Aortic stenosis (AS) is the third most prevalent cardiovascular disease following hypertension and coronary artery disease. The primary objective of this cross-sectional study is to examine gender, racial, and socioeconomic disparities in AS-related health care utilization in patients aged ≥50 years using data from the Healthcare Cost and Utilization Project Nationwide Inpatient Sample. AS was identified among inpatient discharges with International Classification of Diseases, Ninth Revision, Clinical Modification , code 424.1. Using stratum-specific weighted totals, means, proportions, and regression models, we examined time trends and disparities for inhospital AS prevalence according to gender, race, and income over the 2002 to 2012 period, predictors of AS (gender, race, income, age, health insurance, co-morbidities, and hospital-level characteristics), and AS’s role as a predictor of inhospital death, length of stay, and total charges. Inhospital AS prevalence increased from 2.10% in 2002 to 2.37% in 2012, with similar trends observed within gender, race, and income strata. Women were less likely to have AS compared with men (adjusted odds ratio [OR adj ] 0.84; 95% confidence interval [CI] 0.83 to 0.86). Blacks (OR adj 0.68; 95% CI 0.66 to 0.71), Hispanics (OR adj 0.79; 95% CI 0.76 to 0.84), and Asians/Pacific Islanders (OR adj 0.68; 95% CI 0.64 to 0.74) were less likely than whites to have AS diagnosis that was directly associated with income. AS was inversely related to inhospital death but positively linked to total charges overall and longer hospital stays among men, whites, and middle-income patients. However, shorter stays with AS were observed among blacks. In conclusion, among older inpatients, AS prevalence was ∼2% and was higher among males, whites, and higher income groups. Although inhospital death was lower and total charges were higher in AS, length of stay’s association with AS varied by gender, race, and income.
Aortic stenosis (AS) is currently the third most prevalent cardiovascular disease following hypertension and coronary artery disease. Although considered a degenerative disorder typically affecting the elderly people, AS has gained importance in recent years because of population aging. AS prevalence in subjects ≥65 years was estimated at 2% to 7%, with 5.5% of subjects >85 years diagnosed with AS ; each year, ∼22,000 deaths were attributed to valvular disease in the United States, with AS being particularly problematic because it carries a high mortality risk once symptoms have appeared. Furthermore, the multifactorial etiology of AS points to prominent roles played by traditional cardiovascular risk factors such as male gender, advanced age, smoking, hypertension, hypercholesterolemia, diabetes, metabolic syndrome, and atherosclerosis. The equitable distribution of health care resources requires a better understanding of the epidemiology of AS, whereas identifying subgroups within the general population at high risk for AS. Previous research suggested that most AS-affected subjects (∼70%) are >60 years, with a balanced gender ratio. AS is also less prevalent among African-Americans compared with Caucasians. Little is known about the prevalence of AS in other racial groups and whether socioeconomic disparities exist in the prevalence of AS. In this population-based cross-sectional study, we examined gender, racial, and socioeconomic disparities in AS-related health care utilization using data collected for the Nationwide Inpatient Sample (NIS).
Methods
The work described has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans. A federal state-industry partnership sponsored by the Agency for Healthcare Research and Quality (AHRQ), the Healthcare Cost and Utilization Project (HCUP), developed several databases and software tools, of which the NIS is the largest publicly available all-payer database in the United States that can provide national estimates of health care utilization, access, charges, quality, and outcomes. NIS consists of a 20% stratified probability sample of community hospitals from the State Inpatient Databases including all inpatient data currently contributed to HCUP. The sampling frame of hospitals (or hospital discharges) was divided into strata using five hospital characteristics, namely ownership/control, bed size, teaching status, urban/rural location, and US region. NIS samples were selected with probabilities proportionate to the number of hospitals (or hospital discharges) in each stratum. NIS data were collected from ∼1,000 hospitals from 1988 to 2012, and the number of states covered by NIS has grown from 8 to 44. Starting in 2012, NIS became a sample of discharge records from all HCUP-participating hospitals, rather than a sample of hospitals, of which all discharges were kept. To facilitate trend analyses, AHRQ created new discharge trend weights for the NIS time period of 1993 until 2011, calculated in the same way as the redesigned 2012 NIS weights. The NIS contains clinical, nonclinical, and resource use data elements typical of a discharge abstract. Inpatient hospital discharges corresponding to patients who satisfy the criteria of being either male or female and ≥50 years were selected from the 2012 NIS for most analyses and 2002 to 2012 NIS for trend analyses. Accordingly, the study sample consists of 45,380,630 discharges ≥50 years (weighted mean age ± SEM: 70.6 ± 0.1 years, weighted proportion female ± SEM: 54.6 ± 0.1%, weighted number of discharges: 217,060,201), out of a total number of 87,039,711 discharges (2002 to 2012) (weighted mean age ± SEM: 47.9 ± 0.2 years, weighted proportion female ± SEM: 58.5 ± 0.1%, weighted number of discharges: 411,487,801). For the 2012 NIS, the study sample aged ≥50 years consisted of 3,991,678 discharges (weighted mean age ± SEM: 70.1 ± 0.0 years, weighted proportion female ± SEM: 53.5 ± 0.1%, weighted number of discharges: 19,958,391).
Our plan was to examine time trends in the prevalence of AS diagnosis over a 10-year study period (2002 to 2012), predictors of AS including demographic (gender, race, and socioeconomic status), clinical (age, health insurance, and co-morbidities), and hospital (geographical area, hospital location and teaching status, and hospital control) characteristics, AS diagnosis as a predictor of inhospital death, hospital length of stay, and total charges overall, and according to gender, race, and socioeconomic categories. Accordingly, a dichotomous outcome variable was defined as the presence or absence of AS among inpatient discharges using an International Classification of Diseases, Ninth Revision, Clinical Modification , code of 424.1 as a criterion based on any of the multiple diagnostic variables. Trends in AS prevalence were evaluated over a 10-year period (2002 to 2012) using calendar year. The key predictors of AS diagnosis were patient gender (male or female), patient race (white, black, Hispanic, Asian or Pacific Islander, Native American, or other), and household socioeconomic status defined in quartiles. Potential confounders of the hypothesized relations were clinical and hospital characteristics identified as simultaneously related to the outcome and predictor variables but not on the causal pathway between them. “Clinical characteristics” included patient admission type (elective and nonelective), age, health insurance (Medicare, Medicaid, Private (including HMO), self-pay, no charge or other), and a 12-item Charlson’s co-morbidity index defined using diagnostic variables. Of note, from 2002 to 2008, only 15 diagnoses were available, whereas after 2009, 25 diagnoses became available. “Hospital characteristics” included geographical area (North, South, Midwest, East, or West), location, and teaching status (rural, urban/nonteaching, or urban/teaching) and also control (“government, nonfederal,” “private, nonprofit,” and “private, investor-own”). Finally, adverse health care utilization outcomes were evaluated based on “inhospital death,” “hospital length of stay (in days),” and “total charges” (in US dollars).
All statistical analyses were conducted using STATA, version 14 (STATA Corp., College Station, Texas). Summary statistics included mean (±SE) for continuous variables and frequencies with percentages for categorical variables. Bivariate relations were evaluated using the chi-square test, Fisher’s exact test, independent-sample t test, or Wilcoxon’s rank-sum test, as appropriate. Linear or logistic regression models were constructed to assess disparities in AS prevalence and AS diagnosis as a predictor of adverse health care utilization outcomes, before and after adjustment for confounders. All analyses were performed accounting for complex survey design, through STATA’s svy commands. Two-sided statistical tests were conducted at an alpha level of 0.05.
Results
Table 1 presents the distribution of the study sample according to gender, race, and socioeconomic status in the 2012 NIS. Further analyses suggested disparities in AS prevalence according to these demographic characteristics were statistically significant (p <0.0001). Therefore, time trend analyses for AS prevalence between 2002 and 2012 were stratified according to gender, race, and socioeconomic status. Statistically significant variations over time were noted for AS prevalence in the total sample and in most subgroups except for the nonwhite racial groups. Although no clearly defined linear trend was observed, AS prevalence increased from 2.10% in 2002 to 2.37% in 2012, with similar trends observed within gender, race, and socioeconomic subgroups ( Table 2 ).
| Variable | |
|---|---|
| Sex (N=19,957,316): | % ± SEM |
| Male | 46.4 ± 0.075 |
| Female | 53.6 ± 0.075 |
| Race (N=18,942,347): | % ± SEM |
| White | 74.80 ± 0.48 |
| Black | 12.59 ± 0.31 |
| Hispanic | 7.29 ± 0.29 |
| Asian or Pacific Islander | 1.93 ± 0.10 |
| Native American | 0.58 ± 0.069 |
| Other | 2.79 ± 0.22 |
| Socioeconomic Status (N=19,522,160): | % ± SEM |
| 1st quartile | 30.25 ± 0.47 |
| 2nd quartile | 25.15 ± 0.36 |
| 3rd quartile | 23.6 ± 0.32 |
| 4th quartile | 21.0 ± 0.50 |
| 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | P trend | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Overall | 2.10 ± 0.04 | 2.17 ± 0.04 | 2.21 ± 0.04 | 2.34 ± 0.05 | 2.28 ± 0.04 | 2.32 ± 0.05 | 2.17 ± 0.04 | 2.21 ± 0.04 | 2.18 ± 0.04 | 2.30 ± 0.04 | 2.37 ± 0.02 | <0.0001 |
| Male | 2.10 ± 0.04 | 2.18 ± 0.05 | 2.26 ± 0.05 | 2.36 ± 0.05 | 2.31 ± 0.05 | 2.36 ± 0.05 | 2.21 ± 0.05 | 2.26 ± 0.05 | 2.19 ± 0.04 | 2.34 ± 0.05 | 2.43 ± 0.02 | <0.0001 |
| Female | 2.09 ± 0.04 | 2.16 ± 0.04 | 2.19 ± 0.04 | 2.33 ± 0.05 | 2.26 ± 0.04 | 2.28 ± 0.04 | 2.13 ± 0.04 | 2.17 ± 0.04 | 2.17 ± 0.04 | 2.28 ± 0.04 | 2.32 ± 0.02 | <0.0001 |
| White | 2.34 ± 0.05 | 2.39 ± 0.05 | 2.48 ± 0.05 | 2.61 ± 0.06 | 2.58 ± 0.06 | 2.63 ± 0.06 | 2.43 ± 0.06 | 2.48 ± 0.05 | 2.41 ± 0.04 | 2.57 ± 0.05 | 2.62 ± 0.02 | <0.0001 |
| Black | 1.30 ± 0.06 | 1.33 ± 0.06 | 1.32 ± 0.07 | 1.47 ± 0.07 | 1.43 ± 0.06 | 1.44 ± 0.06 | 1.35 ± 0.05 | 1.29 ± 0.04 | 1.25 ± 0.04 | 1.33 ± 0.04 | 1.37 ± 0.02 | 0.058 |
| Hispanic | 1.66 ± 0.11 | 1.57 ± 0.11 | 1.51 ± 0.07 | 1.73 ± 0.1 | 1.60 ± 0.09 | 1.61 ± 0.09 | 1.74 ± 0.09 | 1.64 ± 0.09 | 1.54 ± 0.08 | 1.73 ± 0.11 | 1.73 ± 0.05 | 0.53 |
| Asian or Pacific Islander | 1.71 ± 0.15 | 1.70 ± 0.12 | 1.75 ± 1.54 | 2.14 ± 0.15 | 2.04 ± 0.13 | 1.78 ± 0.14 | 1.75 ± 0.11 | 1.69 ± 0.09 | 1.97 ± 0.15 | 1.82 ± 0.10 | 1.97 ± 0.08 | 0.10 |
| Native American | 1.12 ± 0.16 | 1.22 ± 0.22 | 1.16 ± 0.20 | 1.07 ± 0.13 | 1.35 ± 0.17 | 2.18 ± 0.6 | 2.28 ± 0.73 | 1.94 ± 0.33 | 1.65 ± 0.25 | 1.45 ± 0.12 | 2.14 ± 0.03 | 0.17 |
| Other | 1.76 ± 0.17 | 1.89 ± 0.17 | 1.73 ± 0.16 | 1.88 ± 0.13 | 2.21 ± 0.21 | 1.91 ± 0.10 | 1.92 ± 0.11 | 1.86 ± 0.13 | 1.89 ± 0.09 | 2.19 ± 0.21 | 2.32 ± 0.16 | 0.06 |
| 1st quartile | 1.57 ± 0.77 | 1.79 ± 0.05 | 1.78 ± 0.05 | 1.95 ± 0.06 | 1.81 ± 0.05 | 1.89 ± 0.06 | 1.77 ± 0.05 | 1.80 ± 0.05 | 1.69 ± 0.04 | 1.85 ± 0.04 | 1.94 ± 0.02 | 0.0002 |
| 2nd quartile | 1.89 ± 0.052 | 2.10 ± 0.05 | 2.17 ± 0.05 | 2.27 ± 0.06 | 2.22 ± 0.05 | 2.27 ± 0.05 | 2.11 ± 0.05 | 2.16 ± 0.05 | 2.16 ± 0.04 | 2.25 ± 0.05 | 2.32 ± 0.02 | <0.0001 |
| 3rd quartile | 2.09 ± 0.046 | 2.38 ± 0.05 | 2.44 ± 0.05 | 2.54 ± 0.06 | 2.47 ± 0.05 | 2.51 ± 0.05 | 2.32 ± 0.05 | 2.39 ± 0.05 | 2.37 ± 0.05 | 2.83 ± 0.09 | 2.53 ± 0.03 | <0.0001 |
| 4th quartile | 2.28 ± 0.053 | 2.51 ± 0.07 | 2.64 ± 0.06 | 2.72 ± 0.07 | 2.79 ± 0.08 | 2.82 ± 0.08 | 2.63 ± 0.07 | 2.68 ± 0.07 | 2.70 ± 0.06 | 2.83 ± 0.09 | 2.91 ± 0.04 | <0.0001 |
∗ Numbers presented in the table are percentages with their standard error.
In the multivariable analysis, AS diagnosis was less common among women compared with men (adjusted odds ratio [OR adj ] 0.84; 95% confidence interval [CI] 0.83 to 0.86); black (OR adj 0.68; 95% CI 0.66 to 0.71), Hispanic (OR adj 0.79; 95% CI 0.76 to 0.84), and Asian or Pacific Islander (OR adj 0.68; 95% CI 0.64 to 0.74) patients were less likely than white patients to be diagnosed with AS. Whereas AS diagnoses were more likely with increasing socioeconomic status and elective admission, patients having Charlson co-morbidity score of ≥1, urban location, and private ownership of hospitals were significantly related to AS diagnosis. Taking “Medicare” as a referent, recipients of “Medicaid” (OR adj 0.86; 95% CI 0.82 to 0.90) or “other” (OR adj 0.83; 95% CI 0.77 to 0.89) types of insurance were less likely to be diagnosed with AS. Compared with patients seeking health care services in the Northeast, those at Midwestern and Southern hospitals were less likely to be diagnosed with AS ( Table 3 ).
| Variable | Unadjusted OR (95% CI) | Adjusted OR (95% CI) |
|---|---|---|
| Sex: | ||
| Male | Ref. | Ref. |
| Female | 0.95 (0.94-0.97) | 0.84 (0.83-0.86) |
| Race: | ||
| White | Ref. | Ref. |
| Black | 0.51 (0.49-0.54) | 0.68 (0.66-0.71) |
| Hispanic | 0.65 (0.62-0.69) | 0.79 (0.76-0.84) |
| Asian or Pacific Islander | 0.75 (0.69-0.81) | 0.68 (0.64-0.74) |
| Native American | 0.81 (0.62-1.07) | 1.07 (0.87-1.31) |
| Other | 0.88 (0.77-1.02) | 0.98 (0.87-1.10) |
| Socioeconomic Status: | ||
| 1st quartile | Ref. | Ref. |
| 2nd quartile | 1.20 (1.17-1.23) | 1.06 (1.03-1.08) |
| 3rd quartile | 1.31 (1.27-1.35) | 1.09 (1.06-1.12) |
| 4th quartile | 1.51 (1.46-1.57) | 1.16 (1.12-1.19) |
| Admission type: | ||
| Elective | 0.94 (0.91-0.98) | 1.29 (1.26-1.34) |
| Non-Elective | Ref. | Ref. |
| Age (years): | 1.07 (1.07-1.07) | 1.07 (1.07-1.07) |
| Health insurance: | ||
| Medicare | Ref. | Ref. |
| Medicaid | 0.26 (0.24-0.27) | 0.86 (0.82-0.90) |
| Private (including HMO) | 0.39 (0.38-0.41) | 0.99 (0.97-1.02) |
| Self-pay | 0.25 (0.23-0.27) | 0.92 (0.84-1.00) |
| No charge | 0.24 (0.19-0.30) | 1.01 (0.81-1.28) |
| Other | 0.35 (0.33-0.38) | 0.83 (0.77-0.89) |
| Charlson Comorbidity Index: | ||
| 0 | Ref. | Ref. |
| 1 | 1.68 (1.64-1.72) | 1.60 (1.57-1.64) |
| 2+ | 2.24 (2.19-2.29) | 1.98 (1.94-2.02) |
| Geographical area: | ||
| Northeast | Ref. | Ref. |
| Midwest | 0.83 (0.79-0.88) | 0.89 (0.84-0.94) |
| South | 0.70 (0.66-0.74) | 0.86 (0.82-0.89) |
| West | 0.83 (0.79-0.88) | 0.95 (0.90-1.01) |
| Hospital location & teaching status: | ||
| Rural | Ref. | Ref. |
| Urban/non-Teaching | 1.19 (1.13-1.25) | 1.21 (1.15-1.28) |
| Urban/Teaching | 1.24 (1.17-1.30) | 1.38 (1.30-1.46) |
| Hospital control: | ||
| Government, nonfederal | Ref. | Ref. |
| Private, non-profit | 1.45 (1.35-1.56) | 1.18 (1.11-1.25) |
| Private, investor-own | 1.11 (1.02-1.21) | 1.03 (0.96-1.10) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
