Table 4.

Qualitative themes on workshop implementation: overarching theme and themes on strengths and weaknesses.

Theme and subtheme			Exemplar quotes	PARIHSa construct	Linkages to Table 3 rows
Overarching theme
	1. Importance of outreach and marketing to prompt workshop enrollment		Local staff, low-enrollment center: “If it was more advertised and marketed, I think that would be good.” Regional leader: “I feel like it’s an easy sell. We have the promotional materials to send out.” National leader: “I’ve learned that it’s important to have marketing materials for caregivers. That’s a big lesson.”	Spans all PARIHS constructs and implementation strengths and weaknesses	N/Ab
Implementation strengths
	2. Belief in positive impact of workshop encouraged uptake
		2a. Positive caregiver experiences during rollout	Caregiver: “You learned a lot of really good tools that I use in my daily life. And you could also communicate with the other class members.” Local staff, high-enrollment center: “When the caregivers took the course, I got such positive feedback from them, it made me a believer.” Regional leader: “Anybody that has participated in it has really given lots of positive feedback. And probably the most telling thing is the fact that staff continue to make the referrals.”	Evidence—caregiver experiences are positive	Row 1
		2b. Value of evidence from prior researchc	Local staff, low-enrollment center: “You want something that's, you know, evidence-based.” Regional leader: “The research definitely does matter. Because like I said, you're pitching an additional task to [caregivers] who are super busy.”	Evidence—research evidence is convincing	Row 2
	3. Successful outreach to some caregiver groups
		3a. Use of stories and testimonials from trusted sources	Caregiver: “[Caregivers will enroll] if they have good caregiver coordinators that push the program and say, “This is something that you really need to do.” Or the therapists, whoever they trust.” Local staff, high-enrollment center: “My suggestion [for other centers] would be short little testimonials of people who've taken the course and it would say, ‘See what Mary said about the course.’” Regional leader: “I always tell people when promoting it—caregivers when they link up they’ve really enjoyed it.”	Evidence—materials help caregiver determine that program is likely to meet their needs; Facilitation—staff have necessary skills	N/A
		3b. Multiple contact episodes and materials	Caregiver: “First I heard verbally about it. Then they sent me a flier regarding it and said, ‘Here is what it is – read it and see if this is what you're looking for.’ Then he followed up with a phone call. I think the more information, the better.” Local staff, low-enrollment center: “I would say the biggest thing [hindering enrollment] is that we don’t have a follow-up plan—a reminder. Because sometimes that reminder helps. To talk about it and you know, give them a little push.” Regional leader: “We’ve sent out like a large volume of fliers—and I’ve pitched it to people in person and over the phone and things like that. You can't just do it once. You have to repeatedly send stuff out or bring it up.”	Context—staff have sufficient resources; Facilitation—staff have necessary skills	Row 3
Implementation weaknesses and the needs they suggest
	4. Missed opportunities for improved outreach
		4a. Detailed information on workshop content and structure	Caregiver—who knew about the workshop but had not enrolled: “Provide more actual information on the content, not just a link on the computer.” Local staff, low-enrollment center: “I don't see how the system works. I would like that. One caregiver was telling me that she didn't feel there was enough information, but without seeing it, I couldn't respond to her.” Regional leader: “Some of the CSCsd do not understand the details. I think it would be helpful to train the coordinators on what actually is in [it]. Because—marketing wise—you're not going to refer people as readily to something that you don't have knowledge about.”	Evidence—caregiver cannot determine if program is likely to meet their needs; Evidence—staff cannot observe program	N/A
		4b. Expanded online mechanisms for outreach and enrollment	Local staff, low-enrollment center: “My suggestion would be a link on the site [portal to electronic health records]—on the same page, if caregivers could just click on right there.” Local staff, high-enrollment center: “If the general caregivers [caregivers of older, pre-9/11 era veterans] could access it from the caregiver website, that might be really good.” National leader: “There's a barrier in that you have to pick up the phone and find your CSC. So the recruitment process is interrupted by the fact that it's not a complete online experience, even though the rest of their experience will be online.”	Context—lack of technology tools	N/A
		4c. Partnership with communities and community groups	Only 1 caregiver learned about the workshop through a community group: “Crossroads [VA-funded nonprofit] is where he goes once a month. They just said, ‘Well, try this and maybe they can help you.’ So I picked up a brochure.”e Only 1 local staff member, at a high-enrollment center, described doing outreach to the community: “I do community outreach events where I put out that fact sheet.”e National leader: “We have tremendous access to community-based organizations that are working with caregivers all over the country. And we've wanted to promote the program through that [but have not been able to].”	Context—limited awareness among external networks and communities	N/A
		4d. Increased outreach to certain caregiver groups and their health care teams	Pre-9/11 era caregiver: “I don’t think it’s widely known that you have this. None of the doctors told me about it.” Local staff, low-enrollment center: “I definitely think outreach and mailers could be made more regularly to the general [pre-9/11 era] caregivers. It just kind of goes by the wayside.” National leader: “One of the goals is to increase the number of general [pre-9/11 era] caregivers in the program. I don't think the coordinators have as much contact with them.”	Context—limited awareness among some stakeholder groups; Context—limited staff resources and time	Row 3
	5. Missed opportunities to support staff in outreach efforts
		5a. Training and mentoring for new staff	Local staff hired after the rollout began, low-enrollment center: “I had no one to collaborate with, no one to talk to. I think new CSCs need to be more knowledgeable of the program.” Regional leader: “The initial rollout was either really early in my assuming responsibility for this role or shortly thereafter. Initially it flew right past me.”	Facilitation—lack of training and mentoring from experts	Rows 4 and 5
		5b. Improved data management capabilities to Generate outreach contacts Track caregivers Target follow-up outreach	Local staff, low-enrollment center: “A quarterly flier might just remind people of the opportunity to participate in this program. At the national level can they generate mailers?” Local staff, high-enrollment center: “It would be very beneficial to have a way to pull up the list of people that we have referred that have never taken any steps forward, so that we can hit a button and they get a reminder email.” Regional leader: “The website where you can see where the caregiver is in the process, I don't always check it. It probably would be helpful. But you don't always have the time to check individually.” Regional leader: “Once we make referrals, we’re not involved. Unless there was some sort of process where if the caregiver didn’t follow through, [workshop organizers] came back to the coordinator and said, ‘Will you see if they're still interested?’ There’s no mechanism for that.”	Context—lack of information technology capabilities	Row 6

^aPARIHS: Promoting Action on Research Implementation in Health Services.

^bN/A: not applicable.

^cCaregiver participants did not make comments about research evidence.

^dCSC: caregiver support coordinator.

^eThese are not exemplar quotes. They are the only cases in which respondents described using these outreach mechanisms and thus highlight their relative absence from use.