Peer support can offer value while still costing money. An economic estimate compares costs with outcomes, and the comparison determines what the result means.
When researchers call peer support services cost-effective, they mean the benefits are judged worthwhile relative to the costs and to a particular alternative. A program can improve outcomes, require additional spending and still be considered cost-effective. To understand the claim, start with three questions: Compared with what? Whose costs? And which outcomes?[1]
Those questions matter when a recovery center describes its value to a community or funder. They also help people seeking peer recovery support separate a claim about the value of a service from the amount they might pay to use it.
A cost-effectiveness study and an online calculator developed by Sierra Castedo de Martell and colleagues offer a concrete way to see the difference. Their work concerns a particular form of recovery support after substance use treatment. It is useful for understanding how an economic estimate is built, including what changes when its assumptions change.[1][2]
What Peer Recovery Support Did This Research Compare?
The underlying study compared one year of peer recovery support after specialty substance use disorder treatment with specialty treatment alone. Both sides of the comparison included the benefits of treatment. The question was what adding peer recovery support would cost and contribute.[1]
Castedo de Martell told AddictionHelp.com that the calculator is geared toward one-to-one peer recovery support services, often shortened to PRSS, in recovery community centers. These are recovery community organizations with a physical location. She explained that the tool was developed around how those centers provide support, rather than around clinical services.[3]
That service description is part of the finding. An estimate for one-to-one support after treatment does not establish the cost-effectiveness of every service called peer support, an entire clinical program or Alcoholics Anonymous.
The study used a computer model to combine evidence about costs and outcomes. It did not follow a new group of people for the rest of their lives. Its simulated population began at age 38 after specialty treatment, with health effects projected through age 82. The inputs came from different sources and populations, including important evidence about PRSS from Texas.[1]
This makes a practical difference when considering a local program. A center serving people who have never received specialty treatment has a different starting point from the modeled population. Ask whether the estimate actually represents that route into recovery. The study did not include a comparison with receiving no services.[1]
Whose Costs Count in Peer-support Cost-effectiveness?
A cost-effectiveness estimate depends on whose resources the analysis counts. The study considered both health-system costs and a broader societal perspective, which included participant time. Those perspectives answer different questions from a center’s operating budget or a participant’s bill.[1]
| Economic question | What the answer helps explain |
|---|---|
| What changes for the health system? | How the cost of providing support compares with changes in medical costs and health outcomes. |
| What changes for society more broadly? | Costs and benefits beyond medical spending, including the value assigned to participants’ time. |
| Can this organization afford the program? | The money the organization needs, receives and spends. A budget impact analysis addresses this more directly. |
| What will I pay? | The actual fees, coverage and other expenses involved in using the service. A cost-effectiveness ratio does not provide that answer. |
Castedo de Martell, PhD, MPH, a study author and calculator developer who is Senior Researcher at the recovery-support organization The Phoenix, emphasized the distinction in her response to AddictionHelp.com:
“Biggest thing is it doesn’t tell participants seeking help for SUD what kind of impact to expect on their own pocketbooks.”[3]
SUD means substance use disorder. She explained that cost-effectiveness analysis can help tell the larger story funders want to understand, while organizations need other methods to examine their own budgets. She also identified a need for more research into the economic effects on individual participants.[3]
For example, an analysis might place a value on medical care avoided elsewhere in the system. That value does not automatically become money available to pay the recovery center’s staff. Likewise, counting participants’ time in an analysis does not mean participants receive a payment for that time. The economic value, the organization’s cash flow and the person’s expenses are separate questions.
Costs Avoided Versus Costs Added in a Modeled Peer-Support Example
The calculator provides a useful illustration of how a program can produce estimated cost offsets while still adding costs overall. The following figures come from a synthetic scenario using the calculator’s default inputs, not from a recovery center’s financial records.[2]
The scenario represents 1,000 people, with 70% retained to one year or successful completion.[2] Of those retained, 83% are entered as being in recovery.[2] Other inputs are an average age of 38, 106 engagements per participant, 27.95 minutes per engagement including paperwork, and hourly peer pay of $35.86.[2] The pay figure is a historical model input, not a current wage recommendation.[2]
In a November 6, 2026 check of the calculator’s calculation function, those inputs produced approximately $1.78 million in gross averted societal costs.[2] That is the model’s estimate of costs avoided before considering the full net cost comparison. After the model accounted for the added costs, the scenario still had approximately $1.10 million in additional societal costs compared with treatment alone.[2]
These figures describe different parts of the same calculation. Reporting only the $1.78 million could make a reader think the program paid for itself.[2] The net result shows why that conclusion would be wrong for this scenario: the modeled offsets did not cover all the added costs.
The calculator also related the net cost to the added health benefit, producing roughly $5,100 per additional quality-adjusted life-year from the societal perspective.[1][2] A quality-adjusted life-year combines length of life and health-related quality of life into a common measure. One year in full health counts as one. In this type of analysis, the measure lets researchers compare added spending with added health benefit. It is not a dollar amount paid to a participant.[1][2]
To judge whether that added health benefit is worth the added cost, economists compare the cost per additional quality-adjusted life-year with a willingness-to-pay threshold: a benchmark for how much spending is considered worthwhile for that benefit. The calculator shows comparisons with thresholds of $50,000, $100,000 and $200,000 per quality-adjusted life-year, as well as a benchmark based on another specialty treatment episode.[1][2] The roughly $5,100 ratio in this example falls below each of those thresholds.[1][2][1][2]
The published study’s base-case model likewise found that adding peer recovery support after specialty treatment was cost-effective compared with treatment alone, while still increasing total costs. Its ratios were approximately $3,400 per additional quality-adjusted life-year from the societal perspective and $5,900 from the health-system perspective, below all thresholds the study tested.[1][2] Those are the published base-case results. The roughly $5,100 figure above comes from the separate calculator default-input scenario, so the figures should not be treated as the same calculation.[1][2][1][2]
The time periods also matter. The underlying method generally used 2019 costs, a one-year cost base and health effects projected over a modeled lifetime. A lifetime health benefit should not be read as a benefit all realized within the program year.[1]
A simple change shows how much an estimate can depend on assumptions. In the saved calculator check, reducing the number of engagements from 106 to nine, while holding retention and recovery inputs unchanged, changed the net cost from positive to negative. The modeled health effects and gross offsets stayed the same.[2]
That exercise tests the arithmetic of a cheaper service with the same assumed outcomes. It does not show that nine engagements work as well as 106. Before using such a result to justify a service change, the missing question is whether outcomes would actually stay the same.
Why a Recovery Center’s Input Definitions Matter
A calculator can only interpret the numbers it receives according to its definitions. Recovery centers need to know who was counted, when they were counted and what each measure means.
Consider the default recovery input above: 83% of the people retained, rather than 83% of everyone who started.[2] With 1,000 entrants and 70% retained, that means 83% of 700, or 581 people.[2] Confusing those denominators would change the scenario substantially. This is an illustration of the inputs, not an observed recovery rate for a center.[2]
A public Be Well Texas workshop uploaded in April 2024 shows why assembling those inputs takes work. In the HarborCare walkthrough, presenters discussed combining records from a statewide initiative for the 2023 service period. Their retention measure approximated January intakes still engaged in December. The recovery information could not be cleanly matched to that same one-year cohort. Contact figures combined individual and group contacts, and the discussion raised questions about contact time, documentation and benefits in labor costs.[4]
The workshop’s useful lesson is the need to define and reconcile the inputs. Those data choices are not verification that the one-to-one post-treatment model can be applied unchanged to mixed individual and group services, and the walkthrough does not establish observed state-budget savings.
Castedo de Martell also told AddictionHelp.com that better information about where people arrive from and where they go during or after PRSS could support more precise models. She acknowledged the burden of collecting that information for recovery community centers.[3]
For someone reviewing a center’s estimate, a useful request is: “Show me how your records match each input.” That makes it possible to distinguish a directly tracked measure from an approximation, and to see which assumptions deserve a second calculation.
Questions to Ask About a Peer-support Cost-effectiveness Claim
A useful explanation should let you connect the estimate to the service, the people using it and the decision at hand. Ask:
- What was the alternative? Treatment alone, another support service and no services are different comparisons.
- Who and what were studied? Identify the population, route into recovery, setting and actual service. A broad label such as “peer support” is not enough.
- Which outcomes were measured, and which were modeled? Ask how recovery was defined, who was included in the denominator and how far into the future benefits were projected.
- Whose costs were included? Find out whether the estimate describes health-system spending, wider societal costs, the organization’s budget or participants’ expenses. Ask which year’s prices were used.
- Is the headline figure gross or net? Costs avoided and the final cost after providing the service are different numbers. If a cost-effectiveness ratio is reported, ask what added outcome it buys.
- Which assumptions change the answer? Look for calculations showing what happens with different retention, service use or labor costs. Ask whether an assumed outcome has evidence behind it.
If you are seeking support for yourself or someone close to you, follow up with the question the economic model cannot answer: “What would using this service involve and cost for me?” Ask the program about its actual fees, insurance or other funding, eligibility, meeting frequency and services offered. The recovery coach guide addresses the separate task of finding a coach and understanding costs.
A clear economic claim should help you understand the value being proposed. A clear service conversation should help you decide whether that support fits your needs and circumstances. You can ask for both.
About the Research and Reporting
Castedo de Martell contributed attributable responses to AddictionHelp.com for this feature. She is a study author and tool developer, rather than an independent evaluator of the calculator. The published work disclosed that she held an unpaid board role at one participating feedback organization and a part-time evaluation contract at another during the study. The work received Recovery Research Institute pilot support through the National Institute on Drug Abuse, with additional NIDA support for manuscript preparation.[1]
The numerical example uses a bounded check of the university calculator’s current calculation function. It is not a full software validation or a reproduction of every published study result. The HarborCare account draws on the public workshop transcript, not a separate AddictionHelp.com interview or an audit of the program.
Get Treatment Help
If you or someone you love is struggling with addiction, getting help is just a phone call away, or consider trying therapy online with BetterHelp.
Exclusive offer: 20% Off BetterHelp*Following links to the BetterHelp website may earn us a commission that helps us manage and maintain AddictionHelp.com. *Get 20% off your first month of BetterHelp. Offer valid for new BetterHelp users only. Offer cannot be combined with insurance.
