Regular Research Articles
Are Wandering and Physically Nonaggressive Agitation Equivalent?

https://doi.org/10.1097/JGP.0b013e3181629943Get rights and content

Objective

The authors examined equivalence of wandering and physically nonaggressive agitation (PNA) as concepts.

Design

A cross-sectional correlational design was used.

Setting

Participants were recruited from 22 nursing homes and 6 assisted living facilities in two states.

Participants

Ambulatory residents meeting DSM-IV criteria for dementia (N = 181) were studied.

Measurements

Video-tapes for up to twelve 20-minute observations per participant were coded for wandering using an empirically derived taxonomy of ambulation patterns. Separate raters coded the same tapes for six PNA behaviors on the agitation behavior mapping instrument.

Results

Most participants (73.5%) wandered; all showed PNA behaviors. Factor analyses yielded an one-factor solution for wandering (explained variance = 43.66%) and a two-factor solution for PNA (explained variance = 53.45%). Overall wandering correlated significantly with PNA Factor 1 (df =179, r = 0.68, p <0.001) and Factor 2, but at a lower value (df = 179, r = 0.26, p <0.01).

Conclusion

Findings depict wandering and PNA as overlapping, but nonequivalent phenomena. Evidence supporting construct validity of wandering was more robust than that for PNA. Results have implications for accuracy in scientific and clinical detection and labeling of wandering and agitation.

Section snippets

METHODS

In a cross-sectional correlational study, we videotaped participants for up to 12 observations (mean: 9.0; SD: 2.0, range: 1–12) designed to occur once per hour between 8:00 A.M. and 8:00 P.M. The study was approved by appropriate institutional review boards; each site received a federal single project assurance.

RESULTS

Nearly three-fourths (73.5%) of participants (N = 181) wandered, while 100% displayed PNA. Those who wandered showed random wandering most often (71.3%) and pacing least often (8.8%). For PNA, participants displayed general restlessness most often (98.3%) and inappropriate dressing or disrobing least often (8.8%).

Descriptive statistics for overall wandering (2.91 episodes/hour) and PNA (3.52 behaviors/hour) are shown in Table 1. Overall, rates were similar, differing on average by less than 1

Structure

Factor analyses indicate that wandering, measured by pattern, is a coherent construct, though it may have multiple dimensions when measured in other ways.8, 19, 20 In contrast, PNA has two factors. Factor 1 contains three behaviors (pacing/aimless wandering, trying to get to a different place, handling things inappropriately), the first two corresponding with a possible wandering subscale. Basis for a link between wandering and handling things inappropriately is not clear, although

CONCLUSIONS

According to Trochim's13 approach to construct validity, i.e., examining patterns of relationships among aspects of a given phenomenon, the factor analysis and pattern of correlations among wandering behaviors provide beginning evidence for construct validity representing a specific clinical phenomenon. Wandering patterns, individually and overall, operated in a manner consistent with expectations and with existing evidence, where available.

A similar level of evidence for PNA is not supported.

References (23)

  • DL Algase

    Wandering in dementia

    Annu Rev Nurs Res

    (1999)
  • J Cohen-Mansfield

    Agitation in the elderly: Definitional and theoretical conceptualizations

  • A Kolanowski et al.

    Contextual factors associated with disturbing behaviors in institutionalized elders

    Nurs Res

    (1994)
  • KY Lai et al.

    Wandering behavior in people with dementia

    J Adv Nurs

    (2003)
  • JC Peatfield et al.

    Wandering: an integrative review

    J Gerontol Nurs

    (2002)
  • JD Price et al.

    Subjective barriers to prevent wandering of cognitively-impaired people

    Cochrane Database Syst Rev

    (2001)
  • C Siders et al.

    Evidence for implementing nonpharmacological interventions for wandering

    Rehabil Nurs

    (2004)
  • DL Algase et al.

    The Algase Wandering Scale: initial psychometrics of a new caregiver reporting tool

    Am J Alzheimer's Dis Other Dementias

    (2001)
  • DP Devanand et al.

    Behavioral syndromes in Alzheimer's disease

    Int Psychogeriatr

    (1992)
  • J Cohen-Mansfield

    Instruction Manual for the Cohen-Mansfield Agitation Inventory (CMAI)

    (1991)
  • J Cohen-Mansfield

    Guidelines and Suggestions for Administering the Agitation Behavior Mapping Instrument (ABMI)

    (1986)
  • This study was supported by a grant (RO1 NR04569) from the National Institutes of Nursing Research and Aging (to DA).

    Beel-Bates is currently at Grand Valley State University, Grand Rapids, MI.

    View full text