Fitbit Data for Detecting Infections After Appendicitis Surgery

(i-DETECT Trial)

Recruiting in Palo Alto (17 mi)

+3 other locations

Overseen by Arun Jayaraman, PT, PhD

Age: < 65

Sex: Any

Travel: May Be Covered

Time Reimbursement: Varies

Trial Phase: Academic

Recruiting

Sponsor: Ann & Robert H Lurie Children's Hospital of Chicago

Disqualifiers: Non-ambulatory, Mobility limitations, Language barriers, others

No Placebo Group

Trial Summary

What is the purpose of this trial?

The purpose of this study is to analyze Fitbit data to predict infection after surgery for complicated appendicitis and the effect this prediction has on clinician decision making.

Do I need to stop taking my current medications for this trial?

The trial information does not specify whether you need to stop taking your current medications.

What data supports the effectiveness of the treatment Infection-Prediction Algorithm for detecting infections after appendicitis surgery?

The study on machine learning for predicting postoperative sepsis after appendectomy suggests that algorithms can help identify factors associated with infections, indicating potential for similar methods to predict infections after appendicitis surgery. ^{1 2 3 4 5}

Is it safe to use Fitbit data for detecting infections after appendicitis surgery?

Research shows that using Fitbit devices to monitor recovery after appendicitis surgery in children is safe. These devices help track physical activity and detect complications early without any reported safety concerns. ^{6 7 8 9 10}

How does the Infection-Prediction Algorithm treatment differ from other treatments for detecting infections after appendicitis surgery?

The Infection-Prediction Algorithm uses data from a Fitbit wearable device to monitor physical activity, heart rate, and sleep, allowing for early detection of abnormal recovery patterns after appendicitis surgery. This approach is unique because it provides objective, real-time monitoring at home, unlike traditional methods that rely on subjective assessments by caregivers. ^{2 6 7 8 9}

Research Team

Arun Jayaraman, PT, PhD

Principal Investigator

Shirley Ryan AbilityLab

Fizan Abdullah, MD, PhD

Principal Investigator

Ann & Robert H Lurie Children's Hospital of Chicago

Hassan Ghomrawi, PhD, MPH

Principal Investigator

University of Alabama at Birmingham

Eligibility Criteria

This trial is for pediatric patients who have undergone surgery for complicated appendicitis. Specific eligibility criteria are not provided, but typically participants would be children with recent appendectomies.

Inclusion Criteria

I had surgery to remove my appendix due to severe infection or rupture.

I am between 3 and 18 years old.

Exclusion Criteria

My child has a health condition that affects their recovery.

My child cannot walk or has difficulty moving around.

My child has been advised to limit physical activity for more than 48 hours after surgery.

Trial Timeline

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Postoperative Monitoring

Participants' Fitbit data (PA, HR, Sleep) is collected and analyzed using ML methods to predict infection during the recovery period.

4 weeks

Continuous data collection

Follow-up

Participants are monitored for safety and effectiveness after treatment, including daily diary/survey submissions for symptoms and healthcare utilization.

4 weeks

Daily virtual submissions

Implementation of Algorithm

Clinicians receive daily reports and near real-time alerts based on Fitbit data to assess the impact on clinical decision-making and healthcare utilization.

4 weeks

Treatment Details

Interventions

• Infection-Prediction Algorithm (Behavioural Intervention)

Trial Overview The study is testing an infection-prediction algorithm using data from Fitbits to foresee infections after appendectomy in young patients and how these predictions influence doctors' decisions.

Participant Groups

Group I: Aim 2 - Implementation of Algorithm Experimental Treatment 1 Intervention

2a. Exploratory \& Inductive analysis * one transcript will be coded to generate initial themes, using qualitative analytic software 2b. Time to first contact with the healthcare system \& Healthcare use * Cox regression model will be used to model the time to first contact, adjusted for covariates * All comparisons between the two groups will be tested using a chi-square test. Cost will be modeled as a continuous variable and is expected to be skewed, as is typical of cost data. We will use a general linear model (GLM) to model cost outcomes.

Group II: Aim 1 - Validation Active Control 1 Intervention

1a. Development and Internal validation * analyze Fitbit data (PA, HR, sleep) by applying ML methods to create an infection algorithm indicating onset of infection. 1b. External Validation * Once the ML classifier has been internally validated (using Lurie Children's data only) for its ability to detect the presence or absence of postoperative infection using LOSO cross-validation, where each subject is iteratively held out from the training data and used as a test set. External validation will involve applying this classifier to a newer cohort at LCH and cohorts at Loyola University Hospital and CDH and evaluating its performance.

Who Is Running the Clinical Trial?

Ann & Robert H Lurie Children's Hospital of Chicago

Lead Sponsor

Trials

275

Recruited

5,182,000+

Dr. Michael D. Kelleher

Ann & Robert H Lurie Children's Hospital of Chicago

Chief Medical Officer since 2012

MD from University of Chicago

Dr. Thomas Shanley

Ann & Robert H Lurie Children's Hospital of Chicago

Chief Executive Officer since 2019

MD from University of Chicago Pritzker School of Medicine

Loyola University Chicago

Collaborator

Trials

23

Recruited

13,100+

Northwestern University

Collaborator

Trials

1,674

Recruited

989,000+

Dr. Jeffrey Sherman

Northwestern University

Chief Medical Officer

MD from Northwestern University

Dr. Alicia Löffler

Northwestern University

Chief Executive Officer since 2010

PhD from the University of Massachusetts at Amherst, post-doctoral training at Caltech

Central DuPage Hospital

Collaborator

Trials

15

Recruited

3,900+

University of Chicago

Collaborator

Trials

1,086

Recruited

844,000+

Pete Salzmann

University of Chicago

Chief Executive Officer since 2018

MD from University of Chicago’s Pritzker School of Medicine, MBA from Stanford University’s Graduate School of Business

Anh Nguyen

University of Chicago

Chief Medical Officer

MD from Rutgers New Jersey Medical School, MBA from University of Chicago

Findings from Research

Machine learning algorithms applied to a large dataset of 223,214 appendectomies showed moderate accuracy in predicting postoperative sepsis, with an overall occurrence rate of 0.96%.

Key risk factors identified for postoperative sepsis included preoperative congestive heart failure, transfusion, and acute renal failure, which could help in early intervention and reducing complications.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Application of machine learning to the prediction of postoperative sepsis after appendectomy. Bunn, C., Kulshrestha, S., Boyda, J., et al.[2022]

In a study of 476 patients undergoing appendectomy, a computer-aided diagnosis system was developed to differentiate between acute appendicitis and normal appendixes, but it showed no significant improvement over traditional clinical diagnosis.

The abdominal pain index derived from the analysis had a sensitivity of 82% but a low specificity of 39%, meaning while it could correctly identify many cases of appendicitis, it also misclassified a significant number of patients, potentially leading to unnecessary surgeries.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

A feasibility study of computer aided diagnosis in appendicitis. Van Way, CW., Murphy, JR., Dunn, EL., et al.[2009]

Approximately 250,000 cases of appendicitis occur annually in the U.S., with the highest incidence in males aged 10-19 years, indicating a significant public health concern.

The lifetime risk of developing appendicitis is about 8.6% for males and 6.7% for females, with a notable decline in appendicitis rates of 14.6% from 1970 to 1984, although the reasons for this decrease remain unclear.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

The epidemiology of appendicitis and appendectomy in the United States. Addiss, DG., Shaffer, N., Fowler, BS., et al.[2022]

A new method using machine learning and artificial intelligence can diagnose acute appendicitis with 90% accuracy, significantly reducing unnecessary surgeries for patients without appendicitis.

The method also shows potential to differentiate between complicated and uncomplicated appendicitis, although it has lower accuracy (51%) for this specific distinction, indicating room for improvement in identifying the severity of the condition.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Diagnosis and classification of pediatric acute appendicitis by artificial intelligence methods: An investigator-independent approach. Reismann, J., Romualdi, A., Kiss, N., et al.[2020]

Over a 15-year period, the use of computed tomography (CT) for diagnosing appendicitis increased significantly, leading to more non-operative management and longer time to surgery, but ultimately contributing to improved patient outcomes.

Laparoscopic surgery became the preferred method for treating appendicitis, resulting in shorter hospital stays and a decrease in 30-day mortality rates, which fell to 0.33% overall during the study period.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Temporal Trends in the Investigation, Management and Outcomes of Acute Appendicitis over 15 Years in the North of England: A Retrospective Cohort Study. Bhaskar, J., McLean, RC., Bhaskar, K., et al.[2022]

In a study of 86 pediatric patients aged 3-17 who underwent laparoscopic appendectomy for complicated appendicitis, those who developed postoperative complications showed a slower increase in daily step counts, indicating a deviation from normal recovery patterns.

The use of consumer-grade wearable devices like Fitbits can provide objective data on recovery trajectories, allowing for earlier identification of complications after hospital discharge, which could improve postoperative monitoring in outpatient settings.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Using Consumer Wearable Devices to Profile Postoperative Complications After Pediatric Appendectomy. Pitt, JB., Zeineddin, S., Carter, M., et al.[2023]

A study involving 162 children aged 3-17 who underwent appendectomy showed that a consumer-grade wearable device, Fitbit, can effectively monitor postoperative recovery by detecting abnormal recovery events with high accuracy (83% for complicated cases and 70% for simple cases).

The use of machine learning algorithms trained on Fitbit data demonstrates potential for early detection of complications, suggesting that wearables could enhance postoperative monitoring and reduce unnecessary emergency visits.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Applying machine learning to consumer wearable data for the early detection of complications after pediatric appendectomy. Ghomrawi, HMK., O'Brien, MK., Carter, M., et al.[2023]

The study involved 26 children aged 3-18 who underwent appendectomy, using the FitBit Inspire HR to track their recovery over 21 days, revealing that those with simple appendicitis showed a faster recovery trajectory compared to those with complicated appendicitis.

Wearable technology like FitBit can effectively characterize postoperative recovery in children and differentiate recovery patterns based on the severity of the condition, potentially aiding in the early detection of complications.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Utility of Wearable Sensors to Assess Postoperative Recovery in Pediatric Patients After Appendectomy. De Boer, C., Ghomrawi, H., Many, B., et al.[2021]

Children aged 3-11 years recover more quickly to a plateau in physical activity after laparoscopic appendectomy compared to those aged 12-18, indicating that age significantly influences postoperative recovery trajectories.

Using consumer wearable devices like Fitbits provides objective data on recovery patterns, suggesting that tailored discharge instructions based on age and other factors could enhance patient counseling and recovery expectations.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Using Wearable Devices to Profile Demographic-Specific Recovery After Pediatric Appendectomy. Zeineddin, S., Figueroa, A., Pitt, JB., et al.[2023]

A study involving 66-118 patients who underwent weight loss surgery showed that post-operative weeks were linked to decreased daily steps, less time spent walking, and improved sleep metrics, indicating significant lifestyle changes after surgery.

The research highlights the effectiveness of consumer-grade wearable devices, like Fitbits, in monitoring behavioral and physiological changes during recovery, suggesting they could be valuable tools for tracking patient progress in a convenient and continuous manner.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Continuous Digital Assessment for Weight Loss Surgery Patients. Ramirez, E., Marinsek, N., Bradshaw, B., et al.[2020]

References

1. United States pubmed.ncbi.nlm.nih.gov

Application of machine learning to the prediction of postoperative sepsis after appendectomy. [2022]

2. United States pubmed.ncbi.nlm.nih.gov

A feasibility study of computer aided diagnosis in appendicitis. [2009]

3. United States pubmed.ncbi.nlm.nih.gov

The epidemiology of appendicitis and appendectomy in the United States. [2022]

4. United States pubmed.ncbi.nlm.nih.gov

Diagnosis and classification of pediatric acute appendicitis by artificial intelligence methods: An investigator-independent approach. [2020]

5. United States pubmed.ncbi.nlm.nih.gov

Temporal Trends in the Investigation, Management and Outcomes of Acute Appendicitis over 15 Years in the North of England: A Retrospective Cohort Study. [2022]

6. United States pubmed.ncbi.nlm.nih.gov

Using Consumer Wearable Devices to Profile Postoperative Complications After Pediatric Appendectomy. [2023]

7. England pubmed.ncbi.nlm.nih.gov

Applying machine learning to consumer wearable data for the early detection of complications after pediatric appendectomy. [2023]

8. United States pubmed.ncbi.nlm.nih.gov

Utility of Wearable Sensors to Assess Postoperative Recovery in Pediatric Patients After Appendectomy. [2021]

9. United States pubmed.ncbi.nlm.nih.gov

Using Wearable Devices to Profile Demographic-Specific Recovery After Pediatric Appendectomy. [2023]

10. Switzerland pubmed.ncbi.nlm.nih.gov

Continuous Digital Assessment for Weight Loss Surgery Patients. [2020]