Primary care site may play an important role in cardiovascular disease prevalence; however, the distribution of risk factors and outcomes across care sites is not known. In this study, a cross-sectional analysis of 21,778 adult participants from the National Health and Nutrition Examination Survey (NHANES; 1999 to 2008) using multivariate logistic regression was conducted to assess the relation between site of usual care and disease prevalence. Patients’ self-reported histories of several chronic conditions (hypertension, diabetes, and hypercholesterolemia), awareness of chronic conditions, and associated cardiovascular events (angina, coronary heart disease, cardiovascular disease, myocardial infarction, and stroke) were examined. After adjustment for demographic and health care utilization characteristics, there were no significant differences in the prevalence of diabetes or hypercholesterolemia among patients receiving usual care at private doctors’ offices, hospital outpatient clinics, community-based clinics, and emergency rooms (ER). However, participants without usual sources of care and those receiving usual care at ERs had significantly lower awareness of their chronic conditions than participants at other sites. The odds of having a history of each of the adverse cardiovascular events ranged from 2.21 to 4.18 times higher for patients receiving usual care at ERs relative to private doctors’ offices. In conclusion, participants who report using ERs as their usual sites of care are disproportionately more likely to have histories of poor cardiovascular outcomes and are more likely to be unaware of having hypertension or hypercholesterolemia. As health care reform takes place and millions more begin seeking care, it is imperative to ensure access to longitudinal care sites designed for continuous disease management.
We examined a nationally representative sample of participants from the National Health and Nutrition Examination Survey (NHANES) to provide a direct comparison of self-reported sites of primary care in reference to patient composition, the prevalence of conditions known to be risk factors for adverse cardiovascular outcomes, markers of adequate disease management, and cardiovascular outcomes. We explored whether the prevalence of cardiovascular outcomes differs by site after adjusting for known patient-level demographic, lifestyle, and access-related risk factors.
Methods
We analyzed 5 waves of NHANES data, from 1999 to 2008, to explore whether an independent relation exists between respondents’ self-reported sources of usual care and the prevalence of selected chronic and cardiovascular events. We examined the prevalence of self-reported hypertension, diabetes mellitus, hypercholesterolemia, angina, coronary heart disease, myocardial infarction [MI], congestive heart failure, and stroke in patients receiving primary care across 4 distinct sites of care, as well as those reporting having no usual sites of care.
NHANES is a cross-sectional survey administered to a nationally representative sample of the noninstitutionalized United States population. Data collection for patients includes a detailed questionnaire of health status, history, and behaviors; an examination of currently prescribed medications available at the time of interview; and selected laboratory tests on patient samples as appropriate. Our analysis was limited to adult participants aged ≥20 years who identified 1 primary or no primary site of usual care, yielding a total of 21,778 participants.
Our dependent variables were the prevalence of selected chronic conditions and cardiovascular events. Online Appendix Table 1 provides the actual wording of the questions used to assess disease status. Survey participants were asked whether a health professional had informed them that they had each of our studied conditions. In reference to diabetes, hypercholesterolemia, and hypertension, we used a list of currently prescribed antidiabetic medications, total cholesterol readings of >200 mg/dl, and average blood pressure readings >140/90 mm Hg, respectively, to identify additional cases.
The primary independent variable was participants’ reports of their sites of usual care. Categories were a composite variable derived from 2 questions asked in the NHANES questionnaire. The first question asked respondents whether they had places where they usually went for their health care. If participants indicated that they had usual sites of care, a follow-up question prompted them to identify the site as 1 of the following: community health center or clinic (CHC), hospital outpatient clinic, emergency room (ER), private doctor’s office or health maintenance organization (HMO), or multiple sites. We excluded patients who indicated using multiple sites of primary care from our sample because of the small number of positive responses (n = 187).
Participant-level covariates included age (20 to 29, 30 to 44, 45 to 54, 55 to 65, or >65 years), race or ethnicity (white, black, Mexican, other Hispanic, or other race), type of insurance (Medicare, Medicaid, private, uninsured, or other), gender, and income as measured by a poverty-to-income ratio. Additional covariates were smoking status (current, former, and never) and number of physician visits in the 12 months before the survey.
Using patient weights provided by NHANES, we evaluated the demographic distribution of participants at each primary care site. We fit a multivariate logistic regression model to estimate the age-adjusted and fully adjusted prevalence of chronic conditions across sites. Fully adjusted logistic regression models included the primary independent and dependent variables as described previously and the aforementioned covariates. Using the results of these logistic regression models, for each site of care, we calculated fully adjusted proportions (prevalence), which are adjusted to the total population distribution of these potential confounders.
We next modeled the 3-level outcome “global burden of cardiovascular disease,” defined as the proportion of participants at each site with self-reported histories of 0, 1, and ≥2 of the following conditions: hypertension, diabetes, hypercholesterolemia, or 1 of several related cardiovascular events (stroke, MI, coronary heart disease, angina, or congestive heart failure). To estimate the global risk for disease prevalence at each site of care, we constructed age-adjusted and fully adjusted multinomial logistic regression models for the 3-level outcome corresponding to the aggregate number (0, 1, or ≥2) of observed conditions.
In secondary analyses, we explored patient awareness of selected chronic conditions, because poor management of these conditions may affect the likelihood of cardiovascular events. Lack of awareness was examined for hypertension and cholesterol separately, defined as the proportion of patients with average blood pressure readings or total cholesterol measurements out of the normal range (>140/90 mm Hg for hypertension, >200 mg/dl for cholesterol) who reported not being informed that they had high blood pressure or high cholesterol by a health professional. We constructed logistic regression models adjusted for all the previously mentioned covariates other than number of visits and compared participants receiving usual care at private doctors’ offices or HMOs to other care sites. Finally, to understand how the observed relation was mediated by number of visits, we constructed additional models including this variable. Results are reported with 2-tailed p values significant at the α <0.05 level and confidence intervals as appropriate. All analyses were performed with SAS version 9.2 (SAS Institute Inc., Cary, North Carolina) and SUDAAN version 10.0 (RTI International, Research Triangle Park, North Carolina), both of which accounted for the weights, strata, and clusters of the complex survey design.
Results
Of the 21,778 NHANES participants from 1999 to 2008 (representing 198,971,878 adults nationally), 18,127 participants (83%) received their usual care at private doctors’ offices or HMOs, hospital-based outpatient clinics, or CHCs; 435 (2%) identified their usual sources as ERs; and 3,216 (15%) reported no usual sources of primary care. Compared to private doctors’ offices or HMOs, patients seeking their care at ERs, hospital outpatient clinics, and CHCs were more likely to be racial or ethnic minorities, have lower incomes relative to the poverty line, and be uninsured ( Table 1 ). Participants without usual sites of care were on average younger than patients with usual sources of care and had the highest rates of being uninsured compared to respondents reporting usual sites of care ( Table 1 ).
| Variable | Private Doctor’s Office/HMO (n = 13,364) | CHC/Clinic (n = 4,241) | Hospital Outpatient (n = 522) | ER (n = 435) | No Usual Site of Care (n = 3,216) | p Value |
|---|---|---|---|---|---|---|
| Age (years) ⁎ | ||||||
| 20–29 | 13% | 18% | 15% | 27% | 35% | <0.001 |
| 30–44 | 29% | 30% | 24% | 36% | 36% | |
| 45–54 | 23% | 22% | 18% | 18% | 18% | |
| 55–64 | 15% | 13% | 19% | 8% | 7% | |
| ≥65 | 21% | 17% | 24% | 10% | 4% | |
| Race | ||||||
| White | 78% | 64% | 51% | 52% | 58% | <0.001 |
| Black | 10% | 13% | 23% | 29% | 10% | |
| Mexican | 4% | 12% | 7% | 6% | 17% | |
| Other Hispanic | 4% | 6% | 12% | 6% | 8% | |
| Other race | 4% | 5% | 7% | 6% | 7% | |
| Insurance type | ||||||
| Medicare | 8% | 8% | 16% | 5% | 2% | <0.001 |
| Medicaid | 4% | 8% | 13% | 9% | 3% | |
| Private | 68% | 55% | 31% | 33% | 42% | |
| Other | 10% | 10% | 15% | 9% | 5% | |
| Uninsured | 9% | 20% | 24% | 44% | 49% | |
| Gender | ||||||
| Male | 43% | 45% | 61% | 59% | 67% | <0.001 |
| Mean poverty-to-income ratio † | 2.94 | 2.16 | 1.93 | 1.72 | 2.11 | <0.001 |
| Smoking status | ||||||
| Current | 20% | 25% | 33% | 49% | 36% | <0.001 |
| Former | 28% | 25% | 26% | 13% | 17% | |
| Never | 52% | 50% | 41% | 37% | 47% | |
| Average number of outpatient visits in the 12 months before the survey | 2.94 | 2.33 | 2.51 | 1.67 | 0.92 | <0.001 |
⁎ Column totals may not sum to 100%, because of rounding.
† Poverty-to-income ratio represents the ratio of family or unrelated individual income to the appropriate poverty threshold.
Table 2 presents adjusted prevalence rates of hypertension, diabetes mellitus, and hypercholesterolemia by reported sites of care. Patients who identified as having no sites of usual care had the lowest adjusted prevalence of hypertension, diabetes, and hypercholesterolemia. After adjustment, few differences in chronic disease prevalence were observed among sites; however, compared to private doctors’ offices, participants using CHCs as their usual sources of care had statistically significantly lower adjusted rates of being diagnosed with hypertension. In addition to reporting the lowest burden of disease for many of the chronic conditions, participants without usual sites of care also had a lower overall disease burden. For example, >1/2 of these participants reported no histories of chronic or cardiovascular disease at all, with only 15% reporting being diagnosed with ≥2 conditions compared to a range of 23% to 27% of patients across the other care sites ( Table 3 ).
| Condition | Private Doctor’s Office/HMO | CHC/Clinic | Hospital Outpatient Clinic | ER | No Usual Site of Care |
|---|---|---|---|---|---|
| Diabetes mellitus | |||||
| Age-adjusted proportion | 8% | 9% | 15% † | 9% | 2% † |
| Fully adjusted proportion | 8% | 8% | 10% | 7% | 3% † |
| Fully adjusted OR (95% CI) | Reference | 0.93 (0.78–1.11) | 1.32 (0.93–1.88) | 0.86 (0.50–1.46) | 0.27 (0.19–0.39) |
| Hypertension | |||||
| Age-adjusted proportion | 41% | 40% | 46% | 43% | 34% † |
| Fully adjusted proportion | 41% | 38% † | 42% | 41% | 37% † |
| Fully adjusted OR (95% CI) | Reference | 0.86 (0.75–0.97) | 1.04 (0.79–1.37) | 1.02 (0.79–1.31) | 0.81 (0.72–0.92) |
| Hypercholesterolemia | |||||
| Age-adjusted proportion | 29% | 27% † | 30% | 18% † | 17% † |
| Fully adjusted proportion | 28% | 28% | 29% | 22% | 22% † |
| Fully adjusted OR (95% CI) | Reference | 0.93 (0.84–1.03) | 1.02 (0.75–1.40) | 0.68 (0.46–1.01) | 0.64 (0.55–0.75) |
⁎ Model adjusted for age, gender, race, income, insurance type, smoking status, and number of outpatient visits in the 12 months before the survey.
† p <0.05 for all comparisons with private doctor’s office or HMO.
| Care Site | 0 Conditions | 1 Condition | ≥2 Conditions |
|---|---|---|---|
| Private doctor’s office/HMO | |||
| Age-adjusted prevalence | 45% | 31% | 24% |
| Fully adjusted prevalence | 46% | 30% | 24% |
| CHC/clinic | |||
| Age-adjusted prevalence | 46% | 30% | 24% |
| Fully adjusted prevalence | 47% | 30% | 23% |
| Hospital-based outpatient clinic | |||
| Age-adjusted prevalence | 42% | 28% | 30% † |
| Fully adjusted prevalence | 46% | 27% | 27% |
| Hospital ER | |||
| Age-adjusted prevalence | 46% | 27% | 26% |
| Fully adjusted prevalence | 45% | 28% | 27% |
| No usual site of care | |||
| Age-adjusted prevalence | 55% † | 33% | 12% † |
| Fully adjusted prevalence | 51% † | 34% † | 15% † |
⁎ Model adjusted for age, gender, race, income, insurance type, smoking status, and number of outpatient visits in the 12 months before the survey.
† p <0.05 for all comparisons with private doctor’s office or HMO.
Our analysis showed that poor hypertension awareness was significantly more common for participants without usual sites of care and those reporting ERs as their primary sites of care than for participants receiving care at private doctors’ offices or HMOs after adjustment patient age, gender, race, insurance type, income, and smoking status ( Table 4 ). Among participants with blood pressure readings >140/90 mm Hg (n = 9,440), 61% of participants without usual sites of care reported not being informed about their high blood pressure. Lack of hypertension awareness was present in nearly half (46%) of participants receiving care in ERs and 39% of those receiving usual care in doctors’ offices or HMOs. In a final model, adjusting for the number of outpatient visits in addition to the other covariates, results persisted only for participants without usual sites of care. No significant adjusted differences between other care sites and private doctor’s offices or HMOs were observed ( Table 4 ). Similar patterns were observed for patient awareness of hypercholesterolemia.
| Variable | Private Doctor’s Office/HMO | CHC/Clinic | Hospital Outpatient Clinic | ER | No Usual Site of Care |
|---|---|---|---|---|---|
| Hypertension (n = 9,440) | |||||
| Partially adjusted proportion of participants unaware of condition ⁎ | 39% | 39% | 33% | 46% † | 61% † |
| Fully adjusted proportion of participants unaware of condition ‡ | 40% | 39% | 35% | 41% | 52% † |
| Fully adjusted odds ratio (95% CI) ‡ | Reference | 0.94 (0.76–1.18) | 0.78 (0.44–1.18) | 1.02 (0.50–2.08) | 1.71 † (1.18–2.48) |
| Hypercholesterolemia (n = 4,432) | |||||
| Partially adjusted proportion of participants unaware of condition ⁎ | 61% | 62% | 62% | 72% † | 71% † |
| Fully adjusted proportion of participants unaware of condition ‡ | 62% | 63% | 62% | 70% | 69% † |
| Fully adjusted odds ratio (95% CI) ‡ | Reference | 1.05 (0.90–1.22) | 1.02 (0.66–1.59) | 1.54 (0.95–2.49) | 1.36 † (1.09–1.69) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
