Key Questions and Analytic Framework

Using the methods of the USPSTF, which are fully described in Appendix A, and with the input of members of the USPSTF, we developed an analytic framework (Figure) and key questions to guide our literature search and review.

This Figure depicts the analytic framework, which outlines the evidence areas covered in the review, including the populations, screenings, interventions, and outcomes. The populations include children and adolescents who are asymptomatic for hypertension in primary care–generalizable settings. An arrow represents screening for hypertension and diagnostic accuracy of the screening test (Key Question 2), and a subsequent arrow represents adverse effects from screening (Key Question 4). Those with high blood pressure are then diagnosed as having either primary or secondary hypertension. The assessment and treatment of secondary hypertension is beyond the scope of this review. For those with primary hypertension, interventions include drug, lifestyle, and combination drug and lifestyle interventions, which are depicted in the middle of the framework. A subsequent arrow assesses the effectiveness of interventions on blood pressure and other intermediate markers in childhood and during adolescence (Key Question 5), and a dotted line represents the effectiveness of these same outcomes in adulthood (Key Question 6). Intermediate outcomes include left ventricular hypertrophy, urinary albumin excretion (microalbuminuria), intima media thickness (measured at carotid and/or femoral arteries), and retinal vascular changes. Key Question 7 assesses the effectiveness of interventions on final health outcomes in adulthood, represented at the end of the framework, and an arrow for Key Question 8 examines adverse effects from interventions. In addition, an overarching arrow directly assesses the impact of screening on final health outcomes in adulthood (Key Question 1), and a dotted line examines the association between hypertension in childhood with hypertension and other intermediate outcomes in adulthood (Key Question 3).

Figure

Analytic Framework. Abbreviation: KQ = key question. * The assessment and treatment of secondary hypertension is beyond the scope of this review.

Key Questions

  1. Is screening for hypertension in children/adolescents effective in delaying the onset of or reducing adverse health outcomes related to hypertension?
  2. What is the diagnostic accuracy of screening tests for elevated blood pressure in children/adolescents?
  3. What is the association between hypertension in children/adolescents and hypertension and other intermediate outcomes in adults?
  4. What are the adverse effects of screening for hypertension in children/adolescents, including labeling and anxiety?
  5. What is the effectiveness of drug, nondrug, and combination interventions for treating primary hypertension in children/adolescents?
  6. What is the effectiveness of drug, nondrug, and combination interventions initiated for the treatment of primary hypertension in children/adolescents for reducing blood pressure and other intermediate outcomes in adults?
  7. What is the effectiveness of drug, nondrug, and combination interventions initiated for the treatment of primary hypertension in children/adolescents for reducing adverse health outcomes in adults related to primary hypertension?
  8. What are the adverse effects of drug, nondrug, and combination interventions for treating primary hypertension in children/adolescents?

Three contextual questions were also requested by the USPSTF to help inform the report. Contextual questions were not reviewed using systematic review methodology.

Contextual Questions

  1. What are the main risk factors for primary hypertension in children/adolescents?
  2. What is the prevalence of secondary hypertension in asymptomatic children/adolescents in primary care settings?
  3. What are the optimal ages at which to initiate screening and optimal time intervals at which to repeat screening children/adolescents for hypertension?

Search Strategies

We searched the Cochrane Central Register of Controlled Trials and the Cochrane Database of Systematic Reviews (through July 2012) and MEDLINE (1946–July 9, 2012) for relevant studies and systematic reviews. Complete search strategies are described in Appendix A1. We also manually reviewed reference lists of included studies.

Study Selection

We selected studies on the basis of inclusion and exclusion criteria developed for each key question (see Appendix A2 for details). All citations identified through searches and other sources were imported into EndNote v.X3 and were independently reviewed by two investigators for inclusion/exclusion. Discrepancies regarding inclusion/exclusion of full-text papers were resolved through consensus. We included studies of screening for hypertension in asymptomatic children and adolescents and studies of benefits and harms of interventions for childhood hypertension. For studies of diagnostic accuracy, we required that studies include a reference standard comparison and provide adequate data to reproduce 2 × 2 tables, if not reported. Longitudinal cohort studies were included to address the tracking of hypertension from childhood to adulthood. We excluded studies of interventions for treatment of obesity and lipid disorders in children, as these populations are covered by other USPSTF publications.27, 28 We also excluded studies focusing on secondary hypertension, both the treatment of elevated blood pressure in these patients and the treatment of the underlying conditions. In addition, we excluded studies with total populations of less than 30 participants. Appendix A3 shows the results of our literature search and selection process. Appendix A4 shows studies that were excluded at the full-text level with reasons for exclusion.

Data Abstraction and Quality Rating

One investigator abstracted details about the patient population, study designs, testing methods, analysis, followup, and results, and a second investigator checked data abstraction for accuracy. For studies of interventions, we also abstracted data on dose in drug studies. By using predefined criteria developed by the USPSTF29 and others for additional criteria for diagnostic accuracy studies,30 two investigators rated the quality of studies as good, fair, or poor and resolved discrepancies by consensus (Appendix A5).

Data Synthesis

We assessed the overall strength of the body of evidence for each key question as good, fair, or poor using methods developed by the USPSTF, based on the number, quality, and size of studies, consistency of results among studies, and directness of evidence.29 The limited number of studies and differences in study design and methods precluded us from conducting meta-analyses. Results are presented in narrative format and, where possible, include ranges and 95 percent confidence intervals (CIs). For studies of diagnostic accuracy, we constructed 2 × 2 tables and calculated sensitivity, specificity, predictive values, and 95 percent CIs, if not already reported. Pooling of results from diagnostic accuracy studies was also not possible due to heterogeneity across studies.

External Review

This draft report was reviewed by content experts, USPSTF members, Agency for Healthcare Research and Quality (AHRQ) Project Officers, and AHRQ's collaborative partners (Appendix A6).