Rates of catheter-associated urinary tract infection (CAUTI) are significantly higher in low- and middle-income countries (LMICs) compared to high-income countries. Reference Rosenthal, Maki and Salomao1,Reference Rosenthal, Duszynska and Ider2 A report from the International Nosocomial Infection Control Consortium (INICC) showed that the CAUTI rate in LMICs was 3.16 CAUTIs per 1,000 UC days. Reference Rosenthal, Duszynska and Ider2 A report from the US Centers for Disease Control and Prevention National Healthcare Safety Network (NHSN) reported 1.3 CAUTIs per 1,000 urinary catheter (UC) days. Reference Dudeck, Edwards and Allen-Bridson3
Recent studies showed that CAUTI is an independent and significant risk factor for mortality in the ICU. Reference Rosenthal, Yin and Lu4–Reference Rosenthal, Yin and Valderrama-Beltran6 Mortality among ICU patients without any healthcare-associated infection (HAI) is 17.1%; mortality is 30.15% among patients with CAUTI and 63.4% among those with for CAUTI plus central-line–associated bloodstream infections plus ventilator-associated pneumonia. Reference Rosenthal, Duszynska and Ider2 CAUTIs are associated with additional costs of $589 per infection. Reference Tambyah, Knasinski and Maki7
Studies identified the following variables as risk factors for CAUTI: female sex, Reference Farsi8 age >50 years, Reference Anggi, Wijaya and Ramayani9 increased days of catheterization, Reference Juanjuan10 increased length of stay in ICU, Reference Lalitha, Paul, Nagraj and Ghosh11 following a urological surgical procedure, Reference Jimenez-Alcaide, Medina-Polo and Garcia-Gonzalez12 mobility issues, Reference Letica-Kriegel, Salmasian and Vawdrey13 diabetes, Reference Li, Song, Xu, Deng, Zhu and Li14 hypertension, Reference Perrin, Vats and Qureshi15 and spinal cord lesions, Reference Jiang, Deng, Zhou, Li and Liu16 among others.
However, no study has concurrently examined different LMICs or different UC types to determine CAUTI risk factors in ICUs. Furthermore, no prospective study has been conducted over 8 years. Additionally, no study has examined simultaneously the relationships between the following 11 variables and their association with CAUTI: (1) age, (2) sex, (3) length of stay (LOS) prior to CAUTI acquisition, (4) UC days prior to CAUTI acquisition, (5) UC device utilization (DU) ratio as a marker of patient illness severity, (6) UC types, (7) hospitalization type, (8) ICU type, (9) facility ownership, (10) World Bank country classification by income level, and (11) period.
In this study, we report CAUTI incidence rates stratified by country, by region, by ICU type, by facility ownership type, by World Bank country classification by income level, and by UC type. We also sought to determine whether the aforementioned 12 variables were CAUTI risk factors.
Materials and methods
Study population and design
This multinational, multicenter, cohort, prospective study was carried out with patients admitted to 623 ICUs of 224 hospitals in 114 cities in 37 countries of Africa, Asia, Eastern Europe, Latin America, and the Middle East between January 1, 2014, and February 12, 2022.
Prospective cohort surveillance of healthcare-associated infections
Participants in the study were hospitals that were members of INICC. Each patient’s data were compiled at the time of ICU admission. Infection prevention professionals (IPPs) visited each patient’s bedside daily from the time of admission until discharge. The INICC Surveillance Online System (ISOS) was used to gather data on all prospectively included patients who were admitted to an ICU. IPPs carry a tablet to each hospitalized patient’s bedside in the ICU, sign in to the ISOS, and upload patient data in real time. Reference Rosenthal17
The data collected at the time of patient admission contain information about the location, such as the setting, country, city, admission date, ICU type, as well as patient data, sex, age, hospitalization type, and invasive device use. Until patient discharge, IPPs upload data on the patient’s invasive devices and positive cultures.
The institutional review boards of the participating hospitals approved this study. The names of the hospital and the patients remain confidential.
INICC surveillance online system
INICC CAUTI surveillance is carried out using an online platform, the ISOS, which includes CDC/NHSN criteria and methods. 18 In addition, the ISOS collects patient-specific information on all patients, with and without HAI. Reference Rosenthal17 To estimate CAUTI risk factors, data from all patients admitted to the ICU allow matching by various characteristics.
Training of infection prevention professionals
This training of IPPs consists of 4 meetings in which the INICC team reviews how to conduct surveillance and how to upload it to the ISOS. In addition, videos with the same content as the webinar are provided to IPPs. IPPs will also view PDF tutorials that have the same information as the webinar. The IPPs are always able to reach the INICC team by phone, text message, WhatsApp, FaceTime, and/or email with their questions. IPPs upload surveillance data during the training and simultaneously share a screen with the INICC team to let the INICC team check the accuracy of each step of the process. The INICC team trains IPPs to generate a report of surveillance using the ISOS. IPPs generate a report during the training and simultaneously share the screen with the INICC team to let the INICC team check the accuracy of each step of the process. The IPPs also create a PDF report and send it to the INICC team by email. The INICC team will make the same report at the INICC office to compare the 2 reports and find any mistakes in how the graphs were made or processed.
Data validation
When IPPs upload the results of the culture to the ISOS, the ISOS promptly displays a notice and directs the IPP to an ISOS online module where they can check all the CDC HAI criteria to verify the presence and type of HAI. Reference Rosenthal17 Each day, the ISOS checks UC DU. When a bias on patient days or device usage is discovered from admission to discharge, the ISOS notifies the IPPs. If the ISOS detects a lack of use of any kind of device on any given day, it sends a message to the IPPs to remind them to upload any missing devices or upload the discharge of the patient. This approach significantly reduces biases associated with UC days, UC DU ratio, patient days, and discharge conditions. Reference Rosenthal17
Study definitions
Healthcare-associated infection
The CDC definitions of CAUTI in 2014 and all subsequent updates through 2022 were utilized during surveillance. 18 The CDC definitions of catheter-associated urinary tract infections are available in the Supplementary Materials (online). The current and updated CDC definitions of HAIs have been used by all IPPs of all participant hospitals over the 8 years of this study. That is, our IPPs started using the newly revised definitions whenever the CDC updated them. 18 The CDC NHSN definitions exclude patients with suprapubic catheters from CAUTI surveillance. Suprapubic catheter–associated UTIs were defined using the CDC NHSN definitions but were applied to patients with suprapubic catheters.
Indwelling urethral urinary catheter
A drainage tube that is inserted into the urinary bladder through the urethra, is left in place, and is connected to a drainage bag (including leg bags). These devices are also often called Foley catheters. Indwelling urethral urinary catheters that are used for intermittent or continuous irrigation are also included in CAUTI surveillance. 18
Suprapubic catheter
Suprapubic catheterization refers to the placement of a drainage tube into the urinary bladder just above the pubic symphysis. Reference Corder and LaGrange19
Urinary catheter device utilization ratio
Urinary catheter device utilization ratio (UC DU) was calculated as a ratio of UC days to patient days for each location type. As such, the UC DU of a location measures the use of invasive devices and constitutes an extrinsic CAUTI risk factor. The UC DU ratio also served as a marker for the severity of illness of patients, which is an intrinsic risk factor for HAI. 18
World Bank country classifications by income level
The World Bank categorizes nations into 4 income groups based on their economies: low-, lower–middle-, upper–middle-, and high-income countries. The classifications are based on the gross national income (GNI) per capita in the current USD. The GNI of low-income nations is <$1,045 USD. Lower–middle class are those having a GNI between $1,046 and $4,095. Those with a GNI between $4,096 and $12,695 have an upper–middle income. Those with high income have a GNI >$12,695. 20
Facility and institution ownership type
Publicly owned facilities are owned or controlled by a public corporation or a governmental body, where control is the capacity to decide on the corporate strategy. Not-for-profit, privately owned facilities are legal or social organizations established for the exclusive goal of creating goods and services, whose legal position prohibits them from serving as a source of revenue, profit, or other financial gains for the unit(s) that established, controlled, or financed them. For-profit, privately owned facilities are legal organizations created to produce goods and services with the potential to bring in a profit or other financial gains for their owners. 21
Statistical analysis
To estimate the incidence of CAUTI per 1,000 UC days, we divided the number of CAUTIs by number of UC days and multiplied the result by 1,000. We identified CAUTI rates per 1,000 UC days and UC DU ratios stratified by country, by ICU type, by facility ownership type, by World Bank country classification by income level, and by UC type.
To estimate risk for CAUTI, patients with and without CAUTI were compared using multiple logistic regression. Statistically significant variables were associated independently with an increased risk for CAUTI. The Wald test was employed as the test statistic, and a 2-tailed type 1 error rate of .05 was chosen as the level of statistical significance. The adjusted odds ratios (aORs) and associated 95% confidence intervals (CIs) for statistically significant factors were calculated from the results of multiple logistic regression.
We analyzed the following 11 independent variables and its association with the outcome (CAUTI): (1) age; (2) sex; (3) LOS before acquiring a CAUTI; (4) UC days before acquisition of CAUTI; (5) UC DU ratio as a marker of severity of illness of patient; (6) type of UC (suprapubic, external, indwelling urethral); (7) hospitalization type (medical or surgical); (8) ICU type (ie, cardiothoracic, neurologic, neurosurgical, adult oncology, medical, medical-surgical, pediatric, respiratory, surgical, trauma, coronary, or pediatric oncology); (9) facility ownership (publicly owned; not-for-profit, privately owned; for-profit, privately owned; or teaching hospitals) 21 ; (10) income level per country according to World Bank (lower–middle, upper–middle, or high) 20 ; and (11) period (period 1: 2014–2016, period 2: 2017–2019, period 3: 2020–2022). The evaluated outcome was the acquisition of CAUTI according to the CDC NHSN definitions. 18
For analysis of CAUTI risk factors we use data of 37 countries: Argentina, Bahrain, Brazil, Bulgaria, Colombia, Costa Rica, Dominican Republic, Ecuador, Egypt, India, Jordan, Kosovo, Kuwait, Lebanon, Macedonia, Malaysia, Mexico, Mongolia, Morocco, Nepal, Pakistan, Panama, Papua New Guinea, Peru, Philippines, Poland, Romania, Russia, Saudi Arabia, Serbia, Slovakia, Sri Lanka, Thailand, Turkey, United Arab Emirates, Venezuela, and Vietnam. All statistical analyses were performed using R software version 4.1.3 (R Foundation for Statistical Computing, Vienna, Austria).
Results
From January 1, 2014, to February 12, 2022, more than 8 years, a multinational, multicenter, cohort, prospective surveillance study of CAUTIs was conducted across 623 ICUs of 224 hospitals in 114 cities in 37 countries from Africa, Asia, Eastern Europe, Latin America, and the Middle East, currently participating in INICC.
In this cohort study, the length of participation of hospitals was variable, ranging from 1.1 to 226.07 months (mean, 38.47; standard deviation [SD], 42.62). Table 1 shows data on setting and patient characteristics. Table 2 shows CAUTI rate per 1,000 UC days stratified by country, by region, by ICU type, by facility ownership type, by World Bank country classifications by income level, and by urinary catheter type. Low-income countries were not included in this study; only middle-income countries and high-income countries were included. Figure 1 shows the CAUTI rate per 1,000 UC days stratified per country.
Figure 1. Rates of CAUTI per 1,000 urinary catheter-days stratified per country.
Table 1. Setting and Patient Characteristics a
Note. ICU, intensive care unit; UC, urinary catheter; LOS, length of stay; CAUTI, catheter-associated urinary tract infections; SD, standard deviation.
a Data collected from January 1, 2014, to February 12, 2022.
Table 2. CAUTI Rates Stratified by Country, by Region, by ICU Type, by Facility Ownership Type, by World Bank Country Classifications by Income Level, and by Urinary Catheter Type
Note. ICU, intensive care unit; CI, confidence interval; UC, urinary catheter; CAUTI, catheter-associated urinary tract infection; CI, confidence interval.
a Countries are listed alphabetically.
b Rate of catheter associated urinary tract infection per 1,000 urinary catheter days.
c Regions are listed in order of the highest to lowest CAUTI rate.
d ICUs are listed in order of the highest to lowest CAUTI rate.
The pooled CAUTI rate per 1,000 UC days was 2.83. The highest CAUTI rates occurred in patients with suprapubic catheters; in patients hospitalized in Eastern European facilities; in Asian facilities hospitalized at trauma, neurologic, and neurosurgical ICUs; in patients hospitalized at facilities of middle-income countries; and in patients hospitalized at publicly owned facilities. Countries with the lowest CAUTI rates were Costa Rica, Thailand, Kuwait, Lebanon, and Colombia. The countries with the highest CAUTI rates were Morocco, Pakistan, Romania, Papua New Guinea, and Serbia (Table 2).
Using multiple logistic regression, the following variables were identified as significantly associated with CAUTI (Table 3): age, with risk increasing 1% yearly; female sex; LOS prior to acquisition of CAUTI, with risk increasing 4% daily; number of UC days prior to acquisition of CAUTI, with risk increasing 1% per UC day; UC DU ratio; hospitalized at a publicly owned facility; location in an upper–middle-income country, and period. The ICU type with highest risk was neurologic, followed by neurosurgical, adult oncology, medical, trauma, surgical, medical-surgical, pediatric, respiratory, and coronary ICU. After adjusting by all confounders, type of UC, surgical hospitalization, teaching hospital, and for-profit, privately owned facility were not associated with CAUTI risk.
Table 3. Multiple Logistic Regression Analysis of Risk Factors for CAUTI
Note. ICU, intensive care unit; UC, urinary catheter; DU, device utilization; LOS, length of stay; CAUTI, catheter-associated urinary tract infection; aOR, adjusted odds ratio; CI, confidence interval.
Discussion
We identified CAUTI rates stratified by ICU type, by country, by region, by income level according to the World Bank, by facility ownership, and by type of urinary catheter. The following variables were independently and significantly associated with risk for CAUTI: age, female sex, public owned facility, middle-income country, neurologic ICU, and period.
The pooled rates of CAUTI in our study are similar to those pooled CAUTI rates reported by the INICC. Reference Rosenthal, Duszynska and Ider2 The CAUTI rate in LMICs is 3.16 CAUTIs per 1,000 UC days according to the last INICC report. Reference Rosenthal, Duszynska and Ider2 However, pooled rates of CAUTI in our study are higher than those of the CDC NHSN, which reports 1.3 CAUTI per 1,000 UC days. Reference Dudeck, Edwards and Allen-Bridson3
In our study, we identified age as a risk factor for CAUTI. Similarly, in a study by Liu et al Reference Liu, Li and Huang22 at a neurosurgical ICU, age >60 years was identified as a risk factor for CAUTI. We identified sex as a risk factor for CAUTI. Likewise, in a study by Perrin et al Reference Perrin, Vats and Qureshi15 at a neurologic ICU, female sex was identified as a risk factor for CAUTI. The incremental risk of CAUTI increased by 5% per day if the UC remains in place. Consistently, a study conducted by Al-Hazmi Reference Al-Hazmi23 showed the role of LOS prior to CAUTI acquisition as a significant risk factor for CAUTI. The UC DU was identified as a risk factor for CAUTI. Similarly, as stated by Burton et al, Reference Burton, Edwards, Srinivasan, Fridkin and Gould24 the dominant risk factor for acquiring a CAUTI was the duration of UC days.
We identified that the CAUTI rate using suprapubic catheters was higher than that for patients with indwelling urethral catheters. However, when we applied multiple logistic regression to identify a type of UC as an independent risk factor for CAUTI, both types of UC have similar risk, as shown by the overlap of the 95% CIs. Conversely, according to Gibson et al, Reference Gibson, Neill, Tuma, Meddings and Mody25 suprapubic catheters had a lower CAUTI incidence rate compared to indwelling urethral catheters.
In this study, the highest CAUTI rate was seen in patients hospitalized in trauma, neurologic, and neurosurgical ICUs. When applying multiple logistic regression, the study noted that admission into the neurologic, trauma, and neurosurgical ICUs had the highest risk for CAUTI. The UC DU ratio was the highest for the corresponding ICUs according to data collected by hospitals participating in the National Healthcare Safety Network (NHSN) and reported to the US Centers for Disease Control and Prevention (CDC), and a higher UC DU ratio is associated with a higher risk of CAUTI. Reference Dudeck, Edwards and Allen-Bridson3
For patients hospitalized at publicly owned facilities, the CAUTI rate was the highest; for those at for-profit, privately owned facilities, it was intermediate; and for those at teaching hospitals, it was the lowest. Applying multiple logistic regression, patients admitted to publicly owned facilities had a significantly higher risk factor for CAUTI than patients admitted to other types of facilities. This finding was not consistent with a previous study that found that teaching hospitals had a CAUTI rate similar to publicly owned facilities and for-profit, privately owned facilities. Reference Rosenthal, Jarvis and Jamulitrat26
Furthermore, for patients hospitalized in Eastern European facilities, the CAUTI rate was the highest; for those in Asian facilities, it was intermediate; and patients in Middle Eastern and Latin American facilities had the lowest CAUTI rate. We identified and showed those countries with higher CAUTI rates and those with the lowest CAUTI rate, and this was associated with the income of the country. Meanwhile, if the income was lower, the CAUTI rate was higher. 20 The current study found that the CAUTI rate in lower–middle-income countries and upper–middle-income countries was similar, but both were higher than for those hospitalized in high-income countries. When applying multiple logistic regression, the risk of CAUTI was higher in upper-middle-income countries compared with high-income countries. This finding is consistent with a previous study in which lower–middle income countries had higher CAUTI rates than upper–middle income countries, showing that lower income is associated with a higher CAUTI rate, but in this particular study, high-income countries were not included. Reference Rosenthal, Jarvis and Jamulitrat26 Analyzing the period, we discovered that the risk for CAUTI decreased over time, which is consistent with more recent improvements in infection prevention techniques than previously.
Some of the CAUTI risk factors identified in our study are unlikely to change, such as age, sex, the income level of the country, facility ownership, and ICU type. However, some of the risk factors for CAUTI we identified can be modified, for example, LOS prior to acquisition of a CAUTI, and UC utilization. Based on our findings, we should focus on strategies to reduce UC utilization, to reduce LOS, and to implement an evidence-based set of CAUTI prevention recommendations, such as those published by HICPAC. Reference Gould, Umscheid and Agarwal27 Also, the very high rate of CAUTI prevalent in LMICs Reference Rosenthal, Maki and Salomao1,Reference Rosenthal, Duszynska and Ider2 can be reduced by utilizing a strategy of monitoring compliance with recommendations and providing performance feedback to healthcare personnel, as demonstrated in several LMICs. Reference Rosenthal, Guzman and Safdar28–Reference Rosenthal, Todi and Alvarez-Moreno33
Our study had several limitations. First, this study is not representative of all hospitals in LMICs because it is a component of a surveillance system in which hospitals voluntarily participate for free. Second, because the hospitals that participate in our surveillance system are likely the ones that have a higher-quality CAUTI surveillance and prevention program, the CAUTI rates in our study were presumably lower than the CAUTI rates in other hospitals that did not participate in our study. Third, we did not stratify hospital by impact of bed size, services offered, and specialty services. Instead, we stratified them by type of ICU, facility ownership, income level according to the World Bank, by country, and by region, which are more relevant for LMICs, as previously demonstrated. Reference Rosenthal, Jarvis and Jamulitrat26,Reference Rosenthal, Lynch and Jarvis34 Lastly, severity illness scores were not collected by the IPPs of the participating institutions; instead, we used the UC DU ratio as a marker for severity of patient illness.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/ice.2023.215
Acknowledgments
The INICC Advisory Board, Country Directors, and Secretaries have generously supported this exceptional global infection control network, for which we are grateful. We also thank the numerous healthcare professionals who helped with surveillance in their hospitals.
Financial support
No financial support was provided relevant to this article.
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
