Approximately 1% to 2% of patients with hypertension will have a hypertensive emergency at some time in their life. However, no data are available on the frequency of hospitalizations for a hypertensive emergency after the publication of the Seventh Joint National Committee (JNC7) on the prevention, detection, evaluation, and treatment of high blood pressure. We sought to explore the changes in the frequency of hospitalizations and in-hospital mortality for hypertensive emergencies before and after the JNC7 report. Using the Nationwide Inpatient Sample from 2000 to 2007, adult patients (aged ≥18 years) who were hospitalized with a diagnosis of a hypertensive emergency were identified through appropriate “International Classification of Diseases, 9th revision, Clinical Modification” codes. A total of 456,259 hospitalizations with the diagnosis of hypertensive emergency occurred from the start of calendar year 2000 to the end of calendar year 2007. After adjusting for the United States census for 2000 and American Community Survey estimates for 2007 for adults, the frequency of hospitalizations with a hypertensive emergency increased in United States adults from 101/100,000 in 2000 to 111/100,000 in 2007, an average increase of about 1.11%. Despite the increase in hospitalizations, the all-cause in-hospital mortality rate decreased from 2.8% in the pre-JNC7 era to 2.6% in the post-JNC7 era (odds ratio 0.91, 95% confidence interval 0.86 to 0.96). In conclusion, the results of the present study have shown that although the number of patients with hypertensive emergency increased from 2000 to 2007, the mortality rates decreased significantly after the JNC7 guidelines.
Hypertension is a very common condition, affecting approximately 65 million subjects in the United States. As noted in the report by the Joint National Committee on Prevention, Detection, Evaluation and Treatment of High Blood Pressure (JNC7), each 20-mm Hg increment in systolic blood pressure or 10-mm Hg increment in diastolic blood pressure doubles the risk of cardiovascular events, independent of other factors. It is generally acknowledged that JNC7 recommendations are simple and, hence, easy to institute. Accordingly, more patients with hypertension are receiving antihypertensive therapy and are probably reaching the desired control following the JNC7 recommendations. We used the Nationwide Inpatient Sample (NIS) data set, the largest in-patient care database in the United States to address the following: (1) trends in hypertensive emergency-related hospitalizations in the United States from 2000 to 2007 (before and after the JNC7 report); (2) trends in hypertensive emergency-related hospitalizations in relation to patient age and gender; (3) the predictors of mortality during hypertension emergency-related hospitalizations; and (4) the cost of care of hypertensive emergency hospitalizations during this period.
Methods
The data were obtained from the NIS data set for 2000 to 2007. The NIS is a nationally representative survey of hospitalizations conducted by the Healthcare Cost and Utilization Project in collaboration with the participating states. It is the largest all-payer inpatient data set in the United States and includes a 20% sample of United States community hospitals that approximates 20% of all United States community hospitals.
Each entry also contains information on demographic details, including age, gender, race, insurance status, primary and secondary procedures, hospitalization outcome, total cost, and length of stay. The NIS database also contains clinical and resource use information, with safeguards to protect the privacy of patients, physicians, and hospitals. The NIS database results have been shown to correlate well with other hospitalization discharge databases in the United States. We used this data set to gather information on the number of hospitalizations with a primary diagnosis of a hypertensive emergency using the “International Classification of Disease, 9th Revision” codes ( Appendix 1 ). Any secondary diagnosis of a hypertensive emergency, if used with a primary diagnosis of stroke (codes 433, 434), intracranial hemorrhage (codes 431, 432.9), aortic dissection (code 441.0), eclampsia (codes 642.4 to 642.7), acute myocardial infarction (code 410.x0), or hypertensive encephalopathy (code 437.2), was also included ( Figure 1 ) .
Because patients discharged from a hospital could be sampled more than once, the estimates were for discharges, not patients. The specific years of 2000 to 2007 were selected to determine whether the trends and outcomes changed after the JNC7 report. Thus, these years represent both the pre- (2000 to 2003) and post- (2004 to 2007) JNC7 periods. An analysis of these periods helped us draw, albeit limited, inferences on the effect of the JNC7 guidelines. The records of patients aged ≥18 years were included. The patients were divided into age groups (18 to 49, 50 to 64, and ≥65 years). Those aged ≥65 years were selected to determine the trends in hospitalizations for the elderly. The race information was missing for about 20% of the patients, and the race was coded as unknown for these patients.
The data were analyzed using Stata IC, version 11.0 (StataCorp, College Station, Texas). Continuous variables were summarized using the mean values and standard errors. All calculations were performed using the weighted estimates approximating the nationwide population estimates. The census bureau estimates of the United States civilian resident population, as of July 1, for 2000 through 2007, were used to calculate the age- and gender-specific frequency of hospitalization for hypertensive emergency per 100,000 United States residents (available at: http://factfinder.census.gov ). Because of the complex sample design of the NIS, the estimates were subjected to weighted analyses, and the relative standard error of an estimate was examined to determine its reliability. If the relative standard error of any estimate was >30%, the estimate was either not reported or was flagged as unreliable. Nationally representative estimates for the number and frequency of hospitalizations for patients with hypertensive emergency as a primary diagnosis were calculated by age and year.
The age-standardized prevalence of hospitalization for hypertensive emergency was estimated using the United States Census according to age group for 2000 through 2007. The prevalence estimates were age standardized to the year 2000 population. The mean hospital charge for each hypertensive emergency hospitalization was calculated for each year. The hospital charges for hypertensive emergency hospitalization were adjusted for inflation since the year 2000.
The risk factors for in-hospital mortality were examined using multivariate logistic regression analysis adjusted for the survey design. Univariate logistic regression analysis was used to identify the risk factor variables. All variables that were significant on univariate analysis at p <0.05 were included in the final multivariate model. The variables were checked for multiple co-linearity using tolerance and the variance inflation factor. For the variables used in the final model, both the tolerance and the variation inflation factor were very close to unity. The final model for mortality adjusted for age, gender, race, hospital characteristics, insurance type, and Charlson’s co-morbidity index.
Results
An estimated 456,259 ± 12,386 SE hospitalizations for a hypertensive emergency occurred in the United States from the beginning of the calendar year 2000 to the end of the calendar year 2007. These comprised 0.16% of all admissions in 2000 and 0.19% of all admissions in 2007. The frequency of hospitalizations for a hypertensive emergency increased from 101/100,000 United States adults in 2000 to 111/100,000 in 2007, an average increase of about 10%. In absolute terms, the number of hospitalizations for a hypertensive emergency increased from 49,018 to 61,043/year during the 8 years of follow-up ( Table 1 ). The absolute number of hypertension emergency hospitalizations increased from 2000 to 2003 ( Table 1 ), followed by much slower increase until 2007 ( Figure 2 ) . The prevalence of hospitalizations per 100,000 population, however, showed a small decrease from 2004 to 2007 ( Table 1 ). The significant change was seen only in patients aged ≥65 years, with no significant changes in those aged 18 to 49 or 50 to 64 years ( Table 1 ).
| Variable | 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 |
|---|---|---|---|---|---|---|---|---|
| Total | 49,018 | 53,844 | 55,850 | 59,218 | 59,609 | 57,253 | 60,424 | 61,043 |
| Age >18 years | 101 | 112 | 113 | 119 | 116 | 111 | 113 | 111 |
| Age group (years) | ||||||||
| 18–49 | 10 | 11 | 11 | 12 | 12 | 12 | 12 | 12 |
| 50–64 | 31 | 33 | 33 | 35 | 34 | 32 | 35 | 34 |
| ≥65 ⁎ | 60 | 68 | 69 | 72 | 70 | 67 | 66 | 65 |
| Gender | ||||||||
| Male | 39.9 | 40 | 40.2 | 41.4 | 43.3 | 43.3 | 43.3 | 41.6 |
| Female | 60.1 | 60 | 59.8 | 58.6 | 58.8 | 58.9 | 56.7 | 56.7 |
| Race | ||||||||
| White ⁎ | 41.9 | 39.1 | 37 | 37.6 | 36.4 | 39.8 | 35.9 | 34.1 |
| Black ⁎ | 25.4 | 25.3 | 26 | 26.3 | 29 | 25.3 | 27.9 | 29.1 |
| Hispanic | 6.0 | 6.2 | 6.1 | 6.9 | 6.1 | 6.7 | 6.5 | 6.3 |
| Asian | 1.8 | 2.5 | 2.0 | 2.1 | 1.6 | 1.6 | 1.4 | 1.5 |
| Native American | 0.3 | 0.2 | 0.2 | 0.1 | 0.3 | 0.3 | 0.4 | 0.2 |
| Other | 1.5 | 1.4 | 1.5 | 1.6 | 1.7 | 1.7 | 1.7 | 1.6 |
| Unknown | 23.2 | 25.3 | 27.1 | 25.3 | 24.7 | 24.7 | 26.2 | 27.2 |
| Insurance | ||||||||
| Medicare | 49.1 | 50.4 | 51.1 | 52 | 51.4 | 52.1 | 50.3 | 50.2 |
| Medicaid | 10.9 | 11.3 | 10.6 | 11.6 | 11.7 | 11.4 | 11.5 | 10.6 |
| Private ⁎ | 29.5 | 28 | 27.7 | 26 | 25.2 | 25.2 | 25.1 | 26 |
| Self pay ⁎ | 7.2 | 7.3 | 7.5 | 7.5 | 8.5 | 8.4 | 9.3 | 9.5 |
| Geographic location | ||||||||
| Northeast | 18.9 | 16 | 17.4 | 19.1 | 16.6 | 17.2 | 16.4 | 14.7 |
| Midwest | 16.5 | 16.1 | 18.5 | 18.6 | 20.5 | 17.7 | 18.3 | 21.4 |
| South | 50.5 | 52.8 | 52.3 | 48.6 | 51.1 | 52.2 | 53.5 | 52.2 |
| West | 14.1 | 15.1 | 11.8 | 13.7 | 11.8 | 13.0 | 11.8 | 11.8 |
| Charlson’s co-morbid index | ||||||||
| <3 ⁎ | 91.4 | 90.4 | 90 | 89.1 | 89.1 | 83.6 | 74.4 | 71.8 |
| ≥3 ⁎ | 8.6 | 9.6 | 10 | 10.9 | 10.9 | 16.4 | 25.6 | 28.2 |
| Disposition | ||||||||
| Died | 2.9 | 3.0 | 2.8 | 3.0 | 2.9 | 3.2 | 3.0 | 2.5 |
| Skilled nursing facility ⁎ | 11.3 | 11.8 | 11.4 | 12.7 | 13.3 | 15.1 | 13.4 | 14.1 |
| Other hospital | 3.7 | 3.4 | 3.9 | 3.5 | 3.4 | 3.4 | 3.1 | 3.5 |
| Home care ⁎ | 5.4 | 5.5 | 6.0 | 6.4 | 8.2 | 7.7 | 8.6 | 8.3 |
| Home ⁎ | 75.5 | 74.7 | 74.4 | 72.6 | 70.6 | 69 | 70.1 | 69.7 |
| Length of stay (days) | 4.9 ± 0.1 | 5.0 ± 0.1 | 5.1 ± 0.1 | 5.1 ± 0.1 | 5.3 ± 0.1 | 5.2 ± 0.1 | 5.1 ± 0.1 | 4.9 ± 0.1 |
| Mean total charges ($) ⁎ † | 15,434 | 16,826 | 19,329 | 22,074 | 23,197 | 24,871 | 24,404 | 25,435 |
⁎ Significant difference at p <0.001 from 2000 to 2007.
