key: cord-0988494-qbfd3crr
authors: Jeong, Sun Young; Choi, JeongHwa; Kim, Jae Yeun; Ga, Hyuk
title: Development and Application of a Surveillance Method for Healthcare-Associated Infections in Long-Term Care Hospitals in Korea
date: 2020-12-28
journal: Ann Geriatr Med Res
DOI: 10.4235/agmr.20.0067
sha: c2260dd7a83906f6dbb6a50cba0ec25a4c17609d
doc_id: 988494
cord_uid: qbfd3crr

BACKGROUND: This study developed a surveillance method for healthcare-associated infections (HAIs) in long-term care hospitals (LTCHs) and investigated the current status of HAIs in LTCHs in Korea. METHODS: We applied the HAI-related surveillance criteria for long-term care facilities developed by McGeer in six LTCHs. RESULTS: The 197 confirmed HAIs corresponded to incidence rates of 30.38/100 inpatients and 1.57/1,000 days of hospitalization and included 84 cases of respiratory tract infection (43.8%), 78 cases of systemic infection (40.6%), 24 cases of gastrointestinal tract infection (12.5%), and 6 cases of skin and soft tissue mucosal infection (2.1%). The subtypes included 78 cases of unexplained febrile illness (40.6%); 40 cases of pneumonia (20.8%); 27 cases of lower respiratory tract infection (14.1%); 21 cases of gastroenteritis (10.9%); 9 cases of influenza-like illness (4.7%); 8 cases of common cold or pharyngitis (4.2%); 4 cases of cellulitis, soft tissue, or wound infection (2.1%); 3 cases of Clostridium difficile infection (1.6%); 1 case of conjunctivitis (0.5%); and 1 case of fungal oral/perioral and skin infection (0.5%). CONCLUSION: Establishing an HAI surveillance method for LTCHs and identifying HAI rates and risk factors among LTCH patients may help prevent HAIs in LTCHs in Korea.

than half of these patients do not present with fever, 6) and HAIs manifest differently from those observed in acute care facilities. Therefore, the diagnostic criteria for HAIs in LTCHs differ from those used in acute care facilities. 6, 7) The US Centers for Disease Control and Prevention (CDC) developed a national standard for infection surveillance in long-term care (LTC) facilities 8) using the HAI-related surveillance criteria for LTC facilities developed by McGeer via a Delphi technique using a panel comprising infection specialists, gerontologists, and infection control personnel. 7, 9) LTC facilities in the United States are defined as those that provide medical and non-medical support and care to older adults and vulnerable individuals who cannot live independently in the community. These facilities include nursing homes, skilled nursing facilities, LTCHs, intermediate/chronic care facilities for the developmentally disabled, assisted-living facilities, and residential care facilities. 10) Particularly, LTCHs provide medical treatment and rehabilitation treatment to individuals with chronic and complex problems requiring long-term hospital-level care. 10) LTCHs in Korea have mixed functions that are equivalent to LTCHs in the United States, which provide long-term treatment, and to LTC facilities, which focus on providing medical and non-medical support and care. 11) The McGeer HAI diagnostic criteria were developed for patients in LTC facilities who have difficulty undergoing blood and imaging tests. In 2012, the surveillance criteria for urinary tract infections (UTIs) and respiratory tract infections (RTIs) were revised to increase the specificity for these infections, and the definitions for norovirus gastroenteritis and Clostridium difficile infection were newly added to minimize the gap between the diagnostic criteria used by acute care and LTC facilities. 9, 12) In Korea, the Korean Society for Healthcare-associated Infection Control and Prevention (KOSHIC) and the Korea CDC implemented an HAI surveillance method in 2006 primarily for intensive care unit and surgical site infections surveillance. 13) However, LTCHs have lower percentages of severely ill and surgical patients but a higher percentage of older long-term inpatients compared to acute care facilities. Thus, it is inappropriate to conduct HAI surveillance based on the nationwide HAI criteria focused on acute care facilities. Furthermore, surveillance data should be obtained using valid and reliable surveillance criteria so that the data can be compared to those of other healthcare facilities in Korea and HAI surveillance results from other countries. Thus, it is necessary to develop and assess the surveillance criteria for HAIs in LTCHs in Korea.

This study aimed to develop an HAI surveillance method appropriate for LTCHs that primarily admit older patients requiring LTC and to investigate the current status of HAIs in LTCHs in Ko-rea to systematize and standardize HAI surveillance and contribute to prevention of HAI in such facilities.

This study was approved by the Institutional Review Board of Konyang University (No. KYU-2019-282-01). Written informed consent was obtained from the participating institution and the patient's physician. After checking the willingness to participate from the heads of the nursing department in LTC hospitals that completed the "Infection control in LTC hospitals" program administered by the Human Resource Development Institute for Health and Welfare, six LTC hospitals (198 to 522 beds) from the Seoul, Incheon, and Daejeon areas were selected via snowball sampling. One of these hospitals had to withdraw from the surveillance program during the study and was replaced by another LTC hospital in July 2000.

One or two infection control personnel from each participating organization were designated as infection surveillance personnel. A total of nine such personnel were provided group education and training that included briefings about the purpose, outline, infection surveillance criteria and method, case-based training, and questions and answers.

Based on the HAI surveillance standard for LTC facilities developed by the US CDC, 9) the surveillance report was modified and its content validity was reviewed by two infection control nurses, two nursing professors, one infectious disease professor, one gerontology professor, and three infection control personnel in LTCHs through one face-to-face and two e-mail meetings. The standard 8) was updated 9) in 2012 by the CDC to enhance the specificity and positive predictive value through a 2009 review of literature by the Society for Healthcare Epidemiology of America and Long-Term Care Special Interest Group and expert review based on the HAI surveillance criteria developed by McGeer 7) in 1991, in consideration of the nature of LTCHs in which blood tests, body fluid tests, and imaging tests cannot be performed. HAIs were classified into five categories: namely RTI, UTI, skin and soft tissue infections (SSTIs), gastrointestinal tract infection (GTI), and systemic infection. These categories were further classified into 16 subtypes as follows: RTI (pneumonia, lower RTI, influenza-like illness, common cold, or pharyngitis), UTI (catheter or no catheter), SSTIs (cellulitis/soft tissue/wound infections, scabies, fungal oral/perioral and skin infections, herpes, conjunctivitis), GTI (gastroenteritis, norovirus gastroenteritis, C. difficile infection), and systemic infection (primary bloodstream infection, unexplained febrile illness). Although systemic infection was not included in the 2012 CDC criteria, 9) we included it in the present study to improve the sensitivity of surveillance as LTCHs have little experience in surveillance studies, and the McGeer criterion was newly applied (Supplement A). In terms of fever, the axillary criterion was also added, which is commonly used in Korea. The activities of daily living (ADL) index was changed to the ADL scoring system in the patient assessment form that is used as the standard in LTC hospitals and is submitted to the Health Insurance Review and Assessment (HIRA) service monthly. 1) In addition, while the McGeer criteria consisted of seven items rated on a 4-point scale, for a total score of 28, the criteria for use in Korea consisted of 10 items rated on a 5-point scale, for a total score of 50 (Table 1) .

Surveillance was conducted based on reviews of medical charts and/or direct inspection of patients at least weekly for 3 months from July 1 to September 30, 2019, in five of the six LTCHs, and for 2 months from August 1 to September 30, 2019, in the other hospital. When the infection surveillance personnel could not perform active surveillance, nurses in the wards entered the infection criteria (Table 1 ) and symptoms to be monitored (Table 2 ) on the major symptoms form and reported them to the surveillance personnel, who then determined whether the patient had an HAI; thus, both passive and active surveillance were performed. Each case was recorded in a checklist for infection surveillance criteria and patient state. The authors and infection surveillance personnel then determined whether each case met the diagnostic criteria.

Data for the denominators were the electronic records of the monthly numbers of inpatients, lengths of hospital stays, and the numbers of days of catheter use or were computed by summing the daily numbers.

The characteristics of the six participating LTCHs are presented in Table 3 . The hospitals were based in Daejeon (n = 3), Seoul (n = 2), and Incheon (n = 1). A mean of 257 beds were under surveillance (range, 198-522 beds), with 1,801 total beds ( Table 3) . The surveillance periods were 3 months for five hospitals and 2 A new 3-point increase in total ADL score (range, 0-50) from baseline based on the following 10 ADL items, each scored from 0 (independent) to 5 (total dependence): dressing, face washing, teeth brushing, bathing, eating, position changing, sitting up, transfer from bed to chair, locomotion out of the room, and toilet use. Behavior fluctuation (e.g., coming and going or changing in severity during the assessment).

Residents have difficulty in focusing their attention (e.g., unable to keep track of discussions or are easily distracted).

e)

The thinking of the resident is incoherent (e.g., rambling conversation, unclear flow of ideas, or unpredictable switches in subject).

f)

The level of consciousness of the resident is different from baseline (e.g., hyper-alert, sleepy, drowsy, difficult to arouse, nonresponsive).

Cited from ADL items within the monthly recorded inpatient data set, which is composed of 10 evaluation questions about basic ADLs of long-term care inpatients.

www.e-agmr.org 

The incidence of HAI per 100 inpatients was 30.4%. The HAI rate per 1,000 days of hospital stay was 1.57, with rates of 0.81, 1.10, 3.20, 1.62, 1.20, and 1.60, respectively, in each of the six hospitals (Table 4) .

A total of 192 HAIs were confirmed during the study period (Table 5 ). These included 84 cases of RTI (43.8%), 78 cases of systemic infection (40.6%), 24 cases of GTI (12.5%), and 6 cas- Only 260 beds were monitored, excluding single rooms and rehabilitation wards. 

Of the 192 patients who developed HAI, 103 (53.6%) were men and 89 (46.4%) were women. The mean age was 78.0 years, and the highest number of patients were in their 80s (n = 88, 45.8%). The time from admission to HAI diagnosis was 398.6 ± 517.5 days ( Table 6 ). adequate manpower and resources for infection surveillance, 7, 9) for use in Korea. The modified version differed from the CDC's HAI surveillance method standard for LTC facilities in that the diagnostic criteria included systemic infections due to inadequate HAI surveillance experience and data in LTCHs, axillary temperature criterion for fever because of its common use in LTCHs in Korea, and 10 items rated on a 5-point scale based on the ADL scoring system reported to the HIRA by LTCHs in Korea. 1) In addition, considering the reality in LTCHs in which infection control personnel cannot spare adequate time for infection surveillance, we performed passive surveillance, in which nurses providing direct care to patients checked for symptoms included in the HAI surveillance diagnostic criteria, such as fever, increased leukocyte count, altered consciousness, reduced functioning, urinary www.e-agmr.org and bowel dysfunctions, skin and mucosal lesions, and digestive symptoms, and periodically reported to the infection control personnel. While active infection surveillance is the most effective method to investigate HAIs, combining passive infection surveillance amid the inadequate infection surveillance infrastructure in LTCHs in Korea was one difference from the nationwide infection surveillance method focusing on acute care facilities. 13) During the recent coronavirus disease 2019 (COVID-19) pandemic, infection control has become critical in LTCHs that generally provide care to older patients. The COVID-19 guidelines for LTCHs developed by the Korean Geriatrics Society are a good example of infection control. 14) In addition, as caregiving staff, including nurses, actively monitor suspected COVID-19 symptoms, such as respiratory symptoms, fever, and digestive symptoms, we recommend a systematic combination of passive infection surveillance, in which the infection surveillance personnel are given a report, and active infection surveillance, which involves a direct investigation by the infection surveillance personnel.

A total of 192 cases of HAIs were reported by six LTCHs, corresponding to an incidence rate of 30.38%. This rate differed from that previously reported in LTC facilities (3.4% and 4.1%) 15, 16) using the 2012 updated version of the McGeer criteria. 9, 12) This difference could be attributed to the addition of unexplained febrile illness in our study. Another explanation may be the high prevalence of frailty among LTCH patients who are more physically vulnerable compared to those in other nursing homes. 17) Rothan-Tondeur et al. 16) reported bronchitis (35.5%) to be the most common type of HAI, followed by gastroenteritis (23.8%), UTI (16.7%), otorhinolaryngological infection (8.7%), and pneu- Values are presented as number (%) or mean±standard deviation. monia (7.0%). In our study, the most common type of HAI was unexplained febrile infection (40.6%), followed by pneumonia (20.8%), lower RTI (14.1%), and gastroenteritis (10.9%). The high rate of unexplained febrile illness could explain the higher infection rate compared to that reported previously. In the future, the rate of unexplained febrile illness may be lowered by identifying the cause of fever. In our study, the incidence of HAI per 1,000 days of hospital stay was 1.57. The National Healthcare Safety Network (NHSN), an infection surveillance module for LTC facilities in the United States, includes UTI, C. difficile and multi-resistant bacteria infection, and hand hygiene and protective device compliance surveillance. 18) Notably, it was difficult to directly compare the results obtained from the surveillance method used in our study to that used for US infection surveillance due to the differences in these methods. Yet, the NHSN reported an incidence of 0.59 per 1,000 days of hospital stays for UTI and 0.98 per 1,000 days of hospital stays for C. difficile infection between 2013 and 2015 19) while we observed no cases of UTI and three cases of C. difficile infection (0.0024 per 1,000 days of hospital stay). These two types of infections may be difficult to diagnose, as microbial testing is needed for diagnosing them; however, LTCHs are not equipped with the resources for their own microbial testing, and the process of requesting it at an external laboratory is time-and labor-intensive. The McGeer criteria for UTI did not require microbial testing at the time of development in 1991; however, microbial testing was added in the 2012 update to increase the specificity to decrease unnecessary use of resources. 12) As the resource utilization group system was used as the payment model in LTCHs in Korea, 2) it was practically difficult to prescribe microbial testing. Thus, it was difficult to apply the McGeer criteria for UTI based on microbial testing.

In our study, RTI was the second-most common type of HAI after unexplained febrile illness. According to the European data applying the same McGeer criteria, 20) UTI was the most common infection in LTC facilities, followed by RTI and skin infection. In Korea, the diagnosis of UTI is limited by the difficulty in obtaining prescriptions for microbial tests; thus, in our study, RTI was a more frequently diagnosed HAI among patients under long-term care in LTCHs because the diagnosis does not require microbial tests. Furthermore, respiratory symptoms are the main symptoms of the current COVID-19 pandemic, 21) highlighting the importance of RTI prevention in LTCHs. Furthermore, Quach et al. 22) reported that patients aged 65 years or older who visited the emergency department of an acute care facility had risks of developing RTIs and digestive tract infections. Since patients are frequently transferred between LTCHs and acute care facilities, collective efforts are needed to prevent HAIs. This study has some limitations. First, the findings are not generalizable because only six LTCHs that chose to participate were monitored for a relatively short period of 2-3 months. Second, there is a potential bias as some HAIs are underdiagnosed because of the difficulty in obtaining prescriptions for microbial tests for diagnoses. However, all cases were reviewed twice and the authors and the infection surveillance personnel in each LTCH tried to improve the accuracy and sensitivity of cases with periodic online Q&A discussions and sharing of ambiguous cases. Third, LTCH-specific medical fee regulation of the resource utilization group payment system could have constricted microbiological laboratory studies such as urine or blood culture. In the future, improvements in manpower, health insurance systems, and surveillance systems are needed for more LTCHs to continuously participate in HAI surveillance.

In conclusion, the surveillance method applied in our study can be used to assess the infection control methods and level of infection control activities in LTCHs. Particularly, the HAI diagnostic criteria, surveillance form, and checklist that were developed based on the characteristics of inpatients under LTC could contribute substantially to the systematization and standardization of HAI surveillance in LTCHs. Standardizing the surveillance method and investigating the infection rates and infection risk factors will help to prevent HAIs in LTCHs.

Medical guidelines for long-term care hospitals for senior citizens

Categorization of long term care hospital in Korea using cluster analysis

Characteristics and mortality risk factors in geriatric hospital patients visiting one region-wide emergency department

Clinical finding and antimicrobial resistance in bacteremia associated geriatric hospital. Korean

The burden of healthcare-associated infections in European long-term care facilities

Clinical practice guideline for the evaluation of fever and infection in older adult residents of long-term care facilities: 2008 update by the Infectious Diseases Society of America

Definitions of infection for surveillance in long-term care facilities

APIC-HICPAC surveillance definitions for home health care and home hospice infections

Surveillance definitions of infections in long-term care facilities: revisiting the McGeer criteria

Identifying core competencies of infection control nurse specialists in Hong Kong

National Healthcare Safety Network. Long term care facility component: tracking infections in long-term care facilities

Centers for Disease Control and Prevention; c2018

A significant step forward: new definitions for surveillance of infections in long-term care

Korean Society for Healthcare-associated Infection Control and Prevention. KONIS Manual

Korean National Healthcare-associated Infections Surveillance System; c2016

The Korean Geriatrics Society COVID-19 Strategy for Older Adults

Healthcare-associated infections point prevalence survey and antimicrobials use in acute care hospitals (PPS 2016-2017) and long-term care facilities (HALT-3): a comprehensive report of the first experience in Molise Region, Central Italy, and targeted intervention strategies

Infections in nursing homes: is it time to revise the McGeer criteria?

Use of the frailty index and FRAIL-NH scale for the assessment of the frailty status of elderly individuals admitted in a long-term care hospital in Korea

National Healthcare Safety Network. Long term care facility component: tracking infections in long-term care facilities

GA: Centers for Disease Control and Prevention; c2020

The National Healthcare Safety Network long-term care facility component early reporting experience

European point prevalence survey on healthcare associated infections and antibiotic use in long-term care facilities

Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China

Risk of infection following a visit to the emergency department: a cohort study

The authors thank the directors of hospitals and heads of nursing departments who participated in the surveillance of healthcare-associated infections.

The researchers claim no conflicts of interest. 

Supplementary materials can be found via http://doi.org/10. 4235/agmr.20.0067