Maternal and Child Health Journal

, Volume 16, Supplement 1, pp 129–142

Factors Associated with Parent Report of Access to Care and the Quality of Care Received by Children 4 to 17 Years of Age in Georgia

Authors

    • Division of Public HealthGeorgia Department of Community Health
  • David Goodman
    • Division of Reproductive HealthCenters for Disease Control and Prevention
  • Katherine Kahn
    • Division of Public HealthGeorgia Department of Community Health
  • Brendan Noggle
    • Division of Public HealthGeorgia Department of Community Health
  • Cherie Long
    • Division of Public HealthGeorgia Department of Community Health
  • Suparna Bagchi
    • Division of Public HealthGeorgia Department of Community Health
  • Danielle Barradas
    • Division of Reproductive HealthCenters for Disease Control and Prevention
  • Brian Castrucci
    • Division of Public HealthGeorgia Department of Community Health
Article

DOI: 10.1007/s10995-012-1002-2

Cite this article as:
Ogbuanu, C., Goodman, D., Kahn, K. et al. Matern Child Health J (2012) 16: 129. doi:10.1007/s10995-012-1002-2

Abstract

We examined factors associated with health care access and quality, among children in Georgia. Data from the 2007 National Survey of Children’s Health were merged with the 2008 Area Resource File. The medically underserved area variable was appended to the merged file, restricting to Georgia children ages 4–17 years (N = 1,397). Study outcomes were past-year access to care, defined as utilization of preventive medical care and no occasion of delay or denial of needed care; and quality of care received, defined as compassionate, culturally-effective, and family-centered care which was categorized as higher, moderate, or lower. Analysis included binary and multinomial logit modeling. In our study population, 80.8 % were reported to have access to care. The quality of care distribution was: higher (39.4 %), moderate (30.6 %), and lower (30.0 %). Younger age (4–9 years) was positively associated with having access to care. Compared to children who had continuous and adequate private insurance, children who were never/intermittently insured or who had continuous and inadequate private insurance were less likely to have access. Compared to children who had continuous and adequate private insurance, there were lower odds of perceiving received care as higher/moderate versus lower quality among children who were never/intermittently insured or who had continuous and inadequate/adequate public insurance. Being in excellent/very good health and living in safe/supportive neighborhoods were positively associated with quality; non-white race/ethnicity and federal poverty level were negatively associated with quality. Assuring continuous, adequate insurance may positively impact health care access and quality.

Keywords

ChildrenHealth care accessHealth care qualityHealth care utilizationNational Survey of Children’s Health

Introduction

The Institute of Medicine (IOM) has defined health care access as the timely use of personal health services to achieve the best possible health outcomes [1], and quality care as safe, effective, patient-centered, timely, efficient, and equitable care [2]. Patient-centered care involves providing care that is respectful of, and responsive to, individual patient preferences, needs, and values, and ensuring that patient values guide all clinical decisions.

Factors associated with access to care include insurance [39], having a personal health care provider [10], having a usual source of care [11], and race/ethnicity [9, 1215]. Factors associated with quality include race/ethnicity [13, 14, 16, 17], insurance [16, 17], parent’s education, primary household language, child’s overall health status, child’s age, and region of residence [17]. Although published studies on health care access [18, 19] and quality [8, 17] exist at the national level, there is a need for state-level assessment.

The Georgia 2010 Title V 5-Year Needs Assessment, which included the collection, analysis and review of quantitative and qualitative data, identified children’s health care access as a major health concern [20]. Indicators identified by partners and stakeholders which suggested opportunities to improve both access and quality of care for Georgia children were: insurance problems, geographic inaccessibility, unavailability of providers, and the challenge working parents face trying to see providers during regular office hours [20].

In 2007, 88.3 % of Georgia children, aged 0 to 17 years, had a preventive medical visit in the prior year; but only 58.5 % received care that was accessible, continuous, comprehensive, coordinated, compassionate and culturally effective [21]. The proportion receiving coordinated, ongoing, comprehensive care within a medical home was lower (47.3 %) for children with special health care needs (CSHCN) [20].

This state-level study describes the prevalence of and factors associated with access to preventive medical care and receipt of quality health care for children in Georgia. This study also provides an example of how the National Survey of Children’s Health (NSCH) data can be used for improving programs and practice at the state level; serving as a guide for state planners in Georgia; and providing an analytic model for states which identified children’s health care access and quality as priorities during the 2010 Title V Needs Assessment [22]. There are increasing demands and requirements for measures to monitor quality of care for children [23, 24], giving this study added timeliness.

Methods

Study Design

Data were drawn from a dataset created by merging the 2007 NSCH public use file (PUF) with selected 2007 variables from the 2008 Area Resource File (ARF). The medically underserved area (MUA) variable for Georgia, which was downloaded from the Health Resources and Services Administration (HRSA) website [25], was then appended to the merged dataset. Because the merge required access to the restricted-use county of residence variable, the merge and subsequent analyses were conducted at the National Center for Health Statistics’ Research Data Center. The dataset was limited to Georgia children 4–17 years of age (N = 1,397) because the current American Academy of Pediatrics guidelines stipulate several well-child visits from age 0–3 years, but only one visit yearly thereafter, until age 21 years [26].

Details of the 2007 NSCH are described in the foreword of this special issue. The ARF, produced by the HRSA Bureau of Health Professions, provides county-level data on several indicators, including health professions supply and health facility counts by type [27]. Medically underserved areas/populations are areas or populations designated by HRSA as having a low ratio of primary care physicians per 1,000 population, high infant mortality, poverty, and/or elderly population [25]. An index of medical service ≤62.0, based on these factors, qualifies for designation as an MUA [28].

Child-level socio-demographic and health-related variables were obtained from the 2007 NSCH PUF while the number of federally qualified health centers (FQHCs), number of rural health clinics (RHCs), health professional shortage areas (HPSAs) for primary care, and rural–urban designation were obtained from the 2008 ARF.

Variables

Dependent Variables

We investigated access to health care and quality of health care received. For the purpose of these analyses, access to health care is a dichotomous outcome derived from two survey items: utilization of any preventive medical care, and delay or non-receipt of needed care; both in the prior 12 months. Children who had at least one preventive medical care visit and who did not experience any delay or non-receipt of needed care were coded as having access, while those with any other combination (except missing observations) were coded as not having access. We only assessed preventive medical care, because it is the minimum care all children must have, and we combined timeliness of care with utilization based on the IOM’s definition of access [1].

Quality of health care received was derived from five questions related to the receipt of compassionate, culturally-effective and family-centered care. Specifically, whether the health care provider: (1) spends enough time with child, (2) listens carefully to parent, (3) is sensitive to family values and customs, (4) provides specific needed information, and (5) makes parent feel like a partner in child’s care. This definition of health care quality is consistent with the IOM’s emphasis on patient-centered care [2]. Children were classified as having received higher, moderate, or lower quality care based on their parents’ responses to these questions. Children in the higher category had a parental response of always to all five questions; children in the moderate category had a parental response of a combination of always and usually to all five questions; and children in the lower category had a parental response of sometimes or never to at least one of the five questions.

Control Variables

Control variable selection was guided by Andersen’s behavioral model of health services utilization (external environment, predisposing, enabling and need domains; Fig. 1) [10, 18, 19, 29, 30]. The external environment domain is comprised of physical, political, and economic factors that can influence health services utilization. For this study, we included seven binary measures of neighborhood conditions and resources, including the presence of paths, parks, recreation centers, litter, and rundown housing in the neighborhood.
https://static-content.springer.com/image/art%3A10.1007%2Fs10995-012-1002-2/MediaObjects/10995_2012_1002_Fig1_HTML.gif
Fig. 1

Andersen’s behavioral model of health services utilization. FQHC federally qualified health center, RHC rural health clinic, CSHCN children with special health care needs

The predisposing domain includes factors that may incline one to use health care services, such as characteristics encompassing one’s social networks, interactions, and culture. We investigated 12 variables in this domain. Social support is a composite variable created from four survey items: in the neighborhood, (1) people help each other out; (2) we watch out for each other’s children; (3) there are people I can count on; and (4) there are adults whom I trust to help my child if he/she got hurt or scared while playing outside. Children whose parents responded with “1-definitely agree” on all questions were classified as having strong social support. Children whose parents responded with a combination of “1-definitely agree” and “2-somewhat agree” on all questions were classified as having moderate social support. Children whose parents responded with “3-somewhat disagree” or “4-definitely disagree” on at least one of the four questions were classified as having weak social support. For sensitivity testing of the social support variable, we created the social capital index variable from the same four items by summing up the responses, with 4 being the highest social capital index and 11+ being the lowest. Additional variables, including the number of adults/children in the household [10, 12, 18], family structure [8, 18, 31], primary household language [8, 10, 12, 32], parental employment [8, 12, 33], and length of stay of mother (biological, step, foster, or adoptive) in the US [34] were included in this study, consistent with previously published literature.

The enabling domain includes community and personal enabling resources. Nine variables were investigated in this domain. For household income, the percentage of the 2007 federal poverty level (FPL) was used and grouped into 5 categories: at or below 100 %, >100–200 %, >200–300 %, >300–400 %, and above 400 %. Insurance coverage, a composite variable, was created from questions covering current coverage; gaps in the previous 12 months; adequacy in terms of benefits, providers, and out-of-pocket costs; and insurance type (public or private). Two variables were investigated in the need domain: child’s special health care needs status and overall health status.

Data Analysis

Descriptive statistics were computed and bivariable analysis conducted using Chi-square tests to identify significant associations between the independent and dependent variables. Descriptive statistics of study variables were also obtained from the national dataset to identify differences between Georgia and US child populations on factors potentially related to access and quality. Binary logit analysis was conducted for access to health care, while multinomial logistic regression models, using the genlogit approach, were run for quality of health care. Significance testing was set at alpha =0.05. For each outcome, separate models were run by domain as a step toward building the final, overall model. A p-value of 0.3 was used as the criterion for entry and retention in the models. For domain-specific and full models, we utilized a manual backward elimination approach. The sample sizes for the final access and quality models were 1,333 and 1,250, respectively. The subpopulations in these final models were similar to the eligible population of 4–17 year olds (n = 1,397) based on the distribution of the demographic variables. Multiply imputed household income data, available from the NSCH [35], were used when household income was not reported by the respondent (11.5 % of our study population). Analysis was conducted using SUDAAN 10.0.1 to account for the complex survey design and obtain appropriate variance estimates. This study was approved by the Institutional Review Board of the Georgia Department of Public Health.

Results

Study Population

Of Georgia children ages 4–17 years in 2007, 80.8 % were reported to have access to care. Parental perception of quality of health care received was as follows: 39.4 % received higher quality care, 30.6 % received moderate quality care, and 30.0 % received lower quality care (Table 1). The prevalence of most study variables in Georgia was comparable to that of the US with a few exceptions. Compared to the US, Georgia had a smaller proportion of children living in neighborhoods with paths, parks, rundown housing, and the presence of vandalism; a smaller proportion of non-Hispanic Whites, and Hispanics; a larger proportion of non-Hispanic Blacks; a smaller proportion of children living in families with non-English primary household language, and living in counties with 2 or more FQHCs/RHCs; and a much larger proportion of children living in HPSAs affecting whole counties (Table 1).
Table 1

Characteristics of the study population of children aged 4–17 years in Georgia (n = 1,397), and in the United States (73, 348), National Survey of Children’s Health, 2007

Characteristics

GA weighted prevalence (95 % CI)

US weighted prevalence (95 % CI)

Outcomes

Access to health carea

 Yes

80.8 (77.3–83.8)

80.2 (79.5–81.0)

Quality of health careb

 Higher quality care

39.4 (35.5–43.5)

37.6 (36.7–38.5)

 Moderate quality care

30.6 (27.0–34.5)

27.7 (26.9–28.5)

 Lower quality care

30.0 (26.2–34.0)

34.7 (33.8–35.7)

Independent variables

External environment

 Paths (Yes)*

58.4 (54.3–62.4)

72.4 (71.6–73.2)

 Park (Yes)*

68.1 (64.1–71.9)

79.9 (79.2–80.6)

 Recreation Center (Yes)

65.6 (61.6–69.4)

64.9 (64.0–65.7)

 Library (Yes)

82.9 (79.4–85.8)

86.2 (85.6–86.9)

 Litter (Yes)

16.4 (13.3–20.1)

16.5 (15.8–17.2)

 Rundown Housing (Yes)*

10.5 (8.1–13.5)

14.3 (13.6–14.9)

 Vandalism (Yes)*

5.4 (3.8–7.7)

11.3 (10.7–12.0)

Predisposing characteristics

Age of Child (in years)

 4–9

45.8 (41.7–49.9)

42.1 (41.2–43.0)

 10–13

26.1 (22.6–29.9)

28.2 (27.4–29.0)

 14–17

28.2 (24.8–31.8)

29.8 (28.9–30.6)

Gender

 Male

51.4 (47.3–55.4)

51.2 (50.3–52.1)

 Female

48.7 (44.6–52.7)

48.8 (47.9–49.7)

Total number of adults in the household

 1

15.3 (12.1–19.0)

13.3 (12.7–13.9)

 2

65.0 (60.9–68.9)

66.5 (65.6–67.4)

 3+

19.7 (16.9–23.0)

20.2 (19.5–21.0)

Total number of children in the household

 1

22.7 (20.0–25.6)

21.4 (20.8–22.1)

 2

38.9 (35.1–42.9)

39.8 (38.9–40.7)

 3+

38.4 (34.2–42.8)

38.8 (37.9–39.8)

Highest educational level in the household

 Less than high school

10.7 (8.0–14.2)

9.3 (8.7–9.9)

 High school graduate

25.6 (21.7–29.9)

23.7 (22.9–24.6)

 More than high school

63.7 (59.2–67.9)

67.0 (66.1–68.0)

Child’s race/ethnicity

 Non-Hispanic White*

48.6 (44.5–52.7)

56.5 (55.6–57.5)

 Non-Hispanic Black*

34.8 (30.7–39.2)

14.8 (14.2–15.5)

 Non-Hispanic Other

6.9 (5.3–8.9)

8.7 (8.1–9.3)

 Hispanic*

9.8 (7.3–12.9)

20.0 (19.0–20.9)

Any household member employed for at least 50 weeks out of 52 weeks (yes)

85.8 (82.2–88.8)

88.1 (87.4–88.7)

Family structure type

 Two parent biological/adopted

59.9 (55.7–64.0)

64.1 (63.2–64.9)

 Two parent step family

10.6 (8.1–13.7)

9.5 (8.9–10.0)

 Single mother, no father present

21.9 (18.4–25.9)

19.8 (19.0–20.5)

 Other

7.6 (5.8–9.9)

6.7 (6.3–7.2)

Immigrant family type

 Foreign-born child

5.4 (3.6–8.0)

5.3 (4.8–5.9)

 US-born child with 2 foreign-born parents

9.2 (6.7–12.4)

11.7 (10.8–12.7)

 US-born child with 1 foreign-born parent

6.9 (4.9–9.6)

8.5 (7.8–9.2)

 Nonimmigrant Family

78.6 (74.5–82.2)

74.5 (73.4–75.6)

Immigrant family type-2

 Foreign born child

5.4 (3.6–8.0)

5.3 (4.8–5.9)

 Non-foreign born child

94.6 (92.0–96.4)

94.7 (94.1–95.2)

Primary household language

 English*

91.9 (89.2–94.1)

87.8 (87.0–88.5)

 Language other than English*

8.1 (6.0–10.9)

12.2 (11.5–13.0)

Length of stay of mother in the US (in years)

 <10

6.1 (4.2–8.9)

5.3 (4.8–5.9)

 10–19*

4.8 (3.3–7.0)

8.0 (7.3–8.7)

 20+

4.5 (3.0–6.7)

6.5 (5.9–7.1)

 Born in the US

84.5 (81.1–87.5)

80.2 (79.3–81.1)

Social support

 Strong support

36.9 (33.1–41.0)

35.2 (34.4–36.1)

 Moderate support

43.3 (39.2–47.4)

44.4 (43.5–45.4)

 Weak support

19.8 (16.6–23.4)

20.4 (19.6–21.2)

Social capital index

 4 (Highest)

36.9 (33.1–41.0)

35.2 (34.4–36.1)

 5–7

37.1 (33.2–41.3)

37.6 (36.7–38.5)

 8–10

17.4 (14.4–20.7)

18.9 (18.1–19.7)

 11+ (lowest)

8.6 (6.4–11.5)

8.3 (7.7–8.9)

Enabling resources

Number of federally qualified health centers—2007

 0*

49.7 (45.6–53.7)

25.2 (24.6–25.9)

 1

12.2 (9.9–14.9)

12.7 (12.2–13.3)

 2+*

38.2 (34.4–42.1)

62.1 (61.3–62.8)

Number of rural health clinics—2007

 0*

79.5 (76.0–82.7)

74.1 (73.2–75.0)

 1*

15.8 (13.0–19.0)

10.2 (9.7–10.8)

 2+*

4.7 (3.1–6.9)

15.7 (14.9–16.5)

Health professional shortage areas (for primary care)—2007

 The whole county*

47.0 (42.9–51.1)

16.8 (16.3–17.4)

 One or more parts*

13.0 (10.2–16.5)

45.6 (44.7–46.5)

 None of the county

40.0 (36.2–43.9)

37.6 (36.7–38.4)

Medically underserved areas (yes)c

87.5 (84.3–90.2)

n/a

Rural–urban designation (rural urban continuum code)

 Rural county

18.9 (15.8–22.4)

15.5 (15.0–16.0)

 Urban county

81.1 (77.6–84.2)

84.5 (84.0–85.0)

Child’s insurance coverage

 Never insured/intermittently insured

17.8 (14.6–21.6)

15.2 (14.5–16.0)

 Continuous and inadequate private*

12.7 (10.5–15.2)

16.0 (15.4–16.7)

 Continuous and inadequate public

3.4 (2.1–5.4)

4.4 (4.0–4.9)

 Continuous and adequate private

42.7 (38.8–46.7)

45.0 (44.1–45.9)

 Continuous and adequate public

23.3 (19.6–27.5)

19.4 (18.7–20.2)

Household federal poverty level (%)d

 At or below 100

19.7 (16.1–24.0)

17.6 (16.8–18.4)

 >100–200

22.6 (19.0–26.7)

21.1 (20.2–21.9)

 >200–300

18.4 (15.3–21.9)

17.7 (17.0–18.5)

 >300–400

11.4 (9.4–13.6)

13.7 (13.1–14.4)

 Above 400

28.0 (24.8–31.4)

29.9 (29.1–30.7)

Child has a usual source of care (yes)

93.8 (91.1–95.7)

92.9 (92.3–93.4)

Child has a personal doctor or nurse (yes)

89.8 (86.9–92.1)

91.6 (91.1–92.2)

Need variables

Child with special health care needs (yes)

22.3 (19.1–25.9)

21.9 (21.2–22.7)

Child’s health status

  

 Excellent

57.9 (53.7–62.0)

59.7 (58.7–60.6)

 Very good

27.8 (24.0–31.9)

23.6 (22.8–24.3)

 Good/fair/poor

14.3 (11.5–17.8)

16.8 (16.0–17.6)

Source: National Center for Health Statistics and Maternal and Child Health Bureau, National Survey of Children’s Health, 2007, the Area Resource File, 2008, and the medically underserved area variable

* Significant at α = .05 level

aAccess to health care was created from two variables—utilization of preventive medical care in the previous 12 months and delay or denial of needed care in the previous 12 months. Those with access to health care were those who had made at least 1 preventive medical visit in the previous 12 months and who had experienced no delay or denial of needed care in the previous 12 months

bQuality of health care received was derived from five questions related to how compassionate, culturally-effective and family-centered the care was

cThe medically underserved area variable was downloaded/manually created for Georgia only, from the Health Resources and Services Administration Web site at http://muafind.hrsa.gov/

dThe estimates for Federal Poverty Level (FPL) were derived from analyzing the 5 multiply-imputed FPL variables using the MI_VAR statement in SUDAAN

Access to Health Care

In bivariable analysis, variables associated with access to care (p < 0.05) by domain include, predisposing: child’s age, race/ethnicity, immigrant family type (2-levels); and enabling: insurance coverage and having a personal doctor/nurse (Table 2).
Table 2

Factors associated with access to health carea and the quality of health careb received, among Georgia children aged 4–17 years, National Survey of Children’s Health, 2007 (n = 1,397)

Characteristics

Access to health care

p

Quality of health care

p

Yes (%)

Higher quality (%)

Moderate quality (%)

Lower quality (%)

95 % CI

95 % CI

95 % CI

95 % CI

Independent variables

External environment

Paths

 No

77.2 (71.3–82.2)

0.08

36.7 (30.7–43.1)

31.6 (26.0–37.9)

31.7 (25.9–38.2)

0.53

 Yes

83.3 (79.0–86.8)

 

41.4 (36.2–46.8)

29.7 (25.1–34.8)

28.9 (24.0–34.3)

 

Park

 No

79.7 (73.2–84.9)

0.68

34.0 (27.3–41.5)

30.1 (23.8–37.2)

35.9 (28.7–43.7)

0.12

 Yes

81.2 (77.0–84.7)

 

42.0 (37.2–46.9)

30.7 (26.3–35.4)

27.4 (23.1–32.1)

 

Recreation center

 No

78.1 (71.5–83.6)

0.23

35.2 (29.1–42.0)

28.5 (22.8–34.9)

36.3 (29.4–43.8)

0.10

 Yes

82.4 (78.4–85.9)

 

41.5 (36.5–46.7)

31.7 (27.1–36.8)

26.8 (22.4–31.7)

 

Library

 No

79.2 (68.5–87.0)

0.74

35.7 (26.7–45.8)

23.1 (15.9–32.3)

41.2 (31.0–52.3)

0.07

 Yes

80.9 (77.3–84.1)

 

40.2 (35.8–44.7)

32.1 (28.0–36.4)

27.8 (23.8–32.1)

 

Litter

 No

80.9 (77.2–84.1)

0.83

39.3 (35.2–43.6)

31.4 (27.5–35.7)

29.3 (25.3–33.6)

0.53

 Yes

79.9 (69.4–87.4)

 

40.9 (29.6–53.2)

25.6 (17.3–36.2)

33.5 (23.4–45.4)

 

Rundown housing

 No

80.8 (77.1–84.0)

0.83

39.9 (35.7–44.2)

31.8 (27.9–35.9)

28.4 (24.5–32.5)

0.08

 Yes

79.7 (68.4–87.6)

 

36.3 (24.6–49.8)

19.7 (11.2–32.2)

44.1 (30.7–58.4)

 

Vandalism

 No

8 0.9 (77.3–84.0)

0.65

40.4 (36.3–44.6)

31.2 (27.4–35.2)

28.5 (24.7–32.6)

0.05

 Yes

77.5 (60.4–88.6)

 

24.1 (13.7–38.9)

19.7 (9.4–36.9)

56.2 (38.7–72.2)

 

Predisposing characteristics

Age of child (in years)

 4–9*

85.3 (80.4–89.2)

0.03

42.2 (36.2–48.4)

30.7 (25.3–36.7)

27.2 (21.9–33.2)

0.51

 10–13

76.6 (68.6–83.1)

 

39.2 (31.3–47.6)

28.0 (21.2–35.9)

32.9 (25.5–41.3)

 

 14–17

77.2 (71.0–82.3)

 

35.3 (28.9–42.2)

32.9 (26.5–39.9)

31.9 (25.0–39.6)

 

Gender

 Male

80.3 (75.4–84.4)

0.75

39.8 (34.3–45.7)

27.6 (22.7–33.0)

32.6 (27.3–38.4)

0.21

 Female

81.3 (76.3–85.5)

 

39.1 (33.6–44.8)

33.8 (28.6–39.5)

27.1 (22.0–33.0)

 

Total number of adults in the household

 1

80.0 (68.9–87.8)

0.99

34.3 (23.7–46.8)

24.8 (16.1–36.2)

40.9 (29.1–53.8)

0.23

 2

80.9 (76.7–84.5)

 

39.6 (34.9–44.6)

33.2 (28.7–38.1)

27.1 (22.8–32.0)

 

 3+

80.7 (72.7–86.8)

 

43.1 (34.8–51.8)

26.5 (19.7–34.7)

30.5 (23.5–38.5)

 

Total number of children in the household

 1

82.5 (78.0–86.2)

0.72

37.9 (32.1–44.0)

28.3 (23.2–34.1)

33.8 (27.8–40.5)

0.49

 2

81.0 (75.6–85.5)

 

36.5 (30.9–42.5)

33.2 (27.7–39.2)

30.3 (24.7–36.6)

 

 3+

79.5 (72.5–85.0)

 

43.6 (35.8–51.6)

29.2 (22.5–36.8)

27.3 (20.5–35.2)

 

Highest educational level in the household

 Less than high school*

75.4 (59.2–86.6)

0.72

42.3 (27.1–59.2)

15.1 (7.1–29.3)

42.6 (28.0–58.5)

<0.01

 High school graduate

82.0 (73.3–88.3)

 

34.9 (26.0–45.0)

25.9 (18.1–35.5)

39.2 (29.7–49.7)

 

 More than high school

80.9 (77.3–84.1)

 

41.1 (36.9–45.4)

34.7 (30.6–39.1)

24.2 (20.6–28.2)

 

Child’s race/ethnicity

 Non-Hispanic White*

79.2 (75.0–82.8)

0.02

47.7 (42.7–52.8)

34.1 (29.4–39.1)

18.2 (14.6–22.4)

<0.01

 Non-Hispanic Black

87.5 (81.4–91.8)

 

34.8 (27.4–43.0)

23.7 (17.6–31.0)

41.5 (33.6–49.9)

 

 Non-Hispanic Other

81.2 (68.8–89.5)

 

27.5 (17.1–41.1)

41.0 (28.2–55.1)

31.5 (20.2–45.6)

 

 Hispanic

62.9 (46.2–77.0)

 

18.9 (9.3–34.6)

30.8 (17.2–48.9)

50.3 (34.5–66.0)

 

Any household member employed for at least 50 weeks out of 52 weeks

 No

79.1 (67.7–87.3)

0.75 

39.6 (27.7–53.0)

21.1 (11.6–35.2)

39.3 (27.8–52.1)

0.13

 Yes

80.8 (77.1–84.0)

 

39.5 (35.4–43.7)

32.1 (28.3–36.2)

28.4 (24.5–32.8)

 

Family structure type

 Two parent biological/adopted

81.7 (77.6–85.2)

0.63 

39.8 (35.2–44.7)

32.8 (28.2–37.6)

27.4 (23.1–32.3)

0.55

 Two parent step family

80.3 (67.0–89.1)

 

44.4 (30.8–58.8)

28.9 (18.2–42.7)

26.7 (15.9–41.2)

 

 Single mother—no father present

76.6 (67.6–83.7)

 

35.0 (26.2–45.0)

26.7 (19.2–35.9)

38.3 (28.9–48.6)

 

 Other

84.7 (70.0–93.0)

 

44.4 (30.7–59.0)

25.2 (14.3–40.5)

30.4 (19.8–43.7)

 

Immigrant family type

 Foreign-born child*

55.9 (34.9–74.9)

0.20

24.7 (11.2–46.3)

18.5 (8.1–36.9)

56.7 (35.1–76.1)

<0.01

 US-born child with 2 foreign-born parents

80.5 (63.4–90.7)

 

16.7 (8.6–29.8)

38.2 (23.2–55.9)

45.1 (29.8–61.3)

 

 US-born child with 1 foreign-born parent

81.9 (61.8–92.7)

 

34.7 (20.4–52.5)

43.6 (27.3–61.4)

21.8 (11.8–36.7)

 

 Nonimmigrant family

83.7 (79.8–87.0)

 

44.8 (39.6–50.1)

31.5 (26.9–36.5)

23.7 (19.2–28.8)

 

Immigrant family type-2

 Foreign born child*

55.9 (34.9–74.9)

0.04

24.7 (11.2–46.3)

18.5 (8.1–36.9)

56.7 (35.1–76.1)

0.11

 Non-foreign born child

83.3 (79.6–86.4)

 

41.5 (36.8–46.3)

33.0 (28.6–37.7)

25.5 (21.4–30.2)

 

Primary household language 

 English*

82.6 (79.4–85.5)

0.02

41.6 (37.5–45.9)

30.5 (26.8–34.5)

27.8 (24.0–32.0)

<0.01

 Language other than English

59.4 (42.6–74.2)

 

7.0 (2.7–16.9)

31.3 (17.2–49.8)

61.7 (44.2–76.6)

 

Length of stay of mother in the US (in years)

 <10*

57.0 (36.9–75.1)

0.19

13.1 (4.9–30.9)

20.1 (7.5–43.8)

66.8 (45.0–83.1)

<0.01

 10–19

76.0 (54.7–89.2)

 

8.9 (3.6–20.3)

39.3 (22.0–59.7)

51.8 (33.1–70.1)

 

 20+

76.6 (53.3–90.4)

 

27.9 (15.1–45.6)

30.3 (15.2–51.3)

41.9 (23.4–62.9)

 

 Born in the US

82.3 (78.7–85.3)

 

42.8 (38.3–47.5)

30.9 (26.8–35.2)

26.3 (22.2–30.9)

 

Social support

 Strong support*

80.4 (74.4–85.2)

0.09

52.8 (46.4–59.2)

26.8 (21.4–33.0)

20.3 (15.9–25.6)

<0.01

 Moderate support

84.0 (79.1–88.0)

 

36.9 (30.9–43.4)

34.8 (29.1–41.0)

28.3 (22.3–35.2)

 

 Weak support

72.8 (63.1–80.8)

 

19.2 (13.6–26.4)

27.6 (19.3–37.8)

53.2 (43.3–62.9)

 

Social capital index

 4 (highest)*

80.4 (74.4–85.2)

0.30

52.8 (46.4–59.2)

26.8 (21.4–33.0)

20.3 (15.9–25.6)

<0.01

 5–7

83.5 (78.1–87.8)

 

40.0 (33.2–47.2)

31.6 (25.8–38.0)

28.5 (22.1–35.9)

 

 8–10

80.4 (71.2–87.1)

 

21.9 (15.9–29.3)

39.4 (29.8–49.9)

38.7 (29.1–49.3)

 

 11+ (lowest)

68.0 (51.6–80.9)

 

12.6 (6.8–22.1)

23.3 (12.6–38.9)

64.1 (48.8–77.0)

 

Enabling resources

Number of federally qualified health centers—2007

 0

83.0 (77.9–87.0)

0.09

41.4 (35.5–47.7)

30.2 (24.8–36.2)

28.4 (23.0–34.5)

0.43

 1

70.2 (59.3–79.3)

 

44.7 (33.9–56.0)

30.2 (20.6–41.9)

25.1 (16.9–35.6)

 

 2+

81.2 (75.7–85.7)

 

35.2 (29.7–41.2)

31.2 (26.1–36.9)

33.6 (27.7–40.0)

 

Number of rural health clinics—2007

 0

80.7 (76.8–84.1)

0.98

38.4 (34.1–42.9)

30.4 (26.5–34.8)

31.1 (26.8–35.8)

0.38

 1

81.3 (72.2–87.9)

 

46.5 (36.1–57.2)

31.3 (22.1–42.2)

22.2 (15.3–31.1)

 

 2+

79.8 (64.5–89.6)

 

32.9 (16.5–55.0)

30.7 (16.4–49.9)

36.4 (19.3–57.8)

 

Health professional shortage areas (for primary care)—2007

 The whole county

78.6 (73.1–83.2)

0.37

39.7 (34.0–45.7)

30.1 (24.7–36.1)

30.2 (24.6–36.4)

0.31

 One or more parts

80.7 (69.5–88.4)

 

49.0 (35.9–62.3)

21.9 (13.7–33.1)

29.1 (18.5–42.6)

 

 None of the county

83.4 (78.4–87.4)

 

35.9 (30.6–41.6)

34.0 (28.7–39.9)

30.0 (24.7–36.0)

 

Medically underserved areasc

 No

79.9 (68.9–87.7)

0.85

39.9 (28.4–52.7)

21.7 (12.9–34.0)

38.4 (26.3–52.1)

0.19

 Yes

80.9 (77.2–84.1)

 

39.4 (35.2–43.7)

31.9 (28.0–36.1)

28.7 (24.9–32.9)

 

Rural–urban designation (rural urban continuum code)

 Urban county

81.0 (77.2–84.3)

0.72

41.3 (31.9–51.3)

25.4 (18.1–34.5)

33.3 (24.3–43.7)

0.41

 Rural county

79.5 (71.0–86.0)

 

39.0 (34.7–43.5)

31.8 (27.7–36.1)

29.2 (25.2–33.7)

 

Child’s insurance coverage

 Never insured/intermittently insured*

62.7 (51.1–73.0)

<0.01

32.0 (21.8–44.1)

24.2 (16.0–34.8)

43.9 (33.1–55.3)

<0.01

 Continuous and inadequate private

73.9 (63.6–82.1)

 

32.9 (24.7–42.2)

34.8 (25.9–45.0)

32.3 (24.1–41.7)

 

 Continuous and inadequate public

86.6 (68.5–95.1)

 

31.2 (12.6–59.0)

10.8 (3.9–26.8)

57.9 (33.8–78.8)

 

 Continuous and adequate private

84.8 (80.9–88.0)

 

48.1 (42.6–53.7)

35.1 (29.9–40.6)

16.8 (12.9–21.6)

 

 Continuous and adequate public

90.9 (84.8–94.7)

 

35.4 (26.6–45.4)

27.1 (19.2–36.8)

37.5 (28.1–48.0)

 

Household federal poverty level (%)d

 At or below 100*

77.0 (66.4–85.1)

0.32

44.0 (33.0–55.6)

19.0 (11.3–30.2)

37.0 (27.1–48.1)

<0.01

 >100–200

75.9 (66.5–83.4)

 

32.1 (23.6–41.9)

27.3 (19.4–36.9)

40.7 (30.8–51.4)

 

 >200–300

83.4 (75.1–89.4)

 

30.2 (21.7–40.4)

37.9 (28.5–48.3)

31.9 (22.8–42.6)

 

 >300–400

85.6 (77.7–91.0)

 

44.3 (35.1–54.0)

37.2 (28.4–46.9)

18.5 (12.2–27.0)

 

 Above 400

83.5 (78.6–87.5)

 

46.0 (40.2–52.0)

33.6 (27.8–39.9)

20.4 (15.6–26.1)

 

Child has a usual source of care

 No

64.1 (44.7–79.7)

0.08 

29.1 (13.8–51.2)

18.3 (6.0–44.0)

52.6 (32.0–72.3)

0.11

 Yes

81.9 (78.5–84.8)

 

40.0 (36.0–44.2)

31.2 (27.5–35.2)

28.8 (25.0–32.9)

 

Child has a personal doctor or nurse

 No*

64.4 (50.0–76.5)

0.02 

37.7 (24.4–53.2)

18.0 (9.9–30.5)

44.4 (30.6–59.0)

0.04

 Yes

82.6 (79.2–85.5)

 

39.6 (35.5–43.8)

31.8 (27.9–35.9)

28.7 (24.8–32.9)

 

Need variables

Child with special health care needs

 No*

79.9 (75.9–83.4)

0.29 

38.9 (34.5–43.6)

28.4 (24.5–32.7)

32.7 (28.3–37.5)

0.02

 Yes

83.7 (77.1–88.6)

 

41.2 (33.1–49.9)

37.9 (29.9–46.7)

20.8 (14.9–28.3)

 

Child’s health status

 Excellent*

82.0 (77.9–85.5)

0.54 

45.4 (40.6–50.3)

29.4 (25.2–34.0)

25.2 (21.1–29.8)

<0.01

 Very good

80.8 (73.5–86.5)

 

30.4 (23.1–38.9)

38.0 (30.0–46.7)

31.6 (23.9–40.6)

 

 Good/fair/poor

75.6 (63.5–84.6)

 

32.8 (22.4–45.3)

20.4 (12.5–31.7)

46.8 (34.9–59.0)

 

Source: National Center for Health Statistics and Maternal and Child Health Bureau, National Survey of Children’s Health, 2007, the Area Resource File, 2008, and the medically underserved area variable

* Significant at α = .05 level

aAccess to health care was created from two variables—utilization of preventive medical care in the previous 12 months and delay or denial of needed care in the previous 12 months. Those with access to health care were those who had made at least 1 preventive medical visit in the previous 12 months and who had experienced no delay or denial of needed care in the previous 12 months

bQuality of health care received was derived from five questions related to how compassionate, culturally-effective and family-centered the care was

cThe medically underserved area variable for Georgia was downloaded from the Health Resources and Services Administration Web site at http://muafind.hrsa.gov/

dThe estimates for Federal Poverty Level (FPL) were derived from analyzing the 5 multiply-imputed FPL variables using the MI_VAR statement in SUDAAN

In full models (with all domains combined), the significant factors were, predisposing: child’s age; and enabling: insurance coverage (Table 3). Children ages 4–9 years had almost twice the odds of having access to care compared to children ages 14–17 years. Children who were never/intermittently insured (AOR 0.27; 95 % CI 0.16–0.47) and those with continuous and inadequate private insurance (AOR 0.50; 95 % CI 0.28–0.92) had lower odds of having access to care, compared to children who had continuous and adequate private insurance. Those with continuous and inadequate public insurance (AOR 0.85; 95 % CI 0.27–2.65) and continuous and adequate public insurance (AOR 1.46; 95 % CI 0.74–2.87) did not differ from children with continuous and adequate private insurance in having access to care (Table 3).
Table 3

Adjusted odds ratios, association of environmental, predisposing, enabling, and need domain variables with access to health carea and the quality of health careb received among Georgia children aged 4–17 years, National Survey of Children’s Health, 2007

Characteristics

Access to health care

Quality of health care

Yes versus no

Higher quality care versus lower quality care

Moderate quality care versus lower quality care

AOR (95 % CI)

AOR (95 % CI)

AOR (95 % CI)

Independent variables

External environment

Park

 No

Reference

Reference

 Yes

1.61 (0.99–2.60)

1.23 (0.74–2.05)

Vandalism

 No

2.68 (1.01–7.13)*

2.95 (1.05–8.29)*

 Yes

Reference

Reference

Predisposing characteristics

Age of child (in years)

 4–9

1.94 (1.21–3.12)*

 10–13

1.16 (0.67–2.00)

 14–17

Reference

Gender

 Male

Reference

Reference

 Female

1.42 (0.91–2.23)

1.67 (1.05–2.65)*

Child’s race/ethnicity

 Non-Hispanic White

Reference

Reference

Reference

 Non-Hispanic Black

1.78 (0.97–3.26)

0.38 (0.23–0.63)*

0.38 (0.21–0.67)*

 Non-Hispanic Other

1.11 (0.52–2.36)

0.34 (0.16–0.75)*

0.78 (0.39–1.59)

 Hispanic

0.67 (0.32–1.38)

0.15 (0.05–0.41)*

0.40 (0.16–0.99)*

Social support

 Strong support

4.71 (2.50–8.87)*

1.85 (0.95–3.59)

 Moderate support

2.61 (1.42–4.81)*

1.96 (1.03–3.72)*

 Weak support

Reference

Reference

Enabling resources

Number of federally qualified health centers

 0

Reference

 1

0.57 (0.31–1.07)

 2+

1.00 (0.62–1.60)

Child’s insurance coverage

 Never insured/intermittently insured

0.27 (0.16–0.47)*

0.33 (0.16–0.68)*

0.36 (0.17–0.76)*

 Continuous and inadequate private

0.50 (0.28–0.92)*

0.43 (0.22–0.85)*

0.60 (0.29–1.23)

 Continuous and inadequate public

0.85 (0.27–2.65)

0.23 (0.07–0.81)*

0.14 (0.04–0.53)*

 Continuous and adequate private

Reference

Reference

Reference

 Continuous and adequate public

1.46 (0.74–2.87)

0.36 (0.17–0.76)*

0.43 (0.21–0.88)*

Household federal poverty level (%)c

 At or below 100

Reference

Reference

 >100–200

0.33 (0.14–0.77)*

0.75 (0.29–1.94)

 >200–300

0.30 (0.13–0.70)*

0.95 (0.37–2.42)

 >300–400

0.45 (0.17–1.16)

0.99 (0.36–2.69)

 Above 400

0.28 (0.12–0.67)*

0.73 (0.29–1.83)

Need

Child with special health care needs

 No

Reference

Reference

 Yes

2.62 (1.41–4.88)*

2.76 (1.47–5.18)*

Child’s health status

 Excellent

2.54 (1.23–5.25)*

2.44 (1.02–5.81)*

 Very good

1.42 (0.64–3.12)

2.62 (1.05–6.50)*

 Good/fair/poor

Reference

Reference

Full Model: All variables from the 4 domains were entered into the model simultaneously

Source: National Center for Health Statistics and Maternal and Child Health Bureau, National Survey of Children’s Health, 2007, the Area Resource File, 2008, and the medically underserved area variable

* Significant at α = .05 level

aAccess to health care was created from two variables—utilization of preventive medical care in the previous 12 months and delay or denial of needed care in the previous 12 months. Those with access to health care were those who had made at least 1 preventive medical visit in the previous 12 months and who had experienced no delay or denial of needed care in the previous 12 months. The full model for access to health care had a sample size of 1,333. The variables in the final access model as shown in the table are: age of child, race/ethnicity, number of federally qualified health centers and insurance coverage

bQuality of health care received was derived from five questions related to how compassionate, culturally-effective and family-centered the care was. The full model for the quality of health care received had a sample size of 1,250. The variables in the final quality model as shown in the table are: park, vandalism, gender, race/ethnicity, social support, insurance coverage, FPL, CSHCN status, and child’s health status

cThe estimates for Federal Poverty Level (FPL) were derived from analyzing the 5 multiply-imputed FPL variables using the MI_VAR statement in SUDAAN

Quality of Health Care

In bivariable analysis, variables associated with quality of care (p < 0.05) by domain include, predisposing: parental highest educational level, race/ethnicity, immigrant family type (4-levels), social support and social capital index; enabling: insurance coverage, household poverty status, having a personal doctor/nurse; and need: CSHCN status, and child’s overall health status (Table 2).

In full models, the significant associations by domain were, environmental: among children who lived in neighborhoods without signs of vandalism, there were greater odds of perceiving received care as higher (AOR 2.68; 95 % CI 1.01–7.13) or moderate (AOR 2.95; 95 % CI 1.05–8.29) versus lower quality when compared to those living in neighborhoods with vandalism (Table 3); predisposing: among the minority race/ethnicity groups, there were lower odds of perceiving received care as higher/moderate versus lower quality when compared to non-Hispanic Whites; enabling: compared to children with continuous and adequate private insurance, there were lower odds of perceiving received care as higher/moderate versus lower quality among children who were never/intermittently insured or who had continuous and inadequate/adequate public insurance (Table 3); need: special health care needs status was associated with higher odds of perceiving received care as higher/moderate versus lower quality.

Discussion

The majority of Georgia children ages 4–17 years were reported to have access to care; however, the quality of care received was almost equally distributed among the three categories. We found differences between the factors associated with access to care and quality of care. Uniquely associated with access to care was younger age (4–9 years). Many more factors, across all domains, were uniquely associated with quality of care. The single factor associated with both access to care and quality of care was insurance coverage.

Unlike other studies [12, 13], race/ethnicity was not associated with having access to care. This difference may be due to the referenced studies being based on national datasets and the identified difference between the racial/ethnic compositions of Georgia and US child populations. Younger age was associated with greater odds of having access to care. Parents may be more motivated to take younger children for their well-child visits in order to get all the scheduled vaccines. It may also be that parents of older children take them for well-child visits based on their convenience, but not necessarily within 12 months after the previous visit. There is evidence that parents can be encouraged to take their children for well-child visits and screenings at least yearly, by setting up reminder/recall scheduling systems [36] around the birth date of the child. Appointment delays for child preventive care visits can be as long as 4–5 weeks, and so in addition to using reminder/recall systems, an open-access scheduling system, where demand for office visits is matched with the supply—offering patients same-day access to appointments regardless of the nature of their problem [36]—may also be beneficial.

Similar to the published literature [35, 8, 37], this study identified that Georgia children without insurance or who have coverage gaps lack access to care. Also similar to the published literature, this study identified that Georgia children with continuous and adequate public insurance were not different from their private counterparts in their odds of having access to care [7]. However, those with continuous and inadequate private insurance fared worse than their adequately insured counterparts. Modifying deductibles, co-payments, and benefit/provider restrictions by private insurance companies may provide opportunities for enhancing children’s access to preventive medical care.

PeachCare for Kids, Georgia’s Children’s Health Insurance Program (CHIP) extends coverage for children ages 0 to 19 years, up to 235 % of the FPL [38]. The 2010 Affordable Care Act extends federal funding for CHIP through September 30, 2015, providing states with additional funding to ensure more children have access to this program, and increasing outreach and enrollment grants to help reach more eligible children [39]. Recently, the National Healthy Mothers, Healthy Babies Coalition (HMHB) partnered with the Maternal and Child Health (MCH) Program, Georgia Division of Public Health, to submit an application for the CHIP Reauthorization Act outreach and enrollment grants. They propose to use text4baby, a free mobile information service designed to promote MCH, as a vehicle to increase enrollment and retention in Georgia’s Medicaid and PeachCare for Kids programs. It is expected that these changes will result in increasing the proportion of Georgia children with continuous insurance and access to preventive care.

This study identified disparities in the perception of the quality of care received by Georgia children. Minority race/ethnicity was associated with lower odds of perceiving the care received as higher/moderate versus lower quality. Among children who lacked insurance, had coverage gaps, or had continuous but inadequate coverage, there were lower odds of perceiving the care received as high quality. While these findings have been previously described from analysis of National data [8, 13, 17], the finding in Georgia is striking that even those with continuous and adequate public insurance fared worse than their privately-insured counterparts. Opportunities to address the identified disparities in quality of care include, increasing the number of adequately insured Georgia children; eliminating insurance gaps which may result from financial or administrative barriers when switching from private to public insurance or vice versa; and ensuring that all providers, including those reimbursed by the Georgia Medicaid and PeachCare for Kids programs, provide care that is compassionate, culturally-sensitive, and family-centered. The Georgia Association of Family Physicians is currently promoting the National Committee on Quality Assurance’s patient-centered medical home certification program [40], enrolling providers who are interested in getting certified and helping with needed training and education. In addition to certification, providers may benefit from re-evaluating their interactions with children and their families during office visits, setting expectations for office visits at the beginning of the encounter, so both sides know what to expect [41, 42].

This study is unique in that it uses the same data and analytic structure to examine two interdependent concepts, and adds contextual information from the ARF. This study is, however, not without limitations. The cross-sectional study design makes it impossible to establish the direction of causality. Health and health care data from the NSCH were based on parental perceptions. These data were not verified by health care professionals. Caution should therefore be exercised when interpreting these results. In addition, parental responses could be subject to reporting errors, considering that they were being asked about events and behaviors in the previous 12 months.

In conclusion, one of the most mutable associated factors that emerged from our study was insurance coverage for Georgia children. Assuring continuous, adequate insurance may positively affect both health care access and quality.

Acknowledgments

The authors gratefully acknowledge the staff of the Research Data Center, Stephanie Robinson and Alex Erhlich, for their support throughout the study period. We are also grateful to Arianne Weldon and Dr Seema Csukas at the Georgia Department of Community Health, Maternal and Child Health Program, for their insights into the program and policy implications of our findings for Georgia.

Copyright information

© Springer Science+Business Media, LLC 2012