Blended psychosocial support for partners of patients with ALS and PMA: results of a randomized controlled trial

Abstract

Objective: To evaluate whether a blended (face-to-face and online) psychosocial support program for caregivers of patients with Amyotrophic Lateral Sclerosis (ALS) and Progressive Muscular Atrophy (PMA), aimed at enhancing feeling of control over caregiving, reduces psychological distress. Methods: A randomized controlled trial using a wait-list control design was conducted. Caregiver-patient dyads were randomly assigned to either the support program (n = 74) or to a wait-list control group (n = 74). The support program, based on Acceptance and Commitment Therapy, consists of 1 face-to-face contact, 6 online guided modules and 1 telephone contact. Participants filled in questionnaires at baseline, 3 and 6 months. Caregivers’ feeling of control over caregiving was assessed using two self-efficacy measures. Primary outcome was caregivers’ psychological distress. Secondary outcomes included caregiver burden, caregiver quality of life and patients’ quality of life and psychological distress. Intention-to-treat and per-protocol analyses were performed using linear mixed models. Results: The support program had no effect on the primary or secondary outcomes, despite a significant positive effect on the intervening variable self-efficacy with regard to control over thoughts. Almost half of the caregivers did not complete the intervention with the most frequently reported reason being lack of time. Caregivers who completed the intervention evaluated the support program positively. Conclusions: The support program did not reduce distress of partners of patients with ALS/PMA but may be beneficial by increasing feeling of control over the caregiving situation. The high level of intervention drop outs may have limited the ability to detect an intervention effect.

Trial registration: Netherlands Trial Registry NTR5734, registered 28 March 2016.

Keywords:

• Caregivers
• Amyotrophic Lateral Sclerosis (ALS)
• progressive muscular atrophy
• psychosocial support
• psychological distress

Introduction

Informal caregivers play a crucial role in the care for patients with Amyotrophic Lateral Sclerosis (ALS) and Progressive Muscular Atrophy (PMA). ALS and PMA are progressive and fatal diseases in which the physical functioning of patients deteriorates and cognitive and behavioral changes can arise, resulting in increasing care needs over time (1). For both patients and their caregivers this might result in feelings of helplessness (2), subsequently making it difficult to experience a feeling of control over their situation (3). Experiencing high care demands and insufficient control over the caregiving situation increases the risk of psychological distress in caregivers (4). In caregivers of patients with ALS distress increases over time but psychosocial evidence-based interventions are lacking (5).

Although caregivers express a need for individual psychosocial support (6), they frequently withdraw from seeking or accepting support (7). They report difficulty in balancing their personal time with caregiving responsibilities (8). Blended care combines face-to-face healthcare with online healthcare and could bridge the gap between the need for support and lack of time (9). Acceptance and Commitment Therapy (ACT) has proven to be effective in reducing psychological distress in caregivers of other patient populations (10,11).

This study evaluates the effectiveness of a blended psychosocial support program, based on ACT, aimed at enhancing feelings of control over caregiving, in order to reduce psychological distress in caregivers of patients with ALS and PMA. The indirect effect of the support program on patients’ wellbeing and caregivers’ adherence and satisfaction with the program were also studied.

Methods

Study design and procedures

The study protocol has been described in detail by de Wit et al. (2018) (12). In this randomized controlled trial (RCT), caregiver-patient dyads were randomly allocated to the support program or to a 6-months wait-list control group, in addition to usual care. Care as usual is provided by multidisciplinary ALS care teams and is mainly focused on patients, but partners who experience problems are able to receive support from a social worker or psychologist.

Both caregivers and patients filled in online questionnaires at baseline (T0), after the intervention (3 months after baseline, T1) and 3 months after finishing the intervention (T2). Questionnaires were completed at home in the absence of people who were involved in the study. Patients could provide consent for the caregiver to participate, with or without own study participation. Patients’ consent was needed since caregivers filled in questionnaires about patients’ functioning. After caregivers and patients provided informed consent and caregivers completed the baseline measurement, dyads were randomized.

Randomization in a 1:1 ratio was stratified by gender, functional impairment (ALSFRS-R; cut off score for severe disabilities ≥ 24) and behavioral changes of the patient (ALS-FTD-Q; cut off score for mild behavioral changes ≥ 22) and was conducted by the researcher (JW) using a computer-generated randomization sequence, allowing concealment for the next allocation. Due to the nature of the intervention, blinding of participants was not possible.

The study was approved by the Medical Ethics Committee of the University Medical Center Utrecht (16-273-D) The trial was registered at the Netherlands National Trial Register (NTR5734). The study is reported in accordance with the CONSORT guidelines (13).

Participants

Participants were recruited through a Dutch national ALS/PMA database and via the Dutch ALS Center website. Patients in the database were treated by ALS care teams throughout the country. Eligible dyads met the following criteria: 1) the caregiver is the partner of the ALS or PMA patient; 2) the caregiver is 18 years or older; 3) caregiver and patient are proficient in Dutch; 4) caregiver and patient have internet access.

Intervention

The blended psychosocial support program consisted of 1 face-to-face contact, 6 online guided modules and 1 closing telephone contact and was guided by a psychologist (see Supplementary Table 5). The program was solely offered to the caregiver and could be worked through in 8–12 weeks. Modules were offered in a fixed sequence in which participants received access to the next module after 1 or 2 weeks. Caregivers were informed that completing each module would take approximately 1 hour and 30 minutes. The content of the support program was focused on ACT and based on an online intervention aimed at partners of patients with cancer (14,15). We adapted this program to the specific needs of caregivers of patients with ALS and PMA. A total of 21 ALS caregivers were interviewed about their support needs (6) and information and exercises were added to the program in line with these needs. Participants started the program with a face-to-face session with a psychologist who briefly explained and demonstrated the program and retrieved information about the situation. This one-hour face-to-face session was held at the residence of the caregiver. After this contact, the caregiver started with the first online module. Every module was focused on a specific theme (e.g. coping with emotions and thoughts, communication) and contained psychoeducation, psychological exercises, mindfulness exercises and practical information and references to relevant websites. After completing each module, participants received feedback on the exercises from the psychologist. Participants could get in contact with other participants through private messaging and they could share advices using a forum. The support program was closed with a telephone call, in which the caregiver could ask for advice. The closing telephone calls lasted about 30 minutes. Participants were guided through the program by two psychologists who were trained to provide the support program and personalized feedback, and who could be contacted at any moment during the program. The psychologists were not involved in the care as usual of the patient or caregivers.

Outcomes

Primary outcome

The primary outcome of this study was psychological distress of the caregiver, assessed with the Hospital Anxiety and Depression Scale (HADS; range 0–42) (16).

Secondary outcomes

The secondary outcome measures of this study included:

1. Caregiver burden, assessed with the Zarit Burden Interview (ZBI; range 0–48) (17).

2. Caregivers’ quality of life, measured with the Care Related- Quality of Life (CarerQoL; range 7–21 + CarerQol Vas; range 0–10) (18).

3. Patients’ quality of life, measured using the single item scale of the McGill Quality of Life Questionnaire (MQOL; range 0–10) (19).

4. Patients’ distress, assessed with the HADS (16).

Intervening variable

The intervening variable in this study was feeling of control over caregiving which was measured as caregivers’ self-efficacy over caregiving:

1. Caregivers’ beliefs about their capacity to obtain respite from caregiving was measured using the Respite care subscale of the Revised Scale for Caregiving Self-Efficacy (RSCSE-Resp) (per subscale, range 0–100) (20).

2. Caregivers’ beliefs about their ability to control distressing thoughts about caregiving was measured using the ‘Controlling upsetting thoughts about caregiving’ subscale of the RSCSE (RSCSE-Contr).

3. Perceived control over fulfilling caregiver tasks was measured with three questions adapted from the Job Content Questionnaire (JCQ) (21) (range 0–9).

Covariates

The analyses of the caregiver outcomes were adjusted for the baseline scores of:

1. Caregivers’ social support, measured with the Multidimensional Scale of Perceived Social Support (MPSS; range 12–84) (22).

2. Caregivers’ satisfaction with the relationship with the patient, assessed using the Satisfaction Scale (range 4–20) (23).

3. Patients’ physical functioning, assessed using the Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised (ALSFRS-R; range 0–48) (24).

4. Patients’ behavioral changes (e.g. apathy, eating disturbances, increased irritability), assessed with the Amyotrophic Lateral Sclerosis-Fronto Temporal Dementia-Questionnaire (ALS-FTD-Q; range 0–100) (25).

The ALSFRS-R and ALS-FTD-Q were completed by caregivers.

The analyses on the patient outcomes were adjusted for the baseline score of the ALSFRS-R and for the baseline score of: 5. Patient’s own feelings of being a burden for the caregiver, measured using one item of the Self-Perceived Burden Scale (SPBS; range 1–5) (26).

Satisfaction with support

Satisfaction with the support program was assessed with the Client Satisfaction Questionnaire (CSQ-8; range 8–32) (27). Additionally, the participant was asked to rate the intervention in general and the different components of the intervention (i.e. the information, exercises, mindfulness, contact with professional) on a scale from 0–10. This scale contained 9 questions including a question about suggestions for further development. All caregivers, including caregivers who prematurely stopped with the support program, were asked to fill in the satisfaction questionnaires.

Adherence

Adherence to the intervention and time spent in the program were measured by log files. Participants who completed four modules were considered completers.

Demographics

Demographics and information about the care situation were administered using The iMTA Valuation of Informal Care Questionnaire (iVICQ) (28).

Statistical analyses

Sample size calculation

A sample size of 58 per group was needed to detect a clinically relevant difference of 3.65 points (SD=7.3) on the HADS (17) between the groups at T2, with an alpha of 0.05 and a power of 80%. Anticipating a drop-out of 20% between baseline and T2, we aimed to include a total of 140 caregiver-patient dyads.

Efficacy analyses

Primary and secondary outcomes were evaluated six months follow- up (T2). First, data were analyzed on an intention-to-treat principle; all caregivers with follow-up data were included whether or not they completed the intervention or the follow-up assessments. Linear mixed models (LMM) were used to investigate differences in the course of the outcome measures between the two groups. All analyses were performed using fixed effects for group, time and their interaction, as well as a random intercept per subject and a random slope for time. All caregiver models included relevant covariates: caregivers’ social support, caregivers’ satisfaction with their relationship, patients’ physical functioning and patients’ behavioral changes. Patient models were corrected for the baseline scores of patients’ self-perceived burden and patients’ physical functioning.

Additionally, per-protocol analyses were performed including only caregivers who attended four (66.7%) or more online modules. These secondary exploratory analyses were aimed to understand how the intervention works and to generate hypotheses about the mechanism of change, therefore, correcting for multiple testing was not applied. Missing data were accounted for by the statistical techniques used (LMM) (29). Analyses were performed with SPSS version 25.

Results

Participants

The recruitment took place from July 2017 to March 2018 and the trial was completed in February 2019. A total of 538 dyads with ALS/PMA were assessed for eligibility and 148 caregivers and 101 patients with ALS/PMA were included in the study (Figure 1 and Supplementary figure 2). After randomization, 74 caregivers started with the support program and 74 caregivers were on a wait-list for the support program. Characteristics of the caregivers and the patients for whom they provided care are shown in Table 1. The majority of caregivers was female (64.9%); their mean age was 61 years. There were no significant differences in demographic and clinical outcome measures between patients who participated in the study (n = 101) and patients who only provided consent for their partner to participate (n = 47).

Figure 1 Flowchart partners in study.

Table 1 Characteristics of the caregivers in the trial and the patients for whom they provide care (n = 148).

	Intervention (n = 74)	Wait-list (n = 74)
Caregiver characteristics
Gender, n (% female)	48 (64.9)	48 (64.9)
Age in years, mean (SD)	61.8 (10.6)	61.3 (9.8)
Education level, n (%)
Low	28 (37.8)	35 (47.3)
Medium	16 (21.6)	19 (25.7)
High	30 (40.5)	20 (27.0)
Occupation, n (%)
Not working	45 (60.8)	45 (60.8)
Paid work/employment	29 (39.2)	29 (39.2)
Country of birth, n (%)
The Netherlands	68 (91.9)	67 (90.5)
Other	3 (4.1)	2 (2.7)
Missing	3 (4.1)	5 (6.8)
Providing care tasks, n (%)
Yes	59 (79.7)	59 (79.7)
Hours providing care per week, median (inter quartile range)	35 (22–60)	38 (21–56)
Receiving support to deal with caregiving situation n (%)
Yes	19 (25.7)	37 (50.0)
Patient characteristics
Gender, n (% female)	26 (35.1)	27 (36.5)
Age in years, mean (SD)	62.3 (11.0)	62.9 (8.9)
Diagnosis, n (%)
ALS	52 (70.3)	54 (73.0)
PMA	22 (29.7)	20 (27.0)
Time since diagnosis in months, median (inter quartile range)	22.5 (8.0–34.3)	16.00(9.8–44.3)
Level of functioning, mean (SD)
ALSFRS-R	31.7 (9.8)	31.0 (9.5)
ALS-FTD-Q	16.3 (11.8)	17.2 (13.9)
Country of birth, n (%)
The Netherlands	72 (97.3)	65 (87.8)
Other	0 (0.0)	2 (2.7)
Missing	2 (2.7)	7 (9.5)

ALS: Amyotrophic Lateral Sclerosis; PMA: Progressive Muscular Atrophy; ALSFRS-R: Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised; ALS-FTD-Q: Amyotrophic Lateral Sclerosis-Frontotemporal Dementia-Questionnaire; Educational level: low: did not complete secondary school-completed low level secondary school; medium: completed medium level secondary school; high: completed upper level secondary school and/or university degree.

A total of 28 (19%) caregivers dropped out of the study (i.e. they indicated they would no longer participate in any part of the study, filling in questionnaires or the support program). The most common reason for drop-out was the death of the patient (n = 16). Drop-outs had partners with a significantly lower physical functioning score (ALSFRS-R; M = 26.9, SD = 9.0) compared to study completers (M = 32.4, SD = 9.4, p < 0.01). There were no further significant differences between drop-outs and study completers.

Support program adherence

Caregivers could stop with the support program, but still participate in the study by filling in questionnaires. The support program was completed by 38 caregivers (51%); 28 caregivers stopped with the intervention prematurely, and 8 dropped out of the study before completing the intervention. The main reasons reported for not completing the intervention were: a lack of time for receiving support (n = 9), no perceived value of the program (n = 4), the program was too confronting (n = 2) and illness of the caregiver (n = 2) (Figure 1). Completers and non-completers did not differ significantly as far as baseline caregiver and patient characteristics were concerned. Caregivers completed the program in an average of 8 weeks and spent a mean of 1 hour and 26 minutes per module (range 7 minutes–4 hours and 18 minutes).

Efficacy support program analyses

There were no significant effects on psychological distress, despite a significant difference between the intervention group and wait-list control group in caregivers’ self-efficacy with regard to controlling thoughts about caregiving (RSCSE- Contr; Table 2). Nor were there any significant differences in the other caregivers’ outcomes and intervening measures (burden, quality of life and the other self-efficacy scales) between the intervention group and the wait-list control group. Furthermore, no significant differences in psychological distress or quality of life of patients were found between the intervention group and the wait-list control group. The per-protocol analyses confirmed the results of the intention-to-treat analysis (Table 3).

Table 2 Linear mixed model treatment outcome Intervention group vs Wait-list control group.

	Wait-list control group			Intervention group
	Mean (SD)			Mean (SD)			Intervention * time interaction
Caregivers^a	Baseline (n = 74)	3 months (n = 60)	6 months (n = 60)	Baseline (n = 74)	3 months (n = 56)	6 months (n = 46)	b (se)	p-value
Primary outcome
HADS	10.84 (7.19)	11.37 (7.70)	11.48 (7.42)	11.38 (7.12)	10.75 (7.62)	11.00 (7.71)	0.09 (0.38)	0.802
Secondary outcomes
ZBI	13.53 (8.04)	13.62 (7.79)	14.20 (7.58)	13.50 (8.08)	13.18 (7.29)	13.89 (7.76)	−0.19 (0.71)	0.711
CarerQoL-7	10.14 (2.30)	10.10 (2.28)	9.97 (2.09)	10.11 (2.22)	10.63 (2.23)	10.09 (2.23)	−0.11 (0.17)	0.534
CarerQoL-VAS	6.41 (1.43)	6.50 (1.66)	6.25 (1.74)	6.43 (1.55)	6.63 (1.45)	6.43 (1.54)	−0.04 (0.13)	0.735
Intervening variables
RSCSE-Resp	73.43 (22.31)	69.10 (20.70)	69.63 (23.57)	67.81 (25.05)	72.07 (23.00)	67.30 (22.34)	−1.41 (1.93)	0.465
RSCSE-Contr	64.54 (25.36)	58.33 (24.32)	59.13 (27.16)	60.78 (25.87)	62.71 (26.01)	68.26 (23.04)	−5.39 (1.99)	0.007
JCQa	6.28 (1.85)	6.63 (2.02)	6.32 (2.06)	6.41 (1.89)	6.96 (1.88)	6.93 (1.97)	−0.21 (0.21)	0.325
Patients^b	Baseline (n = 43)	3 months (n = 33)	6 months (n = 31)	Baseline (n = 45)	3 months (n = 30)	6 months (n = 24)	b (se)	p-value
Secondary outcomes
HADS	10.42 (5.19)	9.82 (5.69)	11.87 (6.98)	9.42 (6.43)	8.00 (6.72)	9.33 (7.75)	0.20 (0.67)	0.767
MQOL	6.79 (1.41)	6.61 (1.85)	6.52 (2.10)	7.07 (1.34)	6.77 (1.94)	6.63 (1.86)	0.22 (0.21)	0.286

^aLinear mixed model including group, time, intervention*time and adjusted for baseline values of MPSS, SQ, ALSFRS-R, ALS-FTD-Q.

^bLinear mixed model including group, time, intervention*time and adjusted for baseline values of ALSFRS-R, SPBS.

ALSFRS-R: Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised; ALS-FTD-Q: Amyotrophic Lateral Sclerosis-Frontotemporal Dementia-Questionnaire; CarerQoL: Care-related Quality of Life; m: mean; HADS: Hospital Anxiety and Depression Scale; JCQa: Job Content Questionnaire 3 adapted items; MQOL: McGill Quality of Life Questionnaire; MPSS: Multidimensional Scale for Received Support; SD: standard deviation; RSCSE-Resp: Revised scale for caregiving self-efficacy subscale respite care; RSCSE-Contr: Revised scale for caregiving self-efficacy subscale controlling thoughts; SE: standard error; SPBS: Self-Perceived Burden Scale; SQ: Satisfaction Questionnaire; ZBI: Zarit Burden Interview.

A significant group-by-time interaction (p < 0.05) means a significant difference between groups in outcome over time.

Table 3 Linear mixed model treatment outcome Intervention completers vs Wait-list control group.

	Wait-list control group			Intervention completers
	Mean (SD)			Mean (SD)			Intervention * time interaction
Caregivers^a	Baseline (n = 74)	3 months (n = 60)	6 months (n = 60)	Baseline (n = 38)	3 months (n = 37)	6 months (n = 33)	b (se)	p-value
Primary outcome
HADS	10.84 (7.19)	11.37 (7.70)	11.48 (7.42)	9.50 (7.35)	8.22 (6.31)	9.30 (7.06)	0.47 (0.41)	0.256
Secondary outcomes
ZBI	13.53 (8.04)	13.62 (7.79)	14.20 (7.58)	13.13 (8.79)	11.92 (7.43)	12.79 (7.37)	0.50 (0.57)	0.376
CarerQoL-7	10.14 (2.30)	10.10 (2.28)	9.97 (2.09)	10.34 (2.41)	11.16 (1.85)	10.55 (2.09)	−0.24 (0.20)	0.235
CarerQoL-VAS	6.41 (1.43)	6.50 (1.66)	6.25 (1.74)	6.61 (1.46)	6.97 (1.21)	6.76 (1.25)	−0.16 (0.14)	0.262
Intervening variables
RSCSE-Resp	73.43 (22.31)	69.10 (20.70)	69.63 (23.57)	71.53 (24.97)	74.00 (19.22)	72.30 (18.32)	−2.65 (2.13)	0.214
RSCSE-Contr	64.54 (25.36)	58.33 (24.32)	59.13 (27.16)	67.53 (23.94)	68.81 (21.02)	71.39 (19.13)	−4.93 (2.23)	0.028
JCQa	6.28 (1.85)	6.63 (2.02)	6.32 (2.06)	7.00 (1.64)	7.19 (1.79)	6.82 (2.14)	0.11 (0.24)	0.661
Patients^b	Baseline (n = 43)	3 months (n = 33)	6 months (n = 31)	Baseline (n = 23)	3 months (n = 19)	6 months (n = 15)	b (se)	p-value
Secondary outcomes
HADS	10.42 (5.19)	9.82 (5.69)	11.87 (6.98)	9.43 (6.71)	8.58 (6.69)	9.73 (7.95)	0.24 (0.75)	0.747
MQOL	6.79 (1.41)	6.61 (1.85)	6.52 (2.10)	7.17 (1.11)	6.68 (2.03)	6.27 (2.02)	0.43 (0.25)	0.089

^aLinear mixed model including intervention, time, intervention*time and adjusted for baseline values of MPSS, SQ, ALSFRS-R, ALS-FTD-Q.

^bLinear mixed model including intervention, time, intervention*time and adjusted for baseline values of the outcome measure, SPBS, ALSFRS-R.

ALSFRS-R: Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised; ALS-FTD-Q: Amyotrophic Lateral Sclerosis-Frontotemporal Dementia-Questionnaire; CarerQoL: Care-related Quality of Life; m: mean; HADS: Hospital Anxiety and Depression Scale; JCQa: Job Content Questionnaire 3 adapted items; MPSS: Multidimensional Scale for Received Support; MQOL: McGill Quality of Life Questionnaire; SD: standard deviation; RSCSE-Resp: Revised scale for caregiving self-efficacy subscale respite care; RSCSE-Contr: Revised scale for caregiving self-efficacy subscale controlling thoughts; SE: standard error; SPBS: Self-Perceived Burden Scale; SQ: Satisfaction Questionnaire; ZBI: Zarit Burden Interview.

A significant group-by-time interaction (p < 0.05) means a significant difference between groups in outcome over time.

Satisfaction with the blended support program

Participants were satisfied with the interventions and scored a mean total of 25.57 (SD 3.8) on the CSQ-8. Participants who completed the program rated the overall program with a mean score of 7.9 (SD 1.2) and people who did not complete the program rated the program with a mean score of 6.9 (SD 1.4). Completers rated the program more positively on all components except the mindfulness.

With regard to the specific components of the program, caregivers were least satisfied with the peer contact and mindfulness exercises (Table 4).

Table 4 Evaluation intervention (n = 53).

Question	Completers (n = 37) Grade 0–10, M (SD)	Non Completers (n = 16) Grade 0–10, M (SD)
Does the support program match your situation?	6.4 (2.4)	5.6 (2.2)
How do you rate the support program in general?	7.9 (1.2)	6.9 (1.4)
How do you rate the information in the support program?	8.1 (1.3)	7.2 (1.3)
How do you rate the exercises in the support program?	7.4 (1.8)	6.9 (1.5)
How do you rate the mindfulness exercises?	5.5 (3.1)	6.1 (2.2)
How do you rate the contact with the professional who provided feedback?	8.6 (1.3)	8.0 (1.1)
How do you rate the user-friendliness of the program?	8.0 (1.2)	7.3 (2.2)
How do you rate the peer contact functions?	6.8 (2.3)	6.5 (2.5)

Discussion

The blended psychosocial support program was appreciated by caregivers but did not lead to reduced psychological distress, caregiver burden or quality of life in caregivers, nor to reduced quality of life or psychological distress in patients. A significant positive intervention effect was found on the intervening variable caregivers’ self-efficacy with regard to being in control over thoughts about caregiving. Almost half of the participants did not complete the intervention which may have contributed to the lack of efficacy of the intervention on the primary and secondary outcomes.

Although an adherence rate of 51% is not rare in online interventions (30), the percentage is considered to be relatively low. The most frequently reported reason for not completing the intervention was lack of time, a commonly reported drop-out reason in caregiver studies (31). Some of these caregivers expressed that they were so occupied by caregiving that any more obligations were experienced as burdensome. Caregivers of patients with ALS/PMA spend many hours a day on caregiving and the care needed increases over time (32); in our study, 80 percent of the partners provided care for the patient and they spent a median of 36.5 hours a week on caregiving tasks. For caregivers whose partner was in an advanced disease stage, investing time in an intervention was not beneficial since the caregiving would be coming to an end before long. Therefore, the timing of an intervention seems to be crucial. A support program can equip caregivers with tools to deal with the caregiver situation in the long term, which in turn can prevent high levels of distress (33) and prolong the stage during which caregivers feel able to support patients at home (34). Offering a support program to all caregivers, before they experience feelings of psychological distress is therefore recommended. But caregivers need to have enough time and mental space to follow the support program, otherwise the program is perceived as burdensome. This is in line with the dual process model which proposes that adaptive coping is a dynamic process of confronting and avoiding stressors associated with loss (35). Caregivers who are caught up in caregiving and who follow the support program may feel as though they have no respite from dealing with loss, which may have detrimental health effects (36). Therefore, proactively informing caregivers about possible supportive interventions at an early stage of the disease and offering this repeatedly, seems to be crucial (33).

The program resulted in an increase in perceived self-efficacy in caregivers. This is in line with our qualitative study in which caregivers expressed that the program helped them to perceive control over the caregiving situation, to accept negative emotions and thoughts, to be there for their partner and feel acknowledged (39). Three types of self-efficacy were measured in the current study (self-efficacy with regard to respite care, controlling thoughts and completing care tasks), but the program only increased the self-efficacy with regard to controlling upsetting thoughts. Research in dementia caregivers showed that psychological interventions can improve self-efficacy with regard to obtaining respite (37). However, respite care and care tasks may be more complicated in ALS due to the severe care needs of the patient (8). Perceived self-efficacy aimed at organizing respite care and completing care tasks may be strongly influenced by the environment of the caregiver and, therefore, more difficult to influence compared to dealing with thoughts. The program explicitly focused on dealing with thoughts by providing information and exercises on how to deal with upsetting thoughts using ACT. With ‘care as usual’, professionals often focus on respite care and dealing with care tasks while dealing with negative thoughts is less common.

Contamination with care as usual might have affected the results in this study. Awareness of the difficulties caregivers face while caring for a person with ALS/PMA and their needs has increased (38). Caregivers nowadays receive more attention from professionals; this may have reduced the contrast between intervention and control condition. In addition, the proportion of caregivers who received professional support focused on dealing with their caregiving situation at the start of the trial differed between the intervention- and wait-list control group (25.7% and 50%, respectively), which might have affected our results.

Caregivers expressed their satisfaction with regard to the different components of the intervention but were ambivalent toward peer contact and mindfulness. These results are in line with the results of our qualitative evaluation of the intervention (39). Since some caregivers strongly valued these components, providing them optional is recommended. Caregivers expressed that the match between their needs and the content of the program could be improved by providing the option to follow each online module at a self-chosen order and time. This might improve the adherence rate and the added value of the intervention (40). E-health seems to be an acceptable and efficient option for receiving support; support becomes available for caregivers who are not able to leave their home due to care responsibilities and the threshold for receiving support is lowered (41). Choosing when and where they access support gives caregivers the opportunity to take control over the received support (42). However, the preferred modus might differ per caregiver; offering customized care in line with the caregiver’s preferences is advised.

Limitations

Some partners expressed that their main motivation for participating in this trial was to take part in ALS research rather than a need for support. Patients with ALS/PMA and their family members are often eager to cooperate in research in order to contribute to finding the cause of the disease (43). These caregivers were possibly less motivated to complete the intervention which might have affected our results. Another limitation might be that the levels of caregivers’ distress in our study were relatively low compared to results of previous studies (44,45), which leaves less room for improvement on this outcome measure. Future studies could consider using a questionnaire focused on symptoms of stress in stead of solely symptoms of anxiety and depression in this non-patient population. Another limitation might be the use of the original HADS for patients instead of a modified version for their physical symptoms (46). Furthermore, the questions used to measure self-efficacy might not have covered the complete caregiver situation as they were focused on a few aspects of caregiving. Caregivers might feel that they are capable of fulfilling care tasks or dealing with upsetting thoughts about caregiving, but they might not feel in control over their caregiving situation in general.

Conclusion

The support program did not reduce distress although it might be beneficial in empowering caregivers by improving feelings of control in their difficult situation. Appropriate timing of this support program seems to be crucial for caregivers in order to complete the program and benefit from the content. Optimal timing may differ per individual and is likely dependent on multiple factors such as the disease phase of the patient, caregivers’ needs, preferences, personal characteristics and their social network. Since our qualitative study revealed that caregivers reported benefits of the program including improved feelings of control (38) further evaluation of the support program is needed. A first next step would be to tailor the elements of the intervention to the needs of the caregiver (with respect to selection and timing of the modules) and to investigate the perceived benefits of a tailored program. A single case experimental design may provide a rigorous alternative to the RCT for investigating effectiveness of such a tailored intervention as in these studies, each participant serves as his/her own comparison, thereby controlling for confounding variables that can impact the outcome (47).

Disclosures statement

The authors declare that no competing interests exist.

Data availability

The data that support the findings of this study are available from the corresponding author, JMAVM, upon reasonable request.

Additional information

Funding

The research was supported by the Netherlands ALS Foundation, project no. 2014-27.