Home > RNJ > 2006 > January/February > Caregiver Perspectives of Memory and Behavior Changes in Stroke Survivors

Caregiver Perspectives of Memory and Behavior Changes in Stroke Survivors
Patricia C. Clark, PhD RN FAHA FAAN Sandra B. Dunbar, RN DSN FAAN Dawn M. Aycock, MSN RN Elizabeth Courtney, MA OT/L Steven L. Wolf, PhD PT FAPTA

Post-stroke memory and behavior changes (MBC) are associated with negative outcomes for stroke survivors and caregivers. This article describes the types of MBC that occur most frequently and caregivers’ responses to these behaviors. Data were obtained through in-person interviews and administration of questionnaires to 132 caregivers of first-time stroke survivors 3-9 months after stroke. MBC were measured with a modified version of a Memory and Behavior Problems checklist. On average, caregivers reported 7.7 ± 3.6 (range 0–17) behaviors. Common stroke survivor MBC included appearing sad or depressed, interrupting the caregiver, and being restless or agitated. These MBC were distressing to caregivers. Caregivers may not recognize some MBC as potential symptoms of depression. In addition, caregiver misunderstanding of the amount of control survivors may have over some behaviors has implications for rehabilitation and caregivers’ responses to these changes.

The residual effects of a stroke are often manifested through changes in the stroke survivor’s memory and behavior as well as in physical function. Often, however, the focus is on physical recovery from stroke with less emphasis on cognitive and behavioral deficits and their influence on functional recovery. Mayo (1998) referred to cognition, depression, and anxiety as the “neglected” outcomes of stroke. Cognitive deficits after stroke have been reported at approximately 35% (Tatemichi, 1994).

Memory and behavior changes (MBC) of stroke survivors affect family caregivers as well as survivors (Bugge, Alexander, & Hagen, 1999; Clark et al., 2004; Clyburn, Stones, Hadjistavropoulos, & Tuokko, 2000; Schulz, Tompkins, & Rau, 1988). In one small study (n = 22), the MBC of survivors reported by more than 50% of caregivers and rated as most bothersome were uncooperativeness, sad or depressed appearance, and anger (Clark & King, 2003). Grant and colleagues found that caregivers expressed aggravation, sadness, frustration, and disappointment when trying to manage the third most commonly reported problem of cognitive, behavioral, and emotional changes exhibited by stroke survivors during the first month after discharge (Grant, Glandon, Elliott, Giger, & Weaver, 2004). Similarly, a larger, longitudinal study (Schulz, Tompkins, & Ran, 1988) found that an increase in negative personality characteristics of the stroke survivor was the only predictor of caregiver depression 7–9 months after stroke after controlling for caregiver health, income, and age.

Stroke survivor MBC and personality changes may be consistent with depressive symptoms. After stroke depression is a common phenomenon. Yet, in one study only 24% (n = 21) of survivors were receiving an antidepressant medication (Clark & King, 2003). These findings are consistent with other studies reporting only 17.4% (n = 143; Kotila, Numminen, Waltimo, & Kaste, 1998) and 31% (n = 53; King, Shade-Zeldow, Carlson, Feldman, & Philip, 2002) of depressed survivors undergoing pharmacological treatment. After stroke depression is consistently associated with negative consequences, including worse recovery for survivors (Hosking, 1996). Some reports show significant effects of stroke survivor depression on the caregiver (Draper, Poulos, Cole, Poulos, & Ehrlich, 1992; Greveson, Gray, French, & James, 1991), but others have found that the survivor’s physical disability has a stronger effect on caregivers (Kotila, Numminen, Waltino, & Kasle, 1998; Thompson, Bundiek, & Sobolew-Whubin, 1990).

Family caregivers have to respond to these MBC with little knowledge of how they may result from the stroke or how they may affect functional recovery. Family caregivers are involved in the rehabilitation of survivors to help them regain functional independence and reintegrate into family roles. This article describes the types of MBC that occur most frequently and are of most concern to caregivers, and caregivers’ responses to these behaviors.



The stroke family caregiver study was conducted in collaboration with the Extremity Constraint Induced Therapy Evaluation (EXCITE) national clinical trial, which is testing constraint-induced therapy for regaining upper extremity function following stroke (Winstein et al., 2003). Institutional review boards at participating sites approved both studies and informed consent was obtained from all participants. The research settings included six rehabilitation centers across the United States that were locations for EXCITE. After a stroke survivor was enrolled in the EXCITE trial, caregivers were asked about their willingness to participate in the study. Trained research assistants collected data through administration of questionnaires and semistructured interviews performed in person or via telephone. In a nonrandom sample, factors associated with caregiver outcomes were examined by using a correlational design. Only baseline data, not affected by the EXCITE intervention, are reported here.


Eligible family caregivers were those (a) caring for a stroke survivor enrolled in the EXCITE clinical trial, (b) 18 years or older, (c) able to read and speak English, (d) not currently under treatment (medication or counseling) for depression to exclude people with chronic ongoing depression, and (e) self-identified as providing the most assistance to the stroke survivor during recovery.

Stroke survivors met extensive eligibility criteria for the EXCITE trial (Winstein et al., 2003) that included (a) 3–9 months after stroke onset from a first time clinical cerebrovascular accident (ischemic or hemorrhagic), (b) specific active range of motion criteria for wrist movement, and (c) a Mini-Mental State Exam (MMSE) score of >24.

Of the 183 stroke survivors enrolled at EXCITE trial sites participating in this study, 16 reported not having a caregiver, and 13 caregivers did not meet eligibility criteria. Of the 154 eligible caregivers, 19 declined participation for reasons such as “being too busy” or “preferring to focus on the survivor’s recovery.” Three caregivers were inaccessible at the time of study enrollment. A total of 132 caregivers were enrolled. They ranged in age from 25 to 85 years with a mean age of 56.7 + 13.7 years, were primarily female, White, non-Hispanic, college-educated, employed, and were spouses or significant others of the survivor (Table 1).

Stroke survivors were 3–9 months after stroke, had a mean age of 62.2 + 12.6 years (range 30 to 83), and were mostly male (64%), primarily White, non- Hispanic (74%), and well educated (63%; i.e., > high school education or higher). The majority suffered an ischemic stroke (88%) and the remaining a hemorrhagic stroke. Fifty-seven percent had a right hemisphere involvement. Using the upper limb motor impairment portion of the Fugl-Meyer Assesment (FMA; Fugl-Meyer, Jaasko, Leyman, Olsson, & Steglind, 1975), with a max score of 66, 72% were identified as high functioning (score ≥ 34). At the time of stroke survivor enrollment in the EXCITE trial, 37% reported taking an antidepressant.


Stroke survivors’ MBC were measured with an adapted version of the Memory and Behavior Problems Checklist (MBPC; Zarit & Zarit, 1990). The MBPC is a 25-item scale that measures two separate dimensions: the frequency of memory and behavior problems and how bothersome or upsetting the behavior is to the caregiver. The MBPC has been used extensively in dementia research and has adequate construct validity and reliability (Zarit & Zarit). Reliabilities reported have ranged from 0.66 to 0.84. (Clark & King, 2003; Zarit & Zarit). Based on the first author’s preliminary data, items reported as not occurring at all in survivors such as wandering were replaced with items such as irritability from the stroke literature (Anderson, Linto, & Stewart-Wynne, 1995; Schulz et al., 1988). A dichotomous category was added to determine whether the caregiver believed the survivor could control the behavior. The revised 19-item scale includes a wide array of memory and behavior items specific to stroke recovery. The internal consistency coefficient for the caregiver study was 0.73. The frequency of MBC and bothersome reaction were rated by using (0–4) Likert-type scales. Not having occurred to occurring daily were used for frequency, and not at all to extremely were used to describe how bothersome behaviors were to caregivers. Caregivers also responded (“yes” or “no”) if they thought the survivor could control the behaviors that occurred.

Caregivers who responded that an MBC had occurred since the stroke were asked to identify at least two strategies they implemented to respond to or manage the change. These strategies were content analyzed and coded into one of 10 categories (Table 2). The intrarrater reliability coefficient for the coding was 0.90. This was calculated following repeat coding of 20% of responses at a later date. A Likert-type scale of not at all to extremely (0–4) was used for rating how effective caregivers perceived the strategies they used.

Stroke survivor physical function was identified using the upper limb motor impairment portion of the FMA (Fugl-Meyer et al., 1975). The FMA measure takes into account evolving synergy patterns and isolated strength, coordination, and hypertonia. Each item is graded on a 3-point ordinal scale (0 = cannot perform, 1 = can perform partially, 2 = can perform fully) and summed to provide a maximum score of 66 (Chae, Labatia, & Yang, 2003). As part of the larger EXCITE trial, scores were dichotomized to classify stroke survivors as low functioning if they scored 33 or less and high functioning if they scored from 34 to 66.

Statistical Analysis

Descriptive statistics were computed to characterize the sample and describe major study variables, and t tests were used to examine differences in MBC based on survivors’ clinical characteristics. Two caregivers could not be contacted to complete the MBC checklist; therefore analysis was performed on 130 participants.


The average number of MBC reported by caregivers was 7.6 ± 3.6 (range 0–17) out of a possible 19 behaviors, and the average frequency of MBC was 18.1 ± 10.4 (range 0–51). The types of MBC occurring most frequently (≥ 50% reported by caregivers) were the survivor (a) appearing sad or depressed, (b) interrupting caregiver when busy, (c) being restless or agitated, (d) crying or becoming tearful, (e) becoming angry, and (f) being irritable (Table 3). The next set of MBC reported by 30%–49% of caregivers were (a) spends long periods of time inactive, (b) asks the same question over and over, (c) wakes caregiver up at night, (d) loses, misplaces, or hides things, (e) forgets what day it is, and (f) is more self-centered. The MBC least reported (<30%) were (a) unable to keep occupied or busy by self, (b) uncooperative, (c) less thoughtful, (d) incontinent of bowel or bladder, (e) problems eating, (f) talks little or not at all, and (g) talks in a threatening manner. The average bothersome rating for all behaviors was 1.4 ± 0.72 (between a little to moderately bothersome). The range was 0–3.60. MBC that were most bothersome (moderately to extremely) to caregivers were (a) appears sad or depressed; (b) cries or becomes tearful, and (c) becomes angry (Table 3). Common MBC identified by most caregivers as ones survivors were unable to control included (a) cries or becomes tearful, (b) appears sad or depressed, and (c) wakes caregiver up at night; however, MBC rated by caregivers as changes survivors could control included (a) interrupts caregiver when busy, (b) anger, and (c) irritability (Table 3).

Caregiver responses to common MBC are listed in Table 3. Use of emotional or social support or ignoring the survivor were the most common responses to MBC, especially those behaviors similar to depressive symptoms. Actively or passively facilitating the survivor was used with memory- and activity-related changes.

Stroke survivors’ age (r = 0.08, p = .34) and education (r = –0.15, p = .09, n = 123) were not related to frequency of MBC. To determine whether there were differences in the MBC of survivors based on gender and clinical characteristics of survivors, t tests were conducted. There were no differences in the frequency of MBC by gender, presence of at least one comorbidity (e.g., heart disease or diabetes), functional level, hemisphere involved, or treatment with an antidepressant (Table 4).


Relatively high physically functioning survivors had numerous MBC as reported by caregivers. These MBC are distressing to caregivers. MBC have been associated with poorer caregiver mental health and may be exacerbated in the presence of family conflict surrounding stroke recovery (Clark et al., 2004).

Many of the most frequently reported MBC are consistent with depressive symptoms found on commonly used depression screening instruments, such as the Center for Epidemiologic Studies Depression Scale (Radloff, 1997), and identified by the National Institute of Mental Health (NIMH, 2002); others may relate to poor impulse control. If caregivers do not recognize these as depressive symptoms or as possibly resulting from stroke, they may not recognize the importance of discussing them with a healthcare provider. Caregivers’ lack of recognition of MBC as possible depression and misunderstanding the amount of control survivors may have over some behaviors have implications for rehabilitation and for understanding caregivers’ responses to these changes. For example, caregivers may require more education and counseling regarding recognition of depressive symptoms and about what is significant to report to healthcare providers. Addressing these MBC may have an effect on the survivors’ abilities to participate effectively in their rehabilitation programs. In addition, some clinicians noted that many of the survivors were participating in the EXCITE trial at the urging of the caregiver. The caregiver’s emotional investment of wanting the survivor to participate in a therapy that may improve upper extremity function, although unintended, may contribute to the survivor feeling inadequate and contribute to depressive feelings.

The fact that many of the common MBC were consistent with depressive symptoms underscores the importance of careful assessment for after stroke depression. Okan and colleagues (Okan et al., 2004) surveyed stroke survivors about barriers to therapy for depression after stroke. The majority (80%) of survivors (n = 84) reported experiencing depression or depressive symptoms. Of those experiencing one or more symptoms of depression, approximately one-third reported receiving treatment. For the majority, physicians did not inquire about depression, and survivors did not mention depression to their doctors. The most common reasons for not seeking treatment included lack of recognition of symptoms and the belief that the depression was mild or would resolve (Okan et al.).

Common MBC identified by most caregivers as changes survivors could control, such as “anger” and “interrupts caregiver when busy,” might be associated with emotional lability from the stroke. Caregivers’ expectations about survivors’ ability to control behaviors such as activity, anger, and irritability may reflect a lack of understanding regarding the potential effects a stroke may have on cognitive and behavioral function.

The lack of difference in the frequency of MBC based on clinical characteristics such as hemisphere involved and level of upper extremity motor function was unexpected. However, this finding may be attributed to the relatively higher level of physical function possessed by survivors as a result of the strict inclusion criteria for EXCITE. The specific inclusion criteria for stroke survivors to participate in the constraint-induced therapy clinical trial required them to demonstrate at least 10º of extension at the wrist and digits, 20º of extension at the elbow, and 45º of flexion at the shoulder. Therefore, stroke survivors’ range of physical ability was limited, and in a sample with a more representative range of function, results may have been stronger. An alternative explanation is that MBC are not strongly associated with physical function when less physical disability of stroke prevails. The lack of difference in frequency of MBC of survivors on antidepressant medication compared with those who were not was also unanticipated because there is evidence that treatment of after stroke depression has resulted in improvement of depression and better function (Robinson, 2003; Whyte & Mulsant, 2002). The interpretation of the findings may be limited by the lack of data about length of time that survivors had been on antidepressant medication. However, an alternative explanation is that survivors treated for depressive symptoms may be undertreated and therefore not gaining optimal benefit.

A common strategy used by caregivers to respond to MBC was emotional support. There are few reports that categorize caregiver strategies and responses. Bakas, Austin, Okonkwo, Lewis, and Chadwick (2002) did identify caregiver strategies used in response to survivors’ emotions and behaviors. In their study, the responses of caregivers directed toward their own relief in the situation and those directed toward survivors were combined. Although the categorization differs, in reviewing examples they provided of caregiver responses, caregivers reported using similar strategies, such as general emotional and social support, seeking more information, reframing the event, and managing their own responses. This work adds to the findings of Bakas and colleagues (Bakas et al.) by identifying the most frequently used caregiver responses for specific MBC. Negative emotions such as survivors’ irritability or anger have been noted as common behaviors in several studies (Anderson et al., 1995; Hartke & King, 2002; Williams & Dahl, 2002). In this study, caregivers tended to view anger and irritability as behaviors the survivor could control; thus the caregivers frequently used the strategy of ignoring the behavior. Although this study did not address survivors’ responses to caregivers’ actions, it is likely that being ignored may have perpetuated survivors’ frustration and negative behavior. Interventions to help expand the caregiver’s repertoire of responses to MBC that are frequent and bothersome may be effective in reducing negative effects on caregivers.


A limitation of the study is the use of convenience sampling. The sample mainly comprised White, well-educated, female caregivers of relatively high functioning, cognitively intact (MMSE scores >24) stroke survivors. Future studies may need to recruit a more heterogeneous group. However, stroke survivors with more severe physical functioning or poorer cognition may have exhibited more MBC or more frequent MBC. Another limitation is that MBC are only from the perspective of caregivers, and reports from stroke survivors or an independent observer might differ. Nevertheless, these are the MBC with which the caregivers are trying to cope.

Clinical Implications

A valuable implication of this study for nurses is the need for heightened awareness and recognition of emotional and mood changes following stroke and their potential effect on recovery and family members. Rehabilitation nurses may want to use standard screening instruments to help identify stroke survivors who may need further evaluation for depression and involve family members in the screening process through collaborative communication and education. The term depression may have negative connotations, and nurses can help caregivers talk with stroke survivors about feeling sad or the loss of interest in activities. Nurses can educate family members about MBC resulting from stroke and offer ways to identify and seek assistance for these changes. They can provide caregivers with strategies to respond to MBC such as anger that may result in supportive interventions for stroke survivors and possible coping measures for caregivers. Helping caregivers understand that not all MBC may be easy for stroke survivors to control may help the caregiver empathize with the stroke survivors’ feelings. Further studies should be done to assess the outcome of caregiver responses to MBC.


The importance of recognizing and preparing caregivers for not only the physical changes but also MBC after stroke cannot be overstated. The general caregiving literature suggests that greater understanding of specific MBC of survivors has led to the development of targeted interventions, and knowing the common types of MBC in stroke may lead to interventions that will improve outcomes of survivors and caregivers.


This study was supported in part by the following grants: “Family Function, Stroke Recovery, and Caregiver Outcomes” (National Institute of Nursing Research, RO1 NR07612-01, to Patricia C. Clark) and Extremity Constraint-Induced Training Evaluation (EXCITE)” (National Institute of Child Health and Human Development (NICHHD) and the National Institute of Neurological Disorders and Stroke (NINDS), RO1 HD 37606, to Steven L. Wolf). Participating EXCITE sites include Emory University, University of Southern California, University of North Carolina at Chapel Hill, Wake Forest University, Ohio State University, University of Florida-Gainesville, and Washington University School of Medicine, St. Louis, MO.

About the Authors

Patricia C. Clark, PhD RN FAHA FAAN, is an associate professor at the school of nursing and adjunct assistant professor at the school of medicine, Emory University, Atlanta, GA.

Sandra B. Dunbar, RN DSN FAAN, is the Charles Howard Candler Professor of Cardiovascular Nursing at Emory University, Atlanta, GA.

Dawn M. Aycock, MSN RN, is a senior research nurse at the school of nursing, Emory University, Atlanta, GA.

Elizabeth Courtney, MA OT/L, is an occupational therapist at the Center for Rehabilitation Medicine, Emory University.

Steven L. Wolf, PhD PT FAPTA, is a professor in the department of rehabilitation medicine, Emory University.

Address correspondence to Patricia C. Clark at 1520 Clifton Road, Room 306, Emory University, Atlanta, GA 30322, or to pclark@gsu.edu.


Anderson, C. S., Linto, J., & Stewart-Wynne, E. G. (1995). A population-based assessment of the impact and burden of caregiving for long-term stroke survivors. Stroke, 26, 843–849.

Bakas, T., Austin, J. K., Okonkwo, K. F., Lewis, R. R., & Chadwick, L. (2002). Needs, concerns, strategies, and advice of stroke caregivers the first 6 months after discharge. Journal of Neuroscience Nursing, 34, 242–251.

Bugge, C., Alexander, H., & Hagen, S. (1999). Stroke patients’ informal caregivers: Patient, caregiver, and service factors that affect caregiver strain. Stroke, 30, 1517–1523.

Chae, J., Labatia, I., & Yang, G. (2003). Upper limb motor function in hemiparesis: Concurrent validity of the Arm Motor Ability test. American Journal of Physical Medicine & Rehabilitation, 82(1), 1–8.

Clark, P. C., Dunbar, S. B., Shields, C. G., Viswanathan, B., Aycock, D. M., & Wolf, S. L. (2004). Influence of stroke survivor characteristics and family conflict surrounding recovery on caregivers’ mental and physical health. Nursing Research, 53, 406–413.

Clark, P. C., & King, K. B. (2003). Comparison of family caregivers: Stroke survivors vs. persons with Alzheimer’s disease. Journal of Gerontological Nursing, 29(2), 45–53.

Clyburn, L. D., Stones, M. J., Hadjistavropoulos, T., & Tuokko, H. (2000). Predicting caregiver burden and depression in Alzheimer’s disease. Journal of Gerontology, 55B, S2–S13.

Draper, B. M., Poulos, C. J., Cole, A. M., Poulos, R. G., & Ehrlich, F. (1992). A comparison of caregivers for elderly stroke and dementia victims. Journal of the American Geriatrics Society, 40, 896–901.

Fugl-Meyer, A. R., Jaasko, L., Leyman, I., Olsson, S., & Steglind, S. (1975). The post-stroke hemiplegic patient: A method for evaluation of physical performance. Scandinavian Journal of Rehabilitation Medicine, 7(1), 13–31.

Grant, J. S., Glandon, G. L., Elliott, T. R., Giger, J. N., & Weaver, M. (2004). Caregiving problems and feeling experienced by family caregivers of stroke survivors the first month after discharge. International Journal of Rehabilitation Research, 27, 105–111.

Greveson, G. C., Gray, C. S., French, J. M., & James, O. F. W. (1991). Long-term outcome for patients and carers following hospital admission for stroke. Age and Ageing, 20, 337–344.

Hartke, R. J., & King, R. B. (2002). Analysis of problem types and difficulty among older stroke caregivers. Topics in Stroke Rehabilitation, 9(1), 16–33.

Hosking, S. G., Marsh, N. V. & Friedman, P. J. (1996). Poststroke depression: Prevalence, course, and associated factors. Neuropsychology, 6, 107–133.

King, R. B., Shade-Zeldow, Y., Carlson, C. E., Feldman, J. L., & Philip, M. (2002). Adaptation to stroke: A longitudinal study of depressive symptoms, physical health, and coping process. Topics in Stroke Rehabilitation, 9(1), 46–66.

Kotila, M., Numminen, H., Waltimo, O., & Kaste, M. (1998). Depression after stroke: Results of the FINNSTROKE study. Stroke, 29, 368–372.

Mayo, N. E. (1998). Epidemiology and recovery of stroke. Physical Medicine and Rehabilitation: State of the Art Reviews, 12, 355–366.

National Institute of Mental Health. (2002). Depression and stroke. Retrieved May 2004 from www.nimh.nih.gov/publicat/depstroke.cfm.

Okan, N., Billings, M., Fitzpatrick, K., Betts-Lion, C., Cole, A., Allen, M., et al. (2004). Barriers to therapy in post-stroke depression. Stroke, 35, 288.

Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401.

Robinson, R. G. (2003). Poststroke depression: Prevalence, diagnosis, treatment, and disease progression. [Review] [103 refs]. Biological Psychiatry, 54, 376–387.

Schulz, R., Tompkins, C. A., & Rau, M. T. (1988). A longitudinal study of the psychosocial impact of stroke on primary support persons. Psychology & Aging, 3, 131–141.

Tatemichi, T. K., Desmond, D. W., Stern, Y., Paik, M., Sano, M. & Bagiella, E. (1994). Cognitive impairment after stroke: Pattern and relationship to functional abilities. Journal of Neurology, Neurosurgery and Psychiatry, 57, 202–207.

Thompson, S. C., Bundiek, N. I., & Sobolew-Whubin, A. (1990). The caregivers of stroke patients: An investigation of factors associated with depression. Journal of Applied Social Psychology, 20, 115–129.

Whyte, E., & Mulsant, B. (2002). Post stroke depression: Epidemiology, pathophysiology, and biological treatment. Society of Biological Psychiatry, 52, 253–264.

Williams, A. M., & Dahl, C. W. (2002). Patient and caregiver perceptions of stroke survivor behavior: A comparison. Rehabilitation Nursing, 27(1), 19–24.

Winstein, C., Miller, J., Blanton, S., Morris, D., Uswatte, G., Taub, E., et al. (2003). Methods for a multi-site randomized trial to investigate the effect of constraint-induced movement therapy in improving upper extremity function among adults recovering from cerebrovascular stroke. Neurorehabilitation and Neural Repair, 17, 137–152.

Zarit, S. H., & Zarit, J. M. (1990). The memory and behavior problems checklist and the burden interview. University Park, PA: Pennsylvania State University, Gerontology Center.