Sarang Deo1, Pankaj Jindal2,3, Sirisha Papineni4,5. 1. Indian School of Business, AC 3, L1, #3113, ISB Campus, Gachibowli, Hyderabad, 500032, India. sarang_deo@isb.edu. 2. Indian School of Business, AC 3, L1, #3113, ISB Campus, Gachibowli, Hyderabad, 500032, India. 3. UCLA Anderson School of Management, Los Angeles, United States. 4. Harvard Medical School, Boston, United States. 5. World Health Partners, New Delhi, India.
Abstract
BACKGROUND: Xpert MTB/RIF (Xpert) has been recommended by WHO as the initial diagnostic test for TB and rifampicin-resistance detection. Existing evidence regarding its uptake is limited to public health systems and corresponding resource and infrastructure challenges. It cannot be readily extended to private providers, who treat more than half of India's TB cases and demonstrate complex diagnostic behavior. METHODS: We used routine program data collected from November 2014 to April 2017 from large-scale private sector engagement pilots in Mumbai and Patna. It included diagnostic vouchers issued to approximately 150,000 patients by about 1400 providers, aggregated to 18,890 provider-month observations. We constructed three metrics to capture provider behavior with regards to adoption of Xpert and studied their longitudinal variation: (i) Uptake (ordering of test), (ii) Utilization for TB diagnosis, and (iii) Non-adherence to negative results. We estimated multivariate linear regression models to assess heterogeneity in provider behavior based on providers' prior experience and Xpert testing volumes. RESULTS: Uptake of Xpert increased considerably in both Mumbai (from 36 to 60.4%) and Patna (from 12.2 to 45.1%). However, utilization of Xpert for TB diagnosis and non-adherence to negative Xpert results did not show systematic trends over time. In regression models, cumulative number of Xpert tests ordered was significantly associated with Xpert uptake in Patna and utilization for diagnosis in Mumbai (p-value< 0.01). Uptake of Xpert and its utilization for diagnosis was predicted to be higher in high-volume providers compared to low-volume providers and this gap was predicted to widen over time. CONCLUSIONS: Private sector engagement led to substantial increase in uptake of Xpert, especially among high-volume providers, but did not show strong evidence of Xpert results being integrated with TB diagnosis. Increasing availability and affordability of a technically superior diagnostic tool may not be sufficient to fundamentally change diagnosis and treatment of TB in the private sector. Behavioral interventions, specifically aimed at, integrating Xpert results into clinical decision making of private providers may be required to impact patient-level outcomes.
BACKGROUND: Xpert MTB/RIF (Xpert) has been recommended by WHO as the initial diagnostic test for TB and rifampicin-resistance detection. Existing evidence regarding its uptake is limited to public health systems and corresponding resource and infrastructure challenges. It cannot be readily extended to private providers, who treat more than half of India's TB cases and demonstrate complex diagnostic behavior. METHODS: We used routine program data collected from November 2014 to April 2017 from large-scale private sector engagement pilots in Mumbai and Patna. It included diagnostic vouchers issued to approximately 150,000 patients by about 1400 providers, aggregated to 18,890 provider-month observations. We constructed three metrics to capture provider behavior with regards to adoption of Xpert and studied their longitudinal variation: (i) Uptake (ordering of test), (ii) Utilization for TB diagnosis, and (iii) Non-adherence to negative results. We estimated multivariate linear regression models to assess heterogeneity in provider behavior based on providers' prior experience and Xpert testing volumes. RESULTS: Uptake of Xpert increased considerably in both Mumbai (from 36 to 60.4%) and Patna (from 12.2 to 45.1%). However, utilization of Xpert for TB diagnosis and non-adherence to negative Xpert results did not show systematic trends over time. In regression models, cumulative number of Xpert tests ordered was significantly associated with Xpert uptake in Patna and utilization for diagnosis in Mumbai (p-value< 0.01). Uptake of Xpert and its utilization for diagnosis was predicted to be higher in high-volume providers compared to low-volume providers and this gap was predicted to widen over time. CONCLUSIONS: Private sector engagement led to substantial increase in uptake of Xpert, especially among high-volume providers, but did not show strong evidence of Xpert results being integrated with TB diagnosis. Increasing availability and affordability of a technically superior diagnostic tool may not be sufficient to fundamentally change diagnosis and treatment of TB in the private sector. Behavioral interventions, specifically aimed at, integrating Xpert results into clinical decision making of private providers may be required to impact patient-level outcomes.
Authors: T Agizew; R Boyd; A F Auld; L Payton; S L Pals; P Lekone; V Chihota; A Finlay Journal: Int J Tuberc Lung Dis Date: 2019-01-01 Impact factor: 2.373
Authors: Karen R Steingart; Hojoon Sohn; Ian Schiller; Lorie A Kloda; Catharina C Boehme; Madhukar Pai; Nandini Dendukuri Journal: Cochrane Database Syst Rev Date: 2013-01-31
Authors: John D Birkmeyer; Therese A Stukel; Andrea E Siewers; Philip P Goodney; David E Wennberg; F Lee Lucas Journal: N Engl J Med Date: 2003-11-27 Impact factor: 91.245
Authors: Andrew F Auld; Katherine L Fielding; Ankur Gupta-Wright; Stephen D Lawn Journal: Trans R Soc Trop Med Hyg Date: 2016-09-16 Impact factor: 2.184
Authors: David W Dowdy; J Lucian Davis; Saskia den Boon; Nicholas D Walter; Achilles Katamba; Adithya Cattamanchi Journal: PLoS One Date: 2013-08-12 Impact factor: 3.240
Authors: Sabine Hermans; Judy Caldwell; Richard Kaplan; Frank Cobelens; Robin Wood Journal: Bull World Health Organ Date: 2017-04-28 Impact factor: 9.408