Introduction
Understanding the multilevel factors associated with controlled blood pressure is important to determine modifiable factors for future interventions, especially among populations living in poverty. This study identified clinically important factors associated with blood pressure control among patients receiving care in community health centers.
Methods
This study includes 31,089 patients with diagnosed hypertension by 2015 receiving care from 103 community health centers; aged 19–64 years; and with ≥1 yearly visit with ≥1 recorded blood pressure in 2015, 2016, and 2017. Blood pressure control was operationalized as an average of all blood pressure measurements during all the 3 years and categorized as controlled (blood pressure <140/90), partially controlled (mixture of controlled and uncontrolled blood pressure), or never controlled. Multinomial mixed-effects logistic regression models, conducted in 2022, were used to calculate unadjusted ORs and AORs of being in the never- or partially controlled blood pressure groups versus in the always-controlled group.
Results
A total of 50.5% had always controlled, 39.7% had partially controlled, and 9.9% never had controlled blood pressure during the study period. The odds of being partially or never in blood pressure control were higher for patients without continuous insurance (AOR=1.09; 95% CI=1.03, 1.16; AOR=1.18; 95% CI=1.07, 1.30, respectively), with low provider continuity (AOR=1.24; 95% CI=1.15, 1.34; AOR=1.28; 95% CI=1.13, 1.45, respectively), with a recent diagnosis of hypertension (AOR=1.34; 95% CI=1.20, 1.49; AOR=1.19; 95% CI=1.00, 1.42), with inconsistent antihypertensive medications (AOR=1.19; 95% CI=1.11, 1.27; AOR=1.26; 95% CI=1.13, 1.41, respectively), and with fewer blood pressure checks (AOR=2.14; 95% CI=1.97, 2.33; AOR=2.17; 95% CI=1.90, 2.48, respectively) than for their counterparts.
Conclusions
Efforts targeting continuous and consistent access to care, antihypertensive medications, and regular blood pressure monitoring may improve blood pressure control among populations living in poverty.
INTRODUCTION
Hypertension is the most common chronic condition in the U.S., affecting nearly 1 in 2 adults.
1
Most do not have their hypertension under control.1
,2
Uncontrolled hypertension is the largest single contributor to all-cause and cardiovascular mortality.3
Annually, hypertension contributes to about 600,000 U.S. deaths4
and an estimated $50 billion in medical spending.Centers for Disease Control and Prevention (CDC). Facts about hypertension. https://www.cdc.gov/bloodpressure/facts.htm. Accessed December 29, 2022.
5
There are many multilevel factors that could be associated with uncontrolled blood pressure (BP). Previous studies have shown that sex, race and ethnicity, income, education, language, rural residence,
6
,7
and health insurance8
, 9
, 10
, 11
, 12
, 13
are factors associated with uncontrolled BP. In addition, community-level indicators (e.g., area deprivation, racial segregation) have also been linked to BP control14
, 15
, 16
(with mixed results) and hypertension prevalence.- Angier H
- Green BB
- Fankhauser K
- et al.
Role of health insurance and neighborhood-level social deprivation on hypertension control following the affordable care act health insurance opportunities.
Soc Sci Med. 2020; 265113439https://doi.org/10.1016/j.socscimed.2020.113439
14
,17
, 18
, 19
, 20
, 21
, 22
, 23
Most of these studies rely on a single BP measurement to assess control, which does not differentiate patients who are always, partially, or never in control.2
,8
,9
,13
,24
Because of the cardiovascular health benefits of being in control,- Chobufo MD
- Gayam V
- Soluny J
- et al.
Prevalence and control rates of hypertension in the USA: 2017–2018.
Int J Cardiol Hypertens. 2020; 6100044https://doi.org/10.1016/j.ijchy.2020.100044
25
it is essential to understand the clinically important factors associated with BP control to identify modifiable factors for future interventions, especially among populations living in poverty.Thus, the objective of this study was to identify the clinically important factors associated with BP control among patients with hypertension who receive care at community health centers (CHCs).
26
CHCs primarily serve patients with socioeconomic disadvantages. Using a sliding payment scale, CHCs serve 1 in 6 Medicaid beneficiaries, 1 in 3 persons with low-income/without insurance, 1 in 3 individuals living below the federal poverty level, and 1 in 6 rural Americans.National Association of Community Health Centers, America's Health Centers. Snapshot 2022. https://www.nachc.org/research-and-data/americas-health-centers-2022-snapshot/. Accessed December 29, 2022.
26
The CHCs serve a population with complex problems such as health insurance instability, exposure to adverse social determinants of health (e.g., food/housing insecurity), and those who are often in worse health (e.g., higher prevalence of multimorbidity) than those seen in other settings.National Association of Community Health Centers, America's Health Centers. Snapshot 2022. https://www.nachc.org/research-and-data/americas-health-centers-2022-snapshot/. Accessed December 29, 2022.
26
, National Association of Community Health Centers, America's Health Centers. Snapshot 2022. https://www.nachc.org/research-and-data/americas-health-centers-2022-snapshot/. Accessed December 29, 2022.
27
Therefore, CHCs represent an ideal setting for understanding clinically important factors associated with BP control in low-income populations.Lewis C, Getachew Y, Abrams M, Doty M. Changes at community health centers, and how patients are benefiting: results from the Commonwealth Fund National survey of federally qualified health centers; 2013–2018. New York, NY: The Commonwealth Fund.https://www.commonwealthfund.org/publications/issue-briefs/2019/aug/changes-at-community-health-centers-how-patients-are-benefiting. Published August 2019. Accessed September 20, 2019.
METHODS
Electronic health record (EHR) data were obtained from the OCHIN community health information network, a multistate collaboration of CHCs.
28
Eligible clinics (n=103) were primary care clinics live on the OCHIN network's EHR platform throughout the study period, 2015–2017. Clinics were located in states that expanded Medicaid eligibility: Alaska, California, Indiana, Massachusetts, Minnesota, Montana, Ohio, Oregon, Washington, and Wisconsin. This study was approved by the Oregon Health & Science University IRB.Study Population
Patients included in this study were aged 19–64 years throughout the study period, with at least 1 visit in each study year (n=141,552). Patients who were pregnant were excluded (n=30,858). The sample was further restricted to those with hypertension in 2015 (n=42,029) and a minimum of 3 BP recordings (1 per year; excluding n=605). Among those, patients with a documented diagnosis of end-stage renal disease, a Medicare-financed visit during the study period (because patients aged <65 years with Medicare differ medically from those with other types of insurance, having multiple chronic conditions and/or disability),
29
or missing clinic site or rural‒urban status were also excluded (excluding n=10,335). The final sample included 31,089 patients.Measures
Average annual BP from all BP measurements available within a year and each visit was used to classify patient control status for each year. For example, if BP was recorded more than once during a visit, all BP measures were included in the average. BP can be highly variable from visit to visit, particularly if the patient is ill or in pain, thus using average BP is more likely to give a better estimate of overall control. Most patients had >1 BP per year (87%). Patients with an elevated blood BP were those with an average yearly value ≥140 mmHg systolic and/or 90 mmHg diastolic BP (Stage 2 hypertension). The 3 average annual BP values were combined to create a single BP control variable for 2015–2017 as (1) controlled BP during all the 3 years (referred to as always controlled), (2) a mixture of controlled and uncontrolled BP during the study years (referred to as partially controlled), or (3) never controlled BP during all the 3 years (referred to as never controlled).
Sociodemographic indicators included age at study start (<50 or ≥50 years), sex (male, female), race (White, Black, Asian, other, unknown), ethnicity (Hispanic, non-Hispanic, unknown), preferred language (English, Spanish, other, unknown), health insurance (continuously insured, discontinuously insured, uninsured), and household income as a percent of the federal poverty level (≤138%, >138%, missing). Earliest recorded weight during the study was used to calculate BMI (<25, 25 to <30, 30 to <40, ≥40 kg/m2, missing), and a modified Charlson comorbidity index score was calculated (excluding the hypertension diagnosis because all included patients had hypertension).
30
Using patient address recorded nearest to the study start, geographic attributes based on the specificity of address geocoding were assigned: 88% of patients had census tract level, 0.9% had ZIP-code tabulation area, and 10.7% had county-level information available. Rurality was classified using Rural-Urban Commuting Area Codes as urban; largely rural; or small, isolated rural.
31
To quantify disadvantages in the patient's community, Social Deprivation Index (SDI) percentile at the study start was utilized.USDA. Rural-urban commuting area codes. U.S. Department of Agriculture. https://www.ers.usda.gov/data-products/rural-urban-commuting-area-codes/. Accessed December 29, 2022.
16
,- Angier H
- Green BB
- Fankhauser K
- et al.
Role of health insurance and neighborhood-level social deprivation on hypertension control following the affordable care act health insurance opportunities.
Soc Sci Med. 2020; 265113439https://doi.org/10.1016/j.socscimed.2020.113439
32
Provider continuity refers to the percentage of all in-person ambulatory visits that occurred with the most visited provider. The number of unique dates with BP measurements was summed for each patient. Measurement date counts, SDI, and provider continuity were categorized by quartiles for analysis.
Patients were required to have hypertension by December 31, 2015. Duration of hypertension was assessed as the time between the first recorded hypertension diagnosis and December 31, 2015. An indicator for patients with an active hypertension medication order at the start of the study was included. In addition, the continuity of medication during the study was assessed using the date of medication order and projected end date (using quantity of order or recorded end date). A gap of 4 weeks was used to indicate an interruption in medication order continuity. Patients with no order gaps during the study were classified as consistent, patients with medications ordered but having 1 or more gaps were classified as inconsistent, and patients with no medication orders during the study were classified as never.
Statistical Analysis
First, differences between the always, partially, and never controlled BP groups were described using frequency and relative frequency counts. Next, univariable (for each indicator) and multivariable analyses, using multinomial mixed-effects logistic regression with a random CHC intercept (to take into account the correlation between patients with the same CHC), were performed to compare the odds of partially controlled or never controlled BP with those of the reference group of always controlled. Multivariate multinomial mixed-effects logistic regression models were adjusted for sociodemographic indicators, clinical metrics, and care access. All analyses were conducted using SAS, Version 9.3, and statistical significance was set at a Type I error of 5%.
RESULTS
Among patients with hypertension, 50.5% had always controlled BP, 39.7% had partially controlled BP, and 9.9% had BP that was never controlled during the study period. Table 1 displays the patient characteristics for the 3 BP control groups. The partially controlled and the never-controlled groups had higher proportions of patients who were male, were Black, were non-Hispanic, spoke English as their primary language, resided in rural areas, and lived with hypertension for 1 year or less than those with always controlled BP. There was a higher proportion of patients with fewer BP measurements and inconsistent receipt of antihypertensive medication in the partially and never-controlled groups than in the always-controlled BP group.
Table 1Characteristics of Patients With Low SES and Hypertension With BP Partially, Never, or Controlled During All 3 Years, 2015 Through 2017
| Characteristics | Always- controlled BP (n=15,690; 50.5%), | Partially controlled BP (n=12,330; 39.7%), | Never-controlled BP (n=3,069; 9.9%), |
|---|---|---|---|
| n (%) | n (%) | n (%) | |
| Sex | |||
| Male | 6,302 (40.2) | 5,892 (47.8) | 1,582 (51.5) |
| Female | 9,388 (59.8) | 6,438 (52.2) | 1,487 (48.5) |
| Age at study start, years | |||
| <50 | 6,378 (40.7) | 5,366 (43.5) | 1,322 (43.1) |
| ≥50 | 9,312 (59.3) | 6,964 (56.5) | 1,747 (56.9) |
| Race | |||
| White | 11,584 (73.8) | 8,729 (70.8) | 1,972 (64.3) |
| Asian | 608 (3.9) | 399 (3.2) | 99 (3.2) |
| Black | 2,638 (16.8) | 2,539 (20.6) | 838 (27.3) |
| Other | 362 (2.3) | 282 (2.3) | 67 (2.2) |
| Unknown | 498 (3.2) | 381 (3.1) | 93 (3.0) |
| Ethnicity | |||
| Non-Hispanic | 10,964 (69.9) | 9,062 (73.5) | 2,309 (75.2) |
| Hispanic | 4,448 (28.3) | 2,997 (24.3) | 698 (22.7) |
| Unknown | 278 (1.8) | 271 (2.2) | 62 (2.0) |
| Preferred language | |||
| English | 11,263 (71.8) | 9,479 (76.9) | 2,375 (77.4) |
| Spanish | 3,618 (23.1) | 2,262 (18.3) | 520 (16.9) |
| Other | 770 (4.9) | 544 (4.4) | 164 (5.3) |
| Unknown | 39 (0.2) | 44 (0.4) | 11 (0.4) |
| % of federal poverty level | |||
| >138% | 2,464 (15.7) | 2,147 (17.4) | 522 (17.0) |
| ≤138% | 13,010 (82.9) | 9,979 (80.9) | 2,477 (80.7) |
| Unknown | 216 (1.4) | 204 (1.7) | 70 (2.3) |
| Health insurance patterns | |||
| Continuously insured | 10,815 (68.9) | 8,518 (69.1) | 2,088 (68.0) |
| Discontinuously insured | 4,114 (26.2) | 3,212 (26.1) | 836 (27.2) |
| Continuously uninsured | 761 (4.9) | 600 (4.9) | 145 (4.7) |
| Rurality of residence | |||
| Urban | 12,571 (80.1) | 9,345 (75.8) | 2,281 (74.3) |
| Large rural | 2,206 (14.1) | 2,182 (17.7) | 616 (20.1) |
| Small/Isolated small rural | 913 (5.8) | 803 (6.5) | 172 (5.6) |
| Social deprivation index score | |||
| <54 | 4,076 (26.0) | 3,140 (25.5) | 706 (23.0) |
| 54 to <76 | 4,011 (25.6) | 3,306 (26.8) | 853 (27.8) |
| 76 to <90 | 3,373 (21.5) | 2,765 (22.4) | 696 (22.7) |
| ≥90 | 4,230 (27.0) | 3,119 (25.3) | 814 (26.5) |
| BMI | |||
| Normal: <25 kg/m2 | 1,851 (11.8) | 1,415 (11.5) | 322 (10.5) |
| Overweight: 25<30 | 4,119 (26.3) | 3,261 (26.4) | 752 (24.5) |
| Obese I and II: 30 to <40 | 6,791 (43.3) | 5,348 (43.4) | 1,385 (45.1) |
| Obese III: ≥40 | 2,870 (18.3) | 2,241 (18.2) | 591 (19.3) |
| Unknown | 59 (0.4) | 65 (0.5) | 19 (0.6) |
| Charlson comorbidity score | |||
| 0 | 3,069 (19.6) | 3,048 (24.7) | 766 (25.0) |
| 1–2 | 5,922 (37.7) | 4,313 (35.0) | 1,064 (34.7) |
| 3–5 | 4,804 (30.6) | 3,712 (30.1) | 927 (30.2) |
| ≥6 | 1,895 (12.1) | 1,257 (10.2) | 312 (10.2) |
| Provider continuity | |||
| <43% | 4,266 (27.2) | 3,387 (27.5) | 894 (29.1) |
| 43 to <60% | 3,805 (24.3) | 2,836 (23.0) | 736 (24.0) |
| 60 to <79% | 3,936 (25.1) | 3,235 (26.2) | 730 (23.8) |
| ≥79% | 3,683 (23.5) | 2,872 (23.3) | 709 (23.1) |
| Time since hypertension diagnosis | |||
| <1 year | 2,479 (15.8) | 2,888 (23.4) | 742 (24.2) |
| 1–3 years | 5,446 (34.7) | 3,829 (31.1) | 912 (29.7) |
| 4–5 years | 4,399 (28.0) | 3,276 (26.6) | 807 (26.3) |
| >5 years | 3,366 (21.5) | 2,337 (19.0) | 608 (19.8) |
| Antihypertensive medication at the study start (January 1, 2015) | |||
| Yes | 11,175 (71.2) | 8,034 (65.2) | 2,016 (65.7) |
| No | 4,515 (28.8) | 4,296 (34.8) | 1,053 (34.3) |
| Blood-pressure measurement, unique dates | |||
| <7 | 2,233 (14.2) | 2,547 (20.7) | 623 (20.3) |
| 7–10 | 4,007 (25.5) | 3,786 (30.7) | 889 (29.0) |
| 11–15 | 4,431 (28.2) | 3,120 (25.3) | 781 (25.4) |
| ≥16 | 5,019 (32.0) | 2,877 (23.3) | 776 (25.3) |
| Antihypertensive medication continuity, 2015–2017 | |||
| Consistent order | 11,221 (71.5) | 8,755 (71.0) | 2,218 (72.3) |
| Inconsistent order | 3,045 (19.4) | 2,951 (23.9) | 782 (25.5) |
| Never received order | 1,424 (9.1) | 624 (5.1) | 69 (2.2) |
a Always controlled BP refers to patients with controlled BP during all the 3 years, partially controlled are those with a mixture of controlled and uncontrolled BP during study years, and never-controlled BP refers to those without BP control during all the 3 years.
b Higher scores on the Social Deprivation Index reflect greater neighborhood disadvantage.
c Provider continuity measures by the percentage of visits with the same provider.
d Number of BP measurements represents unique visit dates with BP records.BP, blood pressure.
In adjusted models (Table 2), non-Hispanic Black patients had more than twice the odds (AOR=2.37; 95% CI=2.05, 2.74) of never having their BP controlled than non-Hispanic White patients and elevated odds of partially controlled BP (AOR=1.38; 95% CI=1.26, 1.52). Female patients were less likely than males to have partially or never-controlled BP (AOR=0.79; 95% CI=0.75, 0.83 and AOR=0.65; 95% CI=0.60, 0.71, respectively). Patients whose preferred language was Spanish had lower odds of having never or partially controlled BP (AOR=0.76; 95% CI=0.67, 0.85 and AOR=0.72; 95% CI=0.59, 0.87, respectively), whereas those whose preferred language was neither English nor Spanish were at increased odds of never-controlled BP (AOR=1.28; 95% CI=1.02, 1.61) compared with English-speaking patients. Patients without continuous insurance were more likely to have never-controlled or partially controlled BP (AOR=1.18; 95% CI=1.07, 1.30; AOR=1.09; 95% CI=1.03, 1.16, respectively) than continuously insured patients. Patients with obesity were more likely to have never-controlled BP (BMI=30–40: AOR=1.27; 95% CI=1.11, 1.46; BMI≥40: AOR=1.36; 95% CI=1.16, 1.59) than patients with lower BMI. Patients with less provider continuity (<43%) had higher odds of never-controlled or partially controlled BP than those seen by the same provider for 79% or more of their visits (AOR=1.28; 95% CI=1.13, 1.45 and AOR=1.24; 95% CI=1.15, 1.34, respectively). Patients with a recent diagnosis of hypertension (AOR=1.34; 95% CI=1.20, 1.49) were more likely to have partially controlled BP than those with a diagnosis dating back 5 years or more. Patients with inconsistent orders for antihypertensive medications during the study were more likely to have never-controlled (AOR=1.26; 95% CI=1.13, 1.41) or partially controlled BP (AOR=1.19; 95% CI=1.11, 1.27) than those with consistent medication orders; the pattern did not hold for those who never had an order for antihypertensive medications (AOR=0.16; 95% CI=0.12, 0.22 for never controlled relative to always controlled). The odds of a patient's BP being never controlled or partially controlled relative to being always controlled declined with an increasing number of visits with a BP measurement.
Table 2Factors Associated With Partially or Never Controlled BP Among Patients With Hypertension Receiving Care in Community Health Centers
| Characteristics | Partially controlled versus always-controlled BP, OR (95% CI) | Never-controlled versus always-controlled BP, OR (95% CI) | ||
|---|---|---|---|---|
| Unadjusted | Adjusted | Unadjusted | Adjusted | |
| Sex | ||||
| Male | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Female | 0.74 (0.71, 0.78) | 0.79 (0.75, 0.83) | 0.62 (0.58, 0.68) | 0.65 (0.60, 0.71) |
| Age at study start, years | ||||
| <50 | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| ≥50 | 0.90 (0.86, 0.95) | 0.97 (0.92, 1.02) | 0.91 (0.84, 0.99) | 1.00 (0.92, 1.09) |
| Race | ||||
| White | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Asian | 0.87 (0.76, 1.00) | 0.78 (0.66, 0.91) | 0.98 (0.77, 1.23) | 0.78 (0.59, 1.02) |
| Black | 1.42 (1.30, 1.55) | 1.38 (1.26, 1.52) | 2.48 (2.16, 2.84) | 2.37 (2.05, 2.74) |
| Other | 1.03 (0.88, 1.21) | 1.02 (0.86, 1.20) | 1.13 (0.86, 1.48) | 1.11 (0.84, 1.46) |
| Unknown | 1.06 (0.92, 1.22) | 1.00 (0.86, 1.17) | 1.24 (0.98, 1.57) | 1.20 (0.93, 1.54) |
| Ethnicity | ||||
| Non-Hispanic | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Hispanic | 0.88 (0.82, 0.94) | 1.07 (0.96, 1.19) | 0.70 (0.62, 0.79) | 0.98 (0.82, 1.16) |
| Unknown | 1.15 (0.96, 1.37) | 1.10 (0.91, 1.33) | 1.10 (0.82, 1.48) | 1.04 (0.76, 1.42) |
| Preferred language | ||||
| English | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Spanish | 0.79 (0.73, 0.85) | 0.76 (0.67, 0.85) | 0.64 (0.57, 0.73) | 0.72 (0.59, 0.87) |
| Other | 0.84 (0.74, 0.96) | 1.01 (0.87, 1.17) | 0.98 (0.80, 1.19) | 1.28 (1.02, 1.61) |
| Unknown | 1.05 (0.68, 1.64) | 1.01 (0.64, 1.60) | 0.84 (0.41, 1.74) | 0.84 (0.40, 1.78) |
| % of federal poverty level | ||||
| >138% | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| ≤138% | 0.91 (0.85, 0.98) | 1.02 (0.95, 1.09) | 0.87 (0.78, 0.97) | 0.96 (0.85, 1.07) |
| Unknown | 1.07 (0.87, 1.32) | 0.88 (0.71, 1.09) | 1.55 (1.14, 2.10) | 1.26 (0.92, 1.72) |
| Health insurance patterns | ||||
| Continuously insured | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Discontinuously insured | 1.06 (1.00, 1.12) | 1.09 (1.03, 1.16) | 1.14 (1.04, 1.25) | 1.18 (1.07, 1.30) |
| Continuously uninsured | 1.10 (0.98, 1.24) | 1.04 (0.91, 1.18) | 1.21 (0.99, 1.47) | 1.21 (0.98, 1.50) |
| Rurality of residence | ||||
| Urban | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Large rural | 1.07 (0.92, 1.25) | 1.14 (0.98, 1.34) | 1.14 (0.88, 1.49) | 1.31 (0.99, 1.72) |
| Small/Isolated small rural | 1.07 (0.91, 1.26) | 1.15 (0.97, 1.36) | 0.87 (0.64, 1.17) | 1.00 (0.73, 1.36) |
| Social deprivation index score | ||||
| <54 | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| 54 to <76 | 1.01 (0.94, 1.08) | 1.04 (0.97, 1.12) | 1.10 (0.98, 1.23) | 1.12 (1.00, 1.26) |
| 76 to <90 | 1.05 (0.97, 1.13) | 1.09 (1.01, 1.17) | 1.05 (0.92, 1.19) | 1.05 (0.92, 1.19) |
| ≥90 | 0.97 (0.90, 1.05) | 1.04 (0.96, 1.13) | 0.97 (0.85, 1.11) | 1.01 (0.88, 1.16) |
| BMI | ||||
| Normal: <25 kg/m2 | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Overweight: 25<30 | 1.05 (0.97, 1.15) | 1.04 (0.96, 1.14) | 1.07 (0.93, 1.24) | 1.06 (0.91, 1.22) |
| Obese I and II: 30 to <40 | 1.05 (0.97, 1.13) | 1.08 (0.99, 1.17) | 1.21 (1.06, 1.39) | 1.27 (1.11, 1.46) |
| Obese III: ≥40 | 1.02 (0.93, 1.12) | 1.10 (1.00, 1.21) | 1.22 (1.05, 1.42) | 1.36 (1.16, 1.59) |
| Unknown | 1.43 (1.00, 2.06) | 1.41 (0.97, 2.05) | 1.97 (1.14, 3.40) | 1.94 (1.11, 3.41) |
| Charlson comorbidity score | ||||
| 0 | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| 1–2 | 0.73 (0.68, 0.78) | 0.81 (0.76, 0.87) | 0.72 (0.64, 0.79) | 0.79 (0.71, 0.88) |
| 3–5 | 0.75 (0.70, 0.81) | 0.85 (0.79, 0.91) | 0.77 (0.69, 0.85) | 0.84 (0.75, 0.95) |
| ≥6 | 0.64 (0.58, 0.70) | 0.77 (0.70, 0.85) | 0.66 (0.57, 0.76) | 0.78 (0.66, 0.91) |
| Provider continuity | ||||
| <43% | 1.07 (1.00, 1.16) | 1.24 (1.15, 1.34) | 1.07 (0.95, 1.21) | 1.28 (1.13, 1.45) |
| 43 to <60% | 0.99 (0.92, 1.07) | 1.12 (1.04, 1.21) | 1.00 (0.89, 1.13) | 1.15 (1.01, 1.30) |
| 60 to <79% | 1.08 (1.01, 1.16) | 1.17 (1.09, 1.26) | 0.98 (0.87, 1.10) | 1.08 (0.95, 1.21) |
| ≥79% | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Time since hypertension diagnosis | ||||
| <1 year | 1.74 (1.61, 1.89) | 1.34 (1.20, 1.49) | 1.67 (1.47, 1.90) | 1.19 (1.00, 1.42) |
| 1–3 years | 1.05 (0.98, 1.13) | 0.98 (0.91, 1.06) | 0.94 (0.83, 1.06) | 0.87 (0.77, 0.99) |
| 4–5 years | 1.06 (0.98, 1.14) | 1.01 (0.93, 1.09) | 0.96 (0.84, 1.09) | 0.91 (0.8, 1.04) |
| >5 years | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Antihypertensive medication at study start (January 1, 2015) | ||||
| Yes | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| No | 1.34 (1.27, 1.41) | 1.32 (1.21, 1.45) | 1.31 (1.21, 1.43) | 1.43 (1.23, 1.65) |
| Blood-pressure measurement, unique dates | ||||
| <7 | 2.08 (1.92, 2.24) | 2.14 (1.97, 2.33) | 2.07 (1.83, 2.34) | 2.17 (1.90, 2.48) |
| 7–10 | 1.70 (1.59, 1.82) | 1.71 (1.59, 1.83) | 1.59 (1.43, 1.77) | 1.61 (1.44, 1.81) |
| 11–15 | 1.27 (1.19, 1.36) | 1.28 (1.20, 1.37) | 1.22 (1.10, 1.37) | 1.24 (1.11, 1.39) |
| ≥16 | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Antihypertensive medication continuity, 2015–2017 | ||||
| Consistent order | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) |
| Inconsistent order | 1.23 (1.16, 1.31) | 1.19 (1.11, 1.27) | 1.33 (1.21, 1.46) | 1.26 (1.13, 1.41) |
| Never received order | 0.55 (0.49, 0.60) | 0.39 (0.35, 0.45) | 0.23 (0.18, 0.30) | 0.16 (0.12, 0.22) |
Note: Bolded estimates were significant at p<0.05.
Multinomial mixed-effects logistic regression models with a random CHC intercept were used to compute adjusted (multivariate) and unadjusted (univariate) ORs and account for the clustering of patients within CHC. The ref category is always controlled BP.
a Always controlled BP refers to patients with controlled BP during all the 3 years, partially controlled are those with a mixture of controlled and uncontrolled BP during the study years, and never-controlled BP refers to those with uncontrolled BP during all the 3 years.
b Higher score on the Social Deprivation Index reflects greater neighborhood disadvantage.
c Provider continuity measures by the percentage of visits with the same provider.
d Number of BP measurements represents unique visit dates with BP records.BP, blood pressure; CHC, community health center.
DISCUSSION
Among CHC patients with hypertension, almost 50.5% had controlled BP over the study period, with hypertension control rates somewhat higher than those reported by studies using National Health and Nutrition Examination Survey (NHANES) data over a similar time period
2
but lower than those reported in Kaiser health systems.33
Fortunately, only 10% of patients with hypertension had BP that was never controlled, with 39.7% having partially controlled BP during the study period.The results showed a number of indicators and potentially modifiable factors associated with protection against partially or never-controlled BP, including continuous health insurance coverage, greater number of visits with a BP measurement, increased continuity of care with the same provider, and being on antihypertensive medications at the start of and without medication gaps during the study period. These factors were likely protective for 2 reasons. First, continuous health insurance provides patients with ample opportunity to receive health care. Indeed, the Patient Protection and Affordable Care Act significantly improved access to health insurance, especially among CHC patients, which in turn improved the receipt of preventive care and recommended medication prescribing.
34
, 35
, 36
, 37
, 38
Continued effort toward affordable and comprehensive health insurance for all is one of the strategies that could improve hypertension control among populations living in poverty. Furthermore, a recent study within the Veteran's Administration found that disruptions of care caused by the coronavirus disease 2019 (COVID-19) pandemic led to worsening hypertension control.- Gemelas J
- Marino M
- Valenzuela S
- Schmidt T
- Suchocki A
- Huguet N.
Changes in diabetes prescription patterns following Affordable Care Act Medicaid expansion.
BMJ Open Diabetes Res Care. 2021; 9e002135https://doi.org/10.1136/bmjdrc-2021-002135
39
Remote, home-based telehealth care that includes home BP monitoring might ameliorate such disruptions but is likely less accessible to patients seen in CHCs.- Baum A
- Barnett ML
- Wisnivesky J
- Schwartz MD.
Association between a temporary reduction in access to health care and long-term changes in hypertension control among veterans after a natural disaster.
JAMA Netw Open. 2019; 2e1915111https://doi.org/10.1001/jamanetworkopen.2019.15111
40
Additional funding and other support to ensure BP self-monitoring capability could assist CHCs in providing telehealth options to their patients. Second, having a greater number of visits with a BP measurement and increased continuity of care with the same provider would provide patients with more opportunities to track and respond to BP readings and discuss lifestyle and medication changes with their providers.Other well-documented risk factors for uncontrolled hypertension, including Black race,
8
male sex, and obesity, persisted and were not attenuated after controlling for patient-level (e.g., insurance, BP visits) or social (e.g., SDI, income, rural residence) indicators. The current results are similar to those of a study reported by He et al.10
using NHANES data, which found even after adjusting for patients’ social context (i.e., education, income, employment, health insurance, access to health care, and home ownership) that mean systolic BP remained higher among Black than among White individuals. In contrast, Egan and colleagues,9
also using NHANES data, found that adjusting for factors such as education, obesity, and access to care reduced disparities in BP control between Black and White individuals. Differences between these contradictory findings might be due, in part, to data inconsistencies and/or lack of longitudinal data. This study was strengthened by the use of longitudinal data and the inclusion of patients who had access to care over a 3-year period.The findings showed no association between SDI and hypertension control. Similarly, income level had little effect, whereas Spanish language was protective against never-controlled BP. Previous studies showed that non-U.S.‒born and non-English‒speaking Hispanic immigrants have lower rates of hypertension than U.S.-born or English-speaking immigrants.
11
,41
CHCs care for a large proportion of migrant farm workers who may have resided in the U.S. for fewer years than English-speaking patients. Studies11
,41
, 42
, - Brown SA
- Becker HA
- García AA
- et al.
The effects of gender and country of origin on acculturation, psychological factors, lifestyle factors, and diabetes-related physiological outcomes among Mexican Americans: the Starr County diabetes prevention initiative.
Chronic Illn. 2022; (In press. Online March 24)https://doi.org/10.1177/17423953221089315
43
, 44
have shown the negative impact of acculturation on Hispanic populations who have recently immigrated to the U.S., which may explain why patients of Hispanic ethnicity were at lower risk for having never-controlled BP in this study.Interestingly, never being on an antihypertensive medication was protective against never having controlled BP, possibly because patients without hypertensive medications had mild hypertension that could be controlled with lifestyle modifications, were misclassified as having hypertension, had few BP measurements, or had white-coat hypertension. However, those without hypertensive medications throughout the study period represented only 7% of the study cohort. In contrast, a recent diagnosis of hypertension increased the risk of partially controlled and to a lesser degree never-controlled BP, possibly because more time was needed to establish ongoing continuous hypertension control and/or provider continuity.
This study has several strengths, including longitudinal assessments of hypertension control and inclusion of clinical-level indicators that are not available in most other epidemiologic studies. In addition, CHCs routinely collect patient-level information that is not available in non-CHC settings, including income level and patients’ language preferences.
Limitations
This study has several limitations, including a lack of data that are not routinely collected in EHRs such as education level. The results cannot be adjusted for differences in BP measurement protocols across CHCs because contextual information on clinician training or the use of protocols (such as assuring 5 minutes of rest) was not available. BP control could also be misclassified, with some patients being normotensive outside of the clinic setting (white-coat hypertension or effect) or normotensive in the clinic and high outside of the clinic (masked hypertension). However, hypertension is generally diagnosed and treated on the basis of clinic measurements. The white-coat and masked hypertension groups could be more prevalent among the 7% of patients with hypertension who did not have an antihypertensive medication. That said, individuals with a documented white-coat syndrome are 2‒3 times more likely to develop hypertension
45
,46
and may benefit from interventions aimed at increasing BP control. In addition, SDI was mostly based on census tract and may not capture neighborhood differences and neighborhood segregation, which in a recent study predicted incident hypertension.47
This study defined hypertension control as BP<140/90 mmHg, yet newer guidelines recommend BP<130/80 mmHg.- Gao X
- Kershaw KN
- Barber S
- et al.
Associations between residential segregation and incident hypertension: the multi-ethnic study of atherosclerosis.
J Am Heart Assoc. 2022; 11e023084https://doi.org/10.1161/JAHA.121.023084
48
These guidelines were released at the end of the study period, and quality performance metrics still define hypertension control as BP<140/90 mmHg.- Whelton PK
- Carey RM
- Aronow WS
- et al.
2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice guidelines.
49
Furthermore, other unmeasured factors may account for variation in BP control. Finally, these observational data cannot be used to determine causality.NCQA. Controlling high blood pressure (CBP). https://www.ncqa.org/hedis/measures/controlling-high-blood-pressure/. Accessed December 29, 2022.
CONCLUSIONS
More than half of patients with hypertension receiving care in CHCs had controlled BP over 3 years. Efforts targeting continuous health insurance coverage and antihypertensive medication access, provider continuity, and frequent BP monitoring may improve BP control in populations with low SES.
ACKNOWLEDGMENTS
The authors acknowledge the participation of the partnering community health centers.
The views presented in this article are solely the responsibility of the authors and do not necessarily represent the views of the funding agencies.
OCHIN data are aggregated in the Accelerating Data Value Across a National Community Health Center Network (ADVANCE) Clinical Research Network. ADVANCE is led by OCHIN in partnership with Health Choice Network, Fenway Health, and the Robert Graham Center. ADVANCE is funded through the Patient-Centered Outcomes Research Institute (Contract Number RI-OCHIN-01-MC). This publication was supported by the National Health, Lung, and Blood Institute Grant Number R01HL136575.
No financial disclosures were reported by the authors of this paper.
CRediT AUTHOR STATEMENT
Nathalie Huguet: Conceptualization, Methodology, Project administration, Writing - original draft, Writing - review & editing, supervision. Beverly B. Green: Conceptualization, Methodology, Writing - original draft, Writing - review & editing. Robert W. Voss: Software, Formal analysis, Data curation, Writing - review & editing. Annie E. Larson: Conceptualization, Methodology, Writing - review & editing. Heather Angier: Conceptualization, Methodology, Writing - review & editing. Miguel Marino: Conceptualization, Methodology, Writing - review & editing. Shuling Liu: Conceptualization, Methodology, Writing - review & editing. Michaella Latkovic-Taber: Software, Formal analysis, Data curation, Writing - review & editing. Jennifer E. DeVoe: Conceptualization, Methodology, Writing - review & editing, Funding acquisition.
REFERENCES
- Hypertension prevalence among adults aged 18 and over: United States, 2017–2018.NCHS Data Brief. 2020; 364 (Accessed December 29, 2022): 1-8
- Trends in blood pressure control among U.S. adults with hypertension, 1999–2000 to 2017–2018.JAMA. 2020; 324: 1190-1200https://doi.org/10.1001/jama.2020.14545
- Global burden of hypertension and systolic blood pressure of at least 110 to 115 mm Hg, 1990–2015 [published correction appears in JAMA. 2017;317(6):648].JAMA. 2017; 317: 165-182https://doi.org/10.1001/jama.2016.19043
Centers for Disease Control and Prevention (CDC). Facts about hypertension. https://www.cdc.gov/bloodpressure/facts.htm. Accessed December 29, 2022.
- Heart disease and stroke statistics-2020 update: a report from the American Heart Association.Circulation. 2020; 141: e139-e596https://doi.org/10.1161/CIR.0000000000000757
Ellis C, Grubaugh AL, Egede LE. The effect of minority status and rural residence on actions to control high blood pressure in the U.S. Public Health Rep. 2010;125(6):801–809. https://doi.org/10.1177/003335491012500607.
- Is insurance instability associated with hypertension outcomes and does this vary by race/ethnicity?.BMC Health Serv Res. 2020; 20: 216https://doi.org/10.1186/s12913-020-05095-8
- Racial/ethnic disparities in hypertension prevalence, awareness, treatment, and control in the United States, 2013 to 2018.Hypertension. 2021; 78: 1719-1726https://doi.org/10.1161/HYPERTENSIONAHA.121.17570
- Hypertension control in the United States 2009 to 2018: factors underlying falling control rates during 2015 to 2018 across age- and race-ethnicity groups.Hypertension. 2021; 78: 578-587https://doi.org/10.1161/HYPERTENSIONAHA.120.16418
- Trends in cardiovascular risk factors in U.S. adults by race and ethnicity and socioeconomic status, 1999–2018.JAMA. 2021; 326: 1286-1298https://doi.org/10.1001/jama.2021.15187
- Nativity, language spoken at home, length of time in the United States, and race/ethnicity: associations with self-reported hypertension.Am J Hypertens. 2014; 27: 237-244https://doi.org/10.1093/ajh/hpt209
- Sociodemographics and hypertension control among young adults with incident hypertension: a multidisciplinary group practice observational study.J Hypertens. 2018; 36: 2425-2433https://doi.org/10.1097/HJH.0000000000001872
- Are demographic characteristics, health care access and utilization, and comorbid conditions associated with hypertension among U.S. adults?.Am J Hypertens. 2008; 21: 159-165https://doi.org/10.1038/ajh.2007.32
- Neighborhood characteristics and hypertension.Epidemiology. 2008; 19: 590-598https://doi.org/10.1097/EDE.0b013e3181772cb2
- Neighborhood disadvantage and chronic disease management.Health Serv Res. 2019; 54: 206-216https://doi.org/10.1111/1475-6773.13092
- Role of health insurance and neighborhood-level social deprivation on hypertension control following the affordable care act health insurance opportunities.Soc Sci Med. 2020; 265113439https://doi.org/10.1016/j.socscimed.2020.113439
- Neighborhood stressors and race/ethnic differences in hypertension prevalence (the Multi-Ethnic Study of Atherosclerosis).Am J Hypertens. 2011; 24: 187-193https://doi.org/10.1038/ajh.2010.200
- Neighborhood walkability and cardiometabolic risk factors in Australian adults: an observational study.BMC Public Health. 2013; 13: 755https://doi.org/10.1186/1471-2458-13-755
- Metropolitan-level racial residential segregation and black-white disparities in hypertension.Am J Epidemiol. 2011; 174: 537-545https://doi.org/10.1093/aje/kwr116
- Spatial social polarisation: using the Index of Concentration at the Extremes jointly for income and race/ethnicity to analyse risk of hypertension.J Epidemiol Community Health. 2015; 69: 1199-1207https://doi.org/10.1136/jech-2015-205728
- Geographic and racial patterns of preventable hospitalizations for hypertension: Medicare beneficiaries, 2004–2009.Public Health Rep. 2014; 129: 8-18https://doi.org/10.1177/003335491412900104
- The role of disadvantaged neighborhood environments in the association of John Henryism with hypertension and obesity.Psychosom Med. 2016; 78: 552-561https://doi.org/10.1097/PSY.0000000000000308
- Relation between neighborhood median housing value and hypertension risk among black women in the United States.Am J Public Health. 2007; 97: 718-724https://doi.org/10.2105/AJPH.2005.074740
- Prevalence and control rates of hypertension in the USA: 2017–2018.Int J Cardiol Hypertens. 2020; 6100044https://doi.org/10.1016/j.ijchy.2020.100044
- A Randomized trial of intensive versus standard blood-pressure control [published correction appears in N Engl J Med. 2017;377(25):2506].N Engl J Med. 2015; 373: 2103-2116https://doi.org/10.1056/NEJMoa1511939
National Association of Community Health Centers, America's Health Centers. Snapshot 2022. https://www.nachc.org/research-and-data/americas-health-centers-2022-snapshot/. Accessed December 29, 2022.
Lewis C, Getachew Y, Abrams M, Doty M. Changes at community health centers, and how patients are benefiting: results from the Commonwealth Fund National survey of federally qualified health centers; 2013–2018. New York, NY: The Commonwealth Fund.https://www.commonwealthfund.org/publications/issue-briefs/2019/aug/changes-at-community-health-centers-how-patients-are-benefiting. Published August 2019. Accessed September 20, 2019.
- The OCHIN community information network: bringing together community health centers, information technology, and data to support a patient-centered medical village.J Am Board Fam Med. 2013; 26: 271-278https://doi.org/10.3122/jabfm.2013.03.120234
- Disability, health, and multiple chronic conditions among people eligible for both Medicare and Medicaid, 2005–2010.Prev Chronic Dis. 2013; 10: E157https://doi.org/10.5888/pcd10.130064
- The Charlson comorbidity index is adapted to predict costs of chronic disease in primary care patients.J Clin Epidemiol. 2008; 61: 1234-1240https://doi.org/10.1016/j.jclinepi.2008.01.006
USDA. Rural-urban commuting area codes. U.S. Department of Agriculture. https://www.ers.usda.gov/data-products/rural-urban-commuting-area-codes/. Accessed December 29, 2022.
- Measures of social deprivation that predict health care access and need within a rational area of primary care service delivery.Health Serv Res. 2013; 48: 539-559https://doi.org/10.1111/j.1475-6773.2012.01449.x
- The Kaiser Permanente Northern California story: improving hypertension control from 44% to 90% in 13 years (2000 to 2013).J Clin Hypertens (Greenwich). 2016; 18: 260-261https://doi.org/10.1111/jch.12803
- The Affordable Care Act improved health insurance coverage and cardiovascular-related screening rates for cancer survivors seen in community health centers.Cancer. 2020; 126: 3303-3311https://doi.org/10.1002/cncr.32900
- Racial/ethnic disparities in health insurance and differences in visit type for a population of patients with diabetes after Medicaid expansion.J Health Care Poor Underserved. 2019; 30: 116-130https://doi.org/10.1353/hpu.2019.0011
- The impact of the Affordable Care Act (ACA) Medicaid expansion on visit rates for diabetes in safety net health centers.J Am Board Fam Med. 2018; 31: 905-916https://doi.org/10.3122/jabfm.2018.06.180075
- Racial trends in clinical preventive services use, chronic disease prevalence, and lack of insurance before and after the Affordable Care Act.Am J Manag Care. 2022; 28: e126-e131https://doi.org/10.37765/ajmc.2022.88865
- Changes in diabetes prescription patterns following Affordable Care Act Medicaid expansion.BMJ Open Diabetes Res Care. 2021; 9e002135https://doi.org/10.1136/bmjdrc-2021-002135
- Association between a temporary reduction in access to health care and long-term changes in hypertension control among veterans after a natural disaster.JAMA Netw Open. 2019; 2e1915111https://doi.org/10.1001/jamanetworkopen.2019.15111
- Evidence and recommendations on the use of telemedicine for the management of arterial hypertension: an international expert position paper.Hypertension. 2020; 76: 1368-1383https://doi.org/10.1161/HYPERTENSIONAHA.120.15873
- Acculturation is associated with hypertension in a multiethnic sample.Am J Hypertens. 2007; 20: 354-363https://doi.org/10.1016/j.amjhyper.2006.09.025
- The effects of gender and country of origin on acculturation, psychological factors, lifestyle factors, and diabetes-related physiological outcomes among Mexican Americans: the Starr County diabetes prevention initiative.Chronic Illn. 2022; (In press. Online March 24)https://doi.org/10.1177/17423953221089315
- Health disparities of cardiometabolic disorders among Filipino Americans: implications for health equity and community-based genetic research.J Racial Ethn Health Disparities. 2022; 9: 2560-2567https://doi.org/10.1007/s40615-021-01190-6
- Hispanic health in the USA: a scoping review of the literature.Public Health Rev. 2016; 37: 31https://doi.org/10.1186/s40985-016-0043-2
- White-coat hypertension and cardiovascular events: a meta-analysis.J Hypertens. 2016; 34: 593-599https://doi.org/10.1097/HJH.0000000000000832
- Long-term risk of sustained hypertension in white-coat or masked hypertension.Hypertension. 2009; 54: 226-232https://doi.org/10.1161/HYPERTENSIONAHA.109.129882
- Associations between residential segregation and incident hypertension: the multi-ethnic study of atherosclerosis.J Am Heart Assoc. 2022; 11e023084https://doi.org/10.1161/JAHA.121.023084
- 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice guidelines.Circulation. 2018; 138: e484-e594https://doi.org/10.1161/CIR.0000000000000596
NCQA. Controlling high blood pressure (CBP). https://www.ncqa.org/hedis/measures/controlling-high-blood-pressure/. Accessed December 29, 2022.
Article info
Publication history
Published online: January 04, 2023
Publication stage
In Press Journal Pre-ProofIdentification
Copyright
© 2022 American Journal of Preventive Medicine. Published by Elsevier Inc.
User license
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0)
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
Not Permitted
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article
Elsevier's open access license policy
