OBJECTIVE. The goal was to determine the cost-effectiveness of ambulatory blood pressure monitoring in the initial evaluation of stage 1 hypertension.
METHODS. Retrospective chart review of data for children referred to Texas Children's Hospital hypertension clinic between January 2005 and August 2006 was performed. We compared the costs of standard evaluations versus the initial use of ambulatory blood pressure monitoring for children with clinic blood pressure measurements suggesting stage 1 hypertension. Charges for clinic visits, laboratory tests, and imaging were obtained from the Texas Children's Hospital billing department.
RESULTS. A total of 267 children were referred. One hundred thirty-nine children did not receive ambulatory blood pressure monitoring; 54 met clinical indications for ambulatory blood pressure monitoring but did not receive it because it was not a covered expense (44 children) or the family refused the study (10 children). One hundred twenty-six children received clinically indicated ambulatory blood pressure monitoring, paid for either through insurance or by the family. Fifty-eight children (46%) had confirmed white-coat hypertension, 62 (49%) stage 1 hypertension, and 6 (5%) stage 2 hypertension. With the observed prevalence of white-coat hypertension, initial ambulatory blood pressure monitoring use yielded net savings after evaluation of 3 patients, with projected savings of $2.4 million per 1000 patients.
CONCLUSIONS. Ambulatory blood pressure monitoring in the initial evaluation of suspected childhood hypertension is highly cost-effective. Awareness of cost saving potential may increase the availability of ambulatory blood pressure monitoring for evaluation of new-onset hypertension.
The Task Force on High Blood Pressure in Children defines hypertension as systolic or diastolic blood pressure (BP) measurements of ≥95th percentile on 3 separate occasions, stratified according to age, gender, and height percentiles. The most-recent task force recommendations define hypertensive disease as prehypertension for BP values that are confirmed to be between the 90th percentile and the 95th percentile, stage 1 hypertension for BP values that are of >95th percentile but do not exceed the 99th percentile value by 5 mm Hg, and stage 2 hypertension for BP values that exceed the 99th percentile value by >5 mm Hg.1 Population screening studies estimate the prevalence of hypertension to be 2% to 3% for all children,1,2 4.5% for school-aged children,3 and as high as 10% for overweight children.4 Among children found to have elevated BP in screening examinations, 44% to 53% have white-coat hypertension (WCH), that is, elevated BP only in the context of a physician visit.5,6 The current recommendation for children who meet the criteria for the diagnosis of stage 1 hypertension is a detailed evaluation, including laboratory testing (electrolyte level, blood urea nitrogen level, and creatinine level measurements, complete blood count, and urinalysis), renal ultrasonography, and echocardiography, to distinguish primary hypertension from secondary hypertension and to assess target-organ damage.1 In contrast, children with WCH require only BP monitoring over time, with no additional diagnostic evaluation.1,2
WCH complicates the evaluation of elevated BP and can lead to excessive diagnostic testing and unnecessary pharmacologic therapy. Although some reports suggest that individuals with WCH may be at increased risk for progression to persistent hypertension7–10 or target-organ damage,11–14 current consensus recommendations are to treat individuals with WCH only with lifestyle management.1
Ambulatory BP monitoring (ABPM) is the standard technique and is recommended by the task force guidelines to differentiate persistent hypertension from WCH. ABPM provides multiple BP measurements over a 24-hour period, as the child participates in his or her regular activities. ABPM can be performed successfully in the pediatric population,15–18 its results are reproducible,19 and children usually tolerate the procedure with minimal complaints.16 The use of ABPM is limited, however, because reimbursement from third-party payers is inconsistent. The goal of our study was to assess the cost-effectiveness of the use of ABPM in the initial evaluations of children with elevated BP, to differentiate true stage 1 hypertension from WCH.
The records of all children evaluated in the Texas Children's Hospital hypertension clinic between January 2005 and June 2006 were reviewed. The clinical data were abstracted to a Statistica (StatSoft, Tulsa, OK) database for analysis. A total of 267 children were referred to the hypertension clinic for the initial evaluation of elevated BP during the study period. Children who were referred to the hypertension clinic with an established diagnosis, for a second opinion, or who were already receiving antihypertensive therapy were excluded from the study. The Baylor College of Medicine institutional review board approved data collection.
The children were grouped according to whether ABPM was performed. It is our clinic practice to perform ABPM for all patients with clinic measurements that suggest stage 1 hypertension. Clinic records were reviewed to determine indications for ABPM. For the purposes of the study, ABPM was considered clinically indicated if a child had clinic BP values on ≥3 occasions that suggested stage 1 hypertension and did not have known cardiac or renal disease or overt evidence of hypertensive target-organ damage in the physical examination, urinalysis, or initial laboratory testing. Children who were to receive antihypertensive medication because of stage 2 hypertension or overt complications of hypertension did not have therapy delayed for the purposes of ABPM.
Clinic BP Measurements
Children were seated in a quiet examination room for ≥10 minutes before measurement of their BP values. All BP values were measured by trained certified staff members, in accordance with published recommendations on technique and cuff size.1 BP was measured with Welch Allen aneroid sphygmomanometers (Welch Allen, Inc, Skaneatles Falls, NY), which were calibrated with mercury devices every 2 months. BP was measured 4 times, at 3- to 5-minute intervals, and the mean of the second through fourth BP values was recorded.
BP parameters were defined in the Fourth Report on the Diagnosis, Evaluation, and Treatment of High Blood Pressure in Children and Adolescents.1 Children with BP measurements of >95th percentile at the physician's office but average BP values of <90th percentile outside the clinical setting had WCH. Children with mean systolic or diastolic BP values of >90th percentile but <95th percentile had prehypertension. Children with BP values of >95th percentile but <5 mm Hg above the 99th percentile value had stage 1 hypertension, and those with BP values >5 mm Hg above the 99th percentile value had stage 2 hypertension.
Ambulatory BP Monitoring
ABPM was performed for 24 hours with an oscillometric device (SpaceLabs 90217; SpaceLabs Medical Inc, Issaquah, WA). Appropriate cuffs were selected, according to recommendations in the Fourth Report on the Diagnosis, Evaluation, and Treatment of High Blood Pressure in Children and Adolescents,1 and placed on the nondominant arm. The accuracy and precision of the monitor were confirmed through measurement with a calibrated aneroid sphygmomanometer at the beginning of the monitoring period. BP was measured every 20 minutes from 6 am to 10 pm and every 30 minutes from 10 pm to 6 am. Children and/or parents kept an activity diary during the test. The systolic and diastolic load reflected the proportion of readings above the 95th percentile. The studies were analyzed by using published pediatric normative values.20 Hypertension was conservatively defined as a 24-hour mean systolic or diastolic BP of >95th percentile, a daytime systolic or diastolic BP of >95th percentile, or a systolic or diastolic BP load of >30%. This definition of hypertension was used to minimize the prevalence of WCH in the context of the cost-effectiveness analysis. Children were diagnosed as having WCH if the aneroid clinic BP measurement was greater than the task force-defined 95th percentile for age, gender, and height but did not meet any ABPM criteria for hypertension.
The costs of the task force-recommended diagnostic evaluation (laboratory testing, renal ultrasonography, and echocardiography) of all patients referred to Texas Children's Hospital hypertension clinic were compared with the costs of screening all patients with ABPM and laboratory testing and performing renal ultrasonography and echocardiography only for patients who met the definition of hypertension according to ABPM criteria. Because all patients had an initial office visit, associated facility fee, blood tests, and urinalysis, the main comparison was the additional expense of performing ABPM for all patients versus the cost savings of not performing renal ultrasonography and echocardiography for patients with isolated WCH. Costs represent hospital costs at our institution, without adjustment for third-party payer agreements regarding reimbursement. We elected to use hospital costs over adjusted costs because reimbursement data varied depending on the third-party payer and existing reimbursement contracts. According to the Texas Children's Hospital billing department, the total costs (test and interpretation) were $442 for screening renal ultrasonography, $2150 for 2-dimensional transthoracic echocardiography, and $437 for ABPM.
Comparison of the demographic data for the 2 groups for statistical significance testing was performed with Student's t test for continuous variables and with the Wilcoxon test for noncontinuous variables. Data are reported as means, with SDs and minimum and maximum values when relevant, or as proportions. A P value of .05 was considered the threshold for statistical significance. Statistical analyses were performed with Statistica 7.0 (StatSoft, Tulsa, OK).
Between January 2005 and August 2006, 267 children were referred to the Texas Children's Hospital hypertension clinic for the initial evaluation of elevated BP. Children who had confirmed diagnoses and/or were already receiving treatment were excluded. Children's ages ranged from 1 month to 20 years, with a mean age of 12.92 ± 4.35 years; 175 boys (66%) and 92 girls (34%) were evaluated. The ethnic diversity reflected that of the Houston, Texas, metropolitan area, with the exception of fewer children of Asian descent (43% white, 34% black, 22% Hispanic, and 1% Asian). A total of 128 children underwent ABPM as part of their evaluation; ABPM was clinically indicated for 126 patients and was performed as part of the clinical study protocol for 2 patients. A total of 139 patients did not receive ABPM; 54 had clinical indications for ABPM but did not receive it. Demographic data for the 2 groups are presented in Table 1.
Of the 126 patients who received clinically indicated ABPM, 58 (46%) had WCH, 62 (49%) had BP values that corresponded to stage 1 hypertension, and 6 (5%) had BP values that corresponded to stage 2 hypertension. Of the 139 children who did not receive ABPM, 54 patients met the clinical indications for receiving ABPM. The remaining 85 children did not meet the criteria for ABPM. The reasons why ABPM was not performed are presented in Table 2.
Review of the source of funding for the children who received ABPM showed that the costs were covered by the insurance company for 85.7% (108 of 126 patients) and were paid in full by the family for 14.3% (18 of 126 patients). For the children who did not receive ABPM, insurance denied coverage for 81.5% (44 of 54 patients) and families refused for 18.5% (10 of 54 patients). Insurance company denial of ABPM was more common for black patients; 43% of the black patients (23 of 54 patients) seen in the clinic were denied coverage for ABPM, compared with 18% of the white patients (14 of 78 patients; P = .0019, χ2 test) and 19% of the Hispanic patients (7 of 36; P = .02, χ2 test).
The sum of the charges for the evaluation of a hypertensive child was $3420 ($220 for the clinic visit plus $120 facility fee plus $488 for laboratory testing plus $442 for renal ultrasonography plus $2150 for echocardiography). In comparison, the sum of the charges for the initial evaluation of a child who proved to have WCH was $1265 ($220 for the clinic visit plus $120 facility fee plus $488 for laboratory testing plus $437 for ABPM). With the use of the observed 46% prevalence of WCH in our population with stage 1 hypertension in casual monitoring and with the assumption that children with ABPM-confirmed WCH do not require renal ultrasonography and echocardiography as part of their diagnostic evaluation, the addition of the charges for ABPM for every patient with stage 1 hypertension would result in a net reduction in charges after 3 patients. With this model, the use of ABPM for all patients would yield a predicted savings of $2.4 million for every 1000 patients referred because of stage 1 hypertension.
ABPM has been proved to be very useful in the evaluation of pediatric hypertension.21–23 It overcomes the limitation of office and intermittent home BP monitoring by providing frequent measurement and recording of BP throughout a 24-hour period. It is able to determine the degree to which the BP exceeds the upper limit of the reference range. The data obtained through ABPM are reproducible in the pediatric population.19 The monitor itself is well tolerated by children and successfully records sufficient data for definitive diagnosis.15–18
A concern with more-universal use of ABPM to differentiate WCH from true hypertension involves the limited pediatric ABPM normative data. Díaz and Garin24 evaluated the use of both task force normative data and central European ABPM normative data in the interpretation of ABPM results for hypertensive patients and found that normative data based on casual BP measurements cannot be applied to ABPM. Lurbe et al15 evaluated 241 healthy children in Spain and Reichert et al18 evaluated 294 healthy children in Germany to provide the initial ABPM reference values. The largest ABPM normative database is from a multicenter German study involving 1141 children.20 Because the normative values are based on a homogeneous population base of central European children, the ability to generalize the data and apply them to the US pediatric population has been questioned. There has been discussion of the need to develop US normative values for ABPM. Harshfield et al25 evaluated a small group of 300 US children in an attempt to establish reference values; however, the validity was limited by the small population base. We used the pediatric normative data established by Soergel et al20 in our evaluation of ABPM data obtained during initial evaluations of elevated BP. As suggested previously, a need to formulate a specific population-based set of pediatric ABPM normative data continues to exist.22,26
In our study, we demonstrated underdiagnosis of WCH in the children who did not receive ABPM. Fifty-four children did not receive ABPM despite casual BP values that suggested stage 1 hypertension. Those children had serial home BP measurements, and 47 (87%) were diagnosed as having stage 1 hypertension, compared with 7 children (13%) diagnosed as having WCH. In contrast, the prevalence of WCH among children with apparent stage 1 hypertension (according to clinic measurements) who received ABPM was 46%. This latter observation is consistent with the prevalence of WCH in pediatric populations reported previously (range: 44%–53%).5,6 Underdiagnosis of WCH and overdiagnosis of stage 1 hypertension result in unnecessary diagnostic testing and possible pharmacologic treatment for children without persistent hypertension. With the high prevalence of WCH in children, ABPM serves as a necessary tool to differentiate true hypertension from WCH in children with mildly elevated BP values and to prevent erroneous diagnosis and treatment of hypertension.
We also demonstrated that the use of ABPM to confirm the diagnosis of WCH in the pediatric population is highly cost-effective. National cost savings could be enormous, on the basis of the estimate of 1.47 million hypertensive children in the United States (2% of 73 460 567 children <18 years of age in households, using 2006 US census data).27 These savings are arguably an underestimate. The cost model included the savings from not doing a renal ultrasound and an echocardiogram on children with WCH only. Initial use of ABPM could also reduce laboratory testing and the number of clinic visits resulting in even greater savings. The current limitation to the use of ABPM for differentiation of WCH and stage 1 hypertension is in large part the availability of coverage from third-party payers. In addition, we found racial disparities in access to ABPM; 43% of the black patients seen in the hypertension clinic did not receive clinically indicated ABPM because of denial of coverage by third-party payers. It is hoped that improved third-party awareness of the use of ABPM in the pediatric population and the predicated cost savings improves access for indicated diagnostic evaluations.
This study was supported by National Institutes of Health grants to Dr Feig (grants DK064587 and DK07122).
- Accepted March 14, 2008.
- Address correspondence to Daniel I. Feig, MD, PhD, MS, Department of Pediatrics, Renal Section, Baylor College of Medicine, 6621 Fannin St, MC3-2482, Houston, TX 77030. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
What's Known on This Subject
Laboratory and imaging tests are recommended for all children with elevated BP. Children with white-coat hypertension are at lower risk of target-organ damage. Most insurance companies do not reimburse for ambulatory BP monitoring for children.
What This Study Adds
This study demonstrates the cost-effectiveness of the use of ambulatory BP monitoring in the initial evaluation of stage 1 hypertension.
- ↵National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents. The Fourth Report on the Diagnosis, Evaluation, and Treatment of High Blood Pressure in Children and Adolescents. Pediatrics.2004;114 (2 suppl 4th report):555– 576
- ↵Sorof JM, Lai D, Turner J, Poffenbarger T, Portman RJ. Overweight, ethnicity, and the prevalence of hypertension in school-aged children. Pediatrics.2004;113 (3):475– 482
- ↵Verdecchia P, Reboldi GP, Angeli F, et al. Short- and long-term incidence of stroke in white-coat hypertension. Hypertension.2005;45 (2):203– 208
- ↵Krmar RT, Berg UB. Long-term reproducibility of routine ambulatory blood pressure monitoring in stable pediatric renal transplant recipients. Am J Hypertens.2005;18 (11):1408– 1414
- ↵Harshfield GA, Alpert BS, Pulliam DA, Somes GW, Wilson DK. Ambulatory blood pressure recordings in children and adolescents. Pediatrics.1994;94 (2):180– 184
- ↵US Census Bureau. 2006 census data. Available at: http://factfinder.census.gov/servlet/ACSSAFFFacts. Accessed October 9, 2008
- Copyright © 2008 by the American Academy of Pediatrics