Video Abstract

Video Abstract

Close modal
OBJECTIVES

To assess the association between neighborhood socioeconomic deprivation and health care utilization in a cohort of children with medical complexity (CMC).

METHODS

Cross-sectional study of children aged <18 years receiving care in our institution’s patient-centered medical home (PCMH) for CMC in 2016 to 2017. Home addresses were assigned to census tracts and a tract-level measure of socioeconomic deprivation (Deprivation Index with range 0–1, higher numbers represent greater deprivation). Health care utilization outcomes included emergency department visits, hospitalizations, inpatient bed days, and missed PCMH clinic appointments. To evaluate the independent association between area-level socioeconomic deprivation and utilization outcomes, multivariable Poisson and linear regression models were used to control for demographic and clinical covariates.

RESULTS

The 512 included CMC lived in neighborhoods with varying degrees of socioeconomic deprivation (median 0.32, interquartile range 0.26–0.42, full range 0.12–0.82). There was no association between area-level deprivation and emergency department visits (adjusted risk ratio [aRR] 0.98; 95% confidence interval [CI]: 0.93 to 1.04), hospitalizations (aRR 0.97; 95% CI: 0.92 to 1.01), or inpatient bed-days (aRR 1.00, 95% CI: 0.80 to 1.27). However, there was a 13% relative increase in the missed clinic visit rate for every 0.1 unit increase in Deprivation Index (95% CI: 8%–18%).

CONCLUSIONS

A child’s socioeconomic context is associated with their adherence to PCMH visits. Our PCMH for CMC includes children living in neighborhoods with a range of socioeconomic deprivation and may blunt effects from harmful social determinants. Incorporating knowledge of the socioeconomic context of where CMC and their families live is crucial to ensure equitable health outcomes.

What’s Known on This Subject:

The conditions in which children grow and age can be characterized using area-level data. These data, characterizing social determinants of health, influence outcomes across conditions. Differential exposures to such determinants perpetuate disparities that may extend to children with medical complexity.

What This Study Adds:

For children with medical complexity in a patient-centered medical home, living in more socioeconomically deprived areas was associated with more missed outpatient visits but not emergency or inpatient utilization. Patient-centered medical homes may blunt effects from harmful social determinants.

Social determinants of health are defined by the World Health Organization as “the conditions in which people are born, grow, work, live, and age, and the wider set of forces and systems shaping the conditions of daily life.”1  These conditions influence health outcomes, narrow or widen disparities, and represent opportunities through which interventions may be targeted.2,3  The link between social determinants of health and health outcomes extends to multiple pediatric conditions. For example, we know that broad, persistent socioeconomic disparities exist in rates of acute injuries, lower respiratory tract infections, and all-cause inpatient bed days (IPBDs).47  Similar disparities have been described for chronic conditions that begin in childhood and extend across the life-course.811 

Little is known about how social determinants of health are experienced by and influence the health of children with medical complexity (CMC). CMC are a unique and important pediatric population with chronic, multisystem disease that results in frequent medical needs, including health care utilization, and functional limitations, including medical technology dependence.1215  This population accounts for <1% of all children and just 2.5% of yearly pediatric hospitalizations, yet contributes nearly 25% of all pediatric IPBDs and up to 33% of child health spending.16  We know that CMC and their families experience immense clinical challenges and corresponding strains on family life.1722  We know less, however, about how these challenges and strains are influenced by underlying socioeconomic context. Moreover, the impacts of social determinants of health-on-health care utilization for CMC remain largely unexplored.23 

Social determinants of health are often rooted in place, within the communities in which patients and families live and grow. Examining area-level socioeconomic context can provide insight into patient- and population-level risks and assets that influence utilization of health care services.24,25  In this study, we sought to evaluate the socioeconomic context of CMC cared for in our medical home using information attached to a child’s home address. Specifically, we sought to enumerate the extent to which CMC lived in socioeconomically deprived areas and assess the association of area-level socioeconomic deprivation with health care utilization. We hypothesized that CMC residing in areas with higher socioeconomic deprivation would have more acute health care utilization (emergency department [ED] visits, hospitalizations, IPBDs) and more missed clinic visits than those residing in areas with less deprivation.

This cross-sectional study included CMC receiving care at Cincinnati Children’s Hospital Medical Center’s (CCHMC) Complex Care Center. The Complex Care Center serves as a patient-centered medical home (PCMH) for >500 children with severe, chronic disease who are technology dependent and/or receive care from 3 or more subspecialties.26  The Complex Care Center provides care at an urban main campus and a suburban satellite campus.

Children listed in CCHMC’s Complex Care Center patient registry throughout the 2-year study period (January 1, 2016–December 31, 2017) were eligible for inclusion if they were <18 years old at beginning of the study period (n = 551). Children were excluded if they were in the custody of their home county (ie, foster care) because their home census tract could not be determined (n = 20). We also excluded international patients who received primary care in the Complex Care Center while receiving subspecialty treatment at CCHMC (n = 17).27  Finally, we excluded 2 patients with critical missing data (eg, address), leaving 512 patients in our cohort.

Demographic and utilization variables were abstracted from the electronic medical record (EMR), and clinical variables were examined through abstraction of diagnosis codes from the EMR. Demographics, including home address, and clinical characteristics were examined as of beginning of study eligibility; utilization was examined between January 1, 2016, and December 31, 2017. Each patient’s address was geocoded and mapped to a census tract. Tracts are census-defined geographic areas that classify small regions with demographic and socioeconomic homogeneity.28  Area-level socioeconomic variables, seen as markers of social determinants of health, are available at the census tract-level and are used in medical and public health research to study relationships between context (social determinants of health) and a range of outcomes.29  The CCHMC institutional review board approved this study with a waiver of informed consent.

The primary exposure of interest was an area-level measure of socioeconomic deprivation of the child’s residential census tract. This was quantified using a preexisting census tract-level “Deprivation Index” that includes variables from the 2015 5-year US Census American Community Survey related to socioeconomic deprivation.30  The 6 variables composing the Deprivation Index are: fraction of households below the federal poverty level, median annual household income, fraction of adults with less than high school education, fraction of population without health insurance, fraction of households receiving public assistance, and fraction of housing units that are vacant. The Deprivation Index ranges from 0 to 1 with higher values indicating greater socioeconomic deprivation. To assign a Deprivation Index value to each child, patient home addresses on January 1, 2016, were spatially joined to census tracts using 2015 TIGER/Line address range files with custom, offline geocoding software.31,32 

Patient-level outcomes for health service utilization included the number of ED visits, hospitalizations, and IPBDs for all causes during the 2-year study period. IPBD is defined as the sum of the differences for each admission between hospital discharge date and time and hospital admission date and time. We also examined the number of missed Complex Care Center appointments (ie, appointments that were scheduled, but the patient was a “no show”).

We assessed patient demographics, including age at beginning of study period, gender, race, ethnicity, and insurance. Race and ethnicity were examined in this study as social constructs with known linkages to differential health outcomes among children, likely through pathways related to social, economic, and environmental exposures.3335  At our institution, race and ethnicity data are reported by families and entered into the EMR by registration staff who undergo standardized training. We classified race for this study as White, Black, or Other; ethnicity was classified as Hispanic or non-Hispanic. We additionally examined the distance from the patient’s home to the nearest Complex Care Center location. Measures of medical complexity included the number of body systems affected by complex chronic conditions (CCCs), as well as the presence of diagnostic codes indicative of medical technology dependence (eg, tracheostomy, feeding tube).36,37 

Our exposure of interest, the area-level measure of socioeconomic deprivation (ie, Deprivation Index), was examined as a continuous variable in primary outcome analyses to maintain power. To aid in interpretability and actionability, we also grouped patients into quartiles based on the Deprivation Index score. Demographic and clinical characteristics were then compared across these quartiles using the Kruskal–Wallis test for continuous variables and χ2 test for categorical variables.

Spearman correlation coefficients were calculated to examine the unadjusted relationship between Deprivation Index (as a continuous variable) and utilization outcomes (ED visits, hospitalizations, IPBDs, and missed Complex Care Center appointments). To examine the independent relationship between Deprivation Index and the count variable outcomes of ED visits, hospitalizations, and missed Complex Care Center appointments, we fit multivariable Poisson regression models that controlled for individual-level demographic and clinical covariates defined a priori (patient age at study entry, gender, race, ethnicity, primary insurer, number of CCCs, and technology dependence). Race and ethnicity were included in models as potential confounders given known relationships with both the exposure and outcomes. Missed appointments were modeled as a rate by including the total number of possible visits as an offset term, since patients had a different number of appointments that were possible to miss. To determine the independent relationship between Deprivation Index and the nonnormally distributed continuous outcome of IPBDs, a multivariable linear regression model was used with the log-transformed number of bed days and including the same covariates as in other outcome models.

A total of 512 CMC cared for in our Complex Care Center were included in this study (Table 1). Median age at the time of study initiation was 7.8 years (interquartile range [IQR]: 4.9–12.6), and the majority were male (57.2%), White (64.7%), non-Hispanic (94.9%), and had primary private insurance (61.9%). Children lived a median of 11.2 miles (IQR: 6.8–16.6) from the nearest Complex Care Center location, had a median of 4 CCCs (IQR 1–6), and 61.5% were technology-dependent.

TABLE 1

Demographic, Clinical Characteristics, and Utilization Outcomes by Area-Level Socioeconomic Deprivation Index in Children With Medical Complexity

Overall Cohort, n = 512High Deprivation (0.42–0.82), n = 131High-Medium Deprivation (0.32–0.42), n = 120Low-Medium Deprivation (0.26–0.32), n = 133Low Deprivation (0.12–0.26), n = 128P*ρ (95% CI)**
Demographics and Clinical Characteristics        
 Age, y, median (IQR) 7.8 (4.9–12.6) 6.3 (4.0–10.0) 7.0 (4.3–12.2) 8.8 (5.9–12.1) 10.6 (6.2–15.2) <.0001 — 
 Male 293 (57.2) 66 (50.4) 71 (59.2) 81 (60.9) 75 (58.6) .32 — 
 Race      <.0001 — 
   White 331 (64.7) 58 (44.2) 79 (65.8) 96 (72.2) 98 (76.6) — — 
   Black 155 (30.3) 65 (49.6) 37 (30.8) 32 (24.1) 21 (16.4) — — 
   Other 26 (5.0) 8 (6.1) 4 (3.3) 5 (3.8) 9 (7.0) — — 
 Hispanic ethnicity 26 (5.1) 10 (7.6) 5 (4.2) 6 (4.5) 5 (3.9) .49 — 
 Insurance        
 Public primary insurance 317 (61.9) 114 (87.0) 78 (65.0) 79 (59.4) 46 (35.9) <.0001 — 
 Secondary insurance 147 (28.7) 14 (10.7) 35 (29.2) 38 (28.6) 60 (46.9) <.001 — 
 Distance to Complex Care Center,a miles, median (IQR) 11.2 (6.8–16.6) 7.7 (5.0–12.3) 11.1 (7.7–20.6) 13.1 (9.3–19.5) 12.5 (9.1–15.7) <.0001  
 Number of CCCs, median (IQR) 4 (1–6) 4 (1–6) 5 (2–7) 4 (2–6) 5 (2–7) .39 — 
Technology-dependent 315 (61.5) 80 (61.1) 75 (62.5) 81 (60.9) 79 (61.7) .99 — 
Utilization outcomes, median (IQR)        
  ED visits 1 (0–2) 1 (0–2) 1 (0–2) 1 (0–2) 1 (0–2) — 0.01 (−0.08 to 0.09) 
  Hospitalizations 1 (0–3) 1 (0–3) 0 (0–2) 1 (0–3) 1 (0–3) — −0.03 (−0.11 to 0.06) 
  IPBDs 0.52 (0–7.95) 0.52 (0–7.63) 0 (0–7.13) 0.51 (0–8.96) 0.57 (0–6.47) — −0.03 (−0.12 to 0.05) 
  Missed Complex Care Center visits 1 (0–3) 2 (0–5) 1 (0–3) 1 (0–2) 1 (0–2) — 0.16 (0.08 to 0.25) 
Overall Cohort, n = 512High Deprivation (0.42–0.82), n = 131High-Medium Deprivation (0.32–0.42), n = 120Low-Medium Deprivation (0.26–0.32), n = 133Low Deprivation (0.12–0.26), n = 128P*ρ (95% CI)**
Demographics and Clinical Characteristics        
 Age, y, median (IQR) 7.8 (4.9–12.6) 6.3 (4.0–10.0) 7.0 (4.3–12.2) 8.8 (5.9–12.1) 10.6 (6.2–15.2) <.0001 — 
 Male 293 (57.2) 66 (50.4) 71 (59.2) 81 (60.9) 75 (58.6) .32 — 
 Race      <.0001 — 
   White 331 (64.7) 58 (44.2) 79 (65.8) 96 (72.2) 98 (76.6) — — 
   Black 155 (30.3) 65 (49.6) 37 (30.8) 32 (24.1) 21 (16.4) — — 
   Other 26 (5.0) 8 (6.1) 4 (3.3) 5 (3.8) 9 (7.0) — — 
 Hispanic ethnicity 26 (5.1) 10 (7.6) 5 (4.2) 6 (4.5) 5 (3.9) .49 — 
 Insurance        
 Public primary insurance 317 (61.9) 114 (87.0) 78 (65.0) 79 (59.4) 46 (35.9) <.0001 — 
 Secondary insurance 147 (28.7) 14 (10.7) 35 (29.2) 38 (28.6) 60 (46.9) <.001 — 
 Distance to Complex Care Center,a miles, median (IQR) 11.2 (6.8–16.6) 7.7 (5.0–12.3) 11.1 (7.7–20.6) 13.1 (9.3–19.5) 12.5 (9.1–15.7) <.0001  
 Number of CCCs, median (IQR) 4 (1–6) 4 (1–6) 5 (2–7) 4 (2–6) 5 (2–7) .39 — 
Technology-dependent 315 (61.5) 80 (61.1) 75 (62.5) 81 (60.9) 79 (61.7) .99 — 
Utilization outcomes, median (IQR)        
  ED visits 1 (0–2) 1 (0–2) 1 (0–2) 1 (0–2) 1 (0–2) — 0.01 (−0.08 to 0.09) 
  Hospitalizations 1 (0–3) 1 (0–3) 0 (0–2) 1 (0–3) 1 (0–3) — −0.03 (−0.11 to 0.06) 
  IPBDs 0.52 (0–7.95) 0.52 (0–7.63) 0 (0–7.13) 0.51 (0–8.96) 0.57 (0–6.47) — −0.03 (−0.12 to 0.05) 
  Missed Complex Care Center visits 1 (0–3) 2 (0–5) 1 (0–3) 1 (0–2) 1 (0–2) — 0.16 (0.08 to 0.25) 

Data in table are presented as n (%) unless otherwise noted. —, statistical test was not completed.

a

Distance in miles from home address to nearest Complex Care Center location.

*

P value comparison of demographics and clinical characteristics across 4 quartiles of area-level socioeconomic deprivation (utilizing the deprivation index) calculated by Kruskal–Wallis test for continuous variables and χ2 test for categorical variables.

**

ρ is the calculated Spearmann correlation coefficient between the deprivation index (as a continuous variable) and outcomes, presented with corresponding 95% confidence interval. P values from comparison of outcomes across the 4 quartiles of area-level socioeconomic deprivation (treating deprivation as a categorical variable) calculated by Kruskal–Wallis test are as follows: ED visits, P = .99; hospitalizations, P = .72; IPBD, P = .65; missed Complex Care Center visits, P = .0024.

Area-Level Socioeconomic Deprivation

The median Deprivation Index of the census tracts in which included CMC resided was 0.32 (IQR: 0.26–0.42, full range: 0.12–0.82). Supplemental Table 2 details census tract-level socioeconomic characteristics composing the Deprivation Index.

There were significant differences in demographic characteristics across the quartiles of Deprivation Index (Table 1). Compared with children residing in the lowest quartile of deprivation, those residing in the highest quartile were younger (median age 6.3 vs 10.6 years, P < .001 across all quartiles) and more frequently Black (49.6% vs 16.4%, P < .001 across all quartiles) and publicly insured (87.0% vs 35.9%, P < .001 across all quartiles). Children living in the highest quartile of deprivation lived closer to a Complex Care Center location than children from areas of lesser deprivation (median 7.7 miles versus 12.5 miles in the lowest quartile, P < .001 across all quartiles). There were no significant differences in measures of medical complexity (number of CCCs and technology dependence) across the quartiles.

ED Utilization

Children had a median of 1 ED visit (IQR: 0–2, full range: 0–17). There was no significant association between Deprivation Index and ED utilization in unadjusted (ρ 0.01, 95% confidence interval [CI]: −0.08 to 0.09; Table 1) or adjusted (adjusted risk ratio [aRR] 0.98; 95% CI: 0.93 to 1.04; Fig 1) analyses.

FIGURE 1

Adjusted analysis of association between utilization outcomes and area-level socioeconomic Deprivation Index in children with medical complexity. Multivariable Poisson regression models were used to calculate aRR and associated 95% CI for the independent relationship between Deprivation Index and outcomes of ED visits and hospitalizations. To determine the independent relationship between Deprivation Index and missed Complex Care Center appointments, the multivariable Poisson regression model included the total number of possible visits as an offset term; thus, for this model, aRR represents an adjusted rate ratio. For IPBDs, a multivariable linear regression model using the log-transformed number of bed days was used to calculate aRR and associated 95% CI. All models adjusted for patient age at study entry, gender, race, ethnicity, primary insurance, number of CCCs, and technology dependence.

FIGURE 1

Adjusted analysis of association between utilization outcomes and area-level socioeconomic Deprivation Index in children with medical complexity. Multivariable Poisson regression models were used to calculate aRR and associated 95% CI for the independent relationship between Deprivation Index and outcomes of ED visits and hospitalizations. To determine the independent relationship between Deprivation Index and missed Complex Care Center appointments, the multivariable Poisson regression model included the total number of possible visits as an offset term; thus, for this model, aRR represents an adjusted rate ratio. For IPBDs, a multivariable linear regression model using the log-transformed number of bed days was used to calculate aRR and associated 95% CI. All models adjusted for patient age at study entry, gender, race, ethnicity, primary insurance, number of CCCs, and technology dependence.

Close modal

Hospitalization and Inpatient Bed Days

Children had a median of 1 hospitalization (IQR: 0–3, full range: 0–18) resulting in a median of 0.51 IPBDs (IQR: 0–7.9, full range: 0–135.5). There was no significant association between Deprivation Index and hospital utilization in unadjusted (hospitalization ρ −0.03, 95% CI: −0.11 to 0.06; IPBD ρ −0.03; 95% CI: −0.11 to 0.05; Table 1) or adjusted (hospitalization aRR 0.97; 95% CI: 0.92 to 1.01; IPBD aRR 1.00; 95% CI: 1.00 to 1.27; Fig 1) analyses.

Missed Complex Care Center Appointments

Children had a median of 1 missed appointment (IQR: 0–3, full range 0–15). The majority of missed visits (56%) were for preventive care. In unadjusted analysis, there was a moderate correlation between Deprivation Index and rate of missed Complex Care Center appointments (ρ 0.16; 95% CI: 0.08 to 0.25; Table 1). This association remained significant in adjusted analysis (aRR 1.13; 95% CI: 1.08 to 1.18; Fig 1); this can be interpreted as a 13% relative increase in the missed clinic visit rate for every 0.1-unit increase in Deprivation Index.

This cross-sectional study demonstrates that CMC receiving care in a PCMH live in areas with substantial variability in socioeconomic deprivation. Contrary to our hypothesis, CMC who resided in areas of higher socioeconomic deprivation were not at greater risk of ED utilization, hospital admissions, or total IPBD than their peers from areas with less deprivation. However, there was a significant association between area-level socioeconomic deprivation and the rate of missed Complex Care Center appointments. Our findings suggest that socioeconomic deprivation negatively influences some aspects of the preventive medical care of CMC. It is possible that the PCMH model applied to the care for CMC may mitigate some of the risk of acute utilization.

The substantial socioeconomic deprivation of the neighborhoods that are home to some of the most medically complex children cared for across our health system deserves attention. Of CMC living in the highest quartile of socioeconomic deprivation, the majority were technology-dependent and had 4 or more body systems affected by CCCs, with one-third having 6 or more body systems affected. These markers of medical complexity not only reflect the medical fragility of these children, but they also reflect the burden of care experienced by their families who are responsible for complex daily medical care and who must navigate a complex health care system.17,38,39  Families living in the most socioeconomically deprived neighborhoods, where over one-third of the population have household incomes below the federal poverty level and where less than one quarter of adults have obtained a high school education, likely experience this burden of care differently. Such social and economic complexities may compete with and magnify the medical complexities faced by CMC and their families.

Black families with CMC disproportionately experienced these social and economic complexities. Like many regions across the United States, Greater Cincinnati remains racially and economically segregated. Such societal inequities, and their extension to a range of health outcomes, are rooted in interpersonal, institutional, and structural racism.4042  These forms of racism could underlie our finding of higher missed visit rates for children living in socioeconomically deprived areas. Perhaps the trust a caregiver of a minoritized race or different socioeconomic background may have in a clinic setting is diminished by implicit or explicit biases of clinic staff. Perhaps a family’s competing priorities related to limited income, housing access, and vocational and education opportunities, because of structural racism, supersede their ability to go to a preventive visit. Although addressing interpersonal, institutional, and structural racism is challenging, efforts to identify areas for improvement and implement equity-minded changes are imperative. Tracking stratified CMC outcomes data to characterize gaps, cocreating improvement strategies with minoritized patients and families to address disparities, and holding our communities, institutions, and care teams accountable for providing equitable care are actionable and critical first steps toward dismantling the impacts of racism.

We did not see links between area-level socioeconomic deprivation and acute ED or hospital utilization. It is possible that these forms of health care utilization are predominantly driven by medical complexity in CMC, and not exacerbated by neighborhood socioeconomic factors. However, it is also possible that the impact of neighborhood deprivation is blunted through receipt of primary care in the PCMH care model of the Complex Care Center. Like many similar programs across the nation,4348  one of the goals of the Complex Care Center is to decrease ED and hospital utilization through enhanced care coordination by a multidisciplinary team that includes clinic nurses, nurse care managers, a social worker, social work care managers, and registered dieticians, in addition to 24-hour, 7-day a week access to clinic medical providers. Future studies across our local health system will be necessary to further delineate if the impact of PCMH-centered care differs across populations of varied socioeconomic deprivation. Such investigation would be useful beyond the population of CMC cared for in our PCMH and across similar populations cared for in other settings. Further, understanding which aspects of the PCMH model are most beneficial and if these benefits are consistent across different populations is crucial as health systems seek to optimize care delivery and health outcomes of CMC. Such findings could influence how health systems prioritize enrollment in multidisciplinary medical homes for CMC and distribute care coordination services to address both the medical and social complexity of CMC.

Although not associated with costly ED and hospital utilization, the impact of area-level socioeconomic deprivation on the health and well-being of CMC and their families should not be overlooked. This is perhaps evident in our finding of increased risk of missed Complex Care Center appointments for CMC from areas of higher socioeconomic deprivation. We are unable to determine the impacts of missed care on the health of the children and the utilization outcomes assessed in our study. However, previous work has demonstrated an association between missed primary care and delayed preventive care, as well as associations with ED visits and hospitalizations for both children with and without chronic disease.4952  Given the complexity of a medically fragile population such as ours, and the degree to which many rely on longitudinal primary and subspecialty care, we expect that the impacts of missed or foregone care may be amplified. Efforts to understand barriers specific to families of CMC, and perhaps even more specifically to families of CMC living with socioeconomic deprivation, may help inform interventions to ensure that children receive the care necessary to promote their health and well-being.53  PCMHs need to reconsider how, when, and where care is provided to ensure that it meets the needs and complexities of CMC and their families. Understanding if missed visits are because of previously cited reasons of forgotten appointments or difficulties with transportation or work schedules,53  or because of additional factors such as a lower perception of visit value or “burnout” from many subspecialty appointments, may help guide the development and implementation of interventions. Beyond screening for potential barriers to visit attendance, possible interventions for these children and their families could include simple reminders of appointments via text or calls, coordination of insurance-approved transportation, flexibility of clinic hours including evenings or weekends, expansion of clinic locations to reduce distances families travel, concierge services to coordinate multiple appointments on a single day, and telemedicine access.

There are several limitations to this work. First, this was a single-center, cross-sectional study of CMC who receive care in a PCMH with resources to support families and reduce unnecessary utilization. Our design does not allow the determination of causation. Further work across different settings is necessary to establish generalizability of our results because the distribution of area-level deprivation experienced in our cohort may not mirror that experienced in other areas or patient populations. Second, we were not able to discern the acuity or necessity of acute utilization events or the consequences of missed visits. Third, we used measures of medical complexity that likely oversimplify the complexity and severity of underlying diagnoses; retrospective data are not well suited to distinguish granular grades of functional status. Fourth, race and ethnicity data are imperfect measures in our data source given the lack of uniformity in category definitions by individuals and variable collection approaches.54  Fifth, we examined patient home address at time of study eligibility and did not account for any changes in home address over the study period. This may have resulted in exposure misclassification. However, we assume that this misclassification would be nondifferential (ie, families likely to move to areas of similar deprivation or just as likely to move to areas with greater deprivation as to move to areas of lower deprivation). That said, should families from areas with more deprivation move to areas with less deprivation, our findings would likely be biased toward the null. Finally, this study examined area-level socioeconomic deprivation. Although we cannot necessarily approximate patient-level socioeconomic status (eg, parental education, family constellation, household income) from area-level markers of deprivation, previous studies have shown correlations between family-reported hardships and neighborhood-level socioeconomic data.55  Area-level factors do, however, directly characterize contextual exposures for children and their families. We believe that measuring and understanding the context in which someone lives may have additional benefits to understanding and addressing individual- or family-level challenges.

Compared with peers residing in areas of lower socioeconomic deprivation, CMC living in areas of greater socioeconomic deprivation were at increased risk of missing PCMH appointments. However, they were not at increased risk of ED utilization, hospital admission, or increased days in the hospital. Although underlying medical complexity may drive acute health care utilization, especially in the context of a CMC-focused PCMH, our findings suggest that the social complexity in which CMC live influences other aspects of their health care, including visits to a specialized medical home. Future studies are needed to understand the degree to which such factors influence other outcomes of relevance to CMC and their families (eg, completion of preventive care, attendance at school, quality of life) and to understand how to adapt care to support the provision of the right care at the right time and place. To ensure equitable care and outcomes for all CMC, health systems and providers should seek to understand and address any barriers to care arising from the socioeconomic context in which CMC live.

Dr Thomson conceptualized and designed the study, performed initial data analysis and interpretation, and drafted the initial manuscript; Drs Butts and Camara contributed to conceptualization and design of the study and drafted the initial manuscript; Ms Rasnick and Dr Brokamp supervised final data analysis and interpretation; Drs Heyd, Steuart, and Callahan contributed to conceptualization and design of the study, as well as interpretation of the data; Dr Beck supervised conceptualization and design of the study, as well as interpretation of data; and all authors reviewed and revised the manuscript for important intellectual content, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.

FUNDING: Dr Thomson was supported by the Agency for Healthcare Research and Quality under grant K08-HS025138. The agency had no role in the design or conduct of the study.

     
  • aRR

    adjusted risk ratio

  •  
  • CCC

    complex chronic condition

  •  
  • CCHMC

    Cincinnati Children’s Hospital Medical Center

  •  
  • CI

    confidence interval

  •  
  • CMC

    children with medical complexity

  •  
  • ED

    emergency department

  •  
  • EMR

    electronic medical record

  •  
  • IPBD

    inpatient bed day

  •  
  • IQR

    interquartile range

  •  
  • PCMH

    patient-centered medical home

1
World Health Organization
.
Social determinants of health
.
Available at: www.who.int/social_determinants/en/. 2015. Accessed May 25, 2015
2
Braveman
P
,
Gottlieb
L
.
The social determinants of health: it’s time to consider the causes of the causes
.
Public Health Rep
.
2014
;
129
(
Suppl 2
):
19
31
3
Braveman
P
,
Egerter
S
,
Williams
DR
.
The social determinants of health: coming of age
.
Annu Rev Public Health
.
2011
;
32
:
381
398
4
Johnson
SB
,
Riis
JL
,
Noble
KG
.
State of the art review: poverty and the developing brain
.
Pediatrics
.
2016
;
137
(
4
):
e20153075
5
Yaeger
JP
,
Moore
KA
,
Melly
SJ
,
Lovasi
GS
.
Associations of neighborhood-level social determinants of health with bacterial infections in young, febrile infants
.
J Pediatr
.
2018
;
203
:
336
344.e1
6
Beck
AF
,
Florin
TA
,
Campanella
S
,
Shah
SS
.
Geographic variation in hospitalization for lower respiratory tract infections across one county
.
JAMA Pediatr
.
2015
;
169
(
9
):
846
854
7
Beck
AF
,
Riley
CL
,
Taylor
SC
,
Brokamp
C
,
Kahn
RS
.
Pervasive income-based disparities in inpatient bed-day rates across conditions and subspecialties
.
Health Aff (Millwood)
.
2018
;
37
(
4
):
551
559
8
Kressin
NR
,
Wang
F
,
Long
J
, et al
.
Hypertensive patients’ race, health beliefs, process of care, and medication adherence
.
J Gen Intern Med
.
2007
;
22
(
6
):
768
774
9
Wong
MD
,
Shapiro
MF
,
Boscardin
WJ
,
Ettner
SL
.
Contribution of major diseases to disparities in mortality
.
N Engl J Med
.
2002
;
347
(
20
):
1585
1592
10
Hudson
DL
,
Puterman
E
,
Bibbins-Domingo
K
,
Matthews
KA
,
Adler
NE
.
Race, life course socioeconomic position, racial discrimination, depressive symptoms and self-rated health
.
Soc Sci Med
.
2013
;
97
:
7
14
11
Beck
AF
,
Moncrief
T
,
Huang
B
, et al
.
Inequalities in neighborhood child asthma admission rates and underlying community characteristics in one US county
.
J Pediatr
.
2013
;
163
(
2
):
574
580
12
Simon
TD
,
Berry
J
,
Feudtner
C
, et al
.
Children with complex chronic conditions in inpatient hospital settings in the United States
.
Pediatrics
.
2010
;
126
(
4
):
647
655
13
Cohen
E
,
Kuo
DZ
,
Agrawal
R
, et al
.
Children with medical complexity: an emerging population for clinical and research initiatives
.
Pediatrics
.
2011
;
127
(
3
):
529
538
14
Berry
JG
,
Hall
M
,
Neff
J
, et al
.
Children with medical complexity and Medicaid: spending and cost savings
.
Health Aff (Millwood)
.
2014
;
33
(
12
):
2199
2206
15
Cohen
E
,
Berry
JG
,
Sanders
L
,
Schor
EL
,
Wise
PH
.
Status complexicus? The emergence of pediatric complex care
.
Pediatrics
.
2018
;
141
(
Suppl 3
):
S202
S211
16
Berry
JG
,
Poduri
A
,
Bonkowsky
JL
, et al
.
Trends in resource utilization by children with neurological impairment in the United States inpatient health care system: a repeat cross-sectional study
.
PLoS Med
.
2012
;
9
(
1
):
e1001158
17
Kuo
DZ
,
Cohen
E
,
Agrawal
R
,
Berry
JG
,
Casey
PH
.
A national profile of caregiver challenges among more medically complex children with special health care needs
.
Arch Pediatr Adolesc Med
.
2011
;
165
(
11
):
1020
1026
18
Boat
TF
,
Filigno
S
,
Amin
RS
.
Wellness for families of children with chronic health disorders
.
JAMA Pediatr
.
2017
;
171
(
9
):
825
826
19
Kuhlthau
K
,
Hill
KS
,
Yucel
R
,
Perrin
JM
.
Financial burden for families of children with special health care needs
.
Matern Child Health J
.
2005
;
9
(
2
):
207
218
20
Murphy
NA
,
Christian
B
,
Caplin
DA
,
Young
PC
.
The health of caregivers for children with disabilities: caregiver perspectives
.
Child Care Health Dev
.
2007
;
33
(
2
):
180
187
21
Thyen
U
,
Kuhlthau
K
,
Perrin
JM
.
Employment, child care, and mental health of mothers caring for children assisted by technology
.
Pediatrics
.
1999
;
103
(
6 Pt 1
):
1235
1242
22
Woodgate
RL
,
Edwards
M
,
Ripat
J
.
How families of children with complex care needs participate in everyday life
.
Soc Sci Med
.
2012
;
75
(
10
):
1912
1920
23
Fuller
AE
,
Brown
NM
,
Grado
L
,
Oyeku
SO
,
Gross
RS
.
Material hardships and health care utilization among low-income children with special health care needs
.
Acad Pediatr
.
2019
;
19
(
7
):
733
739
24
Council on Community Pediatrics
.
Poverty and child health in the United States
.
Pediatrics
.
2016
;
137
(
4
):
e20160339
25
Dreyer
B
,
Chung
PJ
,
Szilagyi
P
,
Wong
S
.
Child poverty in the United States today: introduction and executive summary
.
Acad Pediatr
.
2016
;
16
(
3 Suppl
):
S1
S5
26
Lail
J
,
Fields
E
,
Schoettker
PJ
.
Quality improvement strategies for population management of children with medical complexity
.
Pediatrics
.
2017
;
140
(
3
):
e20170484
27
Cincinnati Children’s Hospital Medical Center
.
International patients and families
.
28
United States Census Bureau
.
Census tracts and block numbering areas
.
Available at: https://www2.census.gov/geo/pdfs/reference/GARM/Ch10GARM.pdf. 1994. Accessed March 14, 2021
29
Beck
AF
,
Sandel
MT
,
Ryan
PH
,
Kahn
RS
.
Mapping neighborhood health geomarkers to clinical care decisions to promote equity in child health
.
Health Aff (Millwood)
.
2017
;
36
(
6
):
999
1005
30
Brokamp
C
,
Beck
AF
,
Goyal
NK
,
Ryan
P
,
Greenberg
JM
,
Hall
ES
.
Material community deprivation and hospital utilization during the first year of life: an urban population-based cohort study
.
Ann Epidemiol
.
2019
;
30
:
37
43
31
Brokamp
C
,
Wolfe
C
,
Lingren
T
,
Harley
J
,
Ryan
P
.
Decentralized and reproducible geocoding and characterization of community and environmental exposures for multisite studies
.
J Am Med Inform Assoc
.
2018
;
25
(
3
):
309
314
32
Brokamp
C
.
Zenodo cole-brokamp/hamilton v0.1
.
Available at: 10.5281/zenodo.1134943. Accessed January 29, 2018
33
Goyal
MK
,
Chamberlain
JM
,
Webb
M
, et al.
Pediatric Emergency Care Applied Research Network (PECARN)
.
Racial and ethnic disparities in the delayed diagnosis of appendicitis among children
.
Acad Emerg Med
.
2021
;
28
(
9
):
949
956
34
Nafiu
OO
,
Mpody
C
,
Kim
SS
,
Uffman
JC
,
Tobias
JD
.
Race, postoperative complications, and death in apparently healthy children
.
Pediatrics
.
2020
;
146
(
2
):
e20194113
35
Willi
SM
,
Miller
KM
,
DiMeglio
LA
, et al.
T1D Exchange Clinic Network
.
Racial-ethnic disparities in management and outcomes among children with type 1 diabetes
.
Pediatrics
.
2015
;
135
(
3
):
424
434
36
Feudtner
C
,
Christakis
DA
,
Connell
FA
.
Pediatric deaths attributable to complex chronic conditions: a population-based study of Washington State, 1980-1997
.
Pediatrics
.
2000
;
106
(
1 Pt 2
):
205
209
37
Feudtner
C
,
Feinstein
JA
,
Zhong
W
,
Hall
M
,
Dai
D
.
Pediatric complex chronic conditions classification system version 2: updated for ICD-10 and complex medical technology dependence and transplantation
.
BMC Pediatr
.
2014
;
14
:
199
38
Goudie
A
,
Narcisse
MR
,
Hall
DE
,
Kuo
DZ
.
Financial and psychological stressors associated with caring for children with disability
.
Fam Syst Health
.
2014
;
32
(
3
):
280
290
39
Romley
JA
,
Shah
AK
,
Chung
PJ
,
Elliott
MN
,
Vestal
KD
,
Schuster
MA
.
Family-provided health care for children with special health care needs
.
Pediatrics
.
2017
;
139
(
1
):
e20161287
40
Bailey
ZD
,
Krieger
N
,
Agénor
M
,
Graves
J
,
Linos
N
,
Bassett
MT
.
Structural racism and health inequities in the USA: evidence and interventions
.
Lancet
.
2017
;
389
(
10077
):
1453
1463
41
National Academies of Sciences, Engineering, and Medicine
.
Communities in Action: Pathways to Health Equity
.
Washington, DC
:
The National Academies Press
;
2017
42
U S Department of Health and Human Services
.
Opioid abuse in the United States and Department of Health and Human Services actions to address opioid-drug-related overdoses and deaths
.
J Pain Palliat Care Pharmacother
.
2015
;
29
(
2
):
133
139
43
Casey
PH
,
Lyle
RE
,
Bird
TM
, et al
.
Effect of hospital-based comprehensive care clinic on health costs for Medicaid-insured medically complex children
.
Arch Pediatr Adolesc Med
.
2011
;
165
(
5
):
392
398
44
Cohen
E
,
Friedman
JN
,
Mahant
S
,
Adams
S
,
Jovcevska
V
,
Rosenbaum
P
.
The impact of a complex care clinic in a children’s hospital
.
Child Care Health Dev
.
2010
;
36
(
4
):
574
582
45
Cohen
E
,
Jovcevska
V
,
Kuo
DZ
,
Mahant
S
.
Hospital-based comprehensive care programs for children with special health care needs: a systematic review
.
Arch Pediatr Adolesc Med
.
2011
;
165
(
6
):
554
561
46
Gordon
JB
,
Colby
HH
,
Bartelt
T
,
Jablonski
D
,
Krauthoefer
ML
,
Havens
P
.
A tertiary care-primary care partnership model for medically complex and fragile children and youth with special health care needs
.
Arch Pediatr Adolesc Med
.
2007
;
161
(
10
):
937
944
47
Mosquera
RA
,
Avritscher
EB
,
Samuels
CL
, et al
.
Effect of an enhanced medical home on serious illness and cost of care among high-risk children with chronic illness: a randomized clinical trial
.
JAMA
.
2014
;
312
(
24
):
2640
2648
48
Noritz
G
,
Madden
M
,
Roldan
D
, et al
.
A population intervention to improve outcomes in children with medical complexity
.
Pediatrics
.
2017
;
139
(
1
):
e20153076
49
Hakim
RB
,
Bye
BV
.
Effectiveness of compliance with pediatric preventive care guidelines among Medicaid beneficiaries
.
Pediatrics
.
2001
;
108
(
1
):
90
97
50
O’Connor
ME
,
Matthews
BS
,
Gao
D
.
Effect of open access scheduling on missed appointments, immunizations, and continuity of care for infant well-child care visits
.
Arch Pediatr Adolesc Med
.
2006
;
160
(
9
):
889
893
51
Pittard
WB
III
.
Well-child care in infancy and emergency department use by South Carolina Medicaid children birth to 6 years old
.
South Med J
.
2011
;
104
(
8
):
604
608
52
Tom
JO
,
Tseng
CW
,
Davis
J
,
Solomon
C
,
Zhou
C
,
Mangione-Smith
R
.
Missed well-child care visits, low continuity of care, and risk of ambulatory care-sensitive hospitalizations in young children
.
Arch Pediatr Adolesc Med
.
2010
;
164
(
11
):
1052
1058
53
Samuels
RC
,
Ward
VL
,
Melvin
P
, et al
.
Missed appointments: factors contributing to high no-show rates in an urban pediatrics primary care clinic
.
Clin Pediatr (Phila)
.
2015
;
54
(
10
):
976
982
54
Cowden
JD
,
Flores
G
,
Chow
T
, et al
.
Variability in collection and use of race/ethnicity and language data in 93 pediatric hospitals
.
J Racial Ethn Health Disparities
.
2020
;
7
(
5
):
928
936
55
Auger
KA
,
Kahn
RS
,
Simmons
JM
, et al
.
Using address information to identify hardships reported by families of children hospitalized with asthma
.
Acad Pediatr
.
2017
;
17
(
1
):
79
87

Competing Interests

CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no potential conflicts of interest to disclose.

Supplementary data