Hand Hygiene Deficiency Citations in Nursing Homes

Abstract

Hand hygiene (HH) is recognized as an effective way to decrease transmission of infections. Little research has been conducted surrounding HH in nursing homes (NHs). In this research, deficiency citations representing potential problems with HH practices by staff as identified in the certification process conducted at almost all US NHs were examined. The aims of the study were to identify potential relationships between these deficiency citations and characteristics of the NH and characteristics of the NH environment. We used a panel of 148,900 observations with information primarily coming from the 2000 through 2009 Online Survey, Certification, And Reporting data (OSCAR). An average of 9% of all NHs per year received a deficiency citation for HH. In the multivariate analyses, for all three caregivers examined (i.e., nurse aides, Licensed Practical Nurses, and Registered Nurses) low staffing levels were associated with receiving a deficiency citation for HH. Two measures of poor quality (i.e., [1] Quality of care deficiency citations and [2] J, K, or L deficiency citations, that is deficiency citations with a high extent of harm and/or more residents affected) were also associated with receiving a deficiency citation for HH. Given the percentage of NHs receiving deficiency citations for potential problems with HH identified in this research, more attention should be placed on this issue.

Keywords

citations infection control nursing homes (NHs)

Introduction

Hand hygiene (HH) is recognized as essential to decreasing transmission of infections (Centers for Disease Control [CDC], 2002); however, little research has been conducted surrounding HH in nursing homes (NHs). In this research, deficiency citations for HH in a nationally representative sample of NHs are examined. These deficiency citations represent potential problems with HH practices by staff as identified in the certification process conducted at almost all U.S. NHs. Using a large panel of observations (N=148,900) from 2000 through 2009, and Generalized Linear Models, the aims of this study were to identify potential relationships between these deficiency citations and characteristics of the NH (e.g., staffing levels, quality, ownership) and characteristics of the NH environment (e.g., competition, reimbursement rates).

Prevention and management of infections in NH settings is an important, yet underexamined, resident safety concern (Rust, Wagner, Hoffman, Rowe, & Neumann, 2008). Healthcare Associated Infections (HAIs) are the leading cause of morbidity and mortality in the 1.7 million NH residents with between 1.6 and 3.8 million infections and almost 388,000 deaths occurring annually in this setting (Richards, 2002). Infections are the reason for one fourth of all hospitalizations from long-term care (LTC) facilities (Ahlbrecht, Shearen, Degelau, & Guay, 1999; Richards, 2007). The costs associated with infections in NH settings have a significant impact on the healthcare system with annual estimates ranging from US$38 to US$137 million for antimicrobial therapy and US$673 million to US$2 billion for hospitalizations (Barker et al., 1994; Hu, 1990). Furthermore, the importance of HH in NHs extends beyond just these facilities; as millions of NH residents every year are transferred to and from acute care settings and can potentially spread pathogens between settings (Mody, 2009).

Although there are a number of ways to decrease transmission of infections (i.e., vaccinations, skin testing, use of antimicrobials), HH is recognized as one of the simplest and the most cost effective ways to do so (CDC, 2002; Fendleret al., 2002; Mody, 2009; Smith, Carusone, & Loeb, 2008). When performed correctly, HH can reduce HAIs by up to 30% (Boyce & Pittet, 2002). Improving HH practices has also been identified as a patient safety goal by TJC (i.e., The Joint Commission) and has also been targeted by the World Health Organization (WHO) as the first global patient safety challenge entitled “Clean Care is Safer Care” (Pittet & Donaldson, 2005; WHO, 2009).

Resident infection rates are often associated with HH techniques and use of appropriate HH techniques have been shown to decrease infection rates among NH residents (Richards, 2007). Infection control standards and guidelines are underdeveloped in the NH industry. Regulations stipulate that infection control activities should be performed; but, provide few details. NH infection control specialists (such as infection control nurses) are not mandated. In addition, the infection control specialists that do exist in NHs often lack training in specialized areas of microbiology or epidemiology (Otero, 1993).

Background

Prior Literature

As noted above, examining HH in NHs is important; however, little research has been conducted surrounding HH in NHs (12 publications during the past 9 years). Our review of these publications is summarized in Table 1. This review shows that few empirical studies have been conducted specifically in NHs; and, those that exist have a small sample size and likely poor generalizability. Moreover, few studies have examined factors of NHs associated with HH practices. Overall, our knowledge of HH practices in NHs is limited, and as noted by Juthani-Mehta and Quagliarello (2010), this topic is “vastly understudied” (p. 935). The study presented here is the first nationally representative examination of HH in NHs and is the first to present characteristics of the NH and characteristics of the NH environment that may influence HH practices. However, we note that the analyses are limited to secondary data and include failure to follow HH requirements only as identified by surveyors as part of the annual inspection through the Centers for Medicare and Medicaid Services (CMS).

Table 1.

HH Studies in LTC Settings.

Author(s)	Setting	Type	N	Measure(s)	Outcome(s)
Ashraf et al. (2010)	NHs	Survey	749 employees; 6 NHs	Compliance and knowledge of handwashing based on the CDC guidelines.	Need for more handwashing education to all employees and decreasing the number of barriers to handwashing.
Rummukainen, Jakobsson, Karppi, Kautiainen, & Lyytikäinen (2009)	LTC facilities	Site visits with structured survey	123 facilities	Use of ABHR and ongoing systematic antimicrobials.	One year after site visits, use of ABHR increased and antimicrobials for prevention of urinary tract infections decreased.
Aiello, Malinis, Knapp, & Mody (2009)	NHs	Cross-sectional survey of HCWs	392 HCWs; 4 NHs	HH; alcohol rub use; fingernails; glove use; knowledge of CDC HH guidelines.	Positive responses to HH practices; appropriate glove use and fingernail characteristics; need for further education of HH guidelines.
Rao et al. (2009)	NHs	Cluster randomized trial	12 NHs; 565 residents	Observed infection control measures (i.e., HH facilities; environmental cleanliness; safe disposal of clinical waste) and then introduced improved infection control measures to the intervention group.	Low compliance rates of HH; improved compliance with safe disposal of clinical waste.
Smith et al. (2008)	LTC facilities	Cross-sectional study	459 HH opportunities; 2 LTC facilities	HH monitoring instrument to examine HH compliance.	Low HH compliance.
Pan et al. (2008)	LTC facility	Observationl study	308 HH opportunities	HH compliance (i.e., Handwashing and Glove Use).	HH compliance low and comparable to acute care settings.
Mackenzie, James, Smith, Barnard, & Robinson (2008)	Mental health care settings for older people	Survey	114 staff	Handwashing assistance provided to patients by staff.	Staff do not often assist patients in handwashing.
Richards (2007)	LTC facilities	Literature review	N/A	Urinary tract infections; respiratory tract infections; gastrointestinal infections; skin infections; antimicrobial-resistant infections.	Evaluations; infection control programs; immunization; antimicrobial prescribing; HH.
Huang & Wu (2008)	NHs	Intervention with self reporting	40 NAs	Impact of a training program in HH for NAs.	An intense training program improved knowledge and compliance of HH and a reduction in resident infection rates.
Mody, Langa, Saint, & Bradley (2005)	NHs	Self-administered questionnaire	35 NHs	Infection control practices; immunization practices; role of Icp and infection control activities.	Varying infection control guidelines; different definitions of infections; varying immunization compliance; small number of full time employed ICP.
Fendler et al. (2002)	ECF	Infection rates	275-bed ECF; 265 employees	Effect of alcohol gel sanitizers on infection rates.	Alcohol gel sanitizer can reduce infection rates.
Goldrick (1999)	Skilled nursing LTC facilities	Descriptive study	136 facilities	Infection control programs; effectiveness of infection control activities.	Lower infection surveillance and control; low handwashing compliance.

Note: ABHR = alcohol-based hand rubs; CDC = Centers for Disease Control; ECF = extended care facility; HH=hand hygiene; ICP = infection control practitioner; NA = Nurse Aides; NH=nursing home; LTC = long-term care.

NH Certification and Deficiency Citations

CMS requires NHs to be certified before they can receive reimbursement for Medicare and/or Medicaid residents. The Medicare and Medicaid programs constitute major payers for care, thus almost all NHs in the United States (i.e., 96%) participate in this process. As part of this certification process deficiency citations can be issued in specific areas of care (these are often called F-tags). That is, if NHs do not meet certain minimum health and safety standards, deficiency citation(s) can be issued. The current certification process is described in detail by CMS (www.cms.hhs.gov).

The Nursing Home Reform Act (NHRA) is considered to be highly influential in this regulatory process. The NHRA was included in the Omnibus Budget Reconciliation Act (OBRA) of 1987 (P.L. 100-203). Of most significance to this research, the NHRA mandated that NHs were to have a functioning infection control program in place (Goldrick, 1999). The programs were to include surveillance, outbreak investigations, isolation procedures, educational programs for both employees and residents, and reporting procedures (Ahlbrecht et al., 1999). However, details on implementing each of these programs were generally lacking. For example, educational programs for nurse aides are likely highly important for infection control (Juthani-Mehta & Quagliarello, 2010). Nevertheless, educational programs are considered to be ill-defined and insufficient, and are not mandated as part of the 75 hrs of training nurse aides receive (Juthani-Mehta & Quagliarello, 2010).

The staff hand washing after direct resident contact deficiency citation is examined in this research (i.e., F-Tag 444). Specifically, CMS guidelines state workers are expected to wash their hands before and after direct resident contact (for which HH is indicated by acceptable professional practice); performing any invasive procedure (e.g., fingerstick blood sampling); entering isolation precaution settings; eating or handling food (hand washing with soap and water); assisting a resident with meals; assisting a resident with personal care (e.g., oral care, bathing); handling peripheral vascular catheters and other invasive devices; inserting indwelling catheters; assisting a resident with toileting (hand washing with soap and water); and changing a dressing. In addition, after coming in contact with a resident’s intact skin (e.g., when taking a pulse or blood pressure, and lifting a resident); personal use of the toilet (hand washing with soap and water); contact with a resident with infectious diarrhea including, but not limited to infections caused by norovirus, salmonella, shigella, and Clostridium difficile (hand washing with soap and water); blowing or wiping nose; contact with a resident’s mucous membranes and body fluids or excretions; handling soiled or used linens, dressings, bedpans, catheters and urinals; handling soiled equipment or utensils; performing personal hygiene (hand washing with soap and water); and, removing gloves or aprons; and, when hands are visibly soiled (hand washing with soap and water). Also: when coming on duty and after completing duty (CMS Manual System, 2009a).

CMS also gives guidelines on recommended techniques for washing hands. This includes the following: with soap and water include wetting hands first with clean, running warm water, applying the amount of product recommended by the manufacturer to hands, and rubbing hands together vigorously for at least 15 seconds covering all surfaces of the hands and fingers; then rinsing hands with water and drying thoroughly with a disposable towel; and turning off the faucet on the hand sink with the disposable paper towel (CMS Manual System, 2009a). In addition, CMS states that except for situations where hand washing is specifically required, antimicrobial agents such as alcohol-based hand rubs (ABHR) are also appropriate for cleaning hands and can be used for direct resident care. Recommended techniques for performing HH with an ABHR include applying product to the palm of one hand and rubbing hands together, covering all surfaces of hands and fingers, until the hands are dry. In addition, gloves or the use of baby wipes are not a substitute for HH (CMS Manual System, 2009a).

From 1997 to 2009, an average of approximately 1,000 NHs per year received a deficiency citation for this F-tag (i.e., F-Tag 444). This was the 24th most frequently used F-tag (from approximately 190 available). This F-tag was examined in the research presented here, as it was the only deficiency citation available that specifically addressed hand washing.

Conceptual Framework

This research was guided by the conceptual framework presented in Figure 1 (Mody, 2009). This conceptual framework is proposed as a means of understanding the potential relationships between HH, deficiency citations, and resident outcomes. It was also used as a means for guiding variable selection for the empirical analyses. This conceptual framework consists of resident (e.g., age and decreased immunity) and facility (e.g., staffing levels and quality) risk factors as well characteristics of the NH environment (e.g., competition and reimbursement rates).

Figure 1.

Conceptual Framework for Examining Deficiency Citations for Hand Hygiene in Nursing Homes.

NH residents are typically older and more susceptible to acquiring infections, and this can lead to hospitalizations or even death (Mody et al., 2011). Risk factors associated with NH residents’ susceptibility to acquiring infections include malnutrition and functional impairments. In addition, while in the NH, there are a number of risk factors for residents coming in contact with and/or spreading infections. These risk factors include care processes such as contact with medical devices and group activities (i.e., physical therapy, dining facilities, and bathing areas) (Richards, 2007). Resident risk factors and potential resident outcomes were initially combined in a conceptual framework by Mody (2009). That is, based on existing literature, these resident risk factors and HH practices were proposed to influence healthcare acquired infections, and in turn healthcare acquired infections were proposed to influence resident outcomes such as death and hospitalization. In this research, we modify this initial conceptual framework to include NH internal, organizational, and external factors, as well as the deficiency citation for HH (F-Tag 444). This is shown in Figure 1.

Internal factors are operating characteristics of the facility, such as staffing levels; organizational factors are characteristics of the facility itself, such as the number of beds; and external factors are characteristics generally outside of the influence of the organization, such as competition from other providers. Including these factors is useful, because many internal, organizational, and external factors are believed to influence care in NHs.

One often-cited important internal characteristic, for example, is the staffing level of nurse aides. High nurse aide staffing levels have been shown to be associated with better quality in many prior NH studies (Castle, 2008). One often-cited important organizational characteristic, for example, is the ownership of the facility. For-profit facilities are thought to provide lower quality of care than not-for-profit NHs (Comondore et al., 2009). One often-cited important external characteristic, for example, is the Medicaid reimbursement rate. High rates have been shown to be associated with better quality in many prior NH studies (Hyer et al., 2009). These factors may also be associated with whether or not a facility receives a deficiency citation for HH.

This conceptual model is used, first because based on data availability the NH is used as the unit of observation. Thus, an organization-based conceptual framework was appropriate. Second, a similar conceptual framework was used previously in NH analyses examining deficiency citations (Castle, Wagner, Ferguson-Rome, Men, & Handler, 2011; Castle, Wagner, Ferguson, & Handler, 2011). A complete list of the factors used in the analyses and their definitions are provided in Table 4. Given the large number of factors examined, and given the paucity of research in this area, this study is presented as an exploratory analysis. Thus, specific hypotheses for the potential relationships between factors and deficiency citations for HH were not developed.

Method

Secondary data (described below) from 2000 to 2009 are used in the analyses. These data were combined into a panel consisting of a total of 148,900 NH observations. Descriptive analyses and multivariate analyses are used to examine characteristics (i.e., internal, organizational, and external factors) associated with receiving deficiency citations for HH (F-444).

Data Sources

Deficiency citations are recorded in the Online Survey, Certification, And Reporting (OSCAR) data. Moreover, the OSCAR also includes aggregate resident information (e.g., number of residents with dementia, with psychiatric conditions, etc.), staffing information (e.g., number of full-time equivalent (FTE) nurse aides, etc.), and facility information (e.g., ownership characteristics, bed size, etc.). Thus, all of the internal and organizational factors examined in this research came from the OSCAR data.

The OSCAR data are publicly available from CMS (i.e., Centers for Medicare & Medicaid Services). The OSCAR is the only readily available data source that represents a national sample of NHs and includes resident, staffing, facility, and deficiency citation information over time. The reliability of many variables is well established (such as ownership characteristics; Kash, Hawes, & Phillips, 2007; Kash, Naufal, Cortés, & Johnson, 2010); but, this should not be overstated as no comprehensive psychometric analyses of the data are available, and opinions vary as to the overall reliability of this data (Kash et al., 2007). Details regarding the variables included in the OSCAR and how the data are collected is provided by Kash et al. (2007).

A limited number of variables included in the analyses came from the area resource file (ARF). Specifically, variables used as external factors (i.e., per capita income in the market and the number of elderly in the market). Extensive details regarding the ARF can be found at www.arfsys.com.

In addition, Medicaid reimbursement levels (a variable used as an external factor) came from primary data collected by the authors. This followed a process previously used by others (Grabowski, Feng, Intrator, & Mor, 2004). This primary data collection included contacting representatives in each state, validating responses with information from other sources (when available), and adjusting the Medicaid reimbursement levels using the overall consumer price index (CPI). Extensive details regarding this data collection methodology is provided by Grabowski et al. (2004).

Analytic Approach

The OSCAR data for each NH are available on a yearly basis (as the certification process occurs approximately yearly). Moreover, each facility has a unique ID number. Using these ID numbers, the OSCAR data for each NH from 2000 through 2009 were combined to create a longitudinal panel data source. The period 2000 through 2009 was used because these data were available to the authors, and the data were used in a longitudinal panel format that allows more accurate inference of the included model parameters (Hsiao, Mountain, & Ho-Illman, 1995). ZIP codes are included in the OSCAR that were used to identify counties in which NHs were located and match facilities with the ARF data.

In the baseline year (i.e., 2000) 16,745 NHs were identified. Subsequently, 10% (N=1,855) of these NHs were excluded because they could not be identified in a subsequent year up to 2009. Thus, combined for the analyses 148,900 NH observations were used in this analysis (i.e., 14,890 NHs × 10 years).

Analyses

Descriptive statistics for the deficiency citation for HH (F-444) are presented. This includes the percentage of NHs receiving this deficiency citation; this citation as a percentage of all deficiency citations given; and, the rank of this citation of all deficiency citations used, for each year from 2000 to 2009.

Twelve categories (labeled “A” through “L”) are used for each deficiency citation. These categories vary in scope and severity (i.e., the greater the letter, the more severe the citation). The severity depends on the extent of harm to the resident and the scope depends on the number of residents affected (these designations are further defined in Table 2). Using all years of data (2000 through 2009), the percentage and number of NHs receiving each category of the deficiency citation for HH (F-444) is presented.

Table 2.

Nursing Homes Receiving Deficiency Citations^a for Hand Hygiene (F-tag 444) From 2000 to 2009.

Year	Percentage of Nursing Homes With an F-Tag 444 Deficiency Citation	F-Tag 444 Deficiency Citations as a Percentage of All Citations Given	Rank of F-Tag 444 Deficiency Citations of All Deficiency Citations Given
2000	7.37	1.17	31
2001	7.38	1.18	32
2002	7.31	1.40	26
2003	8.64	1.46	25
2004	9.06	1.51	23
2005	9.97	1.54	22
2006	10.75	1.55	22
2007	11.56	1.64	20
2008	11.65	1.67	19
2009	11.98	1.75	19
Summary (all years)	9.06	1.63	24

Note: ^aA nursing home deficiency citation is defined as “a finding that a nursing home failed to meet one or more federal or state requirements” (Department of Health and Human Services [DHHS], 2004, p. 34).

Descriptive statistics (means and percentages) for the internal, organizational, and external factors used in the analyses are presented. These are stratified by NHs receiving a deficiency citation for HH (F-444) in 2000, or not. These descriptive statistics are provided for the baseline year of 2000 for parsimony.

To examine internal, organizational, and external factors of NHs associated with receiving deficiency citation for HH (F-444) multivariate analyses were used. Thus, multicollinearity and collinearity levels among the variables using the variance inflation factor (VIF) test were first measured (SAS Institute, 1999).

The multivariate analyses used consisted of Generalized Linear Models. Specifically, generalized estimating equations (GEE) with a logit link were used (Zeger & Liang, 1992). GEE controls for the biases that can occur in data consisting of repeat observations (i.e., in this case, NHs with repeat observations from 2000 to 2009). The logit link was used because the variable of interest was dichotomous (e.g., deficiency citation F-444 or no deficiency citation). SAS^® version 9.13 was used for all statistical analyses.

Results

Results of the Descriptive Analyses

From 2000 to 2009 an average of approximately 9% of all NHs per year received an HH deficiency citation (F-444; see Table 2). An upward trend is evident, with an average yearly percentage increasing from 7.37% (in 2000) to 11.98% (in 2009). On average, these are the 24th most frequently used deficiency citation (from 190 available); although, as shown in Table 2, the use of these citations does vary slightly per year from the 32nd most frequently used (in 2001) to the 19th (in 2008 and 2009). In addition, the use of deficiency citations for HH (F-444) varies by state. In some states, such as Pennsylvania, an average of 6% of facilities were given this deficiency citation in 2009; whereas, in other states such as Michigan, an average of 15% of facilities were given this deficiency citation in 2009. The tercile distribution of these deficiency citations by state is shown in Figure 2.

Figure 2.

Tercile distribution of deficiency citations for Hand Hygiene (F-444) are presented. The figure represents the average number of these deficiency citations given per nursing home in each state.

Table 3 shows the percentage of HH deficiency citations (F-444) from 2000 to 2009 given by scope and severity. This shows that almost no NHs received deficiency citations in the most severe categories (i.e., J, K, and L). Most deficiency citations (i.e., 66.3%) were at the “D” level (representing potential for more than minimal harm [severity] and isolated cases [scope]).

Table 3.

Scope and Severity Classifications Used for Medicare/Medicaid Certification Survey Deficiencies and Percentage of Nursing Homes Receiving Deficiency Citations for Hand Hygiene (F-tag 444) From 2000 to 2009.

	Scope
Severity	Isolated	Pattern	Widespread
Potential for minimal harm	A (NR)	B (n=241 2%)	C (n=43 0.5%)
Potential for more than minimal harm	D (n=5,843 66%)	E (n=2,561 29%)	F (n=126 1.5%)
Other actual harm	G (n=1 0.01%)	H (0)	I (0)
Actual or potential for death/serious injury	J (0)	K (0)	L (n=1 0.0%)

Note: Number in parentheses is the number of citations for hand hygiene (F-tag 444) from 2000 to 2009 (total number of these citations = 8,816). Severity is the extent of harm to the resident. Scope is the number of residents affected (Isolated defined as affecting a single or very limited number of residents; Pattern defined as affecting more than a very limited number of residents; Widespread defined as affecting a large portion or all residents). Thus, A deficiency citations are the least problematic and L are the most problematic. For example, an A-level deficiency citation may be given if one nurse aide was observed to not wash her hands adequately; whereas, a C-level deficiency citation may be given if numerous nurse aides were observed to not wash her hands adequately. If one nurse aide was observed to not wash her hands adequately with a resident known to have norovirus a D-level deficiency citation may be given.

NR, not recorded in the OSCAR.

Descriptive statistics of the variables used in the analysis for the baseline period of the year 2000 are presented in Table 4. A total of 1,828 NHs received this deficiency citation (F-444) in the year 2000. Compared to those NHs that did not receive a deficiency citation for HH, the bivariate analyses show higher rates of both more quality of care deficiency citations (addresses how well the facility renders services provided and supervised by nursing staff) and J, K, or L deficiency citations.

Table 4.

Descriptive Statistics of NHs Receiving and Not Receiving Deficiency Citations for HH (F-tag 444).

	Facilities Receiving F-tag 444 Citations^a		Facilities Not Receiving F-tag 444 Citations^b
Variables	Mean (or %)	Standard Deviation	Mean (or %)	Standard Deviation	Operational Definition of Variable
Internal factors
Nurse aide staffing (FTEs per resident)	0.33	0.13	0.32	0.14	FTE nurse aides per resident.
LPN staffing (FTEs per resident)	0.12	0.09	0.12	0.09	FTE LPNs per resident.
RN staffing (FTEs per resident)	0.06	0.09	0.07	0.11	FTE RNs per resident.
Resident case-mix (ADL score)	0.12	0.26	0.12	0.26	The score for three ADLs (eating, toileting, and transferring) constructed by giving a score of 1 for low assistance, 2 for moderate assistance, and 3 for high need for assistance summed for each ADL.
Restraint use	0.10	0.12	0.10	0.12	Proportion of residents in physical restraints.
Psychiatric condition	0.17	0.15	0.16	0.17	Proportion of residents diagnosed with psychiatric conditions.
Dementia	0.44	0.20	0.43	0.20	Proportion of residents diagnosed with dementia (defined as an ICD-9 code indicating a specific dementing illness).
Organizational factors
Quality of care deficiency citations	3.09*	2.42	1.59	1.84	Deficiency citations representing the sum of 19 different deficiency citations (F-tags are: 309, 310, 311, 312, 314, 316, 317, 318, 319, 321, 322, 323, 324, 325, 328, 329, 330, 333, 353). Within-state rankings (percentiles) used for analyses.
J, K, or L deficiency citations	0.08*	0.46	0.04	0.38	Any deficiency citation at J, K, or L level (see Table 2). Within-state rankings (percentiles) used for the analyses.^~
Medicaid resident occupancy	64%	23%	63%	26%	Percent of residents with Medicaid as payor.
Size (number of beds)	110	74	109	73	Number of beds in the nursing home.
For-profit ownership	70%*	−	65%	−	For-profit or not-for-profit (including government owned) ownership.
Chain member	60%*	−	56%	−	Whether member of a nursing home chain or not.
Occupancy rate	82%	16%	83%	17%	Percent of beds occupied by residents.
External factors
Medicaid reimbursement rate (US$)¹	125.05*	27.14	132.40	31.62	The state average daily (i.e., per diem) payment rate for Medicaid residents (US$). The rates were adjusted to constant 2000 dollars using the Consumer Price Index (CPI).
Competition (Herfindahl Index)^d	2,216	2,477	2,076	2,384	Herfindahl Index. Each facility’s percentage share of beds in the county / squared market shares of all facilities in the county (0-1). Higher values indicate a less competitive market.
Elderly per square mile²	81*	174	146	419	Average number of elderly per square mile (age 65 and above) in the county.
Per capita income (US$)²	26,503*	6,237	27,324	7,966	Average income (US$) for all county residents

Note: ^aN = 1,828 facilities (based on 2000 data); ^bN = 12,372 facilities (based on 2000 data).

ADL = activities of daily living; FTE = full-time equivalent; HH = hand hygiene; LPNs = Licensed Practical Nurses; LTC = long-term care; NH = nursing homes; RNs = Registered Nurses.

Difference between receiving deficiency citation for HH and not significant at p<0.001.

The unit of analysis for this variable is the state (unless otherwise noted, for all other variables the facility is the unit of analysis).

The unit of analysis for this variable is the county.

Variation in the use of nursing home deficiency citations is known to occur from state to state. Using the percentile distribution of deficiency citations within each state gives to a more consistent comparison of deficiency citations across states.

(A). variables primarily came from the Online Survey, Certification, And Reporting (OSCAR); with elderly in the county and per capita income coming from the Area Resource File (ARF); and, Medicaid reimbursement rates coming from primary data; (B). the figures presented are for the baseline 2000 data. The baseline figures are presented for parsimony.

Results of the Multivariate Analyses

Based on the commonly used threshold value of 0.8, the variables showed no problems of collinearity and no VIF score exceeded 2.5. Results from the GEE marginal models with a logit link are shown in Table 5.

Table 5.

Multivariate Results of Nursing Homes Receiving and Not Receiving Deficiency Citations for Hand Hygiene (F-tag 444) From 2000 to 2009.

	AOR	95% CI
Internal factors
Nurse aide staffing	0.965	(0.961	−0.983)	**
LPN staffing	0.978	(0.941	−0.993)	*
RN staffing	0.976	(0.938	−0.989)	**
Resident case-mix (ADL score)	1.007	(1.039	−1.111)	***
Restraint use	1.017	(0.989	−1.046)
Psychiatric condition	1.063	(0.992	−1.035)
Dementia	1.031	(1.002	−1.062)	**
Organizational factors
Quality of care citations	1.137	(1.133	−1.142)	***
J, K, or L citations	0.826	(0.779	−0.876)	***
Medicaid resident occupancy	0.942	(0.927	−0.956)	***
Medicaid resident occupancy × Medicaid reimbursement rate^a	0.980	(0.921	−0.997)	*
For-profit ownership	1.005	(0.945	−1.068)
Size	1.021	(0.994	−1.049)
Chain membership	0.935	(0.887	−0.986)	**
Occupancy rate	1.004	(0.975	−1.035)
External factors
Medicaid reimbursement rate	0.922	(0.911	−0.989)	**
Competition (Herfindahl Index)	1.013	(0.982	−1.045)
Elderly in county	0.976	(0.942	−1.018)
Per capita income	0.993	(0.959	−1.029)
Intercept	0.074	(0.0367	−0.149)	***

Note: N = 148,900 observations.

Statistically significant at p=0.05 level or better; ** Statistically significant at p=0.01 level or better; *** Statistically significant at p=0.001 level or better. Analyses also include 49 state and year dummy variables (not shown). Results reported using generalized estimating equations (GEE).

FTE = full-time equivalent, ADL = activities of daily living, RNs = Registered Nurses, LPNs = Licensed Practical Nurses, AOR = Adjusted Odds Ratio; CI = Confidence Interval.

Adjusted Odds Ratios were manually calculated by using the mean levels of the variables (Medicaid reimbursement rate and Medicaid resident occupancy [i.e., a and b] and the variance/covariance matrix [Ai & Norton, 2003]).

For the internal factors all of the staffing level factors were significant at conventional levels (see Table 5). That is, for nurse aides high staffing levels were associated with low deficiency citations for HH (Adjusted Odds Ratios (AOR) = 0.965; p>=.01); high staffing levels of LPNs were associated with low deficiency citations for HH (AOR = 0.978; p>=.05); and, high RN staffing levels were associated with low deficiency citations for HH (AOR = 0.976; p>=.01).

For the organizational factors examined (see Table 5), facilities with quality of care deficiency citations were significantly associated with a high likelihood of receiving a deficiency citation for HH (AOR = 1.137; p <=.001); however, those with J, K, or L level deficiency citations were significantly associated with a low likelihood of receiving a deficiency citation for HH (AOR = 0.826 p <=.001). Also, facilities that are members of a chain were significantly associated with a low likelihood of receiving a deficiency citation for HH (AOR = 0.935 p <=.01).

For the external factors examined (see Table 5), facilities located in states with higher Medicaid reimbursement rates were significantly associated with a low likelihood of receiving a deficiency citation for HH (AOR = 0.922; p <=.01).

Discussion

HAIs are the leading cause of morbidity and mortality amongst NH residents. NH residents are particularly susceptible to HAIs because they are frail, have multiple chronic comorbidities, and take antibiotics that are often inappropriately prescribed leading to greater susceptibility of antibiotic-resistant organisms (Bradley, 2009; Mody et al., 2011). HH can be an extremely influential practice in infection control management to prevent transmission of infectious disease in NHs (Mody et al., 2011). The research presented here is significant in that it gives a nationally representative picture of deficiency citations for HH (i.e., F-tag 444) in NHs. That is, in this study we examined relationships between these deficiency citations and characteristics of the NH and characteristics of the NH environment.

Overall, the conceptual framework used seemed appropriate for examining these deficiency citations. However, we note that the conceptual framework was modified from the initial work of Mody (2009) and the potential relationships indicating HH influencing healthcare acquired infections, death and hospitalization were not examined in the research presented here (see Figure 1). The majority of factors included in the conceptual framework and resulting analyses were significant in the multivariate analyses. This was especially true for the internal and organizational factors. That is, 5 of the 7 internal factors were statistically significant (i.e., nurse aide staffing, LPN staffing, RN staffing, resident case-mix, and dementia) and 5 of the 8 organizational factors (i.e., quality of care citations, J, K, or L citations, Medicaid resident occupancy, Medicaid resident occupancy × Medicaid reimbursement rate, and chain membership) were statistically significant. The external factors performed less well, with 1 of the 4 factors significant (i.e., Medicaid reimbursement rate) in the multivariate analyses. Thus, further modification of these external factors may be warranted in subsequent iterations and development of this conceptual framework.

With an average of approximately 9% of all NHs per year receiving a HH deficiency citation (see Table 2), our results confirm previous survey-based research that HH measures are not uniformly used in NHs ( Aiello et al., 2009; Ashraf et al., 2010). The results show an increasing percentage of NHs with an F-tag 444 deficiency citation and a reduction in rank of F-tag 444 deficiency citations of all deficiency citations given (see Table 2). Thus, HH deficiency citations are becoming more common over time. We are not able to determine the cause(s) of this trend. We speculate that one reason may be greater awareness of surveyors to issues of infection control. A second potential reason may be a more difficult operating environment for NHs. Resident case-mix has increased over time, while revenues have remained flat (Hyer et al., 2009). While providing more care with the same (or less) resources appropriate infection control may be sacrificed.

The issue of staffing appears very prominent in our findings (see Table 5). For all three caregivers examined (i.e., nurse aides, LPNs, and RNs) low staffing levels were associated with receiving a deficiency citation for HH (F-444). With low staffing levels, these caregivers are likely hurried, and may not have the time to use appropriate HH techniques. Empirical research has identified inadequate staffing levels to exist in many NHs (e.g., Hyer, Temple, & Johnson, 2009) and to be highly associated with quality of care (Castle, 2008; Castle & Ferguson, 2010). Our findings show one further consequence of low staffing: potential poor HH practices. We note that other staffing related reasons for poor HH likely exist. For example, poor HH practices could result from limited time for effective supervision, high turnover, or from less peer support. Given the prominent findings for staffing identified in this research, it would be worth exploring some of these other staffing-related factors further.

Given the current inhospitable financial and reimbursement climate in which NHs operate, staffing levels are unlikely to improve in the near future. Therefore, HH-specific training may be one answer to improve compliance especially for nurse aides who provide 80%-90% of resident care (Beck, Ortigara, Mercer, & Shue, 1999). However, studies show that HH training in LTC facilities may be inadequate (Leinbach & English, 1995). To enhance HH training, facilities could do more to educate staff that contaminated hands are one of the most common ways to transmit infections, provide 1:1 coaching on how to wash hands appropriately, describe the pros and cons of handwashing vs. alcohol-based hand sanitizer products, and the need for handwashing (rather than the use of alcohol-based hand sanitizer products) when hands are soiled or there may be the presence of C. difficile infection for example.

Still, potential training and education solutions to address this deficiency citation necessitate that we further understand why staff are not washing their hands as frequently as the guidelines recommend. It could be that the underlying behavior is influenced by the culture or attitudes of staff. Or it could be that staff does not have sufficient access to alcohol gel (for example). Training and education approaches should be informed by a more detailed understanding of reasons for poor HH.

Institutional policies and procedures could be developed to routinize HH as a required process for all staff, followed by internal quality improvement audits (similar to state surveyor audits) to determine HH compliance. It has also been recognized that healthcare workers should receive feedback to ensure proper HH practices (Ahlbrecht et al., 1999). We speculate that a lack of such feedback may also be reflected in the findings. That is, with low staffing levels of RNs in a leadership position (who would typically provide this feedback), this process may be weak or nonexistent.

Our findings also show that receiving a deficiency citation for HH (F-444) is associated with poor quality in general (i.e., with quality of care deficiency citations). This may be due to the notion that HH practices and quality of care are influenced by similar operating and environmental conditions. Nevertheless, HH deficiency citations are associated with lower levels of the worst lapses in quality of care deficiency citations (i.e., J, K, and L deficiency citations, which represent deficiency citations with a high extent of harm and/or more residents affected). These contradictory findings should be investigated further. We speculate that this may reflect the known orthogonality of quality indicators. That is, NHs often perform poorly in one area at the same time as they perform better in a different area (Castle & Ferguson, 2010).

The findings show that states with lower Medicaid reimbursement rates had higher rates of deficiency citations for HH. Low Medicaid reimbursement rates are also a characteristic of the NH industry (Grabowski et al., 2004). These low rates are associated with many negative NH outcomes such as high staff turnover and poor quality of care (Grabowski et al., 2004). That is, Medicaid reimbursement rates can influence the operation of NHs.

Presumably, the HH lapses resulting in deficiency citations by surveyors represent observed cases of problems, and not lapses of reporting and documentation problems (which is a criticism of many other deficiency citations). As such, the use of deficiency citations for HH (similar to citations given for other observed problems such as medication errors; F-329) likely underestimate the potential HH problems in NHs. One would assume that staff would be following clinical care protocols (including HH) when surveyors are observing care. Thus, the potential problems with HH are likely much larger than we report here.

Interventions to improve HH need to be multifactorial in nature since multimodal interventions are more effective at improving HH compliance rates over single employed approaches. The multifactorial approach should include three areas improving knowledge of HH through education (Laustsen, Bibby, Kristensen, Møller, & Thulstrup, 2009); reinforcing behavior through quality improvement activities (Pittet et al., 2000); motivating the HH behavior through introducing products such as personal bottles or wall mounted–alcohol hand rubs; changing the culture through increasing administrative support; and using reminders (Pittet et al., 2000).

Limitations and Suggestions for Future Research

One limitation of examining deficiency citations for HH is that they do not necessarily represent actual resident outcomes. The literature would suggest that when caregivers have poor HH practices this is detrimental to residents, but this cannot be substantiated in our analysis. Future research examining the influence of HH practices on resident outcomes is needed. Other sources of data could be used for such analyses, including the minimum data set (MDS), which provides a somewhat comprehensive evaluation of each resident (Castle & Ferguson, 2010).

From the information available in the OSCAR, it is not possible to present a more fine-grained analysis of specific breakdowns in HH requirements. That is, if the errors occur with specific residents or at specific locations. A more fine-grained analysis of some of the differences that exist between states may also be useful. As shown in Figure 2, considerable differences exist between states. These may be associated with specific state incentives or policies that could be further investigated.

Some of these more fine-grained analyses would also seem warranted given the recent changes in F-tags. In late 2009, F-tags 441, 442, 443, 444, and 445 were all collapsed in F-441 (CMS, 2009b). This enables aggregate “infection control” deficiency citations to be examined; but, is limited in that it is no longer possible to examine specific components of infection control, such as HH.

Conclusions

As noted above, deficiency citations have several limitations when investigating HH. However, no prior research has presented a longitudinal analysis of a national sample of NHs examining these deficiency citations. With HH deficiency citations increasing over time and with an average of approximately 9% of all NHs per year receiving an HH deficiency citation, our findings provide tentative evidence that HH may be a problem in NHs. By examining relationships between these deficiency citations and characteristics of the NH and characteristics of the NH environment, we identify many of these factors (such as staffing) to be potentially important. This research may foster more interest in further elaborating influential determinants in this area of care; with the important objective of improving resident care and resident outcomes.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Nicholas Castle is a professor with the University of Pittsburgh in the Department of Health Policy & Management. His research examines the quality of nursing homes. Previous research initiatives include examining staffing levels, staff turnover, top management, resident satisfaction, and safety culture. Dr. Castle is a Fellow of the Gerontological Society of America.

Laura Wagner is an assistant professor with the New York University College of Nursing in the Hartford Institute for Geriatric Nursing. Her research examines improving patient safety care processes in nursing homes. Previous research has focused on physical restraint and siderail reduction, improving adverse event reporting and disclosure, and safety culture assessment.

Jamie C. Ferguson-Rome is a project director with the University of Pittsburgh in the Department of Health Policy and Management. Her work includes safety and quality of care in nursing homes and assisted living facilities and the age-friendliness of communities. She can be reached at jamief@pitt.edu.

Steven Handler is an assistant Professor with the University of Pittsburgh in the Department of Biomedical Informatics and Division of Geriatric Medicine. His research focuses on the development, implementation, and assessment of clinical decision support systems to improve medication and patient safety primarily in the nursing home setting.

References

Ahlbrecht

Shearen

Degelau

Guay

D. R.

(1999). Team approach to infection prevention and control in the nursing home setting. American Journal of Infection Control, 27, 64-70.

Norton

E. C.

(2003). Interaction terms in logit and probit models. Economics Letters, 80, 123-129.

Aiello

A. E.

Malinis

Knapp

J. K.

Mody

(2009). The influence of knowledge, perceptions, and beliefs, on hand hygiene practices in nursing homes. American Journal of Infection Control, 37, 164-167.

Ashraf

M. S.

Hussain

S. W.

Agarwal

Ashraf

El-Kass

Hussain

. . . Wolf-Klein

(2010). Hand hygiene in long-term care facilities: A multicenter study of knowledge, attitudes, practices, and barriers. Infection Control and Hospital Epidemiology, 31, 758-762.

Barker

W. H.

Zimmer

J. G.

Hall

W. J.

Ruff

B. C.

Freundlich

C. B.

Eggert

G. M.

(1994) Rates, patterns, causes, and costs of hospitalization of nursing home residents: A population-based study. American Journal of Public Health, 84, 1615–1620.

Beck

Ortigara

Mercer

Shue

(1999). Enabling and empowering certified nursing assistants for quality dementia care. International Journal of Geriatric Psychiatry, 14, 197-211.

Bradley

S. F.

(2009). Infections and infection control in the long-term care setting. In Norman

Yoshikawa

(Eds.), Infectious disease in the aging: A clinical handbook (pp. 387-408). Humana Press, New York, NY.

Boyce

J. M.

Pittet

(2002). Guideline for hand hygiene in health-care settings: Recommendations of the Healthcare Infection Control Practices Advisory Committee and the HICPAC/SHEA/APIC/IDSA hand hygiene task force. [Supplement] Infection Control and Hospital Epidemiology, 23, S3-S40.

Castle

N. G.

(2008). Nursing home caregiver staffing levels and quality of care: A literature review. Journal of Applied Gerontology, 27, 375-405.

10.

Castle

N. G.

Ferguson

J. C.

(2010). What is nursing home quality and how is it measured? Gerontologist, 50, 426-442.

11.

Castle

N. G.

Wagner

L. M.

Ferguson

J. C.

Handler

S. M.

(2011). Nursing home deficiency citations for safety. Journal of Aging and Social Policy, 23, 34-57.

12.

Castle

N. G.

Wagner

L. M.

Ferguson-Rome

J. C.

Men

Handler

S. M.

(2011). Nursing home deficiency citations for infection control. American Journal of Infection Control, 39, 263-269. doi: 10.1016/j.ajic.2010.12.010

13.

Centers for Disease Control (CDC). (2002). Guideline for hand hygiene in health-care settings (RR-16). Morbidity and Mortality Weekly Report, 51.

14.

CMS Manual System. (2009a). Pub. 100-07 State Operations Provider Certification. Department of Health & Human Services (DHHS). Transmittal 52. September. Retrieved February 25, 2010, fromhttp://www.cms.hhs.gov/transmittals/downloads/R51SOMA.pdf

15.

CMS Manual System. (2009b). Pub. 100-07 State Operations. Revisions to Appendix PP—Interpretive Guidelines for Long-Term Care Facilities, “Tag F441.” Department of Health & Human Services (DHHS). Transmittal 51. September. Retrieved August 25, 2011, from http://www.cms.hhs.gov/transmittals/downloads/R51SOMA.pdf

16.

Comondore

Devereaux

Zhou

Stone

Busse

Ravindran

. . . Guyatt

(2009). Quality of care in for-profit and not-for-profit nursing homes: Systematic review and meta-analysis. British Medical Journal, 339, b2732. doi: 10.1136./bmj.b2732

17.

Department of Health and Human Services (DHHS) (2004). Register. HFS 132.45: The author, Wisconsin.

18.

Fendler

E. J.

Ali

Hammond

B. S.

Lyons

M. K.

Kelley

M. B.

Vowell

N. A.

(2002). The impact of alcohol hand sanitizer use on infection rates in an extended care facility. American Journal of Infection Control, 30, 226-233.

19.

Goldrick

B. A.

(1999). Infection control programs in skilled nursing long-term care facilities: An assessment, 1995. American Journal of Infection Control, 27, 4-9.

20.

Grabowski

D. C.

Feng

Intrator

Mor

(2004). Recent trends in state nursing home payment policies. Health Affairs, W4, 363-373.

21.

Hsiao

Mountain

D. C.

Ho-Illman

(1995). Bayesian integration of end-use metering and conditional demand analysis. Journal of Business and Economic Statistics, 13, 315-326.

22.

T.-W.

(1990). Impact of urinary incontinence on healthcare costs. Journal of the American Geriatrics Society, 38, 292-295.

23.

Huang

T. T.

S. C.

(2008). Evaluation of a training programme on knowledge and compliance of nurse assistants’ hand hygiene in nursing homes. Journal of Hospital Infection, 68, 164-170.

24.

Hyer

Temple

Johnson

C. E.

(2009). Florida’s efforts to improve quality of nursing home care through nurse staffing standards, regulation, and Medicaid reimbursement. Journal of Aging and Social Policy, 21, 318-337.

25.

Juthani-Mehta

Quagliarello

V. J.

(2010). Infectious diseases in the nursing home setting: Challenges and opportunities for clinical investigation. Aging and Infectious Diseases, 51, 931-936.

26.

Kash

B. A.

Hawes

Phillips

C. D.

(2007). Comparing staffing levels in the Online Survey Certification and Reporting (OSCAR) system with the Medicaid Cost Report data: Are differences systematic? Gerontologist, 47, 480-489.

27.

Kash

B. A.

Naufal

G. S.

Cortés

Johnson

C. E.

(2010). Exploring factors associated with turnover among registered nurse (RN) supervisors in nursing homes. Journal of Applied Gerontology, 29, 107-127.

28.

Laustsen

Bibby

B. M.

Kristensen

Møller

J. K.

Thulstrup

A. M.

(2009). E-learning may improve adherence to alcohol-based hand rubbing: A cohort study. American Journal of Infection Control, 37, 565-568.

29.

Leinbach

R. M.

English

A. J.

(1995). Training needs of infection control professionals in long-term care facilities in Virginia. American Journal of Infection Control, 23, 73-77.

30.

Mackenzie

James

I. A.

Smith

Barnard

Robinson

(2008). Assessing hand hygiene in older people’s care settings. Nursing Times, 104, 30-31.

31.

Mody

(2009). Infection control programs in nursing homes. In Norman

Yoshikawa

(Eds.), Infectious Disease in the Aging: A Clinical Handbook (pp. 409-422). Humana Press, New York City, NY.

32.

Mody

Bradley

S. F.

Galecki

Olmsted

R. N.

Fitzgerald

J. T.

Kauffman

C. A.

. . . Krein

S. L.

(2011). Conceptual model for reducing infections and antimicrobial resistance in skilled nursing facilities: Focusing on residents with indwelling devices. Aging and Infectious Diseases, 52, 654–661.

33.

Mody

Langa

K. M.

Saint

Bradley

S. F.

(2005). Preventing infections in nursing homes: A survey of infection control practices in southeast Michigan. American Journal of Infection Control, 33, 489-492.

34.

Otero

R. B.

(1993). Current approaches to infection control—Nursing care. Nursing Homes. Find Articles.com. Retrieved January 14, 2010, from http://findarticles.com/

35.

Pan

Domenighini

Signorini

Assini

Catenazzi

Lorenzotti

. . . Guerrini

(2008). Adherence to hand hygiene in an Italian long-term care facility. American Journal of Infection Control, 36, 495-497.

36.

Pittet

Donaldson

(2005). Clean care is safer care: The first global challenge of the WHO World Alliance for Patient Safety. American Journal of Infection Control, 33, 476-479.

37.

Pittet

Hugonnet

Harbarth

Mourouga

V. S.

Sauvan

Touveneau

, . . . Members of the Infection Control Programme. (2000). Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. Lancet, 356, 1307-1312.

38.

Rao

G. G.

Jeanes

Russell

Wilson

Atere-Roberts

O’Sullivan

Donaldson

(2009). Effectiveness of short-term, enhanced, infection control support in improving compliance with infection control guidelines and practice in nursing homes: A cluster randomized trial. Epidemiology of Infections, 137, 1465-1471.

39.

Richards

(2002). Infections in residents of long-term care facilities: An agenda for research report of an expert panel. Journal of the American Geriatrics Society, 20, 570-576.

40.

Richards

C. L.

(2007). Infection control in long-term care facilities. Journal of the American Medical Directors Association, 8, S18-S25.

41.

Rummukainen

Jakobsson

Karppi

Kautiainen

Lyytikäinen

(2009). Promoting hand hygiene and prudent use of antimicrobials in long-term care facilities. American Journal of Infection Control, 37, 168-171.

42.

Rust

T. B.

Wagner

L. M.

Hoffman

Rowe

Neumann

(2008). Broadening the patient safety agenda to include safety in long term care [Special Issue]. Healthcare Quarterly, 11, 31-34.

43.

SAS Institute (1999). Multicollinearity in Logistic Regression. Retrieved from www.uky.edu/ComputingCenter/SSTARS/MulticollinearityinLogisticRegression.htm

44.

Smith

Carusone

S. C.

Loeb

(2008). Hand hygiene practices of health care workers in long-term care facilities. American Journal of Infection Control, 36, 492-494.

45.

World Health Organization (WHO). (2009). First Global Patient Safety challenge Clean Care Is Safer Care. WHO Guidelines on Hand Hygiene in Health Care. WHO Press, Switzerland.

46.

Zeger

S. L.

Liang

K. Y.

(1992). An overview of methods for the analysis of longitudinal data. Statistics in Medicine,11, 1825-1839.