Background

F1000Research

2046-1402

F1000 Research Limited

London, UK

10.12688/f1000research.151896.2

Research Article

Articles

Analysis of the dynamics of transition from non-colonization to colonization and Staphylococcus aureus bacteremia in hemodialysis patients using Markov models.

[version 2; peer review: 1 approved, 1 approved with reservations]

Montoya-Urrego

Daniela

Conceptualization Formal Analysis Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0002-9326-4341 1 Vanegas

Johanna M

Conceptualization Formal Analysis Methodology Supervision Writing – Review & Editing 2 Jiménez

J Natalia

Conceptualization Methodology Supervision Writing – Review & Editing https://orcid.org/0000-0002-9183-1912 1 González-Gómez

Difariney

Conceptualization Formal Analysis Supervision Writing – Review & Editing a 3 1Grupo de investigación en Microbiología Básica y aplicada (MICROBA), Escuela de Microbiología, Universidad de Antioquia, Medellín, Antioquia, Colombia 2Escuela de Ciencias de la Salud, Universidad Pontificia Bolivariana, Medellín, Antioquia, Colombia 3Grupo de investigación Demografía y Salud, Facultad Nacional de Salud Pública, Universidad de Antioquia, Medellín, Antioquia, Colombia

a difariney.gonzalez@udea.edu.co

No competing interests were disclosed.

4 11 2024

2024

837

28 10 2024

2024

This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background

Hemodialysis patients are frequently colonized by Staphylococcus aureus, leading to severe infections with high mortality rates. However, little is known about transition from non-colonization to colonization or bacteremia over time. The aim was to analyze the behavior of S. aureus colonization, identifying the probability of transition from non-colonized to colonized state or bacteremia, and the influence of specific covariates.

Methods

The study was conducted in a dialysis unit associated with a tertiary care hospital in Medellín between October 2017 and October 2019. An initial measurement was taken to evaluate S. aureus colonization, and follow-up measurements were performed 2 and 6 months later. Bacteremia evolution was monitored for 12 months. A two-state recurrent continuous-time Markov model was constructed to model transition dynamics from non-colonization to S. aureus colonization in hemodialysis patients. Subsequently, the model was applied to a third state of bacteremia.

Results

Of 178 patients on hemodialysis, 30.3% were colonized by S. aureus. Transition intensity from non-colonization to colonization was three times higher (0.21; CI: 0.14-0.29) than from colonization to non-colonization (0.07; CI: 0.05-0.11). The colonization risk increased in patients with previous infections (HR: 2.28; CI: 0.78-6.68), hospitalization (HR: 1.29; CI: 0.56-2.99) and antibiotics consumption (HR: 1.17; CI: 0.53-2.58). Mean non-colonized state duration was 10.9 months, while in the colonized state was 5.2 months. In the 3-state model, it was found that patients colonized were more likely to develop S. aureus infection (13.9%).

Conclusion

A more likely transition from non-colonization to colonization was found, which increases with factors such as previous infection. In addition, the development of bacteremia was more likely in colonized than in non-colonized patients. These results underline the importance of surveillance and proper management of S. aureus colonization to prevent serious complications, such as bacteremia, and improve prognosis in this vulnerable population.

Markov models Multistate models Staphylococcus aureus Hemodialysis Colonization Bacteremia.

Fondo a primer proyecto. Vicerrectoría de Investigación. Universidad de Antioquia.

This work was supported by Comité para el Desarrollo de la Investigación CODI, Fondo a primer proyecto, Cod: INV 591-17. Vicerrectoría de Investigación. Universidad de Antioquia.

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Revised Amendments from Version 1

In this revised version, we addressed the reviewers' comments and made significant improvements in several sections. In the Introduction, we expanded the explanation of the role, relevance, and appropriateness of using the Markov model in studying the dynamics of Staphylococcus aureus colonization and infection in hemodialysis patients. In addition, we refined the Methods section to clarify how transition periods were calculated using the transition matrix Q, emphasizing the assumption of temporal homogeneity inherent to Markov models. In the Results section, we added details of the exclusion of some patients, and added colonization and infection data from patients in the Extended data, to provide a clearer presentation of the findings. The Discussion was enriched, now including a detailed explanation of the limitations in our study, related to the clinical implications of our findings, the dialysis unit where the study was conducted, the sample size, and the temporality of our study. New “Extended Data” was added, including the code used for the Markov model analysis and additional data detailing the colonization and infection status of patients at different time points. The References section was updated to include new relevant sources supporting the reviews conducted and discussion points. These updates undoubtedly improve the clarity, reproducibility, and clinical relevance of the study.

Introduction

Staphylococcus aureus is one of the microorganisms that most frequently colonizes and causes invasive infection, such as bacteremia in hemodialysis patients. ¹ In addition to increasing the risk of endogenous infection by up to 77.8% in this group of patients, colonization by S. aureus favors the dissemination of this bacterium at hospital and community level, due to the fact that colonized patients constantly circulate among these environments and act as asymptomatic reservoirs and carriers of the microorganism for long periods of time. ² ^, ³ The situation become more worrying when isolates are methicillin-resistant (MRSA), because treatment options are reduced, the prognosis for the patient worsens and morbidity, mortality and the cost of care increase significantly. ¹

Colonization by S. aureus in hemodialysis patients can be persistent, which refers to the permanent presence of the microorganism over time; or intermittent, in which the microorganism can be present or absent at different times. ³ ^, ⁴ There is little information on the probability of switching between the non-colonized state and the colonized state and vice versa, as well as the period of permanence of patients in each state and the implications for infection. This is important because, if patients remain colonized for a longer period of time, this favors the risk of invasive infections such as bacteremia, since the colonizing bacteria can contaminate catheters or other devices and reach the bloodstream. ⁵

In this sense, Markov models emerge as a valuable and useful statistical tool to study the dynamics of colonization over time, allowing to estimate the transition between states, in this case from a state of non-colonization to one of colonization or infection. ⁶ ^, ⁷ This type of statistical model allows modelling systems that change randomly, and assumes that future states depend only on the current state, not on the events that occurred before it (a property known as the Markovian assumption). ⁶

By focusing on this property, Markov models enable estimation not only of the probability of being in each state at any given time but also of the expected duration in each state before transitioning. This ability is essential for predicting patterns over time, as it enables modeling of state behaviors calculation of transition rates, and the stability or persistence within each state. Thus, Markov models provide a structured approach to understand and predict patient transition between non-colonization, colonization, and infection states, offering important insights into the temporal dynamics of colonization and its clinical implication.

In this study, we proposed to analyze the behavior of colonization by S. aureus, identifying the probabilities of change from non-colonized to colonized states and bacteremia, and the influence of some covariables in these transitions. This will allow designing and directing strategies to avoid progression to states that compromise the patient’s prognosis, such as colonization and infection.

Methods Study population

A cohort of 210 hemodialysis patients was taken from a previous study that was carried out in a dialysis unit in Medellín, Colombia, in which colonization by S. aureus was evaluated at three moments in time: at the beginning of the study, at two and at 6 months; and the development of bacteremia during a 12-month follow-up. ² This study included patients from a dialysis unit associated to a high complexity hospital in Medellín, over 18 years of age, with chronic kidney disease and central venous catheter on hemodialysis. Patients who had only the baseline measurement were excluded from the present study. Informed consent was signed by each patient.

Variables

Colonization was assessed in nostrils and skin around the insertion of the hemodialysis catheter at the beginning of the study, at 2 and 6 months later, in order to capture and identify all potential transitions and the behavior of colonization. ⁸ ^– ¹⁰ Infection was defined as the diagnosis of bacteremia according to the criteria given by the Center for Disease Control and Prevention (CDC), as the presence of fever, chills or hypotension with bacteria identified in the blood and not related to an infection at any other site. ¹¹ Central-line associated bloodstream infections (CLABSI) were also considered in the analysis.

Additionally, demographic and clinical variables were evaluated, such as age, sex, smoking, history of hospitalization, previous infection, antibiotic use, comorbidities, and catheter or fistula use. To apply the Markov model, two possible recurrent states were initially considered: 1: Non-colonized and 2: Colonized. Subsequently, at a second stage, 3 states were considered: 1: Non-colonized, 2: Colonized and 3: Bacteremia. Consequently, individuals can independently transit between states where the probability of transition is not time-dependent. ¹² Transitions are defined as changes from one state to another, and this process can be specified in terms of transition intensities. For each patient a transition history was established based on a maximum of three observations during the follow-up period, in which the state of colonization, non-colonization or infection was determined. It is important to highlight that in the renal unit where the study was conducted, there were no decolonization protocols in place for patients. As a result, this factor did not influence the transitions between states.

Data collection

The clinical and epidemiological information was obtained by a questionnaire designed for this purpose, which was applied to all the hemodialysis patients in the company of a researcher by interview. Similarly, the clinical histories provided by the dialysis center were taken into account to know the clinical history of patients.

Colonization screening

To detect S. aureus colonization, samples were obtained from nostrils and skin around the catheter insertion, using a sterilized cotton swab with sterile 0.9% saline solution. Each swab was transported in AMIES medium (transport medium with activated carbon) and then enriched for 18 to 24 hours at 37°C in trypticase soy broth (TSB- OXOID ^TM, CM0129), prepared according to manufacturer’s instructions, adding 30 g to 1 liter of water. Subsequently, it was plated on mannitol salt agar for the selection of fermenting colonies indicative of S. aureus. Preliminary identification was performed by phenotypic methods based on colony morphology in sheep blood agar and positive catalase and coagulase tests. Identification of isolates and antibiotic susceptibility was determined using the Vitek ^®-2 automated system (bioMérieux) according to Clinical and Laboratory Standards Institute (CLSI) cut-off points. ¹³

Statistical analysis

Categorical variables were described as absolute and relative frequencies. Quantitative variables were expressed as mean and standard deviation or median and interquartile range, according to the assumption of normality. Subsequently, Markov models were applied, in which the probability of transitioning from one state to another depends solely on the current state, with no influence from past states, satisfying the Markov property. ⁶ To estimate the transition rates and probabilities between states, the inter-occurrence times were transformed such that the time of the first measurement was considered as time zero (t0), and subsequent measurements were considered as time 1 (t1) and time 2 (t2). This approach allows estimating the transition intensity functions between states, identifying all the transitions that occurred between observations, estimating the maximum probability of transition from one state to another and estimating the maximum probabilities when covariables are present. ⁶ ^, ¹⁴

In relation to the above, two transition matrices were calculated: the transition count matrix Q and the transition probability matrix P. The transition matrix Q represents the frequency of individuals transitioning from state r to state s during a given time interval, called by q _rs . Each element of this matrix counts the number of transitions observed between specific states during the follow-up period. The transition probability matrix P, that refers to the probability of moving from state r to state s in the time interval p _rs, is calculated by dividing each of the elements of matrix Q by the row total. ¹⁵

In both matrices, the rows are designate by the current state, and the columns represent the state to which the transition may occur. The sum of the probabilities of a row of the transition matrix is equal to 1. The transition probabilities are presented with the 95% confidence interval. For the case of patients with S. aureus all possible transitions between states are allowed, in all states there is a positive probability of reaching it.

Statistical analyses were conducted using R software, version 4.4.1, specifically employing the msm Markovchain package. Additionally, the analysis code can be found in the GitHub repository (Extended data).

Results

Of 210 hemodialysis patients, 32 were excluded from the present study because only had the initial observation, leaving a total of 178 patients included. Of the excluded patients, 25% (n=8) passed away, while the rest had only one measurement for other reasons, such as transferring to another renal unit, voluntarily withdrawing from the study, receiving a kidney transplant, or discontinuing treatment by choice (see extended data). The majority of patients were female (51.7%, n=92), and the mean age was 62 years (SD 15.9). Clinical data revealed that 68.5% (n=122) of patients had been hospitalized in the last year, while 57.3% (n=102) consumed antibiotics in the previous 6 months. The most common comorbidities were diabetes mellitus (44.4%, n=79), heart failure (24.7%, n=44), and coronary artery disease (20.2%, n=36). At baseline, 30.3% (n=54) of patients had S. aureus colonization.

Two recurrent state models for <italic toggle="yes">S. aureus</italic> colonization

The recurrent two-state model (Non-Colonized - Colonized) is shown in Figure 1. During the follow-up period, and considering all three measurements, the transition intensity was 3 times higher for going from a non-colonized to colonized state (0.21; CI: 0.14-0.29) compared to the transition from colonized to non-colonized state (0.07; CI: 0.05-0.11). The probabilities of remaining in the same state decreased over time, contrary to the transition between the two states, which presented higher likelihoods over time ( Table 1).

Figure 1. Transitions between two recurrent states of the Markov model for hemodialysis patients.

The arrows indicate the allowed transitions between states. Patients can remain in one state in consecutive cycles.

Table 1. Transition probability and intensity in the two-state Markov model.

Transition	Transition intensity (IC 95%)	Transition probability
Transition	Transition intensity (IC 95%)	Baseline	2 months	6 months
Non-colonized → Non-colonized	-0.07 (-0.11; -0.05)	0.93	0.89	0.79
Non-colonized → Colonized	0.21 (0.14; 0.29)	0.07	0.11	0.21
Colonized → Non-colonized	0.07 (0.05; 0.11)	0.18	0.32	0.60
Colonized → Colonized	-0.21 (-0.29; 0.14)	0.82	0.68	0.40

Effect of covariates on the dynamics of two-state colonization by <italic toggle="yes">S. aureus</italic>

In the presence of most of the covariates analyzed, such as smoking, previous hospitalization, antibiotic use and comorbidities, the transition intensities from the colonized to the non-colonized state doubled with respect to the transition intensities from the non-colonized to the colonized state, with the exception of previous infection, which presented similar intensities in both transitions ( Table 2).

Table 2. Transition intensity and risk ratios for covariates in recurrent two-state model in <italic toggle="yes">S. aureus</italic> colonization.

Variable	Hazard Ratio of transition		Transition intensity
Variable	Non-colonized → Colonized (IC 95%)	Colonized → Non-colonized (IC 95%)	Non-colonized → Non-colonized (IC 95%)	Non-colonized → Colonized (IC 95%)	Colonized → Non-colonized (IC 95%)	Colonized → Colonized (IC 95%)
Smoking	0.67 (0.29;1.53)	1.62 (0.78;3.33)	-0.07 (-0.11;-0.04)	0.07 (0.04; 0.11)	0.21 (0.15;0.30)	-0.21 (-0.30;-0.15)
Hospitalization	1.29 (0.56;2.99)	1.62 (0.76;3.46)	-0.070(-0.11;-0.05)	0.07 (0.05;0.11)	0.21 (0.15;0.30)	-0.21(-0.30;-0.14)
Previous infection	2.28 (0.78;6.68)	1.22 (0.41;3.63)	-0.18 (-0.45;-0.07)	0.18 (0.07;0.45)	0.19 (0.07;0.50)	-0.19 (-0.50;-0.07)
Antibiotics consumption	1.17 (0.53;2.58)	0.97 (0.47;1.99)	-0.07 (-0.11;-0.05)	0.07 (0.05;0.11)	0.20 (0.14;0.29)	-0.20 (-0.29;-0.14)
Diabetes mellitus	0.58 (0.26;1.33)	1.08 (0.53;2.18)	-0.07 (-0.10;-0.05)	0.07 (0.05;0.10)	0.21 (0.14;0.29)	-0.21 (-0.29;-0.15)
Heart failure	2.0 (0.84;4.78)	1.65 0.72;3.76)	-0.15 (-0.31;-0.07)	0.15 (0.07;0.31)	0.27 (0.13;0.55)	-0.27 (-0.55;-0.13)
Coronary artery disease	1.27 (0.50;3.18)	1.33 (0.52;3.38)	-0.11 (-0.24;-0.05)	0.11(0.05;0.24)	0.23 (0.09;0.55)	-0.23 (-0.55;-0.10)
Arterial hypertension	0.21 (0.05;0.84)	0.76 (0.16;3.70)	-0.07 (-0.11;-0.05)	0.07 (0.05;0.11)	0.21(0.15;0.31)	-0.21 (-0.31;-0.15)
Charlson index	1.01 (0.86;1.18)	1.10 (0.96;1.26)	-0.09 (-0.13;-0.06)	0.09 (0.06;0.13)	0.19 (0.14;0.27)	-0.19 (-0.27;-0.14)
Karnofsky index	0.99 (0.97;1.03)	0.98 (0.95;1.01)	0-.09 (-0.13;-0.06)	0.09 (0.06;0.13)	0.18 (0.13;0.26)	-0.18 (-0.26;-0.13)

On the other hand, no evidence was found of the influence of the covariates studied on transition risk. However, significant tendencies to exceed the risk threshold for transition from the non-colonized state to the colonized state were observed in the presence of previous infection (HR: 2.28; CI: 0.78-6.68), previous hospitalization (HR: 1.29; CI 0.56-2.99), antibiotic consumption (HR: 1.17; CI 0.53-2.58) and heart failure (HR: 2.00; CI 0.84-4.78). On the other hand, the factors that showed a tendency to increase the transition risk from colonized to non-colonized state were history of hospitalization (HR: 1.62, CI 0.76-3.46) and smoking (HR:1.62; CI: 0.78-3.33). However, these risk indices were not significant ( Table 2).

Regarding the average time spent (in months) in one of the two states analyzed, it was found that there was longer duration of colonization in presence of hypertension (5.33; CI: 3.72-7.63) and in patients with previous antibiotic use (4.98; CI: 2.86-8.65).

Three-state model for <italic toggle="yes">S. aureus</italic> colonization

In a second stage of the analysis, a third condition was incorporated: S. aureus bacteremia. Regarding this, 35.2% (n=64) of the patients were found to be in the colonized state, 64.8% (n=118) in the Non-colonized state and 4.39% (n=23) in the infected state.

In the three-state model, the results were similar to those presented in the two-state model. There was a 16.7% probability of moving from the non-colonized state to the colonized state. The probability of moving from the colonized state to the non-Colonized state was 32.7%. Staying in the non-Colonized state had a probability of 80.3%, while in the colonized state had a probability of 53.2%. On the other hand, the probability of a non-colonized person becoming infected was 2.8%, and among the colonized patients, 13.9%. There was also a longer stay in the Non-Colonized state (9.3 months) compared to the Colonized state (25 days) and the Infected state (4 days) ( Figure 2).

Figure 2. Three-state Markov model for <italic toggle="yes">S. aureus</italic> colonization and infection.

The arrows indicate the allowed transitions between states. Patients can remain in one state in consecutive cycles.

Related with the probability of moving from the Non-Colonized to Colonized state, it increased from the baseline measurement to the 12-month measurement (from 0.11 to 0.27), as did the probability of moving from Non-Colonized to Infected (from 0.03 to 0.06). In contrast, the probability of moving from colonized to infected status decreased over time, from 0.11 at baseline to 0.06 at 12 months ( Table 3).

Table 3. Estimation for the probability of transition between the 3-state model.

Transition	Transition probability
Transition	2 months	6 months	12 months
Non-colonized → Non-colonized	0.85	0.72	0.67
Non-colonized → Colonized	0.11	0.23	0.27
Non-colonized → Infected	0.03	0.05	0.06
Colonized → Non-colonized	0.26	0.53	0.63
Colonized → Colonized	0.62	0.40	0.31
Colonized → Infected	0.11	0.08	0.06
Infected → Non-colonized	0.35	0.56	0.64
Infected → Colonized	0.55	0.37	0.30
Infected → Infected	0.10	0.07	0.06

Discussion

In this study, Markov models were used to analyze the behavior of S. aureus in hemodialysis patients, and to predict its change between the states of non-colonization, colonization and bacteremia over time. Transition models have positioned as a powerful tool for studying transitions from one state to another, allowing to shed light on behaviors and outcomes of interest in some diseases and infections, such as smoking, diabetes and cancer. ¹⁶ ^– ¹⁸ Unlike other estimation models, such as Generalized Estimating Equations (GEE), with Markov models it is possible to simultaneously study transitions in both directions, such as from non-colonized to colonized and from colonized to non-colonized. In addition, make it possible to identify factors that may behave differently from one direction to the other. ¹⁹ Particularly in the case of the use of Markovian models to evaluate S. aureus colonization, several studies have been reported. ²⁰ ^– ²² However, the application of Markov models to evaluate the transition of states in hemodialysis patients presents limited evidence.

Considering the above, the results of this study show that, for patients who started in a non-colonized state, the probability of changing to a colonized state was 14.7%, lower than the probability of changing from the colonized to the non-colonized state ( Figure 1). However, the transition intensity was found to be 3 times higher in the switch from the non-colonized to colonized state ( Table 1). This may be due to the fact that this type of patient has specific baseline conditions and clinical characteristics that favor colonization, such as high antibiotic consumption, presence of different comorbidities, constant transit between the community and medical centers, frequent hospitalizations and regular contact with medical personnel and other patients. ⁸ ^, ²³ Similarly, in the few studies that have used Markovian models to understand the behavior of S. aureus in other populations, they reach the same conclusion, demonstrating that previous antibiotic use is related to the acquisition of Methicillin-Resistant S. aureus (MRSA) strains in nursing homes. ²² ^, ²⁴ This similarity may be due to the fact that, in this type of population, as well as in hemodialysis patients, there is an environment with conditions that favor colonization by this type of bacteria.

On the other hand, trends were found that the risk of passing from the non-colonized state to the colonized state increased when the patient had a previous infection, previous hospitalization, and antibiotics consumption; also, that the time spent in the colonized state increased when the patient had arterial hypertension and previously used antibiotics. This is in agreement with what has been reported by other authors, who have found an association between previous hospitalization and the risk of S. aureus colonization, and reinforce the importance of the previous conditions of hemodialysis patients in colonization. ²⁵ ^, ²⁶ Likewise, it has been described that most invasive S. aureus infections are of endogenous origin in these patients, ²⁷ therefore, remaining in a colonized state for a longer period of time increases the risk of developing an infection by this bacterium, such as bacteremia. ⁴ Carrier status increases the spread of the microorganism at the hospital and community level, since patients constantly circulate between these two environments, and have a care link, not only with health personnel but also with their home contacts and general community. ⁸ ^, ²⁷ ^, ²⁸

Regarding the 3-state model, in which infection was considered, it was found that there is a greater probability that a colonized person will develop an infection compared to a non-colonized person. This is in agreement with what has been described in previous studies, which show that colonized persons have a higher risk of infection, especially bacteremia ⁵; it has even been reported that 77.85% of hemodialysis patients who developed bacteremia due to S. aureus were previously colonized. ² In addition, the need to prevent colonization that can lead to infection is emphasized, due to the great implications in morbidity, mortality and worse prognosis of patients. ²⁹ ^, ³⁰ On the other hand, it was found that the time spent in the non-colonization state is longer than the time spent in colonization or infection, this may be due to the fact that patients go from colonization to infection or that they receive antibiotic treatment. This may also explain why the probability of remaining in the same state decreases over time.

Finally, high frequencies of colonization and infection in hemodialysis patients demonstrate the importance of maintaining active epidemiological surveillance in order to take actions to prevent infection. One way is to evaluate the possibility of establishing decolonization protocols that reduce the incidence of infections. ³¹ In a study that also used Markov models to model the decolonization of S. aureus, it was shown that the use of mupirocin as a therapy to decolonize patients’ nostrils was effective and reduced autoinoculation and infection. ²¹ In the same sense, infections by this bacterium generate a high cost in the health system, especially in hemodialysis patients, so being able to predict the behavior of S. aureus from colonization to infection can reduce the economic burden, both for the health system and for the patient and their family. ³² ^, ³³

Regarding limitations, this study was conducted at a single facility, which means it may not accurately represent the epidemiology in other dialysis centers. However, the institution where the research was conducted is one of the largest in the city and serves a diverse population from various areas. On the other hand, although S. aureus colonization can occur in multiple body sites, we only evaluated nasal colonization and the skin around the catheter, taking into account that the nasal region is the most common and is considered a strong indicator of overall colonization, offering high sensitivity for detecting S. aureus colonization. ³⁴ ^, ³⁵ Likewise, it was not possible to perform a separate analysis of the behavior of MRSA and MSSA due to the limited number of patients colonized by MRSA. Additionally, the inclusion of patients at an arbitrary point during hemodialysis treatment introduces variability in the interpretation of colonization transition intensities, especially with respect to how these may change over the time of hemodialysis therapy. Since S. aureus colonization may occur at different stages of treatment, the time from the start of hemodialysis to the time of inclusion in the study could influence the observed colonization patterns.

While the Markov model is really useful to study the transitions between non-colonization, colonization and infection by S. aureus in hemodialysis patients, it is important to clarify that this model does not take into account that this dynamic is also affected by the specific epidemiology of each institution and by the “Colonization pressure”, ³⁶ which varies according to the number of colonized patients in the renal unit. This limitation restricts the generalizability of our study to other contexts. Similarly, factors such as hospital stays, antibiotic use, and use of medical devices can affect the transitions between states, making them not constant over time, as assumed in the Markov model, which may affect the clinical applicability of the model. Nonetheless, multiple studies have demonstrated that colonization significantly increases the risk of subsequent infection. ³⁷ ^, ³⁸ Additionally, for patients who only had the initial measurement, death or receiving a kidney transplant were not considered competing events. Therefore, different types of risk analysis are necessary to address this limitation in future studies. Finally, in stochastic models, the evolution of a system may differ, even when starting from the same initial conditions. This is aggravated by the impossibility of controlling all the factors that influence the study of the phenomenon, which leads us to resort to probabilistic parameters.

Conclusion

The Markov model are a tool that can be used to determine the behavior between states of non-colonization, colonization, and S. aureus bacteremia in hemodialysis patients. The evidence of a more likely transition from non-colonization to colonization, especially favored by factors such as previous infections and antibiotic use, highlights the need to adequately manage S. aureus colonization and to have strategies to prevent it. Likewise, this study showed a higher probability of developing bacteremia in colonized patients, which draws attention to the importance of early intervention of colonization in order to avoid its progression. Finally, the use of modeling tools to address this type of problem allows the continuation and improvement of active epidemiological surveillance, making it possible to identify, design and implement evidence-based strategies aimed at preventing colonization and infection in these patients.

Ethics and consent

This research project conforms to the international ethical standards set forth in the Nuremberg Code, the Declaration of Helsinki, the Belmont Report and the World Health Organization’s Good Clinical Practice recommendations. The research was approved by the Ethics Committee for Human Investigations of the University of Antioquia (CBEIH-SIU) with the approval act No 17-65-689 dated May 3, 2017. Informed consent was signed by each patient.

Data availability statement Underlying data

Ethical and security consideration

The study data have sensitive and private information about the patients who participated. If the reviewers or the reader need access to the data, they can request it by e-mail to the corresponding author ( difariney.gonzalez@udea.edu.co). The data can be sent as long as it does not conflict with the consent signed by the patients.

Extended data

Figshare: Clinical and epidemiological information collection form. https://doi.org/10.6084/m9.figshare.26170027. ³⁹

Figshare: Markov Models. https://doi.org/10.6084/m9.figshare.27313296.v1. ⁴⁰

Figshare: Follow-up Hemodialysis Patients. https://doi.org/10.6084/m9.figshare.27313290.v1. ⁴¹

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

References 1

Schamroth Pravda

Maor

Brodsky

: Blood stream Infections in chronic hemodialysis patients - characteristics and outcomes. BMC Nephrol. 2024 Jan;25(1):3. 38172734

10.1186/s12882-023-03442-5

PMC10763456

Vanegas

Salazar-Ospina

Roncancio

: Staphylococcus aureus colonization increases the risk of bacteremia in hemodialysis patients: a molecular epidemiology approach with time-dependent analysis. Am. J. Infect. Control. 2021;49(2):215–223. 32485274

10.1016/j.ajic.2020.05.031

Patel

Jenkins

Mediavilla

: Clinical and Molecular Epidemiology of Methicillin-Resistant Staphylococcus aureus among Patients in an Ambulatory Hemodialysis Center. Infect. Control Hosp. Epidemiol. 2011;32(9):881–888. 21828968

10.1086/661598

Vanegas

Salazar-Ospina

Gallego

: A longitudinal study shows intermittent colonization by Staphylococcus aureus with a high genetic diversity in hemodialysis patients. Int. J. Med. Microbiol. 2021;311(1):151471. 33373839

10.1016/j.ijmm.2020.151471

Price

Sarween

Gupta

: Risk factors and short-term outcomes for methicillin-resistant Staphylococcus aureus and methicillin-sensitive Staphylococcus aureus colonization among hemodialysis patients. Saudi J. Kidney Dis. Transplant an Off. Publ. Saudi Cent Organ Transplantation, Saudi Arab. 2019;30(6):1351–1363. 10.4103/1319-2442.275479

Putter

Fiocco

Geskus

: Tutorial in biostatistics: competing risks and multi-state models. Stat. Med. 2007 May;26(11):2389–2430. 17031868

10.1002/sim.2712

Powell

Verma

Luo

: Modeling Chronic Obstructive Pulmonary Disease Progression Using Continuous-Time Hidden Markov Models. Stud. Health Technol. Inform. 2019 Aug;264:920–924. 31438058

10.3233/SHTI190358

Scheuch

Freiin von Rheinbaben

Kabisch

: Staphylococcus aureus colonization in hemodialysis patients: a prospective 25 months observational study. BMC Nephrol. 2019;20(1):153. 31060511

10.1186/s12882-019-1332-z

PMC6503363

Grothe

Taminato

Belasco

: Screening and treatment for Staphylococcus aureus in patients undergoing hemodialysis: a systematic review and meta-analysis. BMC Nephrol. 2014;15(1):202. 25519998

10.1186/1471-2369-15-202

PMC4289586

Agrawal

Valson

Bakthavatchalam

: Skin Colonizers and Catheter Associated Blood Stream Infections in Incident Indian Dialysis Patients. Indian J. Nephrol. 2022;32(1):34–41. 35283564

10.4103/ijn.IJN_400_20

Event CDCBI: Central Line-Associated Bloodstream Infection and Non-Central Line Associated Bloodstream Infection. Device-associated Modul. 2018.

Saint-Pierre

Combescure

Daurès

: The analysis of asthma control under a Markov assumption with use of covariates. Stat. Med. 2003 Dec;22(24):3755–3770. 14673936

10.1002/sim.1680

Clinical and Laboratory Standards Institute: M100 Performance Standards for Antimicrobial Susceptibility Testing, 29th. 2019;224.

Núñez

Steyerberg

Núñez

: Regression Modeling Strategies. Rev. Esp. Cardiol. 2011;64(6):501–507. 21531065

10.1016/j.recesp.2011.01.019

Urban

Wallin

: Introduction to Markov Models BT - Learning Landscape Ecology: A Practical Guide to Concepts and Techniques. Gergel

Turner

, editors. New York, NY: Springer New York;2002; pp.35–48.

Yeh

H-W

Ellerbeck

Mahnken

: Simultaneous evaluation of abstinence and relapse using a Markov chain model in smokers enrolled in a two-year randomized trial. BMC Med. Res. Methodol. 2012;12(1):95. 22770436

10.1186/1471-2288-12-95

PMC3599722

Wang

Shtylla

Chou

: Order-of-Mutation Effects on Cancer Progression: Models for Myeloproliferative Neoplasm. Bull. Math. Biol. 2024 Feb;86(3):32. 38363386

10.1007/s11538-024-01257-5

PMC10873249

Bauvin

Delacôte

Wandji

LCN

: Early prediction of the impact of public health policies on obesity and lifetime risk of type 2 diabetes: A modelling approach. PLoS One. 2024 Mar 28;19(3):e0301463. 38547299

10.1371/journal.pone.0301463

PMC10977742

Bible

Albert

Simons-Morton

: Practical issues in using generalized estimating equations for inference on transitions in longitudinal data: What is being estimated?. Stat. Med. 2019 Mar;38(6):903–916. 30411379

10.1002/sim.8014

Cai

Philips

Arora

: Cost-effectiveness of a real-time spatiotemporal mapping surveillance system for meticillin-resistant Staphylococcus aureus prevention. J. Hosp. Infect. 2024 Jan;143:178–185. 37774929

10.1016/j.jhin.2023.09.010

Poyraz

Sater

MRA

Miller

: Modelling methicillin-resistant Staphylococcus aureus decolonization: interactions between body sites and the impact of site-specific clearance. J. R. Soc. Interface. 2022 Jun;19(191):20210916. 35702866

10.1098/rsif.2021.0916

PMC9198502

Batina

Crnich

Döpfer

: Acquisition and persistence of strain-specific methicillin-resistant Staphylococcus aureus and their determinants in community nursing homes. BMC Infect. Dis. 2017 Dec;17(1):752. 29212459

10.1186/s12879-017-2837-3

PMC5719525

Montoya Urrego

Vanegas

Jiménez

: The remarkable genetic relationship between Staphylococcus aureus isolates from hemodialysis patients and their household contacts: Homes as an important source of colonization and dissemination. PLoS One. 2022;17(4):e0267276. 35439257

10.1371/journal.pone.0267276

PMC9017883

Batina

Crnich

Anderson

: Models to predict prevalence and transition dynamics of methicillin-resistant Staphylococcus aureus in community nursing homes. Am. J. Infect. Control. 2016 May;44(5):507–514. 26777286

10.1016/j.ajic.2015.10.042

PMC4851896

Karanika

Zervou

Zacharioudakis

: Risk factors for meticillin-resistant Staphylococcus aureus colonization in dialysis patients: a meta-analysis. J. Hosp. Infect. 2015;91(3):257–263. 26428959

10.1016/j.jhin.2015.07.014

Mehraj

Witte

Akmatov

: Epidemiology of Staphylococcus aureus Nasal Carriage Patterns in the Community. Curr. Top. Microbiol. Inmunol. 2016;398(July):435. 10.1007/82_2016_497

Eiff

von Becker

Machka

: Nasal Carriage as a Source of Staphylococcus aureus Bacteremia. N. Engl. J. Med. 2001 Jan 4;344(1):11–16. 11136954

10.1056/NEJM200101043440102

Davis

Iverson

Baron

: Household transmission of meticillin-resistant Staphylococcus aureus and other staphylococci. Lancet Infect. Dis. 2012;12(9):703–716. 22917102

10.1016/S1473-3099(12)70156-1

Christine

J-M

Mandler

Alosh

: Risk Factor Analysis for 30-day Mortality After Surgery for Infective Endocarditis. Thorac. Cardiovasc. Surg. 2024. (EFirst). 10.1055/s-0044-1779709

Bravo Thompson

Campos Herrera

Macías Ibiricu

: Factors Related to the Development of Infective Endocarditis in Hemodialysis Patients in a Third-Level Hospital in Panama. Cureus. 2024 Jan;16(1):e52385. 38361669

10.7759/cureus.52385

Price

Sarween

Gupta

: Meticillin-resistant Staphylococcus aureus and meticillin-susceptible Staphylococcus aureus screening in a cohort of haemodialysis patients: carriage, demographics and outcomes. J. Hosp. Infect. 2015 May 1;90(1):22–27. 25676114

10.1016/j.jhin.2015.01.001

Hirabayashi

Yahara

Oka

: Comparison of disease and economic burden between MRSA infection and MRSA colonization in a university hospital: a retrospective data integration study. Antimicrob. Resist. Infect. Control. 2024 Feb;13(1):27. 38424606

10.1186/s13756-024-01383-8

PMC10905874

Gidengil

Gay

Huang

: Cost-effectiveness of strategies to prevent methicillin-resistant Staphylococcus aureus transmission and infection in an intensive care unit. Infect. Control Hosp. Epidemiol. 2015 Jan;36(1):17–27. 25627757

10.1017/ice.2014.12

PMC4311265

Sakr

Brégeon

Mège

J-L

: Staphylococcus aureus Nasal Colonization: An Update on Mechanisms, Epidemiology, Risk Factors, and Subsequent Infections. Front. Microbiol. 2018;9:2419. 30349525

10.3389/fmicb.2018.02419

PMC6186810

Mermel

Cartony

Covington

: Methicillin-Resistant Staphylococcus aureus Colonization at Different Body Sites: a Prospective, Quantitative Analysis. J. Clin. Microbiol. 2011 Mar 1;49(3):1119–1121. 10.1128/JCM.02601-10

Christopher

Verghis

Antonisamy

: Transmission dynamics of methicillin-resistant Staphylococcus aureus in a medical intensive care unit in India. PLoS One. 2011;6(7): e20604. 21750700

10.1371/journal.pone.0020604

PMC3130025

Zacharioudakis

Zervou

Ziakas

: Meta-analysis of methicillin-resistant Staphylococcus aureus colonization and risk of infection in dialysis patients. J. Am. Soc. Nephrol. 2014 Sep;25(9):2131–2141. 24652802

10.1681/ASN.2013091028

PMC4147983

Keene

Vavagiakis

Lee

M-H

: Staphylococcus aureus Colonization and the Risk of Infection in Critically Ill Patients. Infect. Control Hosp. Epidemiol. 2016/06/21.2005;26(7):622–628. 16092742

10.1086/502591

Montoya

: Clinical and epidemiological information collection form. 2024. 10.6084/m9.figshare.26170027

Reference Source

Montoya

: Markov Models. figshare.Software.2024. 10.6084/m9.figshare.27313296.v1

Montoya

: Follow-up Hemodialysis Patients. figshare.[Dataset].2024. 10.6084/m9.figshare.27313290.v1

10.5256/f1000research.174101.r337447

Reviewer response for version 2

Abdulrahman

Thanaa Rasheed

1 Referee 1Department of Microbiology, University of Al-Nahrain, Baghdad, Iraq

Competing interests: No competing interests were disclosed.

5 12 2024

2024

This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

recommendation

approve

Greetings

All is well and the author is committed to making corrections

With my respects and best regards

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Partly

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

10.5256/f1000research.174101.r337446

Reviewer response for version 2

Harbarth

Stephan

1 Referee 1Geneva University Hospitals, Geneva, Switzerland

Competing interests: No competing interests were disclosed.

18 11 2024

2024

recommendation

approve-with-reservations

Thank you for the invitation to review the updated manuscript. The authors completed a thorough revision of their manuscript and addressed our previous concerns. They now explain the rationale for using Markov models for this analysis, making this manuscript clearer and more accessible for readers without a background in mathematical modelling. They included extended data and additional information needed for the reader while interpreting the study findings. Finally, they added a limitations section, summarizing all the points we raised in our review process. After reading the revised manuscript we have a few comments to be addressed:

Major comment:

The authors present hazards ratio estimates for transition between different states under “ Effect of covariates on the dynamics of two-state colonization by S. aureus” section, yet it is not clear what methods were used to obtain these estimates, and how does it fit within the modelling framework. This should be better explained in the statistical analysis in the methods section.

Minor comments:

In the definition of bacteremia, it would be more accurate to state that primary bloodstream infection (BSI) and catheter related BSI (CLABSI) according to the CDC criteria were included. As in CLABSI a clear source of infection is identified (catheter) and in the current phrasing is contradictory: “ Infection was defined as the diagnosis of bacteremia according to the criteria given by the Center for Disease Control and Prevention (CDC), as the presence of fever, chills or hypotension with bacteria identified in the blood and not related to an infection at any other site. Central-line associated bloodstream infections (CLABSI) were also considered in the analysis”.

The last sentence in the conclusions is not supported by the study findings: ” Finally, the use of modeling tools to address this type of problem allows the continuation and improvement of active epidemiological surveillance, making it possible to identify, design and implement evidence-based strategies aimed at preventing colonization and infection in these patients.” The authors could argue instead on the utility of modelling studies to evaluate decolonization or other infection control interventions in this patient population to inform future real-world epidemiological and interventional studies.

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Reviewer Expertise:

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Montoya Urrego

Daniela

Escuela de Microbiología, Universidad de Antioquia, Medellín, Antioquia, Colombia

Competing interests: The authors declare that no competing interest exist.

24 12 2024

Dear reviewers,

Thank you for guidance reviewing our submission. The manuscript has been revised and your comments have been directly addressed. We believe that these revisions lead to a significantly improved manuscript and we hope it is now acceptable for publication.

Major comment:

The authors present hazards ratio estimates for transition between different states under “Effect of covariates on the dynamics of two-state colonization by S. aureus” section, yet it is not clear what methods were used to obtain these estimates, and how does it fit within the modelling framework. This should be better explained in the statistical analysis in the methods section.

Author Response: Thank you for your valuable comment. The relationship between individual characteristics—whether time-invariant or time-varying—and their transition rates is often a key focus in multi-state models. Explanatory variables for a specific transition intensity can be investigated by modeling the intensity as a function of these variables. Marshall and Jones described a form of proportional hazards model in which the elements of the transition intensity matrix q _rs of interest can be expressed as:

The updated Q matrix is then used to compute the likelihood. When covariates Z(t) are time-dependent, the likelihood contributions of the form P _rs (t - u) are replaced by:

This requires that the value of the covariate is known at each observation time u. In some cases, covariates are observed at different times than the main response, such as during recurrent disease events or through other biological markers. In such cases, covariates may be treated as step functions, remaining constant between their observation times. If the primary response (the state of the Markov process) is not observed when the covariate changes, it may be treated as a "censored" state.

The msm package allows for the incorporation of both individual-specific and time-dependent covariates into transition intensities. To calculate the transition probabilities P ( t ), upon which the likelihood depends, it is assumed that time-dependent covariates are constant within specific intervals. Models with transition intensities that vary over time are classified as time-inhomogeneous models.

Marshall and Jones also described likelihood ratio tests and Wald tests for covariate selection and hypothesis testing. For instance, these tests can determine whether the effect of a covariate is consistent across all forward transitions in a disease progression model or if the effect on backward transitions is equal to the negative effect on forward transitions.

Finally, covariates can be included in the model to examine their impact on transition rates, enabling the construction of an intensity matrix that depends on covariates. This approach is based on the assumption that covariates remain constant between the observation times of the Markov process.

This has now been clarified in the statistical analysis section.

References:

Marshall G, Jones RH. Multi-state models and diabetic retinopathy. Stat Med. 1995 Sep;14(18):1975–83.

Jackson C. Multi-State Models for Panel Data: The msm Package for R. J Stat Softw. 2011 Jan 4;38(8):1–28.

Minor comments:

Comment 1:

In the definition of bacteremia, it would be more accurate to state that primary bloodstream infection (BSI) and catheter related BSI (CLABSI) according to the CDC criteria were included. As in CLABSI a clear source of infection is identified (catheter) and in the current phrasing is contradictory: “Infection was defined as the diagnosis of bacteremia according to the criteria given by the Center for Disease Control and Prevention (CDC), as the presence of fever, chills or hypotension with bacteria identified in the blood and not related to an infection at any other site. Central-line associated bloodstream infections (CLABSI) were also considered in the analysis”.

Author Response: We fully agree with your observation. This has now been clarified in the methods section to ensure that the definition of bacteremia is more precise.

Comment 2:

The last sentence in the conclusions is not supported by the study findings: ”Finally, the use of modeling tools to address this type of problem allows the continuation and improvement of active epidemiological surveillance, making it possible to identify, design and implement evidence-based strategies aimed at preventing colonization and infection in these patients.” The authors could argue instead on the utility of modelling studies to evaluate decolonization or other infection control interventions in this patient population to inform future real-world epidemiological and interventional studies.

Author Response: Thank you very much for your suggestion. We agree with you, so the final part of the conclusion was modified according to your comment.

10.5256/f1000research.166587.r316810

Reviewer response for version 1

Harbarth

Stephan

1 Referee Kheir

Nasreen

1 Co-referee 1Geneva University Hospitals, Geneva, Switzerland

Competing interests: No competing interests were disclosed.

19 9 2024

2024

recommendation

approve-with-reservations

Thank you for the opportunity to review the interesting study by Montoya-Urrego et. al. The study addresses a relevant research question on the dynamics of Staphylococcus aureus colonization and bloodstream infection among hemodialysis patients in a Colombian healthcare setting. The investigators used transition intensities to quantify the rate at which patients move between non-colonization, colonization and infection states. They derive transition probability matrices and intensities from patient observations, which is the mainstay of multi-state modelling.

The study has important findings for clinicians taking care of hemodialysis patients: first , there is a higher probability for transition from a non-colonized to a colonized state, which highlights the challenge in S. aureus prevention. Second, it identifies previous infections, hospitalization, and antibiotic use as significant risk factors for transition to a colonized state, in accordance with existing literature. Last, it highlights the importance of active S. aureus surveillance in hemodialysis patients. All these findings have implications on infection prevention and control practice and can inform future interventions targeting this patient group, beyond conventional decolonization strategies.

After reading the manuscript we found that while the study rationale and objectives are clearly described, there are a few important issues that need to be better addressed in the sections about methodology, results, and discussion. Furthermore, a separate Limitations section should be added to the manuscript.

Major comments

Applying Markov models to study the dynamics of colonization and infection by S. aureus is an interesting approach, yet there is an important limitation to this approach that should be acknowledged. For the case of S. aureus colonization among hemodialysis patients, the probability of colonization depends also on the number of patients colonized in the same unit, namely ‘colonization pressure’, as well as on other factors such as hospital environment and thus, it is not at all random. This cannot be accounted in such a model for and should be discussed as an important limitation of the chosen modelling approach.

The model’s assumption of constant transition rates over time is problematic from a clinical perspective. Acquisition probability might change according to various factors, such as hospitalization, administration of antibiotics, use of invasive instrumentation etc. As such, the clinical relevance of these findings is questionable. This is another limitation that should be discussed in the manuscript.

The methods section lacks important information that is critical for interpreting the study findings. The authors cite their previous publication (Vanegas et al 2021) that served as a basis for this secondary data analysis; however, essential information should be described in the current manuscript as well, for example the complete lack of decolonization measures for patients undergoing chronic hemodialysis in the studied hemodialysis clinic.

Selection bias might be introduced by exclusion of patients with baseline measurement only (15% of all patients). The missing measurements could be not missing at random, as hemodialysis patients have increased mortality rates and can have renal transplantation, and the follow-up is not expected to be complete. These two events are competing events that should be accounted for in the model if such data exists, otherwise they should be further discussed in the limitations section.

In line with the previous comment, specific information on the status of colonization and infection for the patients at the different time points assessed should be presented in a supplement. This information is crucial to assess the internal validity of the model and should be disclosed to the reader.

The authors should consider providing an anonymized dataset (or a synthetic data set) and their analysis code for transparency, reproducibility, and educational purposes.

The definition used for bacteremia state may be problematic, as it includes only a subgroup of patients with primary bloodstream infection, excluding central-line associated bloodstream infections (CLABSI). The latter constitute a substantial part of S. aureus bloodstream infections in this patient population. This should be addressed, and better clarified in case CLABSIs were indeed included.

The sampling strategy for S. aureus colonization is not comprehensive, as it includes only nostrils and catheter insertion site. The colonization prevalence might have been different if additional sites were screened, see also: McKinnell JA et al, Infect Control Hosp Epidemiol. 2013. As such, the sensitivity of the screening approach in the primary study should be also discussed as possible source of bias.

There is no mention of the distinction between MRSA and MSSA in this analysis (presented in the index publication in 2021). Analysis by MRSA/MSSA might have not been possible due to a limited number of patients with MRSA included in the original dataset, yet this should be discussed in the current study as well.

Minor comments:

The external validity of findings is limited, since this study relies on a small patient sample in a single centre hemodialysis unit of patients receiving hemodialysis through a venous catheter, with three S.aureus sampling events. This should be discussed as a limitation.

As patients are included at an arbitrary timepoint during the hemodialysis treatment, it is unclear how these transition intensities behave during the time course of hemodialysis. This also should be discussed and possible implications on the study findings should be addressed.

The statistical methods could benefit from more detailed explanation including the theoretical relations between P,Q, qrs, with formula and which software was used, and what packages with their respective versions.

The abstract needs to be revised some phrases such as ‘some variables’ and ‘high complexity hospital’ can be substituted with more precise terms and data.

Optional suggestion to extend the current analysis:

The authors could consider an option of adding a complementary simulation analysis to enrich the current study. For example, a sensitivity analysis of the Markov model and/or a scenario-based analysis reflecting on specific infection prevention interventions. Possible suggestions:

Simulation with Non-Fixed Time Intervals: a time-varying/non-homogeneous Markov model can be used to simulate a scenario where transition rates can vary over time. This would better capture periods when the risk of infection or colonization is higher (e.g., hospitalization or administration of antibiotic treatments). The results obtained can be then compared to the homogeneous model to analyze the impact of varying transition rates on the duration of colonization or infection periods.

Simulation of the effect of preventive intervention strategies on colonization. This simulation can be used to evaluate different preventive strategies, such as improved hygiene measures or changes in care protocols (e.g., systematic disinfection of catheters/ decolonization of S. aureus carriers). This analysis could test the impact of such interventions on reducing transitions from non-colonization to colonization accounting for different levels of adherence to these measures and observe their effect on reducing transitions.

Structured feedback:

Is the work clearly and accurately presented and does it cite the current literature?

Partially, the literature review is adequate, yet additional information is needed in the methods, results and a limitations section in the discussion to improve the readability and interpretability of this manuscript.

Is the study design appropriate and is the work technically sound?

Yes, the data is sparse, with few measurement points for small sample, but overall, this manuscript has a nice implementation of this modelling approach.

Are sufficient details of methods and analysis provided to allow replication by others?

No, additional information is needed, and the analysis code should be shared in supplement or through an adequate repository (GitHub ,Zenodo, etc) please see specific comments above.

If applicable, is the statistical analysis and its interpretation appropriate?

Yes.

Are all the source data underlying the results available to ensure full reproducibility?

Yes, However, we still recommend that the number of patients with available data in each of the 3 measurement points to be reported in an aggregate manner in the manuscript.

Are the conclusions drawn adequately supported by the results?

Yes.

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Reviewer Expertise:

Healthcare epidemiology, infection control, MSSA/MRSA

We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however we have significant reservations, as outlined above.

Montoya Urrego

Daniela

Escuela de Microbiología, Universidad de Antioquia, Medellín, Antioquia, Colombia

Competing interests: The authors declare that no competing interests exist.

28 10 2024

Dear reviewers,

Thank you for your valuable feedback on our submission. We have carefully revised the manuscript and addressed all of your comments. We believe these changes have substantially strengthened the paper, and we hope it now meets the criteria for publication.

Major comments:

Comment 1:

Response:

We fully agree with your observation, so it was added to the discussion within the limitations of our study. We appreciate that you have raised this very relevant point.

Comment 2:

Response:

Thank you for your valuable comment. Although Table 2 analyses the influence of different covariates on transition risks, for example antibiotic use, previous hospitalizations, history of infections and presence of comorbidities, we recognize that assuming constant transition rates over time may overlook important clinical factors. We have addressed this limitation in the discussion section of the manuscript.

Comment 3:

Response:

We agree with the observation, thank you for your comment. Our previous publication (Vanegas et al., 2021) is cited because, as the reviewers point out, it forms the foundation for the analysis in this study. However, as suggested, we complete the methods section in more detail with essential information such as the absence of decolonization protocols.

Comment 4:

Response:

Thank you for your observation. We have added the reasons for patients having only one measurement in the results section. Additionally, we have included these events as a limitation in the discussion section, acknowledging that mortality and renal transplantation are competing events. In future studies, we will consider performing different types of risk analysis to account for such competing events. Thank you again for your value comment.

Comment 5 and 6:

The authors should consider providing an anonymized dataset (or a synthetic data set) and their analysis code for transparency, reproducibility, and educational purposes.

Response:

A synthetic data set was added with the information requested by the reviewers (Extended Data).

Comment 7:

Response:

Thank you for your observation. We confirm that central-line associated bloodstream infections (CLABSI) were indeed included in our analysis. This has now been clarified in the methodology section to ensure that the definition of bacteremia is more precise.

Comment 8:

Response:

We recognize that S. aureus colonization can occur in multiple body sites, but the nasal region is the most common and is considered a strong indicator of overall colonization. Studies show that nasal colonization correlates highly with colonization at other body sites, so focusing on the nostrils offers high sensitivity for detecting S. aureus colonization. However, we will take your valuable suggestion into account in future studies to improve the accuracy of detection, and we included this limitation in the discussion section. Thank you for your observation.

References:

Sakr A, Brégeon F, Mège J-L, Rolain J-M, Blin O. Staphylococcus aureus Nasal Colonization: An Update on Mechanisms, Epidemiology, Risk Factors, and Subsequent Infections. Front Microbiol. 2018;9:2419

A. ML, M. CJ, Pauline C, Gail M, Dan M. Methicillin-Resistant Staphylococcus aureus Colonization at Different Body Sites: a Prospective, Quantitative Analysis . J Clin Microbiol. 2011 Mar 1;49(3):1119–21

Comment 9:

Response:

As noted by the reviewers, a separate analysis of the behavior of methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-susceptible Staphylococcus aureus (MSSA) was not possible due to the limited number of MRSA-colonized patients in the dataset. This point was added to the discussion section as a limitation.

Minor comments:

Comment 1:

The external validity of findings is limited, since this study relies on a small patient sample in a single center hemodialysis unit of patients receiving hemodialysis through a venous catheter, with three S. aureus sampling events. This should be discussed as a limitation.

Response:

Thanks for the comment, it has been discussed as a limitation in the manuscript. However, the institution where the research was conducted is one of the largest in the city and serves a diverse population from various areas, which increases external validity.

Comment 2:

Response:

The hemodialysis time of each patient was not taken into account in this study, so it is discussed within the limitations. Thank you for the observation.

Comment 3:

Response:

The explanation of the statistical analysis has been improved in the methods section, including information on the software used. Thank you for your feedback.

Comment 4:

The abstract needs to be revised some phrases such as ‘some variables’ and ‘high complexity hospital’ can be substituted with more precise terms and data.

Response:

The Abstract was adjusted according to your recommendations.

Optional suggestion to extend the current analysis:

Response:

Thank you for your valuable suggestions on possible complementary analyses. We agree that incorporating additional simulations would provide a clearer understanding of the dynamics of the model. In particular, it would be interesting to explore time-varying transition rates and assess the impact of infection prevention interventions. We assure you that these insights will be taken into account for future extensions of this research. For the current study, we aimed to focus on providing an initial exploration of the transitions between non-colonization, colonization, and infection using a simpler Markov model.

Your recommendations will certainly guide us in refining and enriching our analysis in subsequent studies.

10.5256/f1000research.166587.r313718

Reviewer response for version 1

Abdulrahman

Thanaa Rasheed

1 Referee 1Department of Microbiology, University of Al-Nahrain, Baghdad, Iraq

Competing interests: No competing interests were disclosed.

5 9 2024

2024

recommendation

approve-with-reservations

Dear Editor

The answer below for both questions.

Are sufficient details of methods and analysis provided to allow replication by others? Are all the source data underlying the results available to ensure full reproducibility?

The author followed the Markov model in conducting the research in terms of the transition between colonization, non-colonization, and infection, but he did not explain the details of the test, how the period was calculated, and how long each transition was. The details of the work must be added to make it a reliable reference research.

Also, the author could have followed up on the period or generation time of bacteria between the transition from non-colonized to colonized or bacteremia, and whether it can contribute to stopping the transition or introducing a treatment to stop colonization.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Partly

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

Microbiology, medical, clinical and diagnostic bacteriology, moleculer microbiology, immunology, parasitology

Montoya Urrego

Daniela

Escuela de Microbiología, Universidad de Antioquia, Medellín, Antioquia, Colombia

Competing interests: The authors declare that no competing interests exist.

28 10 2024

Dear reviewer,

Comment 1:

Response:

The colonization screening process are detailed in the Methods section, likewise the screening periods established in the methodology were based on a previous study conducted by our research group, also considering that colonization can behave intermittently over time, appearing or disappearing at different moments. Several studies have shown that hemodialysis patients can transition between non-colonization and colonization in short periods of time, ranging from 2 weeks to 2 months after starting dialysis therapy, especially in those with catheters.

References:

Scheuch M, Freiin von Rheinbaben S, Kabisch A, et al.: Staphylococcus aureus colonization in hemodialysis patients: a prospective 25 months observational study. BMC Nephrol. 2019;20(1):153. 31060511 10.1186/s12882-019-1332-z PMC6503363

Grothe C, Taminato M, Belasco A, Sesso R, Barbosa D. Screening and treatment for Staphylococcus aureus in patients undergoing hemodialysis: a systematic review and meta-analysis. BMC Nephrol. 2014;15(1):202

Agrawal V, Valson AT, Bakthavatchalam YD, Kakde S, Mohapatra A, David VG, et al. Skin Colonizers and Catheter Associated Blood Stream Infections in Incident Indian Dialysis Patients. Indian J Nephrol. 2022;32(1):34–41

For this reason, repeated measurements were taken over time to capture these changes in patient status. While we acknowledge that conducting more frequent measurements would have been ideal, it was not possible in this study, because the budget available for the research was only sufficient for three observations over time. However, we appreciate the comment and will take it into account for future research.

Regarding the duration of each transition, this was calculated using the transition matrix Q, in which temporal homogeneity is assumed in the transitions between states. In this study, we used a continuous-time Markov model to analyze transitions between non-colonization, colonization, and infection states. Markov models are particularly well-suited for this type of research because they allow us to model the dynamics of transitions over time based solely on the patient’s current state without considering previous states, this is a Markovian property of homogeneity. This approach assumes that once the patient enters a given state, the probability of moving to a new state depends only on that current state and not on any prior states. This allows us to calculate the probabilities of changing from one state to another based on the patient's current state, maintaining the constancy of the transition probabilities over time. The methods section was enhanced in the manuscript to better explain this part.

Comment 2:

Response:

Given the intermittent nature of colonization, measurements were taken at short intervals to capture and identify all potential transitions, including the onset of bacterial colonization. For example, we observed cases where patients initially showed a non-colonization status at baseline but transitioned to colonization at subsequent measurements. Importantly, all patients started follow-up without bacteremia, and follow-up continued until the first episode of bacteremia, so the occurrence of bacteremia did not influence data on transitions between non-colonization and colonization. Furthermore, it is worth mentioning that the renal unit did not have a decolonization protocol in place, meaning that patients were not treated to control colonization. Treatment was only started when an infection was present. However, antibiotic consumption was included in the state-to-state transition analysis in order to assess the influence of this covariate on transition risk (Table 2).