Teen White-Coat Hypertension Analysis

ISEF Category: Biomedical and Health Sciences

Subcategory: Other  ·  Difficulty: Intermediate  ·  Setup: Home Setup  ·  Time: 1 to 2 Months

The Hook

A blood pressure reading can change just because the room changes. That makes teens a good test case for white-coat hypertension, when stress pushes numbers up, and masked hypertension, when a normal-looking reading hides a higher true level. If you pair home readings with classroom readings, you can ask whether the setting itself changes the number. A Bayesian model helps you separate the setting effect from each person's usual baseline.

What Is It?

White-coat hypertension means a person's blood pressure rises in a formal or stressful setting. Masked hypertension means the opposite, the number looks fine in the office but runs higher somewhere else. Think of blood pressure like the volume on a speaker. The true signal is there, but the setting adds static.

Your project asks whether the classroom acts like a mini white coat for teenagers. A Bayesian hierarchical model is a way to keep both the group pattern and each person's pattern in view. It gives you a range for the setting effect instead of one shaky average.

Why This Is a Good Topic

This works well for a science fair because you can collect real measurements, compare two settings, and analyze the results with math you can explain. The topic connects to heart health, stress, and how doctors should interpret a single blood pressure reading. You can learn sampling, controls, uncertainty, and hierarchical modeling without needing a university lab.

Research Questions

• How does measurement setting change systolic blood pressure in teenagers?
• How does measurement setting change diastolic blood pressure in teenagers?
• Does the home-to-classroom gap change with self-reported anxiety before measurement?
• To what extent do repeated readings reduce the estimated setting effect in a Bayesian hierarchical model?
• Which predicts the next reading better, a pooled average model or a hierarchical model?
• What is the effect of time of day on the size of the home-to-classroom blood pressure gap?

Basic Materials

• Validated automated upper-arm blood pressure cuff with the right cuff size.
• Parent consent forms and student assent forms.
• Spreadsheet or notebook for linked home and classroom readings.
• Quiet chair and table for seated rest before each measurement.
• Timer app or stopwatch.
• Simple anxiety rating scale, such as a one-to-five checklist.

Advanced Materials

• Ambulatory blood pressure monitor.
• Multiple cuff sizes for fit checking.
• Heart rate sensor or pulse oximeter.
• Tablet or phone for secure electronic survey entry.
• REDCap or Qualtrics for data capture.

Software & Tools

• R: Fits hierarchical Bayesian models and draws uncertainty plots.
• RStudio: Helps you clean data, run models, and inspect diagnostics.
• Python: Lets you manage data with pandas and fit models in PyMC.
• JASP: Gives you a free interface for quick statistical checks.
• Google Sheets: Tracks paired readings and flags missing data.

Experiment Steps

1. Define your paired-measurement plan so each teenager has a home reading and a classroom reading under matched rest conditions.
2. Choose the variables you will record besides blood pressure, such as anxiety, time of day, and cuff size, so you can test confounders.
3. Build a data structure that keeps each person's repeated readings linked to the right setting and session.
4. Specify a Bayesian hierarchical model with person-level baselines and a setting-effect term.
5. Decide how you will compare the hierarchical model with a simple average model and check whether the setting effect stays stable.
6. Plan how you will report uncertainty, outliers, and missing data before you collect anything.

Common Pitfalls

• Using a cuff that fits some students but not others, which can mimic a setting effect.
• Taking classroom readings right after class or gym, which mixes stress, activity, and posture with the context effect.
• Letting one reading stand for a person's blood pressure, which makes random noise look like white-coat hypertension.
• Ignoring the order of home and classroom tests, which can make the second reading look lower because the student has already settled.
• Pooling everyone into one average, which hides the student-level variation that the hierarchical model is meant to separate.

What Makes This Competitive

A competitive version does more than compare two averages. It separates person-to-person variation, setting effects, and noise, then asks whether the classroom effect survives after those layers are modeled. If you compare a simple mean model with a hierarchical model and report uncertainty clearly, the project looks much more mature. Add a careful check for order effects or anxiety, and you move closer to the kind of analysis judges remember.

Project Variations

• Compare morning versus afternoon readings to see whether time of day changes the context gap.
• Replace classroom measurements with school nurse office readings to test whether authority alone drives the effect.
• Add a short relaxation condition before one session to see whether reducing stress shrinks the white-coat gap.

Learn More

• NIH/NHLBI Blood Pressure: Plain-language background on blood pressure numbers, screening, and hypertension, found on the NIH and NHLBI websites.
• CDC High Blood Pressure: Quick overview of blood pressure risk and screening, on the CDC website.
• PubMed: Search review articles on white-coat hypertension, masked hypertension, and adolescents.
• Hypertension: Peer-reviewed reviews and pediatric guidelines, searchable through the journal site or PubMed.
• MIT OpenCourseWare, Introduction to Probability and Statistics: Free help for hierarchical models and uncertainty, on the MIT OpenCourseWare site.
• American Heart Association guidelines and measurement resources: Accessible summaries of blood-pressure measurement issues, on the AHA website.