Abstract
Pediatric practice guidelines call for repeating an elevated office blood pressure (BP) at the same visit, but there are few data available to support this recommendation. We compared the visit results in children aged 3 to 17 years with a BP reading ≥95th percentile (n = 186 732) based on the initial BP and the mean of two BP readings, using electronic medical records from 2012–2015. Failure to repeat an initial BP reading ≥95th percentile would lead to a false “hypertensive” visit result in 54.1% of children who would require follow‐up visits. After an initial visit result indicating hypertension, hypertension stage I or stage II was sustained in 2.3% and 11.3% of youth during their next visits, respectively. In conclusion, only approximately half of the pediatric patients would be correctly classified based on their initial BP. The recommendation to repeat high BP during the same visit needs to be emphasized because it saves unnecessary follow‐up visits.
Keywords: blood pressure, children, guidelines, hypertension, screening
1. INTRODUCTION
High blood pressure (BP) is a major contributing risk factor for heart failure, heart attack, stroke, and chronic kidney disease and accounts for an estimated 18% of cardiovascular deaths in the United States.1 High BP is not limited to adults. It is estimated that 4% to 5% of youth have hypertension, and another 13% to 18% of youth have prehypertension.2, 3, 4, 5, 6, 7, 8 Children with elevated BP are 2 to 3 times more likely to develop essential hypertension in young adulthood.9 Approximately 30% to 40% of youth with sustained hypertension show early signs of end organ damage.10, 11 Youth with cardiovascular risk developing early in life will face higher rates of cardiovascular morbidity and early death.12 Therefore, screening for high BP in asymptomatic youth is recommended as an important strategy for early diagnosis and treatment of hypertension to prevent cardiovascular disease.13, 14, 15, 16, 17, 18, 19
While routine medical visits offer a good opportunity to screen for high BP in asymptomatic youth,19 the adherence to current BP screening guidelines is poor. Nationally representative data have shown that BP is not measured in two thirds of all pediatric visits and one third of pediatric preventive visits.20 When BP is measured, hypertension is often overlooked and underdiagnosed in pediatric populations until they transition to adult care.21, 22
Initial BP readings can be high for various reasons including measurement error. Hence, BP readings should be repeated during the same visit and the mean of BP readings used as the visit result.19 Repeating a high BP during the same visit can prevent misclassification of a patient's visit result. This has implications for clinical practice because it is recommended that follow‐up visits should be scheduled after a visit with a BP ≥95th percentile to confirm or rule out hypertension. Misclassification of a patient's visit result as hypertensive may result in unnecessary follow‐up visits, which are a burden for both the patient and the healthcare system. However, studies are needed to quantify the magnitude of misclassification arising from the failure to repeat an initial high BP reading.
To estimate the benefit of adhering to standardized protocols for BP screening as recommended by the American Academy of Pediatrics and others,15, 16, 17, 18, 19 we assessed the misclassification of pediatric patients as hypertensive based on an initial BP reading, which could be avoided by a repeated reading during the same visit. The classification of an individual's BP as normotensive or hypertensive during the same visit was assessed based on: (1) the initial BP, (2) the mean of two BP readings, and (3) the lower of two BP readings. We also examined the occurrence of follow‐up visits after an initial office visit with any hypertensive BP and the proportion of youth with sustained hypertension if follow‐up visits were completed as recommended by the American Academy of Pediatrics and others.15, 16, 17, 18, 19
2. PATIENTS AND METHODS
2.1. Study setting and population
The Kaiser Permanente Southern California (KPSC) Children's Health Study includes children and adolescents who are actively enrolled in a KPSC health plan in 2007 or later. It is an ongoing cohort study that continuously enrolls new patients who reaching the target age range or join the health plan.23 In 2010, KPSC members represented approximately 16.2% of the total coverage area population.24 The cohort follow‐up is conducted through passive surveillance of clinical care information using an electronic medical record (EMR) system. All administrative and clinical data are linked through a unique medical record number and include membership information, medical encounters, and other healthcare information. A detailed description of the cohort design and measurement protocol has been published elsewhere.23
The analytic population of the present study was comprised of a subpopulation of the KPSC Children's Health Study. During the study period, we identified 826 648 youth between the ages of 3 and 17 years who were KPSC members between January 1, 2012, and December 31, 2015, and had a medical encounter with at least one BP reading and valid height. We excluded youth who were pregnant (n = 1696) or whose medical record indicated fever (n = 65 240) when BP was measured. We also excluded youth with a preexisting diagnosis of hypertension (n = 2231) and chronic conditions known to significantly affect BP (n = 1686), such as growth hormone deficiency or overproduction, aortic coarctation, chronic renal disease, congenital adrenal hyperplasia, Cushing syndrome, hyperaldosteronism, and/or hyperthyroidism.25 Of the remaining cohort of 755 795 youth, we identified 186 732 (24.7%) with at least one BP reading ≥95th percentile for sex, age, and height during the 4‐year study period who were used as the final analytical cohort. The study protocol was reviewed and approved by KPSC's institutional review board.
2.2. Body weight and height
Body weight and height were routinely measured and extracted from the EMR. Definitions for overweight and obesity in youth at the first visit with high BP are based on the sex‐specific body mass index (BMI)‐for‐age growth charts developed by the Centers for Disease Control and Prevention.26 Overweight was defined as BMI‐for‐age ≥85th and < 95th percentile (or BMI ≥25 kg/m2), obesity as ≥95th percentile (or BMI ≥30 kg/m2) and severe obesity as BMI‐for‐age ≥1.2 × 95th percentile (or BMI ≥35 kg/m2).27, 28
2.3. Race and ethnicity
We obtained race and ethnicity information from health plan administrative records and birth records. We categorized race/ethnicity as non‐Hispanic white, Hispanic (regardless of race), African American, Asian or Pacific Islander, and other or unknown race/ethnicity.
2.4. Socioeconomic status
Since individual‐level education and household income were not available through EMRs, neighborhood education and neighborhood household income at the first visit with high BP were used to indicate socioeconomic status. These population‐level indicators were estimated by geocoding cohort members’ addresses to 2010 US census block data.29 We also used insurance through government healthcare assistance programs such as Medicaid as an additional proxy for socioeconomic status.
2.5. BP screening and classification
BP was measured routinely at the beginning of almost every outpatient medical visit, as described in detail elsewhere.30 Nurses and medical assistants were trained according to guidelines of the American Association of Critical Care Nurses for pediatric care.31 Digital devices (Welch Allyn Connex series, Welch Allyn Inc.) are the preferred BP measurement devices at KPSC. However, it is possible that in some cases, a wall‐mounted aneroid sphygmomanometer (Welch Allyn Inc.) was used instead of the preferred device. Medical staff were trained to ensure that the bladder inside the cuff encircled 80% to 100% of the circumference of the right arm according to standard recommendations.31 A full range of different cuff sizes were available at the locations where patient BP was taken. All staff members who measure BP as part of their daily job functions are certified in BP measurement during their initial staff orientation and recertified annually. BP readings are manually entered into the EMR system. BP is automatically translated into BP percentiles and available together with BP history of the previous two visits in the progress notes. However, no alerts were used to inform whether a BP should be considered elevated and no best practice recommendations were provided during the study period on recommended follow‐up and/or treatment.
BP readings for all outpatient medical encounters were extracted from the EMR. We classified BP using the recommendations of the Fourth Report on the Diagnosis, Evaluation, and Treatment of High Blood Pressure in Children and Adolescents of the National High Blood Pressure Education Program.15, 19 Depending on sex, age, and height, BP was defined as hypertensive stage 1 if it was ≥95th percentile but <99th percentile plus 5 mm Hg, and hypertensive stage 2 if ≥99th percentile plus 5 mm Hg.
2.6. Study outcomes
The primary study outcome was the misclassification of the final visit outcome after an initial BP reading ≥95th percentile for sex, age, and height during outpatient visits (n = 186 732 youth). Using the mean of two consecutive BPs during an outpatient medical visit as the gold standard,15 we classified an individual's visit result as correct, false‐positive high BP (defined as BP status higher than correct), false negative (defined as BP status lower than correct), or unknown (if initial BP was not repeated, n = 18 172) compared with an assessment based on the initial BP. Similarly, we examined the classification of an individual's visit result as correct, false positive, or false negative if the lower of two BP readings would be used compared with the initial reading. Among youth who were not classified as unknown during their initial visit (ie, had a repeated BP reading), the secondary study outcome was sustained hypertension defined as a total of three consecutive visits with a BP ≥95th percentile for sex, age, and height (n = 168 560). These youths were classified as sustained if they had any three consecutive visits with a BP ≥95th percentile. If all three BPs were ≥95th percentile but <99th percentile plus 5 mm Hg, then sustained high BP was classified as hypertension stage I. If any one BP recording was ≥99th percentile plus 5 mm Hg, then sustained high BP was classified as hypertension stage II. Other youth were classified as either, “not sustained,” or unknown (if no or incomplete follow‐up occurred).
2.7. Statistical analysis
Summary statistics were calculated for sociodemographic characteristics. To investigate the relationship between sociodemographic characteristics (sex, age, race, and baseline weight class) and the repetition of BP measurement, log binomial models from the SAS procedure GENMOD were used. Crude and adjusted relative risk and confidence intervals are reported. Models were adjusted for all other sociodemographic characteristics listed above. After an initial high BP ≥95th percentile for sex, age, and height, the follow‐up was classified as completed if: (1) one follow‐up visit occurred with a BP < 95th percentile, or (2) two follow‐up visits occurred with a BP ≥95th percentile, or (3) one follow‐up visit occurred with a BP ≥95th percentile and one <95th percentile.15, 16, 17, 18, 19 Log binomial models were used to describe the association between patient characteristics and completion of recommended follow‐up. All analyses were performed using SAS statistical software version 9.3 (SAS Institute Inc).
3. RESULTS
3.1. Study population characteristics
Of 755 795 youth aged between 3 and 17 years, 186 732 youth (24.7%) had a first BP reading ≥95th percentile for sex, age, and height between January 1, 2012, and December 31, 2015. The majority of youth was Hispanic (52.4%) and had normal body weight (51.3%). About 26.5% of youth received healthcare services under a state subsidized health plan such as MediCal. Among 186 732 youth with high BP, 167 402 (89.6%) of youth had an initial BP ≥95th percentile indicating hypertension stage I and 19 330 (10.4%) ≥99th + 5 mm Hg percentile indicating hypertension stage II (Table 1). A repetition of BP within the same visit was documented in the EMR for 18.3% of visits indicating hypertension stage I and 50.9% indicating hypertension stage II (P < .001). In crude analysis, a high BP reading ≥95th percentile was more likely to be repeated during the same visit if the patient was male, older, overweight to obese, Asian, or Hispanic. After multivariable adjustment, the likelihood for a repeated reading after an initial BP ≥95th percentile was no longer significant for Hispanic youth. After multivariable adjustment, a high BP reading ≥99th + 5 mm Hg percentile was more likely to be repeated during the same visit if the patient was older, and less likely if the patient was Asian. The presence of obesity was not associated with the likelihood for a repeated BP after an initial BP reading ≥99th + 5 mm Hg percentile.
Table 1.
Cohort characteristics and repetition of a high BP during outpatient medical visits for 186 732 youth in 2012–2015 with a first BP ≥95th percentile (hypertension stage I) or ≥99th percentile plus 5 mm Hg (hypertension stage II)
| Totala | Repetitionb | RR (95% CI) | |||
|---|---|---|---|---|---|
| Yes | No | Crude | Adjusted | ||
| First BP ≥95th percentile and <99th percentile plus 5 mm Hg (n = 167 402) | |||||
| Sex | |||||
| Male | 88 437 (52.8) | 18 210 (20.6) | 70 227 (79.4) | Reference | Reference |
| Female | 78 965 (47.5) | 12 355 (15.6) | 66 610 (84.4) | 0.76 (0.74–0.78) | 0.76 (0.74–0.77) |
| Age, y | |||||
| 3–5 | 46 480 (27.8) | 4184 (9.0) | 42 296 (91.0) | Ref | Ref |
| 6–11 | 53 822 (32.2) | 7420 (13.8) | 46 402 (86.2) | 1.53 (1.48–1.59) | 1.50 (1.45–1.55) |
| 12–17 | 67 100 (40.1) | 18 961 (28.3) | 48 139 (71.7) | 3.14 (3.04–3.24) | 3.07 (2.98–3.17) |
| Race/ethnicity | |||||
| Non‐Hispanic/white | 40 790 (24.4) | 7035 (17.2) | 33 755 (82.8) | Reference | Reference |
| Hispanic | 87 814 (52.5) | 16 271 (18.5) | 71 543 (81.5) | 1.07 (1.05–1.10) | 1.01 (0.99–1.04) |
| African American | 13 626 (8.1) | 2354 (17.3) | 11 272 (82.7) | 1.00 (0.96–1.05) | 0.98 (0.94–1.02) |
| Asian/Pacific Islander | 15 568 (9.3) | 3162 (20.3) | 12 406 (79.7) | 1.18 (1.13–1.22) | 1.21 (1.17–1.26) |
| Other/unknown | 9604 (5.7) | 1743 (18.1) | 7861 (81.9) | 1.05 (1.00–1.10) | 1.04 (1.00–1.09) |
| BMI class | |||||
| Underweight/normal | 86 948 (51.9) | 13 863 (15.9) | 73 085 (84.1) | Reference | Reference |
| Overweight | 31 057 (18.6) | 5992 (19.3) | 25 065 (80.7) | 1.21 (1.18–1.24) | 1.07 (1.04–1.10) |
| Moderately obese | 31 654 (18.9) | 6559 (20.7) | 25 095 (79.3) | 1.30 (1.27–1.33) | 1.12 (1.10–1.15) |
| Severely obese | 17 721 (10.6) | 4138 (23.4) | 13 583 (76.6) | 1.46 (1.42–1.51) | 1.17 (1.14–1.21) |
| Unknown | 22 (0) | 13 (59.1) | 9 (40.9) | 3.71 (2.62–5.25) | 2.37 (1.74–3.24 |
| First BP ≥99th percentile plus 5 mm Hg (n = 19 330) | |||||
| Sex | |||||
| Male | 10 430 (54.0) | 5560 (53.3) | 4870 (46.7) | Reference | Reference |
| Female | 8900 (46.0) | 4280 (48.1) | 4620 (51.9) | 0.90 (0.88–0.93) | 0.97 (0.95–0.98) |
| Age, y | |||||
| 3–5 | 6090 (31.5) | 1332 (21.9) | 4758 (78.1) | Reference | Reference |
| 6–11 | 5318 (27.5) | 2338 (44.0) | 2980 (56.0) | 2.01 (1.90–2.13) | 1.26 (1.12–1.30) |
| 12–17 | 7922 (41) | 6170 (77.9) | 1752 (22.1) | 3.56 (3.39–3.74) | 1.67 (1.67–1.71) |
| Race/ethnicity | |||||
| Non‐Hispanic/white | 4760 (24.6) | 2318 (48.7) | 2442 (51.3) | Reference | Reference |
| Hispanic | 10 065 (52.1) | 5240 (52.1) | 4825 (47.9) | 1.07 (1.03–1.11) | 0.99 (0.97–1.01) |
| African American | 1548 (8) | 796 (51.4) | 752 (48.6) | 1.06 (1.10–1.12) | 0.99 (0.96–1.02) |
| Asian/Pacific Islander | 1768 (9.1) | 862 (48.8) | 906 (51.2) | 1.00 (0.95–1.06) | 1.03 (1.00–1.06) |
| Other/unknown | 1189 (6.2) | 624 (52.5) | 565 (47.5) | 1.08 (1.01–1.15) | 1.01 (0.98–1.05) |
| BMI class | |||||
| Underweight/normal | 8836 (45.7) | 3622 (41.0) | 5214 (59.0) | Reference | Reference |
| Overweight | 3391 (17.5) | 1871 (55.2) | 1520 (44.8) | 1.35 (1.29–1.40) | 1.03 (1.01–1.06) |
| Moderately obese | 4036 (20.9) | 2326 (57.6) | 1710 (42.4) | 1.41 (1.36–1.46) | 1.04 (1.02–1.06) |
| Severely obese | 3056 (15.8) | 2012 (65.8) | 1044 (34.2) | 1.61 (1.55–1.66) | 1.06 (1.03–1.08) |
| Unknown | 11 (0.1) | 9 (81.8) | 2 (18.2) | 2.00 (1.51–2.64) | 1.20 (0.67–2.14) |
Abbreviations: BMI, body mass index; CI, confidence interval.The relative risk (RR) was calculated using a log binomial model with and without adjustment for all variables listed in the table.
Data are provided for total cohorta as column percentage and by blood pressure (BP) repetitionb as row percentage.
3.2. Hypertension stage I
In youth for which the initial high BP percentile was repeated during their visit (n = 30 565), 51.2% of youth with an initial BP ≥95th percentile had a mean of two consecutive BP < 95th percentile and were false‐positive results; these patients would be required to schedule potentially unnecessary follow‐up appointments if the initial BP reading would not have been repeated (Figure 1). Only 48.2% of youth also had a mean BP ≥95th percentile. Another 1.2% of youth had a mean BP ≥95th percentile but an initial BP reading <95th percentile and would be missed based on their initial BP (ie, false negative). In 71.4% of youth with an initial BP ≥95th percentile, the lower of two BP readings was <95th percentile.
Figure 1.
Visit result classification of pediatric patients with blood pressure (BP) indicating hypertension stage I or II based on the initial BP reading compared with the mean of two blood pressure readings. N/A indicates not available.
In youth with a BP ≥95th percentile during an outpatient visit, two more follow‐up visits are recommended to meet a diagnosis of hypertension and further workup should be considered. About 32.2% of youth with a visit indicating hypertension stage I completed the recommended follow‐up within 3 months (Table 2). Completion of follow‐up visits varied slightly but significantly and was somewhat higher in older and lower in minority youth.
Table 2.
Completion of the recommended follow‐up after an initial outpatient visit with high BP in youth with a first BP ≥95th percentile (hypertension stage I) or ≥99th percentile plus 5 mm Hg (hypertension stage II) shown as relative risk
| Totala | Time to complete follow‐up, No. (%) b | RR (95% CI) for completion ≤3 mo | |||
|---|---|---|---|---|---|
| ≤3 mo | >6 mo* | Crude | Adjusted | ||
| First BP ≥95th percentile and <99th percentile plus 5 mm Hg | |||||
| N | 155 901 | 50 155 (32.2) | 55 490 (67.8) | ||
| BP repeated at initial visit | |||||
| No | 135 980 (87.2) | 43 903 (32.3) | 92 077 (67.7) | Reference | Reference |
| Yes | 19 921 (12.8) | 6252 (31.4) | 13 669 (68.6) | 0.97 (0.95–0.99) | 0.92 (0.90–0.94) |
| Sex | |||||
| Male | 81 649 (52.4) | 26 651 (32.6) | 54 998 (67.4) | Reference | Reference |
| Female | 74 252 (47.6) | 23 504 (31.7) | 50 748 (68.3) | 0.97 (0.96–0.98) | 0.97 (0.95–0.98) |
| Age at first BP, y | |||||
| 3–5 | 44 654 (28.6) | 12 780 (28.6) | 31 874 (71.4) | Reference | Reference |
| 6–11 | 50 668 (32.5) | 15 997 (31.6) | 34 671 (68.4) | 1.10 (1.08–1.12) | 1.11 (1.08–1.13) |
| 12–17 | 60 579 (38.9) | 21 378 (35.3) | 39 201 (64.7) | 1.23 (1.21–1.26) | 1.24 (1.22–1.27) |
| Race/ethnicity | |||||
| Non‐Hispanic/white | 38 082 (24.4) | 12 888 (33.8) | 25 194 (66.2) | Reference | Reference |
| Hispanic | 81 807 (52.5) | 26 779 (32.7) | 55 028 (67.3) | 0.97 (0.95–0.98) | 0.96 (0.94–0.98) |
| African American | 12 744 (8.2) | 4013 (31.5) | 8731 (68.5) | 0.93 (0.90–0.96) | 0.93 (0.90–0.95) |
| Asian/Pacific Islander | 14 246 (9.1) | 3934 (27.6) | 10 312 (72.4) | 0.82 (0.79–0.84) | 0.82 (0.80–0.85) |
| Other/unknown | 9022 (5.8) | 2541 (28.2) | 6481 (71.8) | 0.83 (0.80–0.86) | 0.83 (0.80–0.86) |
| BMI closest to first BP | |||||
| Underweight/normal | 47 084 (51.6) | 11 861 (25.2) | 28 924 (61.4) | Reference | Reference |
| Overweight | 15 733 (18.5) | 3862 (24.5) | 9879 (62.8) | 0.97 (0.94–1.01) | 0.98 (0.95–1.02) |
| Moderately obese | 16 533 (19.1) | 4132 (25.0) | 10 247 (62.0) | 0.99 (0.96–1.02) | 1.00 (0.97–1.03) |
| Severely obese | 10 108 (10.8) | 2492 (24.7) | 6393 (63.3) | 0.98 (0.94–1.02) | 0.99 (0.95–1.03) |
| Unknown | 58 (0.0) | 10 (17.2) | 47 (81.1) | 0.68 (0.39–1.20) | 0.73 (0.42–1.29) |
| First BP ≥99th percentile plus 5 mm Hg | |||||
| N | 12 659 | 4013 (31.7) | 8646 (68.3) | ||
| BP repeated at initial visit | |||||
| No | 9384 (74.1) | 2884 (30.7) | 6500 (69.3) | Reference | Reference |
| Yes | 3275 (25.9) | 1129 (34.5) | 2146 (65.5) | 1.12 (1.06–1.19) | 1.03 (0.96–1.09) |
| Sex | |||||
| Male | 6653 (52.6) | 2112 (31.7) | 4541 (68.3) | Reference | Reference |
| Female | 6006 (47.4) | 1901 (31.7) | 4105 (68.3) | 1.00 (0.95–1.05) | 1.00 (0.95–1.05) |
| Age at first BP, y | |||||
| 3–5 | 5090 (40.2) | 1447 (28.4) | 3643 (71.6) | Reference | Reference |
| 6–11 | 3714 (29.3) | 1176 (31.7) | 2538 (68.3) | 1.11 (1.04–1.19) | 1.12 (1.05–1.20) |
| 12–17 | 3855 (30.5) | 1390 (36.1) | 2465 (63.9) | 1.27 (1.19–1.35) | 1.27 (1.18–1.36) |
| Race/ethnicity | |||||
| Non‐Hispanic/white | 3140 (24.8) | 996 (31.7) | 2144 (68.3) | Reference | Reference |
| Hispanic | 6560 (51.8) | 2166 (33.0) | 4394 (67.0) | 1.04 (0.98–1.11) | 1.03 (0.97–1.10) |
| African American | 1019 (8.1) | 313 (30.7) | 706 (69.3) | 0.97 (0.87–1.08) | 0.95 (0.86–1.06) |
| Asian/Pacific Islander | 1158 (9.1) | 318 (27.5) | 840 (72.5) | 0.87 (0.78–0.96) | 0.88 (0.79–0.97) |
| Other/unknown | 782 (6.2) | 220 (28.1) | 562 (71.9) | 0.89 (0.78–1.00) | 0.88 (0.77–0.99) |
| BMI closest to first BP | |||||
| Underweight/normal | 6139 (48.5) | 1894 (30.9) | 4245 (69.1) | Reference | Reference |
| Overweight | 2101 (16.6) | 684 (32.6) | 1417 (67.4) | 1.06 (0.98–1.13) | 0.99 (0.92–1.07) |
| Moderately obese | 2474 (19.6) | 801 (32.4) | 1673 (67.6) | 1.05 (0.98–1.12) | 0.97 (0.90–1.04) |
| Severely obese | 1936 (15.3) | 632 (32.6) | 1304 (67.4) | 1.06 (0.98–1.14) | 0.94 (0.87–1.02) |
| Unknown | 9 (0.0) | 2 (0.0) | 7 (0.0) | 0.72 (0.21–2.45) | 0.67 (0.20–2.26) |
Abbreviations: BMI, body mass index; CI, confidence interval.Data are provided for total cohorta as column percentage and by blood pressure (BP) repetitionb as row percentage.
*Includes patients with no follow‐up at the end of the study period.
For the majority of youth with an initial BP ≥95th percentile during an outpatient visit, the required follow‐up to confirm or rule out hypertension was not completed and their BP status could not be determined (n = 114 206, 67.8%). Among those with the required follow‐up visits (n = 54 354), hypertension was not sustained in 97.7% of youth with an initial BP ≥95th percentile (Figure 2). Only 2.3% of youth continued to have BP ≥95th percentile indicating hypertension stage I.
Figure 2.
Hypertension status in pediatric patients with an initial visit indicating hypertension (HTN) stage I or stage II after completing their follow‐up visits.
3.3. Hypertension stage II
In youth for which the initial high BP ≥99th + 5 mm Hg percentile was repeated during their visit (n = 9840), 35.0% of youth also had a mean BP ≥99th + 5 mm Hg percentile (Figure 1). Among youth with an initial BP ≥99th + 5 mm Hg percentile, the mean of two BP was <95th percentile in 65.0% of youth who would have false‐positive results if the BP would not have been repeated. In 81.0% of youth with an initial BP ≥99th + 5 mm Hg percentile, the lower of two BP readings was <95th percentile.
Approximately 31.7% of patients indicating hypertension stage II completed the recommended follow‐up within 3 months (Table 2 ). This was significantly higher in older than in younger patients (36.1% in those aged 12–17 years vs 28.4% in those aged 3–5 years) and lower in most minority youth. For the majority of youth with an initial BP ≥99th + 5 mm Hg percentile during an outpatient visit, the required follow‐up to confirm or rule out hypertension was not completed and could not be determined (n = 8646, 68.3%). Among those with the required follow‐up visits, hypertension was sustained in 11.3% of youth with an initial BP ≥99th + 5 mm Hg percentile but not sustained in 88.7% (Figure 2). Hypertension was sustained at the hypertension stage II level in 7.3% and at the hypertension stage I level in 4.0% of those with an initial BP ≥99th + 5 mm Hg percentile during an outpatient visit.
4. DISCUSSION
BP in the hypertensive range is common in clinical care as first readings. Almost one of four youth between the ages of 3 and 17 years had at least one BP ≥95th percentile for sex, age, and height. Current clinical practice guidelines for evaluation of elevated BP in children and adolescents recommend repeating an elevated office BP reading at the same visit, but there are few data available to support this recommendation. The present study showed that BP was not repeated during the same visit as recommended in almost 80% of youth. Less than half of the pediatric patients (45.9%) would be correctly classified based on their initial BP; the failure to repeat a high BP at the time of the initial visit resulted in a large proportion of false‐positive results (54.1%) from general BP screening. Our results also indicate that if asymptomatic youth are screened and followed up as recommended within 3 months, hypertension is confirmed in only a small proportion of youth. About 2% of youth with an initial visit indicating hypertension stage I, and 11% of youth with an initial visit indicating hypertension stage II continued to have high BP during their follow‐up visits.
Several issues pose significant challenges for pediatricians to effectively screen high BP in youth. For adults, simple and straightforward BP cutoff values guide clinical decision‐making, while the definition of high BP in youth is much more complex. The definition of normal BP values in children and adolescents varies by sex, age, and height. Values from BP readings need to be converted into BP percentiles available in standardized tables, which takes time. Because BP in youth varies, it is recommended to repeat high BP readings during the same visit and to schedule follow‐up visits to confirm sustained hypertension in three independent medical visits for a diagnosis of hypertension and for further evaluation of underlying causes in youth. Our understanding of the implications of nonadherence to screening guidelines with regard to the potential misclassification of a normotensive patient as having hypertension when conducting studies using EMR data or missing patients with true hypertension is limited.
While screening for high BP in asymptomatic youth is recommended as an important strategy for early diagnosis and treatment of hypertension to prevent cardiovascular disease,13, 14, 15, 16, 17, 18, 19 the implications of general BP screening for clinical care have to be considered. The results of our study show that almost 25% of youth had at least one high BP value recorded in their EMR when a general screening was performed at every visit. These visits are not limited to well‐child visits but include acute care visits. In a recent large study of several healthcare systems, only 5% of youth had a high BP value during well‐child visits.6 Considering the high proportion of youth with a high BP reading resulting from general screening at every visit, following standardized screening procedures and adherence to recommendations, such as the repetition of initial high BP readings, is crucial to avoid unnecessary follow‐up visits but also to prevent the possibility that true hypertension is not overlooked and underdiagnosed in high‐risk youth.21, 22
The failure to repeat a high BP at the time of the initial visit is concerning because the large number of false‐positives results (54.1%) from general BP screening in asymptomatic youth would almost double the number of youth estimated to have a hypertensive BP value and would require additional follow‐up visits. The calculation of the mean BP during the same visit may be a barrier towards BP screening in youth. If the lower of two BP readings would be used as the visit result, an initial BP ≥95th percentile was not confirmed in 71.4% of youth (81.0% of youth with an initial BP ≥99th + 5 mm Hg percentile). Appropriate training for providers and office staff on the importance of using standardized BP screening protocols, especially the importance to repeat an initial high BP during the same visit is necessary. If the proportion of false‐positive results can be reduced, resources can be directed towards youth who benefit from follow‐up by confirming or ruling out true hypertension. Providers may be inclined to dismiss a first high BP, especially if a child is upset. It may be common practice to accept the last BP as the valid BP, and averaging BPs may not occur as recommended. Using the lower versus the mean of BP readings during a visit may simplify the efforts for providers, but studies are needed to compare both methods in their effectiveness to detect youth with true hypertension.
To overcome some of the challenges associated with BP screening in the clinical setting, in which the present study was conducted, EMR tools were developed that translate absolute BP values into BP percentiles and provide easy access to BP history during a visit. Results from the present study indicate that simple tools translating BP into percentiles were not successful to achieve a high adherence to current recommendations to screen for high BP in asymptomatic youth. The adherence to recommendations such as the repetition of high BP readings during the same visit and the scheduling of follow‐up visits was low. Studies are needed to test the efficiency of EMR‐embedded decision support tools that populate alerts to the provider regarding high BP and the recommended best practice for each patient based on their BP and medical history.
BP in youth varies and a high BP value often reverses to normal without any intervention.7 Therefore, the criteria for diagnosing hypertension in youth require sustained hypertension in three independent medical visits. If an initial high BP is not recognized or dismissed, a follow‐up visit may not be scheduled as recommended. In a school‐based study, the proportion of youth with elevated BP values based on one visit was five times higher than that based on three measurements taken within a few weeks.32 In our study, initial hypertension was confirmed during follow‐up visits in only 2.3% of pediatric patients with an initial high BP indicating hypertension stage I and 11.3% of pediatric patients with an initial high BP indicating hypertension stage II. The majority of youth did not meet the criteria for sustained hypertension. Future studies in a clinical environment with high adherence to the current screening guidelines are necessary to estimate the rate of sustained hypertension across different population groups defined by sex, age, race, and other potential risk factors.
Findings from our study indicate that only about one of three patients complete the recommended follow‐up visits after a visit with an initial high BP within 3 months. This finding is consistent with other studies, in which 80% of pediatric patients with high BP did not have a follow‐up meeting as recommended.20, 21, 22 Automated tools and alert systems may support this process, as shown in other settings.22 However, the results from the present study indicate that simple tools translating BP into percentiles, as used here, are not sufficient and need to be replaced by more sophisticated and automated computer‐assisted support tools that populate alerts and provide best practice advice.
5. STUDY STRENGTHS AND LIMITATIONS
Several strength and limitations should be noted. The large and diverse population‐based sample allows us to estimate the prevalence of high BP and the potential misclassification in defined subpopulations in routine outpatient medical visits. Several measurements were taken to ensure high quality and that BP readings followed strict protocols according to published recommendations.19, 31, 33 However, this may not always be the case in clinical practice. We cannot ensure that BPs were accurately measured during all routine care visits. The intensity of in‐house training may have varied by medical center. Deviations of the preferred measurement method may have occurred. It is possible that additional BP measurements were performed but were not documented in the EMR system or captured through data abstraction. This would have led to an underestimation of adherence to current BP screening guidelines. However, because of the current clinical workflows, constant training, and quality checks, these deviations are unlikely to have affected the results. Another potential limitation is the type of device used in the clinical settings. Automated digital BP monitors are the preferred devices, while standard BP values were developed based on standard mercury sphygmomanometers. However, the use of automated devices is frequent in clinical settings.34 Measurement error may explain the high number of youth with initial high BP. It may also be difficult to get accurate BP readings for children regardless of the type of device in real‐life clinical situations when assessing BP at every visit. While we cannot exclude errors from the use of automated digital BP monitors, our results cannot be explained by a slight but systematic overestimation caused by these devices. However, the proportion of youth with high BP observed here was high compared with other studies using automated digital devices.6 Moreover, the proportion of false‐positive results was particularly high when using the lower of two readings as standard. These facts suggest that our observation was not a device issue but a matter of practicality or training to achieve accurate BP readings for children during routine clinic visits. Last, the adherence to recommendations to repeat BPs during the same visit and to schedule follow‐up visits was low. This may limit the generalizability of the magnitude of misclassification and the proportion of youth with sustained high BP. However, we carefully examined a variety of risk factors that may indicate selection bias. For the repetition of high BP readings during the same visit, we observed that patients were more likely to have a repeated BP if they were male, older, and obese. This observation suggests that the magnitude of misclassification may be higher in patients who are younger, female, and not obese. In contrast, scheduling appropriate follow‐up visits after an initial visit with high BP was remarkably independent of known risk factors for hypertension and suggests that high BP is ignored in youth regardless of risk factors.
6. CONCLUSIONS
Our findings suggest that a high initial BP reading in youth is common, while the proportion of youth with sustained hypertension is low. Repeating an initial hypertensive BP during the same visit as recommended by current guidelines is important to rule out false‐positive results and to avoid unnecessary follow‐up visits to confirm whether the hypertensive BP persists. Personnel training needs to emphasize the importance of repeating high BP readings.
CONFLICT OF INTEREST
The authors report no specific funding in relation to this research and have no conflicts of interest to disclose.
DISCLOSURES
The present study was supported by Kaiser Permanente Direct Community Benefit Funds.
Koebnick C, Mohan Y, Li X, et al. Failure to confirm high blood pressures in pediatric care—quantifying the risks of misclassification. J Clin Hypertens. 2018;20:174–182. 10.1111/jch.13159
REFERENCES
- 1. Ezzati M, Oza S, Danaei G, et al. Trends and cardiovascular mortality effects of state‐level blood pressure and uncontrolled hypertension in the United States. Circulation. 2008;117:905‐914. [DOI] [PubMed] [Google Scholar]
- 2. Din‐Dzietham R, Liu Y, Bielo MV, et al. High blood pressure trends in children and adolescents in national surveys, 1963 to 2002. Circulation. 2007;116:1488‐1496. [DOI] [PubMed] [Google Scholar]
- 3. McNiece KL, Poffenbarger TS, Turner JL, et al. Prevalence of hypertension and pre‐hypertension among adolescents. J Pediatr. 2007;150:640‐644. 644 e641. [DOI] [PubMed] [Google Scholar]
- 4. Obarzanek E, Wu CO, Cutler JA, et al. Prevalence and incidence of hypertension in adolescent girls. J Pediatr. 2010;157:461‐467. [DOI] [PubMed] [Google Scholar]
- 5. Thompson M, Dana T, Bougatsos C, et al. U.S. Preventive Services Task Force Evidence Syntheses, formerly Systematic Evidence Reviews. Screening for Hypertension in Children and Adolescents to Prevent Cardiovascular Disease: Systematic Review for the U.S. Preventive Services Task Force. Rockville, MD: Agency For Healthcare Research and Quality (US); 2013. [PubMed] [Google Scholar]
- 6. Lo JC, Sinaiko A, Chandra M, et al. Prehypertension and hypertension in community‐based pediatric practice. Pediatrics. 2013;131:e415‐e424. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Falkner B, Gidding SS, Portman R, et al. Blood pressure variability and classification of prehypertension and hypertension in adolescence. Pediatrics. 2008;122:238‐242. [DOI] [PubMed] [Google Scholar]
- 8. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics–2015 update: a report from the American Heart Association. Circulation. 2015;131:e29‐e322. [DOI] [PubMed] [Google Scholar]
- 9. Bao W, Threefoot SA, Srinivasan SR, et al. Essential hypertension predicted by tracking of elevated blood pressure from childhood to adulthood: the Bogalusa Heart Study. Am J Hypertens. 1995;8:657‐665. [DOI] [PubMed] [Google Scholar]
- 10. McNiece KL, Gupta‐Malhotra M, Samuels J, et al. Left ventricular hypertrophy in hypertensive adolescents: analysis of risk by 2004 National High Blood Pressure Education Program Working Group staging criteria. Hypertension. 2007;50:392‐395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Brady TM, Redwine KM, Flynn JT, et al. Screening blood pressure measurement in children: are we saving lives? Pediatr Nephrol. 2014;29:947‐950. [DOI] [PubMed] [Google Scholar]
- 12. Saydah S, Bullard KM, Imperatore G, et al. Cardiometabolic risk factors among US adolescents and young adults and risk of early mortality. Pediatrics. 2013;131:e679‐e686. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Samuels JA, Bell C, Flynn JT. Screening children for high blood pressure: where the US preventive services task force went wrong. J Clin Hypertens (Greenwich). 2013;15:526‐527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. U.S. Preventive Services Task Force . Screening for high blood pressure: U.S. Preventive Services Task Force reaffirmation recommendation statement. Ann Intern Med. 2007;147:783‐786. [DOI] [PubMed] [Google Scholar]
- 15. Expert Panel on Integrated Guidelines for Cardiovascular Health , Risk Reduction in Children, Adolescents . Expert panel on integrated guidelines for cardiovascular health and risk reduction in children and adolescents: summary report. Pediatrics. 2011;128(suppl 5):S213‐S256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Pickering TG, Hall JE, Appel LJ, et al. Recommendations for blood pressure measurement in humans and experimental animals: part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Hypertension. 2005;45:142‐161. [DOI] [PubMed] [Google Scholar]
- 17. Lurbe E, Cifkova R, Cruickshank JK, et al. Management of high blood pressure in children and adolescents: recommendations of the European Society of Hypertension. J Hypertens. 2009;27:1719‐1742. [DOI] [PubMed] [Google Scholar]
- 18. Hagan JF, Shaw JS, Duncan P. 3rd ed. Pocket Guide, ed. Bright Futures: Guidelines for Health Supervision of Infants, Children, and Adolescents. Elk Grove Village, IL: American Academy of Pediatrics; 2008. [Google Scholar]
- 19. National High Blood Pressure Education Program Working Group on High Blood Pressure in CA . The Fourth Report on the Diagnosis, Evaluation, and Treatment of High Blood Pressure in Children and Adolescents, Revised version May 2005. Bethesda, MD: U.S. Department of Health and Human Services, National Institutes of Health, National Heart, Lung, and Blood Institute; 2005. 05‐5267. [Google Scholar]
- 20. Shapiro DJ, Hersh AL, Cabana MD, et al. Hypertension screening during ambulatory pediatric visits in the United States, 2000‐2009. Pediatrics. 2012;130:604‐610. [DOI] [PubMed] [Google Scholar]
- 21. Hansen ML, Gunn PW, Kaelber DC. Underdiagnosis of hypertension in children and adolescents. JAMA. 2007;298:874‐879. [DOI] [PubMed] [Google Scholar]
- 22. Brady TM, Solomon BS, Neu AM, et al. Patient‐, provider‐, and clinic‐level predictors of unrecognized elevated blood pressure in children. Pediatrics. 2010;125:e1286‐e1293. [DOI] [PubMed] [Google Scholar]
- 23. Koebnick C, Coleman KJ, Black MH, et al. Cohort profile: the KPSC Children's Health Study, a population‐based study of 920 000 children and adolescents in southern California. Int J Epidemiol. 2012;41:627‐633. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Koebnick C, Langer‐Gould AM, Gould MK, et al. Sociodemographic characteristics of members of a large, integrated health care system: comparison with US Census Bureau data. Perm J. 2012;16:37‐41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Koebnick C, Black MH, Wu J, et al. The prevalence of primary pediatric prehypertension and hypertension in a real‐world managed care system. J Clin Hypertens (Greenwich). 2013;15:784‐792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Kuczmarski RJ, Ogden CL, Guo SS, et al. 2000 CDC growth charts for the United States: methods and development. Vital Health Stat. 2002;11:1‐190. [PubMed] [Google Scholar]
- 27. Flegal KM, Wei R, Ogden CL, et al. Characterizing extreme values of body mass index for age by using the 2000 Centers for Disease Control and Prevention growth charts. Am J Clin Nutr. 2009;90:1314‐1320. [DOI] [PubMed] [Google Scholar]
- 28. Ogden CL, Carroll MD, Kit BK, et al. Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA. 2014;311:806‐814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Chen W, Petitti DB, Enger S. Limitations and potential uses of census‐based data on ethnicity in a diverse community. Ann Epidemiol. 2004;14:339‐345. [DOI] [PubMed] [Google Scholar]
- 30. Koebnick C, Black MH, Wu J, et al. High blood pressure in overweight and obese youth: implications for screening. J Clin Hypertens (Greenwich). 2013;15:793‐805. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. American Association of Critical‐Care Nurses . American Association of Critical‐Care Nurses Procedure Manual for Pediatric Acute and Critical Care. St. Louis, MO: Saunders; 2007. [Google Scholar]
- 32. Chiolero A, Cachat F, Burnier M, et al. Prevalence of hypertension in schoolchildren based on repeated measurements and association with overweight. J Hypertens. 2007;25:2209‐2217. [DOI] [PubMed] [Google Scholar]
- 33. Handler J. The importance of accurate blood pressure measurement. Perm J. 2009;13:51‐54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Myers MG. Why automated office blood pressure should now replace the mercury sphygmomanometer. J Clin Hypertens (Greenwich). 2010;12:478‐480. [DOI] [PMC free article] [PubMed] [Google Scholar]