Results
Over the whole study period, there were 3,877 patients who were hospitalized for 27,345 days in 10 PICUs for a total of 8,513 UC-days. The first PICUs to participate in the study began collecting data in June 2003, and the latest data included in this analysis are from December 2010. The data for participating hospitals were summarized and classified according to the number of PICUs, number of patients per PICU, type of hospital, and country. The majority of enrolled patients were from academic teaching hospitals (53%), followed by private community hospitals (53%). The remaining 6% of enrolled patients were from a public hospital in Mexico. All participating hospitals are from countries with developing economies. Seventy-two percent of enrolled patients belonged to countries with lower-middle-income economies (India, 37%; El Salvador, 29%; Philippines, 6%). Twenty-eight percent of enrolled patients were from countries with upper-middle-income economies (Turkey, Colombia, and Mexico; Table 1).
Patient characteristics, such as sex, underlying diseases, previous infection, and duration of UC use, were similar during the baseline and intervention phases (Table 2). With respect to infection prevention and control practices, we found that hand hygiene compliance improved significantly in phase 2, from 48% to 70% (Table 2).
Regarding CAUTI rates, we recorded 9 CAUTIs in phase 1 (baseline period), for an overall baseline rate of 5.9 CAUTIs per 1,000 UC-days. During phase 1, there were 1,513 documented UC-days, for a UC use mean of 0.32.
In phase 2, there were 7,000 UC-days, for a UC use mean of 0.31. After the implementation of the INICC multidimensional infection control approach, we recorded 18 CAUTIs, for an incidence density of 2.6 cases per 1,000 UC-days.
These results showed a CAUTI rate reduction from baseline of 57% (from 5.9 to 2.6 CAUTIs per 1,000 UC-days; RR, 0.43 [95% CI, 0.21–1.0];P = .0344; Table 2).
The microorganism profile shows that Candida species, which accounted for 50% of isolates, was the most isolated uropathogen, with no variation in its frequency in both periods (phase 1 and phase 2). It was followed by Enterococcus species (25% of isolates) in phase 1. The remaining pathogens accounted for a maximum of 13% in both periods (Table 3).
Data are quite sparse, with few CAUTI in each period of follow-up. The most notable result is that, despite over 1,000 admissions to ICUs with over 2 years of participation in the INICC, no CAUTI was recorded in this period. The stratified rate ratios for the follow-up period show a decrease in all time periods, although the confidence intervals are wide and include 1 for all periods. Excluding the baseline period, when the time that an ICU had participated in the INICC at the time of admission was included as a continuous variable, we continued to see a decrease in CAUTI rate of 17% for every 3-month period that an ICU has participated in the INICC (nonsignificant; Table 4).