Quality health care services are more difficult to provide in smaller, poorer, and isolated areas (Institute of Medicine [IOM], 2005). For example, the difficulty of recruiting staff is increased by the smaller scale of services and adequacy of supply inherent in rural areas (Phillips & McLeroy, 2004). Rural NHs face different challenges than urban NHs, and little is known about what influences key quality indicators, such as staffing levels, in these settings. The Mountain West region is the part of the nation where much of the projected rate of growth of the elderly population is occurring and concern for an adequate number of health care workers to care for this population is increasing. HRSA (2004) data indicate that states in the Mountain West face particularly high shortages (Table 1). It is estimated that nearly 30,000 additional nurses will be needed in these states alone by 2015. The direct care workforce falls short in meeting the demands of our aging population (U.S. General Accounting Office, 2001). In 2007, vacancy rates for those providing direct care in nursing facilities were 16.3% for RNs, 11.1% for licensed practical nurses (LPNs), and 9.5% for certified nursing assistants (CNAs); these rates are expected to increase (American Health Care Association [AHCA], 2010). This study provides administrators and health policy makers, particularly those in rural areas, with data to inform health care decisions in preparation for increased resource demand and the growing older adult population.
Staffing Hours and Staffing Mix
Measures of nurse staffing have previously been documented as a marker of quality in NH care (Harrington, Zimmerman, Karon, Robinson, & Beutel, 2000). Critical indicators of nurse staffing in long-term care include staffing hours and staffing mix.
Staffing hours refers to the number of hours worked per resident per day (HPRD) by CNAs, LPNs, RNs, and total nursing staff (Bostick, Rantz, Flesner, & Riggs, 2006). The Centers for Medicare & Medicaid Services (CMS) identified staffing thresholds of 2.80 nurse aide and 1.30 licensed nursing (RN and LPN) HPRD and suggested that staffing levels below these thresholds potentially place residents at risk for poor outcomes (Kramer & Fish, 2001). Schnelle et al. (2004) agreed that nursing assistant staffing above 2.8 hours per day led to better quality. However, rural NHs were not likely to meet the CMS staffing recommendations (Phillips, Hawes, & Williams, 2003; Phillips et al., 2004).
Levels of nursing staff including CNAs, LPNs, RNs, and total nursing hours have been reported. In 33 states with staffing standards beyond federal requirements, the median licensed nursing HPRD were 2.35 compared with 1.26 for all states (Mueller et al., 2006). From a national sample (N = 13,777), Harrington et al. (2000) reported an average of 3.4 total nursing staff hours (CNA = 2.14 hours, LPN = 0.67 hours, RN = 0.59 hours). In one of few studies comparing urban and rural NHs, twice as many RN hours (0.80 versus 0.40) were reported in Utah urban NHs compared with rural homes (Towsley, Dudley, & Beck, 2006).
Staffing mix refers to the ratios of full-time equivalent (FTE) hours worked by CNAs, LPNs, and RNs to all FTE nursing staff. Little research related to staffing mix in NHs exists (Bostick et al., 2006). In California, an inverse relationship was found in NHs between number of serious deficiencies and whether the facility met the state staffing standard and percentages of RNs in the staffing mix (Kim, Harrington, & Greene, 2009).
A growing body of evidence indicates that staffing levels are associated with care quality. Higher nursing assistant staffing has demonstrated higher quality in providing assistive care such as feeding or incontinence care (Schnelle et al., 2004), and more licensed nursing hours (including RN but not aide hours) have been associated with better resident functional status, such as eating, bathing, and transferring (Bliesmer, Smayling, Kane, & Shannon, 1998). RN direct care time has been associated with better care outcomes such as fewer pressure ulcers and hospitalizations (Horn, 2008; Horn, Nuerhaus, Bergstrom, & Smout, 2005), while optimal nursing assistant and RN staffing levels were reported to reduce pressure ulcer costs (Hendrix & Foreman, 2001).
Both organizational and market characteristics that influence a NH’s viability and quality of service (Banaszak-Holl, Zinn, & Mor, 1996; Weech-Maldonado, Neff, & Mor, 2003) may affect nurse staffing; however, little research has examined the influence of these factors on staffing hours or mix. Organizational characteristics of NHs include facility size, type of ownership, chain affiliation, ownership changes, and resident acuity. Larger NHs may be more resilient in facing threats such as low occupancy or decreased reimbursement due to the resources that are available to them (Banaszak-Holl et al., 1996; Weech-Maldonado et al., 2003). Thus, larger NHs may also have more resources for staffing. Between 1997 and 2007, however, NHs with 50 to 150 beds had significantly decreased RN HPRD and RN mix and increased LPN and CNA HPRD compared with smaller NHs that had little or no change in RN HPRD or RN mix. Also during this period, for-profit NHs had larger decreases in RN HPRD and RN mix than nonprofit NHs; government-affiliated NHs reported increased HPRD across all staffing categories (Seblega et al., 2010).
When examining ownership type, for-profit NHs have been associated with less staffing than nonprofit NHs (Harrington, Woolhander, Mullan, Carrillo, & Himmelstein, 2001; Mueller et al., 2006). McGregor et al. (2005) examined ownership type and staffing levels of NHs in British Columbia (N = 167). Adjusted analyses for size and level of care revealed significant differences between not-for-profit and for-profit NHs; not-for-profit homes provided approximately 20 more minutes of direct-care staff (McGregor et al., 2005). Higher staffing levels have been associated with fewer number of beds and not being part of a chain (Mueller et al., 2006). Kash, Hawes, and Phillips (2007) compared staffing levels reported in Online Survey Certification and Reporting (OSCAR) and Medicaid cost report data and deemed Medicaid cost report data as more authentic than OSCAR’s. Characteristics of NHs that “overreported” staffing levels, meaning they reported more hours than actually worked, included lower Medicaid and Medicare census, more competition, and for-profit ownership. Overreporting of RN staffing hours in OSCAR data was associated with for-profit and larger NHs (Kash et al., 2007).
Market characteristics include competition, occupancy rates, and availability of special care units. Previous research has suggested that NHs with higher levels of Medicaid residents have lower levels of RN staffing (Harrington & Swan, 2003; Zinn, 1994). These findings are similar to a study conducted by Donoghue (2006). In a sample of 912 NHs, lower CNA and RN staffing levels were reported among NHs that had more beds designated for Medicaid clients; however, no differences were found between for-profit and nonprofit NHs (Donoghue, 2006). Higher levels of staffing have been associated with lower occupancy (Mueller et al., 2006). Kash et al. (2007) suggested lower Medicaid or Medicare occupancy in NHs was associated with the likelihood of overreporting LPN or CNA hours.
Staffing levels are also dependent on the availability of staff and the ability to recruit and retain staff—a challenging task in rural locales. For example, nationally, CNA turnover was estimated at more than 71%, and rural freestanding NHs had the highest turnover rate at 76.4% (Decker et al., 2003). In rural areas, staff RNs and LPNs had turnover rates of 47%, and 45.8%, respectively (Singh & Schwab, 2000). On the other hand, recruitment and retention have been aided by education opportunities, benefits, work flexibility (Kemper et al., 2008; Proenca & Shewchuk, 1997), and community attachment (Singh & Schawb, 2000). Thus, in this study we addressed two specific aims. We not only examined predictors of staffing hours and mix in rural NHs but also qualitatively examined the challenges NHs face and the facilitators NHs use to recruit and retain staff.