Journal of Gerontological Nursing

Feature Article 

Examining the Impact of Board-Certified Registered Nurses in Skilled Nursing Facilities Using National and State Quality and Clinical Indicators

Heidi Keeler, PhD, RN; Rosa R. Baier, MPH; Cyrus Kosar, MA; Beth Culross, PhD, RN, GCNS-BC; Mary E. Cramer, PhD, RN, FAAN

Abstract

A convenience sample of skilled nursing facilities was selected from a sample of graduates of an online training program for RNs who subsequently achieved board certification in gerontological nursing (RN-BC). Facilities that employed one or more RN-BC were pair-matched using 11 organizational characteristics with facilities that did not employ a RN-BC. Facility data were collected at two time points, and differences between time points and between facility type (RN-BC versus non-RN-BC) were analyzed. Findings showed that there were no statistically significant differences between RN-BC and non-RN-BC facilities with respect to quality ratings and nurse sensitive clinical indicators (e.g., restraint use, urinary tract infections, falls, antipsychotic medication use) between the two time periods; however, in the second time period, RN-BC facilities showed greater improvement versus non-RN-BC facilities in seven of nine outcomes, achieving significance in Overall (4.10 vs. 3.55, p < 0.01) and Survey (3.48 vs. 2.86, p < 0.01) 5-Star ratings. [Journal of Gerontological Nursing, 45(11), 39–45.]

Abstract

A convenience sample of skilled nursing facilities was selected from a sample of graduates of an online training program for RNs who subsequently achieved board certification in gerontological nursing (RN-BC). Facilities that employed one or more RN-BC were pair-matched using 11 organizational characteristics with facilities that did not employ a RN-BC. Facility data were collected at two time points, and differences between time points and between facility type (RN-BC versus non-RN-BC) were analyzed. Findings showed that there were no statistically significant differences between RN-BC and non-RN-BC facilities with respect to quality ratings and nurse sensitive clinical indicators (e.g., restraint use, urinary tract infections, falls, antipsychotic medication use) between the two time periods; however, in the second time period, RN-BC facilities showed greater improvement versus non-RN-BC facilities in seven of nine outcomes, achieving significance in Overall (4.10 vs. 3.55, p < 0.01) and Survey (3.48 vs. 2.86, p < 0.01) 5-Star ratings. [Journal of Gerontological Nursing, 45(11), 39–45.]

The long-term care (LTC) nursing workforce in the United States faces numerous challenges as the percentage of older adults and the concurrent demand for post-acute skilled services grow (Castle & Engberg, 2006; Pruitt, 2013; U.S. Department of Health and Human Services [USDHHS], 2003; USDHHS, Health Resources and Services Administration, Bureau of Health Workforce, & National Center for Health Workforce Analysis, 2017). Skilled nursing facilities (SNFs) are increasingly required to provide post-hospital acute care for patients who experience complex and multiple comorbidities (Thach & Weiner, 2018). Yet, the LTC nursing workforce may not be educationally prepared for implementing evidence-based gerontological best practices in post-acute and skilled care settings (Zarowitz, Resnick, & Ouslander, 2018). A lingering misperception persists that nursing care in the LTC setting is less demanding and requires less “skill” than in a hospital setting. The fact that nurse salaries in hospitals are approximately $10,000 higher than in LTC (Bureau of Labor Statistics, 2018) seems to support this misperception. High rates of turnover add to LTC workforce problems (Banaszak-Holl, Castle, Lin, Shrivastwa, & Spreitzer, 2015), as dissatisfied workers take valuable skills and job knowledge with them on departure.

The SNF setting requires its own unique skill set; thus, an underprepared workforce presents potential risks to quality care and patient safety. The Centers for Medicare & Medicaid Services (CMS, n.d.) recognized the need to improve patient care quality and safety in LTC by increasing nursing education and competencies among the workforce as part of the requirements of participation in its reimbursement programs. The 2016 CMS ruling, and subsequent interpretation, means that LTC leaders need to explore methods for ensuring their nursing workforce is prepared with evidence-based best practice knowledge and skills that will ensure high-quality, patient-centered care (CMS, 2019).

Background

The examination of workforce preparation for nurses reveals a gap in expertise in caring for the complex needs of older adults. Despite the fact that the presence of trained baccalaureate-prepared RNs has a positive impact on outcomes in acute care (Kendall-Gallagher, Aiken, Sloane, & Cimiotti, 2011; Needleman et al., 2011) and LTC (Castle & Anderson, 2011), most RNs employed in LTC are prepared in associate degree programs where the amo unt of gerontological education is lacking or very low (Auerbach, Buerhaus, & Staiger, 2015). Although the need to incorporate gerontological-specific training in formal nursing programs has been acknowledged nationally (Institute of Medicine, 2010), any action taken to address this gap would not impact those who graduated from formal educational programs before such curricula were adopted. Although gerontological content in primary education has been increasing (Berman et al., 2005), the average age of a RN is 51 and the average age of a LPN is 52 (Smiley et al., 2018), which suggests that depending on when primary training occurred, it may not have included gerontological-specific courses or integrated content. This knowledge gap may be lessened using specialty board certification by an accredited body, such as the American Association of Colleges of Nursing, recognized as a “gold” standard for quality preparation in a nursing specialty. Research on nursing specialty certification in acute care has mostly demonstrated a positive impact on patient outcomes (Biel, Grief, Patry, Ponto, & Shirey, 2014; Boyle, 2017; Kendall-Gallagher & Blegen, 2009).

In 2010, the University of Nebraska Medical Center (UNMC) College of Nursing was awarded a grant from the Robert Wood Johnson Foundation (Cramer et al., 2014) to address the educational needs of the LTC nursing workforce. The project involved mostly nurses with an associate degree employed in LTC who participated in a 10-week training course aimed at educationally preparing them for gerontological board certification. Participants had a 98% success rate in achieving board certification and demonstrated higher self-empowerment, enhanced job satisfaction, and improved job competence scores (Cramer et al., 2014). Subsequent to the certification success of the grant, the UNMC collaborated with the American Health Care Association (2012)/National Centers for Assisted Living to transform the project into a nationwide, fully online Gerontological Nursing Preparatory Course (GeroNursePrep) offering 30 American Nurses Credentialing Center continuing nursing education contact hours upon completion.

The increase in empowerment, satisfaction, and competence of board-certified RNs (RN-BC) specializing in gerontological nursing raises the question of what impact does certification have on facility quality scores and clinical outcomes. More research is needed to isolate this impact through targeted studies. As RNs function as care team leaders in LTC settings, those who are well prepared in the specialty of caring for older adults should provide value to quality improvement teams and direction to direct care staff, potentially reducing risk and increasing overall quality of care (Rowe et al., 2016). The purpose of the current study is to explore clinical outcomes and quality ratings in LTC facilities following the introduction of RN-BCs.

Method

Study Sample

A convenience sample was selected of SNFs (N = 110) that employed at least one gerontological RN-BC on staff. This sample was selected using the GeroNursePrep course database. A second sample included SNFs (N = 110) that did not have any gerontological RN-BCs on staff. This sample was selected from the Long-Term Care Facts on Care in the US ( LTCfocUS.org), a database maintained by the Brown University School of Public Health. LTCfocUS.org contains aggregated information for all nursing homes nationwide from resident Minimum Data Set (MDS) assessments and facility Online Survey, Certification, and Reporting (OSCAR) data collected during state inspections. Facilities with a RN-BC were pair-matched using 11 organizational characteristics known to impact quality of care and patient outcomes, including: (a) total number of beds; (b) occupancy rate; (c) profit status; (d) chain membership; (e) nurse staffing level using hours per patient day (HPPD); (f) nursing assistant staffing level using HPPD; (g) payer mix (i.e., percent funded by Medicaid); (h) payer mix (i.e., percent funded by Medicare); (i) acuity index; (j) percentage of patients with dementia; and (k) location in a rural or urban county based on Rural Urban Continuum Code classification (Butler & Beale, 1994). For the current study, MDS and OSCAR-based facility measures recorded on April 2011 (prior to RN-BC exposure) and April 2014 were used.

Design

In this retrospective analysis of a cohort of 220 pair-matched SNFs, the primary exposure variable was presence of a gerontological RN-BC within the SNF. Samples were compared via a propensity score approach described below. The first set of outcome measures included 5-Star ratings from the Nursing Home Compare (NHC) database, which is a consumer tool maintained by CMS. NHC rates SNFs on a scale of 1 to 5 (higher scores equate to better quality) based on performance on inspection surveys, MDS–based quality measures, RN staffing, overall staffing, and overall quality. The primary resident outcome measures for the study included: (a) prevalence of restraint use; (b) percentage of residents with urinary tract infections (UTIs) within 30 days of assessment (closest to April 1); (c) percentage of residents with a fall since admission or a prior assessment; and (d) prevalence of antipsychotic medication use among residents without psychiatric diagnoses. The outcome measures used have been identified as potentially sensitive to nursing care (Heslop & Lu, 2014; Rantz, Flesner, & Zwygart-Stauffacher, 2010).

Statistical Analyses

Differences between SNFs with at least one gerontological RN-BC and those that had no gerontological RN-BC were analyzed. There were two time periods examined: the “pre-period” (i.e., before the introduction of a gerontological RN-BC in either the case or control group) in April 2011, and the “post-period” 3 years later (i.e., after the introduction of a gerontological RN-BC) in April 2014 (Figure 1). These dates were chosen based on pre- and post-obtainment of board certification using the GeroNursePrep program participant lists.

Study design.Note. RN-BC = RN board certified in gerontological nursing; Dx = diagnosis.a Pre-period (April 2011), prior to known RN-BC exposure.b Post-period (April 2014), after known exposure to at least one RN-BCc Post-period (April 2014), after no known exposure to at least one RN-BC.

Figure 1.

Study design.

Note. RN-BC = RN board certified in gerontological nursing; Dx = diagnosis.

a Pre-period (April 2011), prior to known RN-BC exposure.

b Post-period (April 2014), after known exposure to at least one RN-BC

c Post-period (April 2014), after no known exposure to at least one RN-BC.

A propensity score model was used to find a set of 110 SNFs to match with the sample of SNFs that employed at least one gerontological RN-BC. The matching method was nearest neighbor without replacement, where a SNF with a RN-BC was matched to a SNF without a RN-BC that resembles it most closely based on organizational and geographic characteristics. This approach estimates the association between RN-BC participation and outcomes based on differences between SNFs that participated in the program and SNFs that were likely to participate but did not. The control SNFs were selected from the entire set of approximately 15,000 SNFs in the 2011 OSCAR and on the basis of the covariates described above. To evaluate the performance of the propensity score matching, differences between SNFs with and without gerontological RN-BCs were examined in the distribution of covariates during 2011 using t tests for continuous variables and chi-squared tests for categorical variables. If the SNFs with a gerontological RN-BC were adequately matched, no differences would be anticipated when compared with those that had no gerontological RN-BC. t tests were then used to determine how the groups differed in outcomes pre- and post-introduction of gerontological RN-BCs. In addition, the level of change in outcomes between 2011 and 2014 for each sample was compared. No adjustments were made for any covariates given the rich set already used for matching. Because only de-identified, facility-level data were used, this analysis is not considered human subjects research. All analyses were completed in Stata SE version 15.1.

Results

Descriptive characteristics of the SNFs in the pre-period (2011) are presented in Table 1. Overall, the average SNF in the two samples had approximately 96 beds with an 86% occupancy rate. Approximately 55% of SNFs were non-profit and members of chains. On average, 54% of beds were funded by Medicaid, whereas 14% were funded by Medicare. Approximately one half of SNF residents had dementia. Approximately 40% of SNFs were in rural counties. There were no statistically significant differences between groups (i.e., SNF with and without a gerontological RN-BC) with respect to characteristics used, indicative of adequate matching.

Characteristics of Facilities With and Without RN-BC, 2011 (N = 220)

Table 1:

Characteristics of Facilities With and Without RN-BC, 2011 (N = 220)

Table 2 presents differences in quality ratings and clinical outcomes for SNFs with and without gerontological RN-BCs. The differences between time periods and groups are also shown in Table 2. Although the SNFs with gerontological RN-BCs performed better on six of nine publicly reported quality ratings and outcome measures in the pre-period (April 2011) and post-period (April 2014), there were no statistically significant differences between groups in the pre-period. In the post-period, there were two measures that were significantly different between groups. SNFs with a gerontological RN-BC outperformed SNFs without a gerontological RN-BC on Overall (4.10 vs. 3.55, p < 0.01) and Survey (3.48 vs. 2.86, p < 0.01) 5-Star ratings. When looking at differences, SNFs with a gerontological RN-BC showed greater improvement versus those SNFs without a gerontological RN-BC on seven of nine measures. Two measures approached statistical significance: Overall 5-Star rating (0.50 vs. 0.07, p = 0.07) and Survey 5-Star rating (0.21 vs. −0.19, p = 0.10).

Outcomes Among Facilities with at Least One RN-BC Versus Matched Control Facilities, 2011 to 2014

Table 2:

Outcomes Among Facilities with at Least One RN-BC Versus Matched Control Facilities, 2011 to 2014

Discussion

The current study explores the impact of specialized training in gerontological nursing targeting board certification standards on the quality of care in SNFs. Although few statistically significant differences in comparisons were found, the fact that seven of nine measures improved more in SNFs with gerontological RN-BCs is meaningful. The difference in the decrease in antipsychotic medication use—>5% among SNFs with a gerontological RN-BC compared to 3% among SNFs without a gerontological RN–BC—may be clinically meaningful. The two measures that did not show greater improvement among facilities with a gerontological RN-BC—the percentage of residents with physical restraints and UTIs—improved as expected in this group, albeit less so than among facilities without a gerontological RN-BC.

Specialty certification has been shown to impact patient care quality and safety practices that improve outcomes in acute care settings. The certification process can augment non-degree education and create connections for nurses to sources of evidence-based practice (Corazzini, Anderson, Mueller, Thorpe, & McConnell, 2012). Specialty knowledge is one component that influences professional jurisdiction (i.e., the authority and autonomy over direct nursing care provided) (Corazzini et al., 2012). Studies show that nurses with specialty certifications demonstrate more innovative thinking than those without this certification (Dy Bunpin, Chapman, Blegen, & Spetz, 2016). Innovation is an important aspect within the nursing scope of practice and is the driver of numerous improvements in care models, policies, and unit level change (Thomas, Seifert, & Joyner, 2016). Innovative thinking is required in the SNF setting and is needed now more than ever at all levels of nursing to impact direct care, care policies, systems, and training, as staff are adjusting to regulatory demands and changing care structures. Multilevel support of certification within facilities (Lindgren & Lancaster, 2016) may be key in equipping nurses with the empowerment, clinical competence, and professional connections needed to drive quality change (Corazzini et al., 2012).

Limitations

The current study has several limitations. First, the sample size was small and convenience sampling was used. Existing administrative data were relied on for matching and outcome measures; thus, there are likely facility-level characteristics not captured in the data (e.g., culture of quality, turnover rate, experience of leadership) that affect the matching process, as well as outcome measures that are more sensitive to gerontological RN-BC and less “noisy” than aggregate facility-level measures. Second, the variable explored is defined as a SNF that had at least one RN-BC in the facility at some point during the two time periods; the exact date of RN-BC introduction can vary. Therefore, some SNFs may have been exposed to the presence of a gerontological RN-BC for less time than others, potentially diluting the effect of having a gerontological RN-BC on the outcome measures. Finally, it was decided to simplify the variable explored to the presence of at least one gerontological RN-BC; however, it may be unrealistic to assume that one such employee could alter SNF clinical outcomes alone. Indeed, further research is needed to determine the number of certified RN-BCs vis-à-vis patient ratios, the role of the gerontological RN-BC (e.g., direct patient care versus administration), and the length of employment. All of these factors may influence clinical outcomes. It is important that further research be done with a larger sample and with more data on the number of RN-BCs per facility and length of employment to determine if certification has the same impact in LTC. Examining the effect of gerontological board certification in a larger sample and with more rigorous, prospective study design should be the goal of future studies.

Conclusion and Implications

The current study provides some evidence that employing RNs with intentional, specialized training, as demonstrated by attainment of board certification, is related to better facility-level quality ratings and clinical outcomes. Although not all measures were statistically significant, overall results showing greater improvement among facilities with RN-BCs versus controls suggest a potentially broader effect. Positive trends in quality indicators as a result of the training and experience of RN-BCs could ultimately lead to better state survey performance and perhaps even cost savings (e.g., reduced fines) as a result. Thus, SNFs should consider gerontological nursing board certification as a strategy to create a culture of quality, help increase retention, and improve overall resident care.

References

Characteristics of Facilities With and Without RN-BC, 2011 (N = 220)

CharacteristicOverallRN-BCControlp Value
No. of beds (mean, SD) (range)95.5 (50.6) (15 to 300)92.9 (47.1)98.2 (54)0.45
Occupancy rate (mean, SD) (range)0.86 (0.12) (0.47 to 1.0)0.87 (0.11)0.86 (0.1)0.54
For profit (n, %)121 (55)62 (56)59 (54)0.78
Chain member (n, %)120 (54)65 (59)55 (50)0.17
Payer mix (mean, SD) (range)
  Medicaid53.4 (22.0) (0 to 97.7)53.1 (21.6)53.8 (22.4)0.83
  Medicare13.7 (13.4) (0 to 100)13.1 (11.2)14.3 (15.3)0.52
Acuity index (mean, SD) (range)101.3 (23.1) (24.3 to 167.5)103.1 (21.2)99.5 (24.8)0.25
% of patients with dementia (mean, SD) (range)52.0 (17.7) (8.3 to 100)53.3 (17.0)50.7 (18.3)0.27
Nursing FTEs (mean, SD) (range)25.1 (16.6) (5.8 to 93.2)24.1 (14.8)26.0 (18.3)0.41
CNA FTEs (mean, SD) (range)40.0 (26.6) (0 to 165.4)37.9 (24.2)42.0 (28.8)0.26
Rural (n, %)89 (40)43 (39)46 (42)0.67

Outcomes Among Facilities with at Least One RN-BC Versus Matched Control Facilities, 2011 to 2014

Outcome2011 (Pre-Perioda)2014 (Post-Periodb)Pre-Post Difference
RN-BCControlp ValueRN-BCControlp ValueRN-BCControlp Value
5 Star Ratingsc
  Overall3.603.480.464.103.55<0.010.500.070.07
  Survey3.273.050.173.482.86<0.010.21−0.190.10
  Quality3.253.410.334.104.150.720.850.740.60
  Staffing3.493.500.953.863.670.170.370.170.34
  RN staffing3.513.560.773.903.750.310.390.190.36
Clinical measuresd
  % restraint use2.782.600.821.981.880.90−0.80−0.720.94
  % UTI1.692.070.571.140.840.59−0.55−1.230.44
  % falls17.3018.350.5816.7817.340.77−0.52−1.010.85
  % antipsychotic medication use21.8622.240.8516.5719.030.19−5.29−3.210.45
Authors

Dr. Keeler is Assistant Professor, Director, Continuing Nursing Education, and GeroNursePrep Faculty, Dr. Culross is Assistant Professor and Director, Learning Resources Center, and Associate Director, UNMC Interprofessional Academy of Educators, and Dr. Cramer is Professor Emeritus, University of Nebraska Medical Center College of Nursing, Omaha, Nebraska; and Ms. Baier is Associate Director, Center for Long-Term Care Quality & Innovation, and Associate Professor of Practice, Department of Health Services, Policy & Practice, and Mr. Kosar is PhD Candidate, Center for Gerontology and Healthcare Research, Brown University School of Public Health, Providence, Rhode Island.

The authors have disclosed no potential conflicts of interest, financial or otherwise.

Address correspondence to Heidi Keeler, PhD, RN, Assistant Professor, CON-Omaha Division, Director, Continuing Nursing Education, GeroNursePrep Faculty, College of Nursing, CON 5071, 985330 Nebraska Medical Center, Omaha, NE 68198-5330; e-mail: hkeeler@unmc.edu.

Received: May 11, 2019
Accepted: August 06, 2019

10.3928/00989134-20191011-06

Sign up to receive

Journal E-contents