Support from Work and Intent to Stay Among Nurses During COVID-19: An Academic-Practice Collaboration

The COVID-19 pandemic created stressful working conditions for nurses and challenges for leaders. A survey was conducted among 399 acute and ambulatory care nurses measuring availability of calming and safety resources, perceptions of support from work, and intent to stay. Most nurses reported intent to stay with their employer, despite inadequate safety and calming resources. High levels of support from work were significantly influenced nurses' intent to stay. Leadership actions at the study site to provide support are described, providing context for results. Nurse leaders can positively influence intent to stay through consistent implementation of supportive measures.

Unit-level variation in bed alarm use in US hospitals

Bed and chair alarms are widely used in hospitals, despite lack of effectiveness and unintended negative consequences. In this cross-sectional, observational study, we examined alarm prevalence and contributions of patient- and unit-level factors to alarm use on 59 acute care nursing units in 57 US hospitals participating in the National Database of Nursing Quality Indicators®. Nursing unit staff reported data on patient-level fall risk and fall prevention measures for 1,489 patients. Patient-level propensity scores for alarm use were estimated using logistic regression. Expected alarm use on each unit, defined as the mean patient propensity-for-alarm score, was compared with the observed rate of alarm use. Over one-third of patients assessed had an alarm in the "on" position. Patient characteristics associated with higher odds of alarm use included recent fall, need for ambulation assistance, poor mobility judgment, and altered mental status. Observed rates of unit alarm use ranged from 0% to 100% (median 33%, 10th percentile 5%, 90th percentile 67%). Expected alarm use varied less (median 31%, 10th percentile 27%, and 90th percentile 45%). Only 29% of variability in observed alarm use was accounted for by expected alarm use. Unit assignment was a stronger predictor of alarm use than patient-level fall risk variables. Alarm use is common, varies widely across hospitals, and cannot be fully explained by patient fall risk factors; alarm use is driven largely by unit practices. Alarms are used too frequently and too indiscriminately, and guidance is needed for optimizing alarm use to reduce noise and encourage mobility in appropriate patients.

Impact of the Hospital-Acquired Conditions Initiative on Falls and Physical Restraints: A Longitudinal Study

BACKGROUND

The Centers for Medicare & Medicaid Services (CMS) implemented the Hospital-Acquired Conditions (HACs) Initiative in October 2008; the CMS no longer reimbursed hospitals for fall injury. The effects of this payment change on fall and fall injury rates are not well described, nor its effect on physical restraint use.

OBJECTIVE

The aim of this study was to examine the effects of the 2008 HACs Initiative on the rates of falls, injurious falls, and physical restraint use.

DESIGN/SETTING

This was a nine-year retrospective cohort study (July 2006-December 2015) involving 2,862 adult medical, medical-surgical, and surgical nursing units from 734 hospitals.

MEASUREMENTS

Annual rates of change in falls, injurious falls, and physical restraint use during the two years before the payment rule went into effect were compared with one-, four-, and seven-year rates of annual change after implementation, adjusting for unit- and facility-level covariates. Stratified analyses were conducted according to bed size and teaching status.

RESULTS

Compared with prior to the payment change, there was stable acceleration in the one-, four-, and seven-year annual rates of decline in falls as follows: -2.1% (-3.3%, -0.9%), -2.2% (-3.2%, -1.1%), and -2.2% (-3.4%, -1.0%) respectively. For injurious falls, there was an increasing acceleration in the annual declines, achieving statistical significance only at seven years post CMS change as follows: -3.2% (-5.5%, -1.0%). Physical restraint use prevalence decreased from 1.6% to 0.6%. Changes in the rates of falls, injurious falls, and restraint use varied according to hospital bed size and teaching status.

CONCLUSIONS AND RELEVANCE

Since the HACs Initiative, there was at best a modest decline in the rates of falls and injurious falls observed primarily in larger, major teaching hospitals. An increase in restraint use was not observed. Falls remain a difficult patient safety problem for hospitals, and further research is required to develop cost-effective, generalizable strategies for their prevention.

Fall Rates in Urban and Rural Nursing Units: Does Location Matter?

Patient falls remain a leading adverse event in hospitals. In a study of 65 rural hospitals with 222 nursing units and 560 urban hospitals with 4274 nursing units, we found that geographic region, unit type, and nurse staffing, education, experience, and outcomes were associated with fall rates. Implications include specific attention to fall prevention in rehabilitation units, creating better work environments that promote nurse retention, and provide RN-BSN educational opportunities.

Trends in Nursing Care Efficiency From 2007 to 2011 on Acute Nursing Units

Improving the efficiency of health care is a national priority. The purpose of this study was to estimate trends in the efficiency of nursing care. Specifically, the baseline and rate of change in efficiency in the association between select hospital and nursing unit characteristics (e.g., nurse staffing levels) and indicators of patient safety (e.g., fall rates and hospital-acquired pressure ulcer rates) was investigated. A small but significant improvement in efficiency for non-Magnet® hospitals and units with increased RN hours per patient day was found. Trends in efficiency varied by unit type, with medical units showing the greatest improvement. In general, efficiency improved most in health care settings having the greatest opportunity for improvement.

Nurse staffing and patient outcomes: a longitudinal study on trend and seasonality

BACKGROUND

Time trends and seasonal patterns have been observed in nurse staffing and nursing-sensitive patient outcomes in recent years. It is unknown whether these changes were associated.

METHODS

Quarterly unit-level nursing data in 2004-2012 were extracted from the National Database of Nursing Quality Indicators® (NDNQI®). Units were divided into groups based on patterns of missing data. All variables were aggregated across units within these groups and analyses were conducted at the group level. Patient outcomes included rates of inpatient falls and hospital-acquired pressure ulcers. Staffing variables included total nursing hours per patient days (HPPD) and percent of nursing hours provided by registered nurses (RN skill-mix). Weighted linear mixed models were used to examine the associations between nurse staffing and patient outcomes at trend and seasonal levels.

RESULTS

At trend level, both staffing variables were inversely associated with all outcomes (p < 0.001); at seasonal level, total HPPD was inversely associated (higher staffing related to lower event rate) with all outcomes (p < 0.001) while RN skill-mix was positively associated (higher staffing related to higher event rate) with fall rate (p < 0.001) and pressure ulcer rate (p = 0.03). It was found that total HPPD tended to be lower and RN skill-mix tended to be higher in Quarter 1 (January-March) when falls and pressure ulcers were more likely to happen.

CONCLUSIONS

By aggregating data across units we were able to detect associations between nurse staffing and patient outcomes at both trend and seasonal levels. More rigorous research is needed to study the underlying mechanism of these associations.

An Exploration of Nursing Home Organizational Processes

The purpose of this study is to explore the relationship between nursing home staffs' perceptions of organizational processes (communication, teamwork, and leadership) with characteristics (turnover, tenure, and educational preparation) of the nursing home administrator (NHA) and director of nursing (DON). NHAs and DONs rate communication, teamwork, and leadership significantly higher than direct care staff do (registered nurses, licensed practical nurses, certified nurse aides [CNAs]). CNAs have the lowest ratings of communication and teamwork. Turnover of the NHA and DON is significantly and negatively associated with communication and teamwork. Two thirds of DONs surveyed hold less than a baccalaureate degree; this does not influence staffs' ratings of communication, teamwork, and leadership. Findings from this study highlight the need to explore differences in perceptions between administrative and direct care staff and how these may or may not influence staff development and quality improvement activities in nursing homes.

Comparison of Unit-Level Patient Turnover Measures in Acute Care Hospital Settings

High patient turnover is a critical factor increasing nursing workload. Despite the growing number of studies on patient turnover, no consensus about how to measure turnover has been achieved. This study was designed to assess the correlation among patient turnover measures commonly used in recent studies and to examine the degree of agreement among the measures for classifying units with different levels of patient turnover. Using unit-level data collected for this study from 292 units in 88 hospitals participating in the National Database of Nursing Quality Indicators®, we compared four patient turnover measures: the inverse of length of stay (1/LOS), admissions, discharges, and transfers per daily census (ADTC), ADTC with short-stay adjustment, and the number of ADTs and short-stay patients divided by the total number of treated patients, or Unit Activity Index (UAI). We assessed the measures' agreement on turnover quartile classifications, using percent agreement and Cohen's kappa statistic (weighted and unweighted). Pearson correlation coefficients also were calculated. ADTC with or without adjustment for short-stay patients had high correlations and substantial agreement with the measure of 1/LOS (κ = .62 to .91; r = .90 to .95). The UAI measure required data less commonly collected by participating hospital units and showed only moderate correlations and fair agreement with the other measures (κ = .23 to .39; r = .41 to .45). The UAI may not be comparable and interchangeable with other patient turnover measures when data are obtained from multiple units and hospitals. © 2016 Wiley Periodicals, Inc.

Effect of Medicare's nonpayment for Hospital-Acquired Conditions: lessons for future policy

IMPORTANCE

In 2008, Medicare implemented the Hospital-Acquired Conditions (HACs) Initiative, a policy denying incremental payment for 8 complications of hospital care, also known as never events. The regulation's effect on these events has not been well studied.

OBJECTIVE

To measure the association between Medicare's nonpayment policy and 4 outcomes addressed by the HACs Initiative: central line-associated bloodstream infections (CLABSIs), catheter-associated urinary tract infections (CAUTIs), hospital-acquired pressure ulcers (HAPUs), and injurious inpatient falls.

DESIGN, SETTING, AND PARTICIPANTS

Quasi-experimental study of adult nursing units from 1381 US hospitals participating in the National Database of Nursing Quality Indicators (NDNQI), a program of the American Nurses Association. The NDNQI data were combined with American Hospital Association, Medicare Cost Report, and local market data to examine adjusted outcomes. Multilevel models were used to evaluate the effect of Medicare's nonpayment policy on never events.

EXPOSURES

United States hospitals providing treatment for Medicare patients were subject to the new payment policy beginning in October 2008.

MAIN OUTCOMES AND MEASURES

Changes in unit-level rates of HAPUs, injurious falls, CLABSIs, and CAUTIs after initiation of the policy.

RESULTS

Medicare's nonpayment policy was associated with an 11% reduction in the rate of change in CLABSIs (incidence rate ratio [IRR], 0.89; 95% CI, 0.83-0.95) and a 10% reduction in the rate of change in CAUTIs (IRR, 0.90; 95% CI, 0.85-0.95), but was not associated with a significant change in injurious falls (IRR, 0.99; 95% CI, 0.99-1.00) or HAPUs (odds ratio, 0.98; 95% CI, 0.96-1.01). Consideration of unit-, hospital-, and market-level factors did not significantly alter our findings.

CONCLUSIONS AND RELEVANCE

The HACs Initiative was associated with improvements in CLABSI and CAUTI trends, conditions for which there is strong evidence that better hospital processes yield better outcomes. However, the HACs Initiative was not associated with improvements in HAPU or injurious fall trends, conditions for which there is less evidence that changing hospital processes leads to significantly better outcomes.

Reliability and Validity of the NDNQI® Injury Falls Measure

Although remarkable efforts have been made to improve patient fall reporting through the utilization of standardized definitions, injury falls reporting has rarely been examined. This study used an overall intra-class correlation coefficient (ICC) estimate and factor analysis to assess the reliability and validity of the National Database of Nursing Quality Indicators® (NDNQI®) falls with injury measure. Data were collected from an online Fall Injury Level Survey that was administered to 1,159 NDNQI site coordinators (39.7% response rate; 91% registered nurses [RNs]). Estimated overall ICC was .85. Exploratory factor analysis (EFA) with a Promax rotation (root mean square error of approximation [RMSEA] = 0.053) identified three latent factors: No Injury, Minor Injury, and Moderate/Major Injuries. Final confirmatory factor analysis (CFA) assessment (comparative fit index [CFI] = 0.914, Tucker Lewis Index [TLI] = 0.910, RMSEA = 0.048) confirmed an acceptable model fit. Results provided strong evidence that the NDNQI falls with injury measure is reliable and valid in supporting hospitals' fall prevention efforts and future injurious falls research.

Pressure ulcers and prevention among acute care hospitals in the United States

BACKGROUND

Most pressure ulcers can be prevented with evidence-based practice. Many studies describe the implementation of a pressure ulcer prevention program but few report the effect on outcomes across acute care facilities.

METHODS

Data on hospital-acquired pressure ulcers and prevention from the National Database of Nursing Quality Indicators 2010 Pressure Ulcer Surveys were linked to hospital characteristics and nurse staffing measures within the data set. The sample consisted of 1,419 hospitals from across the United States and 710,626 patients who had been surveyed for pressure ulcers in adult critical care, step-down, medical, surgical, and medical/surgical units. Hierarchical logistic regression analysis was performed to identify study variables associated with hospital-acquired pressure ulcers among patients at risk for these ulcers.

RESULTS

The rate of hospital-acquired pressure ulcers was 3.6% across all surveyed patients and 7.9% among those at risk. Patients who received a skin and pressure ulcer risk assessment on admission were less likely to develop a pressure ulcer. Additional study variables associated with lower hospital-acquired pressure ulcer rates included a recent reassessment of pressure ulcer risk, higher Braden Scale scores, a recent skin assessment, routine repositioning, and Magnet or Magnet-applicant designation. Variables associated with a higher likelihood of hospital-acquired pressure ulcers included nutritional support, moisture management, larger hospital size, and academic medical center status.

CONCLUSIONS

Results provide empirical support for pressure ulcer prevention guideline recommendations on skin assessment, pressure ulcer risk assessment, and routine repositioning, but the 7.9% rate of hospital-acquired pressure ulcers among at-risk patients suggests room for improvement in pressure ulcer prevention practice.

Associations between rates of unassisted inpatient falls and levels of registered and non-registered nurse staffing

OBJECTIVE

To enhance understanding of how nurse staffing relates to unassisted falls by exploring non-linear associations between unassisted fall rates and levels of registered nurse (RN) and non-RN staffing on 5 nursing unit types, thereby enabling managers to improve patient safety by making better-informed decisions about staffing.

DESIGN

Cross-sectional analysis of routinely collected data using hierarchical negative binomial regression.

SETTING

8069 nursing units in 1361 U.S. hospitals participating in the National Database of Nursing Quality Indicators(®). Main outcome measure Rate of unassisted falls per inpatient day.

RESULTS

Associations between unassisted fall rates and nurse staffing varied by unit type. For medical-surgical units, higher RN staffing was weakly associated with lower fall rates. On step-down and medical units, the association between RN staffing and fall rates depended on the level of staffing: At lower staffing levels, the fall rate increased as staffing increased, but at moderate and high staffing levels, the fall rate decreased as staffing increased. Higher levels of non-RN staffing were generally associated with higher fall rates..

CONCLUSIONS

Increasing non-RN staffing seems ineffective at preventing unassisted falls. Increasing RN staffing may be effective, depending on the unit type and the current level of staffing.

A Standardized measure: NDNQI nursing care hours indicator

Although nursing care hours (NCH) is commonly used to examine factors related to adverse events among inpatients, the reliability of the NCH measure has rarely been examined. This study assessed the reliability of NCH data from the National Database of Nursing Quality Indicators® by estimating intraclass correlation coefficients (ICCs) with data from the California Office for Statewide Health Planning and Development. Hospital-level aggregated NCH data for critical care units were linked from each of the databases for 48 California hospitals matched in the two databases. Estimated ICCs were .96 for registered nurse (RN) hours and .72 for non-RN hours. Findings provide evidence that NCH data of the national database were substantially reliable for use in national comparable benchmarking reports for hospitals' quality improvement activities and research.

Identifying individual changes in performance with composite quality indicators while accounting for regression to the mean

Almost a decade ago Morton and Torgerson indicated that perceived medical benefits could be due to "regression to the mean." Despite this caution, the regression to the mean "effects on the identification of changes in institutional performance do not seem to have been considered previously in any depth" (Jones and Spiegelhalter). As a response, Jones and Spiegelhalter provide a methodology to adjust for regression to the mean when modeling recent changes in institutional performance for one-variable quality indicators. Therefore, in our view, Jones and Spiegelhalter provide a breakthrough methodology for performance measures. At the same time, in the interests of parsimony, it is useful to aggregate individual quality indicators into a composite score. Our question is, can we develop and demonstrate a methodology that extends the "regression to the mean" literature to allow for composite quality indicators? Using a latent variable modeling approach, we extend the methodology to the composite indicator case. We demonstrate the approach on 4 indicators collected by the National Database of Nursing Quality Indicators. A simulation study further demonstrates its "proof of concept."

Data Envelopment Analysis in the Presence of Measurement Error: Case Study from the National Database of Nursing Quality Indicators® (NDNQI®)

Data Envelopment Analysis (DEA) is the most commonly used approach for evaluating healthcare efficiency (Hollingsworth, 2008), but a long-standing concern is that DEA assumes that data are measured without error. This is quite unlikely, and DEA and other efficiency analysis techniques may yield biased efficiency estimates if it is not realized (Gajewski, Lee, Bott, Piamjariyakul and Taunton, 2009; Ruggiero, 2004). We propose to address measurement error systematically using a Bayesian method (Bayesian DEA). We will apply Bayesian DEA to data from the National Database of Nursing Quality Indicators® (NDNQI®) to estimate nursing units' efficiency. Several external reliability studies inform the posterior distribution of the measurement error on the DEA variables. We will discuss the case of generalizing the approach to situations where an external reliability study is not feasible.

Input data quality control for NDNQI national comparative statistics and quarterly reports: a contrast of three robust scale estimators for multiple outlier detection

Background

To evaluate institutional nursing care performance in the context of national comparative statistics (benchmarks), approximately one in every three major healthcare institutions (over 1,800 hospitals) across the United States, have joined the National Database for Nursing Quality Indicators® (NDNQI®). With over 18,000 hospital units contributing data for nearly 200 quantitative measures at present, a reliable and efficient input data screening for all quantitative measures for data quality control is critical to the integrity, validity, and on-time delivery of NDNQI reports.

Methods

With Monte Carlo simulation and quantitative NDNQI indicator examples, we compared two ad-hoc methods using robust scale estimators, Inter Quartile Range (IQR) and Median Absolute Deviation from the Median (MAD), to the classic, theoretically-based Minimum Covariance Determinant (FAST-MCD) approach, for initial univariate outlier detection.

Results

While the theoretically based FAST-MCD used in one dimension can be sensitive and is better suited for identifying groups of outliers because of its high breakdown point, the ad-hoc IQR and MAD approaches are fast, easy to implement, and could be more robust and efficient, depending on the distributional property of the underlying measure of interest.

Conclusion

With highly skewed distributions for most NDNQI indicators within a short data screen window, the FAST-MCD approach, when used in one dimensional raw data setting, could overestimate the false alarm rates for potential outliers than the IQR and MAD with the same pre-set of critical value, thus, overburden data quality control at both the data entry and administrative ends in our setting.

Reliability of the nursing care hour measure: a descriptive study

BACKGROUND

The nursing care hour has become an international standard unit of measure in research where nurse staffing is a key variable. Until now, there have been no studies verifying whether nursing care hours obtained from hospital data sources can be collected reliably.

OBJECTIVES

To examine the processes used by hospitals to generate nursing care hour data and to evaluate inter-rater reliability and guideline compliance with standards of the National Database of Nursing Quality Indicators(®) (NDNQI(®)) and the National Quality Forum.

DESIGN AND SETTING

Two-phase descriptive study of all NDNQI hospitals that submitted data in third quarter of 2007.

METHODS

Data for phase I came from an online survey created by the authors to ascertain the processes used by hospitals to collect nursing care hours and their compliance with standardized data collection guidelines. In phase II, inter-rater reliability was measured using intra-class correlations between nursing care hours generated from clock hour files submitted to the study team by participants' payroll/accounting departments and aggregated data submitted previously.

RESULTS

Phase I data were obtained from a total of 714 respondents. Nearly half (48%) of all sites use payroll records to obtain nursing care hour data and 70% use one of the standardized methods for converting the bi-weekly hours into months. Unit secretaries were reportedly included in NCH by 17.4% of respondents and only 26.2% of sites could accurately identify the point at which newly hired nurses should be included. The phase II findings (n=11) support the ability of two independent raters to obtain similar results when calculating total nursing care hours according to standard guidelines (ICC=0.76-0.99).

CONCLUSIONS

Although barriers exist, this study found support for hospitals' abilities to collect reliable nursing care hour data.

Assessing Content Validity Through Correlation and Relevance Tools

Content validity elicits expert opinion regarding items of a psychometric instrument. Expert opinion can be elicited in many forms: for example, how essential an item is or its relevancy to a domain. This study developed an alternative tool that elicits expert opinion regarding correlations between each item and its respective domain. With 109 Registered Nurse (RN) site coordinators from National Database of Nursing Quality Indicators®, we implemented a randomized Bayesian equivalence trial with coordinators completing “relevance” or “correlation” content tools regarding the RN Job Enjoyment Scale. We confirmed our hypothesis that the two tools would result in equivalent content information. A Bayesian ordered analysis model supported the results, suggesting that evidence for traditional content validity indices can be justified using correlation arguments.

Hospital and unit characteristics associated with nursing turnover include skill mix but not staffing level: an observational cross-sectional study

BACKGROUND

Nursing turnover is expensive and may have adverse effects on patient care. Little is known about turnover's association with most hospital and nursing unit characteristics, including nurse staffing level and registered nurse skill mix.

OBJECTIVE

To explore associations between nursing unit turnover rates and several hospital- and unit-level variables, including staffing level and skill mix.

DESIGN

Observational cross-sectional study of longitudinal data.

SETTINGS

1884 nursing units in 306 U.S. acute care hospitals.

METHODS

During a 2-year period units reported monthly data on staffing and turnover. Total nursing staff turnover and registered nurse turnover rates were modeled as dependent variables in hierarchical Poisson regression models. The following hospital characteristics were considered as predictors: Magnet(®) status, ownership (government or non-government), teaching status, locale (metropolitan, micropolitan, or rural), and size (average daily census). The U.S. state in which the hospital was located was included as a covariate. Unit-level variables included total nursing hours per patient day, size of nursing staff, registered nurse skill mix, population age group (neonatal, pediatric, or adult), and service line (critical care, step-down, medical, surgical, medical/surgical, psychiatric, or rehabilitation).

RESULTS

Government ownership, Magnet designation, and higher skill mix were associated with lower total turnover and registered nurse turnover. Neonatal units had lower total and registered nurse turnover than pediatric units, which had lower total and registered nurse turnover than adult units. Unit service line was associated only with total turnover. Psychiatric, critical care, and rehabilitation units had the lowest mean turnover rates, but most differences between service lines were not significant. The other explanatory variables considered were not significant.

CONCLUSIONS

Several hospital and unit characteristic variables have significant associations with nursing turnover; these associations should be taken into account in nursing turnover research and need to be explored further. Controlling for hospital ownership, Magnet status, unit service line, and unit population age group, registered nurse skill mix is apparently more important than total nurse staffing level in predicting nursing turnover.

The Box-Cox power transformation on nursing sensitive indicators: does it matter if structural effects are omitted during the estimation of the transformation parameter?

Background

Many nursing and health related research studies have continuous outcome measures that are inherently non-normal in distribution. The Box-Cox transformation provides a powerful tool for developing a parsimonious model for data representation and interpretation when the distribution of the dependent variable, or outcome measure, of interest deviates from the normal distribution. The objectives of this study was to contrast the effect of obtaining the Box-Cox power transformation parameter and subsequent analysis of variance with or without a priori knowledge of predictor variables under the classic linear or linear mixed model settings.

Methods

Simulation data from a 3 × 4 factorial treatments design, along with the Patient Falls and Patient Injury Falls from the National Database of Nursing Quality Indicators (NDNQI^®) for the 3^rd quarter of 2007 from a convenience sample of over one thousand US hospitals were analyzed. The effect of the nonlinear monotonic transformation was contrasted in two ways: a) estimating the transformation parameter along with factors with potential structural effects, and b) estimating the transformation parameter first and then conducting analysis of variance for the structural effect.

Results

Linear model ANOVA with Monte Carlo simulation and mixed models with correlated error terms with NDNQI examples showed no substantial differences on statistical tests for structural effects if the factors with structural effects were omitted during the estimation of the transformation parameter.

Conclusions

The Box-Cox power transformation can still be an effective tool for validating statistical inferences with large observational, cross-sectional, and hierarchical or repeated measure studies under the linear or the mixed model settings without prior knowledge of all the factors with potential structural effects.

Midnight census revisited: Reliability of patient day measurements in US hospital units

BACKGROUND

Patient days are widely used in nurse staffing research and for nursing quality measurement. Nursing hours per patient day (NHPPD) and fall rates incorporate patient days in the denominator and are endorsed by the US National Quality Forum (NQF) as nursing sensitive consensus measures. Measurement error introduced by patient days would affect the accuracy of these nursing quality indicators.

OBJECTIVES

The aim of this study was to assess the reliability of five patient day reporting methods accepted by the National Database of Nursing Quality Indicators (NDNQI). The specific aims were (1) to investigate the agreement of five patient day measurements with a defined quasi-gold standard, (2) to explore method bias by investigating the association of potential confounding variables with the differences between the routine measurements and the quasi-gold standard, and (3) to extrapolate the potential effect of bias of the patient day methods on nursing quality indicators.

DESIGN

A multiple census study with a national convenience sample of hospital units in the US was conducted.

SETTING

260 out of 282 units (92%) from 54 hospitals sent bi-hourly patient census data for seven randomly selected days in September 2008.

METHODS

The multiple census data comprised the quasi-gold standard and was compared with data routinely submitted to the database. Intraclass correlations were calculated for an agreement analysis. A Bayesian regression analysis was conducted to explore the impact of different data collection methods and the degree of short stay patients.

RESULTS

Overall agreement between routine data and the quasi-gold standard was excellent (ICC [95% CI]: 0.967 [0.958-0.974]). A Bayesian regression analysis identified that two methods underestimated patient days and an interaction between the degrees of short stay patients and one of the data collection methods also affected patient day measurement by up to 7.6%.

Indoor smoking ordinances in workplaces and public places in Kansas

INTRODUCTION

The purpose of this study was to investigate the preferences of elected city officials regarding the need for a statewide clean indoor air law and to analyze the content of local smoking ordinances.

METHODS

A survey of elected officials in 57 larger Kansas cities obtained information on the perceived need for statewide legislation, venues to be covered, and motivating factors. Clean indoor air ordinances from all Kansas cities were analyzed by venue.

RESULTS

The survey response rate was 190 out of 377 (50.4%) for elected officials. Over 70% of the respondents favored or strongly favored greater restrictions on indoor smoking. Sixty percent favored statewide legislation. Among these, over 80% favored restrictions in health care facilities, theaters, indoor sports arenas (including bowling alleys), restaurants, shopping malls, lobbies, enclosed spaces in outdoor arenas, and hotel/motel rooms. Officials who had never smoked favored a more restrictive approach. Employee and public health concerns were cited as influential by 76%-79% of respondents. Thirty-eight ordinances, covering over half of the state's population, were examined. They varied considerably in their exemptions.

DISCUSSION

Official's attitudes toward smoking regulations were associated with their smoking status. The examination of existing ordinances revealed a piecemeal approach to smoking regulations.

A multilevel confirmatory factor analysis of the Practice Environment Scale: a case study

BACKGROUND

Practice Environment Scale (PES) data are collected from RNs in nursing units in hospitals that are members of the National Database of Nursing Quality Indicators (NDNQI). Patient and RN information are collected to aid in quality improvement and research at the nursing unit level. The data were collected from the individual RN, but items are worded so that analyses can be conducted at the individual, unit, or hospital level. There is a need to examine the validity of the PES at both the individual and the unit level.

OBJECTIVE

To describe multilevel confirmatory factor analysis via a case study for investigating the validity of the PES, a measure of the nursing practice environment.

APPROACH

The PES was administered to 72,889 RNs from 4,783 nursing units (16 unit types; e.g., critical care and obstetric) in 2007. The PES has 31 items in five different domains. A multilevel confirmatory factor analytic model was fit with a structure on the basis of the five domains. From this model, an estimate was sought between unit loadings and within unit loadings to investigate factorial, convergent, and discriminant validity at both the unit and the RN levels. To investigate criterion-related validity, the five PES domains were correlated with the seven job enjoyment items adapted from the National Database of Nursing Quality Indicators at the unit and RN levels (also using a multilevel model).

RESULTS

The multilevel factor analysis provides evidence of factorial, convergent, discriminant, and criterion-related validity at both the unit and the RN levels.

DISCUSSION

The PES is a valid instrument for use in quality improvement and research both at the unit and individual RN levels.

Evaluation of the National Database of Nursing Quality Indicators (NDNQI) Training Program on Pressure Ulcers

The National Database of Nursing Quality Indicators (NDNQI) Pressure Ulcer Training Program was developed to improve nursing accuracy and reliability in identifying and staging pressure ulcers and differentiating hospital- and unit-acquired from community-acquired pressure ulcers. Of 5,200 individuals who completed the training program within 5 months of release, 937 provided written evaluation comments. Four major themes emerged from content analysis of the evaluation remarks: pressure ulcer and other wound pictures; accuracy and clarity of content; program design and technology; and the educational/informational experience. Reviewers most frequently commented on their positive learning experience. Program components that enhanced the educational experience included the array of pressure ulcer pictures at each stage and pictures of other wounds. Clarity of content, program design, and technological problems were barriers to reviewer satisfaction. Findings suggest the NDNQI Pressure Ulcer Training Program was effective for educating staff nurses on pressure ulcer identification and staging.

Improving quality indicator report cards through Bayesian modeling

Background

The National Database for Nursing Quality Indicators^® (NDNQI^®) was established in 1998 to assist hospitals in monitoring indicators of nursing quality (eg, falls and pressure ulcers). Hospitals participating in NDNQI transmit data from nursing units to an NDNQI data repository. Data are summarized and published in reports that allow participating facilities to compare the results for their units with those from other units across the nation. A disadvantage of this reporting scheme is that the sampling variability is not explicit. For example, suppose a small nursing unit that has 2 out of 10 (rate of 20%) patients with pressure ulcers. Should the nursing unit immediately undertake a quality improvement plan because of the rate difference from the national average (7%)?

Methods

In this paper, we propose approximating 95% credible intervals (CrIs) for unit-level data using statistical models that account for the variability in unit rates for report cards.

Results

Bayesian CrIs communicate the level of uncertainty of estimates more clearly to decision makers than other significance tests.

Conclusion

A benefit of this approach is that nursing units would be better able to distinguish problematic or beneficial trends from fluctuations likely due to chance.

Inter-rater reliability of pressure ulcer staging: ordinal probit Bayesian hierarchical model that allows for uncertain rater response

This article describes a method for estimating the inter-rater reliability of pressure ulcer (PU) staging (stages I-IV) from raters in National Database of Nursing Quality Indicators (NDNQI) participating hospitals. The method models ordinal spanning data utilizing an ordinal probit Bayesian hierarchical model (BHM) across several hospitals in which raters monitor patient's PUs. An ulcer that cannot be accurately assessed because the base of the wound cannot be seen is defined as unstageable. Our novel approach allows for an unstageable PU rating to be included in the analysis. We compare the ordinal probit BHM to an approximate random-effects (standard approach in the literature) model that assumes that the raw ordinal data are continuous.

Non-normal path analysis in the presence of measurement error and missing data: a Bayesian analysis of nursing homes' structure and outcomes

Path analytic models are useful tools in quantitative nursing research. They allow researchers to hypothesize causal inferential paths and test the significance of these paths both directly and indirectly through a mediating variable. A standard statistical method in the path analysis literature is to treat the variables as having a normal distribution and to estimate paths using several least squares regression equations. The parameters corresponding to the direct paths have point and interval estimates based on normal distribution theory. Indirect paths are a product of the direct path from the independent variable to the mediating variable and the direct path of the mediating variable to the dependent variable. However, in the case of non-normal distributions, the point and interval estimates of the indirect path become much more difficult to estimate. We address the issue of calculating indirect point and interval estimates in the case of non-normally distributed data. Our substantive application is a nursing home research problem in which the variables in the path analysis of interest involve variables with normal, Bernoulli, or Poisson distributions. Additionally, one of the Poisson variables is observed with error. This paper addresses estimating point and interval estimation of indirect paths for variables with non-normal distributions in the presence of missing data and measurement error. We handle these difficulties from a fully Bayesian point of view. We present our substantive path analysis motivated from a nursing home structure, process, and outcomes model. Our results focus on the impact job turnover in the nursing homes has on nursing home outcomes.

Summarizing benchmarks in the national database of nursing quality indicators using bootstrap confidence intervals

When summarizing the benchmarks for nursing quality indicators with confidence intervals around the means, bounds too high or too low are sometimes found due to small sample size or violation of the normality assumption. Transforming the data or truncating the confidence intervals at realistic values can solve the problem of out of range values. However, truncation does not improve upon the non-normality of the data, and transformations are not always successful in normalizing the data. The percentile bootstrap has the advantage of providing realistic bounds while not relying upon the assumption of normality and may provide a convenient way of obtaining appropriate confidence intervals around the mean for nursing quality indicators.

Unit type differences in RN workgroup job satisfaction

Using cross-sectional data from the 2004 National Database of Nursing Quality Indicators (NDNQI) RN Satisfaction Survey, differences in RN workgroup job satisfaction were examined among 10 unit types--medical-surgical, step-down, critical care, pediatric, maternal-newborn, psychiatric, emergency department, rehabilitation, surgical services, and outpatient clinics and labs. The national sample included RN workgroups in 2,900 patient care units (55,516 RNs; 206 hospitals in 44 states). Workgroup satisfaction across all unit types was moderate. RN workgroups in pediatric units were the most satisfied, whereas those in surgical services and emergency department unit types were least satisfied. A consistent finding across all unit types was high satisfaction with the specific domains of nurse-to-nurse interaction, professional status, and professional development versus much lower satisfaction with task, decision making, and pay. Findings can be used to inform and develop investigations that examine specific aspects of the work environment for RN workgroups in various unit types.

Reliability Testing of the National Database of Nursing Quality Indicators Pressure Ulcer Indicator

A criterion-referenced Web-based test was designed and administered to 256 individuals at 48 randomly sampled National Database of Nursing Quality Indicators (NDNQI) member hospitals to determine the reliability of the NDNQI pressure ulcer indicator. Overall kappa values for pressure ulcer identification, staging, and sourcing indicate moderate to near perfect reliability. Findings suggest that nurses can accurately differentiate pressure ulcers from other ulcerous wounds in Web-based photographs, reliably stage pressure ulcers, and reliably identify community versus nosocomial pressure ulcers.

Inter-rater reliability of nursing home surveys: a Bayesian latent class approach

In the U.S., federal and state governments perform routine inspections of nursing homes. Results of the inspections allow government to generate fines for findings of non-compliance as well as allow consumers to rank facilities. The purpose of this study is to investigate the inter-rater reliability of the nursing home survey process. In general, the survey data involves 191 binary deficiency variables interpreted as 'deficient' or 'non-deficient'. To reduce the dimensionality of the problem, our proposed method involves two steps. First, we reduce the deficiency categories to sub-categories using previous nursing home studies. Second, looking at the State of Kansas specifically, we take the deficiency data from 1 year, and use Bayesian latent class analysis (LCA) to collapse the sub-categories to a binary variable. We evaluate inter-rater agreement using deficiency data from two separate survey teams on one facility, a matched-pair design. We evaluate the agreement of the two raters on binary data using the weights from the LCA. This allows a two-by-two contingency analysis using a Bayesian beta-binomial model. We elicit informative prior distributions from the nursing home providers. Together, with the experimental data, this provides a posterior distribution of the kappa agreement of the raters for nursing home deficiency citation data.

The NDNQI-Adapted Index of work satisfaction

The valid measurement of nurses' job satisfaction is critical because job satisfaction is important for the retention of qualified nurses to provide patient care in hospitals. Two studies were conducted to adapt the Stamps Index of Work Satisfaction (1997b) to measure work satisfaction at the patient care unit level for use by the National Database of Nursing Quality Indicators (NDNQI). In Study 1 (n = 918 RNs) exploratory factor analysis of data obtained using the NDNQI-Adapted Index replicated the conceptual dimensions of the Stamps measure. Associations with scores on Job Enjoyment were evidence that the Index measured the intended construct. Using theta, the reliability of the composite subscales was .91. The adapted Work Satisfaction subscale scores explained 46% of the variance in Job Enjoyment, with each subscale contributing uniquely (p < .001). In Study 2 (n = 2277 RNs) confirmatory factor analysis using structural equation modeling supported the 7-subscale structure for the Adapted Index (CFI [719] = .88; RMR = .05). Replication of associations between scores on the Index subscales and Job Enjoyment provided further evidence regarding validity of the data, since the Work Satisfaction subscales explained 56% of the variance in Job Enjoyment. The feasibility of using an on-line version of the Adapted-Index for data collection was demonstrated. The findings from the two studies indicate that the adapted Index of Work Satisfaction has a structure similar to the original instrument and is a reliable and valid measure of work satisfaction at the patient care unit level.

Herbal and Vitamin/Mineral Supplement Use by Retirement Community Residents

A survey questionnaire assessed supplement use by free-living residents of a retirement community. Of the 318 respondents (mean age 82.2 years), 20% of women and 20% of men reported using herbal supplements, with 62% of these using them at least once per week. Most herbal supplements users (97%) also used vitamin/mineral supplements. Sixty-eight percent of herbal supplement users felt very much/somewhat informed about taking the supplements. Forty-four percent, however, were not sure whether there was testing before marketing; 33% were not sure about side effects. Half (52%) relied "very much/somewhat" on doctors/nurses as information sources; 40% relied on dietitians.