Physician Versus Computer Coding of Verbal Autopsies

NCT02810366 | Unknown

• 1 other identifier: PCVA vs CCVA
• Study Type: interventional
• Target: 12,500
• Locations: 1 country
• Sites: 3

Brief Summary

The objective of this study is to compare the performance of computer-coded verbal autopsies (CCVA) to physician-coded verbal autopsies (PCVA) at the population level. In order to do so a randomised control trial is being conducted in five districts of India. In each district, 50% of deaths are randomly selected for PCVA and the rest for CCVA. The cause of death distribution for both groups are then compared within each district. If the performance of PCVA and CCVA are comparable, the attained distributions should be similar.

Conditions

Fatal Outcome

Keywords

Verbal Autopsy; Physician-coded verbal autopsy; Computer-coded verbal autopsy; InterVA; Naive Bayes Classifier (NBC); Tariff; InSilicoVA; Randomised Control Trial; SmartVA; King-Lu

Outcome Measures

Primary Outcomes (1)

• Equivalence (CSMF Accuracy) of cause of death distribution between physician versus computer coded verbal autopsies

  Use CSMF Accuracy to measure the equivalence of the cause of death distribution between the physician and computer coded VA arms of this study, in order to assess whether the performance of physician vs. computer coding of VAs are comparable at the population level

  1 year

Secondary Outcomes (1)

• Equivalence of cause of death assignment at the population (CSMF Accuracy) and individual (sensitivity) levels of physician versus lay surveyor collected verbal autopsies

  1 year

Other Outcomes (1)

• Number of households that responded "poorly" to the short verbal autopsy questionnaire with narrative versus a long questionnaire without a narrative, as reported by the surveyor

  1 year

Study Arms (2)

Physician Coded Verbal Autopsy

ACTIVE COMPARATOR

Of the approximately 12,500 VAs collected, 50% in each district will be randomly collected using the "electronic Verbal Autopsy" (eVA) instrument. In addition to "general information" about the deceased (e.g. name, sex, age, etc.), this VA instrument contains a short checklist questionnaire to capture from the respondent the signs and symptoms noted during the final illness, followed by a free-text narrative. Cause of death for these VAs will be assigned by trained physicians using the MDS physician coding system; this includes dual, independent coding of VA records, disagreements resolved by reconciliation, and remaining cases by adjudication by a third physician. The assignment of cause of deaths will be in line with the international classification of disease version 10 (ICD-10).

Other: Physician versus Computer Coded Verbal Autopsy

Computer Coded Verbal Autopsy

EXPERIMENTAL

Of the approximately 12,500 VAs collected, 50% in each district will be randomly collected using the "Extended Symptom List" (ESL) VA instrument. In addition to "general information" about the deceased (e.g. name, sex, age, etc.), this VA instrument contains a long checklist questionnaire to capture from the respondent the signs and symptoms noted during the final illness. This VA instrument does not contain a free-text narrative. The cause of death for these VAs will be independently assigned by five leading computer-coding VA algorithms. The assignment of cause of deaths will be in line with 17 broad cause of death categories.

Other: Physician versus Computer Coded Verbal Autopsy

Interventions

Physician versus Computer Coded Verbal Autopsy OTHER

Comparing the performance of computer coded verbal autopsies (CCVA) to physician coded verbal autopsies (PCVA) at the population level.

Also known as: PCVA vs. CCVA

Computer Coded Verbal Autopsy; Physician Coded Verbal Autopsy

Eligibility Criteria

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

You may qualify if:

• All deaths that have occurred in the last five years

You may not qualify if:

• None

Sponsors & Collaborators

• Centre for Global Health Research, Toronto: lead
• International Institute for Population Sciences: collaborator
• Tata Memorial Hospital: collaborator
• HM Patel Center for Medical Care and Education: collaborator

Study Sites (3)

HM Patel Center for Medical Care and Education

Karamsad, Gujarat, 388325, India

RECRUITING

Tata Memorial Centre

Mumbai, Maharashtra, 400 012, India

RECRUITING

International Institute of Population Sciences

Mumbai, Maharashtra, 400 088, India

COMPLETED

Related Publications (4)

• Leitao J, Desai N, Aleksandrowicz L, Byass P, Miasnikof P, Tollman S, Alam D, Lu Y, Rathi SK, Singh A, Suraweera W, Ram F, Jha P. Comparison of physician-certified verbal autopsy with computer-coded verbal autopsy for cause of death assignment in hospitalized patients in low- and middle-income countries: systematic review. BMC Med. 2014 Feb 4;12:22. doi: 10.1186/1741-7015-12-22.

  PMID: 24495312 BACKGROUND

• Desai N, Aleksandrowicz L, Miasnikof P, Lu Y, Leitao J, Byass P, Tollman S, Mee P, Alam D, Rathi SK, Singh A, Kumar R, Ram F, Jha P. Performance of four computer-coded verbal autopsy methods for cause of death assignment compared with physician coding on 24,000 deaths in low- and middle-income countries. BMC Med. 2014 Feb 4;12:20. doi: 10.1186/1741-7015-12-20.

  PMID: 24495855 BACKGROUND

• Miasnikof P, Giannakeas V, Gomes M, Aleksandrowicz L, Shestopaloff AY, Alam D, Tollman S, Samarikhalaj A, Jha P. Naive Bayes classifiers for verbal autopsies: comparison to physician-based classification for 21,000 child and adult deaths. BMC Med. 2015 Nov 25;13:286. doi: 10.1186/s12916-015-0521-2.

  PMID: 26607695 BACKGROUND

• Jha P, Kumar D, Dikshit R, Budukh A, Begum R, Sati P, Kolpak P, Wen R, Raithatha SJ, Shah U, Li ZR, Aleksandrowicz L, Shah P, Piyasena K, McCormick TH, Gelband H, Clark SJ. Automated versus physician assignment of cause of death for verbal autopsies: randomized trial of 9374 deaths in 117 villages in India. BMC Med. 2019 Jun 27;17(1):116. doi: 10.1186/s12916-019-1353-2.

  PMID: 31242925 DERIVED

Study Officials

• Abhishek Singh, PhD

  Associate Professor, International Institute of Population Sciences

  PRINCIPAL INVESTIGATOR

• Atul Budukh, MD

  Assistant Professor Epidemiology, Tata Memorial Centre

  PRINCIPAL INVESTIGATOR

• Dinesh Kumar, MD

  Associate Professor, HM Patel Center for Medical Care and Education

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Rehana Begum, MD

CONTACT

+919945908671; BegumR@smh.ca

Prabhat Jha, MD, PhD

CONTACT

+14168646042; prabhat.jha@utoronto.ca

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: PARTICIPANT
• Purpose: BASIC SCIENCE
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: April 11, 2016
• First Posted: June 23, 2016
• Study Start: September 1, 2015
• Primary Completion: September 1, 2016
• Study Completion: December 1, 2016
• Last Updated: June 23, 2016

Record last verified: 2016-06

Data Sharing

• IPD Sharing: Will share

Locations

IN (3)