What Can Big Data and a Social Network Analysis Tell Us About Avian Mycobacteriosis and Disease Transmission?
Avian mycobacteriosis has long been considered contagious via fecal-oral transmission.2 However, independent environmental sources may also be important.1,3,4 Both transmission pathways are biologically plausible, but differentiating between them is nearly impossible when case clusters arise in zoos. Social network analysis was used to investigate transmission patterns for avian mycobacteriosis in a dynamic population of >16,000 birds housed at San Diego Zoo Global from 1992–2014. Cases were identified when acid-fast bacilli were observed histopathologically. Network edges were assembled from enclosure histories and temporally aligned to correspond with probable incubation and shedding periods of individually connected birds. Relative risk of mycobacteriosis was estimated for direct and indirect contact with infected birds. Significance was determined by comparing observed estimates to those generated with 1000 random network permutations. Incidence density was 5 cases/1000 bird-years at risk, corroborating the low disease incidence in this population.5 Disease clustered significantly among both directly and indirectly connected birds. Importantly, there was a small, but statistically significant increase in risk of mycobacteriosis from infected birds (versus non-infected birds) that never resided in the same enclosure and were only connected indirectly through another bird. This provides strong evidence that a contagious process is present, because environmental contamination cannot confound the association. However, the continued low incidence in this population suggests that mycobacteriosis has limited transmissibility, which warrants passive management practices. This study demonstrates the value of big data and a statistically powerful social network analysis to disentangle exposure sources and show that contagion likely drives some disease emergence.
The authors thank the Disease Investigations team at San Diego Zoo Global for years of population-level disease surveillance that includes processing and archiving samples as well as maintaining data. The authors thank the bird departments at San Diego Zoo and San Diego Zoo Safari Park for ongoing efforts to record and maintain avian enclosure data. The authors thank Caroline Baratz, Dave Rimlinger, Amy Flanagan, and Michael Mace for assistance with compiling and interpreting enclosure data. Funding was provided in part by the Ellen Browning Scripps Foundation.
1. Falkinham III JO. Surrounded by mycobacteria: nontuberculous mycobacteria in the human environment. J Appl Microbiol. 2009;107:356–367.
2. Fulton RM, Sanchez S. Tuberculosis. In: Swayne DE, Glisson JR, McDougald LR, Nolan LK, Suarez DL, Nair V, eds. Diseases of Poultry. 13th ed. Ames, IA: John Wiley & Sons; 2013:1008–1017.
3. Kauppinen J, Hintikka E, Iivanainen E, Katila M. PCR-based typing of Mycobacterium avium isolates in an epidemic among farmed lesser white-fronted geese (Anser erythropus). Vet Microbiol. 2001;81:41–50.
4. Schrenzel M, Nicolas M, Witte C, Papendick R, Tucker T, Keener L, Sutherland-Smith M, Lamberski N, Orndorff D, Heckard D, Witman P, Mace M, Rimlinger D, Reed S, Rideout BA. Molecular epidemiology of Mycobacterium avium subsp. avium and Mycobacterium intracellulare in captive birds. Vet Microbiol. 2008;126:122–131.
5. Witte CL, Hungerford LL, Papendick R, Stalis IH, Rideout BA. Investigation of characteristics and factors associated with avian mycobacteriosis in zoo birds. J Vet Diagn Invest. 2008;20:186–196.