Multidrug-resistant bacteria may be associated with bed allocation and utilization efficiency in healthcare institutions, based on national monitoring data from China (2014– 2020)

doi:10.21203/rs.3.rs-2738569/v1

Download PDF

Article

Multidrug-resistant bacteria may be associated with bed allocation and utilization efficiency in healthcare institutions, based on national monitoring data from China (2014– 2020)

https://doi.org/10.21203/rs.3.rs-2738569/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 12 Dec, 2023

Read the published version in Scientific Reports →

You are reading this latest preprint version

Multidrug-resistant bacteria have become one of the most paramount health perils in the 21st century. In this study, provincial-level panel data in China were used to analyze the relationship between bed allocation, utilization efficiency, and MDROs detection rates by the panel data quantile regression model. It was shown that the number of beds per 1,000 people had a negative effect on the detection rate of MRSA, VREA, VREM, PRSP, MRCNS, and CTX/CRO-R ECO (regression coefficient < 0, P < 0.05). The hospital bed utilization rate had a positive effect on the detection rate of MRSA, MRCNS, VREM, PRSP, CTX/CRO-R ECO, CR-ECO, CTX/CRO-R KPN, CR-KPN, CR-PAE, and CR-ABA (regression coefficient > 0, P < 0.05). The average hospital stay had a positive effect on the detection rate of MRSA, MRCNS, VREA, VREM, PRSP, CTX/CRO-R ECO, CR-ECO, QNR-ECO, CTX/CRO-R KPN, CR-PAE, and CR-ABA (regression coefficient > 0, P < 0.05). MDROs are associated with bed allocation and utilization efficiency in healthcare institutions. The Department of Medical Administration should reduce the incidence of MDROs by increasing bed medical allocation and improving the efficiency of bed utilization in hospitalized patients.

Health sciences/Risk factors

Health sciences/Health care/Health services

The massive use of antibiotics has inevitably led to the emergence of drug resistance. Multidrug-resistant bacteria (MDROs), which are resistant to three or more drugs simultaneously, pose a considerable challenge to drug selection, infection prevention, and control. It has become one of the most paramount health perils in the 21st century¹. Although partial success in the prevention and control of hospital-associated MDROs infections is suggested by some data², the relatively high risk of death in patients with MDROs infections (64% higher than in patients with non-drug-resistant infections) makes the prevention and control of MDROs transmission impossible to ignore³.

Hand hygiene, screening, and isolation measures are standardized to interrupt the spread of MDROs transmission and nosocomial infections in healthcare institutions^4–11. However, it has been shown in studies in recent years that overcrowding and insufficient reserves of medical staff have promote spread of MDROs in care unit^12–18. As a result, reduced hand hygiene compliance among healthcare workers, increased movement of patients and staff between hospital wards, and overburdened screening and isolation facilities have occurred. Bed allocation and utilization efficiency in healthcare institutions are direct causes of overcrowding and understaffing It was shown in a survey of 24,676 adult patients in the United States of America that the risk of sustained increases in adverse patient safety events in full or overloaded hospitals existed¹⁹. Specifically, there was a 15% increase in adverse event rates for every 10% increase in bed occupancy rate and a 28% increase in adverse event rates for every 0.1% increase in the patient-to-nurse ratio¹⁹.

Some observational studies have analyzed the relationship between bed occupancy and MDROs, but there are conflicting findings^20–28. In this study, provincial-level panel data in China were used to analyze the relationship between bed allocation, utilization efficiency, and MDROs detection rates to provide data to support MDROs prevention and control.

Percentage of targeted species

After eliminating duplicate strains on the principle of retaining the first strain of the same bacteria from the same patient, 20,262,294 bacterial strains were included for analysis from 2014 to 2020. There were 5,917,700 gram-positive strains (29.2%) and 14,344,594 gram-negative strains (70.8%). The total number of bacterial isolates was between 2,227,420 and 3,528,471 annually. There were no changes in the ratio between specimen types during the study period (gram-positive bacteria accounted for about 30%, and gram-negative bacteria accounted for about 70%). The eight selected MDROs accounted for 68.9% of the total bacteria. Among the eight main MDROs, the isolation rate of E. coli was the highest (4,202,679 strains, 20.7% of the total), followed by K. pneumoniae (2,889,256 strains, 14.3% of the total), S. aureus (1,901,629 strains, 9.4% of the total), P. aeruginosa (1,785,500 strains, 8.8% of the total), and A. baumannii (1,457,423 strains, 7.2% of the total). Regarding the three MDROs with the lowest abundances, each accounted for 2.8%: E. faecium (569,071 strains), E. faecalis (575,526 strains), and S. pneumoniae (574,317 strains) (Table 1).

Table 1

Constituent of targeted species, CARSS, 2014–2020
Species	Total		2014		2015		2016		2017		2018		2019		2020
	(n = 20 262 294)		(n = 2 227 420)		(n = 2 400 786)		(n = 2 727 605)		(n = 2 894 517)		(n = 3 234 372)		(n = 3 528 471)		(n = 3 249 123)
	n	%	n	%	n	%	n	%	n	%	n	%	n	%	n	%
Gram-positive bacteria	5 917 700	29.2	634 414	28.5	695 066	28.9	794 073	29.1	859 388	29.7	952 023	29.4	1 043 535	29.6	939 201	28.9
Staphylococcus aureus	1 901 629	9.4	194 749	8.7	223 674	9.3	256 716	9.4	273 872	9.5	309 801	9.6	337 039	9.6	305 778	9.4
Enterococcus faecium	569 071	2.8	55 769	2.5	61 920	2.6	73 469	2.7	78 444	2.7	91 788	2.8	105 437	3	102 244	3.1
Enterococcus faecalis	575 526	2.8	63 566	2.9	67 398	2.8	76 664	2.8	81 403	2.8	90 196	2.8	98 418	2.8	97 881	3
Streptococcus pneumoniae	574 317	2.8	61 770	2.8	64 798	2.7	72 293	2.7	84 374	2.9	101 534	3.1	113 136	3.2	76 412	2.4
Gram-negative bacteria	14 344 594	70.8	1 593 006	71.5	1 705 720	71.1	1 933 532	70.9	2 035 129	70.3	2 282 349	70.6	2 484 936	70.4	2 309 922	71.1
Acinetobacter baumannii	1 457 423	7.2	171 662	3.2	183 124	7.6	208 689	7.7	207 046	7.2	227 091	7.1	239 890	6.8	219 921	6.8
Escherichia coli	4 202 679	20.7	465 136	20.9	509 862	21.2	575 494	21.1	597 909	20.7	660 261	20.4	707 968	20.1	686 049	21.1
Klebsiella pneumoniae	2 889 256	14.3	308 951	13.9	336 738	14	381 198	14	411 487	14.2	465 322	14.4	503 230	14.3	482 330	14.8
Pseudomonas aeruginosa	1 785 500	8.8	202 817	9.1	219 558	9.1	246 242	9	253 083	8.7	283 222	8.8	299 318	8.5	281260	8.7

Distribution Of Mdros In Different Years And Provinces

As shown in Table 2, among gram-positive bacteria, the isolation rate of MRSA had a slow downward trend, from 36% in 2014 to 29.4% in 2020. The isolation rate of MRCNS also gradually decreased, from 79.8% in 2014 to 74.7% in 2020. The isolation rate of VREA decreased from 0.8% in 2014 to 0.2% in 2020. Similarly, the detection rate of VREM decreased from 2.9% in 2014 to 1% in 2020. The detection rate of PRSP decreased from 4.3% in 2014 to 0.9% in 2020. In contrast, the detection rate of ERSP had an increasing trend, from 94% in 2014 to 96% in 2020.

Table 2

Isolation rate of multidrug-resistant bacteria in China, 2014–2020 (%)
	2014	2015	2016	2017	2018	2019	2020	Mann-Kendall test
	2014	2015	2016	2017	2018	2019	2020	Trend	Z value	P value
MRSA	36	35.8	34.4	32.2	30.9	30.2	29.4	Down	-3.0038	0.002667^*
MRCNS	79.8	79.4	77.5	76	75.7	75.4	74.7	Down	-3.0038	0.002667^*
VREA	0.8	0.8	0.6	0.4	0.3	0.2	0.2	Down	-2.7665	0.005666^*
VREM	2.9	2.9	2	1.4	1.4	1.1	1	Down	-2.7665	0.005666^*
PRSP	4.3	4.2	3.9	2.7	1.8	1.6	0.9	Down	-3.0038	0.002667^*
ERSP	94	91.5	94.4	95	95.4	95.6	96	Up	2.7034	0.006864^*
CTX/CRO-R ECO	59.7	59	56.6	54.2	53	51.9	51.6	Down	-3.0038	0.002667^*
CR-ECO	1.9	1.9	1.5	1.5	1.5	1.7	1.6	Down	-0.63511	0.5254
QNR-ECO	54.3	53.5	52.9	51	50.8	50.6	50.7	Down	-2.7034	0.006864^*
CTX/CRO-R KPN	36.9	36.5	34.5	33	32.4	31.9	31.1	Down	-3.0038	0.002667^*
CR-KPN	6.4	7.6	8.7	9	10.1	10.9	10.9	Up	2.8863	0.003898^*
CR-PAE	25.6	22.4	22.3	20.7	19.3	19.1	18.3	Down	-3.0038	0.002667^*
CR-ABA	57	59	60	56.1	56.1	56	53.7	Down	-1.9748	0.04829^*
^*: P < 0.05

Among the gram-negative bacteria, the isolation rate of CTX/ CR-R ECO had a gradual downward trend, from 59.7% in 2014 to 51.6% in 2020. CR-ECO maintained a low level (isolation rate of 1.9–1.6%), and the isolation rate of QNR-ECO had a slow downward trend, from 54.3% in 2014 to 50.7% in 2020. CTX/ CR-R KPN decreased gradually from 36.9% in 2014 to 31.1% in 2020. Still, the detection rate of CR-KPN continued to increase from 6.4% in 2014 to 10.9% in 2020. The detection rate of CR-PAE continued to decrease from 25.6% in 2014 to 18.3% in 2020. The detection rate of CR-ABA fluctuated between 53.7% and 60% and maintained a high level. A statistical description of the detection rates of different MDROs is shown in Appendix 1. It was revealed that the detection rate of some strains deviated in different years. The influence of extreme values was effectively overcome by quantile regression, and further evidence for using quantile regression was provided by the substantial deviation in the detection rate.

The different provinces had different bacterial resistance rates. For example, the detection rate of MRSA in Shanghai, Shaanxi, Jiangsu, Beijing, and Anhui from 2014 to 2020, which are all above the national average values (32.35%), and Yunnan, Xinjiang, Tianjin, Sichuan, and Shandong are all below the national average values. Meanwhile, the detection rate of MRSA in Heilongjiang and Liaoning had a relatively large variation. The distribution of other specific MDROs detected in different regions is shown in Fig. 1.

Allocation And Utilization Efficiency Of Medical Beds In Healthcare Institutions In Various Provinces

As shown in Fig. 2, the number of beds in health institutions per 1,000 people increased annually in the provinces. The hospital bed utilization rate declined relatively more in 2020 and remained the same in other years. The average hospital stay fluctuated differently in the different provinces, and the antibiotic use rate had a slow downward trend. Additionally, we tested the applicability of the Quantile regression. The difference between the minimum and the maximum value and the standard deviation of different MDROs detection rates were considerable, indicating that the MDROs detection rate deviated from the mean in a few provinces. Meanwhile, there were also some differences in the minimum and maximum values of the other variables. The quantile regression model effectively measures the impact of extreme values, which provides further evidence for using quantile regression. The statistical description of the MDROs detection rate and other variables is shown in Appendixes 1 and 2.

Panel Quantile Regression Of Mdros Detection Rates And Medical Resource Allocation

From the number of beds per 1,000 people on five quantile points shown in Table 3, significance at the 0.05 level was shown for MRSA (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), VREA (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), VREM (τ = 0.1, 0.3, 0.5, and 0.7), PRSP (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), MRCNS (τ = 0.7 and 0.9), CTX/CRO-R ECO (τ = 0.5 and 0.9), CTX/CRO-R KPN (τ = 0.3, 0.5, and 0.9), and CR-PAE (τ = 0.1, 0.3, and 0.9). The regression coefficients were all lower than 0, indicating that the detection rates of the eight MDROs mentioned above were negatively affected by the number of beds available per 1,000 people.

Table 3

Association between the detection rate of multi-drug resistant bacteria and the bed allocation and utilization efficiency in healthcare institutions with the panel quantile regression model
MRSA
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	-0.820(-0.06)	13.070(1.02)	11.370(0.72)	-3.957(-0.23)	-7.910(-0.55)
Number of beds per 1,000 population	-1.519**(-1.98)	-1.405**(-1.98)	-2.328***(-2.66)	-2.619***(-2.75)	-1.521*(-1.90)
Hospital bed utilization rate	0.187**(2.02)	0.358***(4.18)	0.311***(2.95)	0.356***(3.11)	0.397***(4.11)
Average hospital stay	-0.983(-1.02)	-2.142**(-2.40)	-0.366(-0.33)	2.059*(1.72)	2.799***(2.79)
Utilization rate of antibiotics	0.612***(6.07)	0.267***(2.86)	0.232**(2.02)	0.099(0.79)	-0.053(-0.50)
Pseudo R²	0.1673	0.1576	0.1245	0.107	0.1737
MRCNS
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	33.410*(1.65)	31.230***(3.43)	51.440***(6.84)	62.600***(9.65)	64.990***(8.86)
Number of beds per 1,000 population	-0.096(-0.09)	-0.289(-0.57)	-0.683(-1.64)	-0.945***(-2.63)	-0.888**(-2.18)
Hospital bed utilization rate	0.449***(3.33)	0.419***(6.89)	0.234***(4.66)	0.147***(3.40)	0.180***(3.68)
Average hospital stay	0.260(0.19)	1.041(1.64)	0.933*(1.78)	0.566(1.26)	0.470(0.92)
Utilization rate of antibiotics	-0.115(-0.78)	-0.055(-0.83)	-0.009(-0.16)	0.076(1.62)	0.008(0.16)
Pseudo R2	0.1412	0.1497	0.1279	0.1299	0.1292
VREA
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	-0.661**(-2.13)	-1.138***(-2.62)	-1.413**(-2.16)	-1.031(-1.22)	1.454(0.84)
Number of beds per 1,000 population	-0.089***(-5.18)	-0.069***(-2.85)	-0.085**(-2.34)	-0.147***(-3.15)	-0.333***(-3.47)
Hospital bed utilization rate	0.000(0.21)	0.003(1.21)	0.001(0.34)	0.001(0.26)	0.000(0.00)
Average hospital stay	0.083***(3.84)	0.112***(3.72)	0.184***(4.06)	0.254***(4.34)	0.212*(1.76)
Utilization rate of antibiotics	0.012***(5.21)	0.010***(3.13)	0.011**(2.21)	-0.001(-0.10)	-0.011(-0.87)
Pseudo R2	0.0935	0.1056	0.0974	0.1146	0.1453
VREM
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	-1.921*(-1.74)	-2.749**(-2.24)	-3.627**(-2.47)	-5.736**(-2.13)	-16.650*(-1.74)
Number of beds per 1,000 population	-0.173***(-2.82)	-0.288***(-4.23)	-0.312***(-3.83)	-0.299**(-2.00)	-0.119(-0.22)
Hospital bed utilization rate	0.014*(1.95)	0.025***(3.07)	0.030***(3.00)	0.035*(1.93)	0.066(1.03)
Average hospital stay	0.222***(2.88)	0.325***(3.80)	0.388***(3.80)	0.646***(3.45)	1.656**(2.49)
Utilization rate of antibiotics	-0.003(-0.38)	-0.005(-0.52)	0.005(0.47)	-0.004(-0.18)	-0.032(-0.46)
Pseudo R²	0.0622	0.1254	0.1287	0.1180	0.2037
PRSP
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	-0.349(-0.21)	-0.066(-0.03)	-3.603(-0.95)	-5.988(-0.92)	-13.220*(-1.66)
Number of beds per 1,000 population	-0.201**(-2.20)	-0.311**(-2.40)	-0.514**(-2.45)	-1.063***(-2.94)	-1.503***(-3.41)
Hospital bed utilization rate	0.029***(2.63)	0.030*(1.92)	0.032(1.27)	0.054(1.23)	0.070(1.32)
Average hospital stay	-0.093(-0.81)	-0.033(-0.20)	0.593**(2.25)	1.118**(2.46)	2.250***(4.07)
Utilization rate of antibiotics	0.012(0.98)	0.020(1.15)	0.010(0.35)	0.015(0.32)	0.016(0.28)
Pseudo R²	0.0514	0.0568	0.0592	0.1177	0.2083
ERSP
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	91.980***(4.70)	79.260***(7.59)	85.120***(15.18)	94.010***(26.87)	91.340***(26.03)
Number of beds per 1,000 population	0.553(0.51)	0.789(1.36)	0.543*(1.75)	0.213(1.10)	0.349*(1.79)
Hospital bed utilization rate	-0.105(-0.81)	0.024(0.34)	0.038(1.02)	-0.006(-0.26)	0.029(1.24)
Average hospital stay	-0.121(-0.09)	0.530(0.73)	0.031(0.08)	0.149(0.61)	0.083(0.34)
Utilization rate of antibiotics	0.089(0.63)	0.053(0.70)	0.074*(1.82)	0.005(0.21)	0.013(0.50)
Pseudo R²	0.0256	0.0293	0.0240	0.0076	0.0076
CTX/CRO-R ECO
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	22.600**(2.22)	17.340**(2.04)	12.900(1.59)	7.972(0.71)	43.780***(3.73)
Number of beds per 1,000 population	-0.444(-0.79)	-0.755(-1.60)	-1.154**(-2.57)	-0.754(-1.22)	-2.526***(-3.88)
Hospital bed utilization rate	0.087(1.28)	0.149***(2.62)	0.234***(4.31)	0.264***(3.53)	0.073(0.93)
Average hospital stay	1.350*(1.91)	1.958***(3.31)	1.909***(3.38)	2.127***(2.74)	2.125***(2.60)
Utilization rate of antibiotics	0.213***(2.88)	0.191***(3.09)	0.242***(4.09)	0.250***(3.07)	0.138(1.61)
Pseudo R²	0.1212	0.1273	0.1509	0.1456	0.1804
CR-ECO
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	-2.096***(-3.49)	-1.476*(-1.70)	-1.342(-0.83)	-3.092(-1.62)	-5.655***(-2.99)
Number of beds per 1,000 population	0.054(1.61)	0.003(0.06)	-0.065(-0.72)	-0.105(-0.99)	-0.040(-0.38)
Hospital bed utilization rate	0.021***(5.22)	0.017***(2.97)	0.026**(2.40)	0.033**(2.55)	0.023*(1.80)
Average hospital stay	0.088**(2.11)	0.107*(1.76)	0.089(0.79)	0.262*(1.97)	0.560***(4.26)
Utilization rate of antibiotics	-0.003(-0.65)	-0.001(-0.08)	-0.002(-0.15)	0.007(0.53)	0.031**(2.21)
Pseudo R²	0.0568	0.0382	0.0456	0.0675	0.1077
QNR-ECO
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	48.300***(6.74)	41.140***(3.88)	29.480**(2.43)	25.080**(2.07)	9.130(0.86)
Number of beds per 1,000 population	-1.601***(-4.03)	-0.749(-1.27)	-0.476(-0.71)	-0.083(-0.12)	0.391(0.66)
Hospital bed utilization rate	0.016(0.33)	-0.109(-1.54)	-0.149*(-1.85)	-0.081(-1.00)	-0.170**(-2.39)
Average hospital stay	0.993**(1.99)	2.710***(3.67)	4.486***(5.32)	4.892***(5.79)	6.672***(9.02)
Utilization rate of antibiotics	-0.100*(-1.91)	-0.090(-1.16)	-0.077(-0.88)	-0.190**(-2.15)	0.010(0.13)
Pseudo R²	0.0937	0.0984	0.1445	0.1881	0.2418
CTX/CRO-R KPN
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	-18.550(-0.92)	16.310(1.44)	-0.964(-0.06)	-24.350*(-1.88)	-45.930(-1.41)
Number of beds per 1,000 population	-1.187(-1.06)	-2.181***(-3.47)	-2.288***(-2.65)	-2.120***(-2.95)	-1.232(-0.68)
Hospital bed utilization rate	0.623***(4.61)	0.312***(4.11)	0.345***(3.31)	0.412***(4.76)	0.476**(2.19)
Average hospital stay	-1.838(-1.31)	-1.364*(-1.73)	0.369(0.34)	3.157***(3.50)	5.589**(2.47)
Utilization rate of antibiotics	0.273*(1.85)	0.254***(3.07)	0.314***(2.77)	0.170*(1.80)	0.025(0.11)
Pseudo R²	0.1607	0.1614	0.1366	0.1431	0.1398
CR-KPN
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	-9.053**(-2.21)	-13.410**(-2.13)	-19.310*(-1.86)	-12.540(-0.77)	-23.490(-0.73)
Number of beds per 1,000 population	0.391*(1.72)	0.580*(1.66)	1.099*(1.91)	0.661(0.73)	1.677(0.94)
Hospital bed utilization rate	0.133***(4.86)	0.163***(3.88)	0.273***(3.95)	0.331***(3.05)	0.501**(2.34)
Average hospital stay	-0.285(-1.00)	-0.244(-0.56)	-0.810(-1.12)	-1.057(-0.93)	0.789(0.35)
Utilization rate of antibiotics	0.003(0.09)	0.048(1.05)	0.079(1.05)	0.000(0.00)	-0.508**(-2.17)
Pseudo R²	0.0606	0.0741	0.0798	0.0502	0.1109
CR-PAE
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	-0.489(-0.07)	2.804(0.35)	-13.890(-1.05)	-30.660**(-2.12)	-20.760**(-2.29)
Number of beds per 1,000 population	-1.426***(-3.42)	-2.222***(-5.04)	-0.967(-1.32)	-0.768(-0.96)	-1.205**(-2.40)
Hospital bed utilization rate	0.165***(3.29)	0.234***(4.40)	0.334***(3.78)	0.357***(3.69)	0.224***(3.70)
Average hospital stay	0.966*(1.85)	0.473(0.86)	0.948(1.03)	2.595**(2.58)	3.050***(4.83)
Utilization rate of antibiotics	-0.054(-0.99)	0.023(0.40)	0.012(0.13)	0.0612(0.58)	0.171**(2.59)
Pseudo R²	0.0862	0.1080	0.1148	0.1449	0.2006
CR-ABA
variable	0.1RQ	0.3RQ	0.5RQ	0.7RQ	0.9RQ
Constant	-79.000*(-1.96)	-35.910*(-1.91)	-22.860(-1.52)	-16.570(-1.12)	-33.460(-1.08)
Number of beds per 1,000 population	2.703(1.21)	1.536(1.47)	0.581(0.70)	-0.292(-0.36)	-0.098(-0.06)
Hospital bed utilization rate	1.048***(3.90)	0.602***(4.79)	0.534***(5.32)	0.302***(3.05)	0.252(1.22)
Average hospital stay	-0.076(-0.03)	2.409*(1.84)	2.498**(2.39)	4.818***(4.68)	7.174***(3.34)
Utilization rate of antibiotics	0.298(1.02)	0.053(0.38)	0.100(0.92)	0.143(1.33)	0.241(1.07)
Pseudo R²	0.1849	0.118	0.112	0.106	0.146
* p < 0.05 ** p < 0.01, The value is “Regression coefficients” in table and “t” in brackets.

Regarding the hospital bed utilization rate, significance at the 0.05 level was observed for MRSA (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), MRCNS (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), VREM (τ = 0.1, 0.3, 0.5, and 0.7), PRSP (τ = 0.1 and 0.3), CTX/CRO-R ECO (τ = 0.3, 0.5, and 0.7), CR-ECO (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), CTX/CRO-R KPN (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), CR-KPN (τ = 0.1, 0.3, and 0.5), CR-PAE (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), and CR-ABA (τ = 0.1, 0.3, 0.5, 0.7, and 0.9). The regression coefficients were all higher than 0, indicating that the detection rates of the 10 MDROs mentioned above were positively affected by the hospital bed utilization rate.

Regarding the average hospital stay, the significance at the 0.05 level is shown for MRSA (τ = 0.7 and 0.9), MRCNS (τ = 0.5), VREA (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), VREM (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), PRSP (τ = 0.1 and 0.3), CTX/CRO-R ECO (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), CR-ECO (τ = 0.1, 0.3, 0.7, and 0.9), QNR-ECO (τ = 0.1, 0.3, 0.5, 0.7, and 0.9), CTX/CRO-R KPN (τ = 0.7 and 0.9), CR-PAE (τ = 0.1, 0.7, and 0.9), and CR-ABA (τ = 0.3, 0.5, 0.7, and 0.9). All the regression coefficients were higher than 0, indicating that the average hospital stay had a positive effect on the detection rates of the 11 MDROs mentioned above. In the sensitivity analysis (Appendixes 3), we obtained the same findings, indicating that the study results are stable.

Factors influencing the occurrence and spread of MDROs include antimicrobial drug use, level of disinfection and isolation, hand hygiene, environmental hygiene, etc. China has taken targeted measures to strengthen the control of MDROs in-hospital transmission since the beginning of the 21st century to address these factors^{29, 30}. From the national surveillance data of MDROs, it can be seen that the detection rates of 11 MDROs, except for ERSP and CR-KPN, had a decreasing annual trend during 2014–2020, indicating that the implemented control measures were effective.

In this study, we used provincial-level panel data from China to analyze the relationship between bed allocation, utilization efficiency, and MDROs detection rates. We showed in our study that the number of beds per 1,000 people had a negative effect on most of the MDROs detection rates, and the hospital bed utilization rate and the average hospital stay had positive effects on most of the MDROs detection rates. This means that the higher the number of beds per 1,000 people, the lower the MDROs detection rate. Additionally, the higher the hospital bed utilization rate or the average hospital stay, the higher the MDROs detection rate.

The number of beds per 1,000 people reflects the distribution of beds and the accessibility of health services. It is an indicator that varies from region to region, depending on the economic level. The hospital bed utilization rate is the ratio of beds in use to beds available per day. It constitutes an assessment of the workload of hospital beds and medical staff. In general, the hospital bed utilization rate exceeds 85%, which may lead to omissions in hospital infection management due to the high-load working state of medical staff. There was a significant increase in MRSA detection rates found by studies in acute hospitals in England between 2001 and 2004 when bed occupancy rates of more than 90% were observed. Cunningham et al²⁰. Reported a significant association between bed occupancy and MRSA rates in UK hospitals, contributing to the spread of nosocomial pathogens when occupancy exceeded 85%. The Department of Health in the Netherlands and UK released further guidance to reduce healthcare-associated infections, such as the availability of beds in nursing homes, hospital isolation facilities, and changes in nurse-to-patient ratios. Health authorities in China require medical institutions to admit patients strictly following the approved bed capacity to prevent the spread of infectious strains in hospitals and to ensure that the hospital bed utilization rate is controlled at 85–93%²³.

Additionally, authorities indicate that the bed spacing must not be less than 0.8 m to avoid problems, such as increased medical and nursing staff workload, increased incidence of errors, and difficulties in in-hospital infection control caused by high bed utilization rates. However, due to bed space constraints, the inability to effectively isolate patients with MDROs may also be another important factor contributing to the spread of MDROs within the hospital. From national surveillance data, the detection rates of MDROs are also relatively high in large cities with dense populations and relatively scarce medical resources, such as Shanghai and Beijing. In economically developed regions, medical resources are allocated relatively well. Nonetheless, because of the large population and the high demand for medical care from patients, the available medical resources, such as beds and medical care, are still unable to meet the needs of many patients, resulting in high incidences of MDROs.

Therefore, the Department of Medical Administration should improve the efficiency of bed utilization by increasing bed and medical and nursing allocation, reducing the incidence of MDROs in hospitalized patients, and ensuring the quality of overall medical care. At the same time, according to the bed utilization index of the department and the nosocomial infection situation, patients waiting for beds will be divided into light, medium, and critical care. Appropriate adjustments will be made, such as light patients can be deferred and critical patients will not be added to beds as far as possible. Still, empty beds for light and medium-care patients will be maintained frequently to avoid the accumulation of heavy beds caused by the addition of beds as far as possible.

This study has some limitations. First, the relationship between MDROs, bed allocation, and utilization efficiency from a macroscopic perspective was analyzed in this study. Still, we may have ignored the infection factors of the patients because of our study design. Second, we chose only bed-related factors in selecting explanatory variables and may have incompletely considered other factors that influence the occurrence of MDROs.

In conclusion, MDROs are associated with bed allocation and utilization efficiency in healthcare institutions. Therefore, the Department of Medical Administration should improve the efficiency of bed utilization by increasing bed and medical and nursing allocation, reducing the incidence of MDROs in hospitalized patients, and ensuring the quality of overall medical care.

Data sources. The detection rate and drug resistance of MDROs were obtained from the surveillance report of bacterial resistance of the China Antimicrobial Resistance Surveillance System (CARSS, http://www.carss.cn ). CARSS was established and organized in 2005 by the Expert Committee on Rational Use of Drugs, National Health Commission of the People’s Republic of China (Formerly Health and Family Planning Committee of China). In 2014, 1,100 hospitals from all provinces and autonomous regions participated in the survey. In 2020, they were increased to 1,432, among which the tertiary hospitals accounted for about 3/4 (74.1–76.2%), and the secondary hospitals accounted for about a quarter (23.8–25.8%). All monitoring hospitals have an adequate working basis for clinical microbiology laboratories that received quality control certification from provincial or/and national clinical testing centers with their indoor quality control standards.

Bacterial identification and drug resistance monitoring were conducted following the CARSS monitoring technical protocol, with uniform methods and technical training. Quality control followed the American Clinical Laboratory Standardization Institute (CLSI) requirements. Routine quality control procedures were performed once a week under stable test conditions. In this study, MDROs included 13 types of bacteria, including methicillin-resistant Staphylococcus aureus (MRSA). Methicillin-resistant coagulase-negative Staphylococcus (MRCNS), vancomycin-resistant Enterococcus faecalis (VREA), vancomycin-resistant Enterococcus faecium (VREM), penicillin-resistant Streptococcus pneumoniae (PRSP), erythromycin-resistant Streptococcus pneumoniae (ERSP), cefotaxime or ceftriaxone resistant Escherichia coli (CTX/CRO-R ECO), carbapenem-resistant E. coli (CR-ECO), quinolone-resistant E. coli (QNR-ECO), cefotaxime or ceftriaxone resistant Klebsiella pneumoniae (CTX/CRO-R KPN), carbapenem-resistant Klebsiella pneumoniae (CR-KPN), carbapenem-resistant Pseudomonas aeruginosa (CR-PAE), and carbapenem-resistant Acinetobacter baumannii (CR-ABA).

The panel data on health resource indicators were selected from the China Health and Health Statistics Yearbook (2014–2020). The extracts include the number of beds in healthcare institutions per 1,000 people, the hospital bed utilization rate, and the average length of stay. The population size is the resident population, and healthcare institutions include medical institutions, such as general hospitals, Chinese medicine hospitals, and specialist hospitals. The hospital bed utilization rate reflects the ratio of beds in use to actual beds per day in healthcare institutions, i.e., the ratio of the total number of bed days occupied to the total number of bed days open in the hospital.

The utilization rate of antibiotics was from the Center for Antimicrobial Surveillance (http://y.chinadtc.org.cn/login), which was established in 2006 and organized by the National Health Commission of the People’s Republic of China (Formerly Health and Family Planning Committee of China). In 2020, more than 4,000 hospitals joined the center, of which general hospitals accounted for more than 60%.

Statistical analysis. In this study, data from all provinces in China were used as the study sample, with 2014–2020 as different time observation points to form a balanced panel data sample. As the Tibet Autonomous Region is one of the most economically disadvantaged provinces in China, its bacterial surveillance data for 2014–2016 are missing, and those from 2017–2020 are poorly stable. Therefore, Tibet Autonomous Region were excluded from the study. Strains and percentages were used in the study to provide a basic description of MDRO, the R (4.1.3, Free Software Foundation) package (“ggplot2” package) to map the temporal and areal distribution of different multidrug-resistant bacteria in 30 Chinese provinces from 2014 to 2020. The Mann-Kendall Trend Test was used to analyze the isolation trend rate of 13 types of MDROs to check whether they constantly increased or decreased over time. We constructed a panel data quantile regression model using Stata/SE 15.0 (15.0, Stata Corp.). The response variables were set as the detection rates of different MDROs. The number of beds in medical and health institutions per 1,000 people, the hospital bed utilization rate, and the average length of stay as explanatory variables, and the rate of antimicrobial drug use as covariates because it is now generally accepted that the rate of antimicrobial drug use affects the occurrence of MDROs. In the sensitivity analysis, we added the explanatory variables Average daily burden of bed days and Medical practitioners per 1,000 people to examine the stability of the results.

Model construction. Panel data are two-dimensional data containing time and individual observation. It integrates the advantages of cross-sectional and time-series data, so the panel data model can effectively reduce design and estimation biases. Its general form is as follows:

$${y}_{it}={x}_{it}\beta +{z}_{i}\alpha +{\epsilon }_{it}$$

where${y}_{it}$ is the value of the explained variable at cross-section $i$ and time$t$; ${x}_{it}$ is the value of the explanatory variable at cross-section $i$ and time$t$; ${\epsilon }_{it}$ is a random error term, ${z}_{i}\alpha$ is heterogeneity or individual effect, where ${z}_{i}$ contains a constant observable term and a series of group variables that may not change over time.

Ordinary least squares (OLS) is a commonly used parameter estimation method in panel data regression models that are mainly used to estimate the influence of a given explanatory variable on the conditional mean of the explained variable. This method is primarily based on data where the explained variable follows a normal distribution and is sensitive to outliers. Quantile regression is an extension of OLS. It uses the weighted residual method to estimate the parameters of the sum of the absolute value according to the conditions of the explanatory variables of variable quantile regression. No distribution assumption is assumed in terms of the random perturbation model. When the conditional distribution of the interpreted variable has a Gaussian distribution, quantile regression is adopted so that more reasonable conclusions are reached. The sensitivity to outliers is much less than that of OLS, so quantile regression has stronger robustness.

Quantile regression of panel data was used to estimate the parameters of the panel data model using the quantile regression method. The influence of explanatory variables on the conditional distribution of explained variables may be better analyzed with this method at different sub-points. At the same time, we controlled individual heterogeneity, especially to capture the characteristics of the tail of the distribution. A fixed-effect panel quantile regression model was selected in this study to eliminate the association between multidrug-resistant bacteria in different years:

${Q}_{{y}_{it}}\left(\tau \left|{x}_{it}\right.\right)={\alpha }_{i}+{X}_{it}^{T}\beta \left(\tau \right)+{\mu }_{it}$, $i=1, 2, 3, \dots , N$, $t=1, 2, 3, \dots , T$ (2)

where the subscript $i$ represents the province (city, autonomous region), $t$ represents the year, $\tau$(0 < $\tau$< 1)indicates the selected decimal point, ${y}_{it}$is the explained variable in $i$th provinces (cities, autonomous regions) in year$t$, ${Q}_{yit}\left(\tau \left|{x}_{it}\right.\right)$ denotes the τ conditional quantile of${y}_{it}$under a given clause ${x}_{it}$; ${\alpha }_{i}$ represents the fixed effect of the $i$th province (city, autonomous region), which does not change with time; ${X}_{it}^{T}$represents the explanatory variable vector of the$i$th province (city, autonomous region) in year $t$; $\beta$represents the parameter vector to be estimated, which is assumed not to change with time. ${\mu }_{it }$indicates the random error term. In this study, five quantile points of 0.1, 0.3, 0.5, 0.7, and 0.9 were selected to construct the regression model.

Statement

No human participants were involved during study.

Acknowledgements

We would like to express their gratitude to EditSprings (https://www.editsprings.cn) for the expert linguistic services provided.

Author Contributions

P.F.L. conceived and designed the study. X.T.W. led on data processing and data management. W.W.L., H.M., X.Y.S., Z.L., and X.J.M. planned the statistical analysis, analyzed the data, generated the tables and figures, and accessed and verified all the data. D.F.P., C.C., and M.S. provided topic-area expertise. X.T.W. drafted the manuscript, and P.F.L. reviewed and revised drafts and approved the final version. All authors had access to the data and accept responsibility for the decision to submit for publication.

Data Availability statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author upon request.

Competing Interests Statement

All authors declare no competing interests.

Arias CA, Murray BE. Antibiotic-resistant bugs in the 21st century–a clinical super-challenge. N Engl J Med. 2009;360(5):439–43.
Price L, MacDonald J, Melone L, Howe T, Flowers P, Currie K, et al. Effectiveness of national and subnational infection prevention and control interventions in high-income and upper-middle-income countries: a systematic review. The Lancet Infectious diseases. 2018;18(5):e159-e71.
Jernigan JA, Hatfield KM, Wolford H, Nelson RE, Olubajo B, Reddy SC, et al. Multidrug-Resistant Bacterial Infections in U.S. Hospitalized Patients, 2012–2017. N Engl J Med. 2020;382(14):1309–19.
Jain R, Kralovic S, Evans M, Ambrose M, Simbartl L, Obrosky D, et al. Veterans Affairs initiative to prevent methicillin-resistant Staphylococcus aureus infections. The New England journal of medicine. 2011;364(15):1419–30.
YJ L, JZ C, HC L, HY L, SY L, HH L, et al. Impact of active screening for methicillin-resistant Staphylococcus aureus (MRSA) and decolonization on MRSA infections, mortality and medical cost: a quasi-experimental study in surgical intensive care unit. Critical care (London, England). 2015;19:143.
Johnson PD, Martin R, Burrell LJ, Grabsch EA, Kirsa SW, O'Keeffe J, et al. Efficacy of an alcohol/chlorhexidine hand hygiene program in a hospital with high rates of nosocomial methicillin-resistant Staphylococcus aureus (MRSA) infection. Med J Aust. 2005;183(10):509–14.
Pittet D, Hugonnet S, Harbarth S, Mourouga P, Sauvan V, Touveneau S, et al. Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. Infection Control Programme. Lancet (London, England). 2000;356(9238):1307–12.
Huang SS, Singh R, McKinnell JA, Park S, Gombosev A, Eells SJ, et al. Decolonization to Reduce Postdischarge Infection Risk among MRSA Carriers. The New England journal of medicine. 2019;380(7):638–50.
JV R, N G, BD C, AG B, JA W, JD E, et al. Screening, isolation, and decolonisation strategies in the control of meticillin resistant Staphylococcus aureus in intensive care units: cost effectiveness evaluation. BMJ (Clinical research ed). 2011;343:d5694.
Robotham JV, Deeny SR, Fuller C, Hopkins S, Cookson B, Stone S. Cost-effectiveness of national mandatory screening of all admissions to English National Health Service hospitals for meticillin-resistant Staphylococcus aureus: a mathematical modelling study. Lancet Infectious Diseases. 2016;16(3):348–56.
Wernitz MH, Keck S, Swidsinski S, Schulz S, Veit SK. Cost analysis of a hospital-wide selective screening programme for methicillin-resistant Staphylococcus aureus (MRSA) carriers in the context of diagnosis related groups (DRG) payment. clinical microbiology and infection. 2005;11(6):466–71.
Balkhy HH, Perl TM, Arabi YM. Preventing healthcare-associated transmission of the Middle East Respiratory Syndrome (MERS): Our Achilles heel. J Infect Public Health. 2016;9(3):208–12.
Cowling BJ, Park M, Fang VJ, Wu P, Leung GM, Wu JT. Preliminary epidemiological assessment of MERS-CoV outbreak in South Korea, May to June 2015. Euro Surveill. 2015;20(25):7–13.
Park HY, Lee EJ, Ryu YW, Kim Y, Kim H, Lee H, et al. Epidemiological investigation of MERS-CoV spread in a single hospital in South Korea, May to June 2015. Euro Surveill. 2015;20(25):1–6.
Hui DS. Severe acute respiratory syndrome (SARS): lessons learnt in Hong Kong. J Thorac Dis. 2013;5 Suppl 2(Suppl 2):S122-6.
Andersen BM, Lindemann R, Bergh K, Nesheim BI, Syversen G, Solheim N, et al. Spread of methicillin-resistant Staphylococcus aureus in a neonatal intensive unit associated with understaffing, overcrowding and mixing of patients. J Hosp Infect. 2002;50(1):18–24.
Leistner R, Thürnagel S, Schwab F, Piening B, Gastmeier P, Geffers C. The impact of staffing on central venous catheter-associated bloodstream infections in preterm neonates - results of nation-wide cohort study in Germany. Antimicrob Resist Infect Control. 2013;2(1):11.
Clements A, Halton K, Graves N, Pettitt A, Morton A, Looke D, et al. Overcrowding and understaffing in modern health-care systems: key determinants in meticillin-resistant Staphylococcus aureus transmission. The Lancet Infectious diseases. 2008;8(7):427–34.
Weissman JS, Rothschild JM, Bendavid E, Sprivulis P, Cook EF, Evans RS, et al. Hospital workload and adverse events. Med Care. 2007;45(5):448–55.
Cunningham JB, Kernohan WG, Sowney R. Bed occupancy and turnover interval as determinant factors in MRSA infections in acute settings in Northern Ireland: 1 April 2001 to 31 March 2003. J Hosp Infect. 2005;61(3):189–93.
Zhang M, Ge J, Lin Z. The Impact of the Number of Hospital Beds and Spatial Heterogeneity on an SIS Epidemic Model. Acta Applicandae Mathematicae. 2020;167(1):1–15.
Kaier K. Economic implications of the dynamic relationship between antibiotic use and hospital-acquired infections. Value Health. 2012;15(1):87–93.
Orendi J. Health-care organisation, hospital-bed occupancy, and MRSA. Lancet (London, England). 2008;371(9622):1401–2.
Jones DR. Hospital bed occupancy demystified and why hospitals of different size and complexity must run at different average occupancy. British Journal of Healthcare Management 2011;17(6):242–8.
Kaier K, Mutters NT, Frank U. Bed occupancy rates and hospital-acquired infections—should beds be kept empty? Clinical Microbiology and Infection. 2012;18(10):941–5.
Fischer D, Schlößer RL, Kempf VAJ, Wichelhaus TA, Klingebiel T, Philippi S, et al. Overcrowding in a neonatal intermediate care unit: impact on the incidence of multidrug-resistant gram-negative organisms. BMC Infect Dis. 2019;19(1):357.
Kong F, Cook D, Paterson DL, Whitby M, Clements AC. Do staffing and workload levels influence the risk of new acquisitions of meticillin-resistant Staphylococcus aureus in a well-resourced intensive care unit? J Hosp Infect. 2012;80(4):331–9.
Borg MA. Bed occupancy and overcrowding as determinant factors in the incidence of MRSA infections within general ward settings. J Hosp Infect. 2003;54(4):316–8.
Hu F, Zhu D, Wang F, Wang M. Current Status and Trends of Antibacterial Resistance in China. Clin Infect Dis. 2018;67(suppl_2):S128-s34.
Hu FP GY, Zhu DM, Wang F, Jiang XF, Xu YC, Zhang XJ, Zhang CX, Ji P, Xie Y, Kang M, Wang CQ, Wang AM, Xu YH, Shen JL, Sun ZY, Chen ZJ, Ni YX, Sun JY, Chu YZ, Tian SF, Hu ZD, Li J, Yu YS, Lin J, Shan B, Du Y, Han Y, Guo S, Wei LH, Wu L, Zhang H, Kong Resistance trends among clinical isolates in China reported from CHINET surveillance of bacterial resistance, 2005–2014. Clin Microbiol Infect. 2016;22(Suppl 1):9–14.

No competing interests reported.

SupplementaryMaterial.pdf

Download PDF

Journal Publication

published 12 Dec, 2023

Read the published version in Scientific Reports →

Editorial decision: Major revision
26 Aug, 2023
Reviews received at journal
24 Aug, 2023
Reviewers agreed at journal
23 Aug, 2023
Reviews received at journal
16 Aug, 2023
Reviewers agreed at journal
08 Aug, 2023
Reviewers invited by journal
05 Jul, 2023
Editor assigned by journal
19 May, 2023
Editor invited by journal
04 Apr, 2023
Submission checks completed at journal
04 Apr, 2023
First submitted to journal
26 Mar, 2023

You are reading this latest preprint version

Multidrug-resistant bacteria may be associated with bed allocation and utilization efficiency in healthcare institutions, based on national monitoring data from China (2014– 2020)

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Results

Percentage of targeted species

Distribution Of Mdros In Different Years And Provinces

Allocation And Utilization Efficiency Of Medical Beds In Healthcare Institutions In Various Provinces

Panel Quantile Regression Of Mdros Detection Rates And Medical Resource Allocation

Discussion

Methods

Statement

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1