ABSTRACT
Introducción: Las infecciones asociadas al cuidado de la salud representan un problema de salud pública y la transmisión horizontal supone un incremento en la morbimortalidad y los costos en la atención. La vigilancia activa es costosa y tiene alto riesgo de omitir la detección de brotes, mientras que la virtual (modelos matemáticos) permite la búsqueda sistemática de alertas de brotes. El objetivo de este estudio es evaluar la relación costo-efectividad del uso de la herramienta SaTScan-Whonet para la detección temprana de infecciones bacterianas, comparada con la vigilancia tradicional en una institución de alta complejidad de Colombia. Metodología: En un hospital universitario de alta complejidad se realizó un estudio retrospectivo, se identificó un brote bacteriano, se caracterizó clínicamente y por biología molecular. Se extrajeron las bases de datos de los sistemas automatizados de identificación y susceptibilidad microbiológica. Se realizaron análisis retrospectivos de SaTScan-Whonet, así como simulaciones diarias para el primer semestre de 2011 de manera prospectiva; también se identificó la fecha para la alerta de detección de brote, tanto en la vigilancia activa como en la virtual. Resultados: Se aislaron 4.584 microorganismos en los servicios de hospitalización tanto UCI como no UCI entre 2010 y 2011 (2.288 y 2.296 respectivamente). Por vigilancia activa se notificó un brote por Enterococcus faecium el 28 de marzo de 2011, que fue caracterizado por biología molecular con la presencia del gen Van A, que confiere resistencia a glucopéptidos. Se identificó de manera retrospectiva una alerta de brote para E. faecium entre el 14 de marzo y el 10 de mayo de 2011 con un intervalo de recurrencia de 609.384. En los análisis prospectivos simulados se identificó la primera alerta de brote de esta bacteria el 13 de abril de 2011 con un intervalo de recurrencia de 3.897 (p = 0,0002655). Conclusión: La utilización de dicha herramienta de manera prospectiva no fue superior a la vigilancia activa en cuanto a oportunidad en la detección. Los análisis retrospectivos tuvieron un alto rendimiento diagnóstico y podrían ser de utilidad para los sistemas de vigilancia y control de los entes reguladores.
Background: Healthcare-associated infections represent a public health problem, and horizontal transmission has led to an increase in morbidity and mortality as well as higher health care costs. Active surveillance is expensive and carries high risk of failing to detect outbreaks. Virtual surveillance (mathematical models) allows a systematic search for alerts to outbreaks. The objective of this study is to evaluate the cost-effectiveness of the SaTScan-Whonet tool for the early detection of outbreaks of bacterial infection, compared with traditional surveillance, in an institution of high complexity in Colombia. Methodology: In a university hospital of high complexity a retrospective study was performed, identifying a bacterial outbreak that was characterised clinically and by molecular biology techniques. Databases of automated systems of identification and microbiological susceptibility were extracted. Retrospective analyses were performed using SaTScan-Whonet and daily simulations during the first semester of 2011 in a prospective manner. The date for the alert to the detection of the outbreak for both active and virtual surveillance was also identified. Results: A total of 4,584 microorganisms were isolated both inside and outside the ICU bet-ween 2010 and 2011 (2,288 and 2,296, respectively). An outbreak of Enterococcus faecium was identified by active surveillance on March 28, 2011. Using molecular biology techniques, the outbreak was characterised, showing the presence of the vanA gene, which confers resistance to glycopeptides. An alert to an Enterococcus faecium outbreak was retrospectively identified between March 14 and May 10, 2011 with a recurrence interval of 609,384. The first alert to outbreak for this bacterium was identified in a prospective simulated analysis on April 13, 2011 with a recurrence interval of 3,897 (P=.0002655). Conclusion: The use of such a tool prospectively is not superior to active surveillance in regard to timely detection of bacterial outbreaks. Retrospective analyses have high diagnostic ability and could be very helpful in systems of surveillance and control of regulatory entities.
Subject(s)
Humans , Drug Resistance, Microbial , Laboratory Equipment , Disease Outbreaks , Recovery Room , Cross Infection , Enterococcus faecium , Intensive Care UnitsABSTRACT
Objective To analyze the pilot results of both temporal and temporal-spatial models in outbreaks detection in China Infectious Diseases Automated-alert and Response System (CIDARS)to further improve the system. Methods The amount of signal, sensitivity, false alarm rate and time to detection regarding these two models of CIDARS, were analyzed from December 6,2009 to December 5,2010 in 221 pilot counties of 20 provinces. Results The sensitivity of these two models was equal(both 98.15%). However, when comparing to the temporal model, the temporal-spatial model had a 59.86% reduction on the signals(15 702)while the false alarm rate of the temporal-spatial model(0.73%)was lower than the temporal model(1.79%), and the time to detection of the temporal-spatial model(0 day)was also 1 day shorter than the temporal model.Conclusion Comparing to the temporal model, the temporal-spatial model of CIDARS seemed to be better performed on outbreak detection.
ABSTRACT
Objective To compare the different thresholds of 'moving percentile method' for outbreak detection in the China Infectious Diseases Automated-alert and Response System (CIDARS). Methods The thresholds of P50, P60, P70, P80 and P90 were respectively adopted as the candidates of early warning thresholds on the moving percentile method. Aberration was detected through the reported cases of 19 notifiable infectious diseases nationwide from July 1,2008 to June 30,2010. Number of outbreaks and time to detection were recorded and the amount of signals acted as the indicators for determining the optimal threshold of moving percentile method in CIDARS. Results The optimal threshold for bacillary and amebic dysentery was P50. For non-cholera infectious diarrhea,dysentery, typhoid and paratyphoid, and epidemic mumps, it was P60. As for hepatitis A, influenza and rubella, the threshold was P70, but for epidemic encephalitis B it was P80. For the following diseses as scarlet fever, typhoid and paratyphoid, hepatitis E, acute hemorrhagic conjunctivitis, malaria, epidemic hemorrhagic fever, meningococcal meningitis, leptospirosis, dengue fever, epidemic endemic typhus,hepatitis C and measles, it was P90. When adopting the adjusted optimal threshold for 19 infectious diseases respectively, 64 840(12.20%)signals had a decrease, comparing to the adoption of the former defaulted threshold(P50)during the 2 years. However, it did not reduce the number of outbreaks being detected as well as the time to detection, in the two year period. Conclusion The optimal thresholds of moving percentile method for different kinds of diseases were different.Adoption of the right optimal threshold for a specific disease could further optimize the performance of outbreak detection for CIDARS.
ABSTRACT
Objective Using simulated outbreaks to choose the optimal model and its related parameters on measles so as to provide technical support for developing an Auto Warning System(AWS).Methods AEGiS-Cluster Creation Tool was applied to simulate a range oftmique outbreak signals.Then these simulations were added to the aetnal daily counts of measles from the National Disease Surveillance System,between 2005 and 2007.Exponential weighted moving average(EWMA),C1-MILD(C1),C2.MEDIUM(C2).C3-ULTRA(C3)and space.time permutation scar statistic model were comprehensively applied to detect these simulations.Tools for evaluation as Youden's index and detection time were calculated to optimize parameters before an optimal model was finally chosen.Results EWMA(λ=0.6,κ=1.0),C1(κ=0.1,H=3σ),C2(k=0.1,H=30),C3(κ=1.0,H=4σ)and space-time permutation scan statistic(maximum temporal cluster size=7 d,maximum spatial cluster size=5 km)appeared to be the optimal parameters among these models.Youden's index of EWMA was 90.8%and detection time being 0.121 d.Youden's index of C1 was 88.7%and detection time being 0.142 d.Youden's index of C2 was 92.9%and detection time being 0.121 d.Youden's index of C3 was 87.9%and detection time being 0.058 d.Youden's index of space-time permutation scan statistic was 94.3%and detection time being 0.176 d.Conelusion Among these five early warning detection models.space-time permutation scan statistic model had the highest efficacy.