Effective infection prevention and control measures in long-term care facilities in non-outbreak and outbreak settings: a systematic literature review

Background Healthcare-associated infections in long-term care are associated with substantial morbidity and mortality. While infection prevention and control (IPC) guidelines are well-defined in the acute care setting, evidence of effectiveness for long-term care facilities (LTCF) is missing. We therefore performed a systematic literature review to examine the effect of IPC measures in the long-term care setting. Methods We systematically searched PubMed and Cochrane libraries for articles evaluating the effect of IPC measures in the LTCF setting since 2017, as earlier reviews on this topic covered the timeframe up to this date. Cross-referenced studies from identified articles and from mentioned earlier reviews were also evaluated. We included randomized-controlled trials, quasi-experimental, observational studies, and outbreak reports. The included studies were analyzed regarding study design, type of intervention, description of intervention, outcomes and quality. We distinguished between non-outbreak and outbreak settings. Results We included 74 studies, 34 (46%) in the non-outbreak setting and 40 (54%) in the outbreak setting. The most commonly studied interventions in the non-outbreak setting included the effect of hand hygiene (N = 10), oral hygiene (N = 6), antimicrobial stewardship (N = 4), vaccination of residents (N = 3), education (N = 2) as well as IPC bundles (N = 7). All but one study assessing hand hygiene interventions reported a reduction of infection rates. Further successful interventions were oral hygiene (N = 6) and vaccination of residents (N = 3). In outbreak settings, studies mostly focused on the effects of IPC bundles (N = 24) or mass testing (N = 11). In most of the studies evaluating an IPC bundle, containment of the outbreak was reported. Overall, only four articles (5.4%) were rated as high quality. Conclusion In the non-outbreak setting in LTCF, especially hand hygiene and oral hygiene have a beneficial effect on infection rates. In contrast, IPC bundles, as well as mass testing seem to be promising in an outbreak setting. Supplementary Information The online version contains supplementary material available at 10.1186/s13756-023-01318-9.


Background
In the United States, there are 65,600 regulated long-term care facilities (LTCF).Around 70% of people turning 65 are expected to need long-term care at some point in their life, and 18% of the older persons will spend over a year in a nursing facility [1].Similar data exist for Europe, where approximately 3 million long-term care beds exist in nursing and residential care facilities in the 26 EU member states for which data are available in 2020 [2].
Healthcare-associated infections (HAI) are a major threat in acute and long-term care [3].Point prevalence studies from Switzerland demonstrated that between 2.0 and 4.4% of nursing home residents are affected by HAI [4].In combination, these numbers indicate that a large proportion of the population will sooner or later be affected by HAI in a long-term care institution and that there is an essential need for effective HAI preventive and control measures in these settings [3].The Covid-19 pandemic underlined the strong need for recommendations to prevent HAI in long-term care [5].
While infection prevention and control (IPC) measures and outcomes are well defined for acute care hospitals in the World Health Organization (WHO) core components for infection prevention [6], data are scarce for long-term care settings.
In a thorough review by Lee et al., published 2019 prior to the Covid-19 pandemic, the authors were unable to identify a set of measures that could be proposed for implementation of effective IPC measures [7].Up to this review, only a few high-quality studies were available [7].
In the current study, we aimed to both, update the findings by Lee et al. and complete by focusing on the Covid-19 pandemic in order to provide an overview of the current literature, identify existing research gaps and propose IPC measures and that could uniformly be recommended in long-term care.For the analysis, we differentiated between non-outbreak and outbreak settings.

Methods
The methods and results are reported according to the Preferred Reporting Items for Systematic Review and Meta-analyses (PRISMA) statement 2020 [8].

PICOS statement
The population of interest was defined as residents and healthcare workers in adult LTCF.Interventions included any IPC measures in accordance with the WHO core components for infection prevention even if they were mainly developed for acute care settings [9].Furthermore, we included oral hygiene as IPC measure as it has been shown to have a beneficial effect on infection rates in other settings [10].No restrictions in terms of comparisons were made.Outcomes were defined as HAIs or HAI prevention measures, mortality or transmission events, as well as healthcare worker attributes such as IPC knowledge or adherence to measures.

Search strategy
In order to cover the most recent scientific evidence, with a specific focus on the Covid- 19  ).Thereby, we built on the search strategy used in the most comprehensive existing review [7], but extended the time frame from 2017 until the 4th of November, 2022.In addition, reference lists of reviewed articles were scanned and the results combined.

Eligibility criteria
We included randomized controlled trials, observational studies (cohort and case-control studies) and quasi-experimental studies (before-after studies) in non-outbreak settings and outbreak reports.Studies were included if they were published in English and reported results from an infection prevention intervention in adult LTCFs.
Article types such as review papers, letters, editorials, expert opinions, ecological studies and study protocols were excluded, as were studies from pediatric long-termcare settings.

Study selection
Four main authors (NB, DF, SPK, and JM) screened searched titles and abstracts of each reference identified by the search.If the study met the eligibility criteria, the full-text article was reviewed independently for definitive inclusion by two authors each.In case of disagreement or in unclear cases, a third author made the decision about final inclusion.

Data extraction
Study data were extracted by the same authors (NB, DF, SPK, and JM), including setting, study design, main topic, type of intervention, and outcomes, using a standardized data collection form.An intervention was rated as successful when a statistically significant effect in the primary outcome was observed.
Included studies were further classified into non-outbreak versus outbreak settings.

Quality assessment
To assess methodological quality and risk of bias, we used the Cochrane risk-of-bias (RoB) 2.0 tool for randomized controlled trials, and the Newcastle Ottawa Quality Assessment Scale for Cohort studies and case-control studies [11,12].Each included study was assessed by one author and classified as high, medium, or low quality.
If the judgement in all key domains was 'low risk of bias' for RCT or achieved one star within every category for observational studies, the study was determined to be high quality.If the judgement in one or more key domains was 'unclear' or had 'some risk of bias' in the RoB 2.0 tool or achieved most but not all stars in the Newcastle-Ottawa-Scale, the study was evaluated to be medium quality.If the study was assessed to be at high risk of bias in one or more key domains for RCTs or failed to meet most of the stars for observational studies, the qualitysummary was deemed to be low in quality.Single-arm trials and outbreak reports were classified as low quality.
In order to avoid duplication and for better readability, most results are either presented in the detailed tables or in the main text.
Detailed descriptions of the respective investigated infection control and prevention measures are given in Tables 1 and 2.

Study characteristics
The literature search yielded 8675 references (Fig. 1).After the screening of titles and abstracts, we selected 150 studies for full-text screening.Seventy-four studies met the inclusion criteria and were included  (Tables 1, 2).Details for study type, study quality, place of study, and type of intervention are summarized in Table 3.
No study evaluated hand hygiene alone in an outbreak setting.

Antimicrobial stewardship
Four studies in non-outbreak-settings on antimicrobial stewardship which also measured the infection rates were included in our review [13][14][15][16].Three could demonstrate a reduction of C.difficile infections through antimicrobial stewardship [14][15][16], while one retrospective quasi-experimental study showed no decrease of MDROincidence or C.difficile infections [13].
In an outbreak setting no studies on this topic were undertaken so far.

Education
Two studies assessed the effect of education in IPC measures [19,20].Both were executed in a non-outbreak setting.One RCT found no difference of methicillinresistant Staphylococcus aureus (MRSA) prevalence in groups with IPC education [19].The other study recorded a successful outcome with a significant improvement of knowledge after education [20].
No studies were conducted to evaluate the effect of education alone in an outbreak setting.

Decolonization
One RCT assessed decolonization measures as main intervention in a non-outbreak setting [17] and found a reduction of MRSA prevalence after decolonization measures were implemented.No study evaluated decolonization measures in an outbreak setting.

Isolation precautions
One high-quality study from the USA evaluated the effect of isolation precautions alone with no significant difference in MDRO prevalence with/without isolation precautions [37].

Vaccination
We included three studies on vaccination in a non-outbreak setting [44][45][46].A high-quality trial from Japan showed a significant reduction in cases of pneumonia in residents of 23 LTCF after the 23-valent pneumococcal vaccine was introduced [46].Two studies were conducted in the non-outbreak setting with COVID-19 vaccination and showed a significant reduction in COVID-19 cases, COVID-19 related hospitalization and mortality [44,45].In outbreak settings, COVID-19 vaccination of residents significantly reduced outbreaks, COVID-19 cases, COVID-19 related hospitalization, and mortality in 3 of 4 studies.One study, executed in the turn of the year 2021 to 2022 showed no reduction in COVID-19 cases, but a reduced case fatality after vaccination [85].

Oral hygiene
Six studies evaluated the effect of improved oral hygiene on overall infection rates, all from a non-outbreak setting [38][39][40][41][42][43].All studies found a reduction of infections (mainly cases of pneumonia) with the intervention.
No publication on the effect of oral hygiene in an outbreak setting was recorded.

Mass testing
We found no study on mass testing in a non-outbreak setting.All studies that analyzed the effect of mass testing were performed in an outbreak setting during an early stage of the COVID-19 pandemic [72][73][74][75][76][77][78][79][80][81] and mostly resulted in the isolation of residents and quarantine of HCWs who tested positive.All of them found a significant number of asymptomatic HCWs and residents with a range of asymptomatic carriers from around 3% up to 43% in different studies.

IPC bundles
Half of the included studies (21% in non-outbreak-setting [18,[31][32][33][34][35][36] and 60% in outbreak setting ) focused on several topics simultaneously within an IPC bundle.In the non-outbreak setting one cRCT study evaluated a bundle of education of health care workers (HCW), surface cleaning, and feedback on HAI rates and could not observe a significant reduction in infection rates [34].
Furthermore, a large RCT in 104 long-term care facilities in Switzerland showed no effect of MRSA decolonization and different isolation precautions (standard vs. contact precautions) on MRSA prevalence [31].
In contrast, four studies could demonstrate a reduction of MDRO prevalence through a multicomponent intervention that included barrier precautions, active surveillance of MDRO and infections, as well as staff education and hand hygiene promotion [18,33,35,36].Koo et al. could at least show an improvement in knowledge for trained topics through an IPC bundle that included education while not evaluating infection rates [32].Twenty-four of 31 included studies on IPC bundles were IPC bundle infection and prevention control bundle performed in an outbreak setting .The included studies contained cohort and case-control studies, as well as outbreak reports.A median of 5 measures were included in an IPC bundle (range 2 to 8) with isolation/ precautions (N = 24, 19.7%), surveillance (N = 13, 10.7%) and hand hygiene (N = 9, 8.2%) being the most represented interventions included in the bundles.All outbreak reports showed containment of the outbreaks.When we differentiated by the transmission route, we found 15 studies where the transmission occurred mainly by respiratory droplets (SARS-CoV-2, Group A streptococci, Influenza-like illnesses) [47, 48, 50, 54-56, 58-61, 63, 65-68] and 8 studies with transmission via direct and/ or indirect contact (gastroenteritis, MDRO, Norovirus etc.) [49,51,53,57,62,64,69,70]. The bundles in these two categories varied slightly.The ones for pathogens transmitted through the respiratory route concentrated on wearing masks and repetitive testing, whereas those for direct or indirect contact transmissions focused more on environmental cleaning measures and contact precautions.

COVID-19
In the non-outbreak setting we found two articles focusing on the effect of vaccination on SARS-CoV-2 infection rates [44,45].Both found a positive effect of the vaccination on infection incidence in nursing home residents and staff as well as a reduced mortality in residents.

Other WHO core components
Other WHO core components for infection prevention, such as IPC programs per se, IPC guidelines, monitoring of IPC practices, reduction of workload, optimized staffing and bed occupancy rates as well as the environment, materials and equipment alone were not evaluated in the studies that were identified by our search.

Main results
In this systematic review, which also covers the SARS-CoV-2 pandemic, we identified 74 studies of different quality evaluating the effect of infection prevention and control measures in long-term care facilities in outbreak or non-outbreak settings, respectively.Hand hygiene, staff education measures, antimicrobial stewardship, vaccination and oral care seem to be consistently effective in preventing healthcare-associated infections or transmission events in long-term care settings.However, studies were mostly of low quality and highly heterogeneous with regard to setting, intervention measures, populations, and outcomes.Therefore, deriving standard of care recommendations or guidelines for LTCFs based on these data remains difficult.
Our current systematic review covers data from nonoutbreak and outbreak settings, especially during the SARS-CoV-2 pandemic, from a variety of countries worldwide.With a large increase in new publications during the COVID-19 pandemic, our study provides an update on the currently available literature on the effectiveness of different infection prevention measures in LTCFs in comparison to previous reviews.This allowed us to draw a more accurate picture of the current evidence on this topic.
For non-outbreak publications, our results regarding the effectiveness of different measures as well as the difficult comparability of the studies are in line with earlier well-made systematic reviews [7,87].In comparison to Lee et al., we identified relatively good quality data on the importance of hand hygiene, antimicrobial stewardship, vaccination and oral hygiene in addition to the already known beneficial effects of education, monitoring and multi-modal strategies.Of note, Lee et al. did not evaluate any antimicrobial stewardship interventions in their review [7].While Uchida et al. focused solely on therapeutic measures [87] we also analyzed studies on educational measures and focused more on the effect of the type of intervention.This allowed us to identify the particular contribution of various measures to a given outcome.
In contrast to others authors [7,[87][88][89][90], we included articles from the non-outbreak setting as well as from the outbreak-setting.While one review on IPC measures in the outbreak setting was conducted before COVID-19 [90], the others were published during the pandemic [88,89].
For the outbreak setting, mainly for studies on SARS-CoV-2, our review indicates that reasonably good data exist for the effectiveness of vaccination, mass testing, and IPC bundles, whereas no statement can be made about other single or combination of measures [71,72].Since outbreaks in general and virus-related outbreaks in particular are often self-limiting [91], it remains difficult to assess and put into context the added value of such transiently applied outbreak control measures.Whether an outbreak could be contained because of the IPC bundle or because of the temporary nature of outbreaks is impossible to discriminate in studies without control group.
It is to be expected that a combination of different measures produces an additive or synergistic effect, although, in our review, combinations of different measures were mostly applied in outbreak settings, with a difficult to evaluate outcome for the reasons mentioned above.Therefore, an additive or synergistic effect cannot be proven in our dataset.
Although education is often part of a bundle of measures, there is very little data on the importance of education alone.However, this should not limit the importance of education, which is extremely important in this context where health care workers are often insufficiently trained in medical and infection prevention and control.

Strengths and limitations
Our study has several limitations.First, generalizability is hampered in that we only included studies published in English and most studies in our review were performed in North America and Europe.As long-term settings vary widely within and across countries, settings and thus effectiveness of interventions may differ across institutions.Second, publication bias may have played a role in that ineffective IPC interventions may not be published, especially in outbreak settings.Furthermore, due to the heterogeneity and the low quality of studies, we were unable to compare effect sizes, let alone to meta-analyze effects across studies, even within similar settings or types of interventions.Last, we did not extend our search beyond PubMed and The Cochrane Central Register of Controlled Trials (CENTRAL), but given the quality and heterogeneity of identified studies, we are confident that searching further databases would not have led to more refined results.Another limitation of our study is the fact that LTC institutions provide medical and nursing care for different and rather heterogeneous resident populations in different countries.Thus, an identical measure could have a different clinical outcome based on the cognitive and or functional status of the persons living in the LTCF.This also applies to common geriatric syndromes such as frailty and/or malnutrition including urinary or stool incontinence.In addition the way how and by whom medical care is provided may have some impact upon the outcomes documented in our selected studies.
Strengths of our study are the inclusion of studies conducted in both non-outbreak and outbreak settings, including the COVID-19 pandemic and outbreaks of other pathogens, the inclusion of antimicrobial stewardship as a topic and the updated search until November 2022.Through this, we were able to recognize a large amount of studies with IPC measures not included in other reviews.

Conclusion and outlook
In conclusion, although we were able to find a good amount of data on IPC measures in the LTCF setting, interpretability and generalizability of these data remains difficult.Especially for outbreak settings, reports of successful control measures often do not add more value than do single case reports in the individual patient care setting.Given that the population at risk for healthcareassociated infections in these settings is large and constantly growing, coordinated action is imperative.In order to move a step forward and to complete the picture, well executed studies on this topic are desperately needed.These include a systematic evaluation of clearly defined single interventions or intervention bundles using high-quality (cluster-)-randomized controlled trials in well-defined settings and patient populations with useful outcome measures.These, due to the special needs of this population, do not only include HAIs, but also other measures such as quality of life, which sometimes might be favored over restrictive measures for infection prevention.In addition, IPC intervention trials and or measures across a clearly defined resident population and interventions that control for geriatric syndromes are urgently needed.Such efforts are only possible if sufficient funding for large, concerted, multi-national initiatives is available.
In general, it can be discussed whether reducing nosocomial infections is of high priority for the long-termcare setting or whether the focus should rather be on maintaining quality of life.Data on the influence of IPC measures on quality of life in long-term-care facilities are scarce or non-existing.From the COVID-19 pandemic, we assume that certain factors, such as visitor restriction, isolation measures and wearing masks for example, had an impact on the well-being of APH residents.
In the meantime, using the available low-quality evidence and extrapolating infection prevention and control measures from acute to long-term care with some common sense seem to be useful approaches.Thereby, the most essential basic IPC measures from the acute care setting, such as standard hygiene measures with hand hygiene and personal protective equipment when needed, combined with a good education for HCW and a functioning surveillance system might be the cornerstones of a successful IPC program in long-term care.Given that LTCFs are very heterogeneous with ever changing activities, defining the needs of every single institution is challenging.However, a standardized IPC program that every institution could adapt to its temporary needs may be a reasonable approach with a high acceptance on the part of the residents, HCW, and IPC team.

Fig. 1
Fig.1PRISMA flow diagram 2020.Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA statement[8]

Fig. 2
Fig. 2 Non-outbreak setting, divided in successful and non-successful intervention by type of intervention

Table 1
Included studies from the non-outbreak setting

Table 2
Included studies from the outbreak setting

Table 3
Characteristics of the included studies with respect of study quality,-type,-place and type of intervention