Historie a Hy, research on wandering has been directed at reducing the effects without considering the motivation or reasons for the behavior. In particular, extensive effort over the last decade has been directed at changing the typical hospital environment through specialized care units to better manage wandering (Chafetz, 1991; Holmes et al., 1994; Sand, 1992; Swanson, 1993; Thomas, 1996; Volicer, Collard, Hurley, & Bishop, 1994; Zeisel, Hyde & Levkoff, 1994). Less attention, however, has focused on why people with dementia have an apparent need to wander. Rabins (1991) summarized this lack of understanding by suggesting, "what is most notable about wandering is how little we know about it." Understanding why an individual has a need to move about, in some cases 77% of one's wakeful time (Thomas, 1995), would be of immense value to nurses and other care providers who want to improve communication and develop effective management strategies. Regrettably, writings on the subject of wandering have emphasized potential interventions without adequately addressing the actual cause(s) of the behavior. When viewed in totality, however, a common underlying theme seems to surface within the literature - which is that wandering appears to reflect a continuity of premorbid behavior.
Several studies (Butler & Barnett, 1991; Coons, 1988; Hall, 1990; Snyder, Rupprecht, Pyrek, Brekhus, & Moss, 1978) suggested that wandering was purposeful in nature. Purposeful wandering was primarily attributed to a reenactment of prior leisure and work roles. Words used to describe purposeful wandering included reminiscence, recreation, leisure, industrious, work and familiar routines. The concept of "agenda behavior" (Rader, Doan, & Schwab, 1985) supports the idea of continuity. Agenda behavior was viewed as a way (or the wanderer to "recapture old satisfying situations which formerly brought feelings of safety and belonging." Studies have revealed (Cohen-Mansfield, Werner, Marx, & Freedman, 1991; Monsour & Robb, 1982) a continuity between the way wanderers historically dealt with stress (i.e., increased psychomotor activity).
The Continuity Theory of aging assumes that a person evolves throughout life based on prior history and experiences (Atchley, 1989). The individual is able to employ concepts of their past as a way to conceive and structure their future. This continuity of experiences was viewed by Atchley as an adaptive technique where the individual continued to practice what s /he was good at which, in turn, provided for feelings of satisfaction. The term habitual response (Arendt, 1977) is related to the concept of continuity. Habitual responses may provide the demented person with a feeling of competence and mastery through involvement in ingrained, familiar behaviors. This struggle to hang on to what is familiar may provide a clue as to why people wander.
Lawton (1992) discussed the continuity of personality types. Temperaments of introversion and extroversion were defined in relation to how people were motivated. Lawton maintained that introverted people typically seek stimulation from within themselves while extroverts seek stimulation from the external world. Further delineation between introversion and extroversion revealed that introverts are more inclined to seek security and extroverts are more prone to seek autonomy. An individual, throughout life, is normally able to self-regulate the amount of stimulation along the security-autonomy continuum to best reflect his/her comfort zone. Lawton suggested, however, that biological aging, especially when accompanied by chronic illness, i.e., dementia, forces the person to conserve his/her resources by giving greater priority to "interiority" and preference for security rather than autonomy. The extrovert, in a move toward "interiority," would assume more introverted characteristics such as an increased need for familiarity and security and motivation through threat of punishment rather than an expectation of reward.
CONTINUITY PATTERNS OF WANDERERS AND NON-WANDERERS
The wanderer, when viewed in this context, may be searching for security through a conditioned dependence on the external environment. Familiar relationships that were developed externally, i.e., social interactions, leisure activities and work experiences, are sought in an attempt to establish comfort within a confusing and irrational world created by the illness. In other words, the wanderer, based on biological and environmental constriction, experiences an elevated need for security. To find security, the wanderer continues a lifelong conditioned response of looking to the external world for familiarity and comfort.
In contrast, it might be speculated that the non-wanderer is capable of achieving security and comfort through internal processes which include inner feelings of stimulation and contentment rather than external activation. Under similar circumstances it could be reasoned that the non-wanderer is less inclined to actively pursue security outside his/her immediate proximity man the wanderer. The non-wanderer, in this case, would have less need, than the wanderer, to move about.
In short, one's biological temperament may play a role in whether or not a person wanders. Realizing the individual's prior personality characteristics and behaviors, within a continuity framework, would provide a better understanding of why people with dementia have a need to wander. In particular, knowing that the wanderer may be predisposed toward receiving a sense of satisfaction through interaction with the external environment - which was often reflected in leisure and social behaviors - would be of immense value to nurses and other care providers when conceptualizing strategies for interaction and program provision within the overall management of the wanderer. The purpose of the present study was to contribute to the body of knowledge needed to effectively manage the wanderer by exploring the relationship between personality types to wandering.
Subjects for the study were selected from a 911-bed nursing facility in the Philadelphia area. The subjects ranged in age from 64 through 94. All subjects had a diagnosis of a chronic form of dementia and were rated as either middle or late dementia as determined by the Global Deterioration Scale for Assessment of Primary Degenerative Dementia (Reisberg, Ferris, Leon, & Crook, 1982). The investigator developed two subject pools (wanderers and non-wanderers) based on the recommendations of the social service and nursing staff. The investigator provided the social service department, via an inservice training session, with the criteria for determining the qualifications of a wanderer and non-wanderer. The criteria was based on the most common characteristics derived from the literature (Table 1). The social service and nursing staff provided the investigator with their recommendations of residents that fit the criteria. Only 100% agreement among staff resulted in the resident's inclusion into the subject pool. There were a total of 31 wanderers and 37 non-wanderers assigned to each pool. A random selection was drawn from the wanderer's pool until a total of 20 was reached. Each wanderer was then matched as closely as possible with a non-wanderer based on age, gender and dementia level.
Criterio ior Wanderers and Non-wanderers
After the population was determined a modified version of the NEO-Five Factor Inventory (NEOFH) (Costa & McCrae, 1992) along with a demographic questionnaire was administered to a surrogate (significant other) of each subject. Selection of a surrogate was based on having a historical familiarity of the subject as detenriined by the social service staff. The most common surrogates were the subject's spouse or offspring. The NEO-FFI consisted of 60 items which were responded to by the surrogate through the use of a 5point Likert Scale. The NEO-FFI measured the five personality factors of neurotiäsm (N), extroversion (E), openness (O), agreeableness (A) and conscientiousness (C).
VAUDITY AND RELIABILITY
Global Deterioration Scale
Concurrent validity of the scale has been demonstrated through significant correlation of scores with 13 of the 19 cognitive items in the Inventory of Psychic and Somatic Complaints in the Elderly (Reisberg, Ferris, & Schneck, 1981). In addition, the Global Deterioration Scale correlated significantly (p<.01, N=43) with CT scan rankings of ventricular dilation (r=.62) and with CT scan cortical assessments of sulcal enlargement (r=.53) (Reisberg, Ferris, & Schneck, 1981).
Reliability and validity were established with the long version NEO Personality Inventory (NEO-PI). It was assumed that since the NECV-FFI was derived from the correlates with the NEO-PI that reliability and validity estimates will be similar if not somewhat lower. When correlated with the NEO-PI valid Max factors, the NEO-FFI scales showed correlations ranging from .75 for the (C) domain to .89 for the (N) domain. Internal consistency for the NEO-FFI scales was calculated using coefficient alpha. Coeffìdents were .86, .77, .73, .68, and .81 for N, E, O, A, and C, respectively. Validity and reliability of the NEO-PI has been well documented (Costa & McCrae, 1988; McCrae & Costa, 1982,1986,1989)
Scores between the two groups were statistically compared using an independent f-test and chi-square test (.05 level of significance) for each of the demographic and personality factors. In addition, a qualitative follow-up was conducted based on a 25% random sample of the two groups, which further analyzed the statistically significant quantitative results.
Of the five personality factors, extroversion and agreeableness scores were found to be statistically significant. Wanderers scored higher in extroversion (t=2.76) with a mean score of 31.45 compared to 24.7 for non-wanderers, and was significant at the ? 5S. 01 level. Wanderers also scored higher on agreeableness (t=1.66) with a mean score of 34.35 compared to 29.6 for non-wanderers, and was significant at the p <.05 level.
Further investigation of the extroversion and agreeableness factors found statistically significant differences between wanderers and nonwanderers on several sub-factors or facets. Wanderers had higher mean extroversion facet scores on warmth (?*?.05), gregariousness (? ^.05), activity (?5*. 05) and positive-emotion (pss.01) (Table 2). No statistically significant differences were found between the two groups on the extroversion facet scores of assertiveness and excitement-seeking. Wanderers also had a statistically significant higher mean score (p^.05) on the agreeableness facet of altruism compared to non-wanderers (Table 2). No statistically significant differences were found between the two groups on the agreeableness facets of trust, straightforwardness, compliance and tender-mindedness.
Two of the demographic variables (i.e., age and diagnosis) were found to significantly differentiate wanderers from non-wanderers (Table 3). As a group, the non-wanderers were older (X=83.7) than the wanderers (X=77.85) (t=2.82; p^.01). Also, the two groups differed in diagnosis (X2=8.64; p^.05). Wanderers were primarily diagnosed with Alzheimer's disease while non-wanderers were primarily diagnosed with vascular dementia. An analysis of covariance indicated that the covariant of age, when correlated to the dependent variables of extroversion (F=.02; p=.889) and agreeableness (F=.67; p=A19), showed no relationship. Likewise, when a two-way ANOVA was performed using group (wanderers and non-wanderers) and diagnosis (Alzheimer's and nonAlzheimer's) as independent variables, no relationship was found to the dependent variables of extroversion (F=.00; p=.95) and agreeableness (F=3.93; p=.06). It was concluded, therefore, that the demographic variables of age and diagnosis were not related to the significant finding and thus did not affect the results.
Differences In Personality Facets Between Wanderers and Non-wanderers
A typological analysis, which used a life history interview, revealed various classifications of behaviors that helped to differentiate wanderers from non-wanderers on each personality facet (Table 4). On the facet of warmth, wanderers best represented the typology of "openly-congenial" whereas the non-wanderers were better represented as "standoffish." On the facet of gregariousness, wanderers best represented the typology of "social-seeking" while non-wanderers were better represented as "social-accepting." On the facet of activity, wanderers best represented the typology of "multiple/continuous" whereas non- wanderers were better represented by "active /inactive." The positive-emotion facet was distinguished by three typologies. Wanderers were best represented as "generally optimistic" and "middle ground" while nonwanderers were primarily represented as "middle ground" and "generally pessimistic."
Test Results at Demographic Variables
The agreeableness facet of altruism was also comprised of three typologies. Wanderers were best represented by "giving/ general" and "giving /familiar" while non-wanderers were best represented by "giving/familiar" and "self-serving."
The results of the study suggested that wanderers differed from nonwanderers in certain personality characteristics. In general, wanderers had a personality make-up that resulted in more continually-active and social-seeking behavior along with a greater positive regard toward oneself and others.
The findings supported an earlier study by Monsour and Robb (1982) which found that wanderers tended to expend more physical energy in pursuit of social and leisure activities than non-wanderers. The personality facet of activity which was categorized as "multiple /continuous" reflected a high energy level that strongly related to Monsour and Robb's "physical energy" concept. The active "social-seeking" behavior found in the present study also supported the suggestion that wanderers employed more physical energy in pursuit of social activities.
However, unlike the Monsour and Robb study, the present study did not directly find a higher physical energy expenditure in leisure activities per se but rather a higher energy expenditure in overall activities of daily living - which could also include leisure activities. The "high-energy" behavior, or as referred to by some family members: "always busy" behavior, was a way of life that permeated many forms of activity involvement. The finding that wanderers exhibited a higher degree of energy in a multitude of tasks than non-wanderers, also paralleled the results of Snyder, Rupprecht, Pyrek, Brekhus, and Moss (1978). In their study, it was revealed that wanderers displayed a greater degree of "industriousness" - developed years ago - than non-wanderers. Industriousness was characterized by previous roles that may have permanently influenced behavior. Examples from the study included: "I have to go do my sewing now" and "I must attend to the children."
The study revealed that non-wanderers also showed an active style of behavior but primarily in scheduled tasks, that were balanced with more sedentary behavior. Scheduled tasks would encompass traditional activities that require planning and organization. Usually a designated block of time would be allocated during the day for this form of activity participation (i.e., two hours for a bowling game).
Sociable and Optimistic
The extroversion facet of warmth, gregariousness and positive-emotion were also found to significantly differentiate wanderers and nonwanderers. Warmth was characterized by friendliness, compassion, and an intimately involved style of personal interaction; while gregariousness was characterized by the desire to be with other people and enjoyment of the sheer quantity of social stimulation. Warmth and gregariousness together make up what is sometimes called "sociability" (McCrae & Costa, 1990). Hussain in Fopma-Loy (1988) found that when three wanderers were behaviorally "mapped" in an institutional setting, more stationary time was spent at points of interest, stimulation or potential reinforcement. Often, these points were around others or in an open room where others were located. From a continuity perspective, the need to establish social contacts is reflective of a personality trait.
Typological Frequencies of Significant Personality Facets
The agreeableness facet of altruism was also found to significantly differentiate wanderers from nonwanderers. The qualitative results revealed that wanderers tended to give to others, both familiar and unfamiliar people, more so than non-wanderers. "Giving" comprised emotional, financial, material goods and time. Non-wanderers, on the other hand, tended to give primarily to familiar people or were viewed as self-centered. These findings suggested that wanderers were not necessarily more giving than non-wanderers but rather gave to a wider variety of people (i.e., strangers, friends and family). The ability to seek out a broader scope of social contacts, in which to give, was consistent with the premise that wanderers are more sociable overall.
Age and Diagnosis
Two demographic variables were significant in differentiating wanderers from non-wanderers (i.e., age and diagnosis) (Table 3). The wanderers had a mean age of 77.85 compared to 83.7 for the non-wanderers. This finding contradicts previous research that has studied differences between wanderers and non-wanderers and concluded that age was a non-factor (Snyder, Rupprecht, Pyrek, Brekhus, & Moss, 1978; Terri, Larson, & Reifler, 1988). A potential reason for this discrepancy is the general lack of clarity used to define wandering. The majority of studies have relied on the subjective evaluation of staff of what they conceptualized as wandering in determining their sample. The present study used the criteria of "timem-motion" in an attempt to better quantify wandering (Thomas, 1995).
Nevertheless, the results of the present study raises questions in regard to the relevance of age. For example, are people who do not wander more likely to be older? If non-wanderers are older is age the reason they are more sedentary? Furthermore, are some non-wanderers actually people who used to wander but because of age no longer have the desire to wander? The current study attempted to control this possibility by assessing the wandering history of each subject. Those who were defined as non-wanderers-but who had a history of wandering - were eliminated from the study. This still does not account for, however, that some non-wanderers may have contracted dementia at a later point in life, and may have been a wanderer had s /he been a few years younger. The variable of age, especially as it relates to the specific selection criteria used in defining wanderers and non-wanderers is certainly an area that would benefit from additional research.
Another intriguing finding from the study was the significant difference (p^.05) on the variable of diagnosis between wanderers and nonwanderers. Wanderers, for the most part, were diagnosed with Alzheimer's disease while non-wanderers tended to have a vascular form of dementia. This difference suggests that the cognitive progression associated with Alzheimer's disease is more likely, compared to other forms of dementia, to produce wandering behavior. This finding was supported by Cooper (1993) who found, through observation of 1,312 patients, that those diagnosed with Alzheimer's disease were more likely to wander than those diagnosed with multi-infarct dementia. However, the inexact science of making accurate diagnostic estimations of dementia type severely complicates the ability to draw conclusions on the effects of diagnosis. Lyman (1989) noted that numerous studies support the fact that differential diagnosis is unequivocal only postmortem. Neuropathological analysis often reveals a misdiagnosis of Alzheimer's disease. Perhaps realizing the limitations of defining specific forms of dementia explains why the majority of research only used the global term of dementia when selecting a sample relating to wandering behavior.
The continuity of behaviors manifested over a lifetime, and which are characteristic of one's personality, provide important insight into better understanding as to why a person with dementia wanders. Oftentimes nurses, in attempting to understand wandering, attribute the behavior to the pathology of a dementing illness. Trying to understand wandering strictly from a biomedical perspective produces the effect of overlooking critical social factors. Ih discussing health care's preoccupation with the biomedical model, Lyman (1989, p. 600) raised the question of "how might we distinguish mental disorder from behavioral quirks and eccentricities developed over a lifetime that are found within the normal range of heterogeneity of aging? In addition, aberrant behaviors, continuous over a lifetime...that are normal for a given person, may be erroneously attributed to a dementing illness."
When viewed within the contextual world of the individual, wandering can be seen as an expression of one's personality - ingrained over a lifetime of development - radier than solely an expression of dementia. The results of the present study suggest that specific personality characteristics can differentiate wanderers from non-wanderers. Nursing personnel, in trying to understand wandering, need to understand the individual from both a current and historical viewpoint. Understanding lifelong behaviors which are rooted in one's personality provides a key to improving care. Viewing wandering as "normal" would encourage staff to respond, not in punitive and controlling ways, but rather with creativity and acceptance. Ultimately, renewed expectations of behavior by staff toward the wanderer could lead to interactions and programs that better accommodate his/her needs.
LIMITATIONS OF THE STUDY
The author recognizes that several limitations exist that need to be considered, such as: 1) the use of a small sample size (N=40), 2) the use of retrospective data based on the recall of significant others, 3) the subjective interpretation of the sample selection criteria by the social service and nursing staff, and 4) the use of multiple f-tests as a means of statistical analysis which capitalizes on chance and may lead to a Type I error. The author attempted to compensate for this potential effect on validity by incorporating a qualitative follow-up which showed support for the quantitative findings. Nonetheless, readers should be aware of this limitation.
RECOMMENDATIONS FOR FUTURE RESEARCH
1) Duplication of the present study. Suggested modification from the present study includes: increasing the population size, measuring across diverse settings, using multiple raters (surrogates), and using multiple personality scales; the results of which could be correlated.
2) Comparing demographic variables. The variables of age and diagnosis were found to differentiate die two groups. Additional research is needed to measure the validity of these findings.
3) Measuring staff attitudes. Results from the present study support the belief that wandering is a continuity of one's personality. The meaning of wandering to nursing personnel raises some interesting ethical questions. For example, is wandering viewed as abnormal behavior? If so, how does this attitude affect care? The staff's perception of quality of life as it pertains to the wanderer will often translate into styles of service provision that range from being extremely safety conscious to a strong promotion of autonomy. Future research could explore the relationship between attitudes and service delivery systems.
- Arendt, H. (1977). The life of the mind. New York, NY: Harcourt Brace Jovanovidi.
- Atchley, R.C. (1989). A continuity theory of normal aging. The Gerontologist, 29, 183-190.
- Butler, J.P., & Barnett, CA. (1991). Window of wandering. Geriatric Nursing, 12, 226-227.
- Chafetz, PJC (1991). Behavioral and cognitive outcomes of SCU care. Clinical Gerontologist, 11, 19-38.
- Cohen-Mansfield, J., Werner, P., Marx, M.S., & Freedman, L. (1991). Two studies of pacing in the nursing home. Journal of Gerontology, 46, M77-M83.
- Coons, D. (1988). Wandering. 77k American Journal of Alzheimer's Care and Related Disorders and Research, 3, 31-36.
- Cooper, J.K. (1993). Risk factor and behavioral differences between vascular and Alzheimer's dementias: The pathway to endstage disease. Journal of Geriatric Psychiatry & Neurology, 6, 29-33.
- Costa, PT., & McCrae, R.R. (1988). Personality in adulthood: A six-year longitudinal study of self reports and spouse ratings on the NEO Personality Inventory. Journal of Personality and Social Psychology, 55, 258-265.
- Costa, RT, & McCrae, R.R. (1992). Revised NEO personality inventory: Professional manual. Odessa, FL: Psychological Assessment Resources, Inc.
- Fopma-Loy, J. (1988). Wandering: Causes, consequences, and care. Journal of Psychosocial Nursing, 26, 8-18.
- Hall, G. (1990). Alterations in thought processes: Progressive degeneration of the cerebral cortex. In K.C. Buckwalter, & M.L. Maas (Eds)., Nursing diagnoses and interventions in the elderly. Menlo Park, CA: Addison- Wesley.
- Holmes, D., Splaine, M., Teresi, J., Ory, M., Barrett, V., Monaco, C, & Ramirez, M. (1994). What makes special care special: Concept mapping as a definitional tool. Alzheimer Disease and Associated Disorders, 8, S41-S53.
- Lawton, M.P. (1992, July). Aging and activity: A theoretical perspective. Paper presented at the International Conference on Leisure and Mental Health, University of Utah.
- Lyman, ICA. (1989). Bring the social back in: A critique of the biomedicalization of dementia. The Gerontologist, 29, 597-605.
- McCrae, R.R. & Costa, P.T. (1982). Consensual validation of personality traits: Evidence from self-reports and ratings. Journal of Personality and Social Psychobgy, 43, 293-303.
- McCrae, R.R., & Costa, P.T. (1986). Personality, coping, and coping effectiveness in an adult sample. Journal of Personality, 54, 385-405.
- McCrae, R.R., & Costa, P.T. (1989). Rotation to maximize the construct validity of factors in the NEO Personality Inventory. Multivariate Behavioral Research, 24, 107-124.
- McCrae, R.R., & Costa, P.T. (1990). Personality in adulthood. New York, NY: Guilford Press.
- Monsour, N., & Robb, S. (1982). Wandering behavior in old age: A psychosocial study. Social Work, 27, 411-415.
- Rabins, RV. (1991, September). Wandering. Paper presented at the Alzheimer's Association Meeting, Chicago, Illinois.
- Rader, ]., Doan, J., & Schwab, M. (1985). How to decrease wandering, A form of agenda behavior. Geriatric Nursing, 6, 196-199.
- Reisberg, B., Ferris, S., Leon, M., & Crook, T. (1982). The global deterioration scale for assessment of primary degenerative dementia. American Journal of Psychiatry, 139, 11361139.
- Reisberg, B., Ferris, S.H., & Schneck, M.K. (1981). The relationship between psychiatric assessments and cognitive test measures in mild to moderately cognitively impaired elderly. Psychopharmacal Bulletin, 17, 99-101.
- Sand, B. (1992). Alzheimer's disease: Special care units in long-term care facilities. Journal of Gerontological Nursing, 18, 28-34.
- Snyder, L., Rupprecht, P., Pyrek, J., Brekhus, S., & Moss, T. (1978). Wandering. The Gerontologist, 18, 272-280.
- Swanson, E. (1993). Catastrophic reactions and other behaviors of Alzheimer's residents: Special unit compared with traditional units. Archives of Psychiatric Nursing, 7, 292-299.
- Terri, L., Larson, E., & Reifler, B. (1988). Behavioral disturbance in dementia of the Alzheimer's type. Journal of the American Geriatrics Society, 36, 1-6.
- Thomas, D. (1995). Wandering: A proposed definition. Journal of Gerontological Nursing, 21(9), 35-41.
- Thomas, D. (1996). A case study on the effects of a retrofitted dementia special care unit on resident behaviors. American Journal of Alzheimer's Disease, 11, 8-14.
Volicer, L., Collard, A., Hurley, A., & Bishop, C. (1994). Impact of special care unit for patients with advanced Alzheimer's disease on patient's discomfort and costs. Journal of the American Geriatrics Society, 42, 597-603.
- Zeisel, J., Hyde, J., & Levkoff, S. (1994). Best practices: An environment-behavior (EB) model for Alzheimer special care units. The American Journal of Alzheimer's Care and Related Disorders and Research, 9, 4-21.
Criterio ior Wanderers and Non-wanderers
Differences In Personality Facets Between Wanderers and Non-wanderers
Test Results at Demographic Variables
Typological Frequencies of Significant Personality Facets