Patient safety has long been a concern of health care providers, driving quality improvement initiatives and inspiring the pursuit of positive patient outcomes. A major patient safety concern in health care is the risk of falls and the related potential for harm related to patient falls. Numerous targeted interventions have led to a reduction in patient falls but often not sufficiently enough to meet identified benchmarks. In addition, health care providers have increasing concerns about patient safety risks related to substance abuse/drug diversion, mental health/coping, and violent or suicidal behaviors. The use of video monitoring (VM) for protecting patient safety in acute care settings is a relatively recent addition to standard interventions, with encouraging results.1–3 However, because this practice is new, there are limited resources available that describe guidelines for implementing VM. The purpose of this study was to evaluate the VM process implemented at a large tertiary care center from the perspective of nursing staff, video monitor technicians (VMTs), and patients/families. The goal was to use the data gained to improve the process and identify lessons learned for dissemination to others considering implementation of VM for patient safety.
Patient safety, particularly related to fall prevention, is one of the top priorities for acute care facilities.4 In 2007, a comprehensive fall prevention program was implemented at a large tertiary care university medical center. The program included use of the Hendrich II Fall Risk Assessment Model,5 identification of fall-risk patients with purple armbands and socks, placement of a purple star outside the door of at-risk patients, and individualized measures such as hourly rounding, use of a bedside sitter or bed alarms, and placement of high-risk or impulsive patients closer to the nurses' station. Despite numerous efforts to ensure consistency in the utilization of fall prevention measures, patient falls remained outside the targeted benchmarks.
Research has indicated that VM can be effective in reducing patient falls1–3 and also helpful in addressing other safety concerns such as patient elopement, drug use/diversion, self-harm/violence, and dislodgement of medical devices.2,3,6 After a thorough review of available literature related to VM, a variety of methodologies for VM were discovered, including use of hardwired versus mobile cameras, centralized versus decentralized placement of monitoring stations, and patient monitoring by staff performing other dedicated duties versus use of trained VMTs to constantly observe the monitor screens.1,2,6–9 The one general consistency found in the technical aspects of VM was the use of real-time video rather than video recordings to protect patient privacy. In addition to variations in the technical aspects of VM, there were also inconsistencies in the monitoring process, and there were no published evidence-based standards for implementing VM.
After a thorough review of VM-associated scholarly articles by the nursing-led evidence-based practice team, the decision was made in 2013 to implement a pilot program with fixed-camera VM on the 2 medical-surgical units having the highest incidence of patient falls. In 2014, the pilot program was expanded to 2 additional units. About one-third of all beds on the 4 pilot units were VM capable. Real-time VM of patients at high fall risk was used in addition to the aforementioned fall prevention interventions already in place. As part of an institutional commitment to protect both patient privacy and safety, VM was deemed an institutional standard of care for patients at safety risk and, as such, was included in the general patient consent for treatment. Thus, specific patient consent for monitoring and a physician's order to implement VM were not required.2,6–8
A job description was developed for the VMT position, and VMTs were hired and trained. In addition, nursing personnel were trained about the VM process and the algorithm that had been developed to identify patients appropriate for monitoring. Video monitor technicians observed patients from a central monitoring room. When observed patients displayed fall-risk behaviors, the VMTs had several intervention options available: verbally redirecting patients via a microphone in the camera, directly calling the nurse or the unlicensed clinical associate assigned to the patient, activating the patient call bell, or overhead paging of staff to the patient room.
After VM was in place on each unit for at least 6 months, an initial evaluation of patient fall rates and bedside sitter use was undertaken. Findings indicated significant reductions in each after implementation of VM—28.5% and 23.2%, respectively.2 However, the perception was that reductions in both falls and sitter hours would likely have been greater if VM was available for all patients at risk for falls. Thus, additional mobile VM cameras were obtained to increase flexibility in monitoring.
Although VM was demonstrated to reduce patient falls and bedside sitter hours, it appeared that the VM process was not being consistently employed across units and that further improvements in the process could be made. Specifically, perceptions of the nursing-led evidence-based practice team and VMT supervisor indicated that there may be inconsistencies in how patients were selected for monitoring, the information that was shared between nursing staff and VMTs, and perceptions of factors that contributed to falls. Thus, a process evaluation study was undertaken to determine the effective components of the process, identify improvements that could be made, and revise the facility procedures for VM.
The specific aims of the study were to (1) identify nursing staff and monitor technician perspectives on the VM process, including areas for improvement of process; (2) identify patient/family perceptions related to patient monitoring, including perceptions of communication about the process, its impact on patient safety, and protection of privacy; and (3) revise and standardize the process for VM and optimize its value for protecting patient safety.
This descriptive study represents the second phase of step 5 in the Larrabee Model for Evidence-Based Practice, the model used to guide nursing-led practice change efforts within the facility.10 A convenience sample of nursing staff, VMTs, and hospitalized patients or family members of patients who were monitored during hospitalization was used. Separate surveys were developed for nursing staff (registered nurses and unlicensed clinical associates) and monitored patients/family members, addressing issues with the monitoring process that were specific to each. As VMTs were most closely involved in the process, structured interviews were conducted to gather their perceptions about the VMT role and opinions about the monitoring process.
After approval by the university institutional review board as an exempt study, surveys along with a cover letter explaining the study were distributed to all registered nurses and clinical associates employed on the medical-surgical units in which VM was available. Patient/family surveys, together with a cover letter, were distributed to monitored patients. Patient and nursing staff data collection occurred over a 1-month time period. Video monitor technicians were recruited through study flyer and e-mail. Only VMTs who had been in the position for 1 year or longer were included in the interviews to ensure that the respondents had numerous experiences with VM situations. A standard set of interview questions were asked of all participating VMTs.
The nursing staff and patient/family surveys and the VMT interview guide were developed on the basis of feedback from nursing staff and VMTs related to the VM process, as well as the documented VM procedure. The questions were revised on the basis of feedback from staff nurses and unit managers on the monitor-capable hospital units, the medical-surgical evidence-based practice team, and the hospital nursing research council. The nursing staff survey included 12 items to determine triggers for implementing and discontinuing VM, patient information shared with VMTs, and perception about aspects of the VM process. The VMT interview guide contained 10 questions that were parallel to questions on the nursing staff survey and 6 role-specific questions about patient assignments and decision making when patients display at-risk behavior. The patient/family survey included 5 questions asking about how well VM was explained, how well the process of protecting privacy was explained, whether VM was troubling to them, how well they felt that VM protected patient safety, and whether they had any suggestions to improve the VM process. The forced-choice survey questions contained a 5-point Likert-type response scale, with options of not at all to very well.
Nursing staff and VMT data
Seventy-three registered nurses (56%) and 57 clinical associates (44%) responded to the survey. Respondents represented all of the monitor-capable units and all shifts of employment. The vast majority of nursing staff respondents (86.1%) indicated caring for monitored patients some shifts, most shifts, or every shift worked. Seven of 14 (50%) VMTs eligible to participate in the study completed the interview process. All employment shifts were represented, and reported VMT experience ranged from 1.3 to more than 3 years.
The majority of nursing respondents (73%, n = 95) and all VMTs interviewed believed that VM was effective in preventing patient falls. In addition, 62% (n = 81) of nursing respondents indicated that VM had triggered interventions to prevent other harmful patient outcomes such as patient removal of lines/tubes, drug diversion/use, family abuse of the patient, flight/elopement risk, and patient self-harm. Video monitor technicians indicated that VM has prompted intervention for the aforementioned patient safety issues and that they had intervened when observing patients being violent with nursing staff, smoking in the hospital room, and having seizures.
When asked about the most effective VMT intervention when a patient safety issue was observed, the majority of nursing staff and VMTs preferred overhead paging of staff to the patient room. The second preference for both nursing staff and VMTs was a direct call to the nurse or the clinical associate. Video monitor technician verbal redirection of the patient and activation of the patient call light were the least preferred interventions; VMTs indicated that they rarely verbally redirected patients via the microphone on the camera, as it often frightened or upset confused patients.
Both nursing staff and VMTs were asked what they perceived as contributing factors if a monitored patient experienced a fall. Each group identified patient impulsivity and quick movement as a contributing factor. However, nursing staff identified VMT distraction as a factor, while VMTs identified slow nursing response time as a factor and problems with the overhead paging system.
Nursing staff and VMTs were asked what information was shared during VM report. Both indicated that nursing staff alerted VMTs about specific patient behaviors to watch for and provided information about lines/drains that could be dislodged. However, 3 of the VMTs reported that they generally did not receive shift report from nurses. The same number indicated the need for more specific patient information, such as location of intravenous lines or invasive devices that may become dislodged, and instruction as to whether the patient was allowed to ambulate with a family member.
As there were no formal guidelines for discontinuing patient VM, and there had been several circumstances of monitor-capable beds not being available when a patient was identified with safety risks, nurses were asked about what influenced their decision to discontinue VM for a patient. Most frequently, they responded that monitoring was discontinued when the patient was more oriented or had stable mental status (20%, n = 26), consistently called for assistance with ambulation (13%, n = 18), and displayed generally improved/compliant behavior (19%, n = 12).
Video monitor technicians were asked additional questions specific to their positions and responsibilities. They reported that orientation generally prepared them for the role, although 2 did not feel prepared to communicate with nurses and were unsure of the medical terminology used by the nurses, and 1 VMT did not feel prepared to deal with patients with drug use/diversion issues. Video monitor technicians reported generally being responsible for 6 to 12 patients on VM (71%, n = 5), and 86% (n = 6) made assignments based on the acuity/activity of the monitored patients.
Video monitoring at this facility is centralized, with the VMTs located outside of the nursing units, potentially impacting the working relationship between nursing staff and VMTs. Thus, VMTs were asked to describe their relationship with the nursing staff on the units. Two technicians indicated that relationships vary by hospital unit and may be strained if there are a number of unpredictable monitored patients and high workloads for both VMTs and nursing staff. The same number stated that it would be helpful if the nursing staff had an observational experience in the VM location.
Video monitor technicians also were asked for suggestions for any potential actions to further reduce the chance of falls among monitored patients. One response was most common—the hospital's overhead paging system should allow more than 1 overhead staff alert to occur simultaneously, identified by 4 of the respondents. These respondents recalled having to use a nonpreferred intervention for a risk situation due to the overhead paging system being used by another VMT or the central hospital paging system. More frequent use of bedrails was also mentioned by 2 VMTs, and 1 indicated that nursing staff should remove patients from VM when it is no longer needed to reduce VMT workload.
In relation to suggestions for improving the process of VM, responses were quite different between the 2 groups. Nursing staff requested VMTs to intervene first with an overhead staff alert, recommended reduction in the VMT to patient ratio, and suggested VMTs paying closer attention to patients. Video monitor technicians put a much higher emphasis on improving communication between the groups and revising the paging system to allow more than 1 overhead staff alert to occur simultaneously.
Patients and family members of monitored patients completed 52 surveys, with representation from all monitor-available units. Nursing staff indicated the reason the patient was being monitored on each survey prior to distribution by labeling the survey with a letter corresponding to a key code. The majority (71%, n = 37) of patients were being monitored for fall risk, 29% (n = 15) for pulling at lines/tubes, 8% (n = 4) for drug diversion/abuse risk, 6% (n = 3) for elopement risk, and 4% (n = 2) for self-harm risk. Patients completed the majority of surveys (58%, n = 30), with family members completing 41% (n = 22).
Respondents indicated that the need to use VM to protect safety was explained between somewhat and adequately, although 33% (n = 17) responded that it was not explained at all. This may reflect inconsistencies between nursing units and individual nursing staff, or nursing staff explaining the VM process during a time when family members were not present. Similarly, the respondents believed that the process of protecting patient privacy during monitoring was explained between somewhat and adequately, although 44% (n = 23) did not believe that privacy protection was explained at all, again potentially indicating inconsistencies among nursing staff or units or explanation of privacy protection being provided while family respondents were not present. Respondents perceived that VM helped protect the patient's safety, with ratings generally being adequately to well, and 42% (n = 22) indicated very well.
Patients and family members did not feel that VM was troubling, with mean responses between not at all and rarely. Only 3 respondents had any concerns about VM. The only specific patient/family suggestion for improving VM was to explain the process more fully (35% of respondents).
Based on nursing staff, VMT, and patient data, it was apparent that VM was perceived by all groups as being helpful in protecting patient safety and preventing falls. These perceptions are supported by previous research at this institution2 as well as other reported studies that identify significant reductions in patient falls as a result of VM.1,3 In addition, as nursing staff have realized other potential benefits of monitoring, they have begun to use it for additional safety risk situations.
It was also apparent from the data that steps in the VM process itself were being applied inconsistently or needed to be delineated more clearly. Overhead paging of staff alerts for high-risk patient situations was perceived as the most effective VMT intervention; however, VMTs indicated that this option was not always available, as only 1 overhead page could be called at a time due to technology limitations. In addition, VMTs were hesitant to verbally redirect patients due to the fact that it seemed to startle or frighten them—although this finding was not brought forward by the nursing staff. A need for increased communication and additional information about monitored patients was identified frequently by VMTs but to a lesser degree by nursing staff. Patient/family data indicated that VM was perceived as protecting patient safety and was not generally perceived as troublesome but indicated that increased communication about the monitoring process was needed.
As a result of this study, actions have been taken to improve and standardize the VM process at our facility. Mobile cameras have been purchased and installed to increase flexibility in monitoring of high-risk patients. A detailed process for requesting and discontinuing mobile cameras has been disseminated on each unit. An improved algorithm was developed and implemented, with specific criteria for initiating VM and initiating observational bedside sitters (Supplemental Digital Content Figure, available at: http://links.lww.com/JNCQ/A477), and a standard form for nursing staff to VMT report has also been developed and implemented. This report is completed by the charge nurse and telefaxed to the VMT central monitoring location at 7 AM and 7 PM each day. A patient information flyer for patients and families has been developed to explain the VM process. Nursing staff were educated about all areas mentioned previously as well as reminded to place the VM flyer at the head of each monitored patient's bed and outside the door to alert patients, visitors, and staff that the patient is being monitored. Finally, areas for documentation of patient activity and patient/family education about VM were added to the electronic health record to ensure recording of necessary information related to patient VM.
If an acute care facility decides to implement VM, it is important to have clear policies/procedures in place for initiating VM, educate nursing staff and patients/family, ensure regular communication between nursing staff and VMTs, and determine when it is appropriate to discontinue VM. Documentation of patient activity and education in the electronic health record is also important. Finally, multiple methods of VMT intervention in a patient safety risk situation are helpful, but some type of overhead alert system seems to be most effective in averting the patient safety risk.
The limitations of this study include use of a convenience sample of nursing staff, VMTs, and patients/family members. In addition, the data are descriptive and subjective in nature, reflecting perceptions of study participants. These factors could influence generalizability of study findings. However, the data can provide information to other facilities as to the important issues in developing a VM program to reduce patient safety risk.
Results of this study were used to develop standardized guidelines for the VM process, including implementation, effective utilization, discontinuation of VM, nurse-to-VMT communication, and patient/family education about VM. Information gained from this study provides considerations for other institutions implementing VM for patient safety. Organizations that are adopting this innovative technology should continue to explore the most effective and efficient processes for VM to contribute to the development of best practice guidelines for the use of VM for patient safety.