Disruptive behaviors are common on inpatient geriatric units. A preliminary, prospective unpublished study was conducted by the authors. The study showed that of 147 consecutive geriatric admissions to acute general medicine wards, 70.6% of patients had behavioral events that interrupted health care delivery. These behaviors resulted in medication or treatment delays (38% increase), medication or treatment omissions (7% increase), injuries to patients (7% of behaviors resulted in injury), increased staff stress levels by selfreport (42% increase), and loss of a significant amount of staff time (average of 45 minutes per staff member per shift). Patients with disruptive behaviors also were found to have 2.6 times higher nursing home placement rate and a two-fold increased inpatient mortality rate compared to those without disruptive behavior even when controlling for seriousness of illness, dementia, depression, and delirium. Although disruptive behavior was found to have important consequences for patients and staff, the behavior was documented less than 1% of the time on medical charts. Despite the impact disruptive behaviors have on geriatric inpatients, little has been written concerning interventions.
Many authors have described interventions which have been or could be used with elderly patients in varying situations (Baltes, Neumann, & Zank, 1994; Bechtel & Printz, 1994; Beck, Modlin, Heithoff, & Shue, 1992; Bulechek & McCloskey, 1987; Gugel, 1994; Matthiesen, Sivertsen, Foreman, & Cronin-Stubbs, 1994; McDowell, Burgio, Dombrowski, Locher, & Rodriguez, 1992; McMahon, 1988; Mintzer et al., 1994; Snyder, Egan, & Burns, 1995; Stock & Milan, 1993; Watson & Deary, 1994; Yurick, Burgio, & Paton, 1995). Yet there are no prospective studies that identify, categorize, and evaluate the effectiveness of interventions used for different types of disruptive behavior in hospitalized geriatric patients. Research in the area of disruptive behavior interventions is needed to identify those interventions that work best in a given clinical context.
The goals of this study were to:
* Identify and categorize disruptive behaviors of hospitalized geriatric patients.
* Identify and categorize interventions used by nursing staff to address specific disruptive behaviors.
* Determine the effectiveness of these interventions.
A major obstacle to the study of disruptive behaviors is terminology (Grobe, 1990; McCloskey et al., 1990). The common synonyms such as aggression and agitation have no well-defined or common meanings among RNs, PhDs, or MDs (CohenMansfield & Billig, 1986). Therefore, using a multidisciplinary team, a taxonomy was constructed for disruptive behaviors that was shown to be reliable and valid across disciplines (using card sort and cluster analysis). Disruptive behaviors are defined as those behaviors that interrupt the delivery of health care from the perspective of the provider (e.g., changing the work routine, stopping or abbreviating care delivery to another patient). The taxonomy has two main categories of disruptive behaviors: hyperactive and hypoactive. Each is subdivided into physical and verbal behaviors. See the Sidebar on this page for definitions.
Using this terminology, a qualitative observational methodology was employed to describe interventions used by nursing staff members on the unit. Real-time observations were made concerning the type of behavior exhibited, how nurses responded to the behavior, the effectiveness of the intervention(s) used to decrease or stop the behavior, and the impact of the behaviors on nursing work flow.
The research was conducted on a 12-bed GEM acute geriatric unit at a large teaching hospital. This unit has 12 beds with a central nursing station. This allows visualization of all areas, especially of an open common area occupied by those patients with special needs, such as those who wander. The nurse-patient ratio is about 4.5:1. The nursing staff has a 70% RN to 30% non-RN mix. Many of the staff are certified in geriatrics, rehabilitation, or both. The GEM admission criteria exclude patients who are severely demented (requiring 24-hour supervision or 100% nursing care), are terminally ill, or will require nursing home placement. The philosophy of the unit is one of no restraints, physical or chemical, except to promote treatment regimens such as artificial feedings. However in emergent situations (e.g., violent delirium) physical restraints may be used temporarily until further medical evaluation is possible to determine the cause. No such episodes occurred during this real-time observation study. A multidisciplinary team functions on the unit, and all members may contribute to the care of patients. However during this observation period, the focus was only on nursing staff who were observed implementing patient interventions for disruptive behaviors.
INTERVENTION CATEGORIES AND DEFINITIONS
The unit of analysis in this qualitative study was an "event," defined in time by the first recognition of disruptive behavior by a caregiver and ending at the conclusion of the caregivers' response or intervention. Events may involve more than one disruptive behavior or more than one response or intervention. For example, patients might pull at an invasive line and then strike the nurse who comes to restrain their hands, demonstrating two disruptive patient behaviors and one nursing response in a single event. "Nonevents" (a lack of disruptive behaviors by patients who had previously exhibited them) also were investigated to determine differences in circumstances surrounding the change in behavior. Possible responses to any behavior included "doing nothing."
Two RN research observers collected all data. These observers had 2.5 years of prior data collection experience on the GEM aunit and were familiar with nursing staff and care routines for this unit. Data were collected for 8 weeks, with an average of five shifts per week. All observations were randomly distributed among day, evening, and night shifts. Real-time observation of behavioral events was followed immediately with interviews of RNs, LPNs, or nurse aides involved. Interviews determined if:
* The caregiver viewed the behavior as disruptive.
* The behavior disrupted the care plan.
* The caregiver viewed the response or intervention as successful.
SINGLE/SEQUENTIAL INTERVENTION SUCCESS: SINGLE DISRUPTIVE BEHAVIOR EVENTS
Other caregiver comments or reactions regarding the event also were cataloged.
Observations and interviews were coded and downloaded into a specially designed database. Coding of information was used to identify internally homogeneous and externally heterogeneous categories. Accepted qualitative analysis methods were used to establish data trustworthiness and credibility including habituation of the observer on the GEM, triangulation, peer debriefing, and member checking (Mariano, 1993). Qualitative data saturation was achieved; no new relevant data emerged, and the relationships among categories were well established.
Disruptive Events and Related Interventions
Because qualitative methodology evolves over time, events that occurred earlier in the study had less information content than later events. All events were grouped and analyzed according to information completeness. Therefore, the tables in this article may not contain data from all events. Research observers collected information on 97 events containing 113 behaviors by 87 patients. An average of 3 disruptive behaviors per shift were documented. Each of these behaviors disrupted the flow of patient care as noted by the observers. Of the 113 behaviors noted, 7 disruptive behaviors had insufficient information to allow categorization. The remaining 106 disruptive behaviors included 43 (41%) hyperactive-physical, 33 (31%) hyperactive-verbal, and 30 (28%) hypoactive-physical. No hypoactive-verbal behaviors were noted. Nursing staff recognized 53 of these behaviors as being disruptive, or 50% of those so identified by the observers.
Discrepancies in recognition varied according to the specific type of behavior manifested. The discrepancy occurred 30% of the time with hyperactive-verbal behavior, 49% of the time with hyperactive-physical behaviors, and 73% with hypoactive-physical behaviors. Reasons provided by nursing staff for not identifying hyperactive behaviors as disruptive were categorized as:
* Lack of perceived time interruption (e.g., "It's not disruptive if I have time to deal with it.").
* Lack of perceived task interruption or interference with the completion of the care plan (e.g., "I wasn't doing anything anyway.").
* Patient's behavior was expected (e.g., "That's normal because they [patients] are demented, confused, anxious, or severely ill.").
The reasons given by nursing staff for not identifying hypoactive-physical behaviors as disruptive were divided into two categories:
* Perceived normal hospital behavior (e.g., "He just goes to bed out of boredom.").
* The patient's behavior allows more time for other task completion (e.g., "It allows more time for me to take care of another patient's needs.").
Researchers and staff agreed that behaviors were disruptive if:
* The provider had to interrupt caregiving priorities or goals for that patient (24 behaviors).
* It took time, interrupted, or interfered with care of another patient (11 behaviors).
* It caused more work or increased time required to complete a caregiving task (11 behaviors).
* A patient's safety was threatened (4 behaviors).
* It increased caregiver stress or frustration (3 behaviors).
A total of 139 interventions were observed during events. Table 1 identifies interventions and divides them into 16 categories. Observers documented that interventions were used in two ways:
* One intervention at a time or one after another (sequential). When one intervention failed, another one was implemented. For example, a patient did not return to bed after being told to do so (verbal command) and was taken by the hand and led to bed (direct physical action to patient).
* Simultaneous interventions where multiple modalities were used at the same time. For example, a sheet was placed over a patient's urinary catheter to decrease visualization (decrease or remove stimulus) while the patient was being directed to not touch the catheter (verbal command).
Success of an intervention was defined as an immediate positive modification of the disruptive behavior following the implementation of an intervention. Success was judged separately by the caregiver and the observer and coded independently: -1 = made the behavior worse, 0 = had no effect, 1 = was effective, or 2 = was not interpretable. An intervention was deemed successful only if the caregiver and the observer both coded 1 . Agreement of the success of an intervention occurred 96% of the time between staff and researchers. No intervention (i.e., doing nothing) was coded by the observer as worsening the behavior for only one episode.
Eighty-four percent of all cases had a single disruptive behavior treated with a single intervention. Table 2 contains single/sequential intervention successes for different types of disruptive behaviors. Only 10 of the 16 different categories of interventions were used in single disruptive behavior events. Unused categories included: chemical restraint, decrease or remove stimulus, direct physical action to environment, physical restraint, praise, and reasoning with significant other. The success rate for all single interventions with disruptive behavior was 45%. Interventions using relaxation techniques, humor, reasoning, and doing nothing were observed to be noneffective.
PERCENT OF TIME INTERVENTIONS WERE USED ACCORDING TO LEVEL OF EDUCATION OF NURSING STAFF MEMBER
Five events contained single disruptive behaviors treated with simultaneous interventions. Two cases were hypoactive-physícal and were successfully treated with a combination of verbal command plus direct physical action to the patient and the environment. The three other cases were hyperactive - physical. One of these cases responded to reasoning plus verbal command; the other two responded to orientation plus direct physical action to the patient. Simultaneous interventions were successful in all of these events with single disruptive behaviors.
While there were no cases of multiple disruptive events treated with single interventions, five events contained multiple disruptive behaviors treated with simultaneous interventions. Two cases showed hyperactive-verbal and hyperactive-physical behaviors. One simultaneous intervention of verbal orientation plus decreased stimulus was unsuccessful, the other intervention that consisted of decreased stimulus plus relaxation techniques plus verbal command was successful. One case of combined hypoactive-physical and hyperactive-verbal behavior responded favorably to physical or mental diversion plus praise plus planning or contracting plus decreased stimulus. One case of two types of hypoactivephysical behavior responded well to a combination of planning or contracting plus reasoning. The final case consisted of multiple hyperactiveverbal behaviors that did not respond to reasoning combined with doing nothing. In cases of multiple disruptive behaviors occurring simultaneously, verbal orientation plus reasoning were associated with failure unless coupled with a physical intervention.
Influence of Educational Level on Interventions
The types of interventions invoked by nursing staff from different educational levels are shown in Table 3. Only the more common interventions (i.e., those used more than five times) have been reported in this article. In addition, no attempt was made to stratify the results by level of RN education nor time in practice for any group.
Patients were identified who had prior disruptive behaviors, but currently did not show any such behaviors. The nursing staff were asked to identify reasons the disruptive behaviors were no longer present. The reasons cited included:
* Improved medical condition (e.g., less pain).
* Decreased stimulus (e.g., moved to quieter room).
* Becoming accustomed to a new setting.
* Mental distraction, spending time with the patients, or providing activities for them.
* Application of planning, contracting, or behavior-modification techniques.
* Removal of items that may harm a patient or that the patient may damage.
* Providing assistance devices (e.g., a walker for a hypoactivephysical patient).
This study was a preliminary investigation of disruptive behaviors and the associated nursing interventions currendy used on an acute geriatric medicine ward. Interpretation of these results should be guarded because of the relatively small sample size, the contextual nature of qualitative techniques, and the extremely complex nature of the subject. However, these findings suggest interesting relationships regarding disruptive behaviors, intervention use, and staff perceptions. The authors were able to pilot a taxonomy for disruptive behavior that was easy for staff and researchers to use. This led to a categorization of currently used interventions in use on the GEM unit.
These categories should be refined and piloted in other locations. There was an unexpected discrepancy between the nursing staff and researchers' identification of disruptive behaviors. An observed disruption was perceived by nursing staff to occur only 50% of the time. This finding has great implications for future studies concerning disruptive behaviors because most studies rely on nursing reports. Disruptive behavior that interferes with care delivery may be underestimated half of the time if nursing reports are used as a "gold standard." The reasons given for nonrecognition (e.g., "had time to deal with it") also provide insight into the results of this study. Staff perceptions about interruptions of their time, interference with tasks to be performed, or attitudes regarding hospitalized patient behavior determine if a patient's behavior is interpreted as disruptive. No cases of "false" identification of disruptive behavior were documented. False identification is a potential problem that may lead to inappropriate interventions including misuse of medications.
There are practical clinical implications in being able to identify factors that contribute to staff recognition of disruptive behaviors. Because time management and task or care completion are important practical issues in any care setting, staffing decisions should take into account hyperactive and hypoactive disruptive behaviors when assigning patient acuity. Interestingly, if staff perceived they had adequate time for task completion, patient behavior that interrupted care delivery was not perceived as disruptive. This suggests that the nursing staff is more sensitive to time management and less sensitive to task progress.
Another critical factor involves the differences in identification of behaviors as disruptive. Although several permutations of the taxonomy occurred, it was apparent that the term "disruptive behavior" is defined differently by each staff member. It is not an objective term. An assumption was made that the defined examples of hyperactive and hypoactive behaviors would be disruptive for nursing care. The results indicate that further study to refine this concept is needed to understand the subjective nature of these terms. Some of the responses from nursing staff regarding their definition of disruption indicate that determination of the meaning of the phenomenon of disruption is a valid topic for further research (Anderson, 1991). If staff anticipate the behavior of a hospitalized patient will be disruptive, this could be an important factor in identification of behavior as disruptive. The expectation of disruption expressed by this staff may have decreased the labeling of behavior as disruptive for care delivery.
Noteworthy is the finding that hypoactive-physical disruptive behaviors occurred almost as frequently as hyperactive-physical and hyperactive-verbal disruptive behavior. The former was recognized far less often. The finding that hypoactive behaviors were not designated as disruptive to nursing care is of concern. This lack of identification could lead to varied consequences, including misdiagnosis of comorbid conditions such as depression (Alexopoulos, Meyers, Young, Mattes, & Kakuma, 1993), lack of recognition of delirium associated with medications or infection (Lyness, 1990), or delay of interventions resulting in poor prognosis for the patient (Francis & Kapoor, 1992). The need to maintain an attitude of assessment for the purpose of suggesting differential diagnoses rather than focusing on behavior as the problem is suggested by this finding.
Hyperactive behaviors increased the nurses' perceptions of disruption to nursing care. The greatest agreement regarding disruptive behavior was in the behavioral category of hyperactive-verbal disruption among patients with repeated requests or complaints. Because all categories of disruptive behavior interfered with patient care by objective observation, these findings have practical impact for changing patient acuity designations and basing work assignments on patient behavior not just medical diagnosis. Intervention trials are needed to identify other potential modifiable factors within the personal or physical environment that may improve patient care and staff satisfaction when dealing with patients who demonstrate disruptive behaviors (Kikuta, 1991; Thomas, 1997).
Following the identification of disruptive behaviors by both staff and researchers, a high level of agreement (96%) was found regarding the success of the interventions used by the GEM nursing staff. Staff used only one method of intervention 84% of the time when dealing with disruptive behaviors. Planning or contracting was found to be successful but was used only by RN staff. An example of a plan or contract would be: "You may go back to bed in an hour" or "I will come back and check on you in 15 minutes." Patients with hypoactive disruptive behaviors had an increased risk of "doing nothing" being the only intervention. The following reasons given by nursing staff for nonintervention with those patients further support the need to assure hypoactive as well as hyperactive patients receive needed care. One nurse said:
The patient was not demanding my attention so I had more time for my other patients.
If I wasn't so busy, I would have taken more time with him, but it was nice to have fewer patients to deal with.
This nursing staff recognized that doing nothing could have negative consequences for hypoactive patients, such as missed therapies or inadequate personal care.
Only 16% of cases used multiple interventions. These findings suggest simultaneous interventions improve success of care rates. When multiple interventions were used, the success rate increased from 45% to 80%. Some interventions such as reasoning and orienting the patient were used frequently, but were usually not effective when used alone. This supports the work of others (Gallagher-Thompson, Brooks, BIiwise, Leader, & Yesavage, 1992; McMahon, 1998) and suggests the effects of verbal orientation in confused older adults is limited and may actually be harmful. This study did not address the use of multidisciplinary teams to identify or treat disruptive behavior, although a multidisciplinary team exists on the GEM unit. The use of multidisciplinary teams in the identification and treatment of disruptive behaviors is another issue needing further study, based on results of use of such teams for other care issues (McDowell et al., 1992).
LIMITATIONS AND IMPLICATIONS FOR FURTHER RESEARCH
Several factors must be considered in relation to the outcomes of this study. Qualitative methods are useful to identify characteristics of a phenomenon but are not easily applicable to other settings. Therefore, replication at other sites is required. This study includes neither detailed demographics and personal characteristics of nursing staff and environmental factors nor information regarding the diagnosis, medications, or current health status of patients. No information about total time spent dealing with disruptive behaviors or time required to implement different types of interventions are currently available.
One perceived weakness of this analysis may be the lack of correlation between medical diagnoses and interventions used by nursing staff. The authors purposely chose not to correlate medical diagnoses with interventions, believing this would oversimplify a complex problem. Medical diagnostic category alone does not convey important conditions that modulate disruptive behavior (e.g., functional impairment, severity of illness, patient personality, interaction of comorbid disease, polypharmacy, environment, staff personality). These observations indicate some interventions used by staff are independent of medical diagnosis altogether. For example, staff educational level (i.e., RN versus non-RN) determined whether planning (i.e., planned time with patient prior to disruptive behavior) was used as an intervention. Non-RNs (LPNs and nurse aides) were not observed to use planning or contracting as an intervention. These observations indicate that nursing staff decisions concerning interventions for disruptive behaviors are complex and depend on poorly delineated factors but are more complex than matching medical diagnosis with intervention strategies. Describing these factors and assigning a relative importance to them was outside the scope of this study. Future research should incorporate sophisticated approaches to define and correlate patient, staff, and environmental variables in association with success rates to further nurses' knowledge regarding appropriate use of specific interventions.
In addition to these factors, future study also should evaluate potential ways to describe disruptive behavior severity and further categorize possible interventions for specific disruptive behavior types. An investigation of effective ways to document and communicate when disruptive behaviors occur also is needed. Many of these factors are currently being evaluated in a long-term study by the authors. Replication at other locations also will be vital.
Although disruptive behavior may affect all members of an interdisciplinary team, nurses are often the first individuals involved with patients exhibiting those behaviors. Based on the results of this study, at least three factors related to nursing practice appear to be critical to maintain the welfare of these patients. The first is the nurses' ability to recognize and document the context of behavior including the precipitating events, operational definitions of that behavior, and the consequences of behavior. As a result, nurses will be able to anticipate that the behavior could occur. The second factor is the initiation of an intervention that makes sense based on the total context of the behavior. Finally, to promote continuity of care, communication is essential with other nurses and other members of the health care team regarding these behaviors. This communication must include the interventions found to be effective because no one nurse can provide total care or have all the knowledge regarding effective interventions for a given patient in all situations. These events require intradisciplinary and interdisciplinary team efforts to promote understanding of the phenomenon of disruptive behavior and translate that understanding into knowledge-based care.
This study represents an attempt to understand interactions between disruptive behaviors and staff interventions on an inpatient geriatric ward. Initially, a taxonomy was developed by the researchers in an attempt to categorize the types of disruptive behavior observed. Preliminary findings suggest half of all behaviors that interfere with care are not noticed or identified by nursing staff as disruptive, and current interventions for disruptive behaviors when used singly produce poor results. By using simultaneous interventions, success rates indicating redirection of the behavior almost doubled. Additionally, multidisciplinary teams were not used as an intervention for disruptive behavior on this geriatric inpatient unit. Further study of disruptive behaviors in this population will lead to improved patient care, better patient outcomes, and increased staff satisfaction.
- Alexopoulos, G.S., Meyers, B.S., Young, R.C., Mattis, S., & Kakuma, T. (1993). The course of geriatric depression with "reversible dementia": A controlled study. American Journal of Psychiatry, 150, 16931699.
- Anderson, J.M. (1991). The phenomenological perspective. In J.M. Morse (Ed.), Qualitative nursing research (Rev. ed., pp. 23-38) Newbury Park, CA: Sage.
- Baltes, M.M., Neumann, E., & Zank, S. (1994). Maintenance and rehabilitation of independence in old age: An intervention program for staff. Psychology and Aging, 9, 179-188.
- Bechtel, G.A., & Printz, V. (1994). Evaluating quality of care using modular nursing on a multispecialty unit. Clinical Nurse Specialist, 8, 81-84.
- Beck, C, Modlin, T., Heithoff, K., & Shue, V. (1992). Exercise as an intervention for behavior problems. Geriatric Nursing, 13, 273-275.
- Bulechek, G.M., & McCloskey, J.C. (1987). Nursing interventions: What they are and how to choose them. Holistic Nursing Practice, 2, 36-44.
- Cohen-Mansfield, J., & Billig, N. (1986). Agitated behaviors in the elderly: A conceptual review. Journal of the American Geriatrics Society, 34, 71 1-721.
- Francis, J., & Kapoor, W.N. (1992). Prognosis after hospital discharge of older medical patients with delirium. Journal of the American Geriatrics Society, 40, 601-606.
- Gallagher-Thompson, D., Brooks, J.O., III, Bliwise, D., Leader, J., & Yesavage, J.A. (1992). The relations among caregiver stress, "sundowning" symptoms, and cognitive decline in Alzheimer's disease. Journal of the American Geriatrics Society, 40, 807-810.
- Grobe, SJ. (1990). Nursing intervention lexicon and taxonomy study: Language and classification methods. Advances in Nursing Science, S, 22-33.
- Gugel, R.N. (1994). Behavioral approaches for managing patients with Alzheimer's disease and related disorders. Medical Clinks of North America, 18, 861-867.
- Kikuta, S.C. (1991). Clinically managing disruptive behavior on the ward. Journal of Gerontological Nursing, /7(8), 4-8.
- Lyness, J.M. (1990). Delirium: Masquerades and misdiagnosis in elderly inpatients. Journal of the American Geriatrics Society, 38, 1235-1238.
- Mariano, C. (1993). Case study: The method. In P.L. Munhall & CO. Boyd (Eds.), Nursing research: A qualitative perspective (pp. 311-337). New York: National League for Nursing Press.
- Matthiesen, V., Sivertsen, L., Foreman, M.D., & Cronin-Stubbs, D. (1994). Acute confusion: Nursing intervention in older patients. Orthopedic Nursing, 13, 21-28.
- McCloskey, J.C., Bulechek, G., Zichi, M., Cohen, M., Craft, M.J., Crossley, J.D., Denehy, J.A., Glick, O.J., Kruckeberg, T, Maas, M., Prophet, CM., & TrippReimer, T. (1990). Classification of nursing interventions. Journal of Professional Nursing, 6, 151-157.
- McDowell, B.J., Burgio, K.L., Dombrowski, M., Locher, J.L., & Rodriguez, E. (1992). An interdisciplinary approach to the assessment and behavioral treatment of urinary incontinence in geriatric outpatients. Journal of the American Geriatrics Society, 40, 370-374.
- McMahon, R. (1988). The "24-hour reality orientation" type of approach to the confused elderly: A minimum standard for care. Journal of Advanced Nursing, 13, 693-700.
- Mintzer, J.E., Lewis, L., Pennypacker, L., Pitner, J., Simpson, W., Bachman, D., Wohlreich, G.M., Meeks, ?., Hunt, S., Herman, K., & Sampson, R. (1994). A new approach for the management of acute psychiatric disorders in elderly demented patients. Journal of South Carolina Medical Association, 90, 373-376.
- Snyder, M., Egan, E.C., & Burns, K.R. (1995). Interventions for decreasing agitation behaviors in persons with dementia. Journal of Gerontological Nursing, 21(7), 34-40.
- Stock, L.Z., & Milan, M. A. (1993). Improving dietary practices of elderly individuals: The power of prompting feedback and social reinforcement. Journal Applied Behavior Analysis, 26, 379-387.
- Thomas, D.W. (1997). Understanding the wandering patient: A continuity of personality perspective. Journal of Gerontological Nursing, 23(1), 16-24.
- Watson, R., & Deary, LJ. (1994). Measuring feeding difficulty in patients with dementia: Multivariate analysis of feeding problems, nursing intervention and indicators of feeding difficulty. Journal of Advanced Nursing, 19, 283-287.
- Yurick, A., Burgio, L., & Paton, S.M. (1995) Assessing disruptive behaviors of nursing home residents. Journal of Gerontological Nursing, 2/(4), 29-34.
INTERVENTION CATEGORIES AND DEFINITIONS
SINGLE/SEQUENTIAL INTERVENTION SUCCESS: SINGLE DISRUPTIVE BEHAVIOR EVENTS
PERCENT OF TIME INTERVENTIONS WERE USED ACCORDING TO LEVEL OF EDUCATION OF NURSING STAFF MEMBER