FULL-LENGTH ORIGINAL RESEARCH
PACES in epilepsy: Results of a self-management randomized controlled trial *†‡1Robert T. Fraser, †1Erica K. Johnson, ‡Steven Lashley, §Jason Barber, *Naomi Chaytor, ¶John W. Miller, **Paul Ciechanowski, §††Nancy Temkin, and ‡‡Lisa Caylor Epilepsia, 56(8):1264–1274, 2015 doi: 10.1111/epi.13052
SUMMARY
Dr. Robert T. Fraser, professor/director NVSU; UW Rehabilitation Medicine/ Epilepsy Center.
Dr. Erica Johnson is a rehabilitation psychologist and researcher at the University of Washington.
Objective: Self-management challenges facing adults with epilepsy include limited understanding of the condition and treatment, associated psychosocial issues, and lack of community integration. Self-management interventions improve patients’ medical, life role, and emotional management. Previous interventions, developed from expert opinion, indicated issues with participant engagement/retention, and limited follow-up periods. PACES in Epilepsy addressed methodologic concerns by utilizing patient needs assessment data (n = 165) to derive self-management content and program features for evaluation via randomized controlled trial (RCT). Methods: Participants were adults with chronic epilepsy (n = 83), without serious mental illness or substantive intellectual impairment, who were recruited from two epilepsy centers. Participants were assigned randomly to intervention or treatmentas-usual groups. Outcomes included the Epilepsy Self-Management Scale (ESMS), Epilepsy Self-Efficacy Scale (ESES), Quality of Life in Epilepsy-31 (QOLIE-31), Patient Health Questionnaire-9 (PHQ-9), and the Generalized Anxiety Disorder-7 (GAD-7), administered at baseline, postintervention (8 weeks), and 6 months postintervention. The intervention was an 8-week group of 6–8 adults co-led by a psychologist and trained peer with epilepsy that met one evening per week at a hospital for 75 min. Topics included medical, psychosocial, cognitive, and self-management aspects of epilepsy, in addition to community integration and optimizing epilepsy-related communication. The treatment group provided satisfaction ratings regarding program features. Results: PACES participants (n = 38) improved relative to controls (n = 40) on the ESMS (p < 0.001) and subscales [Information (p < 0.001); Lifestyle (p < 0.002)]; ESES (p < 0.001); and QOLIE-31 (p = 0.002). At 6-month follow up, PACES participants remained improved on the ESMS (p = 0.004) and Information subscale (p = 0.009); and Energy/Fatigue (p = 0.032) and Medication Effects (p = 0.005) of the QOLIE-31. Attrition in both groups was low (8% in each group) and all program satisfaction ratings exceeded 4.0/5.0, with leadership (4.76), topics (4.53), and location (4.30) as the most highly rated aspects. Significance: A consumer generated epilepsy self-management program appears to be a promising intervention from multiple perspectives, particularly in relation to disability management. KEY WORDS: Epilepsy, Self-management, Self-efficacy, Depression, Quality of life, Consumer-generated intervention.
In 1997, the Centers for Disease Control and Prevention (CDC) began crafting a public health agenda culminating in the 2003 Living Well with Epilepsy II Conference, sponsored by the Epilepsy Foundation and other epilepsy organizations. A number of priority recommendations from this conference related to self-man-
agement in epilepsy as a path to disease burden reduction. In 2007, the CDC supported the development of the Prevention Research Center’s Managing Epilepsy Well Network (MEWN), which seeks to advance epilepsy self-management research and dissemination. The network focuses on site-specific epilepsy management
1264
1265 PACES in Epilepsy RCT
Key Points • • • •
The PACES epilepsy self-management program was evaluated by a randomized controlled trial involving 83 adults with chronic epilepsy. PACES in Epilepsy, an in-person group intervention, appears successful in improving patient self-management skills and aspects of quality of life. This consumer-generated program produced high satisfaction ratings accompanied by good goal attainment. Future research will evaluate distance delivery and sustention of PACES benefits with the addition of follow-up booster contact.
research as well as cooperative research and development activities.1 Much of the literature on self-management intervention is based on the work of Lorig and colleagues at the Stanford Patient Education Research Center.2,3 Within that literature, emphasis is on self-efficacy and problem-solving, with three primary functions: (1) medical management, such as taking medication, following a special diet, or using medical devices; (2) role management, which refers to routine, changing, and creating new meaningful behaviors and life roles (e.g., adopting new recreational activities, changing responsibilities in household, and so on); and (3) emotional management, such as adjusting to a situation that has induced emotions including anger, fear, sadness, depression, or grief.4–6 A number of epilepsy-specific educational programs have been described in the literature, but a 2007 Cochrane review reported that only two were tested in randomized control trial contexts.7 Helgeson examined the efficacy of a psychoeducational treatment program for adults with epilepsy, the Sepulveda Epilepsy Education (SEE) Program.8 The SEE program involved 16 sessions with 20 participants Accepted May 5, 2015; Early View publication June 29, 2015. *Rehabilitation Medicine, University of Washington, Seattle, Washington, U.S.A.; †Health Promotion Research Center, University of Washington, Seattle, Washington, U.S.A.; ‡Neurology Vocational Services Unit, University of Washington, Seattle, Washington, U.S.A.; §Neurological Surgery, University of Washington, Seattle, Washington, U.S.A.; ¶Neurology, University of Washington, Seattle, Washington, U.S.A.; **Psychiatry and Behavioral Sciences, University of Washington, Seattle, Washington, U.S.A.; ††Biostatistics, University of Washington, Seattle, Washington, U.S.A.; and ‡‡Swedish Neuroscience Institute, Seattle, Washington, U.S.A. 1 These authors contributed equally to this article. Address correspondence to Robert T. Fraser, Neurology Vocational Services Unit, University of Washington-Harborview Medical Center, Box 359744, 325 Ninth Avenue, Seattle, WA 98104, U.S.A. E-mail: [email protected] Wiley Periodicals, Inc. © 2015 International League Against Epilepsy
over 2 days, covering both medical and social-emotional aspects of managing one’s epilepsy. The SEE Program had some significant effects on certain medical concerns such as fear of death or brain damage and extent of overall misinformation and misconception of the disability at 4 months post-program. The authors noted a nonsignificant trend toward reduced seizure frequency and improved compliance. No significant changes were noted on measures of psychosocial adjustment and mood. The second study tested the Modular Service Package for Epilepsy (MOSES), a European program with nine modules presented over 2 days to groups of 7 to 10 adults with epilepsy.9 Again, significant improvements in epilepsy knowledge, coping with epilepsy, information seeking activity, and adherence to antiepileptic drug treatment were seen at 6-month follow-up. Significant reduction in seizure frequency was also noted for the intervention compared to the control group. There were no significant differences on measures of activities of daily living, mobility/leisure behavior, self-esteem, mood, and other psychosocial adjustment measures. The review authors noted some concerns about these studies, such as significant baseline differences between groups and higher proportions of people with partial seizures than would be expected in a community sample. Several studies have been completed within the MEWN related to self-management interventions. DiIorio et al.10 developed an Internet-based intervention, WebEase, to address the important role of social learning, behavior theory, and practical issues such as patient needs and transportation barriers. This program addressed three behavioral areas: medication management, stress, and sleep management. This project included an expanded set of outcome measures relative to prior investigations (e.g., epilepsy self-management and knowledge, medication self-adherence, sleep quality, level of stress, quality of life, and epilepsy selfefficacy).11–14 Those completing at least part of the three WebEase modules reported significantly more medication adherence and epilepsy self-efficacy. Dropout rate as related to program completion, however, was substantial and final outcome was assessed within a 12-week period. Another Emory study, Project UPLIFT, evaluated an 8week cognitive-behavioral mindfulness program for preventing depression in people with epilepsy.15 Upon conclusion, the incidence of major depression was significantly lower in the intervention compared to the control group, as were the number of depressive symptoms over the 10-week follow-up. There was no difference in intervention effectiveness between Web-based or telephone module delivery. The field of epilepsy self-management research is still evolving, but a number of observations can be made: 1 The range of medical and psychosocial adjustment needs addressed in these studies is often limited to medical needs, and a few other related issues. There is need for more encompassing interventions focusing not only on Epilepsia, 56(8):1264–1274, 2015 doi: 10.1111/epi.13052
1266 R. T. Fraser et al. medical barriers, but also a range of psychosocial adjustment areas. 2 These programs are generally planned by professionals, although focus groups can be involved. There is a need for more consumer-driven program development, consistent with community-based participatory research that underlies the MEWN. 3 There has been scant reporting as to patient satisfaction with these self-management programs as to content, mode of delivery, and so on. 4 To date, these programs can have a relatively short-term evaluation period. There is a need to study the durability of benefits from epilepsy self-management program over longer periods.
individual or group meetings with medical, epilepsy professional, and peer leadership were preferred by patients of all levels of functioning. Participants initially recommended six to eight 60- to 90-min sessions on a week night. The desired module formula was to emphasize education and emotional coping strategies. In contrast to other studies, the desired location was at a medical institution.
Development of the PACES Program
Site characteristics The University medical facility is located centrally in the Seattle-Metropolitan area and served by numerous bus lines. The intervention was delivered on-site in a conference room, three blocks from the UW Regional Epilepsy Center and within a mile of the Swedish epilepsy clinic. The study interventionists were two rotating Ph.D.-level rehabilitation psychologists and a trained peer mentor with epilepsy who has a bachelor’s degree and 7 years of experience as a neurologic employment specialist.
The first objective of the current study was to involve a diverse group of adults with epilepsy in the planning of a self-management intervention. A comprehensive review of research on medical and psychosocial concerns in epilepsy and needs assessment surveys used by epilepsy researchers and Epilepsy Foundation affiliates over the last 35 years was conducted to develop a survey on eight domains of concern: seizure activity, medical care, emotional functioning, employment, general health and well-being, independent living, and social life.16 A second section of the survey dealt with self-management program delivery preferences such as program emphasis and design, meeting times during the week, and program length. Information on epilepsy characteristics, demographic information, mood measures, and self-reported cognitive concerns was also collected. The survey was then refined by two focus groups of 10 adults with epilepsy, each at the University of Washington-Harborview Medical Center. The survey was then mailed to 250 adults with epilepsy who were being treated at either the UW-Harborview Medical Center or the Swedish Medical Center, and a group of 22 adults with epilepsy in local affiliate support groups. The survey participants in the mailing were screened for marked cognitive or emotional impairment by each center’s epileptologists. The Dillman approach was used in order to optimize participant response.17 A total of 165 returns were received (61% response rate). There was very high response reliability (internal consistency) across all domains in the survey. A comprehensive description of the survey findings is available in Fraser et al., with implications for the development of the PACES program.18 Of note, the survey findings revealed a significant subgroup with depression (PHQ-9 score ≥10) and/or three or more self-reported cognitive problems of moderate severity. Because of this, the PACES material was developed with emphasis on the range of mood and cognitive concerns across all modules. As to program logistics, face-to-face Epilepsia, 56(8):1264–1274, 2015 doi: 10.1111/epi.13052
Methods The protocol for the study was reviewed and approved by the UW and the Swedish Medical Center Institutional Review Boards. All participants provided informed consent.
Participants Participants were recruited through the UW Regional Epilepsy Center and Swedish Epilepsy Center, both in Seattle. Inclusion criteria were age 18 and older; able to speak, read, and write English; a medically established diagnosis of epilepsy at least 6 months from the initial diagnosis; and reasonably cognitively intact as measured by the Montreal Cognitive Assessment (MoCA).19 We implemented a lower cutoff score of 21 due to the cognitive impairment often experienced with epilepsy.20–22 Exclusion criteria were active severe mental illness/psychosis, or known cognitive impairment/IQ 100,000 Unknown Health insurance status Has health insurance
Overall 83
Treatment 41
Control 42
45.2 (12.5) 46 (55%)
44.9 (12.5) 23 (56%)
45.4 (12.6) 23 (55%)
0.863 1.000
67 (81%) 7 (8%) 9 (11%) 3 (4%) 2 (2%)
31 (76%) 5 (12%) 6 (15%) 2 (5%) 0 (0%)
36 (86%) 2 (5%) 3 (7%) 1 (2%) 2 (5%)
0.277 0.265 0.313 0.616 0.494
15 (18%) 43 (52%) 16 (19%) 34 (41%) 5 (6%) 10 (12%) 4 (5%) 0 (0%)
8 (20%) 20 (49%) 9 (22%) 14 (34%) 2 (5%) 6 (15%) 2 (5%) 0 (0%)
7 (17%) 23 (55%) 7 (17%) 20 (48%) 3 (7%) 4 (10%) 2 (5%) 0 (0%)
0.782 0.663 0.588 0.266 1.000 0.520 1.000
1 (1.2) 19 (23%) 25 (30%) 21 (25%) 10 (12%) 8 (10%)
2 (1.3) 9 (22%) 11 (27%) 11 (27%) 7 (17%) 3 (7%)
1 (1.2) 10 (24%) 14 (33%) 10 (24%) 3 (7%) 5 (12%)
0.593 0.658
15.0 (2.5) 2 (2%) 12 (14%) 14 (17%) 13 (16%) 28 (34%) 9 (11%) 5 (6%)
15.1 (2.4) 1 (2%) 5 (12%) 5 (12%) 8 (20%) 14 (34%) 6 (15%) 2 (5%)
14.8 (2.6) 1 (2%) 7 (17%) 9 (21%) 5 (12%) 14 (33%) 3 (7%) 3 (7%)
0.544 0.801
21 (26%) 12 (15%) 27 (33%) 22 (27%)
9 (22%) 5 (12%) 15 (37%) 12 (29%)
12 (29%) 7 (17%) 12 (29%) 10 (24%)
0.756
44 (53%) 30 (36%) 8 (10%) 1 (1%)
18 (44%) 19 (46%) 4 (10%) 0 (0%)
26 (62%) 11 (26%) 4 (10%) 1 (2%)
0.159
57.1 (61.4) 16 (29%) 11 (20%) 8 (15%) 9 (16%) 11 (20%) 29
59.4 (65.3) 10 (36%) 4 (14%) 3 (11%) 4 (14%) 7 (25%) 14
54.7 (58.2) 6 (22%) 7 (26%) 5 (19%) 5 (19%) 4 (15%) 15
0.742 0.572
70 (85%)
35 (88%)
35 (83%)
0.757
p-Valuea
a
All statistical significance values by ANOVA, Mann-Whitney, or Fisher’s exact test as appropriate.
Management (p < 0.001) and Life Style Management subscales (p = 0.002). The ESES (p < 0.001) and QOLIE-31 (p = 0.002) scores were also significant. In addition to the
total QOLIE-31 score, subscales of overall QOL, Emotional Well-Being, and Energy/Fatigue were also significant (all p < 0.01). Depression was also significantly reduced at Epilepsia, 56(8):1264–1274, 2015 doi: 10.1111/epi.13052
1270 R. T. Fraser et al. Table 2. Baseline assessment N Overall health Mean (SD) 1 - Excellent 2 - Very good 3 - Good 4 - Fair 5 - Poor Life satisfaction rating Mean (SD) 1 - Very satisfied 2 - Satisfied 3 - Dissatisfied 4 - Very dissatisfied Happiness rating Mean (SD) 1 - Very happy 2 - Happy 3 - Neutral 4 - Not very happy 5 - Not happy at all MoCA mean (SD) Baseline assessment (N) Subjects PHQ (0–27) GAD (0–21) ESMS (1–5) IM - Information mgmt. LM - Lifestyle mgmt. MM - Medication mgmt. SE - Seizure mgmt. SM - Safety mgmt. ESES (0–10) QOLIE-31 (0–100) SW - Seizure worry OQ - Overall QOL EW - Emotional well-being EF - Energy/fatigue CO – Cognitive ME - Medication effects SF - Social function
Overall 83
Treatment 41
Control 42
2.8 (1.1) 12 (14%) 21 (25%) 31 (37%) 14 (17%) 5 (6%)
3.0 (1.0) 1 (2%) 14 (34%) 14 (34%) 9 (22%) 3 (7%)
2.5 (1.2) 11 (26%) 7 (17%) 17 (40%) 5 (12%) 2 (5%)
0.059 0.010
2.3 (0.8) 11 (13%) 45 (54%) 20 (24%) 7 (8%)
2.4 (0.8) 3 (7%) 24 (59%) 9 (22%) 5 (12%)
2.2 (0.8) 8 (19%) 21 (50%) 11 (26%) 2 (5%)
0.206 0.282
2.6 (0.9) 8 (10%) 35 (42%) 28 (34%) 9 (11%) 3 (4%) 26.1 (2.7)
2.6 (0.9) 2 (5%) 19 (46%) 13 (32%) 6 (15%) 1 (2%) 26.3 (2.5)
2.5 (1.0) 6 (14%) 16 (38%) 15 (36%) 3 (7%) 2 (5%) 26.0 (2.9)
0.519 0.493
82 (99%) 8.4 (6.1) 5.6 (5.2) 3.6 (0.4) 2.5 (0.8) 3.2 (0.6) 4.4 (0.4) 4.2 (0.6) 3.9 (0.6) 7.7 (1.1) 58 (16) 60 (25) 63 (17) 63 (18) 50 (20) 54 (24) 60 (31) 61 (24)
40 (98%) 8.2 (5.9) 5.5 (5.4) 3.6 (0.4) 2.4 (0.8) 3.1 (0.5) 4.5 (0.4) 4.1 (0.6) 4.0 (0.6) 7.5 (1.0) 57 (17) 55 (27) 62 (15) 63 (17) 46 (20) 54 (26) 53 (30) 59 (25)
42 (100%) 8.6 (6.3) 5.6 (5.1) 3.7 (0.4) 2.6 (0.9) 3.3 (0.6) 4.4 (0.4) 4.3 (0.5) 3.8 (0.6) 7.9 (1.1) 60 (14) 65 (23) 63 (18) 64 (18) 53 (20) 55 (22) 66 (30) 64 (24)
p-Valuea
0.662 0.494 0.697 0.904 0.682 0.393 0.281 0.352 0.129 0.160 0.064 0.306 0.102 0.796 0.752 0.146 0.849 0.059 0.450
Note: For PHQ (depression) and GAD (anxiety) measures, higher scores indicate poorer functioning versus all other measures in which higher scores indicate better functioning. a All statistical significance values by ANOVA, Mann-Whitney, or Fisher’s exact test as appropriate.
program completion (p = 0.020); whereas anxiety did not change. At 6 months, the ESMS total (p = 0.004), the ESMS Information Management subscale (p = 0.009), the QOLIE-31 Energy/Fatigue Subscale (p = 0.032), and the QOLIE-31 Medication Side Effects subscale (p = 0.005) remained significantly improved relative to controls. Participant engagement From a number of perspectives, the participants in both the treatment and control groups valued the program and perceived it to be important. During the intervention phase, there were only three dropouts in the intervention group and Epilepsia, 56(8):1264–1274, 2015 doi: 10.1111/epi.13052
two in the control group, seven and five percent, respectively. In addition, all intervention participants completed all eight modules. In a few cases, a participant had to miss a meeting due to a schedule conflict, but then came early the next week to catch up on the module prior to the following group meeting. In terms of participant satisfaction with the program, all modules had average ratings above 4.0 on a five-point Likert scale (1–5), with “5” indicating excellent levels of satisfaction. The two highest rated modules were Epilepsy and Assertive Communication (4.3) and Dealing with Cognitive Issues (also 4.3). Notably, Epilepsy and Assertive Communication involved each participant script-
1271 PACES in Epilepsy RCT Table 3. Treatment effects Scoresa Treatment mean (SD) 8-week assessment (N) PHQ (0–27) GAD (0–21) ESMS (1–5) IM - Information mgmt. LM - Lifestyle mgmt. MM - Medication mgmt. SE - Seizure mgmt. SM - Safety mgmt. ESES (0–10) QOLIE-31 (0–100) SW - Seizure worry OQ - Overall QOL EW - Emotional well-being EF - Energy/fatigue CO - Cognitive ME - Medication effects SF - Social function 6-month assessment (N) PHQ (0–27) GAD (0–21) ESMS (1–5) IM - Information mgmt. LM - Lifestyle mgmt. MM - Medication mgmt. SE - Seizure mgmt. SM - Safety mgmt. ESES (0–10) QOLIE-31 (0–100) SW - Seizure worry OQ - Overall QOL EW - Emotional well-being EF - Energy/fatigue CO - Cognitive ME - Medication effects SF - Social function
38 6.3 (5.5) 5.4 (6.6) 3.9 (0.4) 3.0 (0.8) 3.5 (0.5) 4.6 (0.4) 4.4 (0.5) 4.1 (0.5) 8.1 (0.8) 63 (17) 60 (29) 68 (18) 68 (16) 55 (18) 59 (21) 55 (26) 68 (26) 37 6.5 (6.8) 5.2 (6.1) 3.9 (0.3) 3.0 (0.9) 3.4 (0.5) 4.6 (0.3) 4.3 (0.4) 4.1 (0.5) 8.1 (1.1) 61 (18) 60 (28) 65 (20) 67 (21) 53 (20) 58 (21) 65 (25) 63 (26)
Change From Baselineb Control mean (SD) 40 8.6 (6.0) 6.1 (5.1) 3.7 (0.4) 2.6 (0.7) 3.3 (0.6) 4.4 (0.4) 4.3 (0.5) 3.9 (0.6) 7.7 (1.2) 58 (15) 64 (27) 61 (18) 60 (16) 48 (20) 52 (21) 56 (34) 66 (25) 39 7.4 (5.4) 4.7 (4.3) 3.7 (0.4) 2.7 (0.8) 3.3 (0.6) 4.3 (0.6) 4.4 (0.5) 3.8 (0.7) 8.0 (1.1) 62 (15) 72 (26) 66 (18) 65 (18) 51 (21) 56 (22) 59 (34) 69 (22)
Tx effect
95% CI
p-Value
Favors
1.72 0.96 0.25 0.50 0.28 0.15 0.22 0.08 0.59 7.0 5.9 9.8 8.7 11.1 4.8 5.6 4.7
( 3.16, 0.28) ( 2.72, 0.81) (0.14, 0.35) (0.27, 0.74) (0.10, 0.45) (0.03, 0.27) (0.01, 0.43) ( 0.12, 0.28) (0.27, 0.91) (2.7, 11.3) ( 0.5, 12.3) (3.5, 16.1) (3.8, 13.5) (4.0, 18.2) ( 1.0, 10.5) ( 3.8, 15.0) ( 3.4, 12.9)
0.020 0.282
PACES in epilepsy: Results of a self-management randomized controlled trial *†‡1Robert T. Fraser, †1Erica K. Johnson, ‡Steven Lashley, §Jason Barber, *Naomi Chaytor, ¶John W. Miller, **Paul Ciechanowski, §††Nancy Temkin, and ‡‡Lisa Caylor Epilepsia, 56(8):1264–1274, 2015 doi: 10.1111/epi.13052
SUMMARY
Dr. Robert T. Fraser, professor/director NVSU; UW Rehabilitation Medicine/ Epilepsy Center.
Dr. Erica Johnson is a rehabilitation psychologist and researcher at the University of Washington.
Objective: Self-management challenges facing adults with epilepsy include limited understanding of the condition and treatment, associated psychosocial issues, and lack of community integration. Self-management interventions improve patients’ medical, life role, and emotional management. Previous interventions, developed from expert opinion, indicated issues with participant engagement/retention, and limited follow-up periods. PACES in Epilepsy addressed methodologic concerns by utilizing patient needs assessment data (n = 165) to derive self-management content and program features for evaluation via randomized controlled trial (RCT). Methods: Participants were adults with chronic epilepsy (n = 83), without serious mental illness or substantive intellectual impairment, who were recruited from two epilepsy centers. Participants were assigned randomly to intervention or treatmentas-usual groups. Outcomes included the Epilepsy Self-Management Scale (ESMS), Epilepsy Self-Efficacy Scale (ESES), Quality of Life in Epilepsy-31 (QOLIE-31), Patient Health Questionnaire-9 (PHQ-9), and the Generalized Anxiety Disorder-7 (GAD-7), administered at baseline, postintervention (8 weeks), and 6 months postintervention. The intervention was an 8-week group of 6–8 adults co-led by a psychologist and trained peer with epilepsy that met one evening per week at a hospital for 75 min. Topics included medical, psychosocial, cognitive, and self-management aspects of epilepsy, in addition to community integration and optimizing epilepsy-related communication. The treatment group provided satisfaction ratings regarding program features. Results: PACES participants (n = 38) improved relative to controls (n = 40) on the ESMS (p < 0.001) and subscales [Information (p < 0.001); Lifestyle (p < 0.002)]; ESES (p < 0.001); and QOLIE-31 (p = 0.002). At 6-month follow up, PACES participants remained improved on the ESMS (p = 0.004) and Information subscale (p = 0.009); and Energy/Fatigue (p = 0.032) and Medication Effects (p = 0.005) of the QOLIE-31. Attrition in both groups was low (8% in each group) and all program satisfaction ratings exceeded 4.0/5.0, with leadership (4.76), topics (4.53), and location (4.30) as the most highly rated aspects. Significance: A consumer generated epilepsy self-management program appears to be a promising intervention from multiple perspectives, particularly in relation to disability management. KEY WORDS: Epilepsy, Self-management, Self-efficacy, Depression, Quality of life, Consumer-generated intervention.
In 1997, the Centers for Disease Control and Prevention (CDC) began crafting a public health agenda culminating in the 2003 Living Well with Epilepsy II Conference, sponsored by the Epilepsy Foundation and other epilepsy organizations. A number of priority recommendations from this conference related to self-man-
agement in epilepsy as a path to disease burden reduction. In 2007, the CDC supported the development of the Prevention Research Center’s Managing Epilepsy Well Network (MEWN), which seeks to advance epilepsy self-management research and dissemination. The network focuses on site-specific epilepsy management
1264
1265 PACES in Epilepsy RCT
Key Points • • • •
The PACES epilepsy self-management program was evaluated by a randomized controlled trial involving 83 adults with chronic epilepsy. PACES in Epilepsy, an in-person group intervention, appears successful in improving patient self-management skills and aspects of quality of life. This consumer-generated program produced high satisfaction ratings accompanied by good goal attainment. Future research will evaluate distance delivery and sustention of PACES benefits with the addition of follow-up booster contact.
research as well as cooperative research and development activities.1 Much of the literature on self-management intervention is based on the work of Lorig and colleagues at the Stanford Patient Education Research Center.2,3 Within that literature, emphasis is on self-efficacy and problem-solving, with three primary functions: (1) medical management, such as taking medication, following a special diet, or using medical devices; (2) role management, which refers to routine, changing, and creating new meaningful behaviors and life roles (e.g., adopting new recreational activities, changing responsibilities in household, and so on); and (3) emotional management, such as adjusting to a situation that has induced emotions including anger, fear, sadness, depression, or grief.4–6 A number of epilepsy-specific educational programs have been described in the literature, but a 2007 Cochrane review reported that only two were tested in randomized control trial contexts.7 Helgeson examined the efficacy of a psychoeducational treatment program for adults with epilepsy, the Sepulveda Epilepsy Education (SEE) Program.8 The SEE program involved 16 sessions with 20 participants Accepted May 5, 2015; Early View publication June 29, 2015. *Rehabilitation Medicine, University of Washington, Seattle, Washington, U.S.A.; †Health Promotion Research Center, University of Washington, Seattle, Washington, U.S.A.; ‡Neurology Vocational Services Unit, University of Washington, Seattle, Washington, U.S.A.; §Neurological Surgery, University of Washington, Seattle, Washington, U.S.A.; ¶Neurology, University of Washington, Seattle, Washington, U.S.A.; **Psychiatry and Behavioral Sciences, University of Washington, Seattle, Washington, U.S.A.; ††Biostatistics, University of Washington, Seattle, Washington, U.S.A.; and ‡‡Swedish Neuroscience Institute, Seattle, Washington, U.S.A. 1 These authors contributed equally to this article. Address correspondence to Robert T. Fraser, Neurology Vocational Services Unit, University of Washington-Harborview Medical Center, Box 359744, 325 Ninth Avenue, Seattle, WA 98104, U.S.A. E-mail: [email protected] Wiley Periodicals, Inc. © 2015 International League Against Epilepsy
over 2 days, covering both medical and social-emotional aspects of managing one’s epilepsy. The SEE Program had some significant effects on certain medical concerns such as fear of death or brain damage and extent of overall misinformation and misconception of the disability at 4 months post-program. The authors noted a nonsignificant trend toward reduced seizure frequency and improved compliance. No significant changes were noted on measures of psychosocial adjustment and mood. The second study tested the Modular Service Package for Epilepsy (MOSES), a European program with nine modules presented over 2 days to groups of 7 to 10 adults with epilepsy.9 Again, significant improvements in epilepsy knowledge, coping with epilepsy, information seeking activity, and adherence to antiepileptic drug treatment were seen at 6-month follow-up. Significant reduction in seizure frequency was also noted for the intervention compared to the control group. There were no significant differences on measures of activities of daily living, mobility/leisure behavior, self-esteem, mood, and other psychosocial adjustment measures. The review authors noted some concerns about these studies, such as significant baseline differences between groups and higher proportions of people with partial seizures than would be expected in a community sample. Several studies have been completed within the MEWN related to self-management interventions. DiIorio et al.10 developed an Internet-based intervention, WebEase, to address the important role of social learning, behavior theory, and practical issues such as patient needs and transportation barriers. This program addressed three behavioral areas: medication management, stress, and sleep management. This project included an expanded set of outcome measures relative to prior investigations (e.g., epilepsy self-management and knowledge, medication self-adherence, sleep quality, level of stress, quality of life, and epilepsy selfefficacy).11–14 Those completing at least part of the three WebEase modules reported significantly more medication adherence and epilepsy self-efficacy. Dropout rate as related to program completion, however, was substantial and final outcome was assessed within a 12-week period. Another Emory study, Project UPLIFT, evaluated an 8week cognitive-behavioral mindfulness program for preventing depression in people with epilepsy.15 Upon conclusion, the incidence of major depression was significantly lower in the intervention compared to the control group, as were the number of depressive symptoms over the 10-week follow-up. There was no difference in intervention effectiveness between Web-based or telephone module delivery. The field of epilepsy self-management research is still evolving, but a number of observations can be made: 1 The range of medical and psychosocial adjustment needs addressed in these studies is often limited to medical needs, and a few other related issues. There is need for more encompassing interventions focusing not only on Epilepsia, 56(8):1264–1274, 2015 doi: 10.1111/epi.13052
1266 R. T. Fraser et al. medical barriers, but also a range of psychosocial adjustment areas. 2 These programs are generally planned by professionals, although focus groups can be involved. There is a need for more consumer-driven program development, consistent with community-based participatory research that underlies the MEWN. 3 There has been scant reporting as to patient satisfaction with these self-management programs as to content, mode of delivery, and so on. 4 To date, these programs can have a relatively short-term evaluation period. There is a need to study the durability of benefits from epilepsy self-management program over longer periods.
individual or group meetings with medical, epilepsy professional, and peer leadership were preferred by patients of all levels of functioning. Participants initially recommended six to eight 60- to 90-min sessions on a week night. The desired module formula was to emphasize education and emotional coping strategies. In contrast to other studies, the desired location was at a medical institution.
Development of the PACES Program
Site characteristics The University medical facility is located centrally in the Seattle-Metropolitan area and served by numerous bus lines. The intervention was delivered on-site in a conference room, three blocks from the UW Regional Epilepsy Center and within a mile of the Swedish epilepsy clinic. The study interventionists were two rotating Ph.D.-level rehabilitation psychologists and a trained peer mentor with epilepsy who has a bachelor’s degree and 7 years of experience as a neurologic employment specialist.
The first objective of the current study was to involve a diverse group of adults with epilepsy in the planning of a self-management intervention. A comprehensive review of research on medical and psychosocial concerns in epilepsy and needs assessment surveys used by epilepsy researchers and Epilepsy Foundation affiliates over the last 35 years was conducted to develop a survey on eight domains of concern: seizure activity, medical care, emotional functioning, employment, general health and well-being, independent living, and social life.16 A second section of the survey dealt with self-management program delivery preferences such as program emphasis and design, meeting times during the week, and program length. Information on epilepsy characteristics, demographic information, mood measures, and self-reported cognitive concerns was also collected. The survey was then refined by two focus groups of 10 adults with epilepsy, each at the University of Washington-Harborview Medical Center. The survey was then mailed to 250 adults with epilepsy who were being treated at either the UW-Harborview Medical Center or the Swedish Medical Center, and a group of 22 adults with epilepsy in local affiliate support groups. The survey participants in the mailing were screened for marked cognitive or emotional impairment by each center’s epileptologists. The Dillman approach was used in order to optimize participant response.17 A total of 165 returns were received (61% response rate). There was very high response reliability (internal consistency) across all domains in the survey. A comprehensive description of the survey findings is available in Fraser et al., with implications for the development of the PACES program.18 Of note, the survey findings revealed a significant subgroup with depression (PHQ-9 score ≥10) and/or three or more self-reported cognitive problems of moderate severity. Because of this, the PACES material was developed with emphasis on the range of mood and cognitive concerns across all modules. As to program logistics, face-to-face Epilepsia, 56(8):1264–1274, 2015 doi: 10.1111/epi.13052
Methods The protocol for the study was reviewed and approved by the UW and the Swedish Medical Center Institutional Review Boards. All participants provided informed consent.
Participants Participants were recruited through the UW Regional Epilepsy Center and Swedish Epilepsy Center, both in Seattle. Inclusion criteria were age 18 and older; able to speak, read, and write English; a medically established diagnosis of epilepsy at least 6 months from the initial diagnosis; and reasonably cognitively intact as measured by the Montreal Cognitive Assessment (MoCA).19 We implemented a lower cutoff score of 21 due to the cognitive impairment often experienced with epilepsy.20–22 Exclusion criteria were active severe mental illness/psychosis, or known cognitive impairment/IQ 100,000 Unknown Health insurance status Has health insurance
Overall 83
Treatment 41
Control 42
45.2 (12.5) 46 (55%)
44.9 (12.5) 23 (56%)
45.4 (12.6) 23 (55%)
0.863 1.000
67 (81%) 7 (8%) 9 (11%) 3 (4%) 2 (2%)
31 (76%) 5 (12%) 6 (15%) 2 (5%) 0 (0%)
36 (86%) 2 (5%) 3 (7%) 1 (2%) 2 (5%)
0.277 0.265 0.313 0.616 0.494
15 (18%) 43 (52%) 16 (19%) 34 (41%) 5 (6%) 10 (12%) 4 (5%) 0 (0%)
8 (20%) 20 (49%) 9 (22%) 14 (34%) 2 (5%) 6 (15%) 2 (5%) 0 (0%)
7 (17%) 23 (55%) 7 (17%) 20 (48%) 3 (7%) 4 (10%) 2 (5%) 0 (0%)
0.782 0.663 0.588 0.266 1.000 0.520 1.000
1 (1.2) 19 (23%) 25 (30%) 21 (25%) 10 (12%) 8 (10%)
2 (1.3) 9 (22%) 11 (27%) 11 (27%) 7 (17%) 3 (7%)
1 (1.2) 10 (24%) 14 (33%) 10 (24%) 3 (7%) 5 (12%)
0.593 0.658
15.0 (2.5) 2 (2%) 12 (14%) 14 (17%) 13 (16%) 28 (34%) 9 (11%) 5 (6%)
15.1 (2.4) 1 (2%) 5 (12%) 5 (12%) 8 (20%) 14 (34%) 6 (15%) 2 (5%)
14.8 (2.6) 1 (2%) 7 (17%) 9 (21%) 5 (12%) 14 (33%) 3 (7%) 3 (7%)
0.544 0.801
21 (26%) 12 (15%) 27 (33%) 22 (27%)
9 (22%) 5 (12%) 15 (37%) 12 (29%)
12 (29%) 7 (17%) 12 (29%) 10 (24%)
0.756
44 (53%) 30 (36%) 8 (10%) 1 (1%)
18 (44%) 19 (46%) 4 (10%) 0 (0%)
26 (62%) 11 (26%) 4 (10%) 1 (2%)
0.159
57.1 (61.4) 16 (29%) 11 (20%) 8 (15%) 9 (16%) 11 (20%) 29
59.4 (65.3) 10 (36%) 4 (14%) 3 (11%) 4 (14%) 7 (25%) 14
54.7 (58.2) 6 (22%) 7 (26%) 5 (19%) 5 (19%) 4 (15%) 15
0.742 0.572
70 (85%)
35 (88%)
35 (83%)
0.757
p-Valuea
a
All statistical significance values by ANOVA, Mann-Whitney, or Fisher’s exact test as appropriate.
Management (p < 0.001) and Life Style Management subscales (p = 0.002). The ESES (p < 0.001) and QOLIE-31 (p = 0.002) scores were also significant. In addition to the
total QOLIE-31 score, subscales of overall QOL, Emotional Well-Being, and Energy/Fatigue were also significant (all p < 0.01). Depression was also significantly reduced at Epilepsia, 56(8):1264–1274, 2015 doi: 10.1111/epi.13052
1270 R. T. Fraser et al. Table 2. Baseline assessment N Overall health Mean (SD) 1 - Excellent 2 - Very good 3 - Good 4 - Fair 5 - Poor Life satisfaction rating Mean (SD) 1 - Very satisfied 2 - Satisfied 3 - Dissatisfied 4 - Very dissatisfied Happiness rating Mean (SD) 1 - Very happy 2 - Happy 3 - Neutral 4 - Not very happy 5 - Not happy at all MoCA mean (SD) Baseline assessment (N) Subjects PHQ (0–27) GAD (0–21) ESMS (1–5) IM - Information mgmt. LM - Lifestyle mgmt. MM - Medication mgmt. SE - Seizure mgmt. SM - Safety mgmt. ESES (0–10) QOLIE-31 (0–100) SW - Seizure worry OQ - Overall QOL EW - Emotional well-being EF - Energy/fatigue CO – Cognitive ME - Medication effects SF - Social function
Overall 83
Treatment 41
Control 42
2.8 (1.1) 12 (14%) 21 (25%) 31 (37%) 14 (17%) 5 (6%)
3.0 (1.0) 1 (2%) 14 (34%) 14 (34%) 9 (22%) 3 (7%)
2.5 (1.2) 11 (26%) 7 (17%) 17 (40%) 5 (12%) 2 (5%)
0.059 0.010
2.3 (0.8) 11 (13%) 45 (54%) 20 (24%) 7 (8%)
2.4 (0.8) 3 (7%) 24 (59%) 9 (22%) 5 (12%)
2.2 (0.8) 8 (19%) 21 (50%) 11 (26%) 2 (5%)
0.206 0.282
2.6 (0.9) 8 (10%) 35 (42%) 28 (34%) 9 (11%) 3 (4%) 26.1 (2.7)
2.6 (0.9) 2 (5%) 19 (46%) 13 (32%) 6 (15%) 1 (2%) 26.3 (2.5)
2.5 (1.0) 6 (14%) 16 (38%) 15 (36%) 3 (7%) 2 (5%) 26.0 (2.9)
0.519 0.493
82 (99%) 8.4 (6.1) 5.6 (5.2) 3.6 (0.4) 2.5 (0.8) 3.2 (0.6) 4.4 (0.4) 4.2 (0.6) 3.9 (0.6) 7.7 (1.1) 58 (16) 60 (25) 63 (17) 63 (18) 50 (20) 54 (24) 60 (31) 61 (24)
40 (98%) 8.2 (5.9) 5.5 (5.4) 3.6 (0.4) 2.4 (0.8) 3.1 (0.5) 4.5 (0.4) 4.1 (0.6) 4.0 (0.6) 7.5 (1.0) 57 (17) 55 (27) 62 (15) 63 (17) 46 (20) 54 (26) 53 (30) 59 (25)
42 (100%) 8.6 (6.3) 5.6 (5.1) 3.7 (0.4) 2.6 (0.9) 3.3 (0.6) 4.4 (0.4) 4.3 (0.5) 3.8 (0.6) 7.9 (1.1) 60 (14) 65 (23) 63 (18) 64 (18) 53 (20) 55 (22) 66 (30) 64 (24)
p-Valuea
0.662 0.494 0.697 0.904 0.682 0.393 0.281 0.352 0.129 0.160 0.064 0.306 0.102 0.796 0.752 0.146 0.849 0.059 0.450
Note: For PHQ (depression) and GAD (anxiety) measures, higher scores indicate poorer functioning versus all other measures in which higher scores indicate better functioning. a All statistical significance values by ANOVA, Mann-Whitney, or Fisher’s exact test as appropriate.
program completion (p = 0.020); whereas anxiety did not change. At 6 months, the ESMS total (p = 0.004), the ESMS Information Management subscale (p = 0.009), the QOLIE-31 Energy/Fatigue Subscale (p = 0.032), and the QOLIE-31 Medication Side Effects subscale (p = 0.005) remained significantly improved relative to controls. Participant engagement From a number of perspectives, the participants in both the treatment and control groups valued the program and perceived it to be important. During the intervention phase, there were only three dropouts in the intervention group and Epilepsia, 56(8):1264–1274, 2015 doi: 10.1111/epi.13052
two in the control group, seven and five percent, respectively. In addition, all intervention participants completed all eight modules. In a few cases, a participant had to miss a meeting due to a schedule conflict, but then came early the next week to catch up on the module prior to the following group meeting. In terms of participant satisfaction with the program, all modules had average ratings above 4.0 on a five-point Likert scale (1–5), with “5” indicating excellent levels of satisfaction. The two highest rated modules were Epilepsy and Assertive Communication (4.3) and Dealing with Cognitive Issues (also 4.3). Notably, Epilepsy and Assertive Communication involved each participant script-
1271 PACES in Epilepsy RCT Table 3. Treatment effects Scoresa Treatment mean (SD) 8-week assessment (N) PHQ (0–27) GAD (0–21) ESMS (1–5) IM - Information mgmt. LM - Lifestyle mgmt. MM - Medication mgmt. SE - Seizure mgmt. SM - Safety mgmt. ESES (0–10) QOLIE-31 (0–100) SW - Seizure worry OQ - Overall QOL EW - Emotional well-being EF - Energy/fatigue CO - Cognitive ME - Medication effects SF - Social function 6-month assessment (N) PHQ (0–27) GAD (0–21) ESMS (1–5) IM - Information mgmt. LM - Lifestyle mgmt. MM - Medication mgmt. SE - Seizure mgmt. SM - Safety mgmt. ESES (0–10) QOLIE-31 (0–100) SW - Seizure worry OQ - Overall QOL EW - Emotional well-being EF - Energy/fatigue CO - Cognitive ME - Medication effects SF - Social function
38 6.3 (5.5) 5.4 (6.6) 3.9 (0.4) 3.0 (0.8) 3.5 (0.5) 4.6 (0.4) 4.4 (0.5) 4.1 (0.5) 8.1 (0.8) 63 (17) 60 (29) 68 (18) 68 (16) 55 (18) 59 (21) 55 (26) 68 (26) 37 6.5 (6.8) 5.2 (6.1) 3.9 (0.3) 3.0 (0.9) 3.4 (0.5) 4.6 (0.3) 4.3 (0.4) 4.1 (0.5) 8.1 (1.1) 61 (18) 60 (28) 65 (20) 67 (21) 53 (20) 58 (21) 65 (25) 63 (26)
Change From Baselineb Control mean (SD) 40 8.6 (6.0) 6.1 (5.1) 3.7 (0.4) 2.6 (0.7) 3.3 (0.6) 4.4 (0.4) 4.3 (0.5) 3.9 (0.6) 7.7 (1.2) 58 (15) 64 (27) 61 (18) 60 (16) 48 (20) 52 (21) 56 (34) 66 (25) 39 7.4 (5.4) 4.7 (4.3) 3.7 (0.4) 2.7 (0.8) 3.3 (0.6) 4.3 (0.6) 4.4 (0.5) 3.8 (0.7) 8.0 (1.1) 62 (15) 72 (26) 66 (18) 65 (18) 51 (21) 56 (22) 59 (34) 69 (22)
Tx effect
95% CI
p-Value
Favors
1.72 0.96 0.25 0.50 0.28 0.15 0.22 0.08 0.59 7.0 5.9 9.8 8.7 11.1 4.8 5.6 4.7
( 3.16, 0.28) ( 2.72, 0.81) (0.14, 0.35) (0.27, 0.74) (0.10, 0.45) (0.03, 0.27) (0.01, 0.43) ( 0.12, 0.28) (0.27, 0.91) (2.7, 11.3) ( 0.5, 12.3) (3.5, 16.1) (3.8, 13.5) (4.0, 18.2) ( 1.0, 10.5) ( 3.8, 15.0) ( 3.4, 12.9)
0.020 0.282
