Introduction
Patients may be involved in disease transmission in healthcare settings Reference Wolfensberger, Mang and Gibson1,Reference Blanco, O’Hara and Harris2 when they transfer pathogens to the environment, spread pathogens to other patients, are involved in cross-contamination with healthcare workers, and when infection is spread to other body sites from endogenous sources. Reference Landers, Abusalem, Coty and Bingham3 In this chain of transmission, patients’ hands play an important role. 4–Reference Mody, Foxman and Bradley8
The importance of hand hygiene in healthcare settings has been known since the mid-1800s 4 , but for the majority of its documented history, the focus of hand hygiene has primarily been on the healthcare workers and not necessarily the patients. Reference Mody, Washer and Kaye5 Recently, however, more focus has been placed on the patient and how adequate patient hand hygiene could be effective in reducing the spread of pathogens. Reference Landers, Abusalem, Coty and Bingham3,Reference Mody, Washer and Kaye5–Reference Cassone, Zhu and Mantey10 Engaging patients in their hand hygiene also improves rates of healthcare worker hand hygiene. Reference Fox, Wavra and Drake11 Therefore implementation of such strategies has great potential in improving quality of care in the hospital setting.
Prior studies in nursing homes that care for older frail populations have shown that patients with certain functional dependencies (ie, bathing, ambulation, or toileting) have a higher risk of hand colonization by multi-drug resistant organisms (MDRO) than their non-dependent counterparts. Reference Patel, Mantey and Mody7 This raises the question of whether this observation can be generalized beyond nursing homes and in settings where MDRO rates tend to be lower but acuity of care is higher. Reference Mody, Foxman and Bradley8
Using data collected as part of the Patient Hand Hygiene Initiative project Reference Mody, Washer and Kaye5 , we compared patient functional dependence level to patient hand MDRO contamination. Specifically, we explored whether Activities of Daily Living (ADL) assessments, such as the Katz ADL score Reference Katz, Downs, Cash and Grotz12 , are associated with hand contamination in newly admitted patients to acute-care hospitals. We hypothesize that patients with functional dependencies are more likely to be contaminated by MDROs on their hands. This information may be used for patient risk stratification and inform targeted patient hand hygiene approaches to decrease the risk of hospital-acquired infections. We were also interested in exploring whether environmental contamination is associated with varying levels of dependencies. This information could be useful in developing efficient MDRO surveillance strategies. Reference Patel, Mantey and Mody7,Reference Cassone, Mantey and Perri13
Methods
Study population and design
This is a secondary cross-sectional analysis of data collected during a prospective cohort study of 399 newly admitted hospitalized patients conducted in two tertiary-care hospitals between February and July of 2017. Reference Mody, Washer and Kaye5 This current study investigates the association between functional disability and hand MDRO colonization at baseline in patients, and is limited to patient data collected during baseline visits. Eligible patients were older than 18 years old and had been admitted in the last 24 hours to a general medicine floor. Baseline swabbing of patient hands and rooms was performed upon enrollment into the study (within 24 hours from admission). Collection of original data analyzed in this study was approved by the institutional review boards at both institutions (IRB# HUM00123279).
Data collection
Clinical and demographic data were obtained through chart reviews and admission documentation, as previously described. Reference Mody, Washer and Kaye5 Information collected included age, sex, height, weight, race, admitting location, history of MDROs (previous 90 days), admission diagnoses, Charlson comorbidity index score, Reference Charlson, Pompei, Ales and MacKenzie14 and dependence in activities of daily living (bathing, dressing, toileting, transferring, continence, and feeding) as measured by the Katz Activities of Daily Living (ADL) scale. Reference Katz, Downs, Cash and Grotz12 These data were collected by the trained research study team member from medical charts including clinician notes and physical/occupational therapist assessments.
Microbiological methods
Trained research staff used premoistened Culturette swabs (Remel, Lenexa, Kansas) to sample the patient’s nares and dominant hand, along with the following high-touch surfaces: bed control/bed rail, call button/television remote, bedside tray tabletop, phone, toilet seat, and bathroom door handle. MDRO identification and resistance testing (including methicillin-resistant Staphylococcus aureus [MRSA], vancomycin-resistant Enterococcus [VRE], and resistant gram-negative bacteria [RGNBs]) was carried out using standard microbiological methods. Reference Mody, Washer and Kaye5
Outcomes definition
Outcomes were assessed on the date of patient enrollment into the study and analyzed independently for each target organism (MRSA, VRE, RGNBs, as well as any MDRO) and defined as (1) no colonization, (2) contamination on hands with or without colonization of nares, and (3) colonization on nares but not on hands.
Environmental contamination was defined as the presence of at least one room surface testing positive for an MDRO (or for MRSA, VRE, and RGNBs, respectively). No contamination was established when all surfaces tested negative.
Exposure definition
Our main exposure of interest was the level of functional dependence in activities of daily living using Katz scale and scored from 0 to 6. Functional status was categorized as functionally independent patients (scored 0), moderately dependent patients (score 1–3), and severely dependent patients (score of 4 or more). We were also interested in the role of individual dependencies with patient hand contamination with MDROs.
Statistical analyses
Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, North Carolina). Unadjusted analyses were performed via multinomial logistic regression comparing hand MDRO contamination status, as well as environmental contamination status, to the level of functional dependence in ADL (independence, moderate dependence, and severe dependence). This was performed for each individual MDRO assessed in the study (MRSA, VRE, and RGNB), as well as contamination by any MDRO. Unadjusted analyses were also performed comparing hand contamination by MDROs to each individual functional dependence in ADL. Patients are often saddled with multiple dependencies. Thus, we also calculated the average number of additional dependencies in ADLs a patient had in addition to the dependency that was being examined. Wilcoxon-ranked sum test was used to compare the order of ADLs when arranged by resulting OR to a hierarchy of ADLs (most difficult to least difficult) established in the literature. Reference Gerrard15 Figures were created using R version 4.2.0 (R Foundation, Vienna, Austria). Reference Wickham, Chang and Henry16–Reference Powell18
Results
Baseline characteristics
Participant demographics and clinical characteristics comparing low and high-dependence patients for relevant covariates are listed in Table 1. Presence of devices, comorbidities, antibiotic use on admission, and history of MDROs in past 90 days, were significantly different between functionally independent and functionally dependent patients. One-hundred and seventeen patients in our cohort were being treated for infections, with most common being skin & soft tissue (32.7%), urinary tract infections (22.4%), and pneumonia (22.4%). Infection rates increased with increasing levels of dependency (Table 1).
Table 1. Baseline characteristics of acute-care patients by level of dependence in ADLs
Abbreviations: ADL, Activities of Daily Living; MDRO, multi-drug resistant organism.Unless otherwise indicated, data is presented as n (%).
Hand contamination by MDROs and functional dependence in ADLs
Rates of hand contamination as well as nares colonization in independent, moderately dependent, and severely dependent patients are shown in Table 2. Severe dependence was associated with higher rates of hand contamination with MRSA (OR: 3.78 (95% CI, 1.39–10.23), VRE (OR: 5.55 (95% CI, 1.09–28.22), and any MDROs (OR: 2.63 (95% CI, 1.21–5.72), while no association was seen for RGNB hand colonization (Table 3).
Table 2. Distribution of patient hand contamination status upon study testing by level of dependence
Abbreviations: MDRO, multi-drug resistant organism; RGNB, resistant gram-negative bacteria; MRSA, methicillin-resistant Staphylococcus aureus; VRE, vancomycin-resistant enterococci.
Data is presented as n (%).
Table 3. Association between functional status and patient colonization
Abbreviations: MDRO, multi-drug resistant organism; RGNB, resistant gram-negative bacteria; MRSA, methicillin-resistant Staphylococcus aureus; VRE, vancomycin-resistant enterococci.
OR: odds ratios. 95% CI: 95% confidence intervals; *P < .05; **P < .01.
Among individual ADLs, feeding dependence had the highest unadjusted odds of MDRO contamination at 4.76 (95% CI, 1.54–14.71), followed by continence (OR 2.71, 95% CI, 1.14–6.42), dressing (OR 2.58, 95% CI, 1.25–5.31), toileting (OR 2.45, 95% CI, 1.14–5.24), bathing (OR 2.32, 95% CI, 1.13–4.75), and transferring (OR 1.81, 95% CI, .89–3.69) (Figure 1A). The average number of additional dependencies was highest for feeding (4.50 additional dependencies), followed by continence and toileting (respectively 3.83 and 3.64), and lowest for transferring, bathing, and dressing (2.76, 3.20, and 3.36, respectively) (Figure 1A). The ordered list of dependencies from highest odds of hand contamination to lowest was compared to an established hierarchy of activities (from most difficult to least) Reference Gerrard15 using the Wilcoxon-Ranked sum test (Figure 1B), with a result of W = 18, and a P value = 1, indicating the rank orders are not significantly different.
Figure 1. Odds Ratios of Dependency in Individual ADLs and MDRO Hand Contamination and Progression of Dependence vs. Hand Contamination. (A) Forest plot of the odds ratios and 95% confidence intervals of dependence in each individual ADL compared to hand contamination with or without nares contamination (red/dark squares) and nares only contamination (green/light squares) by MDROs. Odds ratios greater than one indicate a positive association between dependence in that ADL and contamination by an MDRO. Odds ratios are indicated on the left axis. Overlayed (gray dashed line) is the average of how many additional dependencies a patient tended to have in addition to the dependency being measured. Mean additional dependencies are measured on the right axis. (B) Difficulty ranking was measured using Rasch analysis in Gerrard et alReference Gerrard15, ranking from easiest tasks (top) to most difficult tasks (bottom). Hand contamination (with or without nares contamination rankings were determined by ordering odds ratios from highest to lowest. Wilcoxon-ranked sum was also performed (P = 1), indicating that the two lists are not significantly different.
Environmental contamination by MDROs and level of functional dependence in ADLs
Rates of MDRO contamination of environmental sites ranged from 6.9% to 9.3% for each individual surface, with the exception of the toilet seat (20.2%). MDRO environmental contamination rates did not increase for any organism category among moderately dependent patients compared to independent patients. For severely dependent patients, the odds ratio was significantly higher in the case of MRSA (OR 2.60, 95% CI, 1.16–5.82). All other explored relationships with environmental contamination were not significant (Supplemental Table).
Discussion
The key role of patient hand hygiene in preventing the spread of pathogens has been largely underappreciated until recently. Reference Landers, Abusalem, Coty and Bingham3,Reference Mody, Washer and Kaye5–Reference Patel, Mantey and Mody7 The scarcity of available data on risk for patient hand colonization with MDROs presently hinders the implementation of targeted prevention and intervention efforts. While limited information has been obtained in alternative settings such as nursing homes and rehabilitation centers Reference Hedin, Blomkvist, Janson and Lindblom19 and points to dependence as a potential major risk factor, Reference Patel, Mantey and Mody7 more knowledge needs to be gathered in all care settings, especially acute care. We show that hand contamination rates in hospitalized patients with severe functional dependence are significantly higher than in functionally independent patients. Additionally, we found that this trend is reflected in contamination of the patient’s room.
Gerrad et al. suggest a hierarchy of ADLs exists, Reference Gerrard15 from easiest to most difficult for a patient to accomplish, being: feeding, continence, transferring, toileting, dressing, and bathing. Our findings show dependency in feeding, the “easiest” ADL to perform and usually last to be lost, to have highest odds of MDRO contamination. This is because a patient dependent on feeding is very likely dependent on most or all other ADLs, with their own contribution to risk of contamination. Additionally, such a patient requires more hands-on care from healthcare workers and/or caregivers potentially resulting in opportunities for MDRO transmission.
Our study characterizes patients most likely to benefit from hand hygiene interventions. Targeting high-risk populations and developing personalized interventions based on patient’s functionality is important. Future studies should develop and test such interventions that take into account a patient’s ability to perform hand hygiene and across various healthcare settings. Future epidemiologic investigations should target other high-risk populations such as those with prolonged stays in intensive care units (and thus with short-term but higher levels of dependencies), and in long-term care. The choice of appropriate interventions should be based on each specific population’s characteristics. For example, one study found that difficulty in opening hand-cleaning agent containers might be an overlooked but important and promptly modifiable determinant of patient hand contamination. Reference Knighton, McDowell, Rai, Higgins, Burant and Donskey20 Specifically, they found that some patients were unable to open the pocket-sized containers of hand sanitizer due to either arthritis or difficulty with coordination. These and other barriers should be further explored in the light of our results, and accessibility to appropriate hand hygiene ensured for all patients.
Our study has several limitations, including a relatively small sample size limiting our ability to weigh the potential impact of important conditions such as patients admitted from high MDRO–burden facilities (nursing homes, ICUs, etc.), as opposed to patients admitted from their home, or patients with known history of recent MDRO colonization. Additional further observations, including repeated observations uncovering the time relationship between dependence and contamination, would provide a more detailed picture of the relationship between functional dependence and hand contamination. Second, sampling from two hospitals only may limit generalizability. Another aspect that should be further investigated is whether different risks apply to patients who are chronically dependent, versus those who are recovering from major surgery or other acute conditions and whose dependence is only temporary. This distinction would provide further information about effective prevention strategies. Additionally, other variables that may affect likelihood of contamination may be under-represented in our specific population from two tertiary-care hospitals and thus may have not been considered in this study. Since this study is focused primarily on patient hand colonization, it is not powered to investigate important aspects of overall patient colonization (including other commonly colonized sites). For example, it is possible that overall patient colonization, rather than specific hand colonization, may be driving the observed association with patient dependence status. Our limited available data on patients with nares colonization alone did not show such association and may point to a specific role of patient hands in high risk, severely dependent patients. Analyses of larger datasets will be needed to quantify the specific impact of hand colonization versus overall patient colonization in driving risk among dependent patients. Finally, while we focused on some of the most important pathogens in the healthcare environment, many other organisms that cause infections in acute-care facilities will need to be considered. This is also the case for RGNB, which we included in this study but for which the tested sites do not provide an optimal representation of overall patient colonization.
In conclusion, our study shows a significant, positive association between hand contamination by MDROs and a patient’s functional dependence. Patient engagement and hand hygiene interventions should be tested to determine if a targeted approach based on functional dependence can decrease transmission and burden of MDROs in healthcare.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/ice.2024.70.
Acknowledgments
The authors acknowledge and thank Julia Mantey, MPH, for her contributions to data analysis.
Author contribution
Manuscript drafted by T.B, G.V, and M.C. Selection of database by T.B., G.V, K.E.G, L.M., and M.C. Data analyses performed by T.B and G.V. Interpretation of the data performed by T.B., G.V, K.E.G, L.M., and M.C. Revisions by T.B. G.V, K.E.G, L.M., and M.C. All authors approve the final version of the manuscript.
Financial support
This work was supported by a University of Michigan School of Public Health Certificate in Hospital Infection Prevention and Control Internship to T.B.; a Centers for Disease Control and Prevention contract [BAA 200-2016-91954 to L. M.]; and a National Institutes of Health K24 patient-oriented research and mentorship grant [AG050685 to L.M.].
The sponsors had no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.
Competing interests
The authors report no conflict of interest relevant to the planning, conduct, and reporting of this study.
