With current fiscal constraints, hospitals must constantly scrutinize patient interventions for cost-effectiveness. Constant observation (CO) is 1-on-1 observation typically performed by unlicensed staff as a mechanism to protect high-risk patients from harm.1,2 Constant observation, often referred to as sitters, companions, or 1-on-1, is a nonreimbursable service provided by many hospitals to reduce liability from patient falls and injury.1,2 Reports of CO in the recent literature has demonstrated increased use.1-3 The associated increasing staffing cost is unsustainable and unpredictable in many systems.1-3 In addition, this higher level of staffing is not consistently achieving the high-quality desired outcome.1,3 Our organization faced these challenges with increased use of CO. An analysis of this issue, the cost and contributing factors, resulted in restructuring of the delivery of care to reduce CO expenditures while meeting patient care needs.
The Joint Commission (TJC) restraint and seclusion standards discourage the use of chemical and physical restraints as a routine practice to manage confused patients.4 Other issues including liability concerns led to increased use of CO in hospitals.4 There is scant evidence that the use of CO improves patient outcomes. Tzeng et al5 reported that a CO program had no significant impact on fall rates on 2 medical units. Another study reported a cost-benefit analysis of the impact of a sitter program resulted in a slight increase in falls and decrease in patient satisfaction.6 In contrast, research has demonstrated that some strategies used to reduce CO result in significant cost savings with no negative impact on patient falls, use of restraints, or patient satisfaction.2,7 In 1 organization, the implementation of a CO decision-making algorithm that included a sitter justification form; letters for the physician, nurse, patient, and family; and scripting resulted in major cost reduction without affecting fall rates.1 Similarly, Adams and Kaplow2 instituted safety huddles, intentional hourly rounding, environmental rounds, criteria for CO use, and alternative equipment to CO including alarms. This led to more than 50% reduction in CO with estimated cost savings of $1.2 million and no significant increase in fall rate and restraint use. At another hospital, a psychiatric liaison nurse (PLN) was employed to consult on patients requiring CO.3 The PLN provided support and education to bedside nurses and collaborated with them to implement alternatives to CO. This intervention was successful in reducing CO shifts by 42% with a 4-month cost savings of $97,056, while simultaneously reducing the hospital-wide fall rate by 25%.3 Similarly, a CO reduction program managed by 2 clinical nurse specialists (CNSs) who were consulted for confused, agitated, and suicidal patients resulted in a reduction of dollars spent on CO in the 1st year by $81,254 (65%), in the 2nd year by $13,300, and decreased the length of stay for suicidal patients during the 1st year.8
A number of factors contribute to the high use of CO. Rochefort et al4 found that the use of CO was higher among patients with specific mental health conditions, particularly for patients with dementia, delirium, schizophrenia, psychoses, and other cognitive conditions. Rochefort et al4 also observed that a shortage of RNs was associated with high sitter costs because RNs compensated for their reduced surveillance capacity by assigning a sitter. During an interview with Wilkes et al,9 nurses mentioned inadequate staff education and physical environment as barriers in providing safe care to patients with behavior disturbances. This finding is well established in literature as common contributing causes for high use of CO.
Physician’s orders for CO contribute to the complexity of the issue. When physicians’ orders are neither renewed nor discontinued in timely fashion, there are subsequent, prolonged inappropriate usages of CO.1 One institution eliminated the requirement of a physician’s order for sitters. Following the change, there was a decrease in the number of falls and the rate of fall-related fracture.10 The author posited that delegating direct observation to unlicensed staff is consistent with the RN scope of practice and should not require a physician’s order.10 It is evident from the literature that nurses encourage physicians to write CO orders, believing that it is an effective method for keeping their patients safe.7 This practice shifts responsibility to provide patient surveillance from the nurses to a 1:1 provider. Because there is no rigorous evidence to support the use of CO for patient safety, it is essential to transition this paradigm to a new framework that encompasses both patient safety and fiscal responsibility.
About Our Project
Our facility is a 751-bed Magnet®-designated academic medical center. The hospital is also nationally accredited as a level I trauma center, recognized as an advanced comprehensive stroke center by TJC, and our cancer program has been National Cancer Institute designated. We provide specialized care to patients with complex medical conditions including transplant, oncology, neurology, stroke, burn, and cardiology.
Historically, the hospital utilized CO on all patients who were at risk of suicide, confused, and/or agitated. It was noted by nursing leaders that there was a lack of follow-through by physicians and nurses to critically evaluate the continuing need for CO and to discontinue it when appropriate. An internal survey conducted in 2007 revealed that CO was commonly used for the management of patients with confusion (42.2%); at risk for harm to self or other (21.6%); for fall prevention (10.3%); and various other reasons (25.6%). In fiscal year (FY) 2011, the hospital spent $1.2 million on CO, a 17% increase from the previous year, with no reduction in fall rates. Constant observation had become the 1st line and at times the primary intervention for patient confusion and as a result often prevented the identification and treatment of the underlying cause.
Previous strategies within our organization to reduce CO use were largely unsuccessful, often causing conflict between medical and nursing staff members. In 2012, nursing leadership reorganized the CO practice from a physician-driven to a nurse-driven intervention with oversight provided by CNSs. The term safety watch (SW) was coined to label this practice change, thereby increasing awareness of nurse accountability in providing surveillance for patients.
Safety watch was designed to provide targeted interventions for patients at increased risk of fall or injury. Safety watch empowered the nurses to assess safety needs on an ongoing basis and adjust nursing actions based on the patient’s needs from hour to hour. As a component of the program, nurses were provided with a variety of tools to maintain the safety of confused patients (Table 1). If alternatives were not effective, a nurse-initiated SW was implemented (Table 2). Safety watch provided a framework for nurses to increase the level of observation and assessment and included 15- or 30-minute rounding, 2:1 observation (a single staff member observing 2 patients), or 1-on-1 observation. Charge nurses and managers were encouraged to initially utilize unit resources. If the safety needs of patients could not be met by the standard staffing, float pool resources were requested. Nurse managers were engaged to increase unit-level accountability and transparency by reporting their unit CO utilization data. Nurses were discouraged from calling physicians for CO orders except in the case of suicidal patients. Nurses were educated and mentored to engage family members to personally observe their loved ones, thus providing an additional invaluable resource for patient safety. Clinical nurse specialists rounded daily to mentor staff nurses to better manage confused patients. This unique protocol, along with CNS support, empowered nurses to initiate an appropriate level of observation, allowed unit leadership to staff effectively, and supported the team to work collaboratively to keep every patient safe. Although every 15-minute safetyrounds might have previously sounded impossible, with a culture change to team accountability, these safety checks were achievable. An important distinction between the new protocol and the standard CO was that SW was not intended or appropriate for suicidal patients. Comparisons between the former method of CO and the new SW are outlined in Table 2. The SW program was instituted at the beginning of FY 2012 (July 2011) in all adult and pediatric inpatient units and has become hospital policy for management of patient at risk for harm to self and others.
The specific aim of this project was to evaluate the impact of SW on fall rates, restraint usage, and CO expenditures.
Secondary financial and outcome data were used to evaluate the effectiveness of SW. Financial data were collected from monthly float pool operational expense reports and included the salaries and overtime of individual CO providers. The fall rate and restraint use were compiled from data submitted to Nursing Database Nursing Quality Indicators®.
Constant observation hours were extracted from the hospital’s electronic staffing system and were converted to standardized CO full-time equivalents (FTEs) and CO hours per 100 patient-days. Both financial and clinical outcomes were measured from preprogram FY 2011 (July 2010-–June 2011) to postprogram FY 2012 (July 2011-–June 2012) (Table 3; Figures 1-3). The SW Project was a redesign of internal processes to improve efficiencies and management of resources. Because this was a quality improvement initiative, institutional review board approval was not sought.
Constant observation expenditures were reduced by 41.3% in FY 2012, resulting in a cost saving of $533,917 (Table 3). In FY 2011, the total CO hours were 75,328.7 compared with FY 2012 CO hours of 43,253.7. Despite higher patient census in FY 2012, the CO hours were decreased by 42.6% (Figure 1). The CO hours per 100 patient-days declined from 48.4 to 26.4 in FY 2011 and FY 2012, respectively, resulting in an overall reduction in use of CO by 45.3% and elimination of 15.4 FTEs after program. In addition, patient fall rates were reduced from 3.2 per 1,000 patient-days in FY 2011 to 2.9 in FY 2012, a 10.1% improvement (Figure 2). The percentage of patients with a physical restraint was also reduced from 4.93 in FY 2011 to 3.41 in FY 2012, which represented a 30.8% reduction (Figure 3).
The fiscal impact of SW has been noteworthy. The purpose of SW was to decrease CO cost without negatively impacting patient safety. The pre- and post-SW periods showed dramatic reductions in CO utilization and cost while improving patient fall rates and restraint use. The previous physician-driven process of CO encouraged nurses to use CO as the primary method of keeping patients safe without being held accountable for overall cost. In contrast, the nurse-driven protocol shifted accountability to the nursing teams, thus highlighting their role with fiscal responsibilities. Similar to other programs, SW boosted collaborative decision making. Unit coordinators and staff nurses were compelled to discuss the necessity of sitter and alternatives before initiating SW. This collaboration enhanced patient care, nurse autonomy, and appropriate utilization of resources. Nursing staff became more cognizant and creative about how to best use safety resources. During the daily unit coordinators’ meeting, nursing staff discussed patients who needed SW and shared available staffing. During family visits, family members were purposefully engaged in patient observation, allowing the 1-on-1 providers to be redirected to other patient care tasks, adjustments that were not possible with the previous physician-directed order. Anecdotally, we found that families were supportive and competent in keeping patients calm and safe.
Although we swiftly changed our culture, the journey was not without challenges. Anticipating pushback from physicians regarding their long-standing authority of ordering CO, we opted to take a nonconfrontational approach. Nurses were simply encouraged to institute 1:1 based on SW rather than asking physicians for a CO order. Managers and charge nurses were given scripting to assure physicians that patient safety would be maintained. Clinical nurse specialist rounding was critical in mentoring and educating nursing staff and physicians while also addressing safety concerns on an individual basis. After several months of successful reduction in CO usage, presentations were made to the Medical Staff Performance Improvement Committee showing cost and quality outcomes. We continue to monitor outcomes and provide continuous feedback to all stakeholders, which has been critical to the success of the program. In the future, we plan to more proactively involve the families in SW. Clinical nurse specialist daily rounds have led to the identification of many patients who suffer from delirium and have highlighted the need for additional quality improvement projects.
Safety watch has been successful in reducing the cost of CO without a negative impact on quality. Fall rates and restraint use both decreased following the implementation of SW in our organization. By changing strategies and exploring the root causes of CO use, we concluded that nurses, not physicians, should provide continuous oversight in everyday patient surveillance and safety. We successfully empowered nurses to initiate SW, a flexible patient safety observation program that can be adjusted to the patient’s needs, replicated at other facilities and settings, and discontinued safely at the earliest opportunity.
1. Spiva L, Feiner T, Jones D, Hunter D, Petefish J, Vanbrackle L. An evaluation of a sitter reduction program intervention. J Nurs Care Qual. 2012; 27 (4): 341–345.
2. Adams J, Kaplow R. A sitter reduction program in an acute health care system. Nurs Econ. 2013; 31 (2): 83–89.
3. Rausch B, Bjorklund P. Decreasing the cost of constant observation. J Nurs Adm. 2010; 40 (2): 75–81.
4. Rochefort CM, Ward L, Ritchie JA, Girard N, Tamblyn RM. Patient and nurse staffing characteristic associated with high sitter use costs. J adv Nurs. 2011; 68 (8): 1758–1767.
5. Tzeng H, Yin C, Grunawalt J. Effective assessment of use of sitter by nurses in inpatient care settings. J Adv Nurs. 2008; 64 (2): 176–184.
6. Boswell D, Ramsey J, Smith M, Wagers B. The cost-effectiveness of a patient-sitter program in an acute care hospital: a test of the impact of sitters on the incidence of falls and patient satisfaction. Qual Manag Health Care. 2001; 10 (1): 10–16.
7. Harding AD. Observation assistants: sitter effectiveness and industry measures. Nurs Econ. 2010; 28 (5): 330–336.
8. Heyman EN, Lombardo BA. Managing costs: the confused, agitated or suicidal patient. Nurs Econ. 1995; 13 (2): 107–118.
9. Wilkes L, Jackson D, Mohan S, Wallis M. Close observation by ‘specials’ to promote the safety of the older person with behavioural disturbances in the acute care setting. J Aust Nurs Prof. 2010; 36 (1-2): 131–142.
10. Week Sk. Reducing sitter use: decision outcomes. Nurs Manag. 2011; 42 (12): 37–38.