A Fast Online Monitoring Approach for Surgical Risks

Jiaqi Liu; Xin Lai; Jiayin Wang; Paul B.S. Lai; Xuanping Zhang; Xiaoyan Zhu; Jiaqi Liu; Xin Lai; Jiayin Wang; Paul B.S. Lai; Xuanping Zhang; Xiaoyan Zhu

doi:10.3934/mbe.2020177

Mathematical Biosciences and Engineering

2020, Volume 17, Issue 4: 3130-3146. doi: 10.3934/mbe.2020177

Previous Article Next Article

Research article

A Fast Online Monitoring Approach for Surgical Risks

1.
School of Computer Science and Technology, Xi'an Jiaotong University, Xi'an, China
2.
Department of Surgery, The Chinese University of Hong Kong, Hong Kong, China

Received: 26 December 2019 Accepted: 24 March 2020 Published: 16 April 2020

Risk monitoring has been widely used in health care, further, control charts are often used as monitoring methods for surgical outcomes. Most of the methods can only detect step shifts of position parameters, but cannot take measures on scale parameters. In this paper, we proposed four methods based on EWMA control charts, namely SESOP, STSSO, SESOP-MFIR and STSSO-MFIR, to improve the existing monitoring methods. Specifically, SESOP standardizes variable on the basis of an EWMA charting method; STSSO replaces the statistics of the original EWMA charting method with the score test statistics; for SESOP-MFIR and STSSO-MFIR, we upgrade their control limits from asymptotic to time-varying based on SESOP and STSSO, which enhance the timeliness of the earlier shifts monitoring. In order to verify the improvement of surgical outcomes monitoring, we respectively carry out simulation experiment and a practical application on ESOP and our four methods. SESOP can raise the overall efficiency of detecting shifts; STSSO led to a significant increase in the monitoring stability, especially for small volatilities; the optimization brought by SESOP-MFIR and STSSO-MFIR are more obvious, that the speed of detecting earlier shifts can even be reduced to half of the existing methods. Then, we apply these methods to the SOMIP program of Hong Kong, SESOP-MFIR and STSSO-MFIR have the best performance and can detect early shifts in time. According to the results, the methods we proposed can monitor both early shifts and scale parameters and improve the performance of surgical outcome monitoring in different degrees compared to those existing methods.
- EWMA charting method,
- risk monitoring,
- FIR features,
- variable standardization,
- score test statistics
Citation: Jiaqi Liu, Xin Lai, Jiayin Wang, Paul B.S. Lai, Xuanping Zhang, Xiaoyan Zhu. A Fast Online Monitoring Approach for Surgical Risks[J]. Mathematical Biosciences and Engineering, 2020, 17(4): 3130-3146. doi: 10.3934/mbe.2020177

Related Papers:

Abstract

Risk monitoring has been widely used in health care, further, control charts are often used as monitoring methods for surgical outcomes. Most of the methods can only detect step shifts of position parameters, but cannot take measures on scale parameters. In this paper, we proposed four methods based on EWMA control charts, namely SESOP, STSSO, SESOP-MFIR and STSSO-MFIR, to improve the existing monitoring methods. Specifically, SESOP standardizes variable on the basis of an EWMA charting method; STSSO replaces the statistics of the original EWMA charting method with the score test statistics; for SESOP-MFIR and STSSO-MFIR, we upgrade their control limits from asymptotic to time-varying based on SESOP and STSSO, which enhance the timeliness of the earlier shifts monitoring. In order to verify the improvement of surgical outcomes monitoring, we respectively carry out simulation experiment and a practical application on ESOP and our four methods. SESOP can raise the overall efficiency of detecting shifts; STSSO led to a significant increase in the monitoring stability, especially for small volatilities; the optimization brought by SESOP-MFIR and STSSO-MFIR are more obvious, that the speed of detecting earlier shifts can even be reduced to half of the existing methods. Then, we apply these methods to the SOMIP program of Hong Kong, SESOP-MFIR and STSSO-MFIR have the best performance and can detect early shifts in time. According to the results, the methods we proposed can monitor both early shifts and scale parameters and improve the performance of surgical outcome monitoring in different degrees compared to those existing methods.

References

[1]	W. H. Woodall, S. L. Fogel, S. H. Steiner, The monitoring and improvement of surgical outcome quality, J. Qual. Technol., 47 (2015), 383-399. doi: 10.1080/00224065.2015.11918141
[2]	K. Paynabar, J. Jin, A. B. Yeh, Phase I risk-adjusted control charts for monitoring surgical performance by considering categorical covariates, J. Qual. Technol., 44 (2012), 39-53. doi: 10.1080/00224065.2012.11917880
[3]	M. Douglas, Z. Song, Y. Liu, J. Zhang, A comparative study of memory-type control charts based on robust scale estimators, Qual. Reliab. Eng. Int., 34 (2018), 1079-1102. doi: 10.1002/qre.2309
[4]	J. S. Hunter, The exponentially weighted moving average, J. Qual. Technol, 18 (1986), 203-210. doi: 10.1080/00224065.1986.11979014
[5]	A. Ingolfsson, E. Sachs, Stability and sensitivity of an EWMA controller, J. Qual. Technol., 25 (1993), 271-287. doi: 10.1080/00224065.1993.11979473
[6]	S. V. Crowder, M. D. Hamilton, An EWMA for monitoring a process standard deviation, J. Qual. Technol., 24 (1992), 12-21. doi: 10.1080/00224065.1992.11979369
[7]	R. B. Crosier, A new two-sided cumulative sum quality control scheme, Technometrics, 28 (1986), 187-194. doi: 10.1080/00401706.1986.10488126
[8]	L. Shu, W. Jiang, K. L. Tsui, A comparison of weighted CUSUM procedures that account for monotone changes in population size, Stat. Med., 30 (2011), 725-741. doi: 10.1002/sim.4122
[9]	J. J. Pignatiello, G. C. Runger, Comparisons of multivariate CUSUM charts, J. Qual. Technol., 22 (1990).
[10]	J. Lovegrove, O. Valencia, T. Treasure, C. Sherlaw-Johnson, S. Gallivan, Monitoring the results of cardiac surgery by variable life-adjusted display, Lancet, 350 (1997), 1128-1130. doi: 10.1016/S0140-6736(97)06507-0
[11]	T. Treasure, O. Valencia, C. Sherlaw-Johnson, S. Gallivan, Surgical performance measurement, Health Care Manag. Sci., 5 (2002), 243-248. doi: 10.1023/A:1020373921395
[12]	C. W. Champ, W. H. Woodall, Signal probabilities for runs supplementing a shewhart control chart, Commun. Statist. Simulat. Comput., 26 (1997), 1347-1360.
[13]	J. M. Lucas, Combined Shewart-CUSUM quality control schemes, J. Qual. Technol., 14 (1982), 51-59. doi: 10.1080/00224065.1982.11978790
[14]	L. S. Nelson, The Shewhart control chart-tests for special causes, J. Qual. Technol., 16 (1984), 237-239. doi: 10.1080/00224065.1984.11978921
[15]	J. Neuburger, K. Walker, C. Sherlaw-Johnson, J. Meulen, D. A. Cromwell, Comparison of control charts for monitoring clinical performance using binary data, BMJ Qual. Safety, 26 (2017), 919-928. doi: 10.1136/bmjqs-2016-005526
[16]	S. H. Steiner, W. H. Woodall, Debate: what is the best method to monitor surgical performance, BMC Surg., 16 (2016), 15. doi: 10.1186/s12893-016-0131-8
[17]	X. Zhang, W. H. Woodall, Reduction of the effect of estimation error on in-control performance for risk-adjusted Bernoulli CUSUM chart with dynamic probability control limits, Qual. Reliab. Eng. Int., 33 (2016), 381-386.
[18]	L. Liu, X. Lai, J. Zhang, F. Tsung, Online profile monitoring for surgical outcomes using a weighted score test, J. Qual. Technol., 1 (2018), 88-97.
[19]	D. A. Cook, S. H. Steiner, R. J. Cook, V. T. Farewell, A. P. Morton, Monitoring the evolutionary process of quality: risk-adjusted charting to track outcomes in intensive care, Crit. Care. Med., 31 (2003), 1676-1682. doi: 10.1097/01.CCM.0000065273.63224.A8
[20]	S. H. Steiner, R. J. Cook, V. T. Farewell, T. Treasure, Monitoring surgical performance using risk adjusted cumulative sum charts, Biostatistics, 1 (2000), 441-452. doi: 10.1093/biostatistics/1.4.441
[21]	J. Yue, X. Lai, L. Liu, P. Lai, A new VLAD-based control chart for detecting surgical outcomes, Stat. Med., 36 (2017), 4540-4547. doi: 10.1002/sim.7362
[22]	X. Zhang, W. H. Woodall, Dynamic probability control limits for risk-adjusted Bernoulli CUSUM charts, Stat. Med., 34 (2015), 3336-3348. doi: 10.1002/sim.6547
[23]	D. A. Cook, M. Coory, R. A. Webster, Exponentially weighted moving average charts to compare observed and expected values for monitoring risk-adjusted hospital indicators, BMJ Qual. Safety, 20 (2011), 469-474. doi: 10.1136/bmjqs.2008.031831
[24]	T. Treasure, S. Gallivan, C. Sherlaw-Johnson, Monitoring cardiac surgical performance: A commentary, J. Thorac. Cardiov. Sur., 128 (2004), 823-825. doi: 10.1016/j.jtcvs.2004.03.023
[25]	O. Grigg, V. T. Farewell, A risk-adjusted sets method for monitoring adverse medical outcomes, Stat. Med., 23 (2004), 1593-1602. doi: 10.1002/sim.1763
[26]	S. H. Steiner, EWMA control charts with time-varying control limits and fast initial response, J. Qual. Technol., 31 (1999), 75-86. doi: 10.1080/00224065.1999.11979899
[27]	J. M. Lucas, R. B. Crosier, Fast initial response for CUSUM quality-control schemes: give your CUSUM a head start, Technometrics, 24 (1982), 199-205. doi: 10.1080/00401706.1982.10487759
[28]	A. Haq, J. Brown, E. Moltchanova, Improved fast initial response features for exponentially weighted moving average and cumulative sum control charts, Qual. Reliab. Eng. Int., 30 (2014), 697-710. doi: 10.1002/qre.1521
[29]	S. Knoth, Fast initial response features for EWMA control charts, Stat. Pap., 46 (2005), 47-64. doi: 10.1007/BF02762034
[30]	M. J. Silvapulle, P. Silvapulle, A score test against one-sided alternatives, J. Am. Stat. Assoc., 90 (1995), 342-349. doi: 10.1080/01621459.1995.10476518
[31]	S. K. Sinha, Bootstrap tests for variance components in generalized linear mixed models, Can. J. Stat., 37 (2009), 219-234. doi: 10.1002/cjs.10012

Reader Comments

Your name:*

Email:*
© 2020 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)