Skip to main content
PMC home page
PLOS ONE logoLink to PLOS ONE
. 2024 May 20;19(5):e0303132. doi: 10.1371/journal.pone.0303132

Frequency and mortality rate following antimicrobial-resistant bloodstream infections in tertiary-care hospitals compared with secondary-care hospitals

Cherry Lim 1,2,*, Viriya Hantrakun 1, Preeyarach Klaytong 1, Chalida Rangsiwutisak 1, Ratanaporn Tangwangvivat 3, Chadaporn Phiancharoen 3, Pawinee Doung-ngern 3, Somkid Kripattanapong 3, Soawapak Hinjoy 3, Thitipong Yingyong 3, Archawin Rojanawiwat 4, Aekkawat Unahalekhaka 4, Watcharaporn Kamjumphol 4, Kulsumpun Khobanan 4, Pimrata Leethongdee 4, Narisorn Lorchirachoonkul 5, Suwimon Khusuwan 6, Suwatthiya Siriboon 7, Parinya Chamnan 8, Amornrat Vijitleela 9,10, Traithep Fongthong 11, Krittiya Noiprapai 11, Phairam Boonyarit 11, Voranadda Srisuphan 11, Benn Sartorius 2,12,13, John Stelling 14, Paul Turner 2,15, Nicholas P J Day 1,2, Direk Limmathurotsakul 1,2,16,*
Editor: Ali Amanati17
PMCID: PMC11104583  PMID: 38768224

Abstract

There are few studies comparing proportion, frequency, mortality and mortality rate following antimicrobial-resistant (AMR) infections between tertiary-care hospitals (TCHs) and secondary-care hospitals (SCHs) in low and middle-income countries (LMICs) to inform intervention strategies. The aim of this study is to demonstrate the utility of an offline tool to generate AMR reports and data for a secondary data analysis. We conducted a secondary-data analysis on a retrospective, multicentre data of hospitalised patients in Thailand. Routinely collected microbiology and hospital admission data of 2012 to 2015, from 15 TCHs and 34 SCHs were analysed using the AMASS v2.0 (www.amass.website). We then compared the burden of AMR bloodstream infections (BSI) between those TCHs and SCHs. Of 19,665 patients with AMR BSI caused by pathogens under evaluation, 10,858 (55.2%) and 8,807 (44.8%) were classified as community-origin and hospital-origin BSI, respectively. The burden of AMR BSI was considerably different between TCHs and SCHs, particularly of hospital-origin AMR BSI. The frequencies of hospital-origin AMR BSI per 100,000 patient-days at risk in TCHs were about twice that in SCHs for most pathogens under evaluation (for carbapenem-resistant Acinetobacter baumannii [CRAB]: 18.6 vs. 7.0, incidence rate ratio 2.77; 95%CI 1.72–4.43, p<0.001; for carbapenem-resistant Pseudomonas aeruginosa [CRPA]: 3.8 vs. 2.0, p = 0.0073; third-generation cephalosporin resistant Escherichia coli [3GCREC]: 12.1 vs. 7.0, p<0.001; third-generation cephalosporin resistant Klebsiella pneumoniae [3GCRKP]: 12.2 vs. 5.4, p<0.001; carbapenem-resistant K. pneumoniae [CRKP]: 1.6 vs. 0.7, p = 0.045; and methicillin-resistant Staphylococcus aureus [MRSA]: 5.1 vs. 2.5, p = 0.0091). All-cause in-hospital mortality (%) following hospital-origin AMR BSI was not significantly different between TCHs and SCHs (all p>0.20). Due to the higher frequencies, all-cause in-hospital mortality rates following hospital-origin AMR BSI per 100,000 patient-days at risk were considerably higher in TCHs for most pathogens (for CRAB: 10.2 vs. 3.6,mortality rate ratio 2.77; 95%CI 1.71 to 4.48, p<0.001; CRPA: 1.6 vs. 0.8; p = 0.020; 3GCREC: 4.0 vs. 2.4, p = 0.009; 3GCRKP, 4.0 vs. 1.8, p<0.001; CRKP: 0.8 vs. 0.3, p = 0.042; and MRSA: 2.3 vs. 1.1, p = 0.023). In conclusion, the burden of AMR infections in some LMICs might differ by hospital type and size. In those countries, activities and resources for antimicrobial stewardship and infection control programs might need to be tailored based on hospital setting. The frequency and in-hospital mortality rate of hospital-origin AMR BSI are important indicators and should be routinely measured to monitor the burden of AMR in every hospital with microbiology laboratories in LMICs.

Introduction

Understanding and monitoring the burden of antimicrobial resistant (AMR) bacterial infection is important to design strategies for interventions [1]. A recent modelling study estimated that there are 1.27 million deaths attributable to AMR infections comparing to non-AMR infections in 2019 globally [2]. The study also highlighted the limited availability of data in LMICs, [2] where most of the currently available data were from university hospitals and tertiary-care hospitals (TCHs) [3, 4].

Multiple parameters are required for monitoring and evaluating the AMR burden in hospital settings. The proportions (%) of patients with growth of AMR strains of bacterial species (over total number of patients with growth of bacterial species; i.e. AMR proportion [7]) are commonly used to represent AMR burden [5, 6]. However, the AMR proportion alone cannot reflect the burden of AMR in absolute terms. For example, consider two hospitals with comparable activity, size and target population. Both hospitals may have the same proportion of bloodstream infections (BSI) due to methicillin-resistant Staphylococcus aureus (MRSA), which is 20%. However, the MRSA burden in hospital A would be considerably lower than that in hospital B, as hospital A has 10 MRSA cases out of 50 S. aureus BSI, whereas hospital B has 20 MRSA cases out of 100 S. aureus BSI. The frequencies of patients with AMR infections within a population during a reporting period (i.e. AMR frequencies [7]) are other important parameters. The AMR frequencies, typically expressing per 1000, 1,000 or 100,000, are also commonly used to monitor, evaluate and compare the AMR burden between hospitals or survey sites regardless of the size or services of the hospital [8–10].

AMR proportions and AMR frequencies are reported to be different by type and size of hospitals in some settings in high-income countries (HICs). In Spain, Oteo et al. reported that the proportion of MRSA is higher in hospitals with >500 beds than in those with <500 beds [11]. In Germany, Said et al. reported that the proportion of carbapenem-resistant Acinetobacter baumannii (CRAB) is higher in TCHs and secondary-care hospitals (SCHs) compared to outpatient clinics [12]. In the U.S., Gandra et al. reported that the proportion of AMR infections is not different between TCHs and small community hospitals [13]. The point prevalence survey of healthcare-associated infections (HAIs) in European acute care hospitals shows that HAI prevalence is highest in hospitals with ≥650 beds and lowest in those with <200 beds, [14] suggesting that frequency of HAI (per patients who were admitted to the hospital) is associated with hospital size.

We recently developed the AutoMated tool for Antimicrobial resistance Surveillance System (AMASS), an offline application to generate standardized AMR surveillance reports from routinely available microbiology and hospital data files, and independently tested the application in seven hospitals in seven countries [15]. The automatically generated reports stratify infections into community-origin and hospital-origin based on the recommendations of World Health Organization Global Antimicrobial Resistance and Use Surveillance System (WHO GLASS), and provide additional metrics on mortality involving AMR and non-AMR BSI [15]. Collaborating with Ministry of Public Health (MoPH) Thailand, we previously obtained and analysed microbiology and hospital admission data files of 60 hospitals from 2012 to 2015 in Thailand, and reported the burden of melioidosis, an infectious disease caused by Burkholderia pseudomallei, in Thailand [16].

The aim of this study was to examine the burden of AMR BSI in TCHs and compare that with SCHs using AMASS [15] on the microbiology and hospital admission data from 2012 and 2015 in Thailand. We also examine the burden of culture-confirmed notifiable bacterial diseases under evaluation using the reports generated by the AMASS.

Methods

Study setting

In 2012, Thailand had a population of 64.4 million, consisted of 77 provinces, and covered 513,120 km2. In each province, there is at least one SCH or TCH, [17] equipped with a microbiology laboratory capable of performing bacterial culture using standard methodologies for bacterial identification and susceptibility testing provided by the Bureau of Laboratory Quality and Standards, MoPH, Thailand [18]. The health systems in each province were integrated into 12 groups of provinces, known as health regions, plus Bangkok as health region 13. For example, SCHs generally referred patients to TCHs within the same health region. In 2012, there were 96 public hospitals (68 SCHs and 28 TCHs) in health regions 1 to 12 in Thailand.

Study design

We conducted a retrospective, multicentre surveillance study of all SCHs and TCHs in health regions 1 to 12 in Thailand. From the hospitals that agreed to participate, data were collected from microbiology and hospital admission data files between January 2012 and December 2015 as previously described [16]. Variables in the microbiology data file included patient hospital number (HN), specimen type, specimen collection date, culture result, and antibiotic susceptibility testing result, and each row contained information for each specimen. Variables in the hospital admission data file included HN, admission date, discharge date, and in-hospital discharge outcome, and each row contained information for each admission. Each data set was analysed using the AMASS v2.0, [19] and HN was used as a record linkage between the two data files of each hospital. The AMASS analysed the data and generated reports based on the recommendation of WHO GLASS [20]. A deduplication process was automatically conducted in which only the first isolate of a species per patient per specimen type per survey period was included in the report [15]. The AMASS v2.0 included an additional report on notifiable bacterial diseases (Annex A) and blood culture contamination rate (Annex B) (S1 Text in S1 File). The statistics in the AMR surveillance reports (in PDF and CSV format) were then extracted for analysis. The reports generated by the AMASS were validated using two methods: (a) checking data verification log files generated by the AMASS whether all information was imported accurately (e.g. total number of specimens, total number of hospital admissions, number of missing values, total number of isolates per organism in the raw microbiology data file, and total number of antibiotics being tested) [16] and (b) comparing the summary data and reported generated by the AMASS with data generated from manual calculations obtained from complete line listing of several organisms.

For AMR infections, we analysed the following organisms: CRAB and carbapenem-resistant Pseudomonas aeruginosa (CRPA), third-generation cephalosporin-resistant Escherichia coli (3GCREC) and Klebsiella pneumoniae (3GCRKP), carbapenem-resistant E. coli (CREC), carbapenem-resistant K. pneumoniae (CRKP) and MRSA which are in the WHO GLASS list of priority AMR bacteria [21] and are of local importance. Only blood culture isolates were included in the analysis.

Definitions

AMR BSI is defined as a case of infection in patients with blood culture positive for CRAB, CRPA, 3GCREC, 3GCRKP, CREC, CRKP or MRSA. Non-AMR BSI is defined as cases of infection in patients who had blood culture positive for carbapenem-susceptible A. baumannii (CSAB), carbapenem-susceptible P. aeruginosa (CSPA), third-generation cephalosporin-susceptible E. coli (3GCSEC) or K. pneumoniae (3GCSKP), or methicillin-susceptible S. aureus (MSSA).

Community-origin BSI was defined for patients with first positive blood specimens in the hospital taken within the first two calendar days of admission with calendar day one equal to the day of admission [20]. Patients with first positive blood specimens taken after the first two calendar days were categorized as cases of hospital-origin BSI. The classification of community-origin and hospital-origin BSI was performed within AMASS and based on specimen dates and hospital admission dates extracted from the microbiology and hospital data files, respectively.

The proportion of AMR (%) was calculated as the percentage of patients with new AMR BSI over all patients with new BSIs for each pathogen of interest during the reporting period [19]. The frequency of AMR BSI for each pathogen of interest was calculated as the total number of new patients with AMR BSI during the reporting period per 100,000 admissions (for community-origin BSI), per 100,000 patient-days at risk (for hospital-origin BSI), and per 100,000 tested population (for community-origin and hospital-origin BSI). In-hospital mortality (%) following AMR BSI for each pathogen of interest was calculated as the percentage of patients with new AMR BSI who died in the hospitals. In-hospital mortality rates following AMR BSI for each pathogen of interest were calculated as the total number of patients with new AMR BSI who died in the hospitals during the admission following AMR BSI per 100,000 admissions (for community-origin BSI) and per 100,000 patient-days at risk (for hospital-origin BSI).

In the AMASS v2.0, blood culture contamination is defined as isolation of one or more common commensal organisms; including Arcanobacterium spp., Arthrobacter spp., Bacillus spp. (except B. anthracis), Brevibacillus spp., Brevibacterium spp., Cellulomonas spp., Cellulosimicrobium spp., Corynebacterium spp. (except C. diphtheriae, C. jeikeium, C. pseudotuberculosis, C. striatum, C. ulcerans, and C. urealyticum), Cutibacterium spp., Dermabacter spp., Dermacoccus spp., Diphtheroids spp., Exiguobacterium spp., Geobacillus spp., Helcobacillus spp., Janibacter spp., Knoellia spp., Kocuria spp., Kytococcus spp., Leifsonia spp., Microbacterium spp., Micrococcus spp., Nesterenkonia spp., Paenibacillus spp., Propionibacterium spp., Pseudoclavibacter spp., Staphylococcus spp. (except S. aureus and S lugdunensis), Trueperella spp., Virgibacillus spp., and Viridans group Streptococci [18]. The blood culture contamination rate is defined as the ratio of the number of blood cultures with common commensal organisms over the total number of blood cultures.

The AMASS v2.0 also generated a summary report of culture-confirmed notifiable bacterial diseases caused by 11 pathogens including Brucella spp., B. pseudomallei, Corynebacterium diphtheriae, Neisseria gonorrhoeae, Neisseria meningitidis, Non-typhoidal Salmonella spp., Salmonella enterica serovar Paratyphi, Salmonella enterica serovar Typhi, Shigella spp., Streptococcus suis, and Vibrio spp.. The summary report included the total number and in-hospital mortality of patients with culture-confirmed notifiable bacterial diseases.

Statistical analysis

We compared proportion, frequency, mortality and mortality rate of AMR BSI between SCHs and TCHs for each pathogen of interest. We preliminarily compared AMR proportions and mortality using Chi-square or Fisher’s exact test when small samples (i.e. one or more expected values was <5 observations) and measurements from continuous variables using the Krustal-Wallis test. Then, we estimated the differences in proportions and mortality of AMR BSI between SCHs and TCHs using mixed-effect logistic regression models for patients nested within hospital using the STATA command xtlogit. We estimated the differences in frequency and mortality rate of AMR BSI between SCHs and TCHs using mixed-effect Poisson regression models for patients nested within hospital using the STATA command xtpoisson. We used mixed-effect models to estimate the fixed effects of hospital type while taking account for the random effects of patient within the same hospital. Multivariable mixed-effect models were also performed to control for other variables, including blood culture utilization rates of the hospitals and the health region where the hospitals located. These variables were included in the multivariable models because they could be associated with the hospital type and the outcome variables. We adjusted for blood culture utilization rate because if blood culture utilization rate was low, the observed AMR proportions could be higher than the true susceptibility profiles of pathogenic organisms [22]. Additionally, the observed AMR frequency per 100,000 admissions and per 100,000 patient-days could be lower than the true incidence rate of AMR infections [22]. We adjusted for health regions to control for potential variations in the AMR proportions and frequency across different regions. These regional differences could be influenced by other factors such as habits of antibiotic use and economic levels which were not explored in this study. There were no other potential confounders that were evaluated in the analysis. We then calculated total number and in-hospital mortality of patients with culture-confirmed notifiable bacterial diseases under evaluation. We used STATA (version 17.0; College Station, Texas) for the final statistical analyses and R version 4.0.5 for figures.

Ethical considerations

Ethical permission for this study was obtained from the Institute for the Development of Human Research Protection, the Ethics Committee of the Faculty of Tropical Medicine, Mahidol University (MUTM 2014-017-01). Written approval was given by the directors of the hospitals to use their routine hospital admission database for research. Consent was not sought from the individual patients as this was a retrospective study. This approach was approved by the Ethical and Scientific Review Committees.

Results

Baseline demographics

Ninety-five (99%) hospitals, out of 96 Thai hospitals that we approached, agreed to participate in the study. Twenty-five hospitals were excluded because either the microbiology or hospital admission data file was not available. Next, twenty-one hospitals were excluded because, in the microbiology data files, the antimicrobial susceptibility tests results were not available, readable or interpretable. Forty-nine hospitals were included in the analysis. A total of 35 hospitals (71%) had four years (from 2012 to 2015) of data available for analysis, four hospitals (8%) had three years, four hospitals (8%) had two years, and six hospitals (12%) had one year of data for analysis (S1 Table in S1 File).

Of 49 hospitals included in this study, 15 (31%) and 34 (69%) were TCHs and SCHs respectively (Fig 1). The median bed number was 672 (range 522–1,000) in TCHs and 335 (range 150–549) in SCHs (p<0.001). The median number of hospital admissions per year was 48,836 (range 30,409–98,428) in TCHs and 25,827 (range 7,221–62201) in SCHs (p<0.001). The total number of admissions was 3,134,815 in TCHs and 2,867,762 in SCHs. The blood culture utilization rate (median blood culture utilization rate 69 vs. 60 per 1,000 patient-days, respectively, p = 0.12) and the blood culture contamination rate (median blood culture contamination rates 4.1% and 3.6%, respectively, p = 0.94) were not statistically different between TCHs and SCHs. The all-cause in-hospital mortality among all patients admitted to TCHs was higher than those to SCHs (median in-hospital mortality 3.7% vs. 2.9%, respectively, p = 0.05).

Fig 1. Baseline demographics of 15 tertiary-care hospitals and 34 secondary-care hospitals in Thailand.

Fig 1

Open in a new tab

Each black dot represents the value reported by each hospital. Comparison was made using the Kruskal-Wallis test. The unit of analyses was a hospital. Blood culture (BC) utilisation rate (per 1,000 bed-days) and BC contamination rate (%) were estimated from 7 TCHs and 23 SCHs of which the microbiology data obtained included the culture results of “no growth”.

Proportion of AMR BSI

For community-origin BSI, there were differences in the proportions (%) of BSI being caused by AMR strains for some pathogens between TCHs and SCHs. The proportions of CRAB (36.9% vs. 25.0%, p = 0.041), 3GCREC (37.7% vs. 31.2%, p = 0.020) and 3GCRKP (24.4% vs. 19.2%, p = 0.053) were higher in TCHs than those in SCHs (Fig 2A and S2 Table in S1 File). The proportions of CRPA (18.4% vs. 17.1%), CREC (0.9% vs. 1.6%) and CRKP (2.3% vs. 2.7%) and MRSA (11.7% vs. 11.2%) were not significantly different between TCHs and SCHs.

Fig 2. Proportion (%), frequency (number of patients per 100,000 admissions), mortality (%) and mortality rate (number of deaths per 100,000 admission) of community-origin AMR BSI in 15 tertiary-care hospitals and 34 secondary-care hospitals in Thailand.

Fig 2

Open in a new tab

Each black dot represents the value reported by each hospital. The sizes of black dots are based on the total number of patients with blood culture positive for the bacterial species (row 1) and the total number of patients with blood culture positive for the AMR pathogens (row 2–4). Comparison was made using mixed-effect logistic or Poisson regression models of patients nested within hospitals.

For hospital-origin BSI, the proportions of AMR BSI for all pathogens, including CRAB (75.5% vs. 68.3%), CRPA (36.7% vs. 36.7%), 3GCREC (53.9% vs. 53.5%), 3GCRKP (63.7% vs. 57.3%), CREC (3.0% vs. 3.2%), CRKP (9.2% vs. 8.6%) and MRSA (37.4% vs. 28.0%), were not significantly different between TCHs and SCHs (Fig 3A and S2 Table in S1 File).

Fig 3. Proportion (%), frequency (number of patients per 100,000 patient-days at risk), mortality (%) and mortality rate (number of deaths per 100,000 patient-days at risk) of hospital-origin AMR BSI in 15 tertiary-care hospitals and 34 secondary-care hospitals in Thailand.

Fig 3

Open in a new tab

Each black dot represents the value reported by each hospital. The sizes of the black dots are based on the total number of patients with blood culture positive for the bacterial species (row 1) and the total number of patients with blood culture positive for the AMR pathogens (row 2–4). Comparison was made using mixed-effect logistic or Poisson regression models of patients nested within hospitals.

Similar findings were also observed in the multivariable models except that the proportion of CRAB for community-origin BSI was not significantly different between TCHs and SCHs (p = 0.66, S2 Table in S1 File).

Frequency of AMR BSI

We next calculated the frequency of AMR BSI in SCHs and TCHs. For community-origin BSI, of all pathogens under evaluation, 3GCREC had the single highest frequency of AMR BSI per 100,000 admissions in both TCHs and SCHs (Fig 2B and S3 Table in S1 File). The frequencies per 100,000 admissions of community-origin BSI caused by CRAB (14.2 vs. 4.8, p<0.001), and 3GCRKP (30.5 vs. 18.8, p = 0.0017) in TCHs were relatively higher than those in SCHs. The frequencies of CRPA (7.1 vs. 4.9), 3GCREC (142.3 vs. 108.4), CREC (2.9 vs. 4.7), CRKP (2.4 vs 2.2) and MRSA (14.0 vs. 10.1) were not significantly different between TCHs and SCHs.

Overall, the number of patient-days at risk of hospital-origin BSI included in the analysis was 12,341,585 in TCHs and 9,988,198 in SCHs. For hospital-origin BSIs, of all pathogens under evaluation, CRAB, 3GCREC and 3GCRKP were the top three organisms with the highest frequency in both TCHs and SCHs (Fig 3B and S4 Table in S1 File). Strikingly, the frequency of hospital-origin BSI per 100,000 patient-days at risk in TCHs was about twice that in SCHs for most of the pathogens under evaluation, including CRAB (18.6 vs. 7.0, incidence rate ratio [IRR] 2.77; 95%CI 1.72–4.43, p<0.001), CRPA (3.8 vs. 2.0, IRR 2.14; 95%CI 1.23–3.74, p = 0.0073), 3GCREC (12.1 vs. 7.0, IRR 1.80; 95%CI 1.29–2.50, p<0.001), 3GCRKP (12.2 vs. 5.4, IRR 2.23; 95%CI 1.57–3.17, p<0.001), CRKP (1.6 vs. 0.7, IRR 2.10; 95%CI 1.02–4.35, p = 0.045) and MRSA (5.1 vs. 2.5, IRR 1.88; 95%CI 1.17–3.01, p = 0.0091), except CREC (0.5 vs. 0.4, p = 0.39).

The frequencies of AMR BSI per 100,000 tested patients were calculated for 7 (47%) TCHs and 23 (68%) SCHs which provided results for blood cultures without growth (S5 Table in S1 File). Similar findings of difference between TCHs and SCHs were also observed, but wider 95%CI and larger p values were observed.

Similar findings were also observed in the multivariable models except that the frequency of CRAP and CRKP for hospital-origin BSI per 100,000 patient-days at risk was not significantly different between TCHs and SCHs (p = 0.068 and p = 0.11, respectively) (S3-S5 Tables in S1 File).

All-cause in-hospital mortality (%) following AMR BSI

Of 19,665 patients with AMR BSI, 10,858 (55.2%) and 8,807 (44.8%) were classified as community-origin and hospital-origin BSI, respectively (S6 Table in S1 File). Of 10,858 patients with community-origin AMR BSI, 2,873 (27.5%) died. Of 8,807 patients with hospital-origin AMR BSI, 3,874 (38.2%) died.

For both community-origin and hospital-origin BSIs, of all pathogens under evaluation, mortality (%) following AMR BSI caused by CRAB was higher than AMR BSI caused by other pathogens (Figs 2C and 3C). For both community-origin BSI and hospital-origin BSI, the differences in the mortality (%) following AMR BSI caused by all pathogens between TCHs and SCHs were not statistically significant.

Similar findings were also observed in the multivariable models except that the all-cause in-hospital mortality following AMR BSI caused by CRAB for hospital-origin BSI was higher in TCHs than that in SCHs (55.1% vs. 51.9%, p = 0.04, S6 Table in S1 File).

All-cause in-hospital mortality rate following AMR BSI

We next calculated all-cause in-hospital mortality rate following AMR BSI in SCHs and TCHs. For community-origin BSI, of all pathogens under evaluation, 3GCREC had the single highest mortality rate following AMR BSI per 100,000 admissions in both TCHs and SCHs (Fig 2D and S7 Table in S1 File). The mortality rate following community-origin CRAB and 3GCRKP BSI per 100,000 admissions in TCHs was relatively higher than that in SCHs (6.8 vs. 2.3; mortality rate ratio [MRR] 2.92, 95%CI: 1.65–5.19, p<0.001 and 8.8 vs. 4.9; MRR 1.79, 95%CI 1.28–2.52, p<0.001, respectively). The differences in the mortality rate following community-origin CRPA, CREC, CRKP and MRSA BSI per 100,000 admissions between TCHs and SCHs were not statistically significant.

For hospital-origin BSI, of all pathogens under evaluation, CRAB had the single highest mortality rate per 100,000 patient-days at risk in both TCHs and SCHs (Fig 3D and S8 Table in S1 File). Strikingly, the mortality rate following hospital-origin AMR BSI caused by most of pathogens under evaluation per 100,000 patient-days at risk was also about two to nearly three times those in SCHs; including CRAB (10.2 vs. 3.6; MRR 2.77, 95%CI 1.71–4.48, p<0.001), CRPA (1.6 vs. 0.8, p = 0.020), 3GCREC (4.0 vs. 2.4, p = 0.009), 3GCRKP (4.0 vs. 1.8, p<0.001), CRKP (0.8 vs. 0.3, p = 0.042) and MRSA (2.3 vs. 1.1, p = 0.023), though not for CREC (0.2 vs. 0.2, p = 0.56).

Similar findings were also observed in the multivariable models except that the mortality rate following community-origin 3GCREC BSI per 100,000 admissions in TCHs was relatively higher than in SCHs (33.4 vs. 23.7; adjusted MRR 1.21; 95%CI 1.03–1.41, p = 0.017) and the mortality rate following hospital-origin CRPA and CRKP BSI per 100,000 admissions in TCHs was not significantly different between TCHs and SCHs (adjusted MRR 1.97; 95%CI 0.83–4.63, p = 0.12, and adjusted MRR 1.81; 95%CI 0.80–3.63, p = 0.098, respectively).

Notifiable bacterial diseases

Utilizing microbiology data and hospital admission data, we also calculated total number and in-hospital mortality of patients with culture-confirmed notifiable bacterial diseases indicated in the National Notifiable Disease Surveillance system (Report 506) of Thailand, [23] in the 49 hospitals from 2012 to 2015 (S9 Table in S1 File). The disease with the highest total number of cases was non-typhoidal Salmonella (NTS) infection (n = 11,264 patients), followed by melioidosis (an infection caused by B. pseudomallei; n = 6,164 patients) and Vibrio spp. infections (n = 2,143 patients).

The disease with the highest total number of all-cause in-hospital deaths was melioidosis (n = 1,524 patients), followed by NTS infection (n = 1,005 deaths), Vibrio spp. infection (n = 172 deaths), Streptococcus suis infection (n = 85 deaths), Corynebacterium diphtheriae infection (n = 9 deaths), Shigella spp. infection (n = 7 deaths), Salmonella enterica serovar Paratyphi infection (n = 4 deaths), S. enterica serovar Typhi infection (n = 3 deaths), and Neisseria meningitidis infection (n = 3 deaths). None of 60 and 4 patients with culture-confirmed N. gonorrhoeae infection and Brucella spp. infection died in the hospital.

Discussion

This study compared proportions, frequencies, all-cause in-hospital mortality and all-cause in-hospital mortality rate following AMR BSI between TCHs and SCHs in a LMIC. We show that the observed frequency of and all-cause in-hospital mortality rates following AMR BSI (per 100,000 admissions and per 100,000 patient-days) were considerably higher in TCHs than those in SCHs. The differences were also observed in the multivariable model adjusted for some confounders. The results support the needs to design antimicrobial stewardship (AMS) and infection prevention and control (IPC) in LMICs base on hospital size and type.

Our study highlights the capability of hospitals and national authorities in LMICs to estimate AMR frequencies as another crucial parameter to monitor the effectiveness of AMR interventions. For example, effectiveness of a nationwide intervention in Israel is shown by the reduction in frequency of nosocomial carbapenem-resistant Enterobacteriaceae infections from 55.5 to 11.7 cases per 100,000 patient-days [24]. An 80% reduction in MRSA BSI in England (defined as reported cases per 100,000 population per year and as reported cases per 100,000 bed-days) after a major public health infection prevention campaign also demonstrates the potential impact of coordinated interventions [25, 26].

The higher proportion (%) and frequencies of community-origin AMR BSI for CRAB and 3GCRKP in TCHs compared to those in SCHs could be due to higher proportion of healthcare-associated infections in TCHs. It is likely that TCHs have a higher proportion of patients who are transferred from other hospitals (including from SCHs), who are receiving health care at end-stage renal facilities or long-term care facilities, and who are recently discharged from the hospitals. Routine hospital admission data used in our study could not identify those conditions; therefore, those patients with BSI were categorized as community-origin BSIs in our reports. These findings suggest that TCHs may need to strengthen AMS and IPC, particularly on new inpatients who are at high risk of healthcare-associated infections.

No or minimal difference in proportion (%) of hospital-origin AMR BSI between TCHs and SCHs could be because the AMR prevalence in a country is likely driven by the contagion of AMR organisms within and between hospitals [27]. Although it is possible that higher use of antibiotics in TCHs may drive the emergence of AMR organisms (e.g. emergence of CREC and CRKP) and lead to the higher proportion of AMR BSI compared to those in SCHs, [28] that was not observed in our setting during the study period.

The strikingly higher frequencies of hospital-origin AMR BSI in TCHs than those in SCHs are likely caused by higher proportion of patients who had severe conditions or compromised immune systems, or required complex surgery, prolonged intubation or urinary catheters [11, 14]. The proportion of ICU beds in TCHs is also higher. Those conditions are likely driving both AMR and non-AMR hospital-acquired infections in our setting as shown by no or minimal difference in proportion of hospital-origin AMR BSI between TCHs and SCHs.

No or minimal difference in all-cause in-hospital mortality (%) following AMR BSI between TCHs and SCHs could be because care and antibiotics to be used against AMR BSI are not different between TCHs and SCHs in Thailand.

The higher all-cause in-hospital mortality rate following hospital-origin AMR BSI in TCHs than those in SCHs is, therefore, caused by higher frequencies of AMR BSI. These findings suggest that healthcare workers in TCHs will need to strengthen AMS and IPC, particularly among those are at high risk of hospital-acquired infections (HAI), more than those in SCHs. This is also because patients in TCHs are likely to be more complex than those in SCHs, and, as such, are at higher risk of HAI than those in SCHs.

For hospital-origin AMR BSI in SCHs, although CRAB, 3GCREC and 3GCRKP are the top three organisms with the highest frequencies, CRAB has the highest all-cause in-hospital mortality rate due to the higher all-cause in-hospital mortality (%) following CRAB BSI compared to those following 3GCREC BSI and 3GCRKP BSI.

Our findings of a high number of cases and all-cause in-hospital deaths following nontyphoidal Salmonella disease, [29, 30] melioidosis, [16, 31] S. suis infection, [32, 33] and Vibrio spp. Infection [34, 35] are consistent with previous research. Our study demonstrates that national statistics on multiple national notifiable bacterial diseases in LMICs could be improved by integrating information from readily available databases.

The strength of this study is that routine data used to compare AMR burdens between different hospital settings are from multiple sites in Thailand. Moreover, we have shown that the use of an automated surveillance system can readily generate useful statistics to understand AMR within a hospital and between hospitals in a country, and this empowers collaborative work and analyses across different settings both nationally and globally. The collaborative effort is essential to inform global burdens of AMR, which in turn are important statistics to support public health strategies to control spread of AMR and set priorities in resource allocation locally.

Our study has some limitations. Firstly, the findings from the study may not be generalisable to all LMICs. The differences in proportions, frequencies, all-cause in-hospital mortality and all-cause in-hospital mortality rates following AMR BSI observed in SCHs and TCHs could be confounded by many factors such as difference in patient characteristics, diagnostic stewardship (particularly blood culture utilization [22, 36]) and patient management (including AMS and IPC). For example, the differences in frequency and in-hospital mortality rate observed in our study could be confounded by the differences in the case mix between SCHs and TCHs. It is also possible that the AMS and IPC practices in Thailand are better in SCHs compared to TCHs due to various reasons. These data were not available because our study was not designed to collected those data. The aim of this study was to examine the burden of AMR BSI observed in TCHs and compare that observed in SCHs using AMASS [15]. To understand the size of effect caused by the difference in hospital type alone, further studies using detailed clinical data, AMS and IPC data, and rigorous statistical approaches are needed. It is also possible that the AMS and IPC has already been tailored base on hospital size and type already. Regardless, our findings emphasize the need for further evaluation and improvement of all actions against AMR infections (including AMS and IPC) in TCHs, going beyond the current practices. Secondly, the sample size of SCHs and TCHs included in this study is small, and this may have limited the power to detect differences for organisms that are less predominate such as CRKP in the study setting. Thirdly, in this study we only included patients who were hospitalised and in-hospital mortality. Patients who had blood cultures taken either at community hospitals or the study hospitals but not hospitalised at the study hospitals were not included in the analysis. Fourthly, the mortality and mortality rate associated with AMR BSI reported are all-cause in-hospital mortality and mortality rate, and could be underestimated because some people, in the study area, preferred to die at home and were discharged against advice. The all-cause in-hospital mortality could also overestimate the impact of AMR infection. This is because a proportion of patients with AMR infection might die of other causes after the successful treatment of the AMR infection. This secondary data analysis was not designed to answer causal question but to demonstrate the potentials of using AMASS as a tool to empower routine AMR surveillance data understanding and sharing. Fifthly, the distinction between primary and secondary BSIs cannot be made without detailed, linked clinical and microbiology data. The BSIs in this study include both primary and secondary BSIs. Future surveillance systems should consider innovative methods that can integrate such information to ensure that these distinctions can be accurately made in LMICs. In addition, future studies with robust and comprehensive data collections that can capture all aspect of patient details, including the previous admission history and the previous culture results, the current diagnostic stewardship, AMS and IPC, and other confounding factors, are still needed.

In conclusion, we observed that the burden of AMR infections in TCHs is higher than that in SCHs in Thailand. This might also be occurring in other LMICs. Our results support similar evaluations in other LMICs and highlight the importance of tailoring infection control strategies based on hospital size and type, particularly when significant differences in the burden are observed. The frequency and in-hospital mortality rate of hospital-origin AMR BSI are important indicators and should be routinely measured to monitor the burden of AMR in every hospital with microbiology laboratories in LMICs.

Supporting information

S1 File

(DOC)

pone.0303132.s001.doc (308KB, doc)

Acknowledgments

We gratefully acknowledge the directors, epidemiological and laboratory team of general and regional hospitals for providing microbiological and hospital admission data and their administrative supports. The general and regional hospital network are comprised of Chorkaew Yangyuen (Samutprakarn Hospital), Aphinya Singkhongsin, Chanchira Chaichaem (Pranangklao Hospital), Phkaiwan Kropsanit (Pathumthani Hospital), Winai Suphapphot, Chaiwat Khaokaeo (Sena Hospital), Sasi Sichot (Phra Na Khon Sri Ayutthaya Hospital), Ratri Chalaemphak, Benchawan Khaisongkhram (Angthong Hospital), Praphon Chinthanu, Phutthakhun Wongsuwan (Banmi Hospital), Pritsana wongnoi (Pranarai Maharaj Hospital), Witthaya Yotngoen (Singburi Hospital), Khongsak Sueachoi (Inburi Hospital), Pranom Chantharat (Jainad Narendra Hospital), Sangsan Sinbamrung, Duangkamon Chiratrachu (Phra Phutthabat Hospital), Waranya Sichanta (Saraburi Hospital), Panatda Thipruecha (Chonburi Hospital), Piyaphatcha Phongprasoet, Panatda Inphrom (Rayong Hospital), Pakkawi Siphueak, Ratchani Thamchamrat (Prapokklao Hospital), Phuangphikun PhonPrasit (Trat Hospital), Bunga Chanlee (BuddhaSothorn Hospital), Wiphawadi Dongchan (Chao Phya Abhaibhubejhr Hospital), Haruethai Khunothai (Nakhonnayok Hospital), Atchara Ampere (Somdet Phrayupharacha SaKaeo), Saifon Sutchai, Prayut Kaeomalang, Nonglak Prayunsoet (Maharat Nakhon Ratchasima Hospital), Ratana Chiracharuporn (Buriram Hospital), Suriya Senthong (Surin Hospital), Sunthon Romniyaphet (Sisaket Hospital), Jintana Kanchanabat, Praweennuch Watanachaiprasert, Thanasith Sananmuang (Sunpasitthiprasong Hospital), Somphon Chankaeo (Yasothon Hospital), Wiraphon Khwamman (Chaiyaphum Hospital), Kraison Bunsam, Phonnatcha Katiwong (Amnat Charoen Hospital), Kanyaphak Phanchampa (Buengkan Hospital), Sutthiphong Phonbun (Nongbualamphu Hospital),Marisa Uton, Thitiphan Khunphu (Sirindhorn Hospital of Khon Kaen Province), Kriangkri Kongsuk, Ritthikorn DongLuang (Khon Kaen Hospital), Suthep Thipsawang, Kochnipa Kwawong, Nida Thanaphatphairot (UdonThani Hospital), Suphattra Likrachang, Kirana Phakdiburikun (Loei Hospital), Suphakon Saenthamphon (Nongkhai Hospital), Suchitra Nasingkhan, Kanchanaphon Taratai (Mahasarakham Hospital), Witthaya Ratmaet (Roiet Hospital), Natthasorn Chawaninthawisut (Kalasin Hospital), Phuwanat PhothiChai (Sakonnakhon Hospital), Phinthip Saiklang (Nakhonphanom Hospital), Yutphon Mankhong, Yothin Tairayawong (Mukdahan Hospital), Warawan Inthip (Nakornping Hospital), Dr. Thiraphong Tatiyaphonkunthiraphong (Lamphun Hospital), Thirin Ketwichit (Lampang Hospital), Yaowalak ChanDaeng (Uttaradit Hospital), Phana Thatsanawaythit, Itsareeya Boonrat (Phrae Hospital), Sopha Itsaranarongphan (Nan Hospital), Sanong Chaisue, Phanarat Phuangmali (ChiangKham Hospital), Chirawan Sithongphim (Phayao Hospital), Satorn Charatdamrongwat (Chiangrai Prachanukroh Hospital), Jintana Phothip, Duangdi Chomphu (Srisangwan Hospital), Ladda Raden (Sawanpracharak Hospital), Phitya Hema, Kanthika Ocharot, Mongkhon Uising (Uthaithani Hospital), Narong Mahayot (Kampangpetch Hospital), Onraphin Thiwai (Maesot Hospital), Preeyada Triprawat (Somdejprajaotaksin Maharaj Hospital), Kreangkrai Chatsut, Yuppharet Kaewprasern (Srisangworn Sukhothai Hospital), Meena Nakhon (Sukhothai Hospital), Aphinya Innoi, Thoranin Rakthanabodee (Buddhachinaraj hospital), Siwaphorn Phongchin (Phichit Hospital), Jintana Phonphraram (Phetchabun Hospital), Somphon Niamlang (Damnoensaduak Hospital), Thanya Surakhamsang (Banpong Hospital), Thanyalak Borirak (Photharam Hospital), Nopphon Siangchin (Ratchaburi Hospital), Ratchani Watthanayaem, Suwan Manutchan (Pahonpol Payuha Sena Hospital), Ekachai Photnanthawong (Makarak Hospital), Dr. Pornsak Thirathonbun (SomdejPrasangkharach17 Hospital), Narong Wongkanha (Chaophraya Yommarat Hospital), Pitchayakhanid Yaemsoun, Pongphon Roeknaowarat (Nakhonpathom Hospital), Witthaya Sithong (Samutsakorn Hospital), Nonsi Sonthiyat (Kratumban Hospital), Narongchai SiamPhairi (Somdej Prabuddha Lertla Hospital), Chanthana Kalanuwat (Prajomklao Hospital), Saran Songsaeng (HuaHin Hospital), Chirawan Bunchusi (Maharaj Nakhonsithammarat Hospital), Chitchanin Niyomthai, Chutima Phayayam (Krabi Hospital), Boribun Chensamut (Takuapa Hospital), Aphisit Suwannarat, Suwanni Khwankhao, Phityaporn Chunchu (Phang Nga Hospital), Phatcharin Yatraksa (Vachira Phuket Hospital), Jittima Thongnak (Koh Samui Hospital), Ratchanok Withunphan (Suratthani Hospital), Chuenkhwan Kaeowichit, Phimnisa Phet (Ranong Hospital), Kritsanee Wichitakun (Chumporn Ketudomsak Hospital), Sakda Khaophong (Songkla Hospital), Wichian Patangkaro, Nattawan Chanmueang (Hat Yai Hospital), Nittaya Sakunsanti (Satun Hospital), Suriwan Phaksuphara, Umaporn Sina (Trang Hospital), Witthaya Wunchum (Phatthalung Hospital), Sirinthon Wongyoksuriya (Pattani Hospital), Mawin Deae, Ananni Sama (Betong Hospital), Wichai Wanmueang, Suphattra Mahachot (Yala Hospital), Arun Phutkaeo, Yukalipli Kaseng, Khodiyo Yamai (Narathiwat Ratchanakarin Hospital), Praphai Krirat, Naruemon Bunsiri (Sungai Kolok Hospital). We thank Saman Sayumphuruchinan (ED, MoPH), Wanwisa Khammak (Strategy and Planning Division, MoPH), Sittikorn Rongsumlee (MORU) and Prapass Wanapinij (MORU) for administrative and data management supports.

Data Availability

All relevant data are within the manuscript and its Supporting Information files. The study used openly anonymous AMR surveillance reports generated from each hospital which are available at: https://doi.org/10.6084/m9.figshare.20318193.

Funding Statement

The study was supported by the DDC, MoPH, Thailand, and Defense Threat Reduction Agency (DTRA), U.S.. This research was funded in part by the Wellcome Trust (224681/Z/21/Z and Wellcome Trust Institutional Translational Partnership Award-MORU). CL is supported by the Wellcome Trust (106680/B/14/Z). BS is supported by a grant from the UK Department of Health and Social Care using UK aid funding managed by the Fleming Fund (R52354 CN001). For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

Decision Letter 0

Ali Amanati

1 Dec 2023

PONE-D-23-31373Frequency and mortality rate following antimicrobial-resistant bloodstream infections in tertiary-care hospitals compared with secondary-care hospitalsPLOS ONE

Dear Dr. Limmathurotsakul,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jan 15 2024 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ali Amanati

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We suggest you thoroughly copyedit your manuscript for language usage, spelling, and grammar. If you do not know anyone who can help you do this, you may wish to consider employing a professional scientific editing service.

Whilst you may use any professional scientific editing service of your choice, PLOS has partnered with both American Journal Experts (AJE) and Editage to provide discounted services to PLOS authors. Both organizations have experience helping authors meet PLOS guidelines and can provide language editing, translation, manuscript formatting, and figure formatting to ensure your manuscript meets our submission guidelines. To take advantage of our partnership with AJE, visit the AJE website (http://learn.aje.com/plos/) for a 15% discount off AJE services. To take advantage of our partnership with Editage, visit the Editage website (www.editage.com) and enter referral code PLOSEDIT for a 15% discount off Editage services.  If the PLOS editorial team finds any language issues in text that either AJE or Editage has edited, the service provider will re-edit the text for free.

Upon resubmission, please provide the following:

The name of the colleague or the details of the professional service that edited your manuscript

A copy of your manuscript showing your changes by either highlighting them or using track changes (uploaded as a *supporting information* file)

A clean copy of the edited manuscript (uploaded as the new *manuscript* file).

3. During our evaluation of the documents provided, we noted that the data were accessed for research purposes in November 2022, but this was after the project period had expired on the IRB documents supplied. Please could you provide the ethics approval extension documents for the study. If the document is in another language, please also provide an English translation. Please note that if these documents are not included when your manuscript is resubmitted, it may be rejected.

4. Thank you for stating the following financial disclosure:

“The study was supported by the DDC, MoPH, Thailand, and Defense Threat Reduction Agency (DTRA), U.S.. This research was funded in part by the Wellcome Trust (224681/Z/21/Z and Wellcome Trust Institutional Translational Partnership Award-MORU). CL is supported by the Wellcome Trust (106680/B/14/Z). BS is supported by a grant from the UK Department of Health and Social Care using UK aid funding managed by the Fleming Fund (R52354 CN001). For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission.”

Please state what role the funders took in the study.  If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

5. We noted in your submission details that a portion of your manuscript may have been presented or published elsewhere. [No. This work has not been previously published elsewhere. A pre-print version of this manuscript has been posted on medRxiv (https://www.medrxiv.org/content/10.1101/2023.02.07.23285611v2).] Please clarify whether this publication was peer-reviewed and formally published. If this work was previously peer-reviewed and published, in the cover letter please provide the reason that this work does not constitute dual publication and should be included in the current manuscript.

6. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

7. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

Additional Editor Comments:

Dear authors

Your manuscript [PONE-D-23-31373] has passed the review stage and is ready for ‎revision. ‎

Editorial comments

To ensure the Editor and Reviewers can recommend that your revised manuscript be ‎accepted, ‎‎please pay careful attention to each comment posted underneath ‎this email. This way we ‎can ‎avoid future clarifications and revisions, moving swiftly to ‎a decision.‎

‎1. Please provide a point-by-point response to the Editor and reviewer's comments

‎2. Please highlight all the amends on your manuscript with yellow color

‎3. Some grammatical and spacing errors need to be revised throughout the ‎manuscript

Editor comments:

There are significant objections to the statistical tests applied in this study.‎

As the second reviewer has pointed out, multivariable modeling ‎needs to be considered, and as the fourth reviewer has ‎pointed out, the significance of the obtained findings may be weakened ‎substantially without controlling potential confounders for investigating differences between hospital and also investigating factors associated with mortality. Please noticed that without considering advanced statistical revision I am unable to consider the manuscript further for peer review.‎

Truly yours

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Partly

Reviewer #4: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

Reviewer #4: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thanks to the research team for their valuable work. Please find my comments below:

1. In lines 220 to 227, you mentioned the hospitals included in the study, stating that out of 96 hospitals, 46 were excluded under certain conditions, and 49 studies were analyzed in your research, while it should have been 50. Please provide an explanation for this discrepancy regarding one study.

2. In the statistical section, it is necessary to clarify which variables have fixed effect(s) and which have random effect(s) in your mixed-effects logistic model.

3. Please be aware that p-values only indicate the level of significance, not the strength of the relationship. Therefore, consider this in all your tables and replace all values of <0.0001 with <0.001. NOTICE: "When p-values are displayed as very small numbers (e.g., 0.0001), it can be challenging for readers to quickly grasp their significance. It's more common to use the conventional format of displaying p-values as three decimal places (e.g., 0.001) for ease of interpretation."

4. The quality of your figures is significantly low, or they did not meet the desired quality in the submitted version, and I could not extract any useful information from them.. Please ensure that the figures meet the required quality standards or provide better versions in the submitted manuscript.

5. In the references section, you've used reports as sources. Please include relevant links for each reference to facilitate readers' access to the sources in your article.

Reviewer #2: I am impressed by the large amount of data collected during the study period 2012 to 2015; however, I recommend rejection due to the limited analysis performed and the similarities with previous studies.

Reviewer #3: Dear Authors

Kindly address the following few concerns

Line 154 correct the grammatical error " local importance" instead of "locally importance."

Line 215, second sentence is unclear, kindly revise.

LINE 398 "This is likely occurring in other LMICs” as part of the conclusion is too much of extrapolation from the current study under review. Rephrasing it to eg “This may be situation in other LMICs, or This might also be occurring in other LMICs” will be more scientifically acceptable.

Additional concerns

1. please include a section/paragraph that describes challenges or weakness or limitations of this study.

2. Are the patients' demographics accessible? We could make more meaningful inference from the comparative analysis with the demographic data. If this is not available, kindly include it in the weakness section.

Reviewer #4: Review of the manuscript frequency and mortality rate following antimicrobial-resistant blood stream infection in tertiary care hospital compared with the secondary care hospital.

This study addresses an important topic comparing the burden of antimicrobial resistant bloodstream infections between tertiary care and secondary care hospitals in Thailand. The study demonstrates the utility of using automated tools like AMASS for analyzing routine surveillance data to generate useful statistics on AMR infections. However, there are some limitations that should be acknowledged:

Major Comments:

1. More details needed on the data analysis plan - which specific metrics were compared between tertiary and secondary care hospitals and what statistical tests were used.

2. The results section overall is well-written, but could benefit from some sub-headings to improve readability by breaking it down into logical sections.

3. One limitation as acknowledged is that the findings may not be generalizable to other LMIC settings, so it may be useful to explicitly state this in the conclusion as well.

4. The paper does not provide information on the specific criteria used to classify community-origin and hospital-origin bloodstream infections (BSI), which may affect the accuracy of the results. Based only on admission date fails to account for healthcare exposures prior to admission. This could inflate community proportions at teaching hospitals receiving many transfers e.g. Patient with MRSA blood stream infection from previous hospitalization. More granular exposure data is needed

5. Designation of the Health care originated or associated blood stream infection relies on the fact that cultured organism should not be isolated from any site other than blood in order to be classified as BSI(blood stream infection) (www.cdc.gov/nhsn/pdfs/pscmanual/4psc_clabscurrent.pdf) during the infection window period. Absence of information about the culture results during the infection window period (with same organism) from sites other than blood can overestimate the proportion of blood stream infections

6. Major potential confounders like antibiotic usage, infection control practices are not adjusted for. This limits the ability to draw causal inferences about differences between hospital types.

7. Crude mortality rates do not adjust for important risk factors and likely oversimplify differences between complex tertiary vs general hospitals. More rigorous statistical approaches are needed to control for confounding

8. Relying on routinely collected data presents various quality issues like missing/inconsistent variables, coding errors, and variability in diagnostic/prescribing practices between sites. This reduces ability to accurately measure and compare burdens.

9. No validation is reported of automated classification (e.g. AMR vs susceptible). Misclassification bias could differentially affect estimates between settings

Minor Comments:

1. Standardize the formatting of abbreviations - write out full forms initially before introducing abbreviations.

2. Carefully proofread the manuscript to correct any minor grammar, spelling or punctuation errors.

3. Sample sizes: with only 15 TCHs and 34 SCHs, the study may be underpowered to detect some differences, especially for less common organisms. Larger multi-year datasets could provide more robust comparisons.

4. The study does not provide information on the specific interventions or strategies that could be implemented to address the differences in AMR burden between TCHs and SCHs.

5. The term AMR is not defined clearly whether it refers to an organism resistant to multiple families of antibiotics or just one

6. Mortality measures: all-cause in-hospital deaths only capture a fraction of outcomes and underestimate AMR impacts. Longer term and attributable mortality should also be examined.

7. Proportional measures like AMR proportions do not account for differences in total case volumes between hospitals. Higher proportions may still represent lower absolute burdens at smaller hospitals. Frequencies provide more clinically meaningful comparisons but are still prone to biases from differences in populations served.

8. Focus is on bacteremia but AMR impacts many infection types. Generalizing findings may overlook reservoir/transmission dynamics across all healthcare-associated infections.

Specific Comments:

1. Line 95-100 : The concept provided is misinterpreted , actually the reason for expressing the rates in 100s, 1000s or even 100000 is to provide the utility of comparison of rates across different locations and cohorts regardless of the size or services of the hospital https://www.cdc.gov/STD/Sassi/Module2/expression_of_rates.html

2. Line 181-192: it is not clear whether the BSI with common commensals are included in the analysis or not. according to CDC, 2 or more blood samples collected with common commensals and clinical picture constitutes BSI www.cdc.gov/nhsn/pdfs/pscmanual/4psc_clabscurrent.pdf

3. Line 194-200: the statistical analysis portion need elaboration as which of the strata of have less than 5 observations. There is mention of many statistical analysis. Author needs to clarify the data type used and rationale of using the statistical analyses.

4. Line 305-319: the text is out of the context of study as the study is about the frequency , proportion and burden of blood stream infections

5. Line 324-326: there is no assessment of AMS program or the infection control program for the participating facilities. So author cannot suggest if the practices are not tailored according to the scope of services already.

6. Line 338-343: this highlight the need of robust and comprehensive data collections that captures all aspects of patient detail including the previous admission history and the previous culture results

7. Line 344-349: no reference or bases provided for the assumptions

8. Line 354-256: contradicting statement to the earlier statement in line 324-326

9. Line 390-392: it could be overestimated as well. Since only the first isolate is taken into account there is no knowledge of how well the patient responding to the antibiotics and could die of cause other than BSI.

In summary, while a pragmatic first step, the methods have limitations for making robust policy inferences about AMR burdens in different hospital contexts. Greater standardization and more sophisticated statistical analyses are needed. Prospective collection of clinical and microbiological data would allow for more rigorous analyses of differences in AMR burdens and outcomes between hospital settings. Overall the manuscript makes a useful contribution on an important topic. Addressing the comments above would further improve it. I hope these suggestions are helpful for the authors. Please feel free to let me know if you need any clarification or have additional questions!

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Reviewer #4: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Attachment

Submitted filename: Referee_answer.pdf

pone.0303132.s002.pdf (132.7KB, pdf)
Attachment

Submitted filename: review PONE-D-23-31373.docx

pone.0303132.s003.docx (18.9KB, docx)
PLoS One. 2024 May 20;19(5):e0303132. doi: 10.1371/journal.pone.0303132.r002

Author response to Decision Letter 0

15 Feb 2024

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Response: Revised as suggested.

2. We suggest you thoroughly copyedit your manuscript for language usage, spelling, and grammar. If you do not know anyone who can help you do this, you may wish to consider employing a professional scientific editing service.

Whilst you may use any professional scientific editing service of your choice, PLOS has partnered with both American Journal Experts (AJE) and Editage to provide discounted services to PLOS authors. Both organizations have experience helping authors meet PLOS guidelines and can provide language editing, translation, manuscript formatting, and figure formatting to ensure your manuscript meets our submission guidelines. To take advantage of our partnership with AJE, visit the AJE website (http://learn.aje.com/plos/) for a 15% discount off AJE services. To take advantage of our partnership with Editage, visit the Editage website (www.editage.com) and enter referral code PLOSEDIT for a 15% discount off Editage services. If the PLOS editorial team finds any language issues in text that either AJE or Editage has edited, the service provider will re-edit the text for free.

Upon resubmission, please provide the following:

The name of the colleague or the details of the professional service that edited your manuscript

A copy of your manuscript showing your changes by either highlighting them or using track changes (uploaded as a *supporting information* file)

A clean copy of the edited manuscript (uploaded as the new *manuscript* file).

Response: Revised as suggested.

3. During our evaluation of the documents provided, we noted that the data were accessed for research purposes in November 2022, but this was after the project period had expired on the IRB documents supplied. Please could you provide the ethics approval extension documents for the study. If the document is in another language, please also provide an English translation. Please note that if these documents are not included when your manuscript is resubmitted, it may be rejected.

Response: The study (MUTM 2014-017-01) was approved on 4 March 2015, and was extended for 5 rounds until 3 March 2020. The documents confirming the extension of the study approval until 2020 in English have been attached. The study was closed as per protocol of the Ethics Committee of the Faculty of Tropical Medicine, Mahidol University, that the study could be closed when all data collection had been completed. Data analysis and manuscript submission could be performed after the study closure. A letter of confirmation of the discussion about data analysis regarding this study (MUTM 2014-017-01) from Ethics Committee of the Faculty of Tropical Medicine, Mahidol University has been attached.

4. Thank you for stating the following financial disclosure:

“The study was supported by the DDC, MoPH, Thailand, and Defense Threat Reduction Agency (DTRA), U.S.. This research was funded in part by the Wellcome Trust (224681/Z/21/Z and Wellcome Trust Institutional Translational Partnership Award-MORU). CL is supported by the Wellcome Trust (106680/B/14/Z). BS is supported by a grant from the UK Department of Health and Social Care using UK aid funding managed by the Fleming Fund (R52354 CN001). For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission.”

Please state what role the funders took in the study. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

Response: We have added a statement in both the manuscript and the cover letter to acknowledge the role of the funders in the study.

5. We noted in your submission details that a portion of your manuscript may have been presented or published elsewhere. [No. This work has not been previously published elsewhere. A pre-print version of this manuscript has been posted on medRxiv (https://www.medrxiv.org/content/10.1101/2023.02.07.23285611v2).] Please clarify whether this publication was peer-reviewed and formally published. If this work was previously peer-reviewed and published, in the cover letter please provide the reason that this work does not constitute dual publication and should be included in the current manuscript.

Response: We confirm that the pre-print version posted on medRxiv was not peer-reviewed or formally published hence this work does not constitute dual publication. The medRxiv (https://www.medrxiv.org/content/about-medrxiv) is a free online archive for unpublished manuscripts (preprints) in medical, clinical and related health science. The current submitted manuscript has not been formally published in any platform. The below statement to clarify this is added in the cover letter:

“A pre-print version that was not peer-reviewed or formally published of this manuscript has been posted on medRxiv (https://www.medrxiv.org/content/10.1101/2023.02.07.23285611v2)”

6. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

Response: DOIs to access our analysis data is now added in the main manuscript as below:

“The anonymous AMR surveillance reports generated from each hospital are open-access and available at https://figshare.com/s/c028f157c18a3cc06a82.”

7. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

Response: The captions for the supporting information files have been added at the end of the manuscript.

Additional Editor Comments:

Dear authors

Your manuscript [PONE-D-23-31373] has passed the review stage and is ready for ‎revision. ‎

Editorial comments

To ensure the Editor and Reviewers can recommend that your revised manuscript be ‎accepted, ‎‎please pay careful attention to each comment posted underneath ‎this email. This way we ‎can ‎avoid future clarifications and revisions, moving swiftly to ‎a decision.‎

‎1. Please provide a point-by-point response to the Editor and reviewer's comments

‎2. Please highlight all the amends on your manuscript with yellow color

‎3. Some grammatical and spacing errors need to be revised throughout the ‎manuscript

Editor comments:

There are significant objections to the statistical tests applied in this study.‎

As the second reviewer has pointed out, multivariable modeling ‎needs to be considered, and as the fourth reviewer has ‎pointed out, the significance of the obtained findings may be weakened ‎substantially without controlling potential confounders for investigating differences between hospital and also investigating factors associated with mortality. Please noticed that without considering advanced statistical revision I am unable to consider the manuscript further for peer review.‎

Truly yours

Response: We appreciate invaluable suggestion and feedback from the editor and reviewers. We have now added multivariable mixed-effect regression models, as suggested. After adjusted for potential confounders, our findings still show differences in the frequency and mortality rates of AMR BSI between tertiary-care and secondary-care hospitals. This additional evidence supports our conclusion that targets of and resources for antimicrobial stewardship and infection control programs in LMICs might need to be tailored based on hospital type and size, as burden of AMR infections might differ by hospital setting. We have added the results from the multivariable models in both the main text of the manuscript and supporting documents.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Partly

Reviewer #4: Partly

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

Reviewer #4: No

________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

________________________________________

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thanks to the research team for their valuable work. Please find my comments below:

1. In lines 220 to 227, you mentioned the hospitals included in the study, stating that out of 96 hospitals, 46 were excluded under certain conditions, and 49 studies were analyzed in your research, while it should have been 50. Please provide an explanation for this discrepancy regarding one study.

Response: We approached 96 hospitals and 95 of them agreed to participate in the study (Line 217). Of these 95 hospitals, 46 were excluded and 49 were analysed. We revised the sentence as below to clarify this:

“Ninety-five (99%) hospitals, out of 96 Thai hospitals that we approached, agreed to participate in the study.”

2. In the statistical section, it is necessary to clarify which variables have fixed effect(s) and which have random effect(s) in your mixed-effects logistic model.

Response: The lines below are now revised and added in manuscript to clarify mixed-effect models.

“We estimated the magnitude of differences in proportions and mortality of AMR BSI between SCHs and TCHs using mixed-effect logistic regression models for patients nested within hospital using the STATA command xtlogit. We estimated the magnitude of differences in frequency and mortality rate of AMR BSI between SCHs and TCHs using mixed-effect Poisson regression models for patients nested within hospital using the STATA command xtpoisson. We used mixed-effect models to estimate the fixed effects of hospital type while taking account for the random effects of patient within the same hospital. Multivariable mixed-effect models were also performed to control for other variables, including blood culture utilization rates of the hospitals and the health region where the hospitals located. These variables were included in the multivariable models because they could be associated with the hospital type and the outcome variables.”

3. Please be aware that p-values only indicate the level of significance, not the strength of the relationship. Therefore, consider this in all your tables and replace all values of <0.0001 with <0.001. NOTICE: "When p-values are displayed as very small numbers (e.g., 0.0001), it can be challenging for readers to quickly grasp their significance. It's more common to use the conventional format of displaying p-values as three decimal places (e.g., 0.001) for ease of interpretation."

Response: P-values in the text have been revised as suggested.

4. The quality of your figures is significantly low, or they did not meet the desired quality in the submitted version, and I could not extract any useful information from them.. Please ensure that the figures meet the required quality standards or provide better versions in the submitted manuscript.

Response: Figures with improved resolutions have been uploaded. The figures with improved resolutions can be downloaded from the PLoS website using the links on the PDF.

5. In the references section, you've used reports as sources. Please include relevant links for each reference to facilitate readers' access to the sources in your article.

Response: The links for references to reports have been added.

Reviewer #2: I am impressed by the large amount of data collected during the study period 2012 to 2015; however, I recommend rejection due to the limited analysis performed and the similarities with previous studies.

Response: Additional multivariable models are now added to control for variables that define the characteristics of the hospitals. The findings from the multivariable models have been added in both the main text of the manuscript and supporting documents. The results from the multivariable models still suggested differences in the observed frequency and mortality rates of AMR BSI between tertiary-care and secondary-care hospitals. The additional evidence supports our conclusion that the targets of and resources for antimicrobial stewardship and infection control programs in LMICs might need to be tailored based on hospital type and size, as burden of AMR infections might differ by hospital setting.

In addition, there are few studies comparing frequency and mortality rate following antimicrobial-resistant infections between tertiary-care hospitals and secondary-care hospitals in LMICs.

Reviewer #3: Dear Authors

Kindly address the following few concerns

Line 154 correct the grammatical error " local importance" instead of "locally importance."

Response: Revised as suggested.

Line 215, second sentence is unclear, kindly revise.

Response: Line 215 has been revised as below:

“Consent was not sought from the individual patients as this was a retrospective study. This approach was approved by the Ethical and Scientific Review committees.”

LINE 398 "This is likely occurring in other LMICs” as part of the conclusion is too much of extrapolation from the current study under review. Rephrasing it to eg “This may be situation in other LMICs, or This might also be occurring in other LMICs” will be more scientifically acceptable.

Attachment

Submitted filename: response_to_reviewers_DL.docx

pone.0303132.s004.docx (42KB, docx)

Decision Letter 1

Ali Amanati

13 Mar 2024

PONE-D-23-31373R1Frequency and mortality rate following antimicrobial-resistant bloodstream infections in tertiary-care hospitals compared with secondary-care hospitalsPLOS ONE

Dear Dr. Limmathurotsakul,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Apr 27 2024 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ali Amanati

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments:

Dear authors

We are pleased to inform you that your manuscript has passed through ‎‎the ‎review stage and is ready for revision. The manuscript's overall ‎‎presentation ‎improved after amendments; ‎‎however, ‎still needs minor revision.‎

Editor comments:

Abstract:‎

‎Add corresponding statistics in the “results” section for the data presented in ‎the abstract section. Address the following shortcomings: ‎

a. “The frequencies of hospital-origin AMR BSI per 100,000 patient-days at risk ‎in TCHs were about twice that in SCHs for most pathogens under evaluation.”; ‎add frequencies and p. value. ‎

b. “All-cause in-hospital mortality (%) following hospital-origin AMR BSI was

not significantly different between TCHs and SCHs.”; add p. value.‎

c. “However, due to the higher frequencies, all-cause in-hospital mortality rates ‎following hospital-origin AMR BSI per 100,000 patient-days at risk were ‎considerably higher in TCHs for most pathogens.”; remove “However”; add ‎frequencies and p. value.‎

d. the following statement need corrections: ‎

‎“For example, the all-cause in-hospital mortality rate ...”; remove “For example, ”‎

e. Conclusion should be improved (both abstract and main text). It is a fact that AMR is varied in different ‎hospital settings. Based on the results of your study, you should highlight the ‎important indicators that found to be clinically more different and suggest ‎appropriate action to address issues.‎

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #3: All comments have been addressed

Reviewer #4: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #3: Yes

Reviewer #4: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #3: Yes

Reviewer #4: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Your statistical section leaves me feeling disconnected for a couple of reasons. Firstly, there's no mention of demographic variables for the patients. Was this data unavailable or deemed unnecessary for your model? Secondly, the confounding variables you adjusted for aren't specified. Were these variables shared between patients and hospitals? Were they chosen based on their relevance to patients or hospitals? The only insights provided are tucked away in the supplementary file, in the footer of the tables.

Reviewer #3: No further comments at this stage. Authors have tried to address largely the concerns raised earlier.

Reviewer #4: All the previous comments have been addressed but the author have not mentioned which AMR proportions and Mortalities were analyzed using either the chi square test or fisher exact test or both.

line 222: author repeatedly mentioned "estimate of magnitude" but there is no mention of coefficient in result section , figures or in the supplemental material. kindly provide the output of analysis in stata showing coefficient.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

Reviewer #4: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2024 May 20;19(5):e0303132. doi: 10.1371/journal.pone.0303132.r004

Author response to Decision Letter 1

18 Mar 2024

Re: PONE-D-23-31373R1

Frequency and mortality rate following antimicrobial-resistant bloodstream infections in tertiary-care hospitals compared with secondary-care hospitals

PLOS ONE

Dear Editor and Reviewers,

We would like to thank the editor and all the reviewers for their time reading and commenting on our manuscript. We have revised the manuscript based on the feedback. In brief, the conclusion in the Abstract and Discussion sections are revised as suggested by the Editor. Also, frequencies, p-values, and mortality rates are included in the Abstract section as recommended by the Editor. The statistical analysis section is revised to improve clarity.

We very much appreciate the suggestions and made changes in our manuscript as below:

Editor comments:

Abstract:‎

‎Add corresponding statistics in the “results” section for the data presented in ‎the abstract section. Address the following shortcomings: ‎

a. “The frequencies of hospital-origin AMR BSI per 100,000 patient-days at risk ‎in TCHs were about twice that in SCHs for most pathogens under evaluation.”; ‎add frequencies and p. value. ‎

Response: The frequencies and p values have been added as below:

“(for carbapenem-resistant Acinetobacter baumannii [CRAB]: 18.6 vs. 7.0, incidence rate ratio 2.77; 95%CI 1.72-4.43, p<0.001; for carbapenem-resistant Pseudomonas aeruginosa [CRPA]: 3.8 vs. 2.0, p=0.0073; third-generation cephalosporin resistant Escherichia coli [3GCREC]: 12.1 vs. 7.0, p<0.001; third-generation cephalosporin resistant Klebsiella pneumoniae [3GCRKP]: 12.2 vs. 5.4, p<0.001; carbapenem-resistant K. pneumoniae [CRKP]: 1.6 vs. 0.7, p=0.045; and methicillin-resistant Staphylococcus aureus [MRSA]: 5.1 vs. 2.5, p=0.0091).”

b. “All-cause in-hospital mortality (%) following hospital-origin AMR BSI was

not significantly different between TCHs and SCHs.”; add p. value.‎

Response: The p-values have been added.

c. “However, due to the higher frequencies, all-cause in-hospital mortality rates ‎following hospital-origin AMR BSI per 100,000 patient-days at risk were ‎considerably higher in TCHs for most pathogens.”; remove “However”; add ‎frequencies and p. value.‎

Response: “However” and “For example” are now removed. The frequencies and p values are reported as below:

“Due to the higher frequencies, all-cause in-hospital mortality rates following hospital-origin AMR BSI per 100,000 patient-days at risk were considerably higher in TCHs for most pathogens (for CRAB: 10.2 vs. 3.6, mortality rate ratio 2.77; 95%CI 1.71 to 4.48, p<0.001; CRPA: 1.6 vs. 0.8; p=0.020; 3GCREC: 4.0 vs. 2.4, p=0.009; 3GCRKP, 4.0 vs. 1.8, p<0.001; CRKP: 0.8 vs. 0.3, p=0.042; and MRSA: 2.3 vs. 1.1, p=0.023).”

d. the following statement need corrections: ‎

‎“For example, the all-cause in-hospital mortality rate ...”; remove “For example, ”‎

Response: “For example” has been removed.

e. Conclusion should be improved (both abstract and main text). It is a fact that AMR is varied in different ‎hospital settings. Based on the results of your study, you should highlight the ‎important indicators that found to be clinically more different and suggest ‎appropriate action to address issues.‎

Response: The conclusion has been improved as below in the Abstract and in the Discussion:

“The frequency and in-hospital mortality rate of hospital-origin AMR BSI are important indicators and should be routinely measured to monitor the burden of AMR in every hospital with microbiology laboratories in LMICs.”

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #3: All comments have been addressed

Reviewer #4: All comments have been addressed

________________________________________

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #3: Yes

Reviewer #4: Yes

________________________________________

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #3: Yes

Reviewer #4: Yes

________________________________________

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #3: Yes

Reviewer #4: Yes

________________________________________

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: Yes

Reviewer #4: Yes

________________________________________

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Your statistical section leaves me feeling disconnected for a couple of reasons. Firstly, there's no mention of demographic variables for the patients. Was this data unavailable or deemed unnecessary for your model? Secondly, the confounding variables you adjusted for aren't specified. Were these variables shared between patients and hospitals? Were they chosen based on their relevance to patients or hospitals? The only insights provided are tucked away in the supplementary file, in the footer of the tables.

Response: We would like to thank the reviewer for highlighting the potential confusing points in our manuscript. No individual patient-level data on demographic variables was collected. Our study is a secondary data analysis, which involved extracting aggregated statistics on AMR burden from routine AMR surveillance report generated by the 49 local hospitals. The sentence below is added to clarify:

“This secondary data analysis was not designed to answer causal question but to demonstrate the potentials of using AMASS as a tool to empower routine AMR surveillance data understanding and sharing.”

The sentence below is added in the “Statistical Analysis” section under the “Methods” to clarify the variables included in the multivariable analysis:

“Multivariable mixed-effect models were also performed to control for other variables, including blood culture utilization rates of the hospitals and the health region where the hospitals located.”

Reviewer #3: No further comments at this stage. Authors have tried to address largely the concerns raised earlier.

Reviewer #4: All the previous comments have been addressed but the author have not mentioned which AMR proportions and Mortalities were analyzed using either the chi square test or fisher exact test or both.

Response: We thank the reviewer for highlighting the confusing point on Chi square test and Fisher exact test. In the final analysis, neither Chi-square tests nor Fisher Exact tests were used. We have revised the sentence below to avoid confusion:

“We preliminarily compared AMR proportions and mortality using Chi-square or Fisher’s exact test when small samples (i.e. one or more expected values was <5 observations) and measurements from continuous variables using the Krustal-Wallis test. Then, we ..”

We also add all statistical tests used for all footnotes for clarity as below:

Footnote of Figure 1: “Comparison was made using the Kruskal-Wallis test.”

Footnote of Figure 2: “Comparison was made using mixed-effect logistic or Poisson regression models of patients nested within hospitals.”

Footnote of Figure 3: “Comparison was made using mixed-effect logistic or Poisson regression models of patients nested within hospitals.”

All footnotes of supplementary tables.

line 222: author repeatedly mentioned "estimate of magnitude" but there is no mention of coefficient in result section , figures or in the supplemental material. kindly provide the output of analysis in stata showing coefficient.

Response: The coefficients (from both univariable and multivariable models) are reported in the main text and supplementary materials in the form of incidence rate ratio (IRR) and mortality rate ratio (MRR), which are the exponentials of the coefficients. We decided not to report odds ratio of the logistic regression models to avoid confusion. We also removed the word “magnitude” to avoid confusion.

________________________________________

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

Reviewer #4: No

All contributing authors have reviewed and concurred with the revised manuscript.

Yours,

Direk Limmathurotsakul

Attachment

Submitted filename: response_to_reviewers_2_DL.docx

pone.0303132.s005.docx (23.5KB, docx)

Decision Letter 2

Ali Amanati

8 Apr 2024

PONE-D-23-31373R2Frequency and mortality rate following antimicrobial-resistant bloodstream infections in tertiary-care hospitals compared with secondary-care hospitalsPLOS ONE

Dear Dr. Limmathurotsakul,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by May 23 2024 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ali Amanati

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments:

Dear authors,

New comments were posted by the reviewer #1. So, the manuscripts require a round of revision.‎ Please provide a point-by-point response to the reviewer comments and highlight all the ‎amends on your manuscript with yellow color.‎

Yours

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #4: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #4: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #4: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #4: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #4: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I appreciate the research team for answering the concerns. The authors' response does address some of the concerns by providing clarification on the nature of the data and the purpose of the study. However, there are still aspects that could be further explicit to improve the clarity and robustness of the study.

Demographic Variables: The authors clarify that individual patient-level data on demographic variables was not collected and that the study relied on aggregated statistics. While this explanation addresses the absence of demographic variables, it would be beneficial for the authors to explain why such data was not collected or why it was not deemed necessary for the model.

Confounding Variables: The authors mention that multivariable mixed-effect models were performed to control for other variables, including blood culture utilization rates of the hospitals and the health region where the hospitals were located. While this adds some clarity regarding the confounding variables, it would be helpful for the authors to elaborate on why these specific variables were chosen and how they were determined to be relevant to the analysis. Additionally, providing information on any other potential confounders that were considered but not included in the analysis would strengthen the robustness of the study.

Reviewer #4: no Comments . all the previous comments have been addressed .

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #4: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2024 May 20;19(5):e0303132. doi: 10.1371/journal.pone.0303132.r006

Author response to Decision Letter 2

14 Apr 2024

Re: PONE-D-23-31373R2

Frequency and mortality rate following antimicrobial-resistant bloodstream infections in tertiary-care hospitals compared with secondary-care hospitals

PLOS ONE

Dear Editor and Reviewers,

We would like to thank the editor and all the reviewers for their time reading and commenting on our manuscript. We have revised the manuscript based on the feedback.

We very much appreciate the suggestions and made changes in our manuscript as below:

Reviewer #1: I appreciate the research team for answering the concerns. The authors' response does address some of the concerns by providing clarification on the nature of the data and the purpose of the study. However, there are still aspects that could be further explicit to improve the clarity and robustness of the study.

Demographic Variables: The authors clarify that individual patient-level data on demographic variables was not collected and that the study relied on aggregated statistics. While this explanation addresses the absence of demographic variables, it would be beneficial for the authors to explain why such data was not collected or why it was not deemed necessary for the model.

Response: We are grateful for the advice. We have revised and elaborated the details of data collected from the study hospitals and used for this study in the method section as follow, “Variables in the microbiology data file included patient hospital number (HN), specimen type, specimen collection date, culture result, and antibiotic susceptibility testing result, and each row contained information for each specimen. Variables in the hospital admission data file included HN, admission date, discharge date, and in-hospital discharge outcome, and each row contained information for each admission.”

We have also revised and elaborated this limitation in the discussion section as follow, “These data were not available because our study was not designed to collected those data. The aim of this study was to examine the burden of AMR BSI observed in TCHs and compare that observed in SCHs using AMASS.15 To understand the size of effect caused by the difference in hospital type alone, further studies using detailed clinical data, AMS and IPC data, and rigorous statistical approaches are needed.”

Confounding Variables: The authors mention that multivariable mixed-effect models were performed to control for other variables, including blood culture utilization rates of the hospitals and the health region where the hospitals were located. While this adds some clarity regarding the confounding variables, it would be helpful for the authors to elaborate on why these specific variables were chosen and how they were determined to be relevant to the analysis. Additionally, providing information on any other potential confounders that were considered but not included in the analysis would strengthen the robustness of the study.

Response: We are grateful for the advice. We have revised and elaborated the details of data collected from the study hospitals in details in the method section as follow, “We adjusted for blood culture utilization rate because if blood culture utilization rate was low, the observed AMR proportions could be higher than the true susceptibility profiles of pathogenic organisms.22 Additionally, the observed AMR frequency per 100,000 admissions and per 100,000 patient-days could be lower than the true incidence rate of AMR infections.22 We adjusted for health regions to control for potential variations in the AMR proportions and frequency across different regions. These regional differences could be influenced by other factors such as habits of antibiotic use and economic levels which were not explored in this study. There were no other potential confounders that were evaluated in the analysis.”

All contributing authors have reviewed and concurred with the revised manuscript.

Yours,

Direk Limmathurotsakul

Attachment

Submitted filename: response_to_reviewers_3_DL.docx

pone.0303132.s006.docx (17.9KB, docx)

Decision Letter 3

Ali Amanati

22 Apr 2024

Frequency and mortality rate following antimicrobial-resistant bloodstream infections in tertiary-care hospitals compared with secondary-care hospitals

PONE-D-23-31373R3

Dear Dr. Direk Limmathurotsakul,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. If you have any questions relating to publication charges, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Ali Amanati

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

I read the revised manuscript ‎

I have no further comments to add. I thank the authors for their detailed ‎‎replies to the reviewers' comments.‎

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

**********

Acceptance letter

Ali Amanati

29 Apr 2024

PONE-D-23-31373R3

PLOS ONE

Dear Dr. Limmathurotsakul,

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now being handed over to our production team.

At this stage, our production department will prepare your paper for publication. This includes ensuring the following:

* All references, tables, and figures are properly cited

* All relevant supporting information is included in the manuscript submission,

* There are no issues that prevent the paper from being properly typeset

If revisions are needed, the production department will contact you directly to resolve them. If no revisions are needed, you will receive an email when the publication date has been set. At this time, we do not offer pre-publication proofs to authors during production of the accepted work. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few weeks to review your paper and let you know the next and final steps.

Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

If we can help with anything else, please email us at customercare@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Ali Amanati

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 File

    (DOC)

    pone.0303132.s001.doc (308KB, doc)
    Attachment

    Submitted filename: Referee_answer.pdf

    pone.0303132.s002.pdf (132.7KB, pdf)
    Attachment

    Submitted filename: review PONE-D-23-31373.docx

    pone.0303132.s003.docx (18.9KB, docx)
    Attachment

    Submitted filename: response_to_reviewers_DL.docx

    pone.0303132.s004.docx (42KB, docx)
    Attachment

    Submitted filename: response_to_reviewers_2_DL.docx

    pone.0303132.s005.docx (23.5KB, docx)
    Attachment

    Submitted filename: response_to_reviewers_3_DL.docx

    pone.0303132.s006.docx (17.9KB, docx)

    Data Availability Statement

    All relevant data are within the manuscript and its Supporting Information files. The study used openly anonymous AMR surveillance reports generated from each hospital which are available at: https://doi.org/10.6084/m9.figshare.20318193.

    Articles from PLOS ONE are provided here courtesy of PLOS

    RESOURCES