American Journal of Infection Control 42 (2014) 12-6
Contents lists available at ScienceDirect
American Journal of Infection Control
American Journal of Infection Control
journal homepage: www.ajicjournal.org
Major article
A multifaceted intervention to reduce rates of catheter-associated urinary tract infections in a resource-limited setting L. Gayani Tillekeratne MD a, *, Darren R. Linkin MD, MSCE a, b, Mariah Obino MBBS c, Afua Omar MBBS c, Mary Wanjiku MBBS c, David Holtzman MD, MSc a, Jennifer Cohn MD, MPH a, b a b c
Hospital of the University of Pennsylvania, Philadelphia, PA Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA School of Medicine, University of Nairobi, Nairobi, Kenya
Key Words: Health care-associated infections Quality improvement interventions Developing settings Infection surveillance and prevention
Background: Health care-associated infections such as catheter-associated urinary tract infections (CAUTIs) are prevalent in resource-limited settings. This study was carried out to determine whether a multifaceted intervention targeting health care personnel would reduce CAUTI rates in a public hospital located in a resource-limited setting. Methods: A one group, pretest-posttest study was carried out from March to July 2012 in a public district hospital in Nairobi, Kenya. Patients admitted to adult medical wards, and who received urinary catheters, were evaluated for symptomatic CAUTIs using a modified definition by the Centers for Disease Control and Prevention. After collecting baseline CAUTI rates for 8 weeks, a multifaceted intervention consisting of lectures, reminder signs, and infection prevention rounds (week 9) was implemented. The postintervention rate of CAUTIs was measured over 7 subsequent weeks. Bivariable analysis was performed to determine whether the intervention was associated with reduced CAUTIs. Results: A total of 125 patients received urinary catheters, with 82 preintervention and 43 postintervention. Mean duration of catheterization did not change between phases (6.9 vs 5.6 days, respectively, P ¼ .322), but catheter utilization ratio decreased from 0.14 to 0.09 (P < .001). There were 13 preintervention CAUTIs (for 30.4 infections per 1,000 catheter-days) and no postintervention CAUTIs (P ¼ .002). Conclusion: In this resource-limited setting, the baseline rate of CAUTIs was high. A low-cost, multifaceted intervention resulted in decreased urinary catheter use and CAUTI rates. Copyright Ó 2014 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
Health care-associated infections (HCAIs) are recognized as the most frequent adverse event in health care, resulting in increased morbidity, mortality, and economic costs for hospitalized patients in the acute care setting.1,2 It is estimated that approximately 1.7 million patients in the United States are affected by HCAIs each year, with approximately 100,000 of these patients dying from complications related to their infections.3 Given the level of morbidity and mortality associated with HCAIs, and studies showing that a large proportion of HCAIs may be preventable using current evidencebased strategies, the United States’ Medicare system stopped reimbursements for the treatment of certain HCAIs starting in October 2008.4 * Address correspondence to L. Gayani Tillekeratne, MD, 5555 Greenridge Drive, Toledo, OH 43615. E-mail address: [email protected] (L.G. Tillekeratne). Conflicts of interest: None to report.
The prevalence of HCAIs globally, especially in developing countries, is less well understood. Available data indicate that rates may be up to 2 to 13 times greater in certain high-risk populations in developing countries.5 Some of this burden may be due to understaffing, which is common in developing countries: a low level of nursing staff is associated with up to a 50% increase in infection risk in the critical care setting.6 In addition, factors such as lack of national infection prevention surveillance, low levels of hand hygiene compliance, and limited funds and resources to allocate toward infection prevention are postulated to play a role in the prevalence of HCAIs in developing countries.7,8 As one of the more prevalent HCAIs in the world, catheterassociated urinary tract infections (CAUTIs) represent a considerable burden of morbidity.9 A recent meta-analysis indicated a CAUTI rate of 6.5 per 1,000 device-days in 18 developing countries’ medical-surgical intensive care units (ICUs), compared with 3.4 per 1,000 device-days in American medical-surgical ICUs.10 The Centers
0196-6553/$36.00 - Copyright Ó 2014 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajic.2013.07.007
L.G. Tillekeratne et al. / American Journal of Infection Control 42 (2014) 12-6
for Disease Control and Prevention (CDC) in the United States, the Society for Healthcare Epidemiology of America, and the World Health Organization have published guidelines on strategies acute care hospitals can utilize to prevent CAUTIs, which include (1) limiting the insertion of urinary catheters to appropriate indications, (2) using aseptic technique in placing urinary catheters and ensuring only trained personnel insert catheters, (3) maintaining urinary catheters aseptically, (4) implementing quality improvement measures and the appropriate clinical infrastructure to prevent CAUTIs, and (5) performing active surveillance for CAUTIs.2,11,12 The implementation of these strategies in the developing world could potentially have a great impact on reducing rates of CAUTIs, without requiring a large expenditure of finances or resources. We developed a multifaceted, low-cost intervention for health care personnel working on the medical wards of a public hospital in Nairobi, Kenya, with the goal of reducing CAUTIs. The intervention targeted nurses and clinical officer interns, who traditionally insert and maintain urinary catheters in this hospital, and was designed to decrease inappropriate urinary catheter use and to improve the care of necessary catheters. The intervention consisted of a lecture series with educational videos, reminder signs posted in the medical wards, and weekly infection prevention rounds by the nurse matrons. It was hypothesized that this multifaceted intervention would result in the reduction of inappropriate urinary catheter use and a decrease in CAUTI rates. If effective and locally acceptable, similar interventions could help decrease the burden of HCAIs in other resource-poor, developing settings. METHODS Setting and study design This 1 group, pretest-posttest study was carried out from March 2012 to July 2012 at a 169-bed, public district hospital in Nairobi, Kenya. The hospital primarily serves patients of a low socioeconomic class. Surveillance activities were carried out in the 4 medical wards of the hospital, which contain a total of 48 male beds and 48 female beds. The University of Pennsylvania Institutional Review Board, Kenyatta National Hospital Ethics and Research Committee, and the Ethical Board of the local hospital all approved this study. Surveillance As part of a quality-improvement activity, active surveillance was carried out on the medical wards to determine the base line rate of symptomatic CAUTIs. Surveillance was conducted over an 8week period from March 2012 to May 2012 and was carried out by members of the study team including 1 infectious diseases physician and 3 local, bilingual research assistants trained in surveillance activities (by L.G.T.). The total number of admitted patients and the total number of patients with urinary catheters were assessed daily during weekdays, at approximately the same time every day. All patients who received a urinary catheter during routine clinical care were monitored daily for signs of symptomatic CAUTIs. Because urine cultures were not routinely performed given patients’ inability to pay for such tests, a modified definition of catheter-associated symptomatic UTIs by the CDC was used: (1) At least 1 of the following signs or symptoms with no other recognized causedfever >38 C; suprapubic tenderness; costovertebral angle pain; or dysuria, frequency, or urgency if the catheter had been removed within 48 hours; and (2) at least 1 of the following: positive dipstick for leukocyte esterase and/or nitrite; pyuria; or microorganisms seen on Gram’s stain of unspun urine.13 The research team, with the assistance of regular clinical personnel,
13
monitored catheterized patients daily for CAUTIs by performing a focused history and physical examination, including check of axillary temperatures. Dipstick urine tests were performed by the research team for patients displaying signs of CAUTIs, with results being shared with the clinical team; in addition, results from urine tests sent for clinical purposes were recorded from the medical chart. Further information regarding patients’ sociodemographic background, clinical course, laboratory data, and treatment history was extracted from patients’ medical records if patients met criteria for symptomatic CAUTI. Intervention During the initial surveillance period, discussions were held with nurses and clinical officer interns regarding the appropriate components of a quality-improvement intervention to decrease rates of CAUTIs. A multifaceted intervention was developed based on these discussions and a review of the literature and consisted of (1) didactic sessions on the placement and management of urinary catheters, to be offered to nurses and clinical officer interns on the medical wards; (2) reminder signs over beds encouraging daily assessment of the need for urinary catheters, as well as reiterating standard medical indications for urinary catheterization; and (3) weekly infection prevention rounds by the nurse matrons evaluating the continued need for urinary catheters in catheterized patients. The didactic sessions were an hour in length and covered standard medical indications for urinary catheter placement, appropriate technique when placing urinary catheters, and appropriate management of urinary catheters.2,11,12 The discussed indications for catheter placement included acute urinary obstruction or retention, need for accurate measurement of input and output in critically ill patients, presence of open sacral or perineal ulcers in patients with urine incontinence, prolonged immobilization, need for continuous urine outflow during perioperative care, and enhancement of comfort in patients receiving endof-life services. During the discussion on appropriate placement of urinary catheters, the use of hand hygiene, sterile gloves, and antisepsis were highlighted. The appropriate management of urinary catheters was described as the use of a closed drainage system with a urine bag, rather than having the urine drain into a bucket underneath the bed; emptying the urine bag every 8 hours or for volumes greater than 400 mL; and leaving the urine bag in a dependent position below the level of the bladder, rather than alongside the patient on the bed. To enhance health care workers’ understanding of these concepts, videos on appropriate technique during urinary catheter placement in both sexes, available online through the New England Journal of Medicine, were shown during the didactic sessions.14 The video on catheter placement in female patients was approximately 15 minutes in length, and the video on catheter placement in male patients was approximately 10 minutes in length. The reminder signs that were posted in the wards, besides encouraging catheter removal, listed the medical indications for urinary catheterization that were covered during the didactic sessions. Following the initial 8 weeks of surveillance, the multifaceted intervention was implemented in the medical wards over a 1-week period. The didactic sessions were offered to nurses and clinical officer interns on 2 different days to maximize participation and drew an average of 22 staff per session. It was estimated that over 90% of targeted staff on the medical wards attended the sessions. Approximately 30 reminder signs were posted in the medical wards, with a sign-to-bed ratio of 1:3. These signs were not removed following the intervention week. Weekly nurse matron rounds began during and continued after the intervention week
14
L.G. Tillekeratne et al. / American Journal of Infection Control 42 (2014) 12-6
and consisted of nurse matrons performing a weekly assessment of the number of urinary catheters in the wards and determining whether any catheters could be removed. Following the week-long implementation of the intervention, active surveillance of CAUTIs was again carried out as part of a quality-improvement activity, with surveillance being performed for 7 additional weeks from May 2012 to July 2012. Outcome measures The primary outcome was the rate of CAUTIs per 1,000 catheterdays. Secondary outcomes included the rate of CAUTIs per 1,000 patient-days and catheter utilization rates (days of catheter use divided by total days of inpatient care).13 These secondary outcomes were measured to detect a decrease in CAUTIs due to a reduction in urinary catheter usage, which may not be apparent from the primary outcome alone. Other exposures recorded included patient sex, age, primary medical condition, indication for urinary catheter placement, presence of urinary incontinence, placement of urinary catheter by nurse versus clinical officer intern, maintenance of urinary catheter by nurse versus clinical officer intern, use of a closed versus open system (with urine draining into a urine bag rather than a bucket under the bed) for urine collection, placement of urinary bag to gravity versus on the patient’s bed, and treatment with antibiotics during hospitalization. Statistical analysis Bivariable analyses were carried out using the t test for continuous variables and the c2 test or Fisher exact test for categorical variables. The primary analysis measured whether the intervention was significantly associated with a decrease in CAUTIs. An additional analysis examined the secondary outcome of CAUTI rates per 1,000 patient-days. RESULTS A total of 125 patients received urinary catheters as part of routine clinical care during the study period, with 82 in the preintervention phase and 43 in the postintervention phase. The baseline characteristics of patients receiving urinary catheters are shown in Table 1. Subjects were generally similar in the pre- and postintervention phases, although the proportion of males was lower in the postintervention phase. Median age was 38 years in the preintervention phase and 32 years in the postintervention phase. HIV was a common comorbid condition, affecting 62.2% of patients in the preintervention phase and 65.1% of patients in the postintervention phase. Meningoencephalitis was the most common primary diagnosis listed in the medical record, with 47.6% of patients in the preintervention phase and 34.9% of patients in the postintervention phase having this diagnosis. The characteristics of catheter placement and maintenance are listed in Table 2. Clinical officer interns placed the majority of catheters in both phases. In this hospital lacking basic incontinence management supplies such as bedpans and urinals, the primary reason for urinary catheter placement during both study periods was decreased mobility and inability to walk to the bathroom. Urinary incontinence was stated as a reason for catheterization in a smaller proportion of patients following the intervention (24.4% vs 4.7%, respectively, P ¼ .003). A larger proportion of catheters were placed to gravity following the intervention (80.5% vs 97.7%, respectively, P ¼ .005). Mean duration of catheterization did not change between the intervention periods, from 6.9 days to 5.6 days (P ¼ .322), but catheter utilization ratio decreased from 0.14 to 0.09 (P < .001).
Table 1 Demographic and clinical characteristics of patients receiving urinary catheters during routine clinical care in the pre- and postintervention periods Preintervention period
Male sex, n (%) Age, y, median (range) Most common primary diagnoses, n (%)* Gastroenteritis Meningoencephalitis Cryptococcal meningitis Pneumonia Tuberculosis Comorbid conditions-HIV
Postintervention period
N ¼ 82
N ¼ 43
P value
38 (46.3) 36 (19-67)
10 (23.3) 32 (18-78)
.002 .919
9 39 8 9 14 51
4 15 2 8 15 28
.722 .095 .256 .111 .002 .693
(11.0) (47.6) (9.8) (11.0) (17.1) (62.2)
(9.3) (34.9) (4.7) (18.6) (34.9) (65.1)
*Many patients had more than 1 primary diagnosis.
There were 13 CAUTIs during the preintervention period for 30.4 infections per 1,000 catheter-days and no CAUTIs in the postintervention period (P ¼ .002, see Table 3). The number of CAUTIs per 1,000 patient-days also decreased from the pre- to postintervention period: 4.6 CAUTIs per 1,000 patient-days versus no CAUTIs per 1,000 patient-days, respectively, P ¼ .028. A total of 79 (96.3%) patients was treated with antibiotics for any medical indication in the preintervention phase, and 40 (93.0%) patients were treated with antibiotics in the postintervention phase. Of patients with CAUTIs, only 1 had a urinary tract infection listed as a reason for antibiotic therapy in the medical record, but all patients with CAUTIs were receiving antibiotic therapy for some medical indication. DISCUSSION Our results showed that the baseline rate of CAUTIs on the medical wards of this public district hospital was high, at 30.4 infections per 1,000 catheter-days. A low-cost, multifaceted intervention consisting of a lecture series with videos, reminder signs, and infection prevention rounds by the nurse matrons was wellreceived by health care personnel, and resulted in a significant decrease in the rate of CAUTIs to zero infections in the postintervention phase. The secondary outcomes of CAUTIs per 1,000 patient-days and catheter utilization rates also decreased significantly following the intervention. Process measures including the placement of catheters to gravity, the use of a closed urinary collection system, and the avoidance of urinary catheter placement for incontinence improved following the intervention. The baseline rate of CAUTIs in this study was greater than that generally reported for developed or even other developing settings. A recent meta-analysis of data from 226 ICUs in developing countries revealed a pooled density of 9.8 infections per 1,000 catheterdays compared with 30.4 infections per 1,000 catheter-days in our study.1 However, the studies included in the meta-analysis were all from the critical care setting and showed a high level of heterogeneity. There is a paucity of data on HCAIs from general medical settings in the developing world, and data that are available vary in quality.5 Our study adds to the current body of literature available on CAUTI rates in general medical settings and indicates that these rates may be quite high. The primary and secondary outcomes of CAUTI rates and catheter utilization rates decreased significantly following the intervention. Two other published studies document the implementation of similar practices in the developing world, with a subsequent reduction in CAUTIs. A study by Rosenthal et al showed that a multidimensional infection prevention strategy
L.G. Tillekeratne et al. / American Journal of Infection Control 42 (2014) 12-6 Table 2 Catheter characteristics in the pre- and postintervention periods with appropriate P values
Indications for catheterization* Urinary incontinence Urinary retention Need for accurate ins/outs Open sacral/perineal ulcer Decreased mobility Catheter characteristics Indwelling catheter To gravity Closed system Catheter placement by RN Clinical officer intern Other/unknown
Preintervention period, n (%)
Postintervention period, n (%)
N ¼ 82
N ¼ 43
20 3 4 2 75
(24.4) (3.7) (4.9) (2.4) (91.5)
2 2 1 1 39
P value
(4.7) (4.7) (2.3) (2.3) (90.7)
.003 .741 .434 .975 .850
60 (73.2) 66 (80.5) 76 (92.7)
38 (88.4) 42 (97.7) 42 (97.7)
.025 .005 .210
20 (24.4) 60 (73.2) 2 (2.4)
3 (7.0) 39 (90.7) 1 (2.33)
.008 .010 .975
*Many patients had more than 1 indication for catheter placement.
including components such as education and training, hand hygiene and aseptic technique, outcome and process surveillance, and feedback resulted in a significant reduction in CAUTIs in adult ICUs in 15 developing countries.15 Another study by Marra et al showed that the implementation of similar measures in 1 ICU and 2 step-down units in Brazil resulted in a significant reduction in CAUTI rates.9 Both these studies were carried out in the critical care or intermediate care settings. To our knowledge, our study is the first that documents the feasibility of such a multifaceted intervention in a general medical setting of a public hospital located in a resource-limited region. In addition, our intervention only included low-cost measures consisting of education, increased surveillance, and reminder systems, with no increased commodity measures. The extent of decrease in CAUTI rates was unexpected given the high level of morbidity within this patient population. However, the improvement in process measures indicates that this may have contributed to the large decrease. This study revealed that baseline practices in urinary catheter placement and management could be improved in this general medical setting. Simple practices such as closing the urinary collection system and placing the catheter to gravity, which could reduce the rate of CAUTIs, were not always followed. The indications for catheter placement were frequently not strictly medical, with urinary incontinence and decreased mobility of the patient being the most common reasons for catheter placement. Following the intervention, a greater percentage of catheters were placed to gravity and used a closed collection system, suggesting that these improved process measures contributed to the decrease in CAUTI rates. In addition, incontinence was listed less frequently as an indication for catheter placement, although decreased mobility continued to be the primary reason for catheterization. In this setting lacking incontinence management supplies such as bedpans and urinals, patients who were unable to ambulate to the bathroom had urinary catheters placed for management of soilage. Given a lack of nursing resources for frequent bed linen changes, health care staff were concerned about poor hygiene and skin breakdown in such patients. This highlights the need for appropriately allocating and prioritizing funds when purchasing hospital supplies in resource-poor settings. Estimates for 2008-2009 from the Kenya Medical Supplies Authority, which supplies medical equipment and pharmaceutical products to public health care facilities in Kenya, indicated that a disposable urinary catheter cost $0.34 and a 2-L urine bag cost $0.20.16 Estimates for reusable plastic bedpans and urinals were not available from this source. However, even if such devices cost twice as much as a urinary
15
Table 3 Primary and secondary outcomes of CAUTI and catheter utilization rates Preintervention period
Mean catheterization duration Catheter utilization ratio CAUTIs per 1,000 catheter-days CAUTIs per 1,000 patient-days
Postintervention period
N ¼ 82
N ¼ 43
P value
6.9 days 0.14 30.4 4.6
5.6 days 0.09 0 0
.322
Contents lists available at ScienceDirect
American Journal of Infection Control
American Journal of Infection Control
journal homepage: www.ajicjournal.org
Major article
A multifaceted intervention to reduce rates of catheter-associated urinary tract infections in a resource-limited setting L. Gayani Tillekeratne MD a, *, Darren R. Linkin MD, MSCE a, b, Mariah Obino MBBS c, Afua Omar MBBS c, Mary Wanjiku MBBS c, David Holtzman MD, MSc a, Jennifer Cohn MD, MPH a, b a b c
Hospital of the University of Pennsylvania, Philadelphia, PA Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA School of Medicine, University of Nairobi, Nairobi, Kenya
Key Words: Health care-associated infections Quality improvement interventions Developing settings Infection surveillance and prevention
Background: Health care-associated infections such as catheter-associated urinary tract infections (CAUTIs) are prevalent in resource-limited settings. This study was carried out to determine whether a multifaceted intervention targeting health care personnel would reduce CAUTI rates in a public hospital located in a resource-limited setting. Methods: A one group, pretest-posttest study was carried out from March to July 2012 in a public district hospital in Nairobi, Kenya. Patients admitted to adult medical wards, and who received urinary catheters, were evaluated for symptomatic CAUTIs using a modified definition by the Centers for Disease Control and Prevention. After collecting baseline CAUTI rates for 8 weeks, a multifaceted intervention consisting of lectures, reminder signs, and infection prevention rounds (week 9) was implemented. The postintervention rate of CAUTIs was measured over 7 subsequent weeks. Bivariable analysis was performed to determine whether the intervention was associated with reduced CAUTIs. Results: A total of 125 patients received urinary catheters, with 82 preintervention and 43 postintervention. Mean duration of catheterization did not change between phases (6.9 vs 5.6 days, respectively, P ¼ .322), but catheter utilization ratio decreased from 0.14 to 0.09 (P < .001). There were 13 preintervention CAUTIs (for 30.4 infections per 1,000 catheter-days) and no postintervention CAUTIs (P ¼ .002). Conclusion: In this resource-limited setting, the baseline rate of CAUTIs was high. A low-cost, multifaceted intervention resulted in decreased urinary catheter use and CAUTI rates. Copyright Ó 2014 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
Health care-associated infections (HCAIs) are recognized as the most frequent adverse event in health care, resulting in increased morbidity, mortality, and economic costs for hospitalized patients in the acute care setting.1,2 It is estimated that approximately 1.7 million patients in the United States are affected by HCAIs each year, with approximately 100,000 of these patients dying from complications related to their infections.3 Given the level of morbidity and mortality associated with HCAIs, and studies showing that a large proportion of HCAIs may be preventable using current evidencebased strategies, the United States’ Medicare system stopped reimbursements for the treatment of certain HCAIs starting in October 2008.4 * Address correspondence to L. Gayani Tillekeratne, MD, 5555 Greenridge Drive, Toledo, OH 43615. E-mail address: [email protected] (L.G. Tillekeratne). Conflicts of interest: None to report.
The prevalence of HCAIs globally, especially in developing countries, is less well understood. Available data indicate that rates may be up to 2 to 13 times greater in certain high-risk populations in developing countries.5 Some of this burden may be due to understaffing, which is common in developing countries: a low level of nursing staff is associated with up to a 50% increase in infection risk in the critical care setting.6 In addition, factors such as lack of national infection prevention surveillance, low levels of hand hygiene compliance, and limited funds and resources to allocate toward infection prevention are postulated to play a role in the prevalence of HCAIs in developing countries.7,8 As one of the more prevalent HCAIs in the world, catheterassociated urinary tract infections (CAUTIs) represent a considerable burden of morbidity.9 A recent meta-analysis indicated a CAUTI rate of 6.5 per 1,000 device-days in 18 developing countries’ medical-surgical intensive care units (ICUs), compared with 3.4 per 1,000 device-days in American medical-surgical ICUs.10 The Centers
0196-6553/$36.00 - Copyright Ó 2014 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajic.2013.07.007
L.G. Tillekeratne et al. / American Journal of Infection Control 42 (2014) 12-6
for Disease Control and Prevention (CDC) in the United States, the Society for Healthcare Epidemiology of America, and the World Health Organization have published guidelines on strategies acute care hospitals can utilize to prevent CAUTIs, which include (1) limiting the insertion of urinary catheters to appropriate indications, (2) using aseptic technique in placing urinary catheters and ensuring only trained personnel insert catheters, (3) maintaining urinary catheters aseptically, (4) implementing quality improvement measures and the appropriate clinical infrastructure to prevent CAUTIs, and (5) performing active surveillance for CAUTIs.2,11,12 The implementation of these strategies in the developing world could potentially have a great impact on reducing rates of CAUTIs, without requiring a large expenditure of finances or resources. We developed a multifaceted, low-cost intervention for health care personnel working on the medical wards of a public hospital in Nairobi, Kenya, with the goal of reducing CAUTIs. The intervention targeted nurses and clinical officer interns, who traditionally insert and maintain urinary catheters in this hospital, and was designed to decrease inappropriate urinary catheter use and to improve the care of necessary catheters. The intervention consisted of a lecture series with educational videos, reminder signs posted in the medical wards, and weekly infection prevention rounds by the nurse matrons. It was hypothesized that this multifaceted intervention would result in the reduction of inappropriate urinary catheter use and a decrease in CAUTI rates. If effective and locally acceptable, similar interventions could help decrease the burden of HCAIs in other resource-poor, developing settings. METHODS Setting and study design This 1 group, pretest-posttest study was carried out from March 2012 to July 2012 at a 169-bed, public district hospital in Nairobi, Kenya. The hospital primarily serves patients of a low socioeconomic class. Surveillance activities were carried out in the 4 medical wards of the hospital, which contain a total of 48 male beds and 48 female beds. The University of Pennsylvania Institutional Review Board, Kenyatta National Hospital Ethics and Research Committee, and the Ethical Board of the local hospital all approved this study. Surveillance As part of a quality-improvement activity, active surveillance was carried out on the medical wards to determine the base line rate of symptomatic CAUTIs. Surveillance was conducted over an 8week period from March 2012 to May 2012 and was carried out by members of the study team including 1 infectious diseases physician and 3 local, bilingual research assistants trained in surveillance activities (by L.G.T.). The total number of admitted patients and the total number of patients with urinary catheters were assessed daily during weekdays, at approximately the same time every day. All patients who received a urinary catheter during routine clinical care were monitored daily for signs of symptomatic CAUTIs. Because urine cultures were not routinely performed given patients’ inability to pay for such tests, a modified definition of catheter-associated symptomatic UTIs by the CDC was used: (1) At least 1 of the following signs or symptoms with no other recognized causedfever >38 C; suprapubic tenderness; costovertebral angle pain; or dysuria, frequency, or urgency if the catheter had been removed within 48 hours; and (2) at least 1 of the following: positive dipstick for leukocyte esterase and/or nitrite; pyuria; or microorganisms seen on Gram’s stain of unspun urine.13 The research team, with the assistance of regular clinical personnel,
13
monitored catheterized patients daily for CAUTIs by performing a focused history and physical examination, including check of axillary temperatures. Dipstick urine tests were performed by the research team for patients displaying signs of CAUTIs, with results being shared with the clinical team; in addition, results from urine tests sent for clinical purposes were recorded from the medical chart. Further information regarding patients’ sociodemographic background, clinical course, laboratory data, and treatment history was extracted from patients’ medical records if patients met criteria for symptomatic CAUTI. Intervention During the initial surveillance period, discussions were held with nurses and clinical officer interns regarding the appropriate components of a quality-improvement intervention to decrease rates of CAUTIs. A multifaceted intervention was developed based on these discussions and a review of the literature and consisted of (1) didactic sessions on the placement and management of urinary catheters, to be offered to nurses and clinical officer interns on the medical wards; (2) reminder signs over beds encouraging daily assessment of the need for urinary catheters, as well as reiterating standard medical indications for urinary catheterization; and (3) weekly infection prevention rounds by the nurse matrons evaluating the continued need for urinary catheters in catheterized patients. The didactic sessions were an hour in length and covered standard medical indications for urinary catheter placement, appropriate technique when placing urinary catheters, and appropriate management of urinary catheters.2,11,12 The discussed indications for catheter placement included acute urinary obstruction or retention, need for accurate measurement of input and output in critically ill patients, presence of open sacral or perineal ulcers in patients with urine incontinence, prolonged immobilization, need for continuous urine outflow during perioperative care, and enhancement of comfort in patients receiving endof-life services. During the discussion on appropriate placement of urinary catheters, the use of hand hygiene, sterile gloves, and antisepsis were highlighted. The appropriate management of urinary catheters was described as the use of a closed drainage system with a urine bag, rather than having the urine drain into a bucket underneath the bed; emptying the urine bag every 8 hours or for volumes greater than 400 mL; and leaving the urine bag in a dependent position below the level of the bladder, rather than alongside the patient on the bed. To enhance health care workers’ understanding of these concepts, videos on appropriate technique during urinary catheter placement in both sexes, available online through the New England Journal of Medicine, were shown during the didactic sessions.14 The video on catheter placement in female patients was approximately 15 minutes in length, and the video on catheter placement in male patients was approximately 10 minutes in length. The reminder signs that were posted in the wards, besides encouraging catheter removal, listed the medical indications for urinary catheterization that were covered during the didactic sessions. Following the initial 8 weeks of surveillance, the multifaceted intervention was implemented in the medical wards over a 1-week period. The didactic sessions were offered to nurses and clinical officer interns on 2 different days to maximize participation and drew an average of 22 staff per session. It was estimated that over 90% of targeted staff on the medical wards attended the sessions. Approximately 30 reminder signs were posted in the medical wards, with a sign-to-bed ratio of 1:3. These signs were not removed following the intervention week. Weekly nurse matron rounds began during and continued after the intervention week
14
L.G. Tillekeratne et al. / American Journal of Infection Control 42 (2014) 12-6
and consisted of nurse matrons performing a weekly assessment of the number of urinary catheters in the wards and determining whether any catheters could be removed. Following the week-long implementation of the intervention, active surveillance of CAUTIs was again carried out as part of a quality-improvement activity, with surveillance being performed for 7 additional weeks from May 2012 to July 2012. Outcome measures The primary outcome was the rate of CAUTIs per 1,000 catheterdays. Secondary outcomes included the rate of CAUTIs per 1,000 patient-days and catheter utilization rates (days of catheter use divided by total days of inpatient care).13 These secondary outcomes were measured to detect a decrease in CAUTIs due to a reduction in urinary catheter usage, which may not be apparent from the primary outcome alone. Other exposures recorded included patient sex, age, primary medical condition, indication for urinary catheter placement, presence of urinary incontinence, placement of urinary catheter by nurse versus clinical officer intern, maintenance of urinary catheter by nurse versus clinical officer intern, use of a closed versus open system (with urine draining into a urine bag rather than a bucket under the bed) for urine collection, placement of urinary bag to gravity versus on the patient’s bed, and treatment with antibiotics during hospitalization. Statistical analysis Bivariable analyses were carried out using the t test for continuous variables and the c2 test or Fisher exact test for categorical variables. The primary analysis measured whether the intervention was significantly associated with a decrease in CAUTIs. An additional analysis examined the secondary outcome of CAUTI rates per 1,000 patient-days. RESULTS A total of 125 patients received urinary catheters as part of routine clinical care during the study period, with 82 in the preintervention phase and 43 in the postintervention phase. The baseline characteristics of patients receiving urinary catheters are shown in Table 1. Subjects were generally similar in the pre- and postintervention phases, although the proportion of males was lower in the postintervention phase. Median age was 38 years in the preintervention phase and 32 years in the postintervention phase. HIV was a common comorbid condition, affecting 62.2% of patients in the preintervention phase and 65.1% of patients in the postintervention phase. Meningoencephalitis was the most common primary diagnosis listed in the medical record, with 47.6% of patients in the preintervention phase and 34.9% of patients in the postintervention phase having this diagnosis. The characteristics of catheter placement and maintenance are listed in Table 2. Clinical officer interns placed the majority of catheters in both phases. In this hospital lacking basic incontinence management supplies such as bedpans and urinals, the primary reason for urinary catheter placement during both study periods was decreased mobility and inability to walk to the bathroom. Urinary incontinence was stated as a reason for catheterization in a smaller proportion of patients following the intervention (24.4% vs 4.7%, respectively, P ¼ .003). A larger proportion of catheters were placed to gravity following the intervention (80.5% vs 97.7%, respectively, P ¼ .005). Mean duration of catheterization did not change between the intervention periods, from 6.9 days to 5.6 days (P ¼ .322), but catheter utilization ratio decreased from 0.14 to 0.09 (P < .001).
Table 1 Demographic and clinical characteristics of patients receiving urinary catheters during routine clinical care in the pre- and postintervention periods Preintervention period
Male sex, n (%) Age, y, median (range) Most common primary diagnoses, n (%)* Gastroenteritis Meningoencephalitis Cryptococcal meningitis Pneumonia Tuberculosis Comorbid conditions-HIV
Postintervention period
N ¼ 82
N ¼ 43
P value
38 (46.3) 36 (19-67)
10 (23.3) 32 (18-78)
.002 .919
9 39 8 9 14 51
4 15 2 8 15 28
.722 .095 .256 .111 .002 .693
(11.0) (47.6) (9.8) (11.0) (17.1) (62.2)
(9.3) (34.9) (4.7) (18.6) (34.9) (65.1)
*Many patients had more than 1 primary diagnosis.
There were 13 CAUTIs during the preintervention period for 30.4 infections per 1,000 catheter-days and no CAUTIs in the postintervention period (P ¼ .002, see Table 3). The number of CAUTIs per 1,000 patient-days also decreased from the pre- to postintervention period: 4.6 CAUTIs per 1,000 patient-days versus no CAUTIs per 1,000 patient-days, respectively, P ¼ .028. A total of 79 (96.3%) patients was treated with antibiotics for any medical indication in the preintervention phase, and 40 (93.0%) patients were treated with antibiotics in the postintervention phase. Of patients with CAUTIs, only 1 had a urinary tract infection listed as a reason for antibiotic therapy in the medical record, but all patients with CAUTIs were receiving antibiotic therapy for some medical indication. DISCUSSION Our results showed that the baseline rate of CAUTIs on the medical wards of this public district hospital was high, at 30.4 infections per 1,000 catheter-days. A low-cost, multifaceted intervention consisting of a lecture series with videos, reminder signs, and infection prevention rounds by the nurse matrons was wellreceived by health care personnel, and resulted in a significant decrease in the rate of CAUTIs to zero infections in the postintervention phase. The secondary outcomes of CAUTIs per 1,000 patient-days and catheter utilization rates also decreased significantly following the intervention. Process measures including the placement of catheters to gravity, the use of a closed urinary collection system, and the avoidance of urinary catheter placement for incontinence improved following the intervention. The baseline rate of CAUTIs in this study was greater than that generally reported for developed or even other developing settings. A recent meta-analysis of data from 226 ICUs in developing countries revealed a pooled density of 9.8 infections per 1,000 catheterdays compared with 30.4 infections per 1,000 catheter-days in our study.1 However, the studies included in the meta-analysis were all from the critical care setting and showed a high level of heterogeneity. There is a paucity of data on HCAIs from general medical settings in the developing world, and data that are available vary in quality.5 Our study adds to the current body of literature available on CAUTI rates in general medical settings and indicates that these rates may be quite high. The primary and secondary outcomes of CAUTI rates and catheter utilization rates decreased significantly following the intervention. Two other published studies document the implementation of similar practices in the developing world, with a subsequent reduction in CAUTIs. A study by Rosenthal et al showed that a multidimensional infection prevention strategy
L.G. Tillekeratne et al. / American Journal of Infection Control 42 (2014) 12-6 Table 2 Catheter characteristics in the pre- and postintervention periods with appropriate P values
Indications for catheterization* Urinary incontinence Urinary retention Need for accurate ins/outs Open sacral/perineal ulcer Decreased mobility Catheter characteristics Indwelling catheter To gravity Closed system Catheter placement by RN Clinical officer intern Other/unknown
Preintervention period, n (%)
Postintervention period, n (%)
N ¼ 82
N ¼ 43
20 3 4 2 75
(24.4) (3.7) (4.9) (2.4) (91.5)
2 2 1 1 39
P value
(4.7) (4.7) (2.3) (2.3) (90.7)
.003 .741 .434 .975 .850
60 (73.2) 66 (80.5) 76 (92.7)
38 (88.4) 42 (97.7) 42 (97.7)
.025 .005 .210
20 (24.4) 60 (73.2) 2 (2.4)
3 (7.0) 39 (90.7) 1 (2.33)
.008 .010 .975
*Many patients had more than 1 indication for catheter placement.
including components such as education and training, hand hygiene and aseptic technique, outcome and process surveillance, and feedback resulted in a significant reduction in CAUTIs in adult ICUs in 15 developing countries.15 Another study by Marra et al showed that the implementation of similar measures in 1 ICU and 2 step-down units in Brazil resulted in a significant reduction in CAUTI rates.9 Both these studies were carried out in the critical care or intermediate care settings. To our knowledge, our study is the first that documents the feasibility of such a multifaceted intervention in a general medical setting of a public hospital located in a resource-limited region. In addition, our intervention only included low-cost measures consisting of education, increased surveillance, and reminder systems, with no increased commodity measures. The extent of decrease in CAUTI rates was unexpected given the high level of morbidity within this patient population. However, the improvement in process measures indicates that this may have contributed to the large decrease. This study revealed that baseline practices in urinary catheter placement and management could be improved in this general medical setting. Simple practices such as closing the urinary collection system and placing the catheter to gravity, which could reduce the rate of CAUTIs, were not always followed. The indications for catheter placement were frequently not strictly medical, with urinary incontinence and decreased mobility of the patient being the most common reasons for catheter placement. Following the intervention, a greater percentage of catheters were placed to gravity and used a closed collection system, suggesting that these improved process measures contributed to the decrease in CAUTI rates. In addition, incontinence was listed less frequently as an indication for catheter placement, although decreased mobility continued to be the primary reason for catheterization. In this setting lacking incontinence management supplies such as bedpans and urinals, patients who were unable to ambulate to the bathroom had urinary catheters placed for management of soilage. Given a lack of nursing resources for frequent bed linen changes, health care staff were concerned about poor hygiene and skin breakdown in such patients. This highlights the need for appropriately allocating and prioritizing funds when purchasing hospital supplies in resource-poor settings. Estimates for 2008-2009 from the Kenya Medical Supplies Authority, which supplies medical equipment and pharmaceutical products to public health care facilities in Kenya, indicated that a disposable urinary catheter cost $0.34 and a 2-L urine bag cost $0.20.16 Estimates for reusable plastic bedpans and urinals were not available from this source. However, even if such devices cost twice as much as a urinary
15
Table 3 Primary and secondary outcomes of CAUTI and catheter utilization rates Preintervention period
Mean catheterization duration Catheter utilization ratio CAUTIs per 1,000 catheter-days CAUTIs per 1,000 patient-days
Postintervention period
N ¼ 82
N ¼ 43
P value
6.9 days 0.14 30.4 4.6
5.6 days 0.09 0 0
.322
