- Email: [email protected]
The impact of infection control cost reimbursement policy on central line–associated bloodstream infections
ARTICLE IN PRESS American Journal of Infection Control 000 (2019) 1−6
Contents lists available at ScienceDirect
American Journal of Infection Control journal homepage: www.ajicjournal.org
Major Article
The impact of infection control cost reimbursement policy on central line−associated bloodstream infections Ji Young Park MD, PhD a, Ki Tae Kwon MD, PhD b,*, Won Kee Lee PhD c, Hye In Kim MD, PhD d, Min Jung Kim MD d, Do Young Song MD, PhD e, Mi Hyae Yu RN f, Hyun Ju Park RN f, Kyeong Hee Lee RN f, Hyun Ju Chae RN f a
Department of Pathology, School of Medicine, Kyungpook National University, Daegu, Republic of Korea Department of Internal Medicine, School of Medicine, Kyungpook National University, Daegu, Republic of Korea c Medial Research Collaboration Center, School of Medicine, Kyungpook National University, Kyungpook National University Hospital, Daegu, Republic of Korea d Division of Infectious Diseases, Daegu Fatima Hospital, Daegu, Republic of Korea e Department of Laboratory Medicine, Daegu Fatima Hospital, Daegu, Republic of Korea f Department of Infection Control, Daegu Fatima Hospital, Daegu, Republic of Korea b
Key Words: Central venous catheter Catheter-associated infection Health insurance reimbursement Staffing
Background: In September 2016, the Korean National Health Insurance Service began reimbursing infection control (IC) costs on the condition that a certain number of doctors and full-time nurses for IC be allocated to supported hospitals. We analyzed the impact of the IC cost reimbursement policy on central line−associated bloodstream infections (CLABSIs). Methods: A before-and-after study that analyzed the CLABSI rate trends between preintervention (January 2016 to February 2017) and intervention (March to December 2017) periods using autoregression time series analysis was performed in intensive care units (ICUs) at a 750-bed, secondary care hospital in Daegu, Republic of Korea. The enhanced IC team visited ICUs daily, monitored the implementation of CLABSI prevention bundles, and educated all personnel involved in catheter insertion and maintenance from March 2017. Results: Autoregressive analysis revealed that the CLABSI rates per month in the preintervention and intervention periods were 0.256 (95% confidence interval, 0.613 to 0.101; P = .15) and 0.602 (95% confidence interval, 0.972 to 0.232; P = .008), respectively. The rates of compliance with maximal barrier precautions significantly improved from the preintervention (36.2%) to the intervention (77.9%) period (x2 test, P < .001). Conclusions: The IC cost reimbursement policy accelerated the decline in CLABSI rates significantly in monitored ICUs. A nationwide study to evaluate the effectiveness of the IC cost reimbursement policy for various health care−associated infections is warranted. © 2019 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
Central line−associated bloodstream infections (CLABSIs) significantly increase morbidity, mortality, and health care costs.1 The key strategies for preventing CLABSIs are minimizing bacterial contamination during the insertion and maintenance of central catheters and removing unnecessary catheters immediately. The central line (CL) bundle was designed to be easily implemented at the front lines of medical facilities and includes handwashing, using maximal barrier precautions (MBPs) during the insertion of central venous catheters, using alcoholic chlorhexidine
*Address correspondence to Ki Tae Kwon, MD, PhD, Department of Internal Medicine, School of Medicine, Kyungpook National University, 807 Hokuk-ro, Buk-gu, Daegu 41404, Republic of Korea. E-mail address: [email protected] (K.T. Kwon). Conflicts of interest: None to report.
antiseptic for skin preparation, avoiding the femoral site if possible, and removing unnecessary catheters.2 CL bundle use has been shown to be effective, sustainable, and cost-effective for both adults and children.3 Maintenance (postinsertion care) bundle implementation after catheter insertion has been proposed as a strategy to optimize CL bundle use.4 The maintenance bundle consists of (1) the daily inspection of the insertion site; (2) site care if the dressing is wet or soiled, or has not been changed for 7 days; (3) the documentation of the ongoing need for the catheter; (4) the proper application of a chlorhexidine gluconate-impregnated sponge at the insertion site; (5) the performance of the hand hygiene procedure before handling the intravenous system; and (6) the application of an alcohol scrub to the infusion hub for 15 seconds before each use.4 Several guidelines recommend CL bundle implementation and proper postinsertion care to prevent CLABSIs.1,3,5,6
https://doi.org/10.1016/j.ajic.2019.09.002 0196-6553/© 2019 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
ARTICLE IN PRESS 2
J.Y. Park et al. / American Journal of Infection Control 00 (2019) 1−6
Multifaceted approaches and comprehensive interventions— including continuous education and periodic assessment and feedback for knowledge of and adherence to guidelines for all personnel involved in the insertion and maintenance of central catheters—are critical for improving compliance with these guidelines and preventing CLABSIs. However, this is difficult to achieve if there are inadequate medical staff responsible for infection control (IC). A 2015 national survey in the Republic of Korea showed that there was a shortage of skilled workers and inferior infrastructure for IC.7 The authors of this survey concluded that support for policy-directed management is necessary to achieve effective infection prevention and control.7 In response to this conclusion, the Korean National Health Insurance Service (NHIS) began reimbursing IC costs from September 2016 on the condition that a certain number of doctors and full-time nurses for IC be allocated to supported hospitals. The Korean Nosocomial Infections Surveillance System reported that the CLABSI rate was 2.99 per 1,000 catheter-days among a total of 171 intensive care units (ICUs) from 2006 to 2012. This rate did not change significantly during this period.8 The enhanced infection control team (ICT)—on the back of the IC cost reimbursement policy—visited ICUs daily, monitored the implementation of CLABSI prevention bundles (CL and maintenance bundles), and educated all personnel involved in CL insertion and maintenance from March 2017. We analyzed the impact of these interventions, which were made possible by the IC cost reimbursement policy, on trends in CLABSI rates. METHODS Setting and design A before-and-after study from January 2016 to December 2017 was conducted in 3 ICUs (medical, surgical, and neurosurgical) at a 750-bed, secondary care hospital in Daegu, Republic of Korea. The medical ICU (MICU) had 24 beds, which had increased from 14 beds in May 2017; the surgical ICU (SICU) had 9 beds, which decreased from 13 beds in April 2017; and the neurosurgical ICU (NSICU) had 16 beds. CLABSI surveillance in the ICUs was performed monthly by the ICT, and the results were reported to the Korean Nosocomial Infections Surveillance System. Educating all personnel about CLABSI prevention bundles and maintaining adherence to them in the ICUs were important roles of the ICT. However, it was difficult to fulfill these roles effectively because only 2 full-time nurses and 1 doctor holding the main clinician post were allocated to the ICT. In September 2016, the Korean NHIS began reimbursing IC costs on the condition that a certain number of doctors and full-time nurses for IC be allocated to supported hospitals. The highest reimbursing standard ratio was 1 infection control nurse (ICN) per 150 hospital beds and 1 infection control doctor (ICD) working 20 hours a week for IC per 300 hospital beds or 1 full-time ICD per 600 hospital beds.9 ICNs and ICDs should be qualified for educational and career standards presented by the NHIS.9 To meet the highest reimbursing standard of this policy, the number of ICNs increased from 2 to 5 in September 2016, and the first full-time ICD was allocated to our hospital in March 2017. The enhanced ICT visited ICUs daily, monitored the implementation of CLABSI prevention bundles (CL and maintenance bundles), and educated all personnel involved in CL insertion and maintenance from March 2017. We analyzed the impact of these interventions, which were made possible by the IC cost reimbursement policy. This study was approved by the institutional review board of Daegu Fatima Hospital (DFE18ORIO029-R1). Interventions Among the 5 components of CL bundles, alcoholic chlorhexidine gluconate use for skin preparation and avoidance of the femoral site
were already implemented well before this study. Before the study, handwashing rates fluctuated and compliance with MBPs (1 wearing a cap,2 mask,3 sterile gown,4 sterile gloves,5 and a sterile full-body drape during the insertion of central venous catheters) was poor. To improve handwashing and MBP compliance rates, the education materials for handwashing and MBPs—such as ICT-generated posters and videos—were distributed, and doctors who were involved in catheter insertion procedures were trained by the ICT. The handwashing and MBP performance compliance during CL insertion were measured by ICU nurses according to the checklist, and the performance of these measurements was monitored by the ICT. The doctors received immediate individual feedback from an ICD in the ICU regarding their performance. To minimize bacterial contamination while central catheters were maintained, the ICT educated the nurses about maintenance bundles, with specific advice about (1) hand hygiene before handling the intravenous system; (2) the disinfection of catheter hubs, needleless connectors, and injection ports using an alcoholic chlorhexidine preparation for 15 seconds before each use; (3) recording dressing dates and documenting the ongoing need for the catheter; (4) monitoring and recording the condition of the catheter site skin and dressings every working shift; (5) site care if the dressing was wet or soiled, or had not been changed for 7 days; and (6) the proper application of chlorhexidine gluconate-impregnated dressings at the insertion site. The ICT also established and provided guidance about the standard aseptic process for changing administrative sets. The ICD had direct communication with doctors concerning the removal of nonessential catheters. Regular meetings between the ICT and ICU staff, including nurses and doctors, were held to implement and coordinate the CLABSI prevention bundles. The compliance rates with all CLABSI prevention bundles and CLABSI rates were monitored, and feedback was provided to all health care personnel involved in the insertion and maintenance of intravascular catheters. Feedback was also provided to the ICU directors, the committee for ICU care, and administrators. When the number of CLABSI-free days reached 100, ICU staff were commended for their efforts as a form of positive reinforcement. Definitions and data collection CLABSIs were defined using standard National Healthcare Safety Network surveillance definitions and identified by the ICT throughout the study period.10 The CLABSI rate was expressed as the number of CLABSIs per 1,000 catheter-days. To compare CLABSI rates in the ICUs before and after the intervention, the data were divided into 2 periods: preintervention (from January 2016 to February 2017) and intervention (from March to December 2017) periods. The count was performed each day, and each patient with a CL was counted as a CL day. MBP and maintenance bundle compliance were calculated by dividing the number of CL insertions during which the recommended practice was followed by the total number of CL insertions observed by ICNs or ICU nurses. Microbiologic data were obtained for each CLABSI case to identify causative organisms. Microorganisms were rieux, Marcy-l’Etoile, identified using the VITEK 2 system (bioMe France). We referenced the national data on IC cost reimbursement in the medical statistics of the Bigdata Hub website provided by Health Insurance Review and Assessment service.11 Statistical analysis The CLABSI rate trends were expressed as CLABSI rates per month and slopes of them during the preintervention and intervention periods. They were calculated and analyzed for statistical significance through autoregression time series analysis and the Wald x2 test was used to compare estimates among the 2 periods. Compliance with each MBP element was compared between the periods using the
ARTICLE IN PRESS J.Y. Park et al. / American Journal of Infection Control 00 (2019) 1−6
2.34 3.07 1.43 1.82
CLABSI ratey
Pearson x2 test with Yates’ continuity correction. A P value of <.05 was considered statistically significant. Data were analyzed using SAS 9.4 (SAS Institute Incorporated, Cary, NC) and Excel 2013 (Microsoft Corporation, Redmond, WA).
3
14 9 2 3 0.51 0.51 0.65 0.42 11,803 5,771 2,155 3,877 5,978 2,935 1,396 1,647 3.41 4.30 1.53 4.68 25 12 4 9 0.49 0.50 0.59 0.40 14,808 5,585 4,435 4,788 7,326 2,790 2,611 1,925
Patient-days Central line-days
CLABSI, central line−associated bloodstream infection; ICU, intensive care unit. *Central line use ratio = central line-days per patient-day. y CLABSI rate = CLABSI events per 1,000 central line-days
Several guidelines recommend multifaceted approaches to CLABSI prevention. These approaches have been effective in both developed and developing countries.1-3,5,6,12-15 However, limited available resources—including human resources as well as financial and managerial support—lead to poor compliance or performance regarding these guidelines.16,17 Given these circumstances, the establishment of
Study site
DISCUSSION
Table 1 CLABSI rates in participating ICUs between preintervention and intervention periods
Compliance with CLABSI prevention bundles
Central line use ratio*
Preintervention (14 months)
CLABSI event
Enterococcus spp were the most common (35.9%), followed by Candida spp (17.9%), Staphylococcus aureus (10.3%), and Acinetobacter baumannii (10.3%). Of the 14 Enterococcus spp, 13 (92.9%) were resistant to penicillin, and 5 (35.7%) were resistant to vancomycin. All Candida spp were susceptible to fluconazole, amphotericin, and voriconazole tested by the VITEK II system. Of the 4 Staphylococcus aureus isolates, 2 were susceptible to oxacillin, and 2 were resistant. All Acinetobacter baumannii and Pseudomonas aeruginosa were resistant to carbapenem.
CLABSI ratey
Study periods
Microbiologic features of CLABSIs in ICUs
The comparisons of compliance rates for hand hygiene, MBPs, and each MBP component between the preintervention and intervention periods are shown in Figure 2. The numbers of compliance measurements between the preintervention and intervention periods were 94 and 317, respectively. The overall MBP compliance rate along with the compliance rates for cap, mask, sterile gown, and full-body drape use improved significantly from the preintervention to intervention period (P < .001). The compliance rates for hand hygiene (84.0%) and sterile glove use before catheter insertion (90.5%) were high during the preintervention period and remained high during the intervention period (96.8% and 98.7%, respectively). The compliance rates for hand hygiene, all MBPs, and each MBP component among the 3 ICUs during the preintervention and intervention periods were not significantly different. The rate of compliance with the maintenance bundle during the intervention period was 95.96%.
Central line-days
Patient-days
Intervention (10 months)
During the 24-month study period, 26,611 patient-days and 13,304 CL days were observed at a total of 49 beds in MICU, SICU, and NICU. The CLABSI rate for overall ICUs during the entire study period was 2.93 per 1,000 catheter-days. The overall CLABSI rate and that of each ICU decreased from the preintervention to the intervention period, even though the CL use ratio increased (Table 1). The most prominent decline in CLABSI rate, from 4.68 to 1.82 per 1,000 catheter-days, was observed in the NSICU (Table 1). The overall CLABSI rates per month as well as the rates per month for each of the 3 ICUs—which were analyzed through autoregressive time series analysis—are shown in Table 2, and the overall CLABSI rate trend is illustrated in Figure 1. The CLABSI rates per month for overall ICUs (0.602; 95% confidence interval, 0.972 to 0.232; P = .008) and the MICU (0.788; 95% confidence interval, 1.479 to 0.098; P = .036) decreased significantly in the intervention period. The difference in CLABSI rate trend between the preintervention and intervention periods was only statistically significant in the SICU (P < .001).
Overall (all ICUs) Medical ICU Surgical ICU Neurosurgical ICU
CLABSI rates in ICUs
Central line use ratio*
CLABSI event
RESULTS
ARTICLE IN PRESS J.Y. Park et al. / American Journal of Infection Control 00 (2019) 1−6
CI, confidence interval; CLABSI, central line−associated bloodstream infection; ICU, intensive care unit; SE, standard error.
P value
.066 .128 <.001 .166 3.39 2.32 14.39 1.92 .008 .036 .086 .240 0.164 0.305 0.284 0.322 0.602 (0.972 to 0.232) 0.788 (1.479 to 0.098) 0.568 (1.21 to 0.075) 0.415 (1.143 to 0.314) 0.165 0.394 0.128 0.288 Overall (all ICUs) Medical ICU Surgical ICU Neurosurgical ICU
0.256 (0.613 to 0.101) 0.160 (1.011 to 0.691) 0.147 (0.423 to 0.129) 0.590 (1.212 to 0.031)
.15 .693 .273 .065
Wald x2 test statistic P value SE Estimate (95% CI)
Preintervention Study site
Table 2 CLABSI rates per month in participating ICUs using autoregressive time series analysis
Study period
Estimate (95% CI)
Intervention
SE
P value
Comparison of estimates between preintervention and intervention periods
4
reasonable medical insurance fees for effective IC programs was suggested based on a national survey that evaluated the current status of IC personnel and infrastructure resources in the Republic of Korea.7 In response to this suggestion, the Korean NHIS began reimbursing IC costs from September 2016 based on the number of ICNs and ICDs per hospital bed. The highest reimbursing standard ratio was 1 ICN per 150 hospital beds and 1 ICD working 20 hours a week for IC per 300 hospital beds or 1 full-time ICD per 600 hospital beds.9 The reimbursement benefit to general hospitals under the highest standard was $2.06 per patient-day in 2017 (Table 3).18 If a 600-bed general hospital hires 1 full-time ICD and 4 full-time ICNs to meet the highest reimbursing standard, a total of $406,026 (0.9ⅹ600ⅹ365ⅹ2.06) per year will be reimbursed to the general hospital assuming a bed operation rate of 90%. In fact, the hospital where this study was conducted had a total of 230,693 patient-days, and $468,307 (2.03ⅹ230,693) was reimbursed in 2017. We believe that this level of reimbursement benefit meets the approximate annual salaries of 1 full-time ICD and 4 full-time ICNs, which are required to maintain the highest reimbursing standard for a general hospital, although there are some variations. This staffing level was similar to or lower than the median ratio of a 2015 European Society of Clinical Microbiology and Infectious Diseases member survey.19 The enhanced ICT achieved their goal of reducing CLABSIs. We believe that strengthening ICT staffing levels will have positive effects on various other IC indicators. The rates of ventilator-associated pneumonia also decreased from 1.23 per 1,000 ventilator-days in 2016 to 0.58 in 2017. Evidence also suggests that ICT activity improves the cost-effectiveness of IC in hospitals.20 The Korean governmental data on the cost of the IC reimbursement policy is shown in Table 3. The overall cost of the IC reimbursement policy was $48,831,297 in 2017 and $59,963,245 in 2018. All tertiary care general hospitals benefited from this policy and took approximately half of the overall cost. However, secondary care general and nongeneral hospitals experienced difficulties hiring ICDs or ICNs to meet these standards, and even though the numbers were much larger, many of them did not benefit from this policy. To address this problem, nongeneral hospitals were more likely to obtain benefits than general hospitals if they met this standard. The reimbursement costs per patients-day according to the hospital grade increased slightly yearly. Although there was an ICT as well as IC programs before this study period, it was difficult to conclude that IC programs were effectively carried out due to insufficient infrastructure, especially in terms of IC staffing levels. At the beginning of the preintervention period, the CLABSI rates were high relative to the national average, and they were decreasing through various efforts, but after the IC staffing level was strengthened, CLABSI rates declined significantly (Fig 1). The role of a full-time ICD was critical because doctor-to-doctor communications were effective, bringing about changes in doctor behavior. In particular, a senior ICD who is looked on respectfully as a figure of authority and leadership is beneficial. In our hospital, similar to the response to the 2015 European Society of Clinical Microbiology and Infectious Diseases member survey,19 a senior infectious diseases doctor undertook the role of the first full-time ICD. CLABSI rates increased at the start of the intervention period. The range of CL use ratios was 0.42 to 0.6, and their values in March and April 2017 were 0.57 and 0.6, respectively, which were the highest and second highest values throughout the entire study periods. These high values might have been owing to the conditions of patients during these periods. Because only patients from the neurosurgical department were admitted to the NSICU, the ICT only needed to communicate with doctors and nurses from the neurosurgical department in the NSICU. In the NSICU, the ICT was able to coordinate CLABSI prevention bundles more effectively than in the MICU and SICU, so the CL use ratio was the lowest and the reduction in the CLABSI rate was the largest,
ARTICLE IN PRESS J.Y. Park et al. / American Journal of Infection Control 00 (2019) 1−6
5
Fig 1. Trends of central line−associated bloodstream infection (CLABSI) rates during preintervention and intervention periods. Dotted lines are the trends of the preintervention and intervention periods, and the slopes are calculated through autoregressive analysis. Graph values are monthly CLABSI rates. Autoregressive analysis revealed decreases of CLABSI rates per month during preintervention and intervention periods of 0.256 (95% confidence interval, 0.613 to 0.101; P=.15) and 0.602 (95% confidence interval, 0.972 to 0.232; P=.008), respectively.
Fig 2. Compliance rates for hand hygiene, MBPs, and MBP components for all intensive care units during preintervention and intervention periods. The asterisks represent statistically significant differences. MBP, maximal barrier precautions. Table 3 Overall cost (calculated by $0.84 per 1,000 KRW of exchange rate) for infection control reimbursement per year according to category of hospitals and grade of reimbursement standards in Korea Hospital category
2017 Number First grade
Tertiary general hospital Secondary general hospital Nongeneral hospital Total
43 301 1,466 1,810
2018
Total reimbursement cost ($)
22,384,163* 20,215,653* 49,154k 42,648,970
Second grade 3,263,289y 2,853,994y 65,044{ 6,182,327
Number Total 25,647,452 23,069,647 114,198 48,831,297
42 311 1,465 1,818
Total reimbursement cost ($) First grade
Second grade
28,107,160z 26,752,004z 206,343# 55,065,507
2,241,704x 2,274,333x 380,701** 4,896,738
Total 30,348,864 29,026,337 587,044 59,962,245
KRW, South Korean Won. The reimbursement costs per patient-days: *$2.03; y$1.67; z$2.06; x$1.69; k$2.44; {$2.06; #$2.44; **$2.09.
although no difference was observed in hand hygiene and MBP performance among the ICUs. In the SICU, the CLABSI rates were the lowest throughout the entire study period, as many patients were admitted for a short period after surgery. It was most difficult for the ICT to implement a multifaceted approach for improving CLABSI prevention
in the MICU because of the array of complex pathologies among the patients admitted to the MICU and transferred from various departments, who were then hospitalized for a long time in the MICU. There were some limitations to our study. First, we conducted this intervention at a single center for a short period; therefore, our
ARTICLE IN PRESS J.Y. Park et al. / American Journal of Infection Control 00 (2019) 1−6
6
findings may not be generalizable to all institutions. Because of the small sample size, differences in estimates between the preintervention and intervention periods were not statistically significant, except for in the SICU. This effect could be limited to secondary care hospitals because several tertiary care hospitals had already reinforced MBP monitoring before the reimbursement policy in association with the development of critical care medicine. Although overall health care− associated infections (HAIs) might also decrease in other hospitals after the implementation of the reimbursement policy, the findings of this research cannot be generalized. Therefore, a nationwide study to evaluate the effectiveness of the IC cost reimbursement policy for the prevention of not only CLABSIs but also various HAIs is warranted. Second, we compared only hand hygiene and MBPs among CLABSI prevention bundles between the preintervention and intervention periods. The maintenance bundle implementation rate was monitored only during the intervention period. Other factors associated with CLABSI rates may have been overlooked. Third, this study was a time series analysis that only captured a short period. Thus, a multicenter study or big data analysis is warranted in the future. CONCLUSIONS The reimbursement policy for IC costs accelerated the decline in CLABSI rates significantly by providing infrastructure and enhancing IC staffing levels to allow for a comprehensive and multifaceted approach to improving the implementation of CLABSI prevention bundles in ICUs. This study showed that national financial and policy support helps control HAIs at a front-line medical facility. However, this policy could have a different effect on each hospital. Therefore, this research cannot be generalized and a nationwide study to evaluate the effectiveness of the IC cost reimbursement policy for various HAIs is warranted. References 1. Ling ML, Apisarnthanarak A, Jaggi N, Harrington G, Morikane K, Thu le TA, et al. APSIC guide for prevention of central line associated bloodstream infections (CLABSI). Antimicrob Resist Infect Control 2016;5:16. 2. Pronovost P, Needham D, Berenholtz S, Sinopoli D, Chu H, Cosgrove S, et al. An intervention to decrease catheter-related bloodstream infections in the ICU. New Eng J Med 2006;355:2725-32. 3. Marschall J, Mermel LA, Fakih M, Hadaway L, Kallen A, O’Grady NP, et al. Strategies to prevent central line-associated bloodstream infections in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol 2014;35:753-71.
4. Guerin K, Wagner J, Rains K, Bessesen M. Reduction in central line-associated bloodstream infections by implementation of a postinsertion care bundle. Am J Infect Control 2010;38:430-3. 5. O’Grady NP, Alexander M, Burns LA, Dellinger EP, Garland J, Heard SO, et al. Guidelines for the prevention of intravascular catheter-related infections. Clin Infect Dis 2011;52:e162-93. 6. Loveday HP, Wilson JA, Pratt RJ, Golsorkhi M, Tingle A, Bak A, et al. epic3: national evidence-based guidelines for preventing healthcare-associated infections in NHS hospitals in England. J Hosp Infect 2014;86(Suppl 1):1-70. 7. Yoon YK, Lee SE, Seo BS, Kim HJ, Kim JH, Yang KS, et al. Current status of personnel and infrastructure resources for infection prevention and control programs in the Republic of Korea: a national survey. Am J Infect Control 2016;44:e189-93. 8. Choi JY, Kwak YG, Yoo H, Lee SO, Kim HB, Han SH, et al. Trends in the incidence rate of device-associated infections in intensive care units after the establishment of the Korean Nosocomial Infections Surveillance System. J Hosp Infect 2015;91:28-34. 9. Health Insurance Review and Assessment Service. Personnel standard for infection control cost reimbursement. Available from: http://www.hira.or.kr/main.do. Accessed April 29, 2019. 10. Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health careassociated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008;36:309-32. 11. Kaier K, Wolkewitz M, Hehn P, Mutters NT, Heister T. The impact of hospitalacquired infections on the patient-level reimbursement-cost relationship in a DRG-based hospital payment system. Int J Health Econ Manag 2019 Jun 5, [Epub ahead of print]. 12. Bion J, Richardson A, Hibbert P, Beer J, Abrusci T, McCutcheon M, et al. ‘Matching Michigan’: a 2-year stepped interventional programme to minimise central venous catheter-blood stream infections in intensive care units in England. BMJ Qual Saf 2013;22:110-23. 13. Berenholtz SM, Lubomski LH, Weeks K, Goeschel CA, Marsteller JA, Pham JC, et al. Eliminating central line-associated bloodstream infections: a national patient safety imperative. Infect Control Hosp Epidemiol 2014;35:56-62. 14. Yaseen M, Al-Hameed F, Osman K, Al-Janadi M, Al-Shamrani M, Al-Saedi A, et al. A project to reduce the rate of central line associated bloodstream infection in ICU patients to a target of zero. BMJ Qual Improv Rep 2016;5. 15. Marsteller JA, Sexton JB, Hsu YJ, Hsiao CJ, Holzmueller CG, Pronovost PJ, et al. A multicenter, phased, cluster-randomized controlled trial to reduce central lineassociated bloodstream infections in intensive care units. Crit Care Med 2012;40:2933-9. 16. Rosenthal VD. Central line-associated bloodstream infections in limited-resource countries: a review of the literature. Clin Infect Dis 2009;49:1899-907. 17. Valencia C, Hammami N, Agodi A, Lepape A, Herrejon EP, Blot S, et al. Poor adherence to guidelines for preventing central line-associated bloodstream infections (CLABSI): results of a worldwide survey. Antimicrob Resist Infect Control 2016;5:49. 18. Health Insurance Review and Assessment Service. Medical statistics in heathcare Bigdata Hub. Available from: https://opendata.hira.or.kr/home.do. Accessed August 2, 2019. 19. Dickstein Y, Nir-Paz R, Pulcini C, Cookson B, Beovic B, Tacconelli E, et al. Staffing for infectious diseases, clinical microbiology and infection control in hospitals in 2015: results of an ESCMID member survey. Clin Microbiol Infect 2016;22, 812.e9e17. 20. Seko T, Tachi T, Kawashima N, Maeda T, Yasuda M, Noguchi Y, et al. Economic evaluation of infection control activities. J Hosp Infect 2017;96:371-6.
Contents lists available at ScienceDirect
American Journal of Infection Control journal homepage: www.ajicjournal.org
Major Article
The impact of infection control cost reimbursement policy on central line−associated bloodstream infections Ji Young Park MD, PhD a, Ki Tae Kwon MD, PhD b,*, Won Kee Lee PhD c, Hye In Kim MD, PhD d, Min Jung Kim MD d, Do Young Song MD, PhD e, Mi Hyae Yu RN f, Hyun Ju Park RN f, Kyeong Hee Lee RN f, Hyun Ju Chae RN f a
Department of Pathology, School of Medicine, Kyungpook National University, Daegu, Republic of Korea Department of Internal Medicine, School of Medicine, Kyungpook National University, Daegu, Republic of Korea c Medial Research Collaboration Center, School of Medicine, Kyungpook National University, Kyungpook National University Hospital, Daegu, Republic of Korea d Division of Infectious Diseases, Daegu Fatima Hospital, Daegu, Republic of Korea e Department of Laboratory Medicine, Daegu Fatima Hospital, Daegu, Republic of Korea f Department of Infection Control, Daegu Fatima Hospital, Daegu, Republic of Korea b
Key Words: Central venous catheter Catheter-associated infection Health insurance reimbursement Staffing
Background: In September 2016, the Korean National Health Insurance Service began reimbursing infection control (IC) costs on the condition that a certain number of doctors and full-time nurses for IC be allocated to supported hospitals. We analyzed the impact of the IC cost reimbursement policy on central line−associated bloodstream infections (CLABSIs). Methods: A before-and-after study that analyzed the CLABSI rate trends between preintervention (January 2016 to February 2017) and intervention (March to December 2017) periods using autoregression time series analysis was performed in intensive care units (ICUs) at a 750-bed, secondary care hospital in Daegu, Republic of Korea. The enhanced IC team visited ICUs daily, monitored the implementation of CLABSI prevention bundles, and educated all personnel involved in catheter insertion and maintenance from March 2017. Results: Autoregressive analysis revealed that the CLABSI rates per month in the preintervention and intervention periods were 0.256 (95% confidence interval, 0.613 to 0.101; P = .15) and 0.602 (95% confidence interval, 0.972 to 0.232; P = .008), respectively. The rates of compliance with maximal barrier precautions significantly improved from the preintervention (36.2%) to the intervention (77.9%) period (x2 test, P < .001). Conclusions: The IC cost reimbursement policy accelerated the decline in CLABSI rates significantly in monitored ICUs. A nationwide study to evaluate the effectiveness of the IC cost reimbursement policy for various health care−associated infections is warranted. © 2019 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
Central line−associated bloodstream infections (CLABSIs) significantly increase morbidity, mortality, and health care costs.1 The key strategies for preventing CLABSIs are minimizing bacterial contamination during the insertion and maintenance of central catheters and removing unnecessary catheters immediately. The central line (CL) bundle was designed to be easily implemented at the front lines of medical facilities and includes handwashing, using maximal barrier precautions (MBPs) during the insertion of central venous catheters, using alcoholic chlorhexidine
*Address correspondence to Ki Tae Kwon, MD, PhD, Department of Internal Medicine, School of Medicine, Kyungpook National University, 807 Hokuk-ro, Buk-gu, Daegu 41404, Republic of Korea. E-mail address: [email protected] (K.T. Kwon). Conflicts of interest: None to report.
antiseptic for skin preparation, avoiding the femoral site if possible, and removing unnecessary catheters.2 CL bundle use has been shown to be effective, sustainable, and cost-effective for both adults and children.3 Maintenance (postinsertion care) bundle implementation after catheter insertion has been proposed as a strategy to optimize CL bundle use.4 The maintenance bundle consists of (1) the daily inspection of the insertion site; (2) site care if the dressing is wet or soiled, or has not been changed for 7 days; (3) the documentation of the ongoing need for the catheter; (4) the proper application of a chlorhexidine gluconate-impregnated sponge at the insertion site; (5) the performance of the hand hygiene procedure before handling the intravenous system; and (6) the application of an alcohol scrub to the infusion hub for 15 seconds before each use.4 Several guidelines recommend CL bundle implementation and proper postinsertion care to prevent CLABSIs.1,3,5,6
https://doi.org/10.1016/j.ajic.2019.09.002 0196-6553/© 2019 Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
ARTICLE IN PRESS 2
J.Y. Park et al. / American Journal of Infection Control 00 (2019) 1−6
Multifaceted approaches and comprehensive interventions— including continuous education and periodic assessment and feedback for knowledge of and adherence to guidelines for all personnel involved in the insertion and maintenance of central catheters—are critical for improving compliance with these guidelines and preventing CLABSIs. However, this is difficult to achieve if there are inadequate medical staff responsible for infection control (IC). A 2015 national survey in the Republic of Korea showed that there was a shortage of skilled workers and inferior infrastructure for IC.7 The authors of this survey concluded that support for policy-directed management is necessary to achieve effective infection prevention and control.7 In response to this conclusion, the Korean National Health Insurance Service (NHIS) began reimbursing IC costs from September 2016 on the condition that a certain number of doctors and full-time nurses for IC be allocated to supported hospitals. The Korean Nosocomial Infections Surveillance System reported that the CLABSI rate was 2.99 per 1,000 catheter-days among a total of 171 intensive care units (ICUs) from 2006 to 2012. This rate did not change significantly during this period.8 The enhanced infection control team (ICT)—on the back of the IC cost reimbursement policy—visited ICUs daily, monitored the implementation of CLABSI prevention bundles (CL and maintenance bundles), and educated all personnel involved in CL insertion and maintenance from March 2017. We analyzed the impact of these interventions, which were made possible by the IC cost reimbursement policy, on trends in CLABSI rates. METHODS Setting and design A before-and-after study from January 2016 to December 2017 was conducted in 3 ICUs (medical, surgical, and neurosurgical) at a 750-bed, secondary care hospital in Daegu, Republic of Korea. The medical ICU (MICU) had 24 beds, which had increased from 14 beds in May 2017; the surgical ICU (SICU) had 9 beds, which decreased from 13 beds in April 2017; and the neurosurgical ICU (NSICU) had 16 beds. CLABSI surveillance in the ICUs was performed monthly by the ICT, and the results were reported to the Korean Nosocomial Infections Surveillance System. Educating all personnel about CLABSI prevention bundles and maintaining adherence to them in the ICUs were important roles of the ICT. However, it was difficult to fulfill these roles effectively because only 2 full-time nurses and 1 doctor holding the main clinician post were allocated to the ICT. In September 2016, the Korean NHIS began reimbursing IC costs on the condition that a certain number of doctors and full-time nurses for IC be allocated to supported hospitals. The highest reimbursing standard ratio was 1 infection control nurse (ICN) per 150 hospital beds and 1 infection control doctor (ICD) working 20 hours a week for IC per 300 hospital beds or 1 full-time ICD per 600 hospital beds.9 ICNs and ICDs should be qualified for educational and career standards presented by the NHIS.9 To meet the highest reimbursing standard of this policy, the number of ICNs increased from 2 to 5 in September 2016, and the first full-time ICD was allocated to our hospital in March 2017. The enhanced ICT visited ICUs daily, monitored the implementation of CLABSI prevention bundles (CL and maintenance bundles), and educated all personnel involved in CL insertion and maintenance from March 2017. We analyzed the impact of these interventions, which were made possible by the IC cost reimbursement policy. This study was approved by the institutional review board of Daegu Fatima Hospital (DFE18ORIO029-R1). Interventions Among the 5 components of CL bundles, alcoholic chlorhexidine gluconate use for skin preparation and avoidance of the femoral site
were already implemented well before this study. Before the study, handwashing rates fluctuated and compliance with MBPs (1 wearing a cap,2 mask,3 sterile gown,4 sterile gloves,5 and a sterile full-body drape during the insertion of central venous catheters) was poor. To improve handwashing and MBP compliance rates, the education materials for handwashing and MBPs—such as ICT-generated posters and videos—were distributed, and doctors who were involved in catheter insertion procedures were trained by the ICT. The handwashing and MBP performance compliance during CL insertion were measured by ICU nurses according to the checklist, and the performance of these measurements was monitored by the ICT. The doctors received immediate individual feedback from an ICD in the ICU regarding their performance. To minimize bacterial contamination while central catheters were maintained, the ICT educated the nurses about maintenance bundles, with specific advice about (1) hand hygiene before handling the intravenous system; (2) the disinfection of catheter hubs, needleless connectors, and injection ports using an alcoholic chlorhexidine preparation for 15 seconds before each use; (3) recording dressing dates and documenting the ongoing need for the catheter; (4) monitoring and recording the condition of the catheter site skin and dressings every working shift; (5) site care if the dressing was wet or soiled, or had not been changed for 7 days; and (6) the proper application of chlorhexidine gluconate-impregnated dressings at the insertion site. The ICT also established and provided guidance about the standard aseptic process for changing administrative sets. The ICD had direct communication with doctors concerning the removal of nonessential catheters. Regular meetings between the ICT and ICU staff, including nurses and doctors, were held to implement and coordinate the CLABSI prevention bundles. The compliance rates with all CLABSI prevention bundles and CLABSI rates were monitored, and feedback was provided to all health care personnel involved in the insertion and maintenance of intravascular catheters. Feedback was also provided to the ICU directors, the committee for ICU care, and administrators. When the number of CLABSI-free days reached 100, ICU staff were commended for their efforts as a form of positive reinforcement. Definitions and data collection CLABSIs were defined using standard National Healthcare Safety Network surveillance definitions and identified by the ICT throughout the study period.10 The CLABSI rate was expressed as the number of CLABSIs per 1,000 catheter-days. To compare CLABSI rates in the ICUs before and after the intervention, the data were divided into 2 periods: preintervention (from January 2016 to February 2017) and intervention (from March to December 2017) periods. The count was performed each day, and each patient with a CL was counted as a CL day. MBP and maintenance bundle compliance were calculated by dividing the number of CL insertions during which the recommended practice was followed by the total number of CL insertions observed by ICNs or ICU nurses. Microbiologic data were obtained for each CLABSI case to identify causative organisms. Microorganisms were rieux, Marcy-l’Etoile, identified using the VITEK 2 system (bioMe France). We referenced the national data on IC cost reimbursement in the medical statistics of the Bigdata Hub website provided by Health Insurance Review and Assessment service.11 Statistical analysis The CLABSI rate trends were expressed as CLABSI rates per month and slopes of them during the preintervention and intervention periods. They were calculated and analyzed for statistical significance through autoregression time series analysis and the Wald x2 test was used to compare estimates among the 2 periods. Compliance with each MBP element was compared between the periods using the
ARTICLE IN PRESS J.Y. Park et al. / American Journal of Infection Control 00 (2019) 1−6
2.34 3.07 1.43 1.82
CLABSI ratey
Pearson x2 test with Yates’ continuity correction. A P value of <.05 was considered statistically significant. Data were analyzed using SAS 9.4 (SAS Institute Incorporated, Cary, NC) and Excel 2013 (Microsoft Corporation, Redmond, WA).
3
14 9 2 3 0.51 0.51 0.65 0.42 11,803 5,771 2,155 3,877 5,978 2,935 1,396 1,647 3.41 4.30 1.53 4.68 25 12 4 9 0.49 0.50 0.59 0.40 14,808 5,585 4,435 4,788 7,326 2,790 2,611 1,925
Patient-days Central line-days
CLABSI, central line−associated bloodstream infection; ICU, intensive care unit. *Central line use ratio = central line-days per patient-day. y CLABSI rate = CLABSI events per 1,000 central line-days
Several guidelines recommend multifaceted approaches to CLABSI prevention. These approaches have been effective in both developed and developing countries.1-3,5,6,12-15 However, limited available resources—including human resources as well as financial and managerial support—lead to poor compliance or performance regarding these guidelines.16,17 Given these circumstances, the establishment of
Study site
DISCUSSION
Table 1 CLABSI rates in participating ICUs between preintervention and intervention periods
Compliance with CLABSI prevention bundles
Central line use ratio*
Preintervention (14 months)
CLABSI event
Enterococcus spp were the most common (35.9%), followed by Candida spp (17.9%), Staphylococcus aureus (10.3%), and Acinetobacter baumannii (10.3%). Of the 14 Enterococcus spp, 13 (92.9%) were resistant to penicillin, and 5 (35.7%) were resistant to vancomycin. All Candida spp were susceptible to fluconazole, amphotericin, and voriconazole tested by the VITEK II system. Of the 4 Staphylococcus aureus isolates, 2 were susceptible to oxacillin, and 2 were resistant. All Acinetobacter baumannii and Pseudomonas aeruginosa were resistant to carbapenem.
CLABSI ratey
Study periods
Microbiologic features of CLABSIs in ICUs
The comparisons of compliance rates for hand hygiene, MBPs, and each MBP component between the preintervention and intervention periods are shown in Figure 2. The numbers of compliance measurements between the preintervention and intervention periods were 94 and 317, respectively. The overall MBP compliance rate along with the compliance rates for cap, mask, sterile gown, and full-body drape use improved significantly from the preintervention to intervention period (P < .001). The compliance rates for hand hygiene (84.0%) and sterile glove use before catheter insertion (90.5%) were high during the preintervention period and remained high during the intervention period (96.8% and 98.7%, respectively). The compliance rates for hand hygiene, all MBPs, and each MBP component among the 3 ICUs during the preintervention and intervention periods were not significantly different. The rate of compliance with the maintenance bundle during the intervention period was 95.96%.
Central line-days
Patient-days
Intervention (10 months)
During the 24-month study period, 26,611 patient-days and 13,304 CL days were observed at a total of 49 beds in MICU, SICU, and NICU. The CLABSI rate for overall ICUs during the entire study period was 2.93 per 1,000 catheter-days. The overall CLABSI rate and that of each ICU decreased from the preintervention to the intervention period, even though the CL use ratio increased (Table 1). The most prominent decline in CLABSI rate, from 4.68 to 1.82 per 1,000 catheter-days, was observed in the NSICU (Table 1). The overall CLABSI rates per month as well as the rates per month for each of the 3 ICUs—which were analyzed through autoregressive time series analysis—are shown in Table 2, and the overall CLABSI rate trend is illustrated in Figure 1. The CLABSI rates per month for overall ICUs (0.602; 95% confidence interval, 0.972 to 0.232; P = .008) and the MICU (0.788; 95% confidence interval, 1.479 to 0.098; P = .036) decreased significantly in the intervention period. The difference in CLABSI rate trend between the preintervention and intervention periods was only statistically significant in the SICU (P < .001).
Overall (all ICUs) Medical ICU Surgical ICU Neurosurgical ICU
CLABSI rates in ICUs
Central line use ratio*
CLABSI event
RESULTS
ARTICLE IN PRESS J.Y. Park et al. / American Journal of Infection Control 00 (2019) 1−6
CI, confidence interval; CLABSI, central line−associated bloodstream infection; ICU, intensive care unit; SE, standard error.
P value
.066 .128 <.001 .166 3.39 2.32 14.39 1.92 .008 .036 .086 .240 0.164 0.305 0.284 0.322 0.602 (0.972 to 0.232) 0.788 (1.479 to 0.098) 0.568 (1.21 to 0.075) 0.415 (1.143 to 0.314) 0.165 0.394 0.128 0.288 Overall (all ICUs) Medical ICU Surgical ICU Neurosurgical ICU
0.256 (0.613 to 0.101) 0.160 (1.011 to 0.691) 0.147 (0.423 to 0.129) 0.590 (1.212 to 0.031)
.15 .693 .273 .065
Wald x2 test statistic P value SE Estimate (95% CI)
Preintervention Study site
Table 2 CLABSI rates per month in participating ICUs using autoregressive time series analysis
Study period
Estimate (95% CI)
Intervention
SE
P value
Comparison of estimates between preintervention and intervention periods
4
reasonable medical insurance fees for effective IC programs was suggested based on a national survey that evaluated the current status of IC personnel and infrastructure resources in the Republic of Korea.7 In response to this suggestion, the Korean NHIS began reimbursing IC costs from September 2016 based on the number of ICNs and ICDs per hospital bed. The highest reimbursing standard ratio was 1 ICN per 150 hospital beds and 1 ICD working 20 hours a week for IC per 300 hospital beds or 1 full-time ICD per 600 hospital beds.9 The reimbursement benefit to general hospitals under the highest standard was $2.06 per patient-day in 2017 (Table 3).18 If a 600-bed general hospital hires 1 full-time ICD and 4 full-time ICNs to meet the highest reimbursing standard, a total of $406,026 (0.9ⅹ600ⅹ365ⅹ2.06) per year will be reimbursed to the general hospital assuming a bed operation rate of 90%. In fact, the hospital where this study was conducted had a total of 230,693 patient-days, and $468,307 (2.03ⅹ230,693) was reimbursed in 2017. We believe that this level of reimbursement benefit meets the approximate annual salaries of 1 full-time ICD and 4 full-time ICNs, which are required to maintain the highest reimbursing standard for a general hospital, although there are some variations. This staffing level was similar to or lower than the median ratio of a 2015 European Society of Clinical Microbiology and Infectious Diseases member survey.19 The enhanced ICT achieved their goal of reducing CLABSIs. We believe that strengthening ICT staffing levels will have positive effects on various other IC indicators. The rates of ventilator-associated pneumonia also decreased from 1.23 per 1,000 ventilator-days in 2016 to 0.58 in 2017. Evidence also suggests that ICT activity improves the cost-effectiveness of IC in hospitals.20 The Korean governmental data on the cost of the IC reimbursement policy is shown in Table 3. The overall cost of the IC reimbursement policy was $48,831,297 in 2017 and $59,963,245 in 2018. All tertiary care general hospitals benefited from this policy and took approximately half of the overall cost. However, secondary care general and nongeneral hospitals experienced difficulties hiring ICDs or ICNs to meet these standards, and even though the numbers were much larger, many of them did not benefit from this policy. To address this problem, nongeneral hospitals were more likely to obtain benefits than general hospitals if they met this standard. The reimbursement costs per patients-day according to the hospital grade increased slightly yearly. Although there was an ICT as well as IC programs before this study period, it was difficult to conclude that IC programs were effectively carried out due to insufficient infrastructure, especially in terms of IC staffing levels. At the beginning of the preintervention period, the CLABSI rates were high relative to the national average, and they were decreasing through various efforts, but after the IC staffing level was strengthened, CLABSI rates declined significantly (Fig 1). The role of a full-time ICD was critical because doctor-to-doctor communications were effective, bringing about changes in doctor behavior. In particular, a senior ICD who is looked on respectfully as a figure of authority and leadership is beneficial. In our hospital, similar to the response to the 2015 European Society of Clinical Microbiology and Infectious Diseases member survey,19 a senior infectious diseases doctor undertook the role of the first full-time ICD. CLABSI rates increased at the start of the intervention period. The range of CL use ratios was 0.42 to 0.6, and their values in March and April 2017 were 0.57 and 0.6, respectively, which were the highest and second highest values throughout the entire study periods. These high values might have been owing to the conditions of patients during these periods. Because only patients from the neurosurgical department were admitted to the NSICU, the ICT only needed to communicate with doctors and nurses from the neurosurgical department in the NSICU. In the NSICU, the ICT was able to coordinate CLABSI prevention bundles more effectively than in the MICU and SICU, so the CL use ratio was the lowest and the reduction in the CLABSI rate was the largest,
ARTICLE IN PRESS J.Y. Park et al. / American Journal of Infection Control 00 (2019) 1−6
5
Fig 1. Trends of central line−associated bloodstream infection (CLABSI) rates during preintervention and intervention periods. Dotted lines are the trends of the preintervention and intervention periods, and the slopes are calculated through autoregressive analysis. Graph values are monthly CLABSI rates. Autoregressive analysis revealed decreases of CLABSI rates per month during preintervention and intervention periods of 0.256 (95% confidence interval, 0.613 to 0.101; P=.15) and 0.602 (95% confidence interval, 0.972 to 0.232; P=.008), respectively.
Fig 2. Compliance rates for hand hygiene, MBPs, and MBP components for all intensive care units during preintervention and intervention periods. The asterisks represent statistically significant differences. MBP, maximal barrier precautions. Table 3 Overall cost (calculated by $0.84 per 1,000 KRW of exchange rate) for infection control reimbursement per year according to category of hospitals and grade of reimbursement standards in Korea Hospital category
2017 Number First grade
Tertiary general hospital Secondary general hospital Nongeneral hospital Total
43 301 1,466 1,810
2018
Total reimbursement cost ($)
22,384,163* 20,215,653* 49,154k 42,648,970
Second grade 3,263,289y 2,853,994y 65,044{ 6,182,327
Number Total 25,647,452 23,069,647 114,198 48,831,297
42 311 1,465 1,818
Total reimbursement cost ($) First grade
Second grade
28,107,160z 26,752,004z 206,343# 55,065,507
2,241,704x 2,274,333x 380,701** 4,896,738
Total 30,348,864 29,026,337 587,044 59,962,245
KRW, South Korean Won. The reimbursement costs per patient-days: *$2.03; y$1.67; z$2.06; x$1.69; k$2.44; {$2.06; #$2.44; **$2.09.
although no difference was observed in hand hygiene and MBP performance among the ICUs. In the SICU, the CLABSI rates were the lowest throughout the entire study period, as many patients were admitted for a short period after surgery. It was most difficult for the ICT to implement a multifaceted approach for improving CLABSI prevention
in the MICU because of the array of complex pathologies among the patients admitted to the MICU and transferred from various departments, who were then hospitalized for a long time in the MICU. There were some limitations to our study. First, we conducted this intervention at a single center for a short period; therefore, our
ARTICLE IN PRESS J.Y. Park et al. / American Journal of Infection Control 00 (2019) 1−6
6
findings may not be generalizable to all institutions. Because of the small sample size, differences in estimates between the preintervention and intervention periods were not statistically significant, except for in the SICU. This effect could be limited to secondary care hospitals because several tertiary care hospitals had already reinforced MBP monitoring before the reimbursement policy in association with the development of critical care medicine. Although overall health care− associated infections (HAIs) might also decrease in other hospitals after the implementation of the reimbursement policy, the findings of this research cannot be generalized. Therefore, a nationwide study to evaluate the effectiveness of the IC cost reimbursement policy for the prevention of not only CLABSIs but also various HAIs is warranted. Second, we compared only hand hygiene and MBPs among CLABSI prevention bundles between the preintervention and intervention periods. The maintenance bundle implementation rate was monitored only during the intervention period. Other factors associated with CLABSI rates may have been overlooked. Third, this study was a time series analysis that only captured a short period. Thus, a multicenter study or big data analysis is warranted in the future. CONCLUSIONS The reimbursement policy for IC costs accelerated the decline in CLABSI rates significantly by providing infrastructure and enhancing IC staffing levels to allow for a comprehensive and multifaceted approach to improving the implementation of CLABSI prevention bundles in ICUs. This study showed that national financial and policy support helps control HAIs at a front-line medical facility. However, this policy could have a different effect on each hospital. Therefore, this research cannot be generalized and a nationwide study to evaluate the effectiveness of the IC cost reimbursement policy for various HAIs is warranted. References 1. Ling ML, Apisarnthanarak A, Jaggi N, Harrington G, Morikane K, Thu le TA, et al. APSIC guide for prevention of central line associated bloodstream infections (CLABSI). Antimicrob Resist Infect Control 2016;5:16. 2. Pronovost P, Needham D, Berenholtz S, Sinopoli D, Chu H, Cosgrove S, et al. An intervention to decrease catheter-related bloodstream infections in the ICU. New Eng J Med 2006;355:2725-32. 3. Marschall J, Mermel LA, Fakih M, Hadaway L, Kallen A, O’Grady NP, et al. Strategies to prevent central line-associated bloodstream infections in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol 2014;35:753-71.
4. Guerin K, Wagner J, Rains K, Bessesen M. Reduction in central line-associated bloodstream infections by implementation of a postinsertion care bundle. Am J Infect Control 2010;38:430-3. 5. O’Grady NP, Alexander M, Burns LA, Dellinger EP, Garland J, Heard SO, et al. Guidelines for the prevention of intravascular catheter-related infections. Clin Infect Dis 2011;52:e162-93. 6. Loveday HP, Wilson JA, Pratt RJ, Golsorkhi M, Tingle A, Bak A, et al. epic3: national evidence-based guidelines for preventing healthcare-associated infections in NHS hospitals in England. J Hosp Infect 2014;86(Suppl 1):1-70. 7. Yoon YK, Lee SE, Seo BS, Kim HJ, Kim JH, Yang KS, et al. Current status of personnel and infrastructure resources for infection prevention and control programs in the Republic of Korea: a national survey. Am J Infect Control 2016;44:e189-93. 8. Choi JY, Kwak YG, Yoo H, Lee SO, Kim HB, Han SH, et al. Trends in the incidence rate of device-associated infections in intensive care units after the establishment of the Korean Nosocomial Infections Surveillance System. J Hosp Infect 2015;91:28-34. 9. Health Insurance Review and Assessment Service. Personnel standard for infection control cost reimbursement. Available from: http://www.hira.or.kr/main.do. Accessed April 29, 2019. 10. Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health careassociated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008;36:309-32. 11. Kaier K, Wolkewitz M, Hehn P, Mutters NT, Heister T. The impact of hospitalacquired infections on the patient-level reimbursement-cost relationship in a DRG-based hospital payment system. Int J Health Econ Manag 2019 Jun 5, [Epub ahead of print]. 12. Bion J, Richardson A, Hibbert P, Beer J, Abrusci T, McCutcheon M, et al. ‘Matching Michigan’: a 2-year stepped interventional programme to minimise central venous catheter-blood stream infections in intensive care units in England. BMJ Qual Saf 2013;22:110-23. 13. Berenholtz SM, Lubomski LH, Weeks K, Goeschel CA, Marsteller JA, Pham JC, et al. Eliminating central line-associated bloodstream infections: a national patient safety imperative. Infect Control Hosp Epidemiol 2014;35:56-62. 14. Yaseen M, Al-Hameed F, Osman K, Al-Janadi M, Al-Shamrani M, Al-Saedi A, et al. A project to reduce the rate of central line associated bloodstream infection in ICU patients to a target of zero. BMJ Qual Improv Rep 2016;5. 15. Marsteller JA, Sexton JB, Hsu YJ, Hsiao CJ, Holzmueller CG, Pronovost PJ, et al. A multicenter, phased, cluster-randomized controlled trial to reduce central lineassociated bloodstream infections in intensive care units. Crit Care Med 2012;40:2933-9. 16. Rosenthal VD. Central line-associated bloodstream infections in limited-resource countries: a review of the literature. Clin Infect Dis 2009;49:1899-907. 17. Valencia C, Hammami N, Agodi A, Lepape A, Herrejon EP, Blot S, et al. Poor adherence to guidelines for preventing central line-associated bloodstream infections (CLABSI): results of a worldwide survey. Antimicrob Resist Infect Control 2016;5:49. 18. Health Insurance Review and Assessment Service. Medical statistics in heathcare Bigdata Hub. Available from: https://opendata.hira.or.kr/home.do. Accessed August 2, 2019. 19. Dickstein Y, Nir-Paz R, Pulcini C, Cookson B, Beovic B, Tacconelli E, et al. Staffing for infectious diseases, clinical microbiology and infection control in hospitals in 2015: results of an ESCMID member survey. Clin Microbiol Infect 2016;22, 812.e9e17. 20. Seko T, Tachi T, Kawashima N, Maeda T, Yasuda M, Noguchi Y, et al. Economic evaluation of infection control activities. J Hosp Infect 2017;96:371-6.