Do high-risk women (e.g. BRCA carriers) regret risk-reducing mastectomy — or regret choosing surveillance instead?
If you act
Having a risk-reducing (prophylactic) mastectomy
19%
If you don't
Choosing intensified surveillance instead of surgery
14%
Percentage who later regret each choice. Bars and full ledger render below.
Health
Last reviewed 2026-07-14
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
3/5
D4 Source comparability
3/5
D5 Gilovich pattern
4/5
D6 Prose quality
5/5
D7 Caveat completeness
5/5
D8 Sample quality
4/5
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
Having a risk-reducing (prophylactic) mastectomy
19%
19% dissatisfied with the procedure a mean 14.5 years on (Frost); in a separate cohort 7% would not choose the operation again and 33% found the cosmetic result not as expected
BRCA carriers / high-risk women after bilateral prophylactic mastectomy
mean 14.5 years post-procedure
Inaction regret
Choosing intensified surveillance instead of surgery
14%
no direct regret survey — proxy: cancer worry rises rather than falls under surveillance, and 14% of surveillance-choosers developed breast cancer (28/201 over a median 5.9 years; 9.9% cumulative at 5 years)
BRCA carriers / high-risk women choosing intensified breast surveillance
crude incidence over a median 5.9-year follow-up; worry measured over 6–8 months
% who regret this choice
Having a risk-reducing (prophylactic) mastectomyChoosing intensified surveillance instead of surgery
19%14%
balanced — Roughly balanced — both choices carry similar regret.
Related decisions
Semantically similar decisions — same territory, different trade-offs.
At long follow-up, roughly one in five women is dissatisfied with a risk-reducing mastectomy. The largest long-term study — 572 women a mean of 14.5 years after bilateral prophylactic mastectomy at the Mayo Clinic — found 19% dissatisfied with the procedure, alongside 74% who reported diminished emotional concern about developing breast cancer (Frost et al., 2000). That dissatisfaction is mostly about the body rather than the decision: about a third reported an unfavourable effect on satisfaction with body appearance, and a quarter on feelings of femininity. An independent Norwegian cohort of BRCA1/2 carriers pins the narrower figure down — only about 7% would not choose the same operation again, though a third said the cosmetic result was not as expected, and lower satisfaction clustered among women who found the decision process difficult or felt unsupported by their clinicians (Hyldebrandt et al., 2025). The regret that exists tends to attach to reconstruction complications and to decisions that felt externally initiated, not to the cancer-risk reduction the surgery reliably delivers.
The surveillance side has no direct regret survey at all, so its rate is a proxy assembled from two things. First, the harm that materialises: in a Canadian cohort, 14% of women who chose surveillance developed breast cancer (9.9% cumulative at five years), against 6% among those awaiting surgery (Macadam et al., 2021). Second, the burden that persists: on the one measure taken comparably on both arms — cancer worry — surgery lowers it and surveillance does not. A prospective cohort found anxiety fell over six to eight months in women who chose mastectomy but rose in women who chose surveillance, ending higher in the surveillance group (Dick et al., 2022). Developing a cancer is not the same as regretting the choice, and many surveillance-detected cancers are caught early, so the 14% inaction figure is a materialised-harm-plus-worry stand-in, not a measured regret rate.
The two numbers land close together, but the closeness is fragile: it depends on which proxy anchors each side. They are not measuring the same thing — the action rate is long-term dissatisfaction with an irreversible operation, while the inaction rate is cancer incidence plus unresolved worry among women who kept their breasts. Swap the broad 19% dissatisfaction for the narrower 7% who would not choose the surgery again, and the balance tips toward surveillance; anchor the inaction side on the 9.9% five-year cumulative incidence instead of the 14% crude rate, and it tips the other way. On the worry axis alone the surveillance side carries more — the Gilovich-typical pattern in which the road not taken keeps generating unease — but risk-reducing mastectomy is permanent and exacts body-image and reconstruction costs that surveillance never imposes, which is what holds the action-side dissatisfaction up to meet it. None of the underlying data is a US survey of the surveillance arm, and no study asks surveillance-choosers directly whether they would decide differently, so the near-balance is a direction with wide error bars, not a settled equivalence.
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]JAMA (Frost et al. 2000, Mayo Clinic) — Long-term satisfaction and psychological and social function following bilateral prophylactic mastectomy
Peer-reviewed
Among 572 women (94% of survivors), mean 14.5 years after bilateral prophylactic mastectomy: 70% satisfied with the procedure, 11% neutral, 19% dissatisfied; 74% reported a diminished level of emotional concern about developing breast cancer; negative effects were reported on satisfaction with body appearance (36%) and on feelings of femininity (25%)
Excerpt
“"Most women (70%) were satisfied with the procedure; 11% were neutral; and 19% were dissatisfied. Among the psychological and social variables, the most striking finding was that 74% reported a diminished level of emotional concern about developing breast cancer. The majority of women reported no change/favorable effects in levels of emotional stability (68%/23%), level of stress (58%/28%), self-esteem (69%/13%), sexual relationships (73%/4%), and feelings of femininity (67%/8%). Forty-eight percent reported no change in their level of satisfaction with body appearance; 16% reported favorable effects. However, 9%, 14%, 18%, 23%, 25%, and 36% reported negative effects in these 6 variables, respectively."
”
Source data from
2000-07-19
Accessed
2026-07-14
Calculation
The action-side anchor. Frost et al. is the largest long-term follow-up of prophylactic mastectomy (n=572, mean 14.5 years, US Mayo Clinic cohort) and the standard reference for durable outcomes. It measures satisfaction, not "regret" directly: 19% were dissatisfied with the procedure at long follow-up, which we take as the action-side proxy for a lingering negative decision outcome. This overlaps with — but is broader than — outright decision regret (see the Hyldebrandt cohort, where only ~7% would not choose the operation again). The 19% dissatisfied largely reflects body-image, femininity, and cosmetic outcomes rather than a wish to have skipped the surgery altogether.
[2]Familial Cancer (Hyldebrandt et al. 2025) — Experiences with risk-reducing mastectomy in Norwegian BRCA1/2 carriers without prior breast cancer
Peer-reviewed
Of 190 BRCA1/2 carriers who had undergone risk-reducing mastectomy: 78.9% (150/190) satisfied and would choose the same procedure again; 13 (6.8%) would not choose the same operation again; 63 women (33%) said the operative result was not as expected; satisfaction with healthcare-system support gave an OR of 5.5 for being satisfied with RRM (p<0.01), and finding the decision difficult lowered the odds (OR 0.2, p=0.02)
Excerpt
“"78.9% (150/190) were satisfied with their decision and would choose the same procedure again. [...] 13 (6.8%) [would not have chosen the same operation]. [...] 63 women (33%) expressed that the result was not as expected. [...] Feeling satisfied with support from the health care system gave an OR of 5.5 for being satisfied with having undergone RRM (p < 0.01). Those who found the decision difficult had lower odds of being satisfied (OR 0.2, p = 0.02)."
”
Source data from
2025-01-01
Accessed
2026-07-14
Calculation
Independent corroboration from a Norwegian cancer-unaffected BRCA1/2 cohort (n=190 who had RRM, of 272 respondents). It supplies the cleaner decision-regret figure: 6.8% would not choose the operation again — a floor well below Frost's 19% dissatisfaction, because "dissatisfied with the result" and "would not do it again" are different constructs. The third of women whose result was "not as expected" tracks the same body-image/reconstruction dissatisfaction Frost measures. The finding that a difficult decision process and weak healthcare support predict lower satisfaction is the source for the prose claim that RRM regret concentrates around reconstruction outcomes and decisions that felt externally driven rather than chosen.
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]Hereditary Cancer in Clinical Practice (Dick et al. 2022) — Psychological factors and the uptake of preventative measures in BRCA1/2 pathogenic variant carriers: results of a prospective cohort study
Peer-reviewed
Prospective cohort (n=98). Anxiety in the intensified-breast-surveillance (IBS) group rose over 6–8 months (mean ~6.2 → ~6.6), while anxiety in the risk-reducing-mastectomy group, elevated just after test-result disclosure (T1 ~8.2, up from ~7.6 at baseline), fell to ~5.7 over the same period; elevated post-disclosure anxiety predicted opting for RRM (OR 1.2, p<0.01)
Excerpt
“"Baseline anxiety levels in women opting for RRM were high but decreased over time, while they increased in women opting for intensified breast surveillance (IBS). [...] Elevated levels of anxiety after genetic test result disclosure (T1) were associated with the decision to undergo RRM (p < 0.01; OR = 1.2, 95% CI = 1.05–1.42)." [Reported group means: RRM T1 8.2 → T2 5.7; IBS T1 6.2 → T2 6.6, so at follow-up the surveillance group's mean anxiety (6.6) exceeded the mastectomy group's (5.7).]
”
Source data from
2022-12-01
Accessed
2026-07-14
Calculation
The construct-matched half of the inaction proxy, and the reason the two sides can be compared at all. On the one axis measured comparably on both arms — psychological burden / cancer worry — surgery lowers it and surveillance does not: the surveillance group's anxiety rose and ended higher than the mastectomy group's. This is the residual worry that is the honest analogue of "regret" for the surveillance choice. It is not a regret rate and yields no percentage; it fixes the direction of the worry burden, not the magnitude.
[2]Current Oncology (Macadam et al. 2021) — Prophylactic Surgery in the BRCA+ Patient: Do Women Develop Breast Cancer While Waiting?
Peer-reviewed
Of 333 BRCA carriers, 201 chose surveillance and 132 chose prophylactic mastectomy with immediate reconstruction. 28/201 surveillance patients (13.9%) developed invasive breast cancer vs 8/132 (6.1%) in the surgery group; 5-year cumulative incidence 9.9% (surveillance) vs 4.9% (surgery). 89% of surveillance-detected tumours were <20 mm
Excerpt
“"During this period, 6% of patients developed breast cancer compared with a 14% incidence of breast cancer in patients choosing surveillance. [...] At five years, the cumulative incidence of breast cancer in the surveillance group was 9.9% compared with 4.9% in the PM/IBR group." [Results: "201 patients chose S, 132 chose PM/IBR"; "Twenty-eight of the 201 surveillance patients and eight of 132 patients in the PM/IBR group developed invasive breast cancer"; median surveillance follow-up 5.9 yr; 89% of tumours <20 mm.]
”
Source data from
2021-02-01
Accessed
2026-07-14
Calculation
The materialized-harm anchor for the inaction proxy. The inaction-side rate is set at 0.14 — the crude cited figure: 28 of 201 surveillance-choosers (13.9%) developed invasive breast cancer in this Canadian cohort over a median 5.9 years. No worry adjustment is applied; an earlier draft nudged this to 0.15 to "fold in" the rising cancer worry, but that increment has no source basis (Dick 2022 yields no regret percentage) and has been removed. The 5-year cumulative incidence of 9.9% is the alternative anchor; choosing it instead would move the rate to ~0.10 and the delta to ~0.09. Developing cancer is not itself "regret" — 89% of the surveillance-detected tumours were under 20 mm, i.e. screen-caught early, and no survey asks surveillance-choosers whether they regret the decision — so this is a materialized-harm-plus-worry proxy, not a regret survey. It is the closest available inaction anchor and is deliberately flagged as a proxy. Boundary call: with action 0.19 and inaction 0.14 the delta is exactly 0.05, sitting right on the rubric's balanced/action_dominates line (|delta| < 0.05 is balanced). The label is deliberately kept at "balanced" rather than tipped to "action_dominates" because the two rates are non-comparable proxies (long-term dissatisfaction vs materialized cancer incidence); a signed delta at the boundary between them carries no interpretable direction, so "neither side clearly dominates" is the honest reading.
Caveats
Both rates are proxies from non-comparable cohorts, and this entry is published with a proxy disclosure for that reason. The action-side 19% is a satisfaction/dissatisfaction figure measured a mean 14.5 years after bilateral prophylactic mastectomy (Frost 2000, US, n=572); the cleaner decision-regret figure from an independent BRCA cohort is much lower — about 7% would not choose the operation again (Hyldebrandt 2025, Norway, n=190), with a third reporting the cosmetic result was not as expected. The inaction-side 14% is not a regret rate at all: it is anchored on the 14% of surveillance-choosers who developed breast cancer (Macadam 2021, Canada, n=201; 28/201 over a median 5.9 years, 9.9% cumulative at 5 years), a materialized harm, with the rising cancer-worry trajectory under surveillance (Dick 2022, Europe, n=98) as the construct-matched corroboration. No published survey directly asks women who chose surveillance whether they regret it, so the inaction rate carries wide uncertainty. The two sides measure different constructs — long-term dissatisfaction with an irreversible surgery versus cancer incidence plus residual worry — so the delta of 0.05 sits at the balance boundary and is not an interpretable magnitude in either direction; it should be read as "neither side clearly dominates," not as a precise equality and not as action being regretted more. That near-balance is also partly an artefact of anchor selection: taking the cleaner decision-regret figure for the action side (Hyldebrandt's 7% who would not choose again, rather than Frost's 19% dissatisfaction) would flip the sign toward inaction, while taking the 9.9% five-year cumulative incidence for the inaction side would widen the gap to about 0.09. The balance is therefore a property of which proxies were chosen, not a finding. The one axis measured comparably on both arms, psychological burden, tilts toward the surveillance side: worry falls after surgery and rises under surveillance, the Gilovich-typical inaction pattern. Against that, prophylactic mastectomy is irreversible and carries body-image, femininity, and reconstruction-complication costs that surveillance does not, which is what keeps the action-side dissatisfaction high enough to offset it. Populations are US, Norwegian, Canadian, and European; none of the regret/satisfaction data is a US surveillance-arm survey. Regret after risk-reducing mastectomy concentrates around reconstruction complications, unmet cosmetic expectations, and decisions that felt externally initiated rather than chosen (Hyldebrandt 2025) — not around the cancer-risk reduction itself, which the surgery reliably delivers.
