Assessing Electronic Health Records for Describing Racial and Ethnic Health Disparities: A Research Note

Abstract

The use of data derived from electronic health records (EHRs) to describe racial and ethnic health disparities is increasingly common, but there are challenges. While the number of patients contained in EHRs can be quite large, they may not be representative of a source population. One way to evaluate the extent of this limitation is by linking EHRs to an external source, in this case with the American Community Survey (ACS). Relying on a stratified random sample of about 200,000 patient records from a large, public, integrated delivery system in North Carolina (2016-2019), we assess linkages to restricted ACS microdata (2001-2017) based on race and ethnicity to understand the strengths and weaknesses of EHR-derived data for describing disparities. Our results suggest that Black-White comparisons will benefit from standard adjustments (e.g., weighting procedures), but that misestimation of health disparities may arise for Hispanic patients due to coverage rates for this group.

Introduction

As survey response rates fall and costs of data collection rise, researchers increasingly look to administrative and clinical data as supplements or substitutes (Meyer et al., 2015). For those interested in health disparities, data derived from electronic health records (EHRs) are one such source. Although researchers do not have direct access to EHRs, for simplicity, we refer to EHR-derived data as EHRs throughout. EHRs are increasingly being used to describe racial and ethnic disparities in health care utilization, treatment, and outcomes. Recent examples include studies of childhood obesity (Sharifi et al., 2016), maternal and neonatal delivery complications (Huennekens et al., 2020), and the receipt of treatment medications among adults seeking care and with a positive SARS-CoV-2 test (Boehmer et al., 2022; Wiltz et al., 2022). However, there are concerns related to the representation of racial and ethnic groups in EHRs, which has implications for studying racial and ethnic disparities in population research. This research note considers lessons learned about the strengths and weaknesses of EHRs for health disparities research, which have long been of interest to demographers (Foster et al., 2024; Montez et al., 2019). By linking EHRs from a large integrated health delivery system in North Carolina to microdata from the American Community Survey (ACS), our findings demonstrate the potential for the use of EHRs for population research.

Background

EHRs have tremendous potential for increasing knowledge and understanding of racial and ethnic health disparities. They contain up-to-date and detailed information pertinent to people’s health (e.g., utilization, laboratory tests and results, procedures, problem lists, visit-specific diagnoses, medications prescribed) for very large patient populations (Casey et al., 2016) and have benefits over other health data sources. Relative to health care administrative data (e.g., insurance claims data), which provide billing diagnoses and types of care regardless of where care is received, EHRs contain results of tests and biologic data such as weight or blood pressure. Compared to vital statistics, which describe health disparities at the beginning and end of life, EHRs cover the years in between. Finally, in contrast to most surveys which rely on subjective reports and are subject to potential recall bias about test results and diagnoses, EHRs contain those test results, whether they exceed clinical thresholds, and diagnoses based on them.

EHRs are a promising source to document racial and ethnic disparities in health and health care and can inform targeted interventions to ameliorate them (Rumball-Smith & Bates, 2018), but there are challenges: the data are not representative of a source population. Information based on EHRs pertains to people who obtain health care from a particular provider (or group of providers). Numbers can be quite large, but as a nonprobability sample of a source population, they are not population representative (Friedman et al., 2013; Goldstein et al., 2016; Groves, 2006). Estimation of health disparities can be biased if racial and ethnic groups are differentially represented. Representativeness may be growing since health facilities are increasingly aggregated into large integrated delivery systems, caring for up to several million people each. Nevertheless, the potential for bias remains.

As a further complication, EHRs are selective of people who choose to seek health care, which may vary between groups depending on insurance coverage and other factors (Goldstein et al., 2016; Weiskopf & Weng, 2013). Numerators and denominators may be affected differently as those who do not need or want health care or who lack insurance may have reduced representation. Denominators may exclude some of those at risk of the disease or health condition recorded in the numerator, potentially overstating prevalence as a result. Analysts typically correct for this by weighting prevalence measures or incorporating inverse proportional weights in regression analyses. Bias is lessened but not eliminated when EHRs incorporate a full range of care, from preventive to acute (Bower et al., 2017; Klompas et al., 2017). Other factors affecting health care utilization are also relevant to inclusion in EHRs, including education, health insurance coverage, and access to transportation (Bower et al., 2017). Race and ethnic differences in access to health care have narrowed over the past two decades, especially for the Hispanic population, but nevertheless persist (Ma et al., 2022; Mahajan et al., 2021).

With full information on the source population, it would be straightforward to identify potential bias and adjust for it. Instead, we take an indirect approach, assessing strengths and weaknesses of EHRs for the study of racial and ethnic health disparities through comparison with the ACS. The ACS is an annual cross-sectional survey of about 1-1.5% of the U.S. population designed as a replacement for the decennial census long-form in 2001. Critically, it is based on a probability sample designed to be representative. Although the sampling fraction is small, the number of individuals included each year is very large, more than 5 million nationally and exceeding 100,000 in North Carolina, the state served by the integrated health system we study.¹ Participation in the ACS is mandatory and response rates have historically topped 90% (U.S. Census Bureau, 2022b), although coverage rates vary by race and ethnicity (U.S. Census Bureau, 2022a), a point to which we will return below.

Data and Methods

This research note leveraged data collected as part of a pilot study to assess the feasibility of integrating the detailed health information available in EHRs with social, economic, and demographic data from the ACS (Udalova et al., 2022). It focused on EHRs from a large integrated health delivery system in North Carolina from 2016-2019. The health system is composed of a large academic health center, 11 community hospitals, and many hundreds of practices across the state including both urban and rural areas. Demographic and clinical data are collected using a single enterprise-level EHR with consistent policies regarding registration and demographic and clinical data collection across hospitals and practices. Care is offered to all residents of the state regardless of ability to pay, including uninsured, resulting in a diverse patient population, including citizens and non-citizens. Over two million patients were seen during 2016-2019, prior to Medicaid expansion in the state.

In the integrated health delivery system, race and ethnicity are generally obtained from a patient or proxy as part of registration during the first visit. During 2016-2019, patients provided the information directly to a clerk or as part of an intake questionnaire. Answers were coded as follows: for race, American Indian or Alaskan Native (AIAN), Asian, Black or African American, Native Hawaiian or Other Pacific Islander (NHPI), Other Race, Patient Refused, Unknown, White or Caucasian (categories listed in alphabetical order on the screen); and for ethnicity, Hispanic or Latino, not Hispanic or Latino, Patient Refused, Unknown. Clerks were instructed to accept whatever answer they were given and specifically not to push if there was resistance and not to fill in based on their own observations. About 10.5 percent of patients were missing or refused to answer the race question (Table 1). This figure compares favorably to averages based on the 56 health care institutions in the National COVID Collaborative Cohort (N3C), in which 11.3% of patients were missing data on race and an additional 8% were recorded as refusals (Cook et al., 2022). Although included in our analyses, a detailed assessment of missing data is beyond the scope of this research note. We return to this in our discussion.

Table 1.

Descriptive statistics for EHR sample (weighted)

	Percent	Standard error
Total	100
Assigned a PIK	97.75	0.0276
Matched to ACS (Conditional on Receiving a PIK)	16.87	0.1367
Sex
Male	41.62	0.1730
Female	58.38	0.1730
Birth Year
Before 1944	4.979	0.0815
1945-1949	10.09	0.1111
1950-1954	11.35	0.1152
1955-1959	11.72	0.1149
1960-1964	11.70	0.1131
1965-1969	10.91	0.1080
1970-1974	10.25	0.1034
1975-1979	9.200	0.0968
1980-1984	9.018	0.0957
1985 and later	10.77	0.1036
Race
White	61.71	0.1464
Black	19.05	0.1064
AIAN	0.473	0.0085
Asian	1.843	0.0154
Other	6.373	0.0376
Missing/Refused	10.54	0.0565
Ethnicity
Hispanic or Latino	5.584	0.0350
Not Hispanic or Latino	84.18	0.0737
Other/Missing	10.23	0.0572
Language
English	93.12	0.0491
Spanish	3.356	0.0265
Unknown	2.814	0.0357
Other	0.7075	0.0164
Health Insurance*
Any Private (Employer, State, Tricare)	59.68	0.1703
Any Medicare	20.32	0.1461
Any Medicaid/Gov’t Assistance	10.55	0.1004
Any Uninsured/Self-Pay	29.60	0.1547
Residence**
In-state	67.30	0.1671
Out-of-state	2.501	0.0558
Missing	30.20	0.1639

Source: Electronic Health Records (2016-2019)

Note:

^*Insurance categories are not mutually exclusive.

^**Residence is based on address provided at most recent visit.

AIAN = American Indian and Alaska Native. DMS #P-7519212.

All results were approved for release by the Disclosure Review Board of the U.S. Census Bureau, authorization number CBDRB-FY23-SEHSD003-021 and CBDRB-FY23-POP001-0074. All numeric values were rounded according to U.S. Census Bureau disclosure protocols to preserve data privacy.

We drew a disproportionate stratified random sample of about 200,000 patients aged 25-74 with at least two visits (e.g., hospitalization, in-patient visit, or out-patient visit) between 2016 and 2019 (Udalova et al., 2022). We oversampled patients who identified as Black, Asian, and Hispanic as well as those for whom race and ethnicity was missing to assess more precise estimates for these groups. Table 1 shows descriptive statistics for the sample drawn, weighted to account for the stratified sample design. Weights were constructed based on the racial and ethnic composition of the original EHR population. Patients in the weighted sample predominantly identify as White (61.71%) or Black (19.05%); smaller fractions identify as AIAN, Asian, or Hispanic (0.47%, 1.84%, and 5.58%, respectively) (Table 1). For this analysis, patients identifying as NHPI were grouped with Other Race.

Structured EHRs were transferred via secure means to the Census Bureau IT environment and a Protected Identification Key (PIK) was assigned.² A PIK is a unique anonymized person identifier used at the Census Bureau to link surveys and administrative records. Personally Identifiable Information (PII) from EHRs is passed through successive modules of the Person Identification Validation System (PVS), comparing Social Security number (SSN), address, name, sex, and full date of birth to a reference file maintained at the Census Bureau (Wagner & Layne, 2014). When a linkage could be made between the incoming record and the reference file, a PIK was appended to the patient record. The match to SSN is exact; matches involving the other identifiers are probabilistic (for additional information regarding PIK rates, see Udalova et al., 2022). Once a PIK was assigned, SSN and name were dropped, leaving information on sex, birth year, race, ethnicity, language, health insurance, and broad categories of residence for analysis. No health information was included in the transfer.

Our first interest was in whether the success of PIK assignment depended on race and ethnicity. There are a couple of reasons why it might (Bond et al., 2014). First, some patients may not be included in the federal and state data sources on which the reference file is based, e.g., recent immigrants. In North Carolina, the Hispanic population has increased rapidly, accounting for almost 10 percent of the overall population in 2019, the end of our window (U.S. Census Bureau, 2019a). Immigration played a major role in this change: about 40 percent of the Hispanic population was foreign-born in 2019 (U.S. Census Bureau, 2019b). A substantial portion of all immigrants in North Carolina were undocumented in 2016 (39%), most of which (56%) were from Mexico (Pew Research Center, 2019). Undocumented immigrants may purposely avoid being included in government data. Second, patients may be included in federal data sources but discrepancies between these sources and EHRs in the way information is recorded may undermine the match.

Once a PIK is assigned to the EHRs, we attempted to match to ACS data between 2001 and 2017. There are three reasons why we might fail to make a match. First, ACS data pertain to a sample of the state population, about 1-1.5% per year. An EHR that received a PIK might not match an ACS record because that person was not part of the ACS sample in any of the study years. Without knowing the population of eligible potential matches, it is not possible to say what the match rate should be, but it likely lies between 12 and 18 percent (Udalova et al., 2022). Match rates will likely be lower for recently arrived immigrants as they are less likely to have been included in the early years of the ACS. Second, there are longstanding differences in the coverage of racial and ethnic groups in the ACS. The Census Bureau measures coverage rates as the ratio of the ACS population of a group to an independent estimate for that group, times 100 (U.S. Census Bureau, 2022d). In 2017, the end of our observation window for the study’s use of ACS data, national coverage rates were 94.9 percent for White Non-Hispanic, 82.5 percent for Black Non-Hispanic, and 86.9 percent for Hispanic individuals (U.S. Census Bureau, 2022a). Although aggregate ACS population estimates are adjusted for under-coverage, some individuals who should have been included will be missing and not available for a match. Third and finally, only ACS records that have received a PIK can be matched to EHRs with PIKs. The challenges with PIK assignments noted for EHRs also apply to PIK assignments for the ACS (Bond et al., 2014).

Results

Table 2 shows unadjusted PIK and ACS match rates for racial and ethnic groups as well as for other social attributes available in the EHRs. Table 3 presents the unadjusted PIK and ACS match rates for racial and ethnic groups from Table 2 alongside PIK and ACS match rates adjusted for sex, birth cohort, language, health insurance, and residence. The adjusted rates are predicted probabilities from logistic regression models for each outcome, weighted to account for sample design. Of interest is the degree to which disparities narrow when we account for the (limited) information in the EHRs relevant to access and utilization. Full results are shown in Appendix Table A.1, which reports regression coefficients from linear probability models (LPM) and average marginal effects (AME) for logistic models. To aid in interpretation, Table 3 also includes information about national ACS PIK and coverage rates available from published sources. While we focus our interpretation of results on White, Black, and Hispanic patients, all groups are included in the tables.

Table 2.

Number and percent of patients receiving a PIK or matched to ACS (conditional on receiving a PIK) across EHR demographic characteristics (unweighted)

	Assigned a PIK		Matched to ACS, Conditional on Receiving a PIK
	N	Row %	N	Row %
Total	187,000	93.97	29,000	15.51
Race
White	55,000	98.65	9,800	17.82
Black	32,500	99.54	4,700	14.46
AIAN	3,200	98.46	600	18.75
Asian	16,500	96.77	2,100	12.73
Other	30,500	77.61	3,400	11.15
Missing/Refused	49,000	96.84	8,200	16.73
Ethnicity
Hispanic or Latino	23,500	71.65	2,400	10.21
Not Hispanic or Latino	111,000	98.49	17,500	15.77
Other/Missing	52,000	98.30	9,000	17.31
Sex
Male	76,000	94.53	11,500	15.13
Female	110,000	93.62	17,000	15.45
Birth Year
Before 1944	7,400	98.01	1,500	20.27
1945-1949	15,500	98.10	3,100	20.00
1950-1954	18,500	97.63	3,500	18.92
1955-1959	20,500	97.16	3,600	17.56
1960-1964	21,500	96.63	3,500	16.28
1965-1969	21,000	94.59	3,200	15.24
1970-1974	20,500	91.93	2,900	14.15
1975-1979	19,500	89.45	2,500	12.82
1980-1984	19,000	89.20	2,300	12.11
1985 and later	23,000	92.00	2,900	12.61
Language
English	160,000	98.40	26,000	16.25
Spanish	11,000	56.41	850	7.73
Unknown	11,000	96.49	1,800	16.36
Other	4,200	91.30	350	8.33
Health Insurance*
Any Private (Employer, State, Tricare)	115,000	98.63	18,500	16.09
Any Medicare	45,500	99.34	8,500	18.68
Any Medicaid/Gov’t Assistance	29,500	90.49	3,600	12.20
Any Uninsured/Self-Pay	58,500	85.40	7,700	13.16
Residence
In-state	129,000	93.48	19,000	14.73
Out-of-state	4,500	95.74	750	16.67
Missing	53,500	95.20	9,400	17.57

Source: Electronic Health Records (2016-2019) and American Community Survey data (2001-2017)

Note:

^*Insurance categories are not mutually exclusive.

AIAN = American Indian and Alaska Native. DMS #P-7519212.

All results were approved for release by the Disclosure Review Board of the U.S. Census Bureau, authorization number CBDRB-FY21-POP001-0087. All numeric values were rounded according to U.S. Census Bureau disclosure protocols to preserve data privacy.

Table 3.

Comparison of PIK and ACS match rates

	PIK rate, EHR (2016-2019)			ACS (2001-2017) match rate, conditional on receiving a PIK			Bond et al. 2014	ACS Coverage Rates
	Unadjusted and unweighted	Adjusted and weighted		Unadjusted and unweighted	Adjusted and weighted		Unadjusted PIK rate, ACS 2010	National, 2017
Race
White	98.65	99.67	(99.62, 99.72)	17.82	17.39	(17.00, 17.77)	93.45	94.9
Black	99.54	99.69	(99.64, 99.74)	14.46	14.53	(14.13, 14.92)	91.4	82.5
AIAN	98.46	99.39	(99.20, 99.58)	18.75	18.49	(17.12, 19.85)	91.03	91.5
Asian	96.77	98.22	(97.98, 98.46)	12.73	14.03	(13.42, 14.63)	90.81	92.6
Other	77.61	98.88	(98.75, 99.01)	11.15	14.7	(14.14, 15.25)	84.88	-
Missing/Refused*	96.84	98.98	(98.87, 99.10)	16.73	16.71	(16.26, 17.16)	-	-
Ethnicity
Hispanic/Latino	71.65	98.63	(98.46, 98.81)	10.21	14.56	(13.85, 15.26)	87.13	86.9
Not Hispanic/Latino	98.49	99.63	(99.58, 99.67)	15.77	16.59	(16.30, 16.87)	93.64	-
Other/Missing	98.30	99.55	(99.49, 99.61)	17.31	16.92	(16.42, 17.43)	-	-

Source: Electronic Health Records (2016-2019) and American Community Survey data (2001-2017)

Note: EHR sample is 199,000; 95% confidence intervals are in parentheses; The unadjusted PIK rates for 2010 ACS were reported in Bond et al. (2014). Coverage rates refer to combined race/ethnicity: non-Hispanic White, non-Hispanic Black, non-Hispanic AIAN, non-Hispanic Asian, and Hispanic; DMS #P-7519212. All results were approved for release by the Disclosure Review Board of the U.S. Census Bureau, authorization number CBDRB-FY21-POP001-0087 and CBDRB-FY23-POP001-0053. All numeric values were rounded according to U.S. Census Bureau disclosure protocols to preserve data privacy.

^*There is no missing data for ACS information as this information is imputed.

Beginning with unadjusted PIK rates, almost all patients who identified as White (98.65%) or Black (99.54%) were assigned PIKs (Table 3). These patients were well covered in administrative data systems, and the information provided to the health care system was of sufficiently high quality for PIK assignment. In contrast, only 77.61 percent of patients who identified their race as something “other” than the categories provided in the EHRs were assigned PIKs. With respect to ethnicity, 71.65 percent of patients who identified as Hispanic received PIKs, in contrast to a high degree of success for patients who did not identify as Hispanic (98.49%). Although broadly similar, we made two observations when comparing these patterns to unadjusted PIK rates for the 2010 ACS (Bond et al., 2014). First, PIK rates were higher for White and Black patients in the EHRs than for the population generally, probably because SSNs were available for many (although not all) of them. Second, PIK rates were lower for “other” race and Hispanic patients in our study, suggesting something distinctive about the population of patients in our EHRs relative to the national sample.

Next, we investigated the likelihood of receiving a PIK adjusting for age, sex, race, ethnicity, language, and insurance status as reported in the EHRs. Table 3 reports predicted probabilities based on logistic regression estimates, with all other variables held at their means. All adjusted PIK rates were quite high, with differences substantially narrowed relative to the unadjusted rates. Adjusted PIK rates for “other” race and for Hispanic patients were within 2 percentage points of the other racial and ethnic categories, indicating the importance of language, health insurance and to a lesser extent age and residence. These results suggest that a substantial portion of the difference in the unadjusted PIK rates reflect differences in health care access among racial and ethnic subpopulations.

Once PIKs are assigned to EHRs, the next step is to match them to individuals in the ACS who have been assigned PIKs. Unadjusted conditional ACS match rates were 17.82 percent for White patients,14.46 percent for Black patients, and 10.21 percent for Hispanic patients. Adjustments for sex, birth cohort, health insurance, and residence were of little consequence to conditional ACS match rates for White and Black patients but make a substantial difference for Hispanic patients. For Hispanic patients, the adjusted conditional ACS match rate was almost 40 percent higher than the unadjusted rate, again suggesting the importance of access and utilization. That there is not more of an improvement for Black patients is likely due to under-coverage in the ACS (Table 3). Although the ACS coverage rates refer to a combined race/ethnicity category rather than separate categories, they are nevertheless instructive. ACS coverage rates for 2017 were highest for Non-Hispanic White individuals (94.9%), lower for Hispanic individuals (86.9%), and lowest for Non-Hispanic Black individuals (82.5%). Additionally, the adjusted ACS match rates in Table 3 were highest for White patients and lowest for Black and Hispanic patients. It is important to remember that population estimates based on ACS data adjust for coverage.

Discussion

Prevalence is measured relative to a population at risk. Research that draws on EHRs to describe health disparities typically uses the total number of patients of a specific race or ethnicity for the denominator. However, because the patients included in EHRs may be differentially selected from the source population, these estimates are vulnerable to bias (Goldstein et al., 2016). Prevalence may be overstated if representation is better in the numerator than the denominator. If the extent of this bias differs by race or ethnicity, the interpretation of health disparities may be misestimated, which has implications for the use of EHRs for population research. The goal of this research note was to reflect on this potential bias by using information gleaned from linking EHRs from a large integrated health delivery system in North Carolina to ACS microdata.

The first step in this linkage, PIK assignment, provided information about the quality of personal information in the EHRs and patient representation in government data systems. In terms of our results, PIK assignments were successful. Availability of SSNs in many of the health records facilitated these assignments, helping to explain why PIK rates for the EHRs were higher than for the 2010 ACS. The exceptions were for patients who identified as “other” race and Hispanic ethnicity in the EHRs. Unadjusted PIK rates for these groups were 77.61 percent and 71.65 percent, respectively. The disproportionate representation of recent immigrants could help explain why PIK rates are so low for Hispanic patients. Problems with PIK assignments for Hispanic patients in our sample may also relate to documentation status.³ More than half of undocumented immigrants in North Carolina are Hispanic (Pew Research Center, 2019). Underutilization of health services among people with undocumented status is well known (Cabral & Cuevas, 2020), possibly for that reason but also due to lack of insurance (except in the case of pregnancy), but those urgently needing medical care may opt to take the risk. EHRs thus may include more undocumented immigrants than government sources, especially those in poor health.

The second step, matching to the ACS conditional on PIK assignment, provided information about representation in the ACS. Our procedures consider ACS data over a large interval, 2001-2017. A conditional ACS match rate that falls short of the average might be due to changes in racial and ethnic composition over that period. In North Carolina, the Hispanic population has increased, more than doubling between 2000 and 2010 and increasing by a further 40 percent the following decade (U.S. Census Bureau, 2020). Because the Hispanic population was smaller in earlier years of the ACS, we expected conditional ACS match rates to be lower as a result, and they were.⁴

Conditional ACS match rates also depend on ACS coverage, measures of which reflect inclusion of addresses in the frame as well as response rates for those found. Despite best efforts, and the fact that participation is mandatory, the ACS falls short of complete coverage. Challenges associated with enumerating the Black population are long standing and well documented (O’Hare, 2019). Lower coverage for the Hispanic population may reflect undocumented status in part. Not only are people with undocumented status less likely to participate in federal data collections, those with whom they live may also avoid participation, even those who are documented or citizens (Hall et al., 2019; Kopparam, 2022).

What does this mean for health disparities research based on EHRs? Our results suggest that differences between disease prevalence rates for Hispanic individuals relative to other groups could be misrepresented by biases in the EHRs. On one hand, our results suggest that undocumented immigrants may be better represented in EHRs than in government systems. Yet, this representation is likely selective, related to the need for health care. Compounding the problem is that a large proportion of Hispanic individuals are uninsured, almost three times that of the state population (31% compared to 11% in 2019, respectively) (North Carolina Justice Center, 2020). Using ACS data to adjust the denominator of the prevalence rate would be an improvement but would not fully correct the problem given that Hispanic individuals missing from government records are less likely to participate in the ACS as well.

The situation is different for Black patients. It is possible that lack of trust and other characteristics that reduce the participation of Black individuals in the ACS also lead these individuals to avoid interacting with the health care system. However, this seems unlikely. Based on high PIK rates, it appears that Black patients are part of government data collection more broadly. Further, the fraction of uninsured for Black individuals is about the same as for North Carolina as a whole (12% compared to 11%) (North Carolina Justice Center, 2020). Moreover, according to BRFSS data for 2019, Black North Carolinians saw a doctor in the past 12 months for routine medical care (checkups) at a rate roughly equivalent to White North Carolinians (84% compared to 80%) and substantially more than Hispanic North Carolinians (64%) (NC State Center for Health Statistics, 2019). In addition to differences in insurance coverage, differences in age structure and the healthy migrant effect for first generation Hispanic individuals help to explain this difference. Given all of this, using ACS data to correct the denominator of prevalence rates for potential underrepresentation seems like a reasonable approach for Black patients.

Of course, our analysis suffers from some limitations. The data come from a single health system, which while large, is one of several in the state of North Carolina. Its mission is to serve the health needs of all North Carolinians, so we have compared patients to the population of the state, but residents do have alternatives and coverage is far from complete. Second, our analysis assumed that EHR-based information about race and ethnicity reflected patient identities. The information was missing for about 10 percent of our sample, a significant fraction even if low compared to many other health systems. A detailed discussion of these patients is beyond the scope of this research note, but we would like to point out that PIK and conditional ACS match rates for this group were quite high. Third, the results may reflect some of the peculiarities of North Carolina, perhaps especially that the recent growth in the Hispanic population and the role immigration is playing in this growth. While our results may not completely generalize to other health systems or states, there are larger lessons learned, particularly about potential bias in the EHRs, the value of linking EHRs to external sources for assessment purposes, and finally the possibility of leveraging EHRs for research on social determinants and population health.

Appendix A

Table A.1.

Likelihood of Patient Receiving a PIK or Being Matched to ACS (Weighted)

	Panel A: Likelihood of patient receiving a PIK			Panel B: Likelihood of patient being matched to ACS (Conditional on receiving a PIK)
	LPM	Logit (AME)		LPM	Logit (AME)
	(1)	(2)		(3)	(4)
Race (White omitted)
Black	.001739**	.0005078		−.0287***	−.02876***
	(.000543)	(.001028)		(0.002887)	(.002888)
American Indian or Alaska Native	−.003379	−.00775**		0.01026	.01103
	(.002138)	(.002469)		(0.007168)	(.007257)
Asian	−.01966***	−.03095***		−.03355***	−.03382***
	(.001531)	(.002023)		(0.003646)	(.003724)
Other	−.0361***	−.01902***		−.02484***	−.02707***
	(.002024)	(.0009497)		(0.003147)	(.003415)
Missing/Refused	−.01107***	−.0169***		−.007349**	−.006844*
	(.001005)	(.001037)		(0.002822)	(.002837)
Ethnicity (Not Hispanic or Latino omitted)
Hispanic or Latino	−.04667***	−.02362***		−.01835***	−.02045***
	(.002696)	(.001326)		(0.003482)	(.00384)
Other/Missing	.001803	−.002226**		0.003325	.003391
	(.0009239)	(.000801)		(0.00282)	(.002838)
Sex (male omitted)	−.000552	.0002036		.01002***	.01002***
	(.0005158)	(.0005224)		(0.002776)	(.002787)
Birth Year (1944 and earlier omitted)
1945-1949	.0008198	.004613		−0.01222	−.01182
	(.001315)	(.003266)		(0.008432)	(.008172)
1950-1954	−.006132***	.004192		−0.01586	−.01584
	(.001383)	(.003385)		(0.008399)	(.008209)
1955-1959	.000422	.01009**		−.02243*	−.02151*
	(.001559)	(.003381)		(0.008895)	(.008913)
1960-1964	.003996**	.01316***		−.03238***	−.0312***
	(.001527)	(.00327)		(0.008932)	(.008968)
1965-1969	.003883*	.01131***		−.04566***	−.04446***
	(.001598)	(.003243)		(0.009023)	(.009056)
1970-1974	−.001623	.00726*		−.05629***	−.05523***
	(.001678)	(.003263)		(0.009062)	(.009084)
1975-1979	−.007128***	.004718		−.05859***	−.05781***
	(.001726)	(.003263)		(0.009237)	(.00928)
1980-1984	−.008131***	.003916		−.06681***	−.06648***
	(.001729)	(.003258)		(0.009208)	(.009228)
1985+	−.002657	.006345		−.05484***	−.05427***
	(.001725)	(.003256)		(0.009138)	(.009209)
Language (English omitted)
Spanish	−.3303***	−.03884***		−.04408***	−.06067***
	(.004517)	(.001803)		(0.003934)	(.004368)
Unknown	−.01176***	−.005569***		−0.003828	−.00387
	(.001803)	(.001121)		(0.005517)	(.005528)
Other	−.0523***	−.02448***		−.06348***	−.07045***
	(.005556)	(.002792)		(0.008197)	(.009028)
Health Insurance*
Any private insurance	.03612***	.02605***		.0171***	.0182***
	(.0009549)	(.0009136)		(0.003709)	(.003818)
Any had Medicare	.02602***	.02604***		.01136*	.0119*
	(.001189)	(.002181)		(0.005064)	(.004976)
Any Medicaid	.01676***	.006137***		−.01455***	−.01656***
	(.00104)	(.0006143)		(0.003844)	(.004362)
Any uninsured/self-pay	−.01398***	−.007856***		−.01009**	−.011***
	(.0006188)	(.0008687)		(0.003141)	(.003285)
Insurance data missing	.02151***	.004712		−0.001292	−.0008151
	(.005751)	(.002915)		(0.01929)	(.02151)
Residence (In-state omitted)
Out-of-state	−.005473*	−.007301**		−0.0001316	−.0001431
	(.002421)	(.002758)		(0.008889)	(.008683)
Missing residence	−.004614***	−.003478***		.0109***	.01043***
	(.0005306)	(.0006258)		(0.003179)	(.0031)
Constant	.9709***			.198***
	(.001742)			(0.009038)
R-squared	0.2921			0.008917
N	199,000	199,000		187,000	187,000

^*p<.05,

^**p<.01,

^***p<.001; two-tailed test

Source: Electronic Health Records (2016-2019) and American Community Survey data (2001-2017)

Note:

^*Insurance categories are not mutually exclusive.

AIAN = American Indian and Alaska Native, AME = average marginal effect, LPM = linear probability model.

Standard errors are in parentheses. DMS #P-7519212. All results were approved for release by the Disclosure Review Board of the U.S. Census Bureau, authorization numbers CBDRB-FY22-POP001-0141 and CBDRB-FY23-SEHSD003-021. All numeric values were rounded according to U.S. Census Bureau disclosure protocols to preserve data privacy