Validation of a Software Based Clinical Decision Support ... · Access to expert, evidence based...
Transcript of Validation of a Software Based Clinical Decision Support ... · Access to expert, evidence based...
Access to expert, evidence based clinical decision making is crucial in
maximizing the outcome of women with breast cancer, but it is a scarce
resource, especially in developing countries. The Navya Expert System is a
patented, software based clinical decision support system that exhaustively
searches and assimilates relevant medical literature and guidelines to make
specific therapeutic recommendations for individual patients based on their
clinical data. This study is a retrospective validation of Navya Expert System’s
output against tumor board decisions of a multidisciplinary group of expert
breast cancer clinicians working in a tertiary care oncology center in India
With structured clinical data as the input, the patented Navya Expert System
queries its structured databases of evidence and guidelines indexed per Navya’s
ontology of decision criteria, and returns as output treatment decisions most
applicable to a given unique patient. The objective of the study was to test the
results of the automated system against tumor board decisions made by expert
treating oncologists. The automated system is inherently scalable and aims to
eventually enable patients and physicians worldwide to increase their access to
evidence based and expert treatment decisions.
Background
P4-16-01
A total of 76 patients involving 224 major and 224 minor therapeutic decisions were
included in the study.
Navya Expert System's output was concordant with the tumor board or expert review
• 224/224 major decisions (100%, 95% CI 99.6%-100%)
• 221/224 minor decisions (98.6%, 95% CI 97.1%-100%)
Navya Expert System's output was concordant with the tumor board alone
• 210/224 major decisions (93.75%, 95% CI 90.6%-96.9%)
• 160/224 minor decisions (71.4%, 95% CI 65.5%-77.3%)
Most common reasons for discordance were
• Non-prescription of HER2 targeted therapy by the tumor board due to financial
constraints
• Non-use of adjuvant radiotherapy for 1-3 node positive patients.(as tumour board
decisions were based on guidelines from 2012)
Of the 64/224 Navya Expert System decisions discordant with the tumor board, only
three Navya decisions were finally deemed discordant with expert clinical
practice
Future Trials
MAJOR DECISIONS MINOR DECISIONS
NACT or Upfront surgery BCT or MRM
Adjuvant chemotherapy or not Anthracyclines or Taxanes or combination
Adjuvant Radiation therapy or not Nodal RT/IMRT or no radiation therapy
Access to Evidence Based Expert Decisions = Maximize Outcomes
Objective Methodology Results
• Navya decisions were compared to tumor board decisions and adjudicated by
experts .
A retrospective trial was conducted to compare the Navya Expert System decisions to
tumor board decisions made at a tertiary care cancer centre (Tata Memorial Centre,
Mumbai) in India. Cases of women with non-metastatic breast cancer treated at Tata
Memorial Center, for whom treatment decisions involved therapeutic complexities
such as two or more treatment options , were included in the trial.
Clinical and pathological data were abstracted from the case files as per Navya’s
decision criteria, eg: Age, ER/PR /Her2neu status, Tumor Size, Number of positive
lymph nodes, Co-Morbidity, etc
Patient preferences, and financial and logistic constraints were not factored into the
decision criteria.
Data was processed through Navya’s automated, patented Evidence Engine and
Guidelines Engine.
Navya treatment decisions were classified as follows:
© Tata Memorial Hospital, Mumbai, India & Navya Network , Cambridge, MA This presentation is the intellectual property of the author/presenter. Contact at [email protected], [email protected] for permission to reprint and/or distribute
Results
Conclusion
A conjoint analysis based decision aid to assess risk/benefit tradeoffs for patients
will be prospectively validated to include as a resource in decision making. This trial
will test the hypothesis that factoring in patient’s preferences in evidence based
expert decisions will result in further improved outcomes.
• Commonly available clinical data + Navya Expert System
= Expert Treatment Decisions
• Navya decisions are concordant with expert clinical practice i.e. concordant with
the tumor board and experts at a tertiary care cancer center
• Navya Expert System can increase global access to evidence based expert
treatment decisions
San Antonio Breast Cancer Symposium - December 9-13, 2014
Validation of a Software Based Clinical Decision Support System for Breast Cancer Treatment in a Tertiary Care Cancer Center in India
NS Nair1, S Gupta1, N Ramarajan2, G Srivastava2, V Parmar1, A Munshi1, S Vanmali1, V Vanmali1, R Hawaldar1, RA Badwe1
1. Tata Memorial Centre, Mumbai, India, 2. Navya Network, Cambridge, MA
DECISION (N)
TUMOR BOARD + EXPERT
ADJUDICATOR Major Conc.(CI)
TUMOR BOARD + EXPERT
ADJUDICATOR Minor Conc.(CI)
TUMOR BOARD Major Conc.
(CI)
TUMOR BOARD Minor Conc.
(CI)
Total 224 100% (0.41) 98.66% (1.5) 93.75% (3.16) 71.42% (5.91)
NACT/ Sx (108) 100% (0.59) 99.07% (1.80) 94.44% (4.32) 79.62% (7.59)
NACT Her2neu -ve (14)
100% (1.65) 92.8% (13.49) 100% (1.65) 64.28% (25.09)
NACT Her2neu+ (12)
100% (1.78) 100% (1.78) 100% (1.78) 8.33% (15.63)
Surgery (82) 100% (0.68) 100% (2.52) 92.68% (0.68) 92.68% (0.68)
Adjuvant chemotherapy (38)
100% (1) 94.73% (7.09) 100% (0.31) 31.57% (14.77)
Adjuvant chemotherapy Her2neu –ve (18)
100% (1.4) 100% (1.43) 100% (1.46) 61.11% (22.52)
Adjuvant chemotherapy Her2neu + (20)
100% (1.38) 90% (13.14) 100% (1.38) 5% (9.55)
Adjuvant radiotherapy (66)
100% (0.76) 100% (0.76) 87.87% (7.87) 77.27% (10.11)
RT N0 (3) 100% (3.57) 100% (3.57) 100% 93.57) 100% (3.57)
RT 1-3 LN + (41) 100% (0.96) 100% (0.96) 80.48% (12.13) 65.85% (14.51)
RT 4+ LN (22) 100% (1.32) 100% (1.32) 100% (1.32) 95.45% (8.70)
Adjuvant hormone therapy (12)
100% (1.78) 100% (1.78) 100 % (1.78) 91.66% (15.63)