Abstract
Objectives. We examined the relation between parents’ level of English proficiency and their children’s access to health care.
Methods. Using the 2001 California Health Interview Survey, we conducted bivariate and multivariate analyses of several measures of children’s access to health care (current health insurance status, usual source of care, emergency room visits, delayed or forgone care, traveling to another country for health care, and perceived discrimination in health care) and their association with parents’ English proficiency.
Results. Compared with English-speaking households, children in non–English-speaking households were more likely to lack health insurance, to not have doctor contact, and to go to other countries for health care and were less likely to use emergency rooms. Their parents were less likely to report their children’s experiencing delayed or forgone care or discrimination in health care.
Conclusion. English proficiency is a strong predictor of access to health insurance for children, and children in non–English-speaking families are especially likely to rely on other countries for their health care. English proficiency may mitigate the effects of race/ethnicity commonly observed in health care access and utilization studies.
The US Census Bureau estimated that 18% of the population aged 5 years and older spoke a language other than English at home, and more than 8% of people living in the United States speak English “less than ‘very well.’”1 Being non-English speaking has been recognized as an obstacle to receiving health care and can affect the health of specific minority populations, as well as the general population.2–4 Those with limited English proficiency report “lack of knowledge” as an access barrier to health services more frequently than do those who speak English well, and they are less likely to understand their diagnosis, medications, treatment, and follow-up instructions than more proficient English speakers.5–10 Studies have shown that, in various circumstances, non-English speakers are less likely to have a usual source of care; to receive preventive services, follow-up appointments, sufficient information, and adequate therapy; and to participate in medical decisionmaking.11–15
Studies of emergency room use have shown that patients with language barriers also may experience higher admission rates and more diagnostic studies, leading to higher costs for emergency services.16–19 In addition, language barriers have been associated with lacking components consistent with family-centered care, fewer physician visits, lower rates of patient satisfaction, and increased reports of quality of care being adversely affected.20–27 Language barriers have more deleterious effects in vulnerable populations, such as children with special health care needs, partly because of the complexity of some children’s needs and the heightened importance of clear communication between parents and health care providers.28 Recent studies on adolescents revealed significant psychosocial deficits in the school environment among those whose primary language at home is not English.29,30
To our knowledge, no previous study has examined levels of parental English proficiency and children’s access to health care. The 2001 California Health Interview Survey (CHIS) collected the information necessary to accomplish this analysis.31 With the assumption that the parent or guardian is the most knowledgeable person about the health and health care of the child and was also the person responsible for helping the child to access health care services, we examined the association between the parent respondent’s level of English proficiency and the child’s access to health care.
Considering the barriers faced by those who speak limited English in the general population in accessing and receiving health care, it was hypothesized that children whose parents have limited English proficiency would report lower rates of health care access than would children with English-speaking parents. Our goal was to describe the demographics of children and the prevalence of selected access characteristics by the English proficiency of the responding parent and to examine the independent effects of English proficiency on health care access while controlling for confounding variables.
The data used in this study were derived from the 2001 CHIS, with revised sample weights released in February 2005. The CHIS, the largest statewide health survey in the United States, collects data on multiple public health issues, including health status, behaviors, and access to care. CHIS 2001 is a random-digit-dialed telephone survey of adults, adolescents, and children. Detailed methodological information is available elsewhere.31
The sampling frame consisted of all California households with a telephone. Each computer-generated telephone number was screened to determine eligibility (e.g., language fluency); 1 adult per household was randomly selected to be interviewed. If the interviewed adult was the parent or guardian of a child or an adolescent residing in the household, the adult was asked to complete the child interview and to give verbal consent for the adolescent to be interviewed. Data were collected from 55 428 households; 12 592 parent proxies for child interviews were completed between November 2000 and October 2001. An additional 800 interviews from supplemental oversamples of ethnic subgroups were added to the most current version of the public use files, yielding a total of 12 797 children with complete data for our analysis. Approximately 12% of the adult interviews were completed in a language other than English, as were 21% of all child (parent proxy) interviews.32 One criterion for the adolescent and child to be selected for the survey was that they had to be “associated” with the selected adult. This meant that, in most cases, the interviewed adult had to be either the parent or guardian. Interviews were conducted in English, Spanish, Chinese (Mandarin and Cantonese dialects), Vietnamese, Korean, and Khmer.33 These languages were selected to include the largest possible number of non–English-speaking California residents.
If a respondent selected English as the language spoken at home, the person was classified as an English speaker at home. Those not speaking English at home were asked whether they spoke English very well, fairly well, or not well. If the interview was conducted in a language other than English, respondents were asked if they spoke English very well, well, not well, or not at all. The parent’s English proficiency was measured as a 4-level variable consisting of English speaker at home, speaks English very well, speaks English well, and speaks English not well or not at all.
We examined 7 measures of health care access. Insurance status was a dichotomous variable (uninsured vs some type of insurance) based on responses to the question on “type of current health coverage source—under 65 years old.” The types of insurance included Medicaid, CHIP, Medicare, employment-based, privately purchased, and other public insurance. Usual source of care was assessed from the following question: “Is there a place that the child would go when he/ she is sick or you need advice about his/her health?” Contact with the doctor was derived from the following question: “During the past 12 months, how many times has your child seen a medical doctor?” An assessment of emergency room visits was based on the following question: “During the past 12 months, did {the child} visit a hospital emergency room?” Delaying or forgoing care was based on the following question: “During the past 12 months, did you delay or not get a test or treatment that a doctor ordered?” Traveling to another country for health care or prescription drugs was derived from 2 questions: Respondents were classified as “Going to other country for medical, dental care, or prescription medicine” if they answered positively to either “During the past 12 months, did you take {the child} to another country, such as Mexico or any other country, for either medical or dental care?” or “During the past 12 months, did you or anyone else go to another country, such as Mexico or any other country, to buy any prescription medicine for {him or her}?” Discrimination in health care was determined by the following question: “Thinking of your experiences with receiving health care in the past 12 months for {the child}; have you felt that you were discriminated against for any reason?” All access measures were coded into dichotomized outcomes.
The covariates included child’s age, gender, citizenship status, area of residence, medical insurance, parent education, and family poverty level (FPL). The FPL variable was created from family income and family size using the US Census Bureau’s poverty thresholds.34
Sample weights, person-level weights, and population weights were used. These weights accounted for, among other variables, non-response, multiple telephone lines, and within-household probability of selection and adjusted for gender, age, race, ethnicity, urbanization, number of children, and number of adolescents in the household. The data were weighted based on the 2000 census data. Thus, our findings can be considered generalizable to children in California.
Chi-square statistics were used to test for differences in the proportion of negative health status indicators and health access and utilization factors among ethnic groups. Logistic regression analyses were used to examine the independent effects of English proficiency on various outcomes. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were computed by using the regression (β) coefficients and standard errors obtained from the logistic regression models.
To account for the complex sample design involving stratification, clustering, and multistage sampling of the CHIS, SUDAAN (Research Triangle Institute, Research Triangle Park, NC) was used to conduct the statistical analyses.35 Taylor series linearization methods were applied for variance estimation as recommended.
Table 1 shows the demographic distributions of the sample population by English proficiency. Among the respondents, there were 7233 (48.9%) English speakers at home, 1954 (16.3%) who spoke English very well, 1312 (12.2%) who spoke English well, and 2184 (22.6%) who spoke English not well or not at all. Significant associations were found for English proficiency and all sociodemographic characteristics examined (P < .05), with the exception of child gender. The children of those who reported speaking English at home tended to be older than the children of the other English proficiency groups. A higher level of English proficiency was related to higher parental education attainment and family income. Nearly 70% of respondents who reported not speaking English well or not speaking English at all had less than a high-school education. More than half of these individuals were poor (< 100% FPL). Nearly one fifth reported fair or poor health status for the child, compared with 3% of children from English-speaking households. This group also indicated a high proportion of Hispanic (91%) and noncitizen children (13.6%) compared with fewer than 0.5% from English-speaking households. More than one fifth reported that their children were uninsured, compared with 4% among English-speaking households. Nearly 80% of children from English-speaking households had private insurance, whereas only about one fifth of the “not well or not at all” group had private insurance. A higher proportion of English-speaking households resided in rural areas.
Table 2 shows the children’s insurance status and health care access and utilization patterns by parental English proficiency and other sociodemographic variables. Parental English proficiency was associated with children having insurance, contact with a doctor in the past 12 months, emergency room visits, going to another country for health care or prescription drugs, and experiencing delayed or forgone care. Less English proficiency was associated with higher level of uninsurance, lacking a doctor’s contact in the past 12 months, and a higher proportion of children seeking health care or prescription drugs. Parents who spoke less English reported less delayed or forgone care. Perception of discrimination in health care was associated only with race/ethnicity, parental education attainment, and insurance status.
Table 3 shows the adjusted odds ratios from logistic regressions for both health insurance status and health services indicators, controlling for parents’ English proficiency; children’s age, health status, citizenship status, insurance status, area of residence, and parental education; and FPL. Compared with English-speaking households, children of parents with limited English proficiency were significantly more likely to lack health insurance (not well or not at all: OR = 2.05, 95% CI = 1.44, 2.93). Children of non–English-speaking households were less likely to visit the emergency room within the past 12 months (well: OR = 0.68, 95% CI = 0.49, 0.93; not well or not at all: OR = 0.5, 95% CI = 0.35, 0.72) and less likely to report having delayed or forgone care (well: OR = 0.54, 95% CI = 0.35, 0.84; not well or not at all: OR = 0.49, 95% CI = 0.34, 0.71). However, children of non–English-speaking households were more likely to have gone to other countries for health care and medications (very well: OR = 6.39, 95% CI = 2.75, 14.88; well: OR = 15.32, 95% CI = 6.11, 38.42; not well or not at all: OR = 10.68, 95% CI = 3.89, 28.29). Parents who did not speak English well or did not speak English at all were less likely to report being discriminated against in health care (OR = 0.48, 95% CI = 0.26, 0.9).
Compared with non-Hispanic Whites, non-Hispanic Blacks and non-Hispanic Asians were less likely to lack health insurance. Asians were more likely to not have contact with a doctor in the past 12 months and were less likely to visit the emergency room and report delayed or forgone care. Hispanics were more likely to have gone to another country for health care and prescriptions. Non-Hispanic Blacks were less likely to report experiencing discrimination in health care. Compared with older children, children aged 0 to 4 years were less likely to lack health insurance, less likely to go without contact with a doctor in the past 12 months, and more likely to have had an emergency room visit in the past 12 months. Compared with children in fair or poor health, children in excellent, very good, or good health were less likely to have had emergency room visits, to lack insurance, to report delayed or forgone care, to go to another country for health care or prescriptions, and to report being discriminated against in health care. Noncitizens were much more likely to lack health insurance, a usual source of care, and contact with a doctor and to report being discriminated against in health care, compared with US-born children. Urban children were more likely to report being discriminated against in health care and less likely to use health care outside of the United States.
Family poverty had a strong impact on insurance coverage. Children in the poorest families (< 100 % FPL) were least likely to seek care outside of the United States. Children whose families earned from 200 to 299% of the FPL were more likely to experience discrimination in health care. Compared with college graduates, every level of parental education conferred higher risk of children lacking health insurance and not having contact with a doctor in the past 12 months. Children whose parents had some college education had a higher risk of visiting the emergency room in the past 12 months compared with those whose parents were college graduates. Compared with children with private insurance, publicly insured children were more likely to lack a usual source of care, more likely to have visited the emergency room in the past 12 months, more likely to report having delayed or forgone care, and more likely to report being discriminated against in health care. Children without insurance were 15 times more likely to lack a usual source of care and 2 times more likely to be without doctor contact in the past 12 months and reporting having delayed or forgone care. They were 4 times as likely to have gone to other countries for health care or prescription and 4 times more likely to report being discriminated against in health care.
Language barriers can present a significant problem in health care delivery. These problems are particularly significant for children, who depend on their parents to navigate the health care system on their behalf. In our study, 23% of children had a respondent parent who spoke English less than very well and would be considered linguistically isolated according to the census definition. Nonetheless, we found that in California, children of parents who do not speak English very well are no more likely than other children to lack a usual source of care or to not have seen a doctor within the past 12 months and are less likely to visit the emergency room or to delay or forgo needed care. However, they were significantly more likely to lack health insurance and more likely to go to other countries for health care.
Poverty, of course, is significantly related to health insurance as well, as are parental education and the citizenship status of the child. Race/ethnicity continues to play an important role in health care access. Similar to earlier findings from another study, Asians were less likely to have contact with the health care system and reported less forgone and delayed care.37 Blacks in this sample were less likely to lack insurance and to report being discriminated against in care. Hispanic children were more likely to have gone to another country for health care or prescriptions. Lacking medical insurance was a potent factor in predicting health care access in our study, demonstrating the effects of legal and employment circumstances that prevent children of immigrant families from having health insurance. The fact that people with limited English proficiency have a lower likelihood of emergency room visits suggests that lacking English proficiency may be a deterrent to necessary emergency visits and may highlight the need for translation services in emergency care.11,16 Interestingly, parents with limited English proficiency were less likely to report forgone or delayed care or to report experiencing discrimination in health care. This finding is consistent with other studies and could be partly attributable to lower health care expectations or sensitivities toward discrimination often expressed by more recent immigrants.28,36,37
We also demonstrated that English proficiency plays a stronger role in realized access (e.g., emergency room utilization and seeking care outside the United States) and a lesser role in potential access (e.g., insurance and usual source of care). An important finding of this study is the large percentage of children relying on health care outside of the United States within the limited English proficiency–speaking population. However, we were not able to differentiate between the use of health care, dental care, and prescription medicine outside of the United States because they were all asked within the same question. Of special concern are those people who do not have the economic resources to do so, as well as those undocumented immigrants who cannot leave the country to seek care because of visa concerns.
Our study has a number of potential limitations. The results of this survey are unlikely to be generalizable to the United States as a whole. Because California has the nation’s highest proportion of foreign-born residents, people in California with low levels of English proficiency may have access to a health care system that is more adapted to a linguistically diverse population. (The 2000 census reported that 20% of Californians speak English less than very well, compared with 8.1% in the United States overall.) Health care providers in other states may be less accustomed to serving these individuals, and our findings in California may underestimate the risks experienced by those with limited English proficiency nationally. We are also not sure how adequately the survey represents the needs of undocumented children, who are likely to have more severe needs. It has been estimated that approximately 10% of Californian children younger than 18 were undocumented or had at least 1 undocumented parent in 2000.38 In this community, limited English proficiency is likely to compound issues of health care access in addition to existing systemic factors of lacking health insurance and financial resources. People who have limited English proficiency have proven to be a challenge to the US health care system. It is a strong predictor of not having insurance and multiple health care access indicators. Language barriers may be the main cause of the lack of knowledge of health and community resources seen in immigrant populations.39
We also did not have access to provider characteristics, which could allow us to assess the effects of provider–patient language concordance. Our study highlights the importance of speaking English well in ensuring access to public benefits such as health insurance. Access to services can be facilitated by translating written notices and communications, providing professional interpreters, and allowing enrollment in settings other than welfare offices. Research shows that bilingual staff members are more likely to be available at community health centers and that immigrant families are more apt to apply for benefits at community clinics or other health-based settings.38 In addition, because a substantial amount of health and educational information is now communicated over the Internet, people with limited English proficiency are further disadvantaged from acquiring the same information as their English-speaking counterparts.
Further research on access to care among immigrant families40 and the role of health services received outside the United States is needed as well. The effects of cultures and health care experiences in the native countries of those with limited English proficiency on their utilization of health care in the United States should be explored. Studies on the relation of health status, health insurance, and utilization in the limited English proficiency–speaking population should be encouraged. We also should examine the public health implications of an untreated limited English proficiency–speaking population who can seek care neither in the United States nor outside of the country.
Note. All χ2 P values were < .05 except for child gender. The family poverty level variable was created from family income and family size using the US Census Bureau’s poverty thresholds. *P < .05
Non-English Speaker at Home English Speaker at Home (N = 7233, Weighted % = 48z.9%) % (SE) Very Well (N = 1954, Weighted % = 16.3%) % (SE) Well (N = 1312, Weighted % = 12.2%) % (SE) Not Well or Not At All (N = 2298, Weighted % = 22.6%) % (SE) Child race/ethnicity Non-Hispanic White 64.4 (0.64) 25.8 (1.20) 7.6 (0.96) 0.5 (0.18) Non-Hispanic Black 12.3 (0.34) 3.6 (0.64) 0.3 (0.17) 0.0 (0.01) Hispanic 13.4 (0.61) 48.6 (1.42) 63.8 (1.54) 91.0 (0.55) Non-Hispanic Asian 3.5 (0.33) 16.4 (0.95) 27.0 (1.38) 8.3 (0.52) Other 6.4 (0.38) 5.6 (0.69) 1.4 (0.45) 0.1 (0.07) Child age, y 0–4 37.0 (0.66) 42.1 (1.67) 42.0 (1.73) 41.1 (1.19) 5–11 63.0 (0.66) 57.9 (1.67) 58.0 (1.73) 58.9 (1.19) Child gender Male 51.1 (0.47) 51.3 (1.69) 49.3 (1.68) 51.9 (1.12) Female 48.9 (0.47) 48.7 (1.69) 50.7 (1.68) 48.1 (1.12) Child health status Fair/poor 3.0 (0.33) 2.7 (0.44) 6.4 (1.00) 18.6 (1.23) Excellent/very good/good 97.0 (0.33) 97.3 (0.44) 93.6 (1.00) 81.4 (1.23) Child citizenship US-born 99.0 (0.15) 95.7 (0.62) 94.4 (0.82) 85.5 (1.13) Naturalized citizen 0.6 (0.09) 1.2 (0.32) 1.0 (0.34) 1.0 (0.28) Noncitizen 0.4 (0.13) 3.1 (0.58) 4.6 (0.71) 13.6 (1.14) Medical insurance Private 78.2 (0.60) 69.9 (1.56) 51.4 (1.76) 21.5 (1.08) Public 17.8 (0.57) 25.2 (1.62) 39.8 (2.04) 58.4 (1.52) None 4.0 (0.32) 4.9 (0.47) 8.8 (1.12) 20.1 (1.06) Area of residence Urban 89.0 (0.36) 92.7 (0.54) 94.7 (0.65) 91.8 (0.59) Rural 11.0 (0.36) 7.3 (0.54) 5.3 (0.65) 8.2 (0.59) Family poverty level < 100% 10.2 (0.56) 15.8 (1.58) 24.6 (1.98) 55.0 (1.52) 100–199% 18.6 (0.71) 19.4 (1.16) 29.8 (1.97) 33.1 (1.40) 200–299% 17.3 (0.64) 19.0 (1.07) 18.2 (1.33) 7.8 (0.78) ≥ 300 % 53.9 (0.79) 45.7 (1.50) 27.4 (1.66) 4.1 (0.44) Parent education < High school 5.5 (0.39) 11.2 (1.27) 25.3 (1.85) 69.3 (1.17) High school 27.8 (0.73) 25.0 (1.37) 30.6 (1.68) 17.2 (0.96) Some college 32.2 (0.77) 27.3 (1.43) 19.7 (1.42) 7.9 (0.72) College graduate 34.4 (0.74) 36.6 (1.38) 24.4 (1.45) 5.6 (0.65) No Insurance % (SE) No Usual Place for Care % (SE) No Doctor Contact in Past 12 Months % (SE) Visited Emergency Room in Past 12 Months % (SE) Delayed/Forgone Care % (SE) Went to Other Countries for Health Care/Prescriptions % (SE) Discriminated Against in Health Care % (SE) Parental English proficiency Is English speaker at home *4.0 (0.32) *1.8 (0.29) *9.6 (0.47) *20.3 (0.80) *8.4 (0.40) *0.2 (0.07) 2.1 (0.25) Speaks very well 4.9 (0.47) 1.6 (0.39) 8.1 (0.88) 22.3 (1.50) 8.0 (0.88) 1.5 (0.36) 1.7 (0.31) Speaks well 8.8 (1.12) 2.5 (0.42) 9.1 (1.11) 15.4 (1.57) 5.1 (0.81) 3.5 (0.68) 2.3 (0.77) Speaks not well or not at all 20.1 (1.06) 6.5 (0.71) 15.3 (1.15) 14.6 (1.14) 6.8 (0.73) 3.3 (0.51) 2.5 (0.47) Child race/ethnicity Non-Hispanic White *3.9 (0.37) *1.6 (0.21) *8.8 (0.45) *19.4 (0.74) *8.4 (0.45) *0.4 (0.08) *2.0 (0.28) Non-Hispanic Black 3.1 (0.82) 2.4 (0.92) 7.1 (1.35) 22.4 (2.52) 6.5 (1.64) 0.3 (0.26) 1.1 (0.46) Hispanic 14.5 (0.74) 4.4 (0.42) 12.2 (0.71) 18.6 (0.86) 7.6 (0.56) 2.9 (0.36) 2.6 (0.35) Non-Hispanic Asian 3.2 (0.52) 1.8 (0.48) 12.1 (1.38) 11.8 (1.37) 3.8 (0.73) 1.2 (0.41) 1.2 (0.41) Other 4.7 (1.27) 2.3 (1.02) 11.4 (2.37) 23.7 (2.83) 11.1 (1.78) 0.2 (0.16) 2.9 (0.72) Child age, y 0–4 *6.6 (0.47) *2.0 (0.29) *4.6 (0.39) *24.5 (0.80) 7.1 (0.50) 1.3 (0.21) 2.0 (0.28) 5–11 9.5 (0.48) 3.5 (0.32) 14.5 (0.55) 14.9 (0.58) 7.9 (0.45) 1.7 (0.23) 2.2 (0.25) Child gender Male 8.2 (0.52) 3.0 (0.32) 14.6 (0.39) *20.1 (0.77) 7.9 (0.47) 1.7 (0.22) 2.3 (0.28) Female 8.5 (0.47) 2.8 (0.33) 14.5 (0.55) 17.2 (0.71) 7.2 (0.46) 1.4 (0.21) 2.0 (0.26) Child health status Fair/poor *19.8 (2.27) *6.9 (1.35) 12.2 (1.98) *29.6 (2.03) 16.2 (1.89) *5.4 (1.14) 5.2 (1.11) Excellent/very good/good 7.5 (0.32) 2.6 (0.22) 10.4 (0.37) 17.9 (0.55) 6.9 (0.31) 1.2 (0.16) 1.9 (0.20) Child citizenship US-born *7.1 (0.31) *2.3 (0.23) *9.9 (0.39) *19.1 (0.51) 7.6 (0.32) *1.3 (0.16) 2.0 (0.19) Naturalized citizen 4.4 (2.36) 3.0 (2.65) 16.1 (5.66) 12.5 (3.50) 8.9 (3.10) 3.7 (2.37) 0.8 (0.62) Noncitizen 36.6 (3.22) 15.3 (2.41) 23.1 (2.68) 10.5 (1.77) 7.6 (1.56) 5.1 (1.37) 5.4 (1.74) Area of residence Urban *8.1 (0.37) 2.9 (0.26) 10.3 (0.42) 18.6 (0.54) 7.5 (0.34) 1.4 (0.17) 2.2 (0.21) Rural 10.9 (1.13) 3.3 (0.69) 12.8 (1.24) 20.1 (1.20) 8.6 (0.74) 2.2 (0.54) 1.9 (0.28) Family poverty level < 100% *16.3 (1.13) *5.4 (0.69) *13.4 (1.19) *20.5 (1.25) 9.3 (0.96) *1.9 (0.39) 2.9 (0.45) 100–199% 12.0 (0.77) 3.9 (0.51) 10.8 (0.71) 18.0 (1.08) 7.3 (0.70) 2.4 (0.49) 2.4 (0.49) 200–299% 7.0 (0.66) 2.2 (0.44) 10.1 (0.84) 17.7 (1.14) 7.9 (0.66) 1.4 (0.32) 2.7 (0.56) ≥ 300% 2.0 (0.24) 1.1 (0.16) 8.9 (0.50) 18.5 (0.74) 6.6 (0.51) 0.8 (0.16) 1.3 (0.18) Parent education < High school *17.0 (1.03) *5.9 (0.66) *14.4 (1.16) 17.1 (1.10) 7.6 (0.76) *2.7 (0.50) *2.6 (0.44) High school 8.5 (0.78) 2.8 (0.41) 10.4 (0.78) 19.3 (1.05) 6.7 (0.65) 1.3 (0.27) 2.3 (0.47) Some college 6.1 (0.50) 2.1 (0.42) 9.9 (0.63) 21.8 (0.95) 9.1 (0.75) 1.4 (0.28) 2.4 (0.30) College graduate 2.4 (0.31) 1.2 (0.22) 7.7 (0.54) 16.9 (0.86) 7.0 (0.56) 0.8 (0.15) 1.4 (0.25) Medical insurance Private 0.9 (0.15) *9.4 (0.42) *18.0 (0.58) *6.5 (0.33) *0.8 (0.12) *2.4 (0.15) Public 2.4 (0.43) 8.8 (0.78) 21.2 (1.04) 8.3 (0.69) 1.7 (0.30) 3.0 (0.50) None 19.2 (1.71) 25.4 (1.92) 14.4 (1.78) 12.5 (1.47) 6.2 (1.18) 4.9 (0.94) No Insurance No Usual Place For Care No Doctor Contact In Past 12 Months Visited Emergency Room In Past 12 Months Delayed/Forgone Care Went To Other Countries For Health Care/Prescriptions Discriminated Against In Health Care Parental English proficiency Is English speaker at home Reference Reference Reference Reference Reference Reference Reference Speaks very well 1.04 (0.79, 1.35) 0.8 (0.42, 1.52) 0.75 (0.56, 0.99) 1.12 (0.89, 1.41) 0.95 (0.7, 1.3) 6.39 (2.75, 14.88) 0.71 (0.39, 1.28) Speaks well 1.46 (0.99, 2.16) 0.83 (0.36, 1.93) 0.73 (0.5, 1.08) 0.68 (0.49, 0.93) 0.54 (0.35, 0.84) 15.32 (6.11, 38.42) 0.73 (0.27, 1.95) Speaks not well or not at all 2.05 (1.44, 2.93) 1.03 (0.41, 2.55) 1.02 (0.69, 1.51) 0.5 (0.35, 0.72) 0.49 (0.34, 0.71) 10.68 (3.89, 29.29) 0.48 (0.26, 0.9) Child race/ethnicity Non-Hispanic White Reference Reference Reference Reference Reference Reference Reference Non-Hispanic Black 0.53 (0.29, 0.95) 1.29 (0.53, 3.13) 0.75 (0.49, 1.16) 1.08 (0.78, 1.5) 0.61 (0.33, 1.12) 0.93 (0.47, 1.86) 0.35 (0.13, 0.96) Hispanic 1.16 (0.8, 1.67) 0.83 (0.42, 1.61) 1.12 (0.88, 1.43) 1.1 (0.88, 1.39) 0.91 (0.7, 1.2) 1.72 (1, 2.96) 0.96 (0.53, 1.74) Non-Hispanic Asian 0.45 (0.3, 0.7) 0.83 (0.34, 2.03) 1.77 (1.23, 2.53) 0.68 (0.5, 0.92) 0.53 (0.34, 0.84) 0.7 (0.31, 1.59) 0.59 (0.19, 1.88) Other 0.87 (0.42, 1.77) 1.17 (0.39, 3.54) 1.39 (0.83, 2.35) 1.18 (0.83, 1.66) 1.21 (0.8, 1.84) 0.4 (0.1, 1.6) 1.13 (0.62, 2.05) Child age, y 0–4 0.69 (0.56, 0.86) 0.71 (0.49, 1.02) 0.29 (0.24, 0.35) 1.88 (1.67, 2.11) 0.9 (0.73, 1.11) 0.8 (0.52, 1.22) 0.85 (0.58, 1.25) 5–11 Reference Reference Reference Reference Reference Reference Reference Child health status Fair/poor Reference Reference Reference Reference Reference Reference Reference Excellent/very good/good 0.68 (0.49, 0.95) 0.79 (0.45, 1.37) 1.26 (0.82, 1.93) 0.41 (0.32, 0.51) 0.35 (0.26, 0.47) 0.37 (0.2, 0.69) 0.41 (0.23, 0.71) Child citizenship US-born Reference Reference Reference Reference Reference Reference Reference Naturalized citizen 0.56 (0.12, 2.56) 1.72 (0.03, 103) 1.5 (0.56, 4.02) 0.78 (0.41, 1.47) 1.29 (0.55, 3.05) 2.59 (0.97, 6.91) 0.44 (0.03, 7.83) Noncitizen 4.23 (3.1, 5.76) 2.86 (1.57, 5.18) 1.39 (1, 1.94) 0.76 (0.51, 1.12) 1.02 (0.6, 1.72) 1.59 (0.77, 3.31) 2.75 (1.22, 6.18) Area of residence Urban 0.77 (0.6, 1) 1.08 (0.64, 1.83) 0.83 (0.64, 1.08) 1.01 (0.85, 1.19) 1.05 (0.85, 1.31) 0.56 (0.34, 0.93) 1.48 (1.02, 2.16) Rural Reference Reference Reference Reference Reference Reference Reference Family poverty level < 100% 3.24 (2.19, 4.78) 1.21 (0.57, 2.54) 1.21 (0.81, 1.8) 1.14 (0.88, 1.47) 1.47 (1, 2.14) 0.35 (0.15, 0.81) 1.08 (0.58, 1.99) 100–199% 3.4 (2.35, 4.91) 1.36 (0.68, 2.74) 1.01 (0.76, 1.33) 0.94 (0.75, 1.19) 1.17 (0.83, 1.64) 0.73 (0.38, 1.41) 1.14 (0.67, 1.96) 200–299 % 2.76 (1.92, 3.97) 1.2 (0.62, 2.31) 1.01 (0.76, 1.34) 0.92 (0.75, 1.13) 1.25 (0.94, 1.65) 0.9 (0.47, 1.7) 1.71 (1.01, 2.92) ≥ 300% Reference Reference Reference Reference Reference Reference Reference Parent education < High school 1.75 (1.13, 2.74) 1.61 (0.82, 3.15) 1.87 (1.26, 2.78) 1.01 (0.74, 1.38) 0.77 (0.54, 1.09) 0.8 (0.4, 1.62) 1.05 (0.58, 1.91) High school 1.69 (1.15, 2.48) 1.26 (0.68, 2.33) 1.39 (1.02, 1.89) 1.09 (0.89, 1.34) 0.71 (0.53, 0.94) 0.84 (0.47, 1.49) 1.15 (0.66, 2) Some college 1.63 (1.11, 2.4) 1.16 (0.65, 2.07) 1.34 (1.1, 1.64) 1.28 (1.06, 1.53) 1.05 (0.79, 1.39) 1.43 (0.81, 2.53) 1.33 (0.84, 2.12) College graduate Reference Reference Reference Reference Reference Reference Reference Medical insurance Private Reference Reference Reference Reference Reference Reference Public 2.06 (1.18, 3.59) 0.71 (0.53, 0.95) 1.25 (1.01, 1.55) 1.43 (1.1, 1.86) 1.5 (0.81, 2.76) 2.51 (1.44, 4.38) None 15.83 (9.58, 26.14) 2.22 (1.69, 2.92) 0.89 (0.64, 1.23) 2.24 (1.61, 3.12) 4.34 (2.14, 8.78) 3.5 (2.1, 5.84)
Z. J. Huang’s participation on the study was made possible by an Intergovernmental Personnel Act Agreement funding from the Office of Data and Program Development, Maternal and Child Health Bureau.
Human Participant Protection This study was a secondary analysis of public-use data and was exempt from review.
