HomeLarge Type Edition
HOME ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 62:673-678 (2007)
© 2007 The Gerontological Society of America

Predictors of Adherence to a Skill-Building Intervention in Dementia Caregivers

Yeon Kyung Chee, Laura N. Gitlin, Marie P. Dennis and Walter W. Hauck

1 Center for Applied Research on Aging and Health and 2 Division of Biostatistics, Department of Pharmacology and Experimental Therapeutics, Thomas Jefferson University, Philadelphia, Pennsylvania.

Address correspondence to Yeon Kyung Chee, PhD, or Laura N. Gitlin, PhD, Center for Applied Research on Aging and Health, Thomas Jefferson University, 130 S. 9th Street, Suite 500, Philadelphia, PA 19107. E-mail: yeon.chee{at}jefferson.edu or laura.gitlin{at}jefferson.edu


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 References
 
Background. Treatment adherence is a widely recognized problem in health services but understudied in caregiver intervention research. This study examines caregiver sociodemographic and psychological characteristics, patient illness severity, and treatment implementation factors as predictors of caregiver adherence to a skills training intervention to help families manage dementia care problems at home.

Methods. The sample consisted of 105 caregivers randomized to the Home Environmental Skill-Building Program at the Philadelphia site of the National Institutes of Health (NIH) Resources for Enhancing Alzheimer's Caregiver Health (REACH I). The intervention, implemented by occupational therapists, consisted of education, problem solving, communication, environmental and task simplification techniques, and home modifications. Adherence was measured by a proportion score representing the percentage of strategies used by participants compared to the total number of strategies prescribed during intervention (Strategy Use).

Results. Regression analysis with intraclass correlation adjustment for interventionist effects revealed that caregivers with better physical health (p <.001), greater treatment exposure (p <.001), more problem areas addressed (p =.012), and for whom more active therapeutic techniques (role play) were used (p =.004) demonstrated greater adherence. Other caregiver characteristics, patient cognitive impairment, and troublesome behaviors were not significantly related to caregiver adherence.

Conclusions. Modifiable caregiver and treatment implementation factors, including active engagement of caregivers, were associated with adherence, whereas patient characteristics were not. Caregivers with poor health may be at risk for not benefiting from intervention, suggesting that efforts, including instruction in preventive care and allocating time to attend to their own health care needs, be directed towards improving their health.

THE past 15 years of caregiver intervention research has resulted in innovative interventions (1–3) showing positive outcomes for families (4,5). With few exceptions, however (6,7), intervention studies have not examined the mechanisms by which benefits are achieved and the factors that predict caregiver adherence (8). Given evidence that adherence enhances medical and health services outcomes (9,10), evaluating predictors of adherence is an important next step in caregiver intervention research that provides a systematic approach to refining interventions to optimize benefits.

This study identified predictors of caregiver adherence to strategies provided in a skill-building intervention. To examine adherence, we used Lichstein and colleagues' treatment implementation (TI) framework (11), which refers to whether a treatment is implemented as intended. TI involves three components: (i) treatment delivery or methodologies for providing an intervention (didactic, hands-on training); (ii) treatment receipt, the extent to which the intervention is received (dose, intensity) and participants acquire knowledge; and (iii) treatment enactment, participant performance, and integration of intervention strategies. We operationalized adherence as the extent to which prescribed strategies were enacted by caregivers (Strategy Use) following the principle that enactment represents a form of adherence in behavioral interventions (12). Although TI is used to monitor psychotherapeutic and behavioral interventions, it has not been evaluated with regard to caregiver adherence (13). We tested the hypothesis that treatment delivery and receipt enhance treatment enactment, or adherence.

We also draw upon conceptual models to identify possible predictors of adherence. Based on the Health Belief Model that takes into account perceived disease risks on adherence (14,15), we consider patient illness severity as a potential predictor. We also use Social Cognitive Theory, which emphasizes individual, interpersonal, and environmental factors as potential predictors (16), and the Transtheoretical Model of Change (17), which posits that enactment of new behaviors is incremental reflecting a person's readiness.


   METHODS
Top
 Abstract
 Methods
Results
 Discussion
 References
 
Participants and Intervention
Data for this study were from the NIH REACH I multisite initiative at the Philadelphia site at which the Environmental Skill-Building Program (ESP) was tested (4,18). Study participants lived with and provided care for a minimum of 4 hours daily for at least 6 months to persons who had a doctor's diagnosis of dementia or a score < 24 on the Mini-Mental State Examination (MMSE) (19), and one or more activity limitations. Of the 255 caregivers enrolled, 129 were randomized to intervention, of whom 105 had treatment documentation for at least 4 of 6 intervention sessions, the minimum dosage believed necessary for treatment benefit (7,9). Of the 24 intervention caregivers not included in the adherence analyses, 8 did not receive ESP (1 refused, 1 placed the patient in a nursing home, 2 were bereaved, and 4 were lost to follow-up prior to initiating ESP), 11 received fewer than 4 sessions, and 5 had incomplete treatment documentation. A comparison of the 105 caregivers with complete documentation and 16 of the 24 caregivers who received intervention but who were not included in these analyses revealed no statistically significant or large differences in caregiver and patient characteristics.

Details of study procedures, ESP, and treatment outcomes are reported elsewhere (4,20). Briefly, trained interviewers obtained a signed Institutional Review Board–approved consent from eligible caregivers and conducted a 21/2 hour baseline home interview. Following baseline, caregivers were stratified by race and relationship to patient and randomized to intervention or usual care. Participants were interviewed again at 6, 12, and 18 months by interviewers masked to study assignment. ESP consisted of a 6-month active phase involving five 90-minute home visits and one 20-minute telephone contact, and a 6-month maintenance phase involving one 90-minute home visit and three 20-minute telephone sessions by occupational therapists who provided caregivers with problem-solving skills, environmental modifications, and stress reduction training. Treatment implementation data were obtained by interventionists immediately following completion of each intervention session. Interventionists were registered, licensed occupational therapists with ≥3 years of clinical experience and completion of 30 hours of training specific to ESP.

Adherence Measure
To assess adherence, at the conclusion of the active phase (6 months postbaseline), interventionists asked caregivers whether each prescribed strategy was in use and, if so, to demonstrate its use. A second assessment of strategy use was conducted by interventionists at the final session of maintenance (e.g., 10th visit) to not penalize caregivers who had been provided strategies at the end of the active phase. A Strategy Use measure was calculated as a proportion score representing the percentage of strategies in use (end of active or maintenance) relative to the total recommended during the intervention.

Predictor Variables
Variables representing five domains were considered potential predictors: caregiver sociodemographics; caregiver psychological factors; patient illness severity; treatment delivery; and receipt. Caregiver and patient characteristics and psychological factors were assessed at baseline by interviewers. Caregiver readiness was assessed by interventionists at the conclusion of the first treatment session. Treatment delivery and receipt were assessed by interventionists following completion of each treatment session.

Caregiver characteristics.-- Caregiver demographics included age, gender, race (white, African American, other), education (<high school, ≥high school), and relationship (spouse, child, other). Caregiver perceived health was measured by a four-item index that consisted of a self-rated health question and three statements using a 5-point response set (In general, would you say your health is ?, 1 = poor to 5 = excellent; I seem to get sick a little easier than other people, I am as healthy as anybody I know, I expect my health to get worse, 1 = definitely false to 5 = definitely true) (Archbold PG, Stewart BJ, Harvath TA, Lucas SA, unpublished manuscript, 1986). Higher scores indicate better caregiver perceived health.

Caregiver psychological factors.-- The 20-item Center for Epidemiologic Studies Depression scale (CES-D) was used to appraise depressive symptomatology (21). Caregiver upset was measured by the Caregiver Assessment of Function and Upset (CAFU) upset subscale with providing Activities of Daily Living (ADL) assistance (0 = not at all to 4 = extremely) (22). A mean upset score was computed across 7 ADLs, with higher scores indicating more upset. Caregiver readiness to learn care strategies was a single item rated by interventionists as 1 = precontemplation ("He or she is just old."); 2 = contemplation (not sure if intervention will work); 3 = preparation (wants/tries to understand); 4 = action/maintenance (actively participates and tries strategies) (17).

Patient illness severity.-- Patient cognitive function was evaluated using the MMSE (19). The number of troublesome behaviors (presence/absence) that occurred in the past month was assessed using the REACH-modified 24-item Revised Memory and Problem Behavior Checklist (RMPBC). Higher scores indicate a greater number of behaviors (23).

Treatment delivery.-- Because intervention delivery required therapeutic engagement, we used the 14-item Therapeutic Engagement Index (TEI) to measure the extent to which this was achieved. For each treatment session, interventionists rated verbal and nonverbal caregiver behaviors along a 5-point scale (0 = not at all to 4 = extremely) that reflected engagement. A mean score was derived across session ratings, with higher scores indicating positive engagement (24). A second set of measures reflected the number of delivery techniques used (yes/no) by interventionists in each session; 4 didactic (providing educational materials) and 8 active (simulation) techniques were considered. Two indices reflecting mean scores of didactic and active items used across sessions were computed.

Treatment receipt.-- We measured receipt as the number of contacts (home and telephone) received (dosage) in the active phase, and the number of problem areas that were addressed in the intervention (wandering, agitation) (7).

Statistical Analysis
Univariate statistics were used to describe caregiver sociodemographics and psychological factors, patient illness severity, treatment delivery, receipt, and enactment. To examine the importance of each predictor within the five domains (caregiver sociodemographics, caregiver psychological factors, patient illness severity, treatment delivery, and receipt), we conducted five separate regression analyses in which adherence (Strategy Use) was the dependent variable. In each regression equation, caregiver sociodemographic variables (age, gender, race, education, relationship to patient, health) were controlled. Then, adherence was regressed on those effects that were statistically significant in the previous five regressions. A final regression analysis included intraclass correlations to account for potential interventionist effects using a mixed effects model with interventionists as the random effect. Significance level was.05.


   RESULTS
Top
 Abstract
 Methods
 Results
Discussion
 References
 
Sample Characteristics
Caregivers were primarily female (73.3%), 61 years old, with a high school education (78.1%), and in good health (M = 13.5 ± 3.8). Most were adult children (47.6%), and 51.4% were African American. Caregivers reported elevated depressive symptoms (M = 14.6 ± 11.9), but low upset with ADL assistance (M = 0.7 ± 0.8). Most caregivers were at a high readiness stage, between preparation and action/maintenance. On average, patients were 80 years old, and mostly female (69.0%) with mean MMSE and RMBPC scores of 11.8 ± 7.2 and 9.4 ± 3.8, respectively (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1. Baseline Characteristics of Study Participants (N = 105).

 
Treatment Implementation Characteristics
ESP appeared to be delivered as intended, with caregivers positively engaged in treatment sessions (M = 41.6 ± 5.0). On average, active therapeutic techniques were used more than didactic (Table 2). Most caregivers received an adequate dose to benefit from ESP. Reasons for receiving fewer than six active phase sessions included bereavement (n = 5), nursing home placement (n = 3), or inactivation (n = 6). An average of 4.6 problem areas per caregiver was addressed.


View this table:
[in this window]
[in a new window]

  		Table 2. Treatment Implementation Characteristics (N = 105).

 
Predictors of Adherence
An average of 27 strategies per caregiver (SD = 20.0; actual range = 2–93) or approximately six strategies per problem area were provided by interventionists. Of the strategies provided, 18 (or a median of 76%) were used by the end of maintenance, with 95% of participants using at least one strategy (Table 2). Separate regression models for each domain revealed that caregiver health, readiness, active techniques, dosage, and problem areas were significant predictors (Table 3). The final regression model shows the relative importance of identified predictors (Table 4). Dosage (receipt) was the strongest predictor (ß =.36), followed by active therapeutic techniques (delivery) (ß =.31), caregiver health (ß =.27), and number of problems addressed (receipt) (ß =.20). Although not statistically significant, caregiver readiness was associated with higher adherence. Other caregiver (age, gender, race, education, relationship to patient, depression, upset), patient illness severity (cognitive status, behavioral occurrence), and treatment delivery (therapeutic engagement, didactic training) characteristics were not statistically significant predictors of strategy use. The intraclass correlations did not reveal interventionist effects on the final model for strategy use.


View this table:
[in this window]
[in a new window]

  		Table 3. Separate Regression Analyses of Caregiver, Patient, and Treatment Implementation Characteristics on Caregiver Adherence (Strategy Use) (N = 105).

 

View this table:
[in this window]
[in a new window]

  		Table 4. Final Regression Model of Caregiver Adherence (Strategy Use) (N = 105).

 

   DISCUSSION
Top
 Abstract
 Methods
 Results
 Discussion
References
 
Adherence to behavioral and lifestyle-changing treatments is a widely recognized concern in health services research, but has not been systematically examined in caregiver studies. With the majority of persons with dementia cared for at home, understanding contributory factors to caregiver adherence is important for developing more effective approaches to mitigate the catastrophic effects of caregiving. Consistent with Lichstein and others, our main hypothesis was that if the intervention was delivered and received as intended, treatment enactment or adherence would be enhanced. We also examined caregiver health and psychological well-being and patient illness characteristics as predictors based on previous adherence research.

We found that caregivers with better physical health used more intervention strategies than those reporting poor health, suggesting that caregivers with poor health may face double jeopardy; declining health and reduced ability to derive intervention benefits. Thus, caregivers with poor health may benefit from learning strategies for taking care of themselves to benefit from skills training. It also suggests the importance of including instruction in preventive health practices and management of stress in caregiver interventions.

Consistent with our hypothesis, we found that treatment implementation factors including use of active techniques (role play), treatment dosage, and number of problem areas were significant predictors of adherence. Previous studies have shown that engaging caregivers by using hands-on techniques is more effective in skill enhancement than are prescriptive or educational approaches alone (25). Also, we considered adequate dosage as a prerequisite for caregivers to enact strategies. As expected, caregivers with a greater number of contacts were more likely to use recommended strategies, possibly because they had more practice opportunities with an interventionist. Additionally, caregivers who were willing and able to identify more problem areas, reflecting engagement in the intervention, used more strategies (7). It should be noted that a caregiver who addressed one problem area may have used a large number of strategies, whereas a caregiver who identified two or more areas may have integrated only one strategy per problem. That is, the number of problems addressed did not impact the number of strategies recommended, and subsequently used.

As expected, although marginally significant, caregivers at a higher stage of readiness used a greater number of strategies. In our study, a high readiness score indicated that a caregiver understood the disease and was ready to try different care approaches. It may be that caregivers who are ready had previously experimented with care strategies and were thus more prepared to try new approaches. As suggested by the Transtheoretical Model, individuals who are behaviorally ready to change and who receive the requisite technical skills to change are successful. The findings imply that intervention strategies should be tailored to caregiver readiness levels to enhance adherence (26). Caregivers with low readiness may initially need a better understanding of the disease and their role prior to learning specific strategies to provide care effectively.

Of importance are the factors that did not predict adherence. Patient cognitive status and number of troublesome behaviors were not statistically significant predictors of adherence. This finding suggests that caregivers adhere to skill-building training independent of the illness severity and that factors related to caregiver appraisal of their own well-being and treatment implementation are of greater importance.

There are several study limitations. Interventionists provided the treatment implementation data, a potential source of response bias. To address this, we added a variance component for interventionists to our regression analyses and found that it did not affect predictive relationships. Nevertheless, it may be desirable to obtain objective observations of treatment delivery, receipt, and enactment in future intervention research. Also, we restricted this study to caregivers who participated in four or more intervention sessions. Because most caregivers had the intended contact number, the relationship of dosage to adherence should be interpreted cautiously. More research is warranted to fully understand factors contributing to adherence among caregivers who did not have the opportunity to use strategies due to institutionalization or death of their relatives. Finally, this sample of mostly high school–educated caregivers volunteered for participation through media announcements and agency referrals. This may represent a highly motivated group who sought new skills. Findings may not generalize to caregivers who did not volunteer but who need intervention. Nevertheless, the majority of participants were African Americans, an underserved caregiver population, who demonstrated high levels of adherence.

Summary
The present study identifies multiple modifiable factors that impact caregiver adherence to a skill-enhancement intervention. The data suggest that interventions should attend to caregiver health, use active instructional techniques, and allow sufficient contact to address multiple problem areas. Given the significance of treatment implementation to adherence, intervention research should include careful measures of treatment components and evaluate their impact on adherence. Future research will need to evaluate whether the predictors identified in this study can be generalized or are intervention specific.


   Acknowledgments
Top
 Abstract
 Methods
 Results
 Discussion
 References
 
The research reported in this article was supported by the National Institute on Aging Grant U01 AG13265 and by the Pennsylvania Department of Health Grant SAP 4100027298.

We acknowledge Laraine Winter for her thoughtful comments on previous drafts.

Walter W. Hauck is now with U.S. Pharmacopeia, Rockville, Maryland.


   Footnotes
Top
 Abstract
 Methods
 Results
 Discussion
 References
 
Decision Editor: Darryl Wieland, PhD, MPH

Received June 9, 2006

Accepted September 25, 2006


   References
Top
 Abstract
 Methods
 Results
 Discussion
 References
 

  1. Bourgeois MS, Schulz R, Burgio L. Interventions for caregivers of patients with Alzheimer's disease: a review and analysis of content, process, and outcomes. Int J Aging Hum Dev. 1996;43:35-92.[Medline]
  2. Schulz R, O'Brien A, Czaja S, et al. Dementia caregiver intervention research: in search of clinical significance. Gerontologist. 2002;42:589-602.[Abstract/Free Full Text]
  3. Sörensen S, Pinquart M, Duberstein P. How effective are interventions with caregivers? An updated meta-analysis. Gerontologist. 2002;42:356-372.[Abstract/Free Full Text]
  4. Gitlin LN, Winter L, Corcoran M, Dennis MP, Schinfeld S, Hauck WW. Effects of the home environmental skill-building program on the caregiver-care recipient dyad: 6-month outcomes from the Philadelphia REACH initiative. Gerontologist. 2003;43:532-546.[Abstract/Free Full Text]
  5. Mittelman MS, Roth DL, Coon DW, Haley WE. Sustained benefit of supportive intervention for depressive symptoms in caregivers of patients with Alzheimer's disease. Am J Psychiatry. 2004;161:850-856.[Abstract/Free Full Text]
  6. Castro CM, Wilcox S, O'Sullivan P, Baumann K, King AC. An exercise program for women who are caring for relatives with dementia. Psychosom Med. 2002;64:458-468.[Abstract/Free Full Text]
  7. Gitlin LN, Corcoran M, Winter L, Boyce A, Marcus S. Predicting participation and adherence to a home environmental intervention among family caregivers of persons with dementia. Fam Relat. 1999;48:363-372.
  8. Gitlin LN, Corcoran M, Martindale-Adams J, Malone C, Stevens A, Winter L. Identifying mechanisms of action: why and how does intervention work? In: Schulz R, ed. Handbook on Dementia Caregiving: Evidenced-Based Interventions for Family Caregivers. New York: Springer; 2000:225–248.
  9. Haynes RB. Improving patient adherence: state of art, with a special focus on medication taking for cardiovascular disorders. In: Burke LE, Ockene IS, eds. Compliance in Healthcare and Research. Armonk, NY: Futura; 2001:3–24.
  10. DiMatteo MR, Giordani PJ, Lepper HS, Croghan TW. Patient adherence and medical treatment outcomes: a meta-analysis. Med Care. 2002;40:794-811.[Medline]
  11. Lichstein KL, Riedel BW, Grieve R. Fair tests of clinical trials: a treatment implementation model. Adv Behav Res Ther. 1994;16:1-29.
  12. Bellg AJ, Borrelli B, Resnick B, et al. Enhancing treatment fidelity in health behavior change studies: best practices and recommendations from the NIH behavior change consortium. Health Psychol. 2004;23:443-451.[Medline]
  13. Burgio L, Corcoran M, Lichstein KL, et al. Judging outcomes in psychosocial interventions for dementia caregivers: the problem of treatment implementation. Gerontologist. 2001;41:481-489.[Abstract/Free Full Text]
  14. Becker MH, Maiman LA. Sociobehavioral determinants of compliance with health and medical care recommendations. Med Care. 1975;13:10-24.[Medline]
  15. Cramer JA, Spilker B, eds. Patient Compliance in Medical Practice and Clinical Trials. New York: Raven Press; 1991.
  16. Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall; 1986.
  17. Prochaska JO, Velicer WF. The transtheoretical model of health behavior change. Am J Health Promot. 1997;12:38-48.[Medline]
  18. Wisniewski SR, Belle SH, Coon DW, et al. The resources for enhancing Alzheimer's caregiver health (REACH): project design and baseline characteristics. Psychol Aging. 2003;18:375-384.[Medline]
  19. Folstein MF, Folstein SE, McHugh PR. "Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189-198.[Medline]
  20. Gitlin LN, Corcoran M. Occupational Therapy and Dementia Care: The Home Environmental Skill-Building Program for Individuals and Families. Bethesda, MD: The American Occupational Therapy Association, Inc.; 2005.
  21. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385-401.
  22. Gitlin LN, Roth DL, Burgio LD, et al. Caregiver appraisals of functional dependence in individuals with dementia and associated caregiver upset: psychometric properties of a new scale and response patterns by caregiver and care recipient characteristics. J Aging Health. 2005;17:148-171.[Abstract/Free Full Text]
  23. Roth DL, Burgio LD, Gitlin LN, et al. Psychometric analysis of the revised memory and behavior problems checklist: factor structure of occurrence and reaction ratings. Psychol Aging. 2003;18:906-915.[Medline]
  24. Chee YK, Dennis MP, Gitlin LN. Provider assessment of interactions with dementia caregivers: evaluation and application of the therapeutic engagement index. Clin Gerontologist. 2005;28:43-59.
  25. Belle SH, Czaja SJ, Schulz R, et al. Using a new taxonomy to combine the uncombinable: integrating results across diverse interventions. Psychol Aging. 2003;18:396-405.[Medline]
  26. Ockene IS, Hayman LL, Pasternak RC, Schron E, Dunbar-Jacob J. Task force #4—adherence issues and behavior changes: achieving a long-term solution. 33rd Bethesda Conference. J Am Coll Cardiol. 2002;40:630-640.[Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
PubMed
Right arrow PubMed Citation

HOME ARCHIVE SEARCH TABLE OF CONTENTS