Table of contents
I remember the first week of the program I did at the Milwaukee Secure Detention Facility in 2019. It had zero hours of group. There were 40 men in the program and two facilitators. The facilitators spent that whole week doing one-on-one intake interviews with each of us. Most of us spent that week just trying to pass the time.[1]
Everyone started together. So everyone waited together.
When I got to a halfway house in Kenosha in December 2016, it was the middle of a Wednesday afternoon. The other residents were already there. One of them showed me around the house. The guys in my group were the ones who told me how the program’s hours worked.[1]
Those were two different ways of running a recovery group. The first one is a closed cohort, and the second one is rolling enrollment.
Most people who go through treatment or a reentry program never hear those words. But the choice shapes how long they wait to start, who is in the room, and how much programming they actually get.
This guide is for program directors, facilitators, and anyone trying to understand why a group runs the way it does. It covers what the two terms mean, what each one does well, what goes wrong in each, and how a program can pick.
Direct Answer
A closed cohort is a group where the same people start on the same day and finish together. Rolling enrollment (also called open or revolving enrollment) lets new members join an ongoing group when they become ready. SAMHSA’s group therapy guide says fixed groups are rare because they take a long-term commitment of resources. It also says revolving groups need more structure and more active leaders.[2] In one survey of 57 treatment programs, 84 percent of directors said most of their groups used open enrollment. They gave shorter waits and steady billing as reasons.[3] Closed cohorts make trust and a set sequence easier. Rolling groups let someone start when they need help, but they need a plan for bringing new members in.
What the two terms mean
A closed cohort is a group that forms with a set list of people and runs for a set length. SAMHSA calls it a fixed membership group. New members get in only at the very start (for example, during the first week of a daily group). That’s because each session builds on the one before it.[2]
Rolling enrollment means people enter the group when they are ready. They don’t wait for a new group to form. SAMHSA calls these revolving membership groups. As clients are admitted and discharged, people come and go. The group has to adjust to changes it can’t always predict.[2]
Rolling enrollment comes in a few versions, and SAMHSA lists three of them:
- Clients stay for as many sessions as they want.
- The group repeats a set of topics, and each client stays until they have finished all of them.
- Clients attend for a set number of weeks.
For example, SAMHSA points to an interpersonal process group in an intensive outpatient program. A person joins it and stays until the work in their treatment plan is done.[2]
So what does a program mean by rolling admission?
It is worth asking. A drop-in continuing care group is rolling. So is a 16-week sequenced curriculum with people entering at different weeks. Those two run very differently, even though they share a label.
How common each one is
Rolling enrollment is the more common of the two in community treatment. In a 2005 survey of 57 substance abuse treatment programs, 84 percent of treatment directors said most of their therapy groups used open enrollment.[3]
That survey is small and old.
But it lines up with SAMHSA’s group therapy guide. The guide says fixed groups are rare because they require a long-term commitment of resources. It notes that most outpatient programs provide only 8 to 20 sessions, and most inpatient programs last 2 to 4 weeks.[2]
Why do so many programs pick rolling? The directors in that survey gave two kinds of reasons. One was money: an ongoing group has no waiting period before a new client can start and be billed, and there is no point where the group stops. The other was clinical: rolling admission cut the wait to get into a group, compared with closed groups that don’t let anyone in after the group starts.[3]
What a closed cohort does well
A closed cohort lets a facilitator teach in order. That means week three can build on week two, because everyone in the room was there for week two. SAMHSA describes this directly: in a time-limited fixed group, learning builds on earlier meetings, so members need to be there from the start.[2]
It also lets people get to know each other over the same stretch of time. The group starts together, works through the hard middle together, and finishes together. When I was at MSDF, I wrote in my journal about the men who would be “graduating with us in July.” The graduation ceremony was something the whole group did together.[1]
Everyone knows the last day, too. That makes the ending easier to plan.
What goes wrong in a closed cohort
The biggest problem is the wait, because there is no way to join a closed group once it has started.[2] If someone is ready today and a new group only starts every few months, that person has to wait for the next start date.
I know what that wait looks like.
My probation officer came up with an Alternative to Revocation for the MSDF program. I was told it was a four-month wait on a four-month program. When I worked out the math in my journal back then, the average wait to get in was about four months. I waited two. Two of my cellmates waited seven. The program was 90 days, but weekends didn’t count toward those 90 days, so it ran close to four months. A four-month wait plus a four-month program came to about eight months on a probation hold.[1]
One time, I got an Alternative to Revocation for an inpatient AODA program inside a jail. It was a one-month program, but there was a one-month wait for a start date.[1]
Is the wait just a scheduling problem? I think it is more than that. SAMHSA notes that research shows a strong tendency toward relapse early in the treatment process. A person early in recovery is at greater risk of returning to use than someone with months of abstinence.[2] The risk is that a person who is ready for help and gets told to wait spends that early stretch without the group.
The second problem is that closed cohorts shrink. When people leave, nobody new comes in to replace them.
In the Earned Release Program at Chippewa Valley, my group had 10 people. Six of us made it to graduation.[1] By the end, a group of 10 was a group of six.
What rolling enrollment does well
Rolling enrollment gets people into the group faster. That was the main clinical reason the directors in the survey gave for using it.[3]
It also means the room almost always has people in it who have been there a while. SAMHSA notes that longtime members of a rolling group often help orient newer members. It lists the energy new members bring as one advantage of this kind of group.[2]
That’s how it worked for me at the halfway house in Kenosha. The program there used to be a standard 90 days. By the time I arrived, it was based on group hours instead. After an assessment, each person got one of three totals: 150, 200, or 250 group hours. Depending on their needs, a person could take up to 23 hours of programming a week. I was given 250. My group mates were the ones who told me that when I arrived.[1]
Because each person had their own hour count, people finished on their own schedules instead of on one shared date. Nobody had to wait for a new group to form, and nobody had to stay longer because the rest of the room started later.
What goes wrong with rolling enrollment
The new person feels behind.
SAMHSA says new members may dread joining because the people already there know each other, know how the group works, and know what has been talked about. The group can be uneasy about the new member, too.[2]
Constant change can also weaken the group. SAMHSA names the effect that membership changes can have on how close the group is. For that reason, it calls preparing people for a rolling group extremely important.[2]
Each time someone joins, the group is back at its beginning for that one person.[2]
It is also harder to teach in order when people keep joining. What happens to someone who joins a sequenced curriculum in week nine? They have missed weeks one through eight. That means the facilitator either has to make each session stand on its own or have a plan for catching the new person up.
And rolling groups can get too big. How big is too big? SAMHSA warns that the temptation to keep adding members is strong if staff are not trained enough and funding is short. It suggests keeping a rolling group to about 15 or fewer. Past 20, it says, group interaction breaks down and the group becomes a class of individuals.[2]
Group size and staffing decide a lot
The format is only part of it.
When I was at MSDF, the 40 men were split between two facilitators, 20 each. One facilitator broke her 20 into two groups of 10, and her clients got about eight hours of programming a week. The other kept all 20 men in one group and ran longer sessions. Her clients got about 16 hours a week. We were all in the same program on the same ATR, but the hours were not the same. Which facilitator you got was the luck of the draw.[1]
To me, that is the part people outside a program don’t see. The same program can look very different from one room to the next, depending on things like staffing and group size.
So when a program decides between rolling and closed, it should also decide a few other things: how many people one facilitator can carry, how many hours each person will get, and what happens when the room gets too full. SAMHSA’s suggestion of about 15 or fewer for a rolling group is a reasonable place to start, but each program still has to set its own number.[2]
What the research can and cannot tell you
Research comparing outcomes in rolling and closed recovery groups is limited. Part of the reason is that open enrollment data are technically hard to analyze.[4]
Researchers Antonio Morgan-Lopez and William Fals-Stewart wrote that this difficulty led federally funded treatment trials to avoid open enrollment groups, even though most real programs use them.[4] The usual statistical approaches were built for groups where the same people stay put. They don’t handle a room where people keep coming and going. In one simulation study, the standard models overestimated treatment effects when dropout ran around 30 to 50 percent and people with different attendance patterns responded differently to treatment.[4]
What does that mean for a program director?
Trials avoided open enrollment groups, so a lot of the evidence behind group curricula may come from closed groups that are easier to analyze.[4] If your program runs rolling groups, that means the curriculum is being used in a different way than it was tested. It can still work. But the program should plan how new members enter, and it should keep track of how people do.
How to choose
There is no single right answer. I think the setting usually decides most of it, and this is how I would sort it:
| Setting | Usually fits | Why |
|---|---|---|
| Short inpatient or residential stay with admissions every week | Rolling | People arrive and leave on their own dates; a closed group would leave beds empty or people waiting |
| Sequenced skills or psychoeducation course with a set length | Closed | Each week builds on the last |
| Reentry program tied to release dates | Rolling, or closed with frequent starts | Release dates do not line up |
| Continuing care or aftercare | Rolling | People step down from different programs at different times |
| Pilot program you want to measure | Closed | Easier to track outcomes for one group from start to finish |
A program can also split the difference. SAMHSA notes that some rolling groups bring new members in at set times, such as once a week for a daily group.[2] That keeps the wait short without a new face in the room every session.
A few questions help sort it out:
- How long will a person who is ready today have to wait to start?
- Does the curriculum need to be taught in order, or can each session stand alone?
- How many people can one facilitator actually carry in one group?
- What happens in the first session for someone who joins late?
- How will the program know whether people are doing well?
How to make rolling enrollment work
Most programs are going to run rolling groups.
So I think the more useful thing to ask is how to run them well, and SAMHSA’s guide gives a few concrete steps. First, meet with each person before they join. SAMHSA recommends individual pregroup sessions for five things: building a working relationship, agreeing on goals, explaining how group works, easing anxiety about joining, and going over the group agreement.[2]
Next, tell the group ahead of time that someone new is coming. Introduce the new person, and ask current members to explain how the group works.[2] Existing members helped me find my way at the halfway house. SAMHSA treats that kind of help as something the leader should set up on purpose.[1][2]
Keep the group small enough that people feel heard and the facilitator knows each person.[2]
If the curriculum is sequenced, have a plan for where a new person starts and how they catch up. A new arrival shouldn’t reset the room, but they also shouldn’t get dropped into week nine without any context.
Where ReturnPath fits
I built ReturnPath as a 16-week facilitator-led curriculum for recovery and reentry programs. Participants do nightly reading and daily Workbook writing on their own, and facilitated sessions work through what came up. ReturnPath can run as a closed cohort or with rolling enrollment. The facilitator guides include synchronization steps for people who join partway through the arc. ReturnPath complements clinical care; it does not replace it.
For more on how that looks in practice, see the guides for recovery workbook treatment groups and halfway house reentry curriculum. If the wait to get into a program is the problem you are dealing with, the Wisconsin piece on program waitlists and short sentences covers how that plays out inside DOC. The programs I describe here, from MSDF to the halfway house in Kenosha, are in my memoir, A Vision of Hope.
Closing
I’ve been through both kinds of groups. I waited months for a closed program to start. When I walked into a rolling one, the guys already there showed me how it worked.[1]
Neither format fixes a program that is understaffed or a group that is too big. A closed cohort with a months-long wait still leaves people waiting at the riskiest time. A rolling group with 20 people and no plan for new members still leaves the new person lost.[1][2]
Which one would I pick? If I were running a program, I would start with two numbers: how long someone has to wait, and how many people one facilitator can carry. In the end, I would pick the format that keeps both of those numbers low.
Frequently Asked Questions
What is the difference between rolling enrollment and a closed cohort?
A closed cohort starts with a set group of people who begin and finish together, and new members are admitted only at the very start. Rolling enrollment lets new people join an ongoing group when they are ready, so membership changes as people arrive and leave.[2]
Which is more common in addiction treatment?
Rolling enrollment. In a 2005 survey of 57 treatment programs, 84 percent of directors said most of their therapy groups used open enrollment.[3] SAMHSA’s group therapy guide describes fixed groups as rare because they require a long-term commitment of resources.[2]
Why do programs use rolling admission?
Directors gave two main reasons. Rolling admission cuts the wait to get into a group, and an ongoing group has no waiting period before a client can start and no point where the group ends.[3]
What are the downsides of rolling enrollment?
New members can feel at a disadvantage because others already know each other and the material. Frequent membership changes can also weaken how close the group is. SAMHSA recommends careful preparation and keeping groups to about 15 or fewer.[2]
How big should a recovery group be?
For rolling groups, SAMHSA suggests about 15 or fewer so people feel heard and the leader knows each member. It warns that past 20 the group stops working as a group and becomes a class of individuals.[2]
Is there research on which format works better?
Direct comparisons are limited. Researchers have noted that open enrollment groups were avoided in federally funded trials because the data are hard to analyze, even though most programs use them. Standard models can also overestimate treatment effects in rolling groups.[4]
References
- Drasen A. A Vision of Hope (memoir). A Vision of Hope Media. https://www.avisionofhopebook.com/books/memoir. Journal entry, April 5, 2019 (MSDF: zero hours of programming the first week while facilitators did intake interviews; 40 men and two facilitators; groups of 10 vs 20; eight vs 16 hours a week; “Luck of the draw”; average four-month wait, “I waited two months,” two cellmates waited seven; 90-day program with weekends not counted; eight months on a probation hold); Alternative to Revocation for MSDF, “a 4-month wait on a 4-month program”; “he won’t be graduating with us in July”; Earned Release Program at Chippewa Valley, groups of 10, “Out of our group, six out of 10 made it to graduation”; Options halfway house in Kenosha, arrived December 20, 2016, midafternoon on a Wednesday, shown around by a resident, group hours of 150, 200, or 250 with up to 23 hours a week, “I was given 250 group hours,” “My group mates informed me of this fact upon arrival”; jail inpatient AODA program, “a one-month program, with a one-month wait for a start date.”
- Substance Abuse and Mental Health Services Administration. Substance Abuse Treatment: Group Therapy, Chapter 4: Group Development and Phase-Specific Tasks. Treatment Improvement Protocol (TIP) Series, No. 41. 2005. Fixed vs revolving membership groups; new members admitted only early in time-limited fixed groups; fixed groups rare, most outpatient programs 8 to 20 sessions and most inpatient 2 to 4 weeks; revolving groups adjust to frequent membership changes, need more structure and active leadership; about 15 or fewer, more than 20 becomes a class; experienced members orient newer ones; new members may dread joining; membership changes and cohesion; new members may enter at fixed intervals; three ongoing revolving formats; pregroup interviews; early relapse risk.
- Morgan-Lopez AA, Fals-Stewart W. Analyzing Data from Open Enrollment Groups: Current Considerations and Future Directions. Journal of Substance Abuse Treatment. 2008;35(1):36-40. Survey of 57 treatment programs (Fals-Stewart 2005): 84 percent of directors said most therapy groups used open enrollment; economic reasons (no wait before billing, group never ends) and clinical reasons (shorter wait for group entry than closed groups).
- Morgan-Lopez AA, Fals-Stewart W. Consequences of Misspecifying the Number of Latent Treatment Attendance Classes in Modeling Group Membership Turnover within Ecologically-Valid Behavioral Treatment Trials. Journal of Substance Abuse Treatment. 2008;35(4):396-409. Research comparing outcomes in rolling and closed groups is limited because open enrollment data are technically hard to analyze; open enrollment groups avoided in federally funded trials despite most programs using them; standard growth models overestimated treatment effects and inflated Type I errors under 30 to 50 percent dropout.