Infection Control & Hospital Epidemiology
317
Fig. 1. Transmission route ranking for outbreak no. 4: Pseudomonas aeruginosa from a contaminated broncho- scope. Panel A: Results for bronchoscopy, showing timing of cases, the proportion of cases who were found in the EHR to have been exposed to bronchoscopy before illness onset, the percent of the control population that was found in the EHR to have been exposed to bronchoscopy, the score, the ranking relative to other exposures, the P value and odds ratio. Panel B: Graphical depiction of relative ranking of bronchoscopy and 2 other ranking exposures (top figure) and the cumulative number of cases (bottom figure), both over time.
total, for the 2011–2016 outbreak requests, potentially 40 or 34 infections (78% or 66% of possible preventable infections, respec- tively) could have been prevented based on intervention imple- mentation at 7 or 14 days.
Discussion
In this study, data mining of theEHRcorrectly identified transmis- sion routes by the eighth patient of 1 outbreak and the second patient in all other outbreaks studied. If run in conjunction with routine molecular typing, up to 40 infections (78% of possible pre- ventable infections) could have been prevented assuming that proper intervention had occurred. To our knowledge, this is the first reported study that combines molecular typing results and automated data mining of the EHR in a hospital outbreak setting to identify routes of bacterial transmission. Our results provide proof of concept that automated data mining can correctly identify routes of exposure in hospital outbreak investigations. Automated data mining has several potential advantages over
traditional approaches to hospital outbreak investigations. First, the EHR can be rapidly scanned for common exposures among patients with complex hospitalizations. Second, automated data mining allows rapid assessment of the strength of association of suspected exposures. In this study, we incorporated a case-control study design to identify outbreak transmission routes, which is similar to the approach that is used in traditional outbreak inves- tigations. We are currently refining this approach to allow the
infection preventionist to easily select and explore the most appro- priate control population within the hospital. For example, to iden- tify the route of transmission during an outbreak that occurs on a single nursing unit, the most appropriate control population may be nonoutbreak patients on the same unit. Both approaches have the potential to substantially decrease the number of hours required for outbreak investigations and to allowinfection preven- tion personnel with limited outbreak investigation expertise to conduct relatively sophisticated investigations. Our study and approach have several limitations. First, only
outbreaks that had been detected by traditional epidemiologic approaches were included. This limitation could have resulted in missing other patients with genetically related isolates who should have been included as cases, thus leading to both an underestimate of the magnitude of the outbreak and having the patients incor- rectly included in our control population. Despite this limitation, data mining still identified the correct transmission routes. We anticipate that we can largely overcome this limitation in the future by implementing WGS surveillance of key hospital pathogens. Second, the intervention delay of 7 or 14 days was based on hypo- thetical timelines that considered the time required to perform WGS, analyze data, and enact interventions (eg, removing a device from use, targeted environmental cleaning, and/or staff education). Regardless, a conservative delay of 14 days for effective interven- tions still demonstrated 34 potential infections prevented across a relatively small number of outbreaks. Third, we did not include
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118 |
Page 119 |
Page 120 |
Page 121 |
Page 122 |
Page 123 |
Page 124 |
Page 125 |
Page 126 |
Page 127 |
Page 128 |
Page 129 |
Page 130 |
Page 131 |
Page 132
