Abstract

Worldwide, home care in the form of home healthcare and home help, has become increasingly more available. This systematic review aims to provide a deeper understanding of factors that are of particular importance for infection control in the home care setting for older adults. Five databases were searched (MEDLINE, Embase, ProQuest, Web of Science, CINAHL) for eligible studies using any research design reporting on individual, medical, behavioral and environmental factors. Retrieved studies were screened and assessed for quality. The Joanna Briggs Institute manual guided the research process and the work of generating a synthesis. Qualitative findings were compiled using meta-aggregation. For quantitative evidence, meta-analyses were conducted when possible. Of 19,484 unique records, 27 studies (7 cohort studies, 9 cross-sectional and 11 qualitative) were included in the review. Risk factors for infection reported in the quantitative studies referred to individual, medical, social, behavioral, environmental, and organisational aspects. Meta-analyses showed associations between urinary catheter use (OR 3.97, 95%CI 2.56鈥�6.15) and limited mobility (OR 1.49, 95%CI 1.31鈥�1.68), respectively, and risk of infection. Pooled ORs of urinary incontinence and risk of infection were not statistically significant. Findings from the qualitative studies covered perceived and observed risk factors to infection control and prevention. The evidence resulted in five synthesised findings covering attitudes, behaviors, home environment, personal interactions, lack of equipment, unsafe disposal of material, pets, unsafe practices and procedures, and lack of training. The combined quantitative and qualitative evidence sheds light on separate yet interconnected elements of risks for infection that may reinforce each other, potentially exposing vulnerable older adults to amplified risks.

Introduction

Worldwide, more people are living longer with chronic conditions. This has increased the demand for care, which in turn has increasingly become more available in the older person鈥檚 home [1, 2]. Providing home care (home healthcare and/or home help) implies challenges, including factors that influence the risk of infection among older adults receiving care in their homes. These factors refer to the individual characteristics, such as older age, previous hospitalisation, poor health [3], a history of prior infections [4], and poor cognition [5]. Furthermore, behaviours, beliefs and attitudes have been reported to affect the risks of infection among older adults with home healthcare, including lack of knowledge and understanding of one鈥檚 illness, as well as hygiene practices and behaviours that prevent infections [6]. Accordingly, in a recent qualitative study with home healthcare nurses, informants reported that their patients鈥� knowledge of and attitudes towards infection prevention and engagement in hygiene practices influenced the patients鈥� behaviours regarding various infection control practices [7].

The practices of those providing home care, often nurse assistants and informal caregivers, have also been shown to influence the risk and spread of infection, especially in terms of good hand hygiene and use of sterile equipment [8]. Home healthcare staff have moreover reported on environmental factors such as clutter, poor lighting, uncleanliness, and pets as obstacles to infection control [6]. Such barriers may negatively affect patients鈥� infection risk through increased stress levels in healthcare professionals, which may in turn inhibit their adherence to adequate infection prevention practices [9, 10].

A systematic review from 2014 on the prevalence of infections and risk factors in home healthcare, showed that intravenous line-associated (catheter-related) infections (n =鈥�19 out of 25 studies) were the most common types of infections [11]. In one of the included studies using a representative sample of Americans with home healthcare (n =鈥�4,394; median age 74.6 years), the infection rate was 11.6% for all types of infections of which 3.6% were urinary tract infections (UTI) [12]. This makes infections in the home care setting a major concern, contributing to morbidity, mortality and considerable healthcare expenditures [13]. Further, in the review, the most common infection risk factors were patients鈥� underlying medical conditions, which was reported in 8 out of 14 studies that examined risk factors. Only one of all identified risk factors was non-medical (lower socio-economic status). Since the publication of the aforementioned systematic review in 2014, a substantial amount of research on infection risk factors has been conducted, of which some studies are referred to above. Several recent studies have examined risks for infection in the home setting beyond medical factors. These include behavioural or environmental factors, which constitute important aspects of infection risks that require due attention and consideration.

To add more knowledge to this subject-area, we conducted a systematic review investigating how facility characteristics, such as facility size, bed-capacity, the physical structure of the facility and staffing patterns, influence the spread of infectious diseases such as COVID-19 in care homes for older adults [14]. A total of sixteen studies were included. The findings suggested that larger facility size and urban location (compared to rural) may be associated with greater risks of an infectious disease outbreak. However, the risk of a larger outbreak seemed to be lower in larger facilities. Staff compartmentalizing seemed to be associated with lower risk of an outbreak whereas staff residing in highly infected areas appeared to be associated with greater risk of an outbreak. The systematic review on the spread of infectious disease in care homes triggered the development and planning of the present systematic review on risk factors for infection. In the current study, the home setting, rather than care homes, is of central focus. The home setting involves environmental challenges in terms of infection risks that are more difficult to overcome than in a care home. Several research studies published in the last few years have reported on multiple factors for the risk of infection in older adults with home care. However, the evidence from these studies has not been compiled and systematically reviewed in the past years, despite the pressing need for a review that synthesises and summarises existing knowledge using rigorous and transparent methods, generating implications for policy and practice.

With the present study, we seek to provide a deeper understanding of factors that are of particular importance for infection control in the home care setting. The importance of better protecting older adults with home care from infections and the potential consequences of hospitalisation and death became urgently apparent during the COVID-19 pandemic. The pandemic affected older adults and presented severe challenges in preventing spread of infection, specifically among older adults who depended on home care providers. This in turn highlighted the need for compiled evidence on factors that influence the risk of infections among older adults with home care.

Novelty of the present study includes the broad range of factors considered, including not only medical, but also individual, behavioural/social, and environmental factors. Additionally, contrary to the aforementioned systematic review by Shang and colleagues in 2014, the present systematic review includes qualitative research combined with quantitative studies and meta-analyses [11]. Another novelty of the current study is the inclusion of home help. The systematic review on infection risk from 2014 reported mainly on studies focusing on home healthcare [11]. Given that home help often involves close contact between caregiver and patient, infection risk is also highly relevant to home help, which in turn poses an increased risk for infection transmission [6]. Consequently, when investigating infection risk in older adults who live at home, it is essential to include home help in addition to home healthcare. In the present systematic review, home help refers to help received by professionals or informal carers to manage everyday life.

Methods

Study aim

This systematic review aims to identify risk factors for infection in older adults who receive home healthcare and/or home help. The structure of this review follows the guidelines for a mixed methods systematic review in the Joanna Briggs Institute (JBI) Manual for evidence synthesis [15].

Review questions

The overarching review question is: What may be understood as risk factors for infection in older adults who receive home healthcare and/or home help?

For the quantitative evidence, the specific review question is: What factors are associated with diagnosis of infection and/or infection symptoms in older adults who receive home healthcare and/or home help?

For the qualitative evidence, the specific review question is: How do home care professionals and clients perceive infection risk factors and barriers to infection control and prevention?

Inclusion criteria

To specify what data to be considered for the mixed methods systematic review, PICO (population, intervention, outcome, context) and additional criteria including phenomena of interest, context, and, type of study, was addressed and presented below.

Population

The review considered older adults who live in their own homes and receive healthcare and or home help by professionals or informal carers. Studies with populations of older adults living in long term institutions or who are hospitalized were excluded. Studies of populations receiving end-of-life care were excluded. We did not restrict the dose or duration of home help or home healthcare. Studies where the mean age of the sample was 65 years or above were considered eligible as 65 years of age often is used to define older adulthood [16, 17]. For some of the qualitative studies the age spans of the clients receiving home health care were not specified. Most of them, however, brought the theme of old age patients into the background and/or discussion sections of the article. We also knew that home health care clients, in their majority, tend to be over 65 years. Based on this information we opted for including the qualitative studies that fulfilled the rest of the PICO criteria.

Intervention/exposure

The review considered studies that assessed medical, individual, behavioural/social, environmental, and organisational factors for risk of infection.

Comparison

There were no restrictions for comparison groups of potentially eligible randomised control trials.

Studies with exclusive focus on the exposure to home parenteral nutrition or home infusion therapy were excluded. To our understanding, such studies in general included a sample of broader age range (sometimes even children). Also, even though these medical interventions take place at home, the primary focus of these studies cannot be considered as the practice of home healthcare delivery.

Outcomes

Studies with the following outcomes were considered: infection, infectious disease, communicable disease, and symptoms of infections (e.g. fever). Quantitative studies of adherence solely to infection prevention and control (IPC) protocols and practices were excluded as they reported on other outcomes.

Phenomena of interest

The qualitative component of this review considered studies that investigated the perceptions and understanding of infection risk as well as experiences of barriers to IPC practices in the home care setting. We considered studies containing client and home care staff perspectives, as well as studies of observations carried out during the delivery of home care. Studies with exclusive focus on home care agencies鈥� perspectives were excluded.

Context

The qualitative component of this review considered studies investigating the context of home help and/or home healthcare services provided by professionals in clients鈥� home setting.

Types of studies

The review considered quantitative (randomised trials and observational studies), qualitative, and mixed methods studies. Mixed methods studies were considered if data from the quantitative or qualitative components could be clearly extracted. There was no restriction in terms of publication date or language. Publications in foreign languages with no title or abstract in English or Swedish, yet potentially eligible, were translated.

Studies with no original data e.g., reviews and opinion pieces were excluded.

Methods approach

The systematic review has applied a convergent segregated approach as defined by the JBI Manual for evidence synthesis. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines have been used (Appendix 1). The study was prospectively registered in PROSPERO (CRD42021261159) and the study protocol has been published [18].

Search strategy

Search terms were developed in consultation with two university librarians at Karolinska Institutet in Stockholm, Sweden. A test search was carried out using the following databases: MEDLINE, Embase, Web of Science, CINAHL, and Sociological Abstracts. This generated 5,841 articles of which the 300 most recently published titles were screened by two researchers (AL, MB). The researchers identified about 30 relevant articles each, of which most overlapped. The articles were read in full and discussed with the librarians who refined the search strategy.

The search was carried out on 20th May 2022, using the search terms outlined in Appendix 2 and referred to home care, home healthcare and home help in various combinations with infection, infectious diseases and communicable diseases. In addition, reference lists of a previous systematic review [11] and the studies selected for critical appraisal were screened for additional studies. An updated search was carried out on 22nd February 2023, to include articles published since the initial search (Appendix 3).

Information sources

Databases searched were MEDLINE (Ovid), Embase, Web of Science (Clarivate) Sociological Abstracts (ProQuest) and CINAHL (EBSCO).

Study selection

Following the search, all identified records were collated and uploaded into Rayyan (rayyan.ai) and duplicates were removed. Titles and abstracts of the remaining records were then screened by two reviewers (MB, GM) for assessment against the inclusion criteria for the review. One doctoral thesis occurred in the searches and was substituted by related scientific papers by the same author. Studies that met the inclusion criteria were retrieved and their full texts were assessed in detail against the inclusion criteria by the two independent reviewers. Librarians assisted with the retrieving of articles not available online. Articles written in languages not spoken by the researchers were translated into English. Full text studies that did not meet the inclusion criteria were excluded and reasons for exclusion are provided in Appendix 4. Any disagreement that arose between the reviewers were resolved through discussion with a third researcher (AL). The same process was repeated for the second search. The titles generated in the second search were screened by researchers EG and AN, and discussed with AL.

Assessment of methodological quality

Qualitative process

Eligible studies were critically appraised by two independent reviewers (MB, AL) for methodological quality using the Critical Appraisal Skills Programme tool (CASP) (casp-uk.net). The tool was slightly adapted, changing Q10 鈥淗ow valuable is the research?鈥� into 鈥淲as the research valuable?鈥�, to allow for a three-level scoring system composed of 鈥淵es鈥�, 鈥淭o some extent鈥� and 鈥淣o鈥� for all questions. Any disagreements that arose between the reviewers were resolved through discussion. All qualitative articles that received one or more 鈥淣o鈥� ratings were further discussed, and it was decided that they should be excluded. The same procedure was repeated by researchers EG and AN for the qualitative studies identified in the updated search. Authors of eligible studies with unsupported findings were contacted. Upon request, authors of one of the eligible studies provided supplementary data files that were inaccessible through the web links referred to in the journal. Authors of some of the other eligible qualitative studies not providing quotations to support their findings were also contacted but they did not reply.

Quantitative process

Eligible quantitative studies were critically appraised, and their quality systematically assessed by reviewer CB and cross-checked by AL. Any disagreement that arose between the reviewers was resolved through discussion. The critical appraisal checklist tools for systematic reviews of aetiology and risk, available in JBI Manual for Evidence Synthesis, were used. Different critical appraisal checklists were used according to the study design of each quantitative study, with separate checklists for cohort studies and cross-sectional studies [19, 20]. No adaptations of the checklist tools were made, but where these were unclear in their questions, an agreed meaning was found between the two reviewers.

Data extraction

Quantitative and qualitative data were extracted from included studies using relevant JBI data extraction tools [15]. Only data that were relevant to the outcomes of this study were extracted.

For qualitative studies, data extracted included specific details about the aim, geographical setting, methods, participants, context, and the phenomenon of interest relevant to the review question. Findings with the studies鈥� corresponding illustrations were also extracted from the analytic text in the results sections. Relevant contradictory findings with regards to IPC practices were also extracted and included. All findings were assigned a level of credibility: 鈥淯nequivocal鈥�, 鈥淐redible鈥� or 鈥淣ot supported鈥�, in accordance with the ConQual approach [21].

For quantitative studies, data extracted included details about the study method/characteristics, the outcome of significance to the review, the data analysis methods including statistical technique, and the study results. This process followed the data extraction for Systematic reviews of aetiology and risk available in JBI Manual for Evidence Synthesis [15].

Meta-analysis of quantitative findings

Meta-analyses were carried out to determine an overall trend and conducted when the same risk factor was reported in three or more studies. Multiple meta-analyses were performed because of the heterogeneity in outcomes and risk factors across the studies, which did not allow for one overall meta-analysis. The meta-analyses were undertaken by combining study-specific odds ratios (ORs) and confidence intervals (CIs) using the DerSimonian and Laird鈥檚 random-effects model. Statistical analyses were performed using the STATA software (version 16.1; StataCorp LLC, College Station, TX, USA). The findings of the meta-analyses are presented as forest plots.

Data synthesis and integration

A convergent segregated approach to synthesis and integration was applied.

The qualitative synthesis of findings was carried out using a meta-aggregative approach. Only findings assessed as 鈥淯nequivocal鈥� or 鈥淐redible鈥� were included in the synthesis. Findings were grouped into categories based on similarity of meaning. Categories contained at least two findings and descriptions were created by consensus process between reviewers. Synthesised findings, containing two or more categories, were created based on similarity of meaning, in a consensus process between reviewers.

The quantitative synthesis of findings was presented in narrative form as suggested by JBI. Where possible, a statistical pooling of recurrent findings was conducted, resulting in meta-analyses to estimate a summary average effect of a specific risk factor.

The synthesised findings based on the quantitative and qualitative evidence were integrated using a configuration approach, whereby the evidence from each separate part was compared and linked to produce one line of argument [15]. Where such configuration was not possible, integration of findings was instead presented in a narrative form.

Methodological quality

Critical assessment of qualitative studies

Table 1 shows the rating of the sixteen eligible qualitative studies assessed for methodological quality using an adapted version of CASP providing 10 questions with answer options Yes, To some extent, and No. CASP was chosen due to its pragmatic approach and assessment of coherence between aim and methods including data collection and analysis choices, and transparency in the reporting of the choices made.

Table 1 Critical appraisal of eligible qualitative studies

Five studies [6, 22,23,24,25] received more than one No rating, indicating poor trustworthiness, and were therefore excluded following discussion. Details of the excluded studies can be found in Appendix 5.

Critical assessment of quantitative studies

The eighteen eligible quantitative studies were assessed for methodological quality using the critical appraisal checklist tools for Systematic reviews of aetiology and risk, available in JBI Manual for Evidence Synthesis [19, 20]. The checklists refer to certain study designs. The critical appraisal checklist tool for cohort studies was used for the six cohort studies [26,27,28,29,30,31], and the critical appraisal checklist tool was used for analytical cross-sectional studies for the remaining 12 studies [3, 5, 30, 32,33,34,35,36,37,38,39,40].

Cohort studies

The checklist tool for cohort studies consisted of 11 questions (Table 2). Each question was rated Yes, Unclear, No, or Not Applicable. No study was excluded due to methodological quality concerns. The potential influence of the studies with one or more Unclear or No ratings on the outcomes of the meta-analyses and overall conclusions, were discussed.

Table 2 Critical appraisal of eligible cohort studies

Cross-sectional studies

The checklist tool for cohort studies consisted of eight questions rated Yes, Unclear, No, or Not Applicable (Table 3). One study received two No and three Unclear ratings, and another study received three No and three Unclear, indicating poor quality. For this reason, these two studies were excluded following discussion. Details of the excluded studies are outlined in Appendix 5.

Table 3 Critical appraisal of eligible cross-sectional studies ¹

Results

Study inclusion

The Prisma flowchart is presented in Appendix 3. The initial search generated 17,984 records of which 80 studies were read in full text. The hand search of reference lists resulted in the identification of one additional eligible study. Excluded studies read in full text are listed in Appendix 4. A total of 17 quantitative studies and nine qualitative studies were critically appraised for methodological quality. This appraisal resulted in the exclusion of one quantitative and three qualitative studies (Appendix 5).

The updated search generated 1,500 records of which 69 were read in full. Sixty-one of these studies were excluded (Appendix 4). A total of seven qualitative studies and one quantitative study were critically appraised for methodological quality, which resulted in the exclusion of the quantitative study and two qualitative studies (Appendix 5).

Thus, out of 19,484 studies a total of 27 studies (16 quantitative studies and 11 qualitative studies) were included in the review.

Characteristics of included studies

Overviews of the characteristics of the included studies are presented in Appendix 6 (qualitative studies), Appendix 7 (cohort studies) and Appendix 8 (cross-sectional studies).

Qualitative studies

Setting and participants

Most qualitative studies explored infection prevention strategies and practices [7, 41,42,43], and perceived barriers and facilitators to IPC [42,43,44,45]. Whilst one study specifically examined the agency perspective [42], the remaining studies focused on the delivery of care in clients鈥� homes. All studies included interviews with home care staff, who were predominantly female. One study also included observations and focus groups [43]. Eight of the studies focused on home care during the Covid-19 pandemic [41, 44,45,46,47,48,49,50]. The studies were published between 2020 to 2022 and most of them were undertaken in Western societies; half of the studies had been conducted in the United States (U.S.).

Quantitative studies

Setting and participants

Five of the 16 quantitative studies used data from databases or registers [3, 5, 27, 36, 38]. The sample sizes ranged from 81 [43] to 156,408 patients [39]. Most of the studies reported the participants鈥� mean age, which ranged between 63 to 85 years (with the exception of two studies [26, 32]). The studies were published between 1995 and 2022. Most studies were conducted in Western societies including five from the U.S. [3, 5, 27, 34, 39].

Study-specific outcomes and tools

Among the sixteen quantitative studies, the most recurrent infections were urological-related of which urinary tract infection (UTI) was the most common infection [3, 5, 26,27,28,29, 34, 37, 38, 51], followed by respiratory-related [3, 5, 26, 29, 31, 32, 35, 38,39,40], and skin-related infections [3, 5, 26, 29, 38]. Two studies had fever events as the main outcome [29, 31] and three studies focused on IPC protocols and practises [34, 36, 39]. Five studies did not focus on a specific infectious disease [3, 26, 36, 38, 39].

The included studies used different sources as exposures. The Outcome and Assessment Information Set (OASIS) was used as source of data collection or secondary analysis in three of the听American studies [3, 5, 34]. The Finnish studies used a similar tool called Resident Assessment Instrument (RAI) [37, 38]. The OASIS and RAI are pre-defined standardised assessment tools administered to home care patients (for USA) [52], or patients receiving long term care (in Finland) [53], at admission and repeated when necessary. Data from these tools were combined with national registers [37], hospital and national discharge records [38] or other databases [34] for the addition of complementary data on patients鈥� characteristics, diagnoses, or agency information relevant to the study. One study used data from specific home care agency databases [28] while three studies were based on self-administered survey data [26, 32, 36]. Two studies measured exposure through direct recordings by the study participants [29, 31]. Other sources of data were clinical records [32, 35, 51], outpatient department records, home visit records, discharge charts [35] and laboratory data [28, 35].

For outcome data, medical records were the most frequent data source [29, 32, 35], where diagnoses were made by qualified physicians with [28, 34, 35, 40] or without [32] referring to the International Classification of Diseases (ICD) codes. Results from laboratory tests were used to support the infection diagnosis in two studies [32, 35] whereas two studies used self-reported or medically diagnosed data on infections regardless of laboratory tests [29, 36]. The remaining studies did not mention laboratory test results. For some studies, the outcome was measured using self-administered surveys completed by a qualified medical professional [26, 31, 36] or by the patients themselves [26].

Findings of the review

The findings present quantitative and qualitative data separately.

Quantitative evidence

Risk factors reported in the quantitative studies referred to individual, medical, social, behavioural, environmental and organisational factors.

Individual

Female sex [5, 27, 37], older age [5] and being of white ethnicity [5, 27] were found to be factors significantly associated with greater odds of UTI and UTI-related hospitalizations in home care patients. Male sex [5, 32, 35], older age [5, 32, 40] and white ethnicity [5] were found to have a significant association with the development of respiratory infections and related hospitalization. Younger age and male sex were found to be individual risk factors related to significantly higher odds of developing a wound-site infection [5]. Studies that did not focus on any specific infections, found that infection-related outcomes were more likely to occur in individuals of male sex, white ethnicity and older age [3]. Older age was also associated with fever events considered a symptom of infection [31].

Medical

The majority of the medical risk factors were specific infectious-related and associated to medical conditions, practices, or devices. Among the included studies, the use and management of urinary catheter was found to be the major risk factor for UTI in home care patients. Presence of a urinary catheter was found statistically associated with UTI events and related hospitalizations in two cohort studies [27, 28] and two cross-sectional [5, 40]. A significant association was also found between time to UTI and frequency of intervals in catheter change, the number of nurses changing the catheter [51], history of urinary catheter use, and history of UTI treatment prospectively [27] and cross-sectionally [5, 37]. Besides, urinary incontinence was found to be associated with greater odds of developing UTI in three cross-sectional studies [5, 37, 38] yet shown to be a protective factor in a retrospective cohort study [37].

Hospitalization due to UTI was associated with the presence of comorbidities [5, 28, 37], hyper-polypharmacy [28], cognitive dysfunction [5, 37], severe difficulties in undertaking basic and instrumental activities of daily living (ADLs) [3, 27, 37], being chair fast [5], and having a history of hospitalization [5, 37]. Based on a retrospective cohort study, UTI was one of the most common diagnose causing fever events [29]. In one retrospective cohort study [27] a few medical factors were associated with a decrease in the odds of infection: a mild ADL dependence, prior stay in an acute care hospital, and presence of urinary incontinence.

Nasogastric tube use [35], recurrent aspiration [40] and receiving respiratory treatments at home [5, 40] were associated with greater odds of having a respiratory infection. The presence of chronic respiratory diseases [35, 40] or comorbidities [5, 40] were also found to be associated with respiratory infection.

Wound infections and decubitus ulcers were the most common skin-related infections reported in the included studies. Skin related medical conditions were found to be significantly associated with an increased risk of wound infections or related hospitalisation. In a cross-sectional study [5], having a stasis ulcer, surgical wounds, skin lesions or open wounds, or severe unhealed pressure ulcers, was associated with developing a wound-site infection. Moreover, limited mobility, progressive conditions, multiple hospitalisations, difficulties managing injectable medications and needing help with changing wound dressing were also associated with infection risk [5]. Fever events were commonly associated with skin and soft tissue infections [29].

Two cross-sectional [3, 38] and one prospective cohort study [26] studied underlying causes of infection-related hospitalizations and risk factors for infectious-related outcomes in home healthcare patients. In their cross-sectional study [3], home healthcare was associated with greater risks of receiving emergency care due to infectious diseases (respiratory, wound, urinary tract, and intravenous catheter-related infections). The other cross-sectional study found that functional and cognitive impairment were associated with higher odds of being hospitalised with an infectious disease diagnosis, along with chronic skin ulcers and daily urinary incontinence [38].

Medically related factors associated with infectious disease were Parkinson鈥檚 disease, dermatosis not caused by pressure ulcers, chronic respiratory failure and inability to perform oral self-care [26]. Two studies that assessed fever events and their risk factors found associations with limited mobility, needing a respiratory device, presence of comorbidities, having cognitive impairment [31] or having comprehensive care needs [29, 31].

Social

Among the included studies, the social factors associated with risk of infections in the older population with home care were mainly related to the availability and help from caregivers. Patients hospitalised due to infections were generally more likely to need help with ambulation with management of medical equipment [3], and with meal preparation [5] as well as need of supervision by a caregiver for their safety [27]. The presence of a caregiver to assist with ADLs was associated with lower odds of being hospitalised for UTI in one retrospective cohort study [27], while having no caregivers to assist was associated with more than doubled odds of being hospitalised for an infection in two cross-sectional studies [3, 5]. Being eligible for Medicaid insurance was found to be associated with hospitalisation for UTI [27] and other infections [3]. Given that Medicaid is offered to people below the federal poverty level [27], Osakwe and colleagues used this information as proxy for socio-economic status.

Behavioural

Smoking was the personal behaviour found to be statistically associated with increased odds of hospitalisation due to respiratory infections [5]. Memory deficit, defined as neuro/emotional/behavioural status, was associated with infection-related outcomes [3].

Environmental

Home care patients鈥� living situation was the most common environmental factor assessed in the included studies. Living in a shared accommodation or congregate setting were found to be associated with increased odds of developing a respiratory infection [5, 32], UTI or wound infection [5]. One study also found increased odds of developing acute lower respiratory infection among patients who reported a different waste collecting system from the public system [32].

Organisational

Three cross-sectional studies [34, 36, 39] specifically examined IPC policies and practices. Harrison et al. showed that policies on replacement of indwelling catheters at fixed intervals, and policies on emptying the drainage bag, lowered the probability of hospital transfer due to UTI in the home care population [34]. Shang and colleagues examined predicted rates of hospital transfers due to respiratory infections among home care patients in different agencies and found that influenza vaccination requirements for home care staff would reduce hospitalizations among patients in for-profit home care agencies, but not in non-profit agencies. If all home care agencies adopted influenza vaccination requirements for staff (called Scenario 2), there would be a 15.75% reduction in the rate of hospital transfer due to respiratory infection in for-profit agencies, compared with the status quo (Scenario 1) (P <鈥�0.05) [39]. The third cross-sectional study found that the presence of a committee for IPC practices and providing staff training on infection prevention were associated with increased incidence of reported infections in agencies with such policies and practices in place [36]. Finally, having a caregiver in need of additional training on adherence to IPC practices was found to be associated with significantly higher risk of developing infections among home health patients, as evidenced by two cross-sectional studies [3, 5].

Meta-analyses

The results from the meta-analyses conducted based on studies that individuated the same risk factor are reported below. These risk factors were urinary incontinence, urinary catheter use and limited mobility. Outcomes were infection development of UTI or urinary incontinence.

Urinary incontinence and development of UTI (or other infections)

Four studies individuated urinary incontinence as a factor related to UTI [5, 27, 37, 38]. The computation of the meta-analysis on urinary incontinence as risk of UTI showed odds ratio (OR) of 1.17 (95%CI [0.85鈥�1.61]) and 93.13% heterogeneity (Fig. 1a).

Fig. 1

a听Pooled ORs of urinary incontinence on the development of UTI.听b听Pooled ORs of urinary incontinence on the development of infection

When adding research not restricted to UTI-infections [3], the OR changed to 1.23 (95%CI [0.92鈥�1.63]), and heterogeneity to 93.29% (Fig. 1b).

Urinary catheter use and development of infection outcomes

Urinary catheter use was a recurrent factor associated with infection outcomes. The meta-analysis pooling the ORs of these studies resulted in an overall OR of 3.97 (95%CI [2.56鈥�6.15], $\tau$ ²鈥�= 0.23) with 92.33% heterogeneity (Fig. 2).

Fig. 2

Pooled ORs of urinary catheter use and risk of infection

Limited mobility and risk of infection

Limited mobility and risk of infections resulted in an overall estimated OR of 1.49 (95%CI [1.31鈥�1.68]) with 73.03% heterogeneity (Fig. 3).

Fig. 3

Pooled analysis of ORs of limited mobility and risk of infection

Qualitative evidence: meta-aggregation

In this section, older adults are referred to as clients as this word was commonly used in the included qualitative studies.

A total of 125 findings from eleven qualitative studies were extracted and grouped into 17 categories based on similarity of meaning. There were 85 unequivocal findings, 26 credible findings, and 14 unsupported findings. The findings and illustrations related to each study can be found in Appendix 9. The categories were aggregated into five synthesised findings presented below. Unsupported findings were not used in the meta-aggregation.

Synthesised finding 1鈥擟lients'characteristics, health state and practices determine their exposure and susceptibility to risk factors of infections. Client attitudes and behaviours are central to IPC (Table 4).

Table 4 Synthesised finding 1: Clients'characteristics, health state and practices determine their exposure and susceptibility to risk factors of infections. Client attitudes and behaviors are central to infection prevention and control

The individual level is represented in this synthesised finding, where the home care client forms the main focus of infection risk factors. It is made up of three categories, containing 15 unequivocal findings and four credible findings from four studies [7, 44, 49, 50].

Category 1. Risk factors related to specific infections, health conditions and client characteristics. Clients are believed to be at risk of infection depending on their history of infections, diagnosis and current health state, such as having a catheter. Each infection is observed to be related to specific risk factors, as illustrated below:

鈥淏ecause somebody with COPD (chronic obstructive pulmonary disease) is obviously at higher risk of developing a lung infection, right? But not necessarily wound infection. So it depends what type of infection you're talking about. Certain people are more at risk for developing respiratory infection because of smoking and I think smoking always puts you at higher risk for infection, in general.鈥� [7] (Appendix 9, Q5).

Category 2. Infection as a result of multiple health issues and interaction with environment. Infections are considered to be a result of several risk factors affecting a client's health state, 鈥渟uch as, older age, diabetes, dehydration or having inadequate nutrition鈥�. [7] (Appendix 9, Q6 & Q9).

Furthermore, infections are understood to be caused by interacting risk factors on the individual level, in the home environment and the reception of care. 鈥淪o a lot of times elderly people that are incontinent, they tend to get urinary tract infections pretty easily. So probably an older person who's incontinent, who's in bed, who maybe doesn't have a lot of help.鈥� [7] (Appendix 9, Q8 & Q9).

Category 3. Clients'lack of understanding of infection risk and low adherence to infection prevention and control practices pose risk of infections. Clients'knowledge, attitudes and behaviour largely influence the risk of infection in home care where the staff has less control over the delivery of care and care environment. Clients'adherence to treatments, hygiene and health practices, as well as attitudes and capacity to maintain a clean environment are all believed to be key aspects of IPC. Illustrations included the following:

鈥淭hen you鈥檙e looking at their attitude towards what you鈥檙e talking about when you ask them to wash their hands. [鈥�] Sometimes they think that one time is enough, and that鈥檚 it. And then how receptive they are to you just teaching them. [鈥�] Sometimes they鈥檒l say, 鈥楿hhuh,鈥� but you know that it鈥檚 just going in one ear, then going out the other.鈥� [7] (Appendix 9, Q29).

"The biggest issue is trying to get him to remember to wear his mask and not to hug people or shake their hands? He鈥檚 just got his hand out there鈥� [49] (Appendix 9, p.187).

Synthesised finding 2. The state of the home environment, the interactions between the client and other members of the household, as well as caretakers'participation in healthcare delivery may all represent risk factors for infection. In the home environment the home care provider exercises limited influence over IPC (Table 5).

Table 5 Synthesised finding 2: The state of the home environment, the interactions between the client and other members of the household, as well as caretakers'participation in health care delivery may all represent risk factors for infection. In the home environment the home health care provider exercises limited influence over infection prevention and control

The household context where the client finds him-/herself and where care is delivered forms another level of infection risks described in this synthesised finding. It is made up of four categories containing 15 unequivocal and 3 credible findings from 4 studies [7, 42, 43, 45].

Category 4. Exposure to environmental risks in the home. Home environments are perceived to present a number of infection risks such as dirt, clutter and hoarding, waste, pest infestations, inadequate ventilation, poor lightning, and inadequate access to clean water. These factors affect the clients'health state as well as the delivery of safe healthcare practices, as illustrated by the example extract below:

鈥淚t鈥檚 hard in home health. Sometimes, I鈥檓 just at a loss. How do you make this happen when these people are living in what they鈥檙e living in? Because a person can look like they got it all together on the outside, and then you get into their home and hoarding situations, infestations of animals. [鈥�]鈥� [42] (Appendix 9, p.1786).

Category 5. Exposure to frequent personal interactions and unsafe caregiving by family members. The home environment often implies close contacts with family and visitors where family members commonly also deliver care to the client. Their knowledge, attitudes and adherence to safe care practices and infection risk prevention therefore influence risk of infection in home care.

"During the delivery of care, more and more visitors arrive. First one adult woman. Then one adult female and adult male with toddlers (approx. 14 months). Then another adult woman with a baby. Toddlers run around the room, sneeze three times successively on chair, wheelchair and side table without covering nose or mouth. [鈥�]鈥� [43] (Appendix 9, Q3).

Specifically, during the Covid-19 pandemic, increased safety measures among family members and family caregivers were observed. Concern also arose regarding the infection risk posed by several different staff working for the same client.

鈥淭here are 5 of us that work with her鈥� What happens if we all get sick?鈥� [45] (Appendix 9, p.1456).

Category 6. Unsafe storage and disposal of material and equipment in the home. The home environment is considered to represent increased infection risk when it is impossible to safely store healthcare supplies due to lack of space or when unsafe waste disposal is carried out.

鈥淪ticking out of the needle container are plastic blister bags that had medication in them. Before the needles can be thrown away, they must first be removed. The client says: 鈥業鈥檒l just empty it鈥�. The nurse asks: 鈥楬ow will you do that?鈥� The client responds: 鈥業鈥檒l just chuck it in the bin鈥�.鈥� [46] (Appendix 9, p.5).

Category 7. Contact with animals in the home. Infection risk is increased by the presence of pets and other animals. They may also hinder staff in the delivery of safe healthcare practices.

鈥淵ou can be absolutely aseptic the whole time you鈥檙e in there. [But] if there鈥檚 dog poop 10 feet away from you when you鈥檙e doing wound care, that鈥檚 a problem. Some of the home environments are not appropriate for certain kinds of patients. There鈥檚 no way I can get them healed in that environment.鈥� [42] (Appendix 9, p.1786).

Synthesised finding 3. Home healthcare staff form a central role in IPC during service delivery and procedures. Staff noncompliance to disinfection and safe practices pose direct infection risks for the client (Table 6).

Table 6 Synthesised finding 3: Home health care staff form a central role in infection prevention and control during service delivery and procedures. Staff noncompliance to disinfection and safe practices pose direct infection risks for the client

The influence of home healthcare staff care delivery practices on infection risk is described in this finding. It describes staff attitudes and behaviours as well as the challenges involved in safe care delivery in the home context. The synthesised finding is made up of three categories containing 16 unequivocal findings and four credible findings from six studies.

Category 8. Staffs'nonadherence or irregular performance of hygiene, disinfection and safe procedures. Infection risks occur when staff do not perform adequate disinfection procedures, when hand hygiene and use of safety equipment is poor or inexistent. Illustrations included the following:

"Initially, the nurse does not wear an apron while providing care. The client herself points out to the nurse that when the research is about the prevention of infections, the nurse should be wearing an apron.[鈥�]鈥� [43] (Appendix 9, Q6).

Related to the Covid-19 pandemic, staffs'increased disinfection and infection prevention practices were observed.

"I take precautions. I wear my mask. I keep sanitizer on me. I wash my hands constantly but I still give them the attention they need, you know, it鈥檚 very hands-on." [50] (Appendix 9, p. 1834).

Category 9. Staffs'irregular use of workwear, electronical devices and personal equipment. Contrary to a hospital environment, home healthcare staff are observed working in personal clothing which may be inadequately sterilised and as such represent infection risks. The frequent irregular use of bags and electronic equipment during the workday of home visits to several clients is further observed as a risk factor.

"That computer, I take it out of that bag, it sits on the table at the client, and it goes back in my bag. And it ends up on the next client's table." [43] (Appendix 9, Q1).

Category 10. Staff experiences of stress and distractions during care. Stress caused by high workload and time pressure, as well as distractions in the form of frequent telephone calls during care delivery were perceived to increase the risk of performing unsafe procedures.

鈥淭he nurses I can say, they are just so stressed up. They have too many patients to see, and they want to make sure that they covered everything. Now they are shortcutting, the tendency of shortcutting is you are giving a high percentage of committing mistakes.鈥� [42] (Appendix 9, p.1784).

Synthesised finding 4. Home care agencies hold key responsibilities in providing staff with sufficient and relevant equipment, information, guidelines, training and support. Low agency priority of these aspects including lack of organised provision of necessary equipment to family care givers represents considerable barriers to safe delivery of home care, while agencies鈥� investments into coordinated IPC efforts are perceived as important positive protective measures (Table 7).

Table 7 Synthesised finding 4: Home health care agencies hold key responsibilities in providing staff with sufficient and relevant equipment, information, guidelines, training, and support. Low agency priority of these aspects represents considerable barriers to safe delivery of home health care, while agencies investments into coordinated infection prevention and control efforts are perceived as important positive measures

This synthesised finding is focused on the influence on IPC from the agency level. The findings describe barriers as well as facilitators related to home care agencies policies, support procedures and investments in infection prevention and control. This finding also included the lack of organised measures to provide family caregivers with necessary equipment for infection prevention. It is made up of five categories containing 30 unequivocal findings and 12 credible findings from eight studies.

Category 11. Lack of access to equipment. Agencies鈥� failure to ensure adequate hand hygiene products and personal protective equipment (PPE) in clients'听homes impede IPC practices. Lack of protective equipment was especially highlighted in relation to the Covid-19 pandemic, however, not in all home care agencies. Family caregivers also struggled to get PPE during the pandemic.

鈥淚鈥檓 worried about getting infected because I don鈥檛 have the right equipment. The agency has not really been providing for their workers, at all.鈥� [45] (Appendix 9, p.1457).

[The client鈥檚 agency]"is giving me like, five masks, and that鈥檚 it. So, I don鈥檛 even think about it, just bring my own bag, I bring my own gloves, my mask, everything.鈥� [50] (Appendix 9, p. 1834).

Category 12. Lack of access to clear information, guidelines and communication. Staff receiving incomplete information on clients'听health state, protocols or guidelines is perceived to be an impediment for safe care delivery in the home care environment. Conflicting messages further contribute to staff confusion and doubt around IPC. Lack of agency support to non-English speaking staff is another barrier to communication and information access.

鈥淚s it really necessary for us to do this hand hygiene 鈥揺ven within the client鈥檚 house 鈥搊r is it unnecessary in the home environment? Because that is all based on research done in the hospital.鈥� [7] (Appendix 9, p.5).

Category 13. Restricted access to Covid-19 testing. During the Covid-19 pandemic, home health care agencies'听incapacity to provide testing and access to information on Covid status of clients and staff hindered effective risk control. However, daily risk screening messages from agencies were recognized by staff.

鈥淚 am with a patient, and she has all this different aides coming in and out. No one tells you if a home health aide had COVID, I do not know if my patient had it鈥︹�� [44] (Appendix 9, p.1365).

Category 14. Limited opportunities provided for staff training and knowledge transfer. The nature of home care where staff often work on their own in clients'听dwellings, represents a challenge to knowledge transfers. Lack of agency investment in training is perceived to hinder the implementation of infection prevention.

鈥淚 actually had no previous experience in home care or formally in infection prevention when I came to this role, which is a challenge. When I look at what resources are out there. [it is] very much geared toward the inpatient world.鈥� [42] (Appendix 9, p.1786).

On the contrary, when agencies prioritise education, it is recognised as an important driver of IPC.

鈥淚n home health, [we have] always used infection control. We were all prepared, and we were always prepared in teaching patients about infection control in their own homes [鈥�]. In the community and in the patients鈥� home we were good at that, and we were prepared for a pandemic.鈥� [41] (Appendix 9, p.4).

Category 15. Importance of home care agency IPC preparedness and coordination. Agencies'听lack of IPC plans and guidelines represent barriers to safe home health care, while cases where agencies have prioritised coordination, teamwork and evaluation are recognized as positive efforts towards IPC.

"We actually are really working just individually right with that team and the team manager to really look at their own outcomes and trying to help them to improve.鈥� [42] (Appendix 9, p.1788).

Synthesised finding 5. Staffs'听beliefs that risk of infection to clients and themselves is inevitable due to the nature and structure of home care services (Table 8).

Table 8 Synthesised finding 5: Staffs'听beliefs that risk of infection to clients and themselves is inevitable due to the nature and structure of home health care services

This synthesised finding presents staffs鈥� reflections on their limited capacity to ensure IPC as a result of the structure of home care services, and the restricted control they exercise over the work conditions. It is made up of two categories containing nine unequivocal findings and three credible findings from seven studies.

Category 16. The structure and nature of home healthcare, in itself, imply infection risk and challenges to IPC. Performing healthcare services in the home environment is perceived to involve lower levels of control over the patient, procedures and environment, than in a hospital setting. Staff working alone or with limited supplies reduces the capacity to avoid situations of risk exposure.

鈥淲ell, it鈥檚 not the hospital where it鈥檚 a controlled environment. You鈥檙e going into patient鈥檚 homes that sometimes aren鈥檛 the cleanest. You just gotta do the best you can and try to be as clean and prevent infections as you can in the home. You鈥檙e working with what you have.鈥� [42] (Appendix 9, p.1784).

Category 17. Staffs'听own exposure to risk is believed to be part of the job. Delivering home care is perceived to involve unavoidable staff exposure to infection risk, especially during the COVID-19 pandemic. This risk is weighed against a call of duty or a need for employment.

鈥淗e needs to stay inside the house, so he tells me, 鈥業 need you to go there, go here.鈥� I really don鈥檛 want to, but I can鈥檛 say no. I鈥檓 the aide; I鈥檓 supposed to do this.鈥� [45] (Appendix 9, p.1456).

鈥淵ou have to contribute certain hours to get benefits... I have to go out there because I have bills to pay.鈥� [45] (Appendix 9, p.1457).

"In November I got infected, but I鈥檓 not really worried, it鈥檚 more natural because of my work, I can鈥檛 do much more.鈥� [46] (Appendix 9, p.9).

Integration of quantitative and qualitative evidence

Congruence between synthesised qualitative findings and quantitative findings

Numerous synthesised qualitative findings were in congruence with several of the quantitative findings, allowing for a configuration of evidence as outlined below.

The first synthesised qualitative finding, known as category 1, and several findings from the quantitative studies showed that patients鈥� individual characteristics, behaviours and health status determine their infection risk. More specifically, category 1 referred to individuals鈥� history of infections, specific medical conditions and the need of medical devices increased their risk of infections. Similarly, findings from the cross-sectional and cohort studies showed increased infection risks for patients with nasogastric tubes, history of hospitalizations and/or previous infections, along with a number of specific medical conditions. The meta-analysis further emphasized this congruity by showing increased risk of infection among patients with urinary catheters, a risk factor also identified in the qualitative studies.

Category 2 showed that the presence of co-morbidities increased infection risks. This provides congruent evidence that multiple medical conditions constitute an increased risk for infections among older adults receiving home care. Moreover, category 2 revealed that interaction between patients鈥� multiple health conditions, such as decreased cognitive capacity and urinary incontinence, and availability of help from caregivers, increased their infection risk. This is in line with the quantitative findings showing that environmental and social risk factors, such as the presence and availability of caregivers, impact infection risks. Finally, category 3 showed that patients鈥� understanding and ability to adhere to infection control practices were key to infection control. This was supported in the meta-analyses indicating that cognitive impairment and restricted mobility increased the risk of infectious disease.

The second synthesised finding, along with quantitative evidence from cohort and cross-sectional studies, showed that interactions between patients and other members of the household posed infection risks. Particularly category 5 revealed that the practices of family members, who often also served as caregivers, were crucial for infection control, in congruence with quantitative evidence that organisational factors such as caregivers鈥� need for training constituted increased risk of infection for home care patients. Category 6 brought forth the risks of infection associated with unsafe disposal of materials in the home, a finding in line with a cross-sectional study that showed increased respiratory infection risks in housing settings where the waste disposal system was not connected to the public network.

The third synthesised finding identified that home care staff鈥檚 behaviours affect infections risks. Category 8 and 9 showed that staff compliance with infection prevention practices, such as disinfection routines and safe workwear usage were linked to infection risks. This is in accordance with findings from the quantitative studies showing that staff training on infection prevention practices, and staff鈥檚 need of training, were associated with increased infection risks.

Synthesised finding 4 showed that agencies鈥� priorities to provide home care staff with necessary guidelines and support were essential for infection control. For instance, category 12 showed that lack of access to clear information and guidelines impeded infection prevention, which is in congruence with quantitative findings showing that the presence and implementation of agency policies that promote best practices for care of patients with urinary catheters decreased hospitalizations for UTI. Moreover, category 14 identified the lack of opportunities for staff training as an infection risk, supported by quantitative findings suggesting that organisational factors such as staff in need of further training increased the infection risks among home care patients.

In accordance with the JBI Manual, we have summarised aspects of the qualitative findings that were not examined in the quantitative studies, and vice versa. These are available in Appendix 10.

Discussion

Using a convergent, segregated approach, this mixed methods systematic review aimed to identify risk factors for infection in older adults who receive home care. Main results show that older adults in home care are at risk for infection due to a range of individual, medical, environmental, and organisational factors. Some of these factors include being in need of medical devices, having limited mobility, and having a history of previous infections and comorbidities. Several infection risk factors are of a structural and organisational nature, such as staff in need of training, lack of access to equipment, working under unclear guidelines or high stress and heavy workload. Moreover, the home environment itself carries risks, for example by being cluttered or by posing difficulties in terms of waste management and lack of sterile surfaces. Finally, the review showed that personal behaviours and attitudes towards infection prevention practices are crucial for infection control, and that these factors interplay with organisational level obstacles to optimal infection prevention practices.

The results of the meta-analysis, which showed a fourfold greater risk of infection due to urinary catheter use (OR 3.97, 95%CI 2.56鈥�6.15) in older adults with home care, adds to previous research on infection risk of urinary catheter use in patients in general. Whilst this finding refers to studies on hospital-acquired infections among older adults with home care, it stresses the importance of infection prevention in their homes to avoid additional co-occurring infections that may cause a rapid decline in the health of this already vulnerable group of older adults [54]. The review further adds to existing research on limited mobility as a potential predictor for risk infection [55] by demonstrating a 1.5 time (OR 1.49, 95%CI 1.31鈥�1.68) greater risk of infection among older adults with limited mobility. Body movement reduces the risk of long periods of urinary stasis, known for increasing the risk of UTI [56]. Urinary flow also reduces bacteria in the bladder [57], highlighting the vulnerability of inactive bedridden or chair-fast home health care patients. Additionally, mobility training has also been shown to reduce incontinence in older women [58]. In the meta-analysis, urinary incontinence was not statistically significantly associated with UTI. The lack of association remained also when adding a study not restricted to UTI but also including respiratory, wound, and intravenous catheter-related infections. The finding is inconsistent with previous research. For instance, a study of healthy postmenopausal women reported that urinary incontinence was more common in women who experience UTIs than those who do not [59]. It is possible that the lack of association in the meta-analysis was driven by a single study, in which the study population sample consisted of groups more likely to be hospitalised for reasons other than UTI [27].

Whilst informal caregivers play an important role in the care of patients who live in their own homes, the findings of this review indicate that inadequately trained caregivers may put the patients at risk of infection. In several qualitative studies [7, 42, 43, 45], unsafe caregiving by family members posed infection risks to home care patients, and similarly, the quantitative findings showed that caregivers in need of training increased infection risks [3, 5]. However, little detail about the type of training required was provided in the included studies. Training on use of devices, administration of medications, recognizing symptoms, or general infection prevention measures were examples suggested. Future well-planned and evaluated interventions that draw upon evidence from both quantitative and qualitative research are needed as evidence of effective interventions is lacking. A systematic review of interventions to improve hand hygiene demonstrated few improvements and concluded that performance feedback, education, and informative signs encouraging hand washing may work, but reported certainty of evidence to be low [60].

Employing an IPC nurse trained in all infection prevention activities and thus able to ensure supervision, support and implementation of routines has been successful [43] but may not be feasible at larger scale. Furthermore, delivery of certain kinds of care should be restricted to registered medical professionals while training of informal caregivers ought to focus on aspects of caregiving that medical and other professionals do not usually provide. Though collaboration between health and social care is essential, having professionals focusing on their area of expertise benefits both the older person who safely receives the best care, as well as the professional who then carries out work that they are confident in undertaking. This should not be confused with some professionals鈥� desire to uphold their role and status to ensure that their professional perspective has a dominant position [61] but allow for different professions to focus on what they are specialised in and to build collaborations based on differences in expertise.

The findings further indicated that structural level factors, such as agency policies and guidelines, play an important role for infection risk and control [34, 36, 39, 41, 43,44,45,46,47,48]. For example, the existence of policies on catheter management was shown to reduce infection risks [34], whereas lack of clear guidelines was considered an obstacle to safe care delivery by home care staff [43, 45, 46, 48]. Agencies鈥� reporting on infections should however be carefully interpreted as the presence of an IPC committee in agencies may increase the number of infections reported [36]. This might be due to agencies with IPC committees having more comprehensive structures for reporting infections, and consequently, better at infection surveillance, compared to agencies without such committee.

Findings of the current review furthermore demonstrate that behavioural and environmental factors are important for infection control in the home care setting on their own yet intertwined with organisational features. Thus, care providers鈥� practices, such as adherence to hygiene routines, are crucial, but so are the structures that inform, impede, or facilitate such practices. Similarly, the broad approach of the current systematic review allow for reporting on both knowledge and attitudes [6] as well as structural features, which may be equally important for infection control in the home setting [41, 42, 46]. Indeed, knowledge on how to perform infection control procedures correctly is not enough if there is a lack of access to necessary protective equipment and testing, or if working conditions are marked by high stress, heavy workload, and lack of materials. Besides, there are national cultural dimensions influencing infection control behaviour shaped by, for example, local outbreaks and vaccination rates, that interplay with governmental initiatives and risk management strategies [62]. Consequently, the interaction of different level risk factors must be considered for a rigorous understanding of IPC in the home environment.

Study limitations

Study strengths include the broad range of risk factors for infection explored and the inclusion of both qualitative and quantitative studies. The search strategy was not restricted to a certain language, geographical area, or time period. Search terms were further developed together with two well-experienced librarians and applied to five different databases. PRISMA and JBI guidelines were used to ensure a systematic and transparent approach, and two reviewers independently assessed the studies.

Study limitations include the lack of inclusion of grey literature, possibly resulting in relevant studies not being considered. Most of the included studies were from high income countries and only one study from the global south (Brazil), limiting the generalizability of the findings. The inclusion of cross-sectional studies restricted the possibility to determine the direction of causality of the observed relationships in the meta-analyses. Though the prospective studies shed light on the direction of causality, combining results from cross-sectional and prospective study designs restricts conclusions on causality and is associated with risks of bias. Additionally, exclusion of studies following quality assessment was not determined a priori but following discussion. Further, the review process took longer time than expected and an updated search was undertaken. Furthermore, many of the included quantitative studies used self-reported rather than objectively assessed data for exposure and outcome measures. Lastly, we initially intended to apply an age limit of 65 years, however, several studies did not report on their participants鈥� exact age. Therefore, it was decided to include studies of which the population receiving home care was reported to be of older age.

Policy and research implications

Recommendations for practice

The integration of qualitative and quantitative findings has several implications for practice and policy. The present review shows that, while staff behaviour is crucial for infection control, agencies must prioritize infection control co-ordination and preparedness, including supply of materials and staff training opportunities to enable optimal IPC practices among care-providers [3, 5, 41, 42, 46, 48]. As reported in the included qualitative studies conducted during or post the Covid-19 pandemic, organisational aspects of infection control are key for home care staff to be able to adhere to safe infection prevention practices and need to be addressed in future policies.

Regarding medical and individual risk factors, urinary catheter use emerged as a common risk factor for UTI and other infections. Given the infection risks associated with catheter use in combination with high demand in the older age home care population, best practice for catheter use is crucial for optimal infection control and prevention. Guidelines for catheter use in surgery recommend using urinary catheters only when needed and removing them as soon as possible after the operation [63]. Similarly, The Royal College of Nurses recommends early removal of catheter where possible [64]. Educating healthcare staff on prevention of urinary tract infection has been shown to reduce catheterization time. Educating healthcare staff about the latest guidelines on how to prevent catheter associated UTIs was further deemed vital for infection control in a recent systematic review [65]. The present systematic review also identified a need for and benefits of educating staff on prevention control. These findings support existing research and guidelines and adds to policy recommendations by showing that education and training on prevention control need to also include informal caregivers. Further, findings also indicate that the home environment itself poses several risk factors for infections, for instance in terms of clutter and the presence of pets. Consequently, home healthcare and home help agencies need to take these barriers to infection control into consideration when training and educating staff and informal caregivers. Cost-effectiveness evaluations may further facilitate resource allocation and healthcare planning. While educating and training staff often are associated with costs, under-provision of training, equipment, and environmental modifications might come at a cost, too. Health economic analyses could provide important input on the optimization of home care delivery. Moreover, to improve detection of UTIs, diagnostic criteria may need to be modified, since UTIs in later life are often asymptomatic and unspecific, and thus difficult to discover [28, 66, 67]. Detection and infection surveillance methods that are adapted and adjusted to the often asymptomatic infections that occur in older adults have the potential to reduce the risk of delayed diagnosis and suboptimal treatment.

Recommendations for further research

Further research on this subject area, applying both specific and broader perspectives, is needed. Specific perspectives to be studied include individual aspects that influence risk of infection such as the role of socio-economic background, ethnicity and geographical area, and the influence of medications on infection risk. This should furthermore be investigated using objective and longitudinal data to better understand the causal relationships between various risk factors and infections among older adults receiving home care. It also has the potential to shed light on possible health disparities within this population. Qualitative studies of older adults鈥� perceptions and experiences with infection risks and control in the home care setting would further enrich the understanding of these individual aspects.

A broader perspective is needed to explore potential interactions between the identified structural and organisational factors with medical and individual risk factors in relation to the risks of working in the home environment. This may include innovative methodologies, interdisciplinary collaborations, and novel interventions. Studying how these separate yet interconnected dimensions of risks may reinforce each other, exposing vulnerable older adults to amplified risks, could inform policies and practices aimed at mitigating such effects.