O’Neil, Margaret E. PT, PhD, MPH; Costigan, Tracy E. PhD; Gracely, Edward J. PhD; Wells, Nora MEd
Between 13% and 19% of children in the United States have special healthcare needs.1–5 A widely accepted definition for children with special healthcare needs (CSHCN) is “children who have or are at increased risk for a chronic physical, developmental, behavioral, or emotional condition and who also require health and related services of a type or amount beyond that required by children in general.”6
Approximately 7% and 9% of CSHCN have disabilities or conditions that are severe enough to cause limitations in functional mobility and participation (eg, play and school).2,7–10 Although children with disabilities make up a small percentage of the US childhood population, their healthcare costs are 3–4 times those of children who are typically developing.9–12 Further, childhood disability results in approximately 66 million restricted activity days including 24 million days out of school, an increase of 26 million physician contacts, and 5 million hospital days.13
Child, family, and insurance characteristics influence access to and utilization of healthcare services for CSHCN.10,14,15 Andersen’s Behavioral Model of Health Services Use provides a framework to explain access and use.16,17 This model includes population characteristics such as predisposing factors (eg, child conditions), enabling factors (eg, family and insurance characteristics), and need factors (eg, attitudes and beliefs) to explain access to care.
Health insurance coverage is a major factor in access to health services.18 Three important insurance characteristics to ensure access are coverage, continuity of coverage, and adequacy of coverage.15 Overall, trends indicate that 92% of CSHCN have health insurance coverage.3 Increases in insurance coverage are associated with increases in numbers of children with public insurance (eg, Medicaid and State Children’s Health Insurance Program).3
There is a high prevalence of CSHCN living in families with parents who have limited education and low incomes, families with single-parent households, and families from minority groups.19,20 These family characteristics are associated with decreased health insurance coverage.19 A disproportionate number of families with CSHCN are at or below the poverty level and the severity of a child’s condition is inversely associated with the family income.21 The prevalence of CSHCN is high for male children, minorities, and older children.19 These child characteristics are associated with decreased access to health services and insurance.3,10
CSHCN (especially those with disabilities) often require a complex array of health and rehabilitation services. Parents have a crucial role in gaining access to these services for their children.22 It is important for a parent to learn the specifics of their child’s health plan to understand the covered benefits, to submit claims, and to appeal a claim if denied.22 Parent-provider partnerships are critical to help parents develop effective advocacy skills for their children and to help parents find community resources and payment for needed services.22
Rehabilitation services for CSHCN often include physical, occupational, and speech therapy. Timely access to health and rehabilitation services is important to promote health, development, function, and participation in home, school, and community activities and to prevent secondary conditions that may be associated with disability.23–26
Family-centered care is the preferred service delivery model in pediatric physical therapy.27–29 Family-centered care approaches emphasize the importance of parent involvement in their child’s health and rehabilitation services. This service delivery model promotes parent-provider partnerships to facilitate communication, care coordination, and access to services.30,31
Families have direct access to physical therapy services in most states in the United States. The American Physical Therapy Association, Section on Pediatrics indicates in its vision statement that direct access will be available to all families and children by 2020.32 This expanded practice privilege comes with increased responsibility for pediatric physical therapists in both patient management and advocacy to support CSHCN and their families in gaining access to services. It is critical that physical therapists understand parents’ concerns and the problems with access to therapy services to ensure direct access.
The purpose of this study was to examine whether child, family, and insurance characteristics predict parents’ perceptions of problems with their health plans in accessing rehabilitation services for their CSHCN. Andersen’s Behavioral Model of Health Service Use provided a framework and rationale for this study and helped define the factors examined in the research model: predisposing factors (child characteristics), enabling factors (family and insurance characteristics), and environmental factors (insurance characteristics).16 Findings from this study will inform physical therapists in their advocacy for families and CSHCN.
A total of 1027 participants from the Family Partners Project were included in this study. Inclusion criteria for participants were that parent/caregiver respondents indicated that their CSHCN needed and received therapy services in the year before the survey and they completed key items on the survey that were part of the research model and analysis plan. A majority of respondents (90%) were mothers. The mean age of mothers was 39 years (SD = 7.6 years). The average age of the children was 8.3 years (SD = 4.4 years). Table 1 provides descriptive information on the child, family, and insurance characteristics examined in this study. Figure 1 illustrates the most frequently reported conditions/diagnoses for the participating children. Developmental delay was the most commonly reported childhood condition (66%) followed by allergies (40.8%) and cerebral palsy (39.2%).
The questionnaire used in the Family Voices Project, “Your Voice Counts!” consists of 89 items including sociodemographic characteristics of the children and families, children’s health status and conditions, characteristics of the child’s primary health plan, access and utilization patterns, and respondent ratings of child’s health plan benefits.1 Some items on the “Your Voice Counts!” questionnaire were similar to those used in the previous national health surveys, and some served as pilot test items for the national surveys for CSHCN.3,4
The Family Partners Project database was used in this secondary data analysis.33 The Family Partners Project was a partnership between Family Voices, a national advocacy organization of families, and friends of CSHCN and Brandeis University, Heller School of Social Policy.34 The purpose of the Family Partners Project was to examine families’ perceptions of their children’s health insurance plans and how these plans met children’s health and rehabilitation needs and family concerns.34
The Family Partners Project was a mail survey conducted in 1998–1999. A total of 6000 questionnaires were mailed and 2220 usable surveys were returned (41% response rate).33 Seventy-two percent (n = 1598) of respondents’ indicated that their children did receive therapy or rehabilitation services in the year before the survey. Sixty-four percent (n = 1027) of this subsample participated in this study based on meeting the inclusion criteria.
Twenty states participated in the survey (2 states from each of 10 federal regions of the Departments of Health and Human Services). States were chosen based on the high prevalence of managed care health plans. Families were recruited from 2 sources: Family Voices35 membership lists and states Title V Program participant lists (Maternal and Child State Health Programs for CSCHN). Random sampling was used to select families from the lists.33
The item from the Family Partners Project questionnaire used to determine whether parents had problems with access to rehabilitation services for their CSHCN was “In the last 12 months, did you have any problems getting the services of physical, occupational, speech or other therapists that your child needed?” Problems with access to therapy services included getting referrals, getting appointments; finding skilled therapists, getting the number of visits necessary to meet child’s needs, amount families had to pay for services, and lack of payment by health insurance plans.
Eight items were used to describe child characteristics, which included demographic information (sex, age, and race); health status, stability of condition (3-point scale), severity of condition (10-point scale), number of health conditions and overall health rating (5-point scale), and health services use (number of other health services received [up to 5 and exclusive of rehabilitation services]). The 13 items used to describe family characteristics were family demographics (relationship of respondent to child, marital status, employment status, family income, and education level), parent’s overall health rating (5-point scale), place of residence and sample group (Family Voices or Title V), family financial effect (out of pocket expenses), financial hardship (4-point scale), time commitment (time spent providing healthcare and time spent providing care coordination), and number of siblings with CSHCN. The 6 items that described insurance plan characteristics included type of plan (private or public), number of managed care characteristics (up to 4), parent’s rating of their plan’s overall performance on providing child’s health benefits (4-point scale), parent’s rating of their plan’s overall performance in providing child’s rehabilitation benefits (4-point scale), parent’s complaints to the plan (5-point scale), and parent’s overall satisfaction with the plan (4-point scale) (Table 1).
A scale was created to identify the number of managed care components in a child’s health plan based on the 4 items in the Family Partners Project questionnaire (health plan uses a provider list, plan pays for out of network services, plan requires a primary care provider, and plan requires a referral for specialty services).36,37 The scale was developed to determine the degree to which a child’s health plan incorporated managed care components because more than one fourth of the respondents (28%) did not know whether their child was in a managed care plan, but they could report characteristics of the plan.
Bivariate analyses were conducted to determine which child, family, and insurance variables would be retained for the hierarchical logistic regression analysis to determine predictors of problems with access to therapy services.38 Variables were retained if results from the bivariate analysis suggested moderate relationships (p ≤ 0.20) among child, family, and insurance characteristics (independent variables) and problems with access to therapy services (dependent variable). Post hoc regression analyses were generated to determine whether multicollinearity was present and posed a problem in the analyses.
Block entry for the hierarchical logistic regression model was based on the Andersen’s model and the family-centered approach to patient care. Therefore, the order of block entry was child, family, and finally insurance characteristics. Model adequacy was established from examination of the Omnibus test, Nagelkerke R2, Hosmer-Lemeshow Goodness of Fit test, and odds ratios (ORs) for individual predictors.
Of the 1027 respondents, 595 (58%) reported they had no problems getting therapy for their children, whereas 432 (42%) reported they had some problems getting these services. Comparisons were made within the respondent sample to determine differences among 3 distinct subgroups: the subgroup examined in this study (n = 1027), the subgroup not included in this study (because of missing data) but who did receive therapy services (n = 571), and the subgroup not in this study and who did not receive therapy services (n = 622). Findings indicate that there were some significant differences among these subgroups on child, family, and insurance characteristics. For example, children in the study had more severe and unstable conditions compared with children in the subgroup who did not receive therapy. Children in the study used more health services compared with children in the subgroup who were not in the study but who did receive therapy. Also, children in the study were younger and more were white in comparison with children in the other 2 groups.
Parents of children in the study reported better health, higher education levels, and higher income levels compared with the other subgroups. Also, parents in the study reported more financial problems because of their child’s health condition and spent more time providing home healthcare for their children compared with parents in the other subgroups.
There were no differences across subgroups on the number of components of managed care in the children’s health plans. More children in the study were in private health plans (employer-paid plans) compared with children in the other subgroups.
Using p ≤ 0.20 as a criterion, results of bivariate analyses reduced the original number of block items for the independent variables from 27 to 17. Six variables on child characteristics were retained and 2 were dropped (rating of child’s overall health and child’s number of conditions). Nine variables on family characteristics were retained and 4 were dropped (relationship of parent to child, marital status, employment status, and time spent coordinating child’s care). Two of the variables on insurance characteristics were retained and 4 were dropped (ratings of plan’s overall benefits, rating of plan’s rehabilitation benefits, overall satisfaction with the plan, and complaints to the plan).
Results of the 3-block hierarchical regression analysis are shown in Table 2 The logistic regression model accounted for a total of 19.1% of the variance in problems with access to therapy services. Given the block order of entry (child, family, and insurance blocks), the child block explained 4.5% of the variance, the family block explained an additional 10.8%, and the insurance block explained another 3.8% of the variance in problems with access to rehabilitation services. Table 3 presents the ORs for the predictor variables in the logistic regression model. For each predictor, the OR is an indicator of the degree to which the odds in favor of an event (ie, the odds of having problems with access to rehabilitation services) will change for an individual for either a unit change in the predictor or in the case of a 2-category predictor, for a change from 1 category to the other. Because the magnitude of an OR is dependent on the scale and type of variable used as a predictor, ORs cannot be compared across items in the model. Probability values (p values) provide an indirect indication of how strongly a predictor is associated with the dependent variable and are roughly comparable to one another in a single study in which the sample size is the same for all predictors.
Results of the regression analysis for the child block (Table 3) indicated that 2 of the 6 items were significant predictors of problems with access to rehabilitation services. Age was a risk factor (child’s age: OR = 1.04, p = 0.04, indicating that for each year that a child ages, there is an increased risk of problems with access to therapy services. Number of services received was a protective factor (number of other services child received: OR = 0.801, p = 0.02), suggesting that with increased services, there is a likelihood of decreased problems with access.
Significant results were found for 4 of the 9 items in the family block (Table 3). For each increase in unit of time that parents spent in providing home healthcare for their child, there was an increased risk of problems with access to therapy services (OR = 1.11, p = 0.046). Families with only 1 CSHCN had increased risk of problems with access to therapy services compared with families with 2 or more CSHCN (OR = 1.37, p = 0.045). For each unit increase in parents’ ratings of financial hardship caused by their child’s health condition, there was increased risk of problems with access to therapy services (OR = 1.63, p < 0.001). Finally, for each increased unit of out-of-pocket expenses to pay for child’s health needs, there was increased risk of problems with access to therapy services (OR = 1.13, p = 0.048).
Results on the insurance block (Table 3) indicated that both items were significant: one item was a protective factor and the other (managed care scale) was a risk factor. Parents who reported having a family-pay (self-pay) insurance plan were likely to have decreased problems with access to therapy services, when compared with families in employer-paid plans (OR = 0.53, p = 0.01). There was no significant difference in access to therapy services in government versus employer-paid plans. For each increase in managed care components in a child’s health plan, there was increased risk of problems with access to therapy services (OR = 1.33, p < 0.001).
Results on the hierarchical regression analysis suggest that this 3-block model has moderate explanatory power in predicting problems with access to rehabilitation services. Findings suggest that family characteristics were not enabling factors for access to rehabilitation services for CSHCN. Family financial hardships and increased time to provide healthcare to CSHCN were significant risk factors in access to rehabilitation services, suggesting that health insurance coverage may be inadequate to meet the needs of these children and their families. Time spent by the parents in providing healthcare may be a marker for increased severity of condition given that children in the study had more severe and unstable conditions compared with children in the other groups. One family characteristic was a protective factor (siblings with SHCN), suggesting that parents may learn to navigate the health system more effectively when they have more than one child for whom they must advocate and obtain services.
Child characteristics were not a strong predictor of access to rehabilitation service, and this may be due to only 2 significant items in this block and 1 being a protective factor. Children who received other health services (in addition to rehabilitation services) were less likely to have problems with access to rehabilitation services. This finding suggests that parents may have better skills working in the health system and are able to successfully access rehabilitation services when their children have been exposed to other services. Rehabilitation services may be delayed if parents cannot access these services directly and this delay could have negative effects on the child’s health outcomes. Increased child age was a risk factor for problems with access to rehabilitation services. This finding is supported by the literature on problems with access to general health services for CSHCN as they age.19 It seems that health insurance practices may limit access to rehabilitation services in a relatively young group of children (mean age = 8.3 years, SD = 4.37 years), which could be detrimental to development, function, and prevention of secondary conditions.
Insurance characteristics were the weakest predictor of access to rehabilitation services for CSHCN. The findings suggest that access to rehabilitation services is less problematic when families can afford self-pay plans, whereas those plans with managed care characteristics pose problems with access. These results match those of studies on access to general health services.36,37
It was expected that the insurance block would be a stronger predictor of problems with access. The explanatory power of this block did not change appreciably when the model was run with the insurance block entered first. Limitations in the research model may be the reason for the weak prediction of the insurance block. The 3-block model may not be mutually exclusive, thus calling for stricter definitions for each block. For example, family financial hardship may contribute to limited ability to afford a health insurance plan with benefits that meet the needs of CSHCN. Also, a child’s disability may limit the type of insurance plans and benefits available due to clauses on preexisting conditions. Further, families often struggle with decisions on full-time versus part-time employment to be able to provide care to their CSHCN. Family decisions on employment will influence types of insurance coverage available, which may have negative effects on access to rehabilitation services. Therefore, the 3-block predictor model may need further refinement to more clearly understand the contributions of child, family, and insurance characteristics in gaining access to rehabilitation services. However, this study provides a starting point for therapists to consider and gives us good information for further development of advocacy skills to support CSHCN and their families.
A majority of children (72%) in the Family Partners Project did receive rehabilitation services and more than one third (39%) of these families reported difficulty in accessing services. Given these findings, future studies are warranted to examine barriers to access to improve rehabilitation service delivery to CSHCN and their families. New directions for future studies on access to rehabilitation services may include modifying current questionnaires for CSHCN to obtain data specific to pediatric physical therapy. Such data would inform service delivery, practice, and advocacy. It is important that therapists “use the power of one’s privileged provider position to advocate for policies and programs that support high quality care.”39,p.225
The authors thank Dr. Marty Krauss and Nora Wells, the principal investigators on the Family Partners Project, for their permission to use the database in this study.
1. Bethell CD, Read D, Blumberg SJ, et al. What is the prevalence of children with special health care needs? Towards an understanding of variations in the findings and methods across three national surveys. Matern Child Health J. 2008;12:1–14.
2. Newacheck PW, Rising JP, Kim SE. Children at risk for special health care needs. Pediatrics. 2006;118:334–342.
3. United States Department of Health and Human Services, Health Resources and Services Administration, Maternal and Child Health Bureau. The National Survey of Children with Special Health Care Needs Chartbook 2005–2006. Rockville, MD: U.S. Department of Health and Human Services; 2007.
4. United States Department of Health and Human Services, Health Resources and Services Administration, Maternal and Child Health Bureau. The National Survey of Children with Special Health Care Needs Chartbook 2001. Rockville, MD: U.S. Department of Health and Human Services; 2004.
5. vanDyck PC, Kogan MD, McPherson M, et al. Prevalence and characteristics of children with special health care needs. Arch Pediatr Adolesc Med. 2004;158:884–890.
6. McPherson M, Arango P, Fox HB, et al. A new definition of children with special health care needs. Pediatrics. 1998;102:137–140.
7. Perrin JM. Health services for children with disabilities. Milbank Q. 2002;80:303–324.
8. Perrin JM, Bloom SR, Gortmaker SL. The increase of childhood chronic conditions in the United States. JAMA. 2007;297:2755–2759.
9. Newacheck PW, Inkelas M, Kim SE. Health services use and health care expenditures for children with disabilities. Pediatrics. 2004;114:79–85.
10. Newacheck PW, Kim SE. A national profile of health care utilization and expenditures for children with special health care needs [erratum appears in Arch Pediatr Adolesc Med. 159:318]. Arch Pediatr Adolesc Med. 2005;159:10–17.
11. Neff JM, Anderson G. Protecting children with chronic illness in a competitive marketplace. JAMA. 1995;274:1866–1869.
12. Neff JM, Sharp VL, Muldoon J, et al. Profile of medical charges for children by health status group and severity level in a Washington state health plan. Health Serv Res. 2004;39:73–89.
13. Newacheck PW, Strickland B, Shonkoff J, et al. An epidemiologic profile of children with special health care needs. Pediatrics. 1998;102:117–123.
14. Davidoff AJ. Insurance for children with special health care needs: patterns of coverage and burden on families to provide adequate insurance. Pediatrics. 2004;114:394–403.
15. Honberg L, McPherson M, Strickland B, et al. Assuring adequate health insurance: results from the National Survey of Children with Special Health Care Needs. Pediatrics. 2005;115:1233–1239.
16. Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Soc Behav. 1995;36:1–10.
17. Dusing SC, Skinner AC, Mayer ML. Unmet need for therapy services, assistive devices and related services: data from the National Survey of Children with Special Health Care Needs. Ambul Pediatr. 2004;4:448–454.
18. Newacheck PW, McManus M, Fox HB, et al. Access to health care for children with special health care needs. Pediatrics. 2000;105:760–766.
19. Newacheck PW, Hung YY, Wright K. Racial and ethnic disparities in access to care for children with special health care needs. Ambul Pediatr. 2002;2:247–254.
20. Newacheck PW, Hung YY, Park MJ, et al. Disparities in adolescent health and medical care: does socioeconomic status matter? Health Serv Res. 2003;38:1235–1252.
21. Newacheck PW, Halfon N. Prevalence and impact of disabling chronic conditions in childhood. Am J Public Health. 1998;88:610–617.
22. Wells N. Learning to care and advocate for children with special health care needs. In: Sobo EJ, Kurtin PS, eds. Optimizing Care for Young Children with Special Health Care Needs: Knowledge and Strategies for Navigating the System. Baltimore, MD: Paul H. Brooks; 2007;115–134.
23. Campbell S. Therapy programs for children that last a lifetime. Phys Occup Ther Pediatr. 1997;7:1–15.
24. King G, Tucker M, Baldwin P, et al. A life needs model of pediatric service delivery. Phys Occup Ther Pediatr. 2002;22:53–77.
25. Fowler EG, Kolobe TH, et al; Section on Pediatrics Research Summit Participants; Section on Pediatrics Research Committee Task Force. Promotion of physical fitness and prevention of secondary conditions for children with cerebral palsy: section on pediatrics research summit proceedings. Phys Ther. 2007;87:1495–1510.
26. Rimmer JH, Rowland JL, Yamaki K. Obesity and secondary conditions in adolescents with disabilities: addressing the needs of an underserved population. J Adolesc Health. 2007;41:224–229.
27. O’Neil ME, Ideishi R, Nixon-Cave K, et al. Care coordination between medical and early intervention services: family and provider perspectives. Fam Syst Health. 2008;26:119–134.
28. O’Neil ME, Palisano R, Westcott S. Relationship of therapists’ attitudes, children’s motor ability, and parenting stress on mothers’ perceptions of therapists’ behaviors during early intervention. Phys Ther. 2001;81:1412–1424.
29. King G, King S, Rosenbaum PL, et al. Family-centered care giving and well-being of parents of children with disabilities: linking process with outcome. J Pediatr Psychol. 1999;24:41–53.
30. O’Neil M, Palisano R. Attitudes toward family-centered care and clinical decision making in early intervention among physical therapists. Pediatr Phys Ther. 1999;12:173–182.
31. King GA, Law M, King SM, et al. Parents’ and service providers’ perceptions of family-centredness of children’s rehabilitation services. Phys Occup Ther Pediatr. 1998;18:21–40.
32. American Physical Therapy Association (APTA). Section on Pediatrics. Strategic Plan 2008–2010. APTA Combined Sections Meeting. Nashville, TN, 2008.
33. Krauss MW, Gulley S, Leiter V, et al. The Family Partners Project: Report on a National Survey of the Health Care Experiences of Families of Children with Special Health Care Needs. Unpublished report from Starr Center for Mental Retardation, Heller School, Brandeis University, Waltham, MA, 2000.
34. Wells N, Krauss MW, Anderson B, et al. What do families say about health care for children with special health care needs? Your voice counts! The Family Partners Project Report to Families. Family Voices at the Federation for Children with Special Needs. Boston, MA, 2000.
35. Family Voices Newsletter. Family Voices Promises to Our Children with Special Health Care Needs. PO Box 769, Algones, NM 87001, 1999.
36. Shenkman E, Wu SS, Nackashi J, et al. Managed care organizational characteristics and health care use among children with special health care needs. Health Serv Res. 2003;38:1599–1624.
37. O’Neil M, Farel A, Palisano RJ. Parents’ perspectives of managed care: implications for pediatric physical therapy. Pediatr Phys Ther. 1999;11:24–32.
38. Hosmer DW, Lemeshow S. Applied Logistic Regression. New York: Wiley; 2000.
39. Sobo EJ. Mastering the health care system for children with special health care needs. In: Sobo EJ, Kurtin PS, eds. Optimizing Care for Young Children with Special Health Care Needs: Knowledge and Strategies for Navigating the System. Baltimore, MD: Paul H. Brooks; 2007:115–134.
© 2009 Lippincott Williams & Wilkins, Inc.