MRSA isolation practices and patient experiences in hospital infection control

¶ … Best Practice to Isolate MRSA Patients in the Hospital Environment

Methicillin-Resistant Staphylococcus Aureus (MRSA) is an anti-microbial organism of concern in the healthcare field; therefore, preventing and controlling its spread within the healthcare environment is a significance function of the infection control experts. One of the preventive measures is isolation of MRSA patients. However, not all are in agreement with such a practice and this makes our topic of discussion raise the question, "Is isolation of MRSA patients in a hospital environment the best practice?"

This reviews aims to evaluate the evidence for the efficiency of isolation measures in managing the prevalence of Methicilin Resistant Staphylococcus Aureus (MRSA) and infection in the hospital environments.

Review Methods: This review relied on scholarly reviewed journals, and selected articles reporting on MRSA related outcomes and illustrating an isolation practice or policy. The paper did not impose quality restrictions on studies employing separation wards or nurse cohorting. In addition, this review utilizes studies, if they showed prospective comparisons of retrospective information.

Results/Discussion: This review accepted 46 studies; 18 using isolation wards, 9 employing nurse cohorting, and 19 had evidence of isolation policies. However, these studies lacked measures to prevent bias, and statistical analysis. In addition, four of the studies provided substantial evidence that intensive control measures including isolation showed efficiency in controlling MRSA.

Conclusion: This review recognizes that major methodological limitations and inadequate reporting in published research showing that the paper includes alternative explanations for reductions in MRSA acquisition linked with interventions.

Introduction

The prevalence of hospital acquired methicillin resistant Staphylococcus auerus (MRSA) continues to rise in the globe hence qualifying as an international health problem (French et al., 2004). The effort to manage the spread of MRSA relies primarily on three measures including hand hygiene among health practitioners (Larson, Quiros and Lin, 2007), restriction of antibiotics, and detection and isolation of infected or colonized patients (Cooper et.al, 2003; Gbanguide-Haore et al., 2008). Perception is that most transmission of MRSA within the hospital environments (French et al., 2004), between patients happens through transiently colonized health workers (Gbanguide-Haore et al., 2008); moreover, airborne transmission through contact with infected or contaminated patients is important for consideration.

The most of the rigorous forms of patient isolation included isolation wards (designed to handle the treatment of known or suspected carriers of MRSA disease) and nurse cohorting, (Talon et al., 2003) (a physical separation of MRSA patients in a part of the ward, with nurses specifically designated to care exclusively for these patients). Into the bargain, other isolation precautions include the use of single bedded patient wards, cohorts of patients on common patient rooms (Talon et al., 2003), which lacked designated staff, and barrier caution (using aprons, gowns, gloves, and masks by healthcare professionals, (Manian and Ponzillo, 2007) as the only physical measure to avoid transmission of MSRA).

However, these control precautions may place significant burden on the hospital resources and the importance of their continued use is in question (Weeber, 2005). Although there are several narrative reviews, the efficiency of isolation measures in reducing transmission and managing MSRA, lacks a systematic evaluation (Hulten et al., 2006). Nevertheless, much of the investigation carried out is quasi-experimental in nature; therefore, there is a need to consider associated threats to valid inferences. This is the primary objective of this paper, to review for evidence on the effectiveness of isolation measures in the management of MRSA within a hospital setting (French et al., 2004).

Method

Search strategy

This review utilizes a search strategy that covers the primary subject areas of the literature review (MRSA, screening, isolation of patients, and control of MRSA infection). The following databases were of significance in this review: Cochrane, TRIP, CINAHL, Medline, National Guidelines Clearing House (NGC), PubMed Clinical Queries, Johanna Briggs and EBSCO. In addition, the review utilized studies published over the year 2000.

Study selection

These review appraised abstracts and it obtained articled if the abstracts mentioned MRSA and an attempt to control it in a hospital environment. Owing to the great number of studies, which it had not anticipated, this review revised the studies that did not impose quality restrictions (Fleming et al., 2006). Therefore, the review imposed the minimum requirement, whereby the accepted studies should include a component of prospective data collection. In addition, if the studies were retrospective, comparisons should show planned and not prompted by part of the data outcome. For the studies incorporating the most rigorous forms of separation (isolation wards and nurse cohorting), lacked such restrictions. This is because such forms of isolation had the greatest influence on hospital resources and organization of services.

Data extraction

This review divided studies into phases appropriately, defining them with major changes in isolation or other elements of infection control policy and extracted data on study design, patient population, isolation facts, screening, other management measures and MRSA related results for patients. In addition, this review documented potential threats to the internal validity of the accepted studies. Therefore, there was consideration of vulnerability on each study selection, performance, and detection and attrition bias (Cooper et al., 2003). Documentation of the measures aimed to prevent bias, noted potential confounders and attempts to record, and adjusts for the measures. In addition, the primary cause of documentation was because of underlying trends, seasonal effects, regression to the average impacts, defined as "the tendency for extreme measurements to be followed by less extreme measurements for imperfectly inked variables that mostly bring about outcomes in wrong conclusions about the impact of interventions." In addition, the review evaluated the appropriateness of any statistical analysis undertaken. This review excludes studies with unclear timing of interventions and main isolation policies.

Type of bias

Cause

Studies vulnerable

Measures to prevent bias

Selection bias

Variation in intervention groups on study entry

Random studies

4-35 interrupted times. Isolation altered presented data allowing for comparison

Performance bias

Variation in care patients apart from interventions under research

Studies with specified element of care was not investigated

One study showed alteration of confounders

Varied length of days

29 interrupted times

Four studies details antibiotic use

Variation in bed occupancy

31 interrupted times

Comparisons in bed occupancy between four

Detection bias

Varied outcome evaluations between intervention groups

All studies

Three studies reported blinding elements of assessors

Variation in diagnosing infections

26 studies

16 of the studies specified diagnostic criteria

Variations in screening measures

10 studies presented colonization data

This suggested that screening efforts could not explain the alterations

Attrition bias

Variations in losses to follow up between the treatment groups

26 studies outcomes with infections

No studies followed up patients after discharge to identify the hospital acquired infections

Data synthesis

This review independently evaluates the strength of evidence in each of the studies by exploring the study design, quality of data, and evidence of plausible alternative explanations of results. In addition, the review characterizes the evidence systematically by basis of "none," "weak," "of intermediate strength," and "stronger." In addition, it considers a formal meta-analysis inadequate due to heterogeneity based on outcome of measures and patient populations.

Results

The electronic search yielded 4,382 abstracts; hand searching did not generate additional papers, the appraisal of abstracts chose 254 papers, including 20 in foreign languages other than English. However, the final review involved 46 studies (Cooper et al., 2003).

Characteristics of the accepted studies

Highest level of separation

Isolation ward

Nurse cohorting

Other isolation precautions

No of studies

18

9

19

Range of study time

3 months-15 years

3.5months-4 years

1 monht-9 years

Entire hospital setting

16

3

7

Unit setting in each hospital

2

6

12

Other Control Measures

Screening

18

9

14

Topical eradication therapy

12

5

8

Hand hygiene

8

2

6

Antibiotic restriction

3

0

2

Study Design

Prospective interrupted time series

1

2

8

Retrospective interrupted time series

15

3

2

Hybrid retrospective and prospective time series

0

2

1

Retrospective cohort study

0

0

1

Non-comparative studies

2

2

3

Study Design

These review doses not find randomized trials and only four retrospective planned comparisons with predefined phases of the study (Mody et al., 2003). In most of the studies, designs faced interruptions time series, meaning time series of results measures recorded before one or more interventions. Moreover, eight of the 38 interrupted time series studies presented only collapsed information, analyzing the time series from both segment in a distinct data top. In addition, one retrospective cohort research used survey data from Dutch healthcare facilities (Talon et. al., 2003). Some ten studies did not weigh against isolation or screening measures with regard to separation or screening. A review of the 36 studies that allowed for comparisons between isolation policies showed that in 27 of the comparison, they had independence on knowledge of the data results. However, short retrospective studies with successful results were vulnerable to this challenge.

Review

The review identified some studies that researched on the use of isolation or cohorting of patients to manage the transmission of MRSA. In the review studies that illustrated the isolation of patients as an intervention to manage the transmission of MSRA; however, no concrete conclusions were available about the impact (King et al., 2006) of isolation from a third of the studies. Concerning these inadequate or lacking conclusions, the evidence provided was weak due to poor design, major confounders, or systematic bias with regression to the evidence of average (Cooper et al., 2003; Allen 2006 and Blumeberg et al. 2006). However, this review chose some studies that offered stronger evidence. In three studies, the use of isolation and other interventions was conflicting (Carrbone et al. 2002; Charlebois 2002; Cohen 2005), one showed reduced infection (Cohen, 2005; Curran et al. 2005), while another showed no infection and another managing infection for some years until the number (Robert et. al., 2006) of colonized patients on admission overwhelmed the hospital (Gbanguide-Haore et al., 2008).

However, another study provided evidence that single room with screening, eradication and a far-reaching hand hygiene (Larson, Quiros and Lin, 2007; Dupeyron et al., 2002) program reduced MRSA infection and colonization in the hospital setting (Graham, Lin and Larson, 2006). Another study conducted in an intensive care unit, single room isolation and patient cohorting, together with screening, feedback and hand hygiene education, efficiently led to reduction of MRSA (Larson, Quiros and Lin, 2007). In addition, a modeling study conducted, several scenarios provided for this review. For this modeling study, the use of an isolation ward for colonized, detected, or infected patients, this reduced incidences of MRSA and in some instance, eradicated the disease completely from the hospital environment. In addition, either improving the detection rate or isolation (Hulten et al., 2006) capacity revealed to decrease the MRSA disease as long as none of them was the limiting factor.

It is important to note that the endemic level of MSRA may at times depend on the opening time of the isolation wards. This is because eradication of MSRA is possible when the isolation wards are opened early (Dupeyron et al., 2002). However, in many modeling scenarios, the failure to control MRSA on long-term basis resulted due to saturation of isolation facilities because there was an increase in numbers of admitted patients under colonization (Graham, Lin and Larson, 2006). Nevertheless, these isolation wards delayed the rate of prevalence, but there was reduction in the endemic levels. Overall, intensive concerted efforts including isolation work well in reducing MRSA even in environments showing high endemic levels of MRSA.

Another study on infection control management within acute care hospitals (Halcomb, Fernandez and Griffiths, 2002), evaluated the alteration in the rate of nosocomial acquisition of MRSA. From this study, it is clear that formulating good quality strategies may help in reducing MRSA. However, it was difficult to determine the clinical efficiency of strategies due to significant confounders. Although there was no study measuring staff compliance with the infection management practices (Cooper et al., 2003), the available information commented that there was a correlation on staff adherence to infection management measures. Some additional studies identified by this review, one conducted in the ICU (Grundmann 2002; Forceville et al. 2002), one in a cohort unit, and another in a geriatric ward (Talon et al., 2003). All these studies fell under observational studies using descriptive structures, illustrating multiple interventions in single wards therefore did not fit for generalization.

In addition, a retrospective analysis of the implementation of several infection control precautions including isolation, in a unit involving elderly people, with endemic levels of MRSA demonstrated statistically a substantial reduction in MRSA bacteraemia over a period of four years (P1, P < 0.0008; P2, P< 0.001; P3, P < 0.001). In a study, researching the impact of the introduction of contact measures, patient isolation and education advancement for healthcare providers, in a hospital with a teaching, (Wang et al., 2004) environment with long-term care and intensive care beds, there was a 17.9% decrease in MRSA infection, colonization (Graham, Lin and Larson, 2006), [MRSA prevalence 0.89 per 1000 patient days, 95% confidence intervals (CI) 0.81-0.97], (Cooper et al., 2003). In addition, there was evaluation of compliance in the using of contact isolation materials (gloves, gowns, masks, soaps) over a period of six months.

The review borrows from a research investigating the link between nosocomial MRSA infection levels and ICU structure and process parameters. The research used the German nosocomial surveillance network (Gustemeir, 2004). The analysis showed that a routine isolation of patients with MRSA led to low rates of MRSA (odds ratio 0.36, 95% CI 0.17-0.79). This review recognizes that some studies had consistent weaknesses when reporting on isolation as a measure for managing MRSA. In addition, the tendency to report the consequent implementation of multiple measures increases the difficulty in making clear conclusions about the impact of isolation as a measure (Cooper et al., 2003). Overall, the current review identified a number of studies that offered substantial value to the existing proof that isolation of patients can contribute in decreased MRSA outbreaks and endemic cases.

Summary of evidence

Most of the evidences offered in each study were descriptive rather than quantitative. In additional, formal meta-analysis of outcomes was inappropriate owing to the diverse settings and outcome precautions. Therefore, this review does not evaluate data using appropriate statistical techniques. In most studies, there was a report of insufficient data allowing for further evaluations. Although some studies illustrated time series, these time series are only appropriate from continuous results, and count data only when the numbers are large to justify a continuation. This calls for a formal time evaluation to varied approaches in analyzing the time series because count data typically arise from epidemics within the hospital setting.

Constraints of this review

Some of the process limitations included restricting number of the searched databases and ignoring an investigation for grey literature. In addition, the time may have led to missing of some evidences. Review of this paper involves an individual; therefore, it needs some evaluation in order to ensure accuracy of quality evaluations on all data extracted. In addition, in the search, some foreign papers came up, but it was not possible to evaluate for quality (Cooper et al., 2003). Another challenge is that, it was not possible to contact authors in order to identify or verify missing information.