| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address correspondence to Helen D. Davies, Veterans Affairs Palo Alto Health Care System, 3801 Miranda Avenue, 116F, Palo Alto, CA 94304. E-mail: hddavies{at}stanford.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Acute care • Behavioral disturbance • Older adults
As the population of older adults continues to swell, the need to address the problems of an increasing number of hospitalized elderly will only become more important. Behavioral incidents have significant negative consequences for staff and patients. In a recent investigation on acute care units, clinical staff self-reported that behavioral problems were often encountered and were considered burdensome (Cassidy et al., 2001). However, data suggest that the actual incidence of agitation and other behavioral problems in these settings are underreported. A survey of nurses' reporting of problematic behaviors revealed that 36% of nurses held the belief that not all incidents needed to be reported (Elnitsky, Nichols, & Palmer, 1997). Traditionally, the literature has focused on the underreporting of the broader category of adverse events rather than specifically studying behavioral problems. Sutton, Standed, and Wallace (1994) found that hospital staffs were prone to underreport accidents, and Stanhope, Crowley-Murphy, Vincent, O'Connor, and Taylor-Adams (1999) found that staff members did not report 55% of adverse incidents.
Accurate reporting of behavioral problems in hospitalized elders is important because, in addition to the burden these behaviors place on clinical staff, patients exhibiting such problems are likely to be at increased risk for incidents that include adverse events and accidents. Yet there are few investigations of the reporting of behavioral problems in older adults in acute care settings. We conducted a small, descriptive investigation to compare the incidence of behavioral problems reported by nursing staff on the Neuropsychiatric Inventory–Questionnaire (NPI-Q; Kaufer et al., 2000) with those reported in patients' medical charts during the same time period.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Measures
Neuropsychiatric Inventory–Questionnaire (NPI-Q)
This questionnaire, developed and cross-validated with the standard NPI (Cummings et al., 1994), provides a brief assessment of neuropsychiatric symptomatology and behavioral problems as well as ratings of staff distress (Kaufer et al., 2000). The NPI-Q measures 12 categories of behavioral disturbance: (a) delusions, (b) hallucinations, (c) anxiety, (d) depression and dysphoria, (e) agitation and aggression, (f) elation and euphoria, (g) disinhibition, (h) irritability and lability, (i) apathy and indifference, (j) motor disturbance, (k) nighttime behavior problems, and (l) problems with appetite and eating. A caregiver (in this study, a member of the nursing staff) reports whether the behavior was observed and ranks the severity of the behavior exhibited on a scale of 1 to 3, with 3 being the most severe. The caregiver also ranks the level of distress caused to them by the behavior, on a scale of 1 to 5, with 5 being the most severe. The NPI-Q takes approximately 10 minutes to administer. One of two trained research assistants, independent from those conducting chart reviews, verbally administered the NPI-Q. In the current study, the caregiver was a member of the nursing staff who attended to the patient during an 8-hr shift on each of his first two days of hospitalization.
Patients' Chart Reviews
Three independent reviewers—a medical practitioner, a research psychologist, and a psychiatric clinical nurse specialist—reviewed the charts and recorded the number of behaviors observed. All notes by all staff were included in the patients' charts, and all notes were reviewed, independent of source reviewed. A consensus meeting was held to compare results, and discrepancies were resolved. The reviewers were blind to the NPI-Q data. All documented evidence of behavioral problems (e.g., delusions, hallucinations, sleeping problems, depression, anxiety, disinhibition, irritability, motor disturbance, and excessive elation or euphoria) was noted. Additionally, all reports of any physically and verbally aggressive behaviors and nonaggressive behaviors indicative of agitation and aggression were noted. Physically aggressive behaviors were defined to include hitting, scratching, grabbing, pushing, tearing, throwing objects, spitting, or kicking. Physically nonaggressive behaviors included wandering, pacing, attempting to exit facility, intruding into inappropriate places, inappropriate robing or disrobing, voiding in inappropriate places, and handling things inappropriately. Verbally aggressive behaviors were defined to include cursing, swearing, screaming, yelling, making odd noises, or temper outbursts. Finally, verbally nonaggressive behaviors included constant requests for attention, complaining, whining, negativism, repetition of sentences, and rambling disjointed sentences. We established the total number of behavioral problems documented across the first 2 days postadmission.
Procedures
The NPI-Q was administered 2 days postadmission to a member of the nursing staff (a licensed vocational nurse or an RN) who attended to the patient during an 8-hr shift on each of his first 2 days of hospitalization and who stated that he or she had significant contact with the patient. We chose to compare the NPI-Q and chart review across the first 2 days postadmission because a significant number of patients on these units are likely to be discharged or transferred at this point. Furthermore, after 2 days we were able to identify nursing staff who had similar levels of exposure to the patients in the study.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
While the occurrence of one behavioral problem may not appear to be a significant burden, the cumulative impact of numerous problem behaviors over a short time span may be disruptive to staff. Indeed, in a previous investigation we found that nursing staff, similar to those in the current study, reported significant burden from dealing with behavioral problems among older hospitalized patients (Cassidy et al., 2001). The reasons for underreporting are unclear. In the Cassidy and colleagues study, staff indicated that contending with behavioral disruptions is "part of their job" and "they happen all the time." Reporting may also add the burden of time to already busy shifts, and staff may need support for recognizing the behavior problems that require documentation. Administration needs documentation of patient acuity, as this helps describe workload and promote the rationale for more resources. Future work should determine which behavior problems have the greatest risk for further escalation, thereby assisting staff in recognizing when documentation is essential.
Documentation alone is only the first step in dealing with such behavioral problems. Staff in acute settings may be less familiar with conditions of the older patient that may lead to behavioral problems. As the nature of the patient population shifts, training in the care of the older patient is quickly becoming a necessity in all settings. Further work also is required to determine appropriate interventions for controlling these problems and to identify the effectiveness of such interventions for both staff and patient.
Finally, the current study has several significant limitations. The small sample size and male-only patient settings limit the generalizability of the findings. Furthermore, documentation strategies may be institution specific. Additionally, patients were followed for only the first 2 days of hospitalization, and documentation strategies, or concern about behavioral problems, may increase with greater length of stay. Indeed, the discrepancies between the NPI-Q and chart ratings of behavioral problems observed during the short duration of the current study may provide a conservative estimate of the actual degree of underreporting of these problems. Replication of this study with a larger, mixed-gender sample, across the full duration of hospitalization will help to shed light on the prevalence of behavioral problems among older adults in acute care settings, the extent of underreporting of these problems, and the impact they have on both staff and patients.
|
Footnotes |
|---|
1 Veterans Affairs Palo Alto Health Care System, Palo Alto, CA.
2 Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford University, Stanford, CA.
3 Veterans Affairs Medical Center San Francisco, San Francisco, CA.
Decision Editor: Linda S. Noelker, PhD
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Minerva BMJ, August 13, 2005; 331(7513): 412 - 412. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
