Aggression, Alzheimer's Disease, Dementia, Depression, Interpersonal Relations, Pain
Conditions
Keywords
aggression, dementia, Alzheimer's disease, pain, depression, caregiver, behavioral health, counseling
Brief summary
This study will evaluate whether a home-based targeted education and skill training (Aggression Prevention Training or APT) will reduce aggression in persons with dementia (PWD) and pain/pain-related features more than usual care plus supportive telephone calls. Half of the participants will receive APT and half will receive supportive telephone calls.
Detailed description
Eighty percent of PWD have behavioral or psychological disturbances, including 40% that are aggressive behaviors. The prevalence of pain in PWD is about 60%, and it is a strong predictor of aggression. The biopsychosocial model of pain posits that pain is bidirectionally related to psychological factors (ie, depression) and social support factors (ie, quality of caregiver/PWD relationship) in addition to biological factors. Thus, depression and quality of the caregiver/PWD relationship can be seen as pain-related features. Caregivers are ideally suited to help address pain, depression, and the caregiver/PWD relationship, thus preventing the development of aggression; but they need tools to assist them in identifying and managing these symptoms. Prior studies of aggression treatment have not examined using a preventive strategy to decrease incidence of aggression in persons with dementia (PWD). Almost all studies have examined use of pharmacologic interventions following development of aggression.This 5-year randomized controlled trial based on the Unmet Needs Model will focus on preventing aggression in PWD with pain and pain-related features by providing the caregiver with targeted education and skill training. PWD and their caregivers will be randomized to APT or to an enhanced usual primary care condition (EU-PC). APT will use active learning tools, including didactics, role-playing, and multimedia \[eg, books and digital versatile discs (DVDs)\] to educate and provide skill training for the caregiver. The 6-8 modules in the intervention will include 4 core modules that address 4 main aggression risk factors: a) recognizing pain, b) treating pain, c) increasing pleasant activities, and d) improving patient-caregiver communication. Caregivers can select 2 to 3 additional elective sessions; elective selection is guided by the needs of the dyad to further enhance skills related to these core topics. EU-PC provides the patient and caregiver educational materials on pain, notifies the primary care provider of the PWD's level of pain and depression, and provides 8 weekly supportive telephone calls to caregivers. PWD and caregiver outcomes will be collected at baseline, 3, 6 and 12 months. Data analysis will include both univariate descriptive statistics and inferential statistics, including regression models, repeated measure modeling and Cox proportional hazards models.
Interventions
BEHAVIORALAggression Prevention Training (APT)
OTHEREnhanced Usual Primary Care (EU-PC)
Sponsors
National Institute of Nursing Research (NINR)
Baylor College of Medicine
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
PREVENTION
Masking
SINGLE (Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Healthy volunteers
No
Inclusion criteria
1. Documented diagnosis of dementia 2. Clinically significant pain, depression, or caregiver/patient relationship difficulties (either self-report or caregiver proxy-report). 3. Receives care from Baylor College of Medicine Geriatric Medicine Associates or Alzheimer's Disease and Memory Disorders Center, or Kelsey Seybold Clinics 4. Has an informal caregiver willing to participate in the study who sees the patient at least 8 hours/week and at least twice/week 5. Speaks English 6. Lives within a 40-mile radius of the coordinating center
Exclusion criteria
1. Advanced dementia based on inability to complete the Mental Impairment Screen-Telephone Version or a Functional Assessment Staging Tool score \> 6. 2. History of aggression during the one month prior to screening or baseline 3. Resides in a long-term care facility
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Number of Participants With Aggression as Per the Cohen Mansfield Agitation Inventory, Aggression Subscale | one year | Aggression is measured on a 7-point Likert scale for frequency and a 5-point Likert scale for disruptiveness. Aggression is considered present if a participant scores over one on both frequency (more than never) and disruptiveness (at least a little) on any of 13 aggressive behaviors, including spitting, verbal aggression, hitting, kicking, grabbing, pushing, throwing, biting, scratching, hurting self/others, destroying property, or making inappropriate verbal or physical sexual advances. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Positive Caregiving Attributes--Positive Aspects of Caregiving Scale | one year | The 9-item Positive Aspects of Caregiving Scale presents statements about a caregiver's mental or affective state in the context of the caregiving experience. Responses are provided on a 5-point agree/disagree scale and designed to assess perception of benefits within the caregiving context, such as feeling useful, feeling appreciated, and finding meaning. Higher scores (range 9-45) represent more positive appraisals. |
| Behavior Problems--Revised Memory and Behavior Checklist | one year | Revised Memory and Behavior Checklist (RMBCL) is a 24-item informant-based measure of observable behavior problems in PWD, including memory-related, disruptive, and depressive behaviors. Scores are computed for the presence/absence of each problem first and then for caregiver reaction or the extent to which caregivers were bothered or distressed by each behavior (0-4). Total score is the sum of reaction scores for all endorsed behaviors. Possible range is 0-96. A higher score indicates a worse outcome. |
| Pain--Philadelphia Geriatric Pain Intensity Scale (Overall Pain as Reported by the PWD) | one year | The Philadelphia Geriatric Pain Intensity Scale will be administered to the PWD to measure pain. It consists of 4 items assessing the extent to which the PWD has been bothered by pain over the past several weeks--at present, when pain was at its worst, when pain was at its least, and overall. These items are rated on a 0- to 5-point Likert scale (not at all to extremely). A fifth item asks for number of days per week that pain was really bad, and a sixth asks for a rating of how much pain has interfered with day-to-day activities. We report only on the item about overall pain. |
| Caregiver Burden--Zarit Burden Interview | one year | The Zarit Burden Interview is a 22-item instrument measuring perceived impact of caregiving on the caregiver's financial status, physical status, physical health, emotional health, and social activities. Questions are answered on a 5-point Likert-type scale (0=never, 4=nearly always). The total scale score ranges from 0-88. |
| Caregiver-Patient Relationship Quality--Mutuality Scale | one year | The Mutuality Scale and is a 15-item instrument measuring the positive quality of the relationship between caregiver and care receiver. Questions are answered by the caregiver on a 5-point Likert-type scale (0=never, 4=a great deal). Its 4 subscales represent domains of shared values, affective closeness, shared pleasurable activities and reciprocity. The total score ranges from 0-4 and is the sum of individual items divided by the number of items answered. High scores indicate a relationship characterized by communication, shared pleasurable activities, common values, and reciprocity. |
| Pain--Philadelphia Geriatric Pain Intensity Scale (Overall Pain as Reported by the Caregiver) | one year | The Philadelphia Geriatric Pain Intensity Scale will be administered to the caregiver to measure caregiver report of PWD pain. It consists of 4 items assessing the extent to which the the caregiver feels the PWD has been bothered by pain over the past several weeks--at present, when pain was at its worst, when pain was at its least, and overall. These items are rated on a 0- to 5-point Likert scale (not at all to extremely). A fifth item asks for number of days per week that pain was really bad, and a sixth asks for a rating of how much pain has interfered with day-to-day activities. We report only on the item about overall pain. |
| Depression--Geriatric Depression Screen (GDS), Caregiver Version | one year | The 30-item GDS will be administered to the caregiver to measure depression. Items are answered with Yes/No. Total score ranges from 0-30. A score of 11 or greater is a possible indicator of depression. |
Participant flow
Recruitment details
2,230 persons with dementia (PWD) were contacted for telephone screening; 253 were interested and eligible for the study
Pre-assignment details
239 signed written informed consent and 233 completed baseline assessments (5 were lost and 1 withdrew after signing consent); 4 were excluded at baseline due to the presence of aggression in the 3 months prior to baseline based on Cohen-Mansfield Agitation Inventory (CMAI); 1 participant withdrew prior to randomization; 228 participants were randomly assigned to a study arm (114 in each group).
Participants by arm
| Arm | Count |
|---|---|
| Aggression Prevention Training (APT) APT will use active learning tools, including didactics, role-playing, and multimedia (eg, books and DVDs) to educate and provide skill training for the caregiver. The 6-8 modules in the intervention will include 4 core modules that address 4 main aggression risk factors: a) recognizing pain, b) treating pain, c) increasing pleasant activities, and d) improving patient-caregiver communication. Caregivers can select 2 to 3 additional elective sessions; elective selection is guided by the needs of the dyad to further enhance skills related to these core topics. Sessions will take place in the patient's home.
Aggression Prevention Training (APT) | 114 |
| Enhanced Usual Primary Care (EU-PC) EU-PC provides the patient and caregiver educational materials on pain, notifies the primary care provider of the PWD's level of pain and depression, and provides 8 weekly supportive telephone calls to caregivers.
Enhanced Usual Primary Care (EU-PC) | 114 |
| Total | 228 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Caregiver death | 1 | 0 |
| Overall Study | Entered Long-term Care | 4 | 7 |
| Overall Study | Lost to Follow-up | 12 | 12 |
| Overall Study | Person with dementia (PWD) death | 7 | 8 |
| Overall Study | Withdrawal by Subject | 12 | 9 |
Baseline characteristics
| Characteristic | Aggression Prevention Training (APT) | Enhanced Usual Primary Care (EU-PC) | Total |
|---|---|---|---|
| Age, Continuous | 77.98 years STANDARD_DEVIATION 9.16 | 78.31 years STANDARD_DEVIATION 8.75 | 78.15 years STANDARD_DEVIATION 8.97 |
| Caregiver-reported Overall Pain | 1.44 units on a scale STANDARD_DEVIATION 1.11 | 1.16 units on a scale STANDARD_DEVIATION 1.12 | 1.30 units on a scale STANDARD_DEVIATION 1.12 |
| Geriatric Depression Scale (GDS) | 11.88 units on a scale STANDARD_DEVIATION 6.25 | 11.10 units on a scale STANDARD_DEVIATION 6.48 | 11.49 units on a scale STANDARD_DEVIATION 6.36 |
| Individual with Dementia-reported Overall Pain | 1.38 units on a scale STANDARD_DEVIATION 1.12 | 1.13 units on a scale STANDARD_DEVIATION 1.24 | 1.25 units on a scale STANDARD_DEVIATION 1.18 |
| Mutuality Scale | 2.93 units on a scale STANDARD_DEVIATION 0.79 | 3.05 units on a scale STANDARD_DEVIATION 0.76 | 2.99 units on a scale STANDARD_DEVIATION 0.78 |
| Positive Aspects of Caregiving | 24.18 units on a scale STANDARD_DEVIATION 9 | 23.88 units on a scale STANDARD_DEVIATION 9.54 | 24.03 units on a scale STANDARD_DEVIATION 9.25 |
| Race/Ethnicity, Customized Black | 24 Participants | 24 Participants | 48 Participants |
| Race/Ethnicity, Customized Hispanic | 9 Participants | 12 Participants | 21 Participants |
| Race/Ethnicity, Customized Non-Hispanic White | 78 Participants | 75 Participants | 153 Participants |
| Revised Memory and Behavior Checklist (RMBC) - Total Frequency | 8.13 units on a scale STANDARD_DEVIATION 3.79 | 7.56 units on a scale STANDARD_DEVIATION 3.43 | 7.85 units on a scale STANDARD_DEVIATION 3.62 |
| Sex: Female, Male Female | 48 Participants | 49 Participants | 97 Participants |
| Sex: Female, Male Male | 66 Participants | 65 Participants | 131 Participants |
| Zarit Burden Inventory | 27.77 units on a scale STANDARD_DEVIATION 15.03 | 27.00 units on a scale STANDARD_DEVIATION 13.29 | 27.39 units on a scale STANDARD_DEVIATION 15.13 |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 7 / 114 | 7 / 114 |
| other Total, other adverse events | 0 / 0 | 0 / 0 |
| serious Total, serious adverse events | 15 / 114 | 18 / 114 |
Outcome results
Primary
Number of Participants With Aggression as Per the Cohen Mansfield Agitation Inventory, Aggression Subscale
Aggression is measured on a 7-point Likert scale for frequency and a 5-point Likert scale for disruptiveness. Aggression is considered present if a participant scores over one on both frequency (more than never) and disruptiveness (at least a little) on any of 13 aggressive behaviors, including spitting, verbal aggression, hitting, kicking, grabbing, pushing, throwing, biting, scratching, hurting self/others, destroying property, or making inappropriate verbal or physical sexual advances.
Time frame: one year
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Aggression Prevention Training (APT) | Number of Participants With Aggression as Per the Cohen Mansfield Agitation Inventory, Aggression Subscale | 29 Participants |
| Enhanced Usual Primary Care (EU-PC) | Number of Participants With Aggression as Per the Cohen Mansfield Agitation Inventory, Aggression Subscale | 23 Participants |
Comparison: Sample-size calculations were performed, based on the ability to detect a small-to-moderate difference (Cohen's h = 0.4) in rate of aggression onset over a 1-year period between APT and EU-PC, assuming 80% power and a type I error rate of 5%. Given an anticipated rate of aggression onset over 1 year of 37% for EU-PC, an ES of h = 0.4 allows detection of aggression onset in APT as high as 19%. Given this effect size and up to 10% attrition, our goal was to include 220 total participants.p-value: 0.2995% CI: [0.78, 2.32]Regression, Cox
p-value: 0.27Chi-squared
Secondary
Behavior Problems--Revised Memory and Behavior Checklist
Revised Memory and Behavior Checklist (RMBCL) is a 24-item informant-based measure of observable behavior problems in PWD, including memory-related, disruptive, and depressive behaviors. Scores are computed for the presence/absence of each problem first and then for caregiver reaction or the extent to which caregivers were bothered or distressed by each behavior (0-4). Total score is the sum of reaction scores for all endorsed behaviors. Possible range is 0-96. A higher score indicates a worse outcome.
Time frame: one year
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Aggression Prevention Training (APT) | Behavior Problems--Revised Memory and Behavior Checklist | 7.86 units on a scale | Standard Deviation 4.37 |
| Enhanced Usual Primary Care (EU-PC) | Behavior Problems--Revised Memory and Behavior Checklist | 7.51 units on a scale | Standard Deviation 4.08 |
Comparison: Differences between APT and EU-PC in change over time (baseline, and 3, 6 and 12 months) for our secondary outcomes were evaluated using individual linear growth curve models (SAS Proc Mixed, SAS Institute, Inc., Cary, NC) with an autoregressive covariance structure type.~The main interest was the interaction between group (APT vs EU-PC) and time.p-value: 0.8Mixed Models Analysis
Secondary
Caregiver Burden--Zarit Burden Interview
The Zarit Burden Interview is a 22-item instrument measuring perceived impact of caregiving on the caregiver's financial status, physical status, physical health, emotional health, and social activities. Questions are answered on a 5-point Likert-type scale (0=never, 4=nearly always). The total scale score ranges from 0-88.
Time frame: one year
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Aggression Prevention Training (APT) | Caregiver Burden--Zarit Burden Interview | 26.12 units on a scale | Standard Deviation 14.72 |
| Enhanced Usual Primary Care (EU-PC) | Caregiver Burden--Zarit Burden Interview | 28.37 units on a scale | Standard Deviation 16.27 |
Comparison: Differences between APT and EU-PC in change over time (baseline, and 3, 6 and 12 months) for our secondary outcomes were evaluated using individual linear growth curve models (SAS Proc Mixed, SAS Institute, Inc., Cary, NC) with an autoregressive covariance structure type.~The main interest was the interaction between group (APT vs EU-PC) and time.p-value: 0.19Mixed Models Analysis
Secondary
Caregiver-Patient Relationship Quality--Mutuality Scale
The Mutuality Scale and is a 15-item instrument measuring the positive quality of the relationship between caregiver and care receiver. Questions are answered by the caregiver on a 5-point Likert-type scale (0=never, 4=a great deal). Its 4 subscales represent domains of shared values, affective closeness, shared pleasurable activities and reciprocity. The total score ranges from 0-4 and is the sum of individual items divided by the number of items answered. High scores indicate a relationship characterized by communication, shared pleasurable activities, common values, and reciprocity.
Time frame: one year
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Aggression Prevention Training (APT) | Caregiver-Patient Relationship Quality--Mutuality Scale | 2.91 units on a scale | Standard Deviation 0.85 |
| Enhanced Usual Primary Care (EU-PC) | Caregiver-Patient Relationship Quality--Mutuality Scale | 2.99 units on a scale | Standard Deviation 0.78 |
Comparison: Differences between APT and EU-PC in change over time (baseline, and 3, 6 and 12 months) for our secondary outcomes were evaluated using individual linear growth curve models (SAS Proc Mixed, SAS Institute, Inc., Cary, NC) with an autoregressive covariance structure type.~The main interest was the interaction between group (APT vs EU-PC) and time.p-value: 0.77Mixed Models Analysis
Secondary
Depression--Geriatric Depression Screen (GDS), Caregiver Version
The 30-item GDS will be administered to the caregiver to measure depression. Items are answered with Yes/No. Total score ranges from 0-30. A score of 11 or greater is a possible indicator of depression.
Time frame: one year
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Aggression Prevention Training (APT) | Depression--Geriatric Depression Screen (GDS), Caregiver Version | 12.06 units on a scale | Standard Deviation 6.98 |
| Enhanced Usual Primary Care (EU-PC) | Depression--Geriatric Depression Screen (GDS), Caregiver Version | 11.43 units on a scale | Standard Deviation 7.42 |
Comparison: Differences between APT and EU-PC in change over time (baseline, and 3, 6 and 12 months) for our secondary outcomes were evaluated using individual linear growth curve models (SAS Proc Mixed, SAS Institute, Inc., Cary, NC) with an autoregressive covariance structure type.~The main interest was the interaction between group (APT vs EU-PC) and time.p-value: 0.22Mixed Models Analysis
Secondary
Pain--Philadelphia Geriatric Pain Intensity Scale (Overall Pain as Reported by the Caregiver)
The Philadelphia Geriatric Pain Intensity Scale will be administered to the caregiver to measure caregiver report of PWD pain. It consists of 4 items assessing the extent to which the the caregiver feels the PWD has been bothered by pain over the past several weeks--at present, when pain was at its worst, when pain was at its least, and overall. These items are rated on a 0- to 5-point Likert scale (not at all to extremely). A fifth item asks for number of days per week that pain was really bad, and a sixth asks for a rating of how much pain has interfered with day-to-day activities. We report only on the item about overall pain.
Time frame: one year
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Aggression Prevention Training (APT) | Pain--Philadelphia Geriatric Pain Intensity Scale (Overall Pain as Reported by the Caregiver) | 1.48 units on a scale | Standard Deviation 1.07 |
| Enhanced Usual Primary Care (EU-PC) | Pain--Philadelphia Geriatric Pain Intensity Scale (Overall Pain as Reported by the Caregiver) | 1.20 units on a scale | Standard Deviation 1.1 |
Comparison: Differences between APT and EU-PC in change over time (baseline, and 3, 6 and 12 months) for our secondary outcomes were evaluated using individual linear growth curve models (SAS Proc Mixed, SAS Institute, Inc., Cary, NC) with an autoregressive covariance structure type.~The main interest was the interaction between group (APT vs EU-PC) and time.p-value: 0.64Mixed Models Analysis
Secondary
Pain--Philadelphia Geriatric Pain Intensity Scale (Overall Pain as Reported by the PWD)
The Philadelphia Geriatric Pain Intensity Scale will be administered to the PWD to measure pain. It consists of 4 items assessing the extent to which the PWD has been bothered by pain over the past several weeks--at present, when pain was at its worst, when pain was at its least, and overall. These items are rated on a 0- to 5-point Likert scale (not at all to extremely). A fifth item asks for number of days per week that pain was really bad, and a sixth asks for a rating of how much pain has interfered with day-to-day activities. We report only on the item about overall pain.
Time frame: one year
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Aggression Prevention Training (APT) | Pain--Philadelphia Geriatric Pain Intensity Scale (Overall Pain as Reported by the PWD) | 1.19 units on a scale | Standard Deviation 1.14 |
| Enhanced Usual Primary Care (EU-PC) | Pain--Philadelphia Geriatric Pain Intensity Scale (Overall Pain as Reported by the PWD) | 1.12 units on a scale | Standard Deviation 1.19 |
Comparison: Differences between APT and EU-PC in change over time (baseline, and 3, 6 and 12 months) for our secondary outcomes were evaluated using individual linear growth curve models (SAS Proc Mixed, SAS Institute, Inc., Cary, NC) with an autoregressive covariance structure type.~The main interest was the interaction between group (APT vs EU-PC) and time.p-value: 0.89Mixed Models Analysis
Secondary
Positive Caregiving Attributes--Positive Aspects of Caregiving Scale
The 9-item Positive Aspects of Caregiving Scale presents statements about a caregiver's mental or affective state in the context of the caregiving experience. Responses are provided on a 5-point agree/disagree scale and designed to assess perception of benefits within the caregiving context, such as feeling useful, feeling appreciated, and finding meaning. Higher scores (range 9-45) represent more positive appraisals.
Time frame: one year
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Aggression Prevention Training (APT) | Positive Caregiving Attributes--Positive Aspects of Caregiving Scale | 25.87 units on a scale | Standard Deviation 8.46 |
| Enhanced Usual Primary Care (EU-PC) | Positive Caregiving Attributes--Positive Aspects of Caregiving Scale | 25.00 units on a scale | Standard Deviation 9.53 |
Comparison: Differences between APT and EU-PC in change over time (baseline, and 3, 6 and 12 months) for our secondary outcomes were evaluated using individual linear growth curve models (SAS Proc Mixed, SAS Institute, Inc., Cary, NC) with an autoregressive covariance structure type.~The main interest was the interaction between group (APT vs EU-PC) and time.p-value: 0.06Mixed Models Analysis