Positive culture data from all hospitals were collected on specimens taken between 1 January 2017 to 31 March 2021. All hospitals reported data on BSI and UTI.
A total of 9,685 positive cultures (1,988 bloodstream and 7,697 urine) from 8,194 patients were included in the final analysis. The overall median age of the patient cohort with positive UTI cultures was 74 (quartile range 60–84), and 65% (4,968/7,697) were female. The median age of the cohort with positive BSI cultures was 66 (quartile range 53–76) and 38% (757/1,988) were female. The mean monthly number of occupied bed days combined in the pre COVID-19 cohort was 75,317 compared to 73,157 for the COVID-19 cohort. All sites reported a notable drop in occupied bed days in April 2020, but by June 2020 numbers had returned to similar pre COVID-19 numbers (Supplementary Figure S1).
Hospital A contributed the most culture positive episodes with 4,792, followed by Hospital B 2,943 episodes, Hospital E 1,614 episodes, Hospital C 230 episodes and Hospital D 106 episodes. The unadjusted incidence rates for all HAIs in the pre-COVID-19 cohort was 25.5 per 10,000 OBDs (95%CI:24.9–26.1) and in the COVID-19 cohort was 25.1 per 10,000 OBDs (95%CI:24.1–26.1). (Table 1) Sensitivity analysis on the influence of each site on combined BSI and UTI infections demonstrated that hospital A had a significant downward influence in the pre-COVID-19 cohort (p = 0.008), and Hospitals B and E had a significant upward influence on the COVID-19 cohort (p = 0.009 and p < 0.001 respectively. (Supplementary Figure S2).
Table 1
Unadjusted incidence rates per 10,000 occupied bed days (OBDs).
|
Pre-COVID-19 cohort (Jan 2017 – Feb 2020)
|
COVID-19 cohort (Mar 2020 – Mar 2021)
|
|
Number
|
Incidence per 10,000 OBDs (95%CI)
|
Number
|
Incidence per 10,000 OBDs (95%CI)
|
Bloodstream cultures
|
1,518
|
5.3 (5.0-5.6)
|
470
|
4.9 (4.5–5.4)
|
Urinary tract cultures
|
5,781
|
20.2 (19.7–20.7)
|
1,916
|
20.1 (19.2–21.1)
|
Total
|
7,299
|
25.5 (24.9–26.1)
|
2,386
|
25.1 (24.1–26.1)
|
Occupied bed days
|
2,864,089
|
--
|
951,042
|
--
|
OBDs – Occupied bed days |
95%CI – 95% Confidence intervals |
Differences in laboratory reporting nomenclature, and small numbers of certain species, resulted in the grouping of several species for analysis, such as Escherichia species, Staphylococcus species and Candida species (Supplementary Table S2). Escherichia species were also the most frequent for BSI and UTI (Table 2).
Table 2
Frequency of most common organisms by healthcare associated infection type*
Bloodstream infection (n = 2,313#)
|
Urinary tract infection (n = 8,573#)
|
Organism
|
Number
|
%
|
Organism
|
Number
|
%
|
Escherichia species
|
339
|
14.7
|
Escherichia species
|
2556
|
29.8
|
Staphylococcus species
|
224
|
9.7
|
Enterococcus species
|
1258
|
14.7
|
Klebsiella species
|
203
|
8.8
|
Candida species
|
1206
|
14.1
|
Enterococcus species
|
200
|
8.6
|
Pseudomonas species
|
683
|
8
|
Candida species
|
176
|
7.6
|
Klebsiella species
|
647
|
7.5
|
VRE species
|
174
|
7.5
|
VRE species
|
423
|
4.9
|
MSSA
|
157
|
6.8
|
Proteus species
|
377
|
4.4
|
Enterobacter species
|
143
|
6.2
|
Enterobacter species
|
329
|
3.8
|
Pseudomonas species
|
140
|
6.1
|
Staphylococcus species
|
202
|
2.4
|
*Not all organisms reported (only top nine species) |
# Includes cultures where more than one organism was reported |
VRE - Vancomycin resistant enterococci |
MSSA – methicillin sensitive Staphylococcus aureus |
Time series analysis of pre COVID-19 cohort and COVID-19 cohort
Combined bloodstream and urinary tract infections
Hospital A demonstrated a significant increase in in the pre-COVID-19 cohort (p < 0.001), and a significant decrease in the COVID-19 cohort (p = 0.004) when combining both BSI and UTI data. Hospital D had a significant decrease in the COVID-19 cohort (p = 0.002). There were no other significant trends identified, however Hospitals C and D had a slight decrease in the COVID-19 cohort, whilst Hospital B demonstrated an increase in the COVID-19 cohort.
Bloodstream infections
When combing all BSI data, although a downward trend is noted in the COVID-19 cohort, it was not significant. Hospital A had significant increase in BSI in the pre COVID-19 cohort (p = 0.028) and a significant decrease in the COVID-19 cohort (p = 0.042). No other significant trends were identified, however Hospitals C, D and E all had downward trends in the COVID-19 cohort.
Urinary tract infections
There is a downward trend in the COVID-19 cohort when combining all hospitals UTI data, however it was not significant. Hospital A had a significant increase in the pre-COVID-19 cohort (p < 0.001) and a significant decrease in the COVID-19 cohort (p = 0.005). Hospitals B, C and D all demonstrated significant decreases in UTI in the pre-COVID-19 cohort (p = 0.026, p = 0.043 and p = 0.041 respectively), whilst hospital B and C showed an increase in the COVID-19 cohort, and Hospitals D and E had a downward trend, none were significant.
Combined infections by state
Combining BSI and UTI data and grouping by state demonstrated that Victoria had a significant increase in the pre-COVID-19 cohort (p = 0.005) and a significant decrease in the COVID-19 cohort (p = 0.011). No significant trends were identified in combined NSW data