Skip to main content Accessibility help
×
Home

Validation Study of Artificial Neural Network Models for Prediction of Methicillin-Resistant Staphylococcus aureus Carriage

  • Cheng-Chuan Hsu (a1), Yusen E. Lin (a1), Yao-Shen Chen (a1) (a2), Yung-Ching Liu (a2) and Robert R. Muder (a3)...

Abstract

Objective.

Use of active surveillance cultures for methicillin-resistant Staphylococcus aureus (MRSA) for all patients admitted to the intensive care unit has been shown to reduce nosocomial transmission. However, the cost-effectiveness and the utility of implementing use of active surveillance cultures nationwide remain controversial. We sought to develop an artificial neural network (ANN) model that would predict the likelihood of MRSA colonization.

Setting.

Two acute care hospitals, one in Pittsburgh (hospital A) and one in Kaohsiung, Taiwan (hospital B).

Methods.

Nasal cultures were performed for all patients admitted to the hospitals. A total of 46 potential risk factors in hospital A and 86 potential risk factors in hospital B associated with MRSA colonization were assessed. Culture results were obtained; 75% of the data were used for training our ANN model, and the remaining 25% were used for validating our ANN model. The culture results were the “gold standard” for determining the accuracy of the model predictions.

Results.

The ANN model predictions were accurate 95.2% of the time for hospital A (sensitivity, 94.3%; specificity, 96.0%) and 94.2% of the time for hospital B (sensitivity, 96.6%; specificity, 91.8%), integrating all potential risk factors into the model. Only 17 potential risk factors were needed for the hospital A ANN model (accuracy, 90.9%; sensitivity, 98.5%; specificity, 83.4%), and only 20 potential risk factors were needed for the hospital B ANN model (accuracy, 90.5%; sensitivity, 96.6%; specificity, 84.3%), if the minimal risk factor method was used. Cross-validation analysis showed an average accuracy of 85.6% (sensitivity, 91.3%; specificity, 80.0%).

Conclusion.

Our ANN model can be used to predict with an accuracy of more than 90% which patients carry MRSA. The false-negative rates were significantly lower than the false-positive rates in the ANN predictions, which can serve as a safety buffer in case of patient misclassification.

Copyright

Corresponding author

Graduate Institute of Environmental Education National Kaohsiung Normal University, 62 Shen-chong Road, Yanchao, Kaohsiung, Taiwan824 (easonlin@nknucc.nknu.edu.tw)

References

Hide All
1.Chambers, HF. The changing epidemiology of Staphylococcus aureus'. Emerg Infect Dis 2001;7:178182.
2.Klevens, RM, Edwards, JR, Tenover, FC, McDonald, LC, Horan, T, Gaynes, R; National Nosocomial Infections Surveillance System. Changes in the epidemiology of methicillin-resistant Staphylococcus aureus in intensive care units in US hospitals, 1992–2003. Clin Infect Dis 2006;42:389391.
3.Klevens, RM, Morrison, MA, Nadle, J, et al. Invasive methicillin-resistant Staphylococcus aureus infections in the United States. JAMA 2007;298:17631771.
4.Hsueh, PR, Teng, LJ, Chen, WH, et al. Increasing prevalence of methicillin-resistant Staphylococcus aureus causing nosocomial infections at a university hospital in Taiwan from 1986 to 2001. Antimicrob Agents Chemother 2004;48:13611364.
5.Warren, DK, Guth, RM, Coopersmith, CM, Merz, LR, Zack, JE, Fraser, VJ. Impact of a methicillin-resistant Staphylococcus aureus active surveillance program on contact precaution utilization in a surgical intensive care unit. Crit Care Med 2007;35:430434.
6.Warren, DK, Guth, RM, Coopersmith, CM, Merz, LR, Zack, JE, Fraser, VJ. Epidemiology of methicillin-resistant Staphylococcus aureus colonization in a surgical intensive care unit. Infect Control Hosp Epidemiol 2006;27:10321040.
7.Shitrit, P, Gottesman, BS, Katzir, M, Kilman, A, Ben-Nissan, Y, Chowers, M. Active surveillance for methicillin-resistant Staphylococcus aureus (MRSA) decreases the incidence of MRSA bacteremia. Infect Control Hosp Epidemiol 2006;27:10041008.
8.Illinois General Assembly. Bill Status of HB0378, 95th General Assembly. Available at: http://www.ilga.gov/legislation/95/HB/09500HB0378.htm. Accessed September 18, 2007.
9.Maryland General Assembly. Hospitals and Nursing Facilities—Healthcare-Associated Infections Prevention and Control Program. Available at: http://mlis.state.md.us/2006rs/billfile/hb0966.htm. Accessed September 18, 2007.
10.Weber, SG, Huang, SS, Oriola, S, et al. Legislative mandates for use of active surveillance cultures to screen for methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococci: position statement from the Joint SHEA and APIC Task Force. Infect Control Hosp Epidemiol 2007;28:249260.
11.Diekema, DJ, Edmond, MB. Look before you leap: active surveillance for multidrug-resistant organisms. Clin Infect Dis 2007;44:11011107.
12.Farr, BM. What to think if the results of the National Institutes of Health randomized trial of methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococcus control measures are negative (and other advice to young epidemiologists): a review and an au revoir. Infect Control Hosp Epidemiol 2006;27:10961106.
13.Nijssen, S, Bonten, MJ, Weinstein, RA. Are active microbiological surveillance and subsequent isolation needed to prevent the spread of methicillin-resistant Staphylococcus aureus'! Clin Infect Dis 2005;40:405409.
14.Hart, A, Wyatt, J. Connectionist models in medicine: an investigation of their potential. In: Hunter, J, Cookson, J, Wyatt, J, eds. AIME-89, Lecture Notes in Medical Informatics. Vol. 38. Berlin: Springer; 1989:115124.
15.Hamid, A, Dwivedi, US, Singh, TN, et al. Artificial neural networks in predicting optimum renal stone fragmentation by extracorporeal shock wave lithotripsy: a preliminary study. BJU Int 2003;91:821824.
16.Schmid, P, Wischnewsky, MB, Sezer, O, Bohm, R, Possinger, K. Prediction of response to hormonal treatment in metastatic breast cancer. Oncology 2002;63:309316.
17.Frize, M, Ennett, CM, Stevenson, M, Trigg, HC. Clinical decision support systems for intensive care units: using artificial neural networks. Med Eng Phys 2001;23:217225.
18.Frize, M, Ibrahim, D, Seker, H, et al. Predicting clinical outcomes for newborns using two artificial intelligence approaches. Conf Proc IEEE Eng Med Biol Soc 2004;5:32023205.
19.Shang, JS, Lin, YS, Goetz, AM. Diagnosis of MRSA with neural networks and logistic regression approach. Health Care Manag Sci 2000;3:287297.
20.Baxt, WG. Application of artificial neural networks to clinical medicine. Lancet 1995;346:11351138.
21.Karchmer, TB, Durbin, LJ, Simonton, BM, Farr, BM. Cost-effectiveness of active surveillance cultures and contact/droplet precautions for control of methicillin-resistant Staphylococcus aureus. J Hosp Infect 2002;51:126132.
22.Chaix, C, Durand-Zaleski, I, Alberti, C, Brun-Buisson, C. Control of endemic methicillin-resistant Staphylococcus aureus: a cost-benefit analysis in an intensive care unit. JAMA 1999;282:17451751.
23.Papia, G, Louie, M, Tralla, A, Johnson, C, Collins, V, Simor, AE. Screening high-risk patients for methicillin-resistant Staphylococcus aureus on admission to the hospital: is it cost effective? Infect Control Hosp Epidemiol 1999;20:473477.
24.Gavalda, L, Masuet, C, Beltran, J, et al. Comparative cost of selective screening to prevent transmission of methicillin-resistant Staphylococcus aureus (MRSA), compared with the attributable costs of MRSA infection. Infect Control Hosp Epidemiol 2006;27:12641266.
25.Furuno, JP, Harris, AD, Wright, MO, et al. Prediction rules to identify patients with methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococci upon hospital admission. Am J Infect Control 2004;32:436440.
26.Roghmann, MC, Siddiqui, A, Plaisance, K, Standiford, H. MRSA colonization and the risk of MRSA bacteraemia in hospitalized patients with chronic ulcers. J Hosp Infect 2001;47:98103.
27.Goetz, A, Posey, K, Fleming, J, et al. Methicillin-resistant Staphylococcus aureus in the community: a hospital-based study. Infect Control Hosp Epidemiol 1999;20:689691.

Metrics

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed