted with data from older adults enrolled in one care coordination program, which limits the generalizability of findings. Fourth, data were missing or incomplete for 241 HQP participants (15.8%); because these participants were dropped from the analysis, it is not known how more complete data may have influenced study findings. And finally, the findings describe associations between nursing contact and participant enrollment but the results of the analysis do not explain the causal factors that may underlie the relationship.

Despite these limitations, this exploratory retrospective cohort study provides important findings to guide subsequent studies of the relationships between nursing contact and enrollment in care coordination services. Research that builds on this work should (a) describe participant satisfaction with nursing contact and their reasons for disenrolling from care coordination programs; (b) examine the relationship between variations in nursing contact and participant outcomes, such as enrollment and rehospitalization rates, in multiple care coordination programs; (c) compare outcomes among older adults who do and do not disenroll from care coordination services; (d) examine nursing contact, enrollment in care coordination and hospice, and palliative care services as older adults and their families confront complex decision making at the end of life; and (e) provide in-depth qualitative analysis of patient experiences in care coordination programs and describe the reasons patients continue engagement in these programs.

Our findings also have implications for refining and developing care coordination practices. Perhaps most importantly, findings underscore the centrality of relationships and continuity in care designed to engage and assist older adults in their self-management of chronic illness. Although our findings suggest that relationships between nurses and patients in the HQP Program were associated with lower disenrollment, the findings do not explain how the nurses interacted with patients and whether styles of interaction between individual nurses and patients contributed to effective relationships and patient care. An important next step in research is to examine individual nursing interactions with patients or family caregivers and identify strategies to develop and maintain therapeutic relationships, and deliver basic elements of care coordination services over time. Our findings also did not account for environmental and community resources that contribute to the effectiveness of care for older adults with chronic illnesses (Martin, Schoster, Woodard, & Callahan, 2012); thus, future studies are also warranted to examine how providers of care coordination integrate their work with other community or environmental resources, such as recreational centers and pharmacies.

Conclusion

Care coordination is a promising strategy to support older adults as they manage chronic illness and to reduce medical costs. Findings indicated that approximately 11% of participants disenrolled from the care coordination program and that in-person and continuity of nursing contact contributed to patients’ decisions to remain enrolled in an effective care coordination program. Further research is needed to determine the reasons that older adults disenroll from care coordination and to develop a workforce with the skill to engage participants and sustain their participation in beneficial programs.

Footnotes

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The investigators retained full independence in the conduct of this research.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was funded by a pilot grant from the NewCourtland Center for Transitions and Health at the University of Pennsylvania School of Nursing. Mark Toles was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, Grant Numbers 1KL2TR001109 and 1UL1TR001111.

Author Biographies

Mark Toles, PhD, RN, is an assistant professor in the School of Nursing at the University of North Carolina at Chapel Hill. His research is focused on improving transitions in care for older adults and their family caregivers. He is a researcher in the NewCourtland Center for Transitions and Health.

Helene Moriarty, PhD, RN is a professor, Diane and Robert Moritz, Jr. Endowed Chair in Nursing Research at Villanova University College of Nursing. In addition, she is a nurse researcher at the Philadelphia VA Medical Center and a member of the NewCourtland Center for Transitions and Health at Penn Nursing. Her research focuses on the health needs of veterans and their families.

Ken Coburn, MD, MPH, is a president, CEO, and medical director of Health Quality Partners (HQP), a non-profit health care quality research and development organization based in Doylestown, Pennsylvania, dedicated to improving health outcomes through health care delivery redesign. His work has centered on designing, testing, and scaling more effective preventive care systems for vulnerable populations.

Sherry Marcantonio, MSW, ACSW, is a senior vice president and chief program architect for HQP’s community-based nurse care management program. She served as the program director for HQP’s Medicare Coordinated Care Demonstration. Her areas of interests center on ensuring high reliability in program and team performance.

Alexandra Hanlon, a biostatistician and research professor in the School of Nursing at the University of Pennsylvania, provides consultation on study design and statistical support to faculty and doctoral students. She received her PhD in statistics from Temple University. As a collaborative research team member, she has received millions in research funding and has co-authored more than 260 peer-reviewed journal articles.

Elizabeth Mauer is an intern in the Biostatistics Consulting Unit of the University of Pennsylvania School of Nursing. She was awarded a BS in neuroscience and minor in philosophy in May of 2011 from the University of Scranton and was awarded an MS in statistics in December, 2014, from Temple University.

Paige Fisher is a PhD student at Louisiana State University Health Sciences Center, where she is studying biostatistics. She has obtained her masters of science in biostatistics at Drexel University in June 2014. She has worked closely with Dr. Alexandra Hanlon at the University of Pennsylvania as a research intern in the Biostatistics Consulting Unit of the Office of Nursing Research.

Melissa O’Connor, PhD, MBA, RN, COS-C, is an assistant professor at Villanova University’s College of Nursing and a National Hartford Center for Gerontological Nursing Excellence Claire Fagin Fellow. Her program of research concerns older adults living in the community with a specialty in skilled home health. She is currently developing decision support methodologies to support home health clinicians.

Connie Ulrich is a nurse bioethicist at the University of Pennsylvania School of Nursing with a secondary appointment in the School of Medicine, Department of Medical Ethics and Health Policy. Her bioethics program of research has focused on ethical issues in clinical practice and research.

Mary D. Naylor, for the past two decades, has led an interdisciplinary program of research designed to improve the quality of care, decrease unnecessary hospitalizations, and reduce health care costs for vulnerable community-based elders. She was elected to the Institute of Medicine in 2005 and was appointed to the Medicare Payment Advisory Commission in 2010. She also serves on the RAND Corporation Health Board and the National Quality Forum Board of Directors.

References

Agency for Healthcare Research and Quality. (2007). Care coordination, quality improvement. Retrieved from http://www.ahrq.gov/research/findings/evidence-based-reports/caregaptp.html

Anderson

(2010). Chronic care: Making the case for ongoing care. Retrieved from http://www.rwjf.org/content/rwjf/en/research-publications/find-rwjf-research/2010/01/chronic-care.html

Bradway

Trotta

Bixby

M. B.

McPartland

Wollman

M. C.

Kapustka

. . . Naylor

M. D.

(2012). A qualitative analysis of an advanced practice nurse-directed transitional care model intervention. The Gerontologist, 52, 394-407. doi:10.1093/geront/gnr078

Brown

R. S.

Peikes

Peterson

Schore

Razafindrakoto

C. M.

(2012). Six features of Medicare coordinated care demonstration programs that cut hospital admissions of high-risk patients. Health Affairs, 31, 1156-1166. doi:10.1377/hlthaff.2012.0393

Centers for Medicare and Medicaid Services. (2001). Medicare coordinated care demonstration. Retrieved from http://www.cms.gov/Medicare/Demonstration-Projects/DemoProjectsEvalRpts/Medicare-Demonstrations-Items/CMS1198864.html

Coburn

K. D.

Marcantonio

Lazansky

Keller

Davis

(2012). Effect of a community-based nursing intervention on mortality in chronically ill older adults: A randomized controlled trial. PLoS Medicine, 9(7), e1001265. doi:10.1371/journal.pmed.1001265

Counsell

S. R.

Callahan

C. M.

Clark

D. O.

Buttar

A. B.

Stump

T. E.

Ricketts

G. D.

(2007). Geriatric care management for low-income seniors: A randomized controlled trial. Journal of the American Medical Association, 298, 2623-2633. doi:10.1001/jama.298.22.2623

Hong

C. S.

Siegel

A. L.

Ferris

T. G.

(2014). Caring for high-need, high cost patients: What makes for a successful care management program? New York, NY: The Commonwealth Fund.

Marek

K. D.

Stetzer

Adams

S. J.

Bub

L. D.

Schlidt

Colorafi

K. J.

(2014). Cost analysis of a home-based nurse care coordination program. Journal of the American Geriatrics Society, 62, 2369-2376. doi:10.1111/jgs.13162

10.

Martin

K. R.

Schoster

Woodard

Callahan

L. F.

(2012). What community resources do older community-dwelling adults use to manage their osteoarthritis? A formative examination. Journal of Applied Gerontology, 31, 661-684. doi:10.1177/0733464810397613

11.

Federal Register. (2013, December). Medicare program; Revisions to payment Policies under the physician fee schedule; Clinical laboratory fee schedule & other revisions to part B for CY 2014. Retrieved from https://www.federalregister.gov/articles/2013/12/10/2013-28696/medicare-program-revisions-to-payment-policies-under-the-physician-fee-schedule-clinical-laboratory

12.

Nagraj

Abel

Paddison

Payne

Elliott

Campbell

Roland

(2013). Changing practice as a quality indicator for primary care: Analysis of data on voluntary disenrollment from the English GP Patient Survey. BMC Family Practice, 14, 89-96.

13.

Naylor

M. D.

(2002). Transitional care of older adults. Annual Review of Nursing Research, 20, 127-147.

14.

Naylor

M. D.

Aiken

L. H.

Kurtzman

E. T.

Olds

D. M.

Hirschman

K. B.

(2011). The care span: The importance of transitional care in achieving health reform. Health Affairs, 30, 746-754. doi:10.1377/hlthaff.2011.0041

15.

O’Connor

Bowles

K. H.

Feldman

P. H.

St Pierre

Jarrin

Shah

Murtaugh

C. M.

(2014). Frontloading and intensity of skilled home health visits: A state of the science. Home Health Care Services Quarterly, 33, 159-175. doi:10.1080/01621424.2014.931768

16.

O’Connor

Hanlon

Bowles

K. H.

(2014). Impact of frontloading of skilled nursing visits on the incidence of 30-day hospital readmission. Geriatric Nursing, 35(2 Suppl.), S37-S44. doi:10.1016/j.gerinurse.2014.02.018

17.

Parry

Kramer

H. M.

Coleman

E. A.

(2006). A qualitative exploration of a patient-centered coaching intervention to improve care transitions in chronically ill older adults. Home Health Care Services Quarterly, 25(3-4), 39-53.

18.

Peikes

Chen

Schore

Brown

(2009). Effects of care coordination on hospitalization, quality of care, and health care expenditures among Medicare beneficiaries: 15 randomized trials. The Journal of the American Medical Association, 301, 603-618. doi:10.1001/jama.2009.126

19.

Peikes

Peterson

Brown

R. S.

Graff

Lynch

J. P.

(2012). How changes in Washington University’s Medicare coordinated care demonstration pilot ultimately achieved savings. Health Affairs, 31, 1216-1226. doi:10.1377/hlthaff.2011.0593

20.

Rantz

Popejoy

L. L.

Galambos

Phillips

L. J.

Lane

K. R.

Marek

K. D.

. . . Ge

(2014). The continued success of registered nurse care coordination in a state evaluation of aging in place in senior housing. Nursing Outlook, 62, 237-246. doi:10.1016/j.outlook.2014.02.005

21.

Rogers

Perlic

Madigan

E. A.

(2007). The effect of frontloading visits on patient outcomes. Home Healthcare Nurse, 25, 103-109.

22.

Safran

D. G.

Montgomery

J. E.

Chang

Murphy

Rogers

W. H.

(2001). Switching doctors: Predictors of voluntary disenrollment from a primary physician’s practice. Journal of Family Practice, 50(2), 130-136.

23.

Schore

Peikes

Peterson

Gerolamo

Brown

(2011). Fourth report to congress of the evaluation on the Medicare coordinated care demonstration. Retrieved from https://www.cms.gov/reports/downloads/Schore_Fourth_Eval_MCCD_March_2011.pdf

24.

Thorpe

K. E.

Howard

D. H.

(2006). The rise in spending among Medicare beneficiaries: The role of chronic disease prevalence and changes in treatment intensity. Health Affairs, 25(5), w378-w388. doi:10.1377/hlthaff.25.w378

25.

Verhaegh

K. J.

MacNeil-Vroomen

J. L.

Eslami

Geerlings

S. E.

de Rooij

S. E.

Buurman

B. M.

(2014). Transitional care interventions prevent hospital readmissions for adults with chronic illnesses. Health Affairs, 33, 1531-1539. doi:10.1377/hlthaff.2014.0160