Impact Evaluation of Use of MATCH AI Predictive Modelling for Identification of Hotspots for TB Active Case Finding

NCT06017843 | Recruiting DMC

• 1 other identifier: CGPH-TB2023-24
• Study Type: interventional
• Target: 180,000
• Locations: 1 country
• Sites: 1

Brief Summary

The aim of this pragmatic, stepped wedge cluster-randomized trial is to measure the comparative yield (number of incident TB cases diagnosed during active case-finding camps) using a site selection approach based on predictions generated via an artificial intelligence software called MATCH-AI (intervention group) versus the conventional approach of camp site selection using field-staff knowledge and experience (control group). The trial will help inform whether a targeted approach towards screening for TB using artificial-intelligence can improve yields of TB cases detected through community-based active case-finding.

Conditions

Tuberculosis

Keywords

Active case finding; Artificial Intelligence; Hotspots

Outcome Measures

Primary Outcomes (1)

• Camp positivity yield

  Counts of bacteriologically confirmed TB (B+) cases diagnosed in each camp

  12 months

Secondary Outcomes (3)

• Camp positivity rate

  12 months

• Camp All-Forms yield

  12 months

• Camp All-Forms TB rate

  12 months

Study Arms (2)

Intervention

EXPERIMENTAL

Camps site selection for active case finding for TB using MATCH-AI

Other: Camps site selection for active case finding for TB using MATCH-AI

Control

NO INTERVENTION

Camps site selection for active case finding for TB using existing approaches.

Interventions

Camps site selection for active case finding for TB using MATCH-AI OTHER

The primary intervention in this study is the roll-out of MATCH-AI, an artificial intelligence software that models sub-district TB prevalence, to guide site selection of ACF camps. The MATCH-AI tool uses a Bayesian modelling approach to predict TB prevalence to a resolution of 10,000 population that are mapped as polygons. The model integrates data from a range of sources including historical TB facility notification data, previous ACF data as well as contextual factors such as demographics, income, population density, health indicators such as vaccination coverage and climate related variables to predict localized TB prevalence. In the intervention arm, camps will be conducted primarily in locations guided by MATCH-AI.

Intervention

Eligibility Criteria

• Age: 15 Years+
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)

You may qualify if:

• All individuals \>15 years of age presenting to camp sites
• Individuals with previous history of TB disease

You may not qualify if:

• Children and adolescents \<15 years of age
• Pregnant women

Sponsors & Collaborators

• Centre for Global Public Health Pakistan: lead
• Mercy Corps Pakistan: collaborator

Study Sites (1)

Mercy Corps Pakistan

Islamabad, Pakistan

RECRUITING

Related Publications (1)

• Zaidi SMA, Mahfooz A, Latif A, Nawaz N, Fatima R, Rehman FU, Reza TE, Emmanuel F. Geographical targeting of active case finding for tuberculosis in Pakistan using hotspots identified by artificial intelligence software (SPOT-TB): study protocol for a pragmatic stepped wedge cluster randomised control trial. BMJ Open Respir Res. 2024 Jul 11;11(1):e002079. doi: 10.1136/bmjresp-2023-002079.

  PMID: 38991950 DERIVED

MeSH Terms

Conditions

Tuberculosis

Condition Hierarchy (Ancestors)

Mycobacterium Infections; Actinomycetales Infections; Gram-Positive Bacterial Infections; Bacterial Infections; Bacterial Infections and Mycoses; Infections

Study Officials

• Faran Emmanuel

  Centre for Global Public Health Pakistan

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Amna Mahfooz, MS(PH)

CONTACT

+923438101441; amna.mahfooz@cgph.org.pk

Faheem Baig

CONTACT

+923345379004; faheem@cgph.org.pk

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: PARTICIPANT
• Masking Details: Individuals visiting camps will be masked to the intervention status.
• Purpose: HEALTH SERVICES RESEARCH
• Intervention Model: CROSSOVER
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Model Details: The MATCH-AI evaluation is designed as a stepped-wedge cluster randomized trial. In this study, clusters (mobile X-ray van teams) will successively switch over in groups of 3 to the intervention in a randomly assigned order until all clusters are eventually exposed to the intervention. In the intervention arm, TB active case finding camps will be conducted primarily in locations guided by MATCH-AI a modelling software that predicts TB hotspots. In the control sites, field-teams will continue to utilize existing approaches such as local knowledge, historical data etc towards camp site-selection.

Study Record Dates

• First Submitted: August 21, 2023
• First Posted: August 30, 2023
• Study Start: September 1, 2023
• Primary Completion: October 31, 2024
• Study Completion: June 30, 2025
• Last Updated: July 25, 2024

Record last verified: 2024-07

Data Sharing

• IPD Sharing: Will not share

Locations

PA (1)