Alzheimer's disease (AD) is the most common of the progressive dementias and leads to severe cognitive impairment in its later stages. Although AD can occur during the middle years, the risk increases with age. Because those who are living to be age 85 and older are the most rapidly growing segment of the population, it is expected the number of individuals affected with AD will increase proportionately. With an estimated 4 million Americans in the United States who have AD, it has become a major health concern (National Institute on Aging, 1995).
Elderly individuals undergo twice the number of hospital admissions and remain hospitalized twice as long as younger counterparts (Graves, 1992; Lewis, 1995; Travis, Moore, & McAuley, 1991). The impact of hospitalization with its institutional environment, separation from familiar caregivers, and challenging tests and procedures may precipitate behavioral changes. When cognitively impaired patients, regardless of age, experience diagnostic testing, their memory deficits and cognitive impairment may exacerbate altered behaviors further (Volicer, Fabiszewski, Rheaume, & Lasch, 1988). To date, no nursing research has been published examining the observed behaviors of older adults with AD undergoing invasive diagnostic procedures, such as a positron emission tomography (PET) scan or lumbar puncture (LP), which are described in this article.
The purpose of this study was to identify and examine the behavioral activity of patients hospitalized for diagnostic procedures in mild, moderate, and severe stages of AD.
The literature reveals no specific studies examining the behaviors of cognitively impaired patients related to invasive diagnostic procedures. In one study by Etienne et al. (1981), the responses of elderly patients with dementia to noninvasive diagnostic tests revealed the change in environment and the impact of psychological testing altered the functional abilities of the patients. It was noted further that patients who were least demented showed the most adverse reactions.
Figure 1. Data collection form.
Although there is a paucity of literature related to the more general effects of hospitalization on elderly individuals, Binder and Robins (1990) did find age and cognitive status were the most important predictors of hospitalization. They demonstrated, by this study, that patients with greater cognitive impairment had more hospital admissions and their length of stay was more prolonged than individuals who were not cognitively impaired. Massey and Riley (1982) cited reactions of older adults to hospitalization and found that sensory loss and decreased flexibility in coping with immediate stress were primary causes of difficulty with relocation and procedures experienced in the hospital.
Robinson and Demuth (1985) reported that both sensory loss and a slower reaction time in elderly individuals can alter patients' adaptive responses to the stimuli produced by the hospital environment and the equipment used for diagnostic procedures. Hall (1991) identified the importance of a supportive environment, consistency, communication, and prevention of fatigue as elements which promote a feeling of wellbeing and niinimize the impact of hospitalization on patients with dementia.
The few studies on hospitalized, cognitively intact elderly individuals demonstrate a need to address the ways in which invasive procedures can impact individuals. Aldwin (1991) concluded the perceived control of a situation had a significant bearing on the coping ability of elderly patients. Recognizing individuals' emotional states and assessing their coping abilities are areas in which nurses can help patients effectively.
The severity of the patient's cognitive impairment may play an important role in determining the response a patient may have to a procedure. This supposition is supported by Cohen, Kennedy, and Eisdorf er (1984) who focused on the need for caregivers to assess and address patient needs at varying stages of the illness. Cohen and colleagues (1984) also emphasized the value of identifying expected patient behaviors to design more sensitive and individualized care plans.
Because the literature reviewed did not address the specific needs of elderly, hospitalized, cognitively impaired patients, the aim of this study was to identify behaviors that occurred when these patients were undergoing invasive procedures and how those behaviors related to the stage of the disease.
A retrospective review of 30 patient medical records was performed, with confidentiality maintained by assigning each patient's medical record a numerical code. A Data Collection Form (Figure 1) was used to document information related to demographics, psychometric test scores, and Mini-Mental State Examination (MMSE) score (Folstein, Folstein, & McHugh, 1975). Data from the nursing notes, which described the behavior prior to, during, and following a PET scan or an LP were reviewed and recorded. The admission assessment was included in the review so the investigators could identify deviations from the patients' usual behaviors.
A pilot study was conducted to evaluate the validity of the Data Collection Form and to determine interrater and intrarater reliability of the nurse investigators.
For the content analysis, investigators reviewed patients' medical records and extrapolated information which reflected the patients' observed behaviors prior to, during, and following a PET scan or LP. It was necessary to determine consistency among the investigators regarding the relevance and perception of the data being collected.
To estimate interrater reliability, the principal investigator selected five medical records. One hospitalization from each record then was selected, and 5 consecutive days of charting were identified: 2 days prior to the procedure, the day of the procedure, and 2 days following the procedure. Each day's nursing notes were formatted with one theme or sentence per line. The investigators were asked to rate the sentences (i.e., unit of analysis) using the following abbreviations: S for significantly altered behaviors or NS for not significantly altered behaviors. For example, a patient was incontinent the night of the PET scan. This sentence was rated NS because this behavior was not an unusual occurrence for the patient. Another patient began to pace after being informed of a scheduled PET scan. This sentence was rated S because the patient had no prior history of pacing.
Coefficient alphas were performed to determine agreement. If the alphas were found to be less than .80, discussions were held to reduce variance, and the pilot then was repeated for interrater reliability. Agreement percentage ranged from 64.5% to 94.5%, with a mean of 84.5%.
STAGES OF DEMENTIA
Two weeks after the initial pilot study, each investigator was asked to rate the same data again in a similar way, and Pearson correlations were calculated using the investigator's first and second ratings. This technique helped determine the degree to which ratings were influenced by temporal factors. Intrarater reliability was found to be consistently greater than .80.
A convenience sample of 30 medical records was selected from a pool of approximately 100 participants in a research program of The National Institute on Aging, Laboratory of Neurosciences, Brain Aging and Dementia Section (NIA/LNS/ BADS) at the National Institutes of Health (NIH). Patients were staged according to their MMSE score (Table 1). Records included in this sample consisted of 10 subjects in each stage of AD who had undergone an LP or a PET scan procedure.
The 19 men and 1 1 women whose medical records were selected were between the ages of 47 and 92, with a mean age of 67. Prior to admission, they had been screened for other potentially confounding diseases or conditions such as diabetes, stroke, and depression, and had been medication-free for at least the previous 2 weeks.
Data analysis and data collection occurred simultaneously. This was considered an appropriate and necessary component of the study because the categories describing behavioral activity were expected to emerge from the data (Patton, 1980).
To ensure categories were developed across all records, consensus conferences were held after the pilot study and following the first five record reviews. Common themes emerged from the nurses' descriptive notes of patient behaviors. These behaviors were discussed and categorized (Table 2). A review of all 30 records then was completed. A chisquare analysis was performed to determine if there were relationships among the categories of behaviors and the stages of dementia.
CATEGORIES OF BEHAVIORS
Figure 2. Frequency of observed behaviors.
A cumulative frequency of observed behaviors was tabulated. Table 2 lists the 20 categories of behaviors in order of frequency of occurrence. Figure 2 shows the number of documented behaviors for each patient group during the 5-day observation period. Day -2 and Day -1 represent the 2 days prior to test, Day 0 is the day of the test, Day +1 and Day 4-2 represent the 2 days following the test.
This study revealed a significant difference in the frequency of behaviors (chi square = 4.18, ? = .041) when all groups and test types were combined. There was no difference in frequency when the type of invasive procedure (i.e., LP or PET scan) was examined separately. This would suggest the type of invasive procedure made no difference in behavioral activity.
When the patients were grouped separately by stage of AD, the mildly impaired subjects showed a significant increase in the number of behaviors on the test day (chi square = 4.051, p = .044). The moderately impaired group showed the most significant curve over the 5 -day observation period (chi square = 9.384, p = . 052).
The severely impaired group exhibited a consistently high level of activity throughout the 5-day observation period. Therefore, the curve over the 4-day period is less striking (chi square = 1.37, p = .850).
The results of these data suggest there were observable behavioral changes in patients with dementia who were hospitalized for invasive procedures. Patients in all three stages of dementia exhibited a change from their usual patterns of behavior. The most frequently observed behaviors were restlessness, confusion, anxiety, agitation, somatic complaints, disrupted sleep patterns, and withdrawal (Table 2). It is thought that patients with moderate cognitive impairment showed the greatest curve in behavioral activity because of their fragmented awareness of the environment and because they retained a degree of insight into the nature of the medical procedures. Therefore, when informed of impending procedures, their level of anxiety increased sooner and lasted longer than those who were either less or more cognitively impaired. Behaviors of patients in the mildly impaired group peaked dramatically on the day of the test, suggesting that because these patients were less cognitively impaired, they had a better understanding of the procedure they were having that day. It could not be established in the severely impaired group if their consistently higher levels of behavioral activity throughout the 5-day period was in response to the hospitalization, the procedures, the effects of the illness, or a combination of any or all of these.
The researchers are aware of the limitations which accompany any content analysis. In this study, the nursing documentation was being reviewed. Although precautions were taken by executing a pilot study, there is always the potential for limited validity of the nurses' documentation of behaviors.
The information derived from this research demonstrates the importance of comprehensively assessing each demented patient's stage of illness and usual behaviors at the time of admission to an acute care setting so care providers may initiate the optimal interventions to support patients undergoing diagnostic tests. Further, it is clear that ongoing reassessments throughout the hospitalization period are essential to facilitate the patient completing all the required diagnostics.
The impact of hospitalization and complex medical procedures may be minimized when the psychological and social needs as well as the physical needs of this patient population are anticipated and met through planned interventions. For example, for patients in the moderate stage of illness, providing teaching related to the procedures may be less anxietyprovoking if completed just prior to and continued step-by-step throughout the procedure. Teaching individuals in the mildly impaired group could be accomplished 1 or 2 days prior to the procedure. With individuals who are severely impaired, teaching will not be remembered. With the expected higher level of activity, patients in the severely impaired group will need maximum observation and support throughout their hospitalization. Observation and support needed may include, but not be limited to, any of the following interventions:
* Observing nonverbal communication for clues to patient's mood and providing gentle reminders.
* Constant reassurance.
* Therapeutic touch.
* Step-by-step instructions.
* Calm and structured environment.
Particular attention to patients' usual behavior and stage of disease are important to the achievement of desired outcomes. Patient and family teaching plans should be individually designed and revised as dictated by changing patient behaviors.
Little research has been conducted on the impact of hospitalization and medical treatment of elderly patients with dementia. Further research could provide caregivers with the information needed to refine their assessments and delineate the most effective intervention strategies for care of patients with dementia who are undergoing invasive procedures in an acute care setting.
- Aldwin, CM. (1991). Does age affect the stress and coping process? Implications of age differences in perceived control. Journals of Gerontology, 46(A), 174-180.
- Binder, E., & Robins, L. (1990). Cognitive impairment and length of hospital stay in older persons. Journal of the American Geriatrics Society, 38, 759-766.
- Cohen, D., Kennedy, G., & Eisdorfer, C. (1984). Phases of change in the patient with Alzheimer's disease: A conceptual dimension for defining health care management. Journal of the American Geriatrics Society, 32, 11, 15.
- Etienne, P., Dastoor, D. Goldapple, E., Johnson, S., Rochefort, E., & Ratner, J. (1981). Effects of medical and psychiatric workup in six demented geriatric patients. American Journal of Psychiatry, 138, 520521.
- Folstein, M., Folstein, S., & McHugh, P. (1975). Mini-mental state: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189-198.
- Graves, E. (1992). 1990 summary: National hospital discharge survey. Vital and Health Statistics of the National Center for Health Statistics, 210, 1-11.
- Hall, G. (1991). This hospital patient has Alzheimer's. American Journal of Nursing, 45(\V), 52.
- Lewis, CB. (1995). Aging the health care challenge (3rd ed.). Philadelphia: Davis.
- Massey, E., & Riley, T. (1982). Managing the patient with dementia. Postgraduate Medicine, 71(5), 207-211, 214-215.
- National Institute on Aging. (1995). Progress report on Alzheimer's disease 1995 (DHHS Publication No. PHS 95-3994). Washington, DC: U.S. Government Printing Office.
- Patton, M. (1980). Qualitative evaluation methods. Beverly Hills, CA: Sage.
- Robinson, S., & Demuth, P. (1985). Diagnostic studies for the aged: What are the dangers? Journal of Gerontological Nursing, 11(b), 6-9.
- Travis, S., Moore, S., & McAuley, W. (1991). Comparison of hospitalization experiences for demented and nondemented elders: Findings of a retrospective chart review. Journal of Gerontological Social Work,l 7(1 -2), 35-46.
- Volicer, L., Fabiszewski, K., Rheaume, Y., & Lasch, K. (Eds.). (1988). Clinical management of Alzheimer's disease. Rockville, MD: Aspen.
STAGES OF DEMENTIA
CATEGORIES OF BEHAVIORS