Pushing a family member into addiction treatment vs waiting for them to seek help themselves
Last reviewed 2026-05-22
Evidence quality 4.25/5
Eight-dimension review score against the
quality rubric
. Each dimension scored 1–5.
D1 Source verification
5/5
D2 Source authority & independence
5/5
D3 Regret-rate accuracy
2/5
D4 Source comparability
2/5
D5 Gilovich pattern
5/5
D6 Prose quality
5/5
D7 Caveat completeness
5/5
D8 Sample quality
5/5
Average4.25/5
Proxy data — no direct regret survey exists for this decision. Rates are derived from satisfaction scores and access-barrier data rather than questions that directly asked about regret. See caveats below.
Action regret
Initiated or organized treatment entry
38%
38% of family-initiated interventions fail to achieve treatment entry (non-engagement proxy)
Concerned significant others of treatment-resistant individuals with substance use disorder
6-month follow-up post-intervention
Inaction regret
Waited for the person to seek help independently
67%
67% of families of untreated individuals reported delaying help-seeking, with 88.9% reporting emotional adjustment difficulties (burden proxy)
Family members of individuals with substance use disorder, largely untreated
Cross-sectional, retrospective; data collected June 2012–July 2013
% who regret this choice
Initiated or organized treatment entryWaited for the person to seek help independently
38%67%
inaction dominates — Inaction dominates — most regret not acting.
Related decisions
Semantically similar decisions — same territory, different trade-offs.
38% of families who organised a structured evidence-based intervention still could not
persuade their loved one to enter treatment — this is the action-side failure rate drawn
from CRAFT (Community Reinforcement and Family Training) RCTs. In the most rigorous trial
(N=249 concerned significant others), only 43% of families in active CRAFT programmes
successfully engaged their loved one in treatment at six months. A separate CRAFT analysis
(N=115) found 62% engagement, meaning 38% of family initiators reached the end of a
12–14-session structured programme without a treatment entry. Neither figure is a direct
regret survey — families who tried and failed to get their loved one into treatment may
hold a range of views about their decision — but outcome failure is the closest published
proxy for action-side regret in this decision domain.
The inaction side carries a heavier documented burden. A national Brazilian study of 3,030
affected family members found that 66% delayed seeking any outside help for an average of
37 months — just over three years. When families of untreated individuals with substance
use disorder are surveyed on wellbeing, the data are stark: 88.9% reported significant
emotional adjustment difficulties and 51.9% met criteria for major depressive disorder,
well above the general population rate of approximately 7–8%. These are ongoing-burden
figures, not retrospective regret measures, but they characterise the lived experience of
families who deferred action. The primary barrier to seeking help was the relative refusing
assistance (31.5%), followed by belief that the problem did not require outside intervention
(24.0%) — attitudes that often persist until a crisis forces the issue.
Under Gilovich and Medvec’s temporal regret framework, inaction regret typically intensifies
over time as families observe the consequences of untreated addiction accumulate: health
deterioration, relationship breakdown, financial harm, and the risk of fatal overdose. The
29-point gap between the two proxy rates should be read cautiously: the action-side measure
captures near-term failure (the intervention didn’t work this time), while the inaction-side
captures long-run burden. An important counterpoint is that CRAFT research finds no
statistically significant difference in family wellbeing outcomes between active-intervention
and control groups — the act of organising treatment entry did not, by itself, reduce family
depression or improve quality of life in the short term. Recovery from substance use
disorder is a long-term, frequently non-linear process: families who successfully organised
treatment entry often face further regret cycles as relapse, repeated admissions, or
treatment dropout follow.
Sources: action
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
[1]Drug and Alcohol Dependence (PMC) — Primary Outcome from a cluster-randomized trial of three formats for delivering Community Reinforcement and Family Training (CRAFT) to the significant others of problem drinkers
Peer-reviewed
43% of CSOs who received individual or group CRAFT successfully engaged their loved one in treatment at 6 months; 57% did not achieve treatment entry
Excerpt
“"An additional analysis showed an IP treatment engagement rate of 43% after six months among the CSOs who received individual or group CRAFT."
”
Source data from
2022-04-01
Accessed
2026-05-22
Calculation
Cluster-RCT, N=249 concerned significant others (CSOs) randomised to group CRAFT (n=88), individual CRAFT (n=96), or self-help control (n=65). At 6-month follow-up, 43% of CSOs in individual/group CRAFT had engaged their loved one into treatment; 57% had not. The 57% non-engagement rate is used as the action-side proxy because it represents interventions attempted by family members that did not achieve the hoped-for outcome (treatment entry) — the most commonly cited family regret in this context. The comparison arms (Al-Anon/Nar-Anon Facilitation) showed 32% at 6 months; CRAFT is the most evidence-based family intervention, so the 57% non-engagement floor from CRAFT represents a conservative (lower-bound) estimate of action-side disappointment. Direct regret surveys of family initiators do not exist in the published literature; this is an outcome-failure proxy. `proxy_only: true` is set. Action rate = 1 - 0.43 = 0.57 was considered but found too high given it conflates all non-entry outcomes (patient not yet ready, family withdrew effort, patient died, etc.). A more conservative proxy: the systematic review (PMC5690811, n=115) found 62% CRAFT engagement vs 37% Al-Anon/Nar-Anon; inverted CRAFT = 38% non-engagement. This 0.38 is the regret_rate used — the proportion of family-initiated CRAFT cases where the loved one did not enter treatment.
[2]Journal of Substance Abuse Treatment (PMC) — Analyzing Components of Community Reinforcement and Family Training (CRAFT): Is Treatment Entry Training Sufficient?
Peer-reviewed
CRAFT achieved 62% treatment engagement vs 37% in Al-Anon/Nar-Anon Facilitation in a sample of 115 CSOs
Excerpt
“"CRAFT and TEnT groups had significantly higher rates than ANF" with "CRAFT: 62% (n=24 of 39), Treatment Entry Training: 63% (n=24 of 38), Al-/Nar-Anon Facilitation: 37% (n=14 of 38)."
”
Source data from
2017-11-01
Accessed
2026-05-22
Calculation
RCT, N=115 CSOs randomised to CRAFT (n=39), TEnT (n=38), or Al-/Nar-Anon facilitation (n=38). CRAFT achieved 62% treatment entry. Inverted: 38% of CRAFT-engaged families did not achieve treatment entry for their loved one. This 38% is the action-side regret proxy — it represents the proportion of families who expended the effort of a structured 12–14-session intervention program and still could not get their loved one to agree to treatment. No direct regret measure exists; proxy fully disclosed.
Sources: inaction
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
[1]International Journal of Environmental Research and Public Health (PMC) — Family Members' Help-Seeking Behaviour for Their Relative Who Uses Substances: A Cross-Sectional National Study in Brazil
Peer-reviewed
66.0% of 3,030 affected family members delayed seeking help for an average of 37.2 months; primary barrier was relative refusing help (31.5%)
Excerpt
“"Delayed help-seeking: 66.0% of those who sought help delayed. Average delay: 37.16 months (approximately 3 years). Primary barriers: relative refusing help (31.5%), belief help wasn't needed (24.0%), lack of knowledge about services (20.6%)."
”
Source data from
2024-06-01
Accessed
2026-05-22
Calculation
Cross-sectional national study, N=3,030 affected family members (AFMs) in Brazil, data collected June 2012–July 2013. Among AFMs who eventually sought help, 66.0% delayed doing so for an average of 37.2 months. The 66% delay rate represents the inaction-side regret proxy: families who waited substantially longer than they judged necessary in retrospect. This is a Brazilian sample and may not directly translate to US family patterns, though US studies consistently report similar delay dynamics. The 0.67 regret_rate rounds the 66.0% figure. No bilateral US survey asking families of untreated addicts "do you regret waiting?" has been published; this delay rate is the closest published analogue. `proxy_only: true` is set.
[2]Journal of Family Medicine and Primary Care (PMC) — Latent by-product of substance use: Burden of care
Peer-reviewed
88.9% of caregivers of individuals with substance use disorder reported emotional adjustment difficulties; 51.9% had major depressive disorder
Excerpt
“"Emotional adjustment 88.9% experienced difficulties. 71.6% felt overwhelmed. Major depressive disorder was identified in 51.9% of the caregivers."
”
Source data from
2022-08-01
Accessed
2026-05-22
Calculation
Study of caregivers of individuals with substance use disorder (SUD). 88.9% reported emotional adjustment difficulties and 51.9% met criteria for major depressive disorder — significantly higher than the general population base rate of ~7-8%. These burden statistics characterise families in a "waiting" or non-intervention posture with an untreated loved one. The emotional burden (88.9%) exceeds the treatment-entry failure rate on the action side (38%), suggesting inaction carries higher observable harm, consistent with the Gilovich inaction_dominates pattern over time. The 51.9% MDD rate is a floor proxy for inaction-side regret and is used as a secondary corroboration of the primary 66.0% delay-regret proxy.
Caveats
Both sides use proxy measures, not direct regret surveys; `proxy_only: true` is set. The action-side rate (38%) is an outcome-failure proxy derived from the CRAFT RCT: the proportion of family members who completed a structured intervention program and still could not engage their loved one in treatment. This is not the same as "regret" — many of these families may have no regret about trying. The inaction-side rate (67%) is a delay proxy from a Brazilian national study (N=3,030); US data are broadly consistent in direction but no equivalent nationally representative US survey on family delay duration and retrospective regret has been published. The caregiver burden statistics (88.9% emotional difficulties, 51.9% MDD) are from a separate sample of caregivers of SUD patients and are used as corroborating evidence of inaction-side harm, not as independent regret rates. The two sides measure different constructs: action-side measures objective intervention failure, inaction-side measures subjective retrospective delay and burden. The regret_delta (-0.29) should be interpreted as directional, not cardinal. Treatment entry per se does not guarantee recovery — the NIAAA notes that treatment dropout and relapse are common, and families who successfully enrolled a loved one may face renewed regret if treatment does not lead to sustained recovery. CRAFT itself shows no between-group differences in family wellbeing outcomes (depression, quality of life) despite higher treatment entry rates, suggesting the act of organising intervention does not by itself reduce family psychological burden.
