 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* Department of Health Psychology
Office of Medical Research
Department of Psychological Sciences
|| Department of Child Health, University of Missouri Health Sciences Center, Columbia, Missouri
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methods. Fifty-one parents of children with special health care needs participated in a pre-/posttreatment assessment of a program designed to enhance comprehensive and coordinated care. Participants were recruited from 3 primary care practices in a central Midwest state and remained in the program for 
12 months.
Results. Parents reported significant increases in satisfaction with care coordination and access to mental health services after the intervention. They also noted decreases in family needs, caregiver strain, parents’ missed work days, children’s school absences, and utilization of ambulatory services. Satisfaction with primary care declined slightly but remained in the "very good" range. Families of children with more complex conditions were more likely to report a decrease in needs after intervention, but other factors, such as geographic location or socioeconomic status, were not related to key outcome variables.
Conclusions. Comprehensive care has a positive effect on children with chronic health conditions and their families, including those who live in more rural areas. Additional study is needed to learn more about rural service delivery strategies that promote implementation of this approach in general practice.
Key Words: medical home • children with special health care needs • rural health services
Abbreviations: CSHCN, children with special health care needs • SES, socioeconomic status • NP, nurse practitioner
Children with special health care needs (CSHCN) rely on a complex and fragmented system of care to meet their medical, educational, mental health, and social service needs.1,2 Both parents and primary care physicians report marked difficulty navigating this system to obtain comprehensive and coordinated services.3–5 The barriers that interfere with quality care include a lack of communication between primary and specialty care providers, long wait times for needed care, poor coordination of medical and nonmedical services, limited access to mental health and other related services, and inadequate health care financing. Some children and families seem more vulnerable to these problems in the health care system than others. For example, unmet child health needs have been related to socioeconomic factors such as minority status, living in poverty, lack of health insurance, and lower parental education.5–7
Children who have chronic health conditions and live in rural areas may be at particular risk for fragmented care, in part because of inadequacies in the rural health delivery system.6,8,9 There are persistent shortages of pediatricians and other primary care providers in rural areas, and specialty care providers are virtually nonexistent. Families must travel long distances to obtain specialized services, at considerable personal and financial cost. Furthermore, children in rural areas are more likely to live in poverty and to experience periods without health insurance compared with their urban peers, increasing their risk for unmet needs for routine and specialty care.10,11 Living in a rural location may compound the problems with access to quality care faced by all families of CSHCN.
In recent years, there has been an increasing focus on improving service delivery for children with chronic conditions through enhancements in primary care practice. National goals in Healthy People 2010 include the development of a medical home for all children, an approach that promotes accessible, continuous, comprehensive, family-centered, coordinated, compassionate, and culturally effective care.12 Pediatricians have endorsed the medical home model, and they have proposed guidelines for the development of medical homes for CSHCN in primary care settings.13–15
Despite this interest in primary care enhancements, very few research studies have examined directly child and family outcomes after medical home interventions. However, available evidence suggests that a comprehensive, coordinated, and community-based approach improves outcomes for CSHCN and their families.1,16–21 These studies have reported positive changes in parent access to and satisfaction with health care services, decreased health needs, fewer hospitalizations, improved child and maternal adjustment, fewer missed work days, and reduced health care costs. Ratings of children’s level of functioning typically remain stable, neither improving nor deteriorating, and children with the most severe health needs seem most likely to benefit.16 It is interesting that most primary care research of this type has been conducted in large urban areas, where needs may be great but health and related resources are also more available.
The purpose of this article was to examine outcomes from a medical home research demonstration project for CSHCN in a central Midwest state that has been described previously.15,22 To our knowledge, this is the first study to evaluate the feasibility and impact of a medical home intervention for CSHCN in this geographic area. We examined program outcomes in 3 broad areas: access to and satisfaction with health care and related services, family functioning, and child functioning. Study hypotheses included the following (1) the medical home intervention will help families navigate the system of care, increasing access to needed services and satisfaction with care coordination; (2) the program will improve family functioning by decreasing unmet needs, reducing caregiver strain, and decreasing missed work days; and (3) the children’s level of functioning will not change, given the chronic nature of their health conditions. The study also compared the outcomes of participants who live in rural versus urban settings and examined other potential moderating factors, such as the complexity of the child’s health condition and socioeconomic indicators. Families that live in rural counties, have children with more complex conditions, or have fewer socioeconomic resources were expected to benefit more from the intervention.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Of the referred children, 149 met eligibility criteria. In the eligible group, 37 parents declined and 29 consented but did not complete baseline research measures. Therefore, the sample size at the beginning of the study was 83 children (56% of those eligible). During the 1-year intervention, 51 (61%) of these 83 families remained in the study. Reasons for dropping out included the following: 4 declined participation, 13 could not be contacted, 4 did not return a survey, 7 were no longer eligible (eg, moved), 3 transferred to other practices, and 1 child died. Parents who dropped out were more likely to be single (
2 [1, N = 83] = 10.53; P = .0012), and the children were more likely to have a higher functional status (Wilcoxon rank sum test, z = 1.94; P = .05) than those who remained in the study. Dropouts did not differ in geographic location, other demographic characteristics, or number of family needs.
The final sample of 51 children is described in Table 1. Each child’s primary diagnosis, as determined by the referring physician, was classified into 1 of 6 categories using the International Classification of Diseases, Ninth Revision, Clinical Modification codes.24 In all, there were 28 different primary diagnoses represented, and 71% of the children had >1 diagnosis (median: 2; range: 1–7). On the basis of parent report, average time since onset of the primary condition was 5.6 years (SD: 4.8 years; range: 0.4–17.2 years). Approximately half of the children resided in a metropolitan county that included a small city of 79000 and an academic health center, and the rest lived in 8 adjacent rural counties (region area = 5144 mi2). Families lived up to 75 miles from the health center and even farther from other tertiary care. Children from ethnic/racial minorities were overrepresented in the sample (33% enrolled compared with 15% living in the region).25 Most families were of lower socioeconomic status (SES; Hollingshead SES mean: 36.17; SD: 19.97),26 and all children had health insurance (33% Medicaid fee-for-service, 32% Medicaid managed care, and 35% commercial).
|
Procedure
The medical home intervention was based on the model of care described by Palfrey and colleagues.15,16 The care team consisted of the primary care physician and office staff, the child and the family, a nurse practitioner (NP), and a parent consultant (a paid family member of a child with special health care needs). The intervention focused on the provision of care coordination, information about resources and services, emotional support and encouragement, and empowerment for families to advocate for their children. To accomplish these goals, every participant received from the NP a set of basic services that included a home visit to conduct a comprehensive assessment of medical and nonmedical needs of the child and family members; a personalized letter that described health, educational, and community resources for meeting these needs; an individualized written health plan for the child; assistance in developing short-term family goals; and at least 1 follow-up to discuss progress toward goals and to problem-solve about any barriers to needed care. In addition, the parent consultant assisted with family-to-family supports, as needed. The intervention lasted 6 months from the time of the first home visit, with periodic follow-up in the second 6 months.
The project’s NP acted as a consultant to the 3 practices, interacting regularly with referring physicians and a designated nurse at each office to problem-solve about ways to improve the medical home for individual children. All medical care was provided by the practice, not by the NP. This differed from the medical home intervention described by Palfrey et al16 in which each practice selected and trained a pediatric NP from their own staff. The 3 practices in this study did not use pediatric NPs because they were not readily available. In the planning stages of this project, participating physicians indicated a preference for the itinerant NP intervention because it allowed practices to adapt to professional shortages by using a consultant to help build the medical home for CSHCN. The project initially used a social worker to consult with the practices, because this type of professional is more available in rural areas, but then shifted to using an NP to enhance care coordination for the medically complex children who were being referred. Five (10%) of the 51 families in the sample received services from the social worker only, but the majority had contact with the NP during the course of the intervention.
In addition to assisting with individual children, project staff met with each practice as a group at least quarterly to answer questions and to provide continuing education on CSHCN topics. A web site was developed for use by families and physicians to promote easy access to additional supports and resources online. A program manual that describes the intervention process in detail is available.
Consent to participate in the demonstration project was obtained from families by telephone, according to a protocol approved by the Institutional Review Board. Parents then completed a packet of questionnaires before the first home visit (time 1). Families were sent the same packet of questionnaires after completion of the program (time 2), with the addition of a measure that assessed satisfaction with the medical home intervention. These measures are described in Table 2. Participants completed the posttreatment assessment at 1 year after program enrollment (mean: 1.07 years; SD: 0.37 years). Families were paid $20 for completing the questionnaires at each assessment. Respondents to the questionnaires were mostly mothers or maternal figures, such as grandmothers (98% at both time 1 and time 2). Medical chart review in the primary care practice was conducted pre- and posttreatment to determine whether children were up to date on immunizations and well-child checks.
|
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
With regard to satisfaction with health services, mothers indicated a significant improvement in mean satisfaction with care coordination services at time 2 (see Table 3). There was a slight decline in satisfaction with primary care services, but mean ratings of primary care were high both before and after the intervention. There was no change in satisfaction with other health and health-related services.
Family Functioning
Mothers reported a reduction in total family needs after the intervention, with less need for social support, financial/material assistance, and help with family relationships (see Table 3). Mothers also described a decrease in family strain on the composite Impact on Family Scale measure, primarily as a result of changes on the Familial/Social subscale rather than on the Personal Strain subscale. Ratings of mean number of missed work days did not change after the intervention. However, when the sample was dichotomized into those who missed > or <10 days, fewer mothers missed 10 or more days of productive activities as a result of their child’s health condition (38% at time 1 vs 22% at time 2; McNemar S [1, N = 50] = 4.57; P = .03).
Child Functioning
Across the 1-year intervention, mothers’ ratings of their child’s functional status did not change (see Table 3). However, ratings of school attendance for children of school age indicated significant improvements in school participation from pre- to posttreatment. Fewer children missed 13 or more days of school after the intervention (28% at time 1 vs 14% at time 2; McNemar S [1, N = 36] = 5.0; P = .025).
Program Acceptability
When asked at time 2 about their perceptions of the medical home intervention, mothers indicated that it was very helpful to extremely helpful (n = 42; mean: 1.48; SD: 0.59). Ninety-five percent of mothers reported that the intervention was helpful to them. Each of the components of the intervention also was rated highly (n range: 25–40; mean: 1.28–1.89; SD: 0.54–0.79). The modal rating was "extremely helpful" for 6 of the components (written care coordination plan, support from the nurse who worked with families, resource referrals, communicating with other parents, family social activities, and family newsletter) and "very helpful" for the other 2 components (communicating with physicians and with educators).
Factors Related to Change Over Time
Analyses were conducted to examine factors that were associated with change in 3 key outcome variables: number of family needs, perceived family burden, and child functioning. With the use of baseline data collected at time 1, moderating factors considered were geographic location (rural versus urban), mother’s education level (high school or less versus more than high school diploma), family SES, minority status (white versus other), type of insurance (Medicaid managed care versus Medicaid fee-for-service versus other), and initial reported need for care coordination (needed versus not needed). No significant associations were found between change in any of the 3 outcome variables and these potential moderating factors (P = .22–.94). Next, the relationships between time 1 level of child functioning and time 2 family needs and perceived family strain were examined. Mothers who had children with lower functioning at time 1 reported fewer family needs at time 2 (Spearman’s r = .29; P = .04).
Additional comparisons of rural versus urban subgroups did not reveal any significant differences in demographic characteristics, number of contacts by the consultant, or change in outcome variables. The relatively small sample size for the rural and urban groups (n = 22 and 29, respectively) may have limited the ability to detect differences, but there was also significant variability within these 2 groups. For example, the urban group (mean: 10.9 needs; SD: 7.3) seemed to have more family needs on average than the rural group (mean: 7.6 needs; SD: 6.6), but this was not statistically significant because the SD was large and needs ranged from 0 to >20 in both groups.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
These improvements likely resulted from the partnership formed between the itinerant NP, the family, and the primary care practices. The written health plan and resource letter served as efficient communication tools to inform primary care physicians about family priorities and about strategies to meet medical and nonmedical needs. In some cases, this information prompted practice changes, such as scheduling longer office visits to problem-solve with family members about their concerns. The project NP also directly connected parents to resources and facilitated more effective communication among health care providers, educators, and other community professionals. Occasionally, the NP arranged multidisciplinary care conferences to discuss complex child problems or challenges to adequate service delivery. These activities seemed to extend the physicians’ ability to build a medical home for CSHCN and their families.
Unexpectedly, parents’ satisfaction with primary care services declined slightly during the intervention but remained in the "very good" range on average. This change may simply reflect regression to the mean rather than a true decline in perceptions about care. However, it is also possible that physician referral to the NP decreased parental perceptions of continuity of care for their child, thus slightly reducing their satisfaction with their primary care provider. Utilization of and satisfaction with other medical services, therapies, and medical equipment did not change after the intervention, and there was no change in indicators of preventive care. Taken as a whole, these findings suggest that having a regular source of primary care promotes adequate medical management of CSHCN,7,32 but the medical home intervention is necessary to enhance mothers’ perceptions of coordinated care across multiple providers and systems of care.
The medical home intervention also had a positive impact on family functioning. Mothers indicated a pattern of improvements on the outcome measures in this domain, including fewer family needs, decreased family strain, and fewer work absences. These positive family effects were underscored by high ratings of program acceptance. The intervention seemed to fill a gap for families by addressing a broad range of parent-identified issues, such as parent-to-parent supports, family relationships, and children’s emotional and behavioral concerns, in addition to child health needs. Although comprehensive, family-centered caregiving has been associated with well-being in parents of CSHCN in previous research,1,16,18,33 this is one of the first studies to examine family outcomes after implementation of the medical home model.
Despite continuing challenges related to their health needs, the children in this sample were more likely to attend school and less likely to be taken to primary and specialty care appointments after intervention. These improvements may be mediated in part by the positive changes observed in family functioning. That is, although health care utilization is affected by the child’s age and health status, greater health service use among CSHCN has also been associated with higher parental stress and lower levels of family support.34–36 Similarly, school attendance has been associated with the quality of family relationships in healthy children and in those with severe psychiatric problems.37,38 By promoting optimal family functioning through the medical home model, primary care providers may be able to achieve these indirect effects on child outcomes, but additional research in this area is warranted to examine this premise.39
Counter to study hypotheses, geographic location did not moderate child and family response to the intervention. Needs and resources seemed to vary within families, rather than by geographic setting. However, the medical home intervention had to be adapted for rural families in ways that increased staff time and cost. For instance, the nurse traveled up to 150 miles round trip for home visits and up to 80 miles round trip to meet with primary care physicians. Effective rural interventions required creative, individualized solutions to overcome workforce shortages (eg, lack of speech/language therapists), financial constraints (eg, underfunded school districts), and general lack of familiarity with CSHCN and pertinent resources (eg, Medicaid nonemergency transportation funds, existing community supports). By traveling to rural communities and tailoring the intervention for each family, the itinerant NP was able to link participants to needed services and supports regardless of location. The participating physicians preferred this consultative model, because it supplemented their own limited staff resources. In the future, this approach might be sustained through specialized training of state agency personnel (eg, Title V service coordinators). It should be noted that the success of this intervention may be attributable to the fact that rural counties were adjacent to a metropolitan area. This model may not be as effective for families and practices in more remote rural areas of the county, where outcomes may vary as a result of the extreme challenges of gaining access to specialized care and where other approaches such as telehealth interventions may be essential.40,41 Additional research with larger and more diverse samples will be required to determine optimal rural service delivery strategies for CSHCN.
The primary factor that did moderate outcomes in this study was the child’s functional status. Mothers of children with lower functional status were more likely to remain enrolled over time and reported a greater decrease in unmet needs compared with mothers with children with less complex needs. Other studies have shown that families of children with more severe health concerns have more unmet needs and benefit more from comprehensive care interventions.16,42 However, Jessop and Stein43 found that families who benefited most from their comprehensive care initiative were those with low illness burden combined with low coping resources, rather than simply those with high illness burden. Additional research is needed to clarify which families benefit most from comprehensive care interventions, taking into account both medical and psychosocial predictors of outcomes.
This study has several limitations. The lack of a control group makes it difficult to know whether pre–post improvements were simply attributable to chance or external factors (eg, extra attention from the project staff, other changes in the family or health care system, social desirability of positive responses after treatment). In addition, no statistical corrections were made for the multiple outcome analyses conducted. This approach was used to increase the likelihood of identifying effects of the intervention, given the relatively few studies of the medical home model. However, 1 to 2 of the analyses conducted in this study may be significant at the .05 level by chance alone. The generalizability of the findings also may be limited for single parents and for children with milder conditions as a result of differential attrition, and outcomes may vary for uninsured children. Finally, parent report must be supplemented with independent sources of information about variables such as health care utilization and school absences.
Much remains to be learned about delivering comprehensive care to CSHCN in primary care settings. The long-term impact of comprehensive care interventions must be evaluated through longitudinal studies that examine the maintenance of treatment effects over time in geographically and culturally diverse settings. Cost savings resulting from the medical home intervention also must be examined. By refining and extending studies of the medical home model, researchers may encourage policy changes that support adequate reimbursement for comprehensive care for CSHCN.
|
ACKNOWLEDGMENTS |
|---|
We thank Dr Judith Palfrey, Dr W. Carl Cooley, and members of their staff for consultation provided during the project development. We also thank Dr Lori Frasier, Dr Alan Gill, Dr Rick Stone, and Dr Joseph Beckmann for leadership in the clinical care of CSHCN and their families.
|
FOOTNOTES |
|---|
Reprint requests to (J.E.F.) Department of Health Psychology, University of Missouri-Columbia, DC116.88, One Hospital Dr, Columbia, MO 65212. E-mail: farmerje{at}health.missouri.edu
Portions of this study were presented at the annual meeting of the Pediatric Academic Societies; May 3–6, 2003; Seattle, WA.
No conflict of interest declared.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. J. Homer, K. Klatka, D. Romm, K. Kuhlthau, S. Bloom, P. Newacheck, J. Van Cleave, and J. M. Perrin A Review of the Evidence for the Medical Home for Children With Special Health Care Needs Pediatrics, October 1, 2008; 122(4): e922 - e937. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Perrin, D. Romm, S. R. Bloom, C. J. Homer, K. A. Kuhlthau, C. Cooley, P. Duncan, R. Roberts, P. Sloyer, N. Wells, et al. A Family-Centered, Community-Based System of Services for Children and Youth With Special Health Care Needs Arch Pediatr Adolesc Med, October 1, 2007; 161(10): 933 - 936. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. O'Shea, S. Nageswaran, D. C. Hiatt, C. Legault, M. L. Moore, M. Naughton, D. J. Goldstein, and R. G. Dillard Follow-up Care for Infants With Chronic Lung Disease: A Randomized Comparison of Community- and Center-Based Models Pediatrics, April 1, 2007; 119(4): e947 - e957. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2005 American Academy of Pediatrics. All rights reserved. |
||
|
||
