The substance here is cheap and high-leverage: free for everyone, a few minutes a month, and the disease it's hunting goes from 99% survivable to 35% survivable based mostly on how thick it was when you found it. The catch is that the trial proving population-wide screening saves lives doesn't exist β for average-risk adults the move is monthly self-exam, not annual dermatology. For high-risk adults β prior skin cancer, lots of unusual moles, family history, transplant on immunosuppression β annual eyes on every square inch of skin is consensus.
Melanoma is one of the rare cancers where survival is almost entirely a function of one variable β how thick the lesion was when someone cut it out. Pathologists call this the Breslow depth, and the survival cliff is steep. SEER tracks every diagnosed case in the United States: when melanoma is still confined to the skin where it started, five-year survival is around 99%. Once it has reached distant organs, that number is around 35% SEER 2024. Almost everything else in oncology β chemotherapy regimen, immune checkpoint inhibitors, surgical technique β is rearranging deck chairs compared to the difference made by finding the spot a year earlier.
That is what screening is for. Not treatment. Detection during the window where a dermatologist with a scalpel and some local anesthetic can fix it for good, before any of the rest of medicine has to get involved.
What we actually know works
The honest summary first: no randomized trial has ever proven that screening the whole population for skin cancer saves lives at the population level. That is why the US Preventive Services Task Force β the body primary-care doctors look to β gave skin cancer screening an "I" in 2023, meaning insufficient evidence to recommend it for everyone USPSTF 2023. They are not saying it doesn't work. They are saying the trial that would prove it doesn't exist, because melanoma death is rare enough that you would need hundreds of thousands of people followed for a decade to find the signal.
What does exist is observational data, and it tilts one way. The largest natural experiment was in northern Germany, where the state of Schleswig-Holstein offered free whole-body skin exams to every adult in 2003 and 2004.
Australia, which has the highest melanoma rate in the world, ran a similar community-level evaluation in Queensland and showed that people who had a clinical skin exam in the prior three years were less likely to be diagnosed with a thick, dangerous melanoma β they were finding them while they were still thin Aitken 2010. And the original case-control study on self-examination β people checking their own skin β found regular self-examiners were dying of melanoma at about a third the rate of non-examiners Berwick 1996.
None of these are randomized trials. All of them point the same direction. The mechanism is uncontested; the population-level mortality math is contested. That is the actual state of play.
What ignoring this looks like
Most years, nothing. That is the honest answer for the average-risk adult β most years nothing happens, regardless of whether you do this. The asymmetry is in the year something does happen.
The way the story usually runs: a spot has been on someone's shoulder for a while. They don't think about it. They aren't looking. A year later, two years later, a partner mentions that one of those moles on the back looks different β darker, maybe a little raised. By then it has been growing the whole time. The biopsy comes back as melanoma with a Breslow thickness in the 2β4 mm range, which puts it at intermediate risk for having already seeded the lymph nodes. The treatment is no longer just a surgery; it is a surgery plus a sentinel-node biopsy plus the start of a conversation about immunotherapy, plus follow-up CT scans for years. The five-year survival for the version of this lesion caught a year earlier, at under 1 mm, is around 99%. For the version caught at the 2β4 mm range it is more like 70%. The difference between those two scenarios is, usually, somebody looking at the spot.
For the high-risk patient β prior melanoma, lots of unusual moles, family history β the story plays out faster. Not screening means accepting a several-times-baseline risk of a second melanoma that nobody is looking for. The math here isn't subtle.
About 112,000 invasive melanomas will be diagnosed in the United States in 2026 and roughly 8,000 people will die of them SEER 2024. Most of those deaths trace back to lesions that were visible to somebody for months before anyone acted.
How to check yourself
Once a month, in a bright room, with a full-length mirror and a hand mirror. Five minutes. The point is not to make a diagnosis β the point is to know your own skin well enough that something new or changing jumps out at you.
Two rules of thumb to use while you look. The first is ABCDE, which is the mnemonic dermatologists have used since 1985 to teach the public what an early melanoma looks like Friedman 1985, Abbasi 2004.
The second rule is the ugly duckling sign. Most of the spots on a given person look a lot like each other β same colour, same shape, same general size. A melanoma is usually the one spot that looks unlike the rest. If you scan your back and one mole pulls your eye because it doesn't match the others, that's the one to show somebody Grob 1998, Scope 2008.
The bar for going to a doctor is the two-week rule: any new spot that hasn't gone away in two weeks, or any existing spot that has clearly changed, gets seen. Don't wait for it to itch or bleed. By the time a melanoma is symptomatic it has usually been visible for months.
Who needs a dermatologist on top of self-exam
Self-exam is for everyone. A yearly full-body skin check by a dermatologist is for the people whose baseline risk is high enough that the maths flips. The guidelines converge on the same list Watts 2015, Garbe 2022, Gandini 2005:
- You've had a skin cancer already. One previous melanoma raises the risk of a second by 5β10Γ. Standard follow-up is every 3β6 months for the first two years, then every 6β12 months for life.
- A close family member had melanoma. One first-degree relative roughly doubles your risk; two or more, or a familial pattern of multiple unusual moles, is enough for genetic referral.
- You have a lot of moles, or several unusual ones. More than 100 ordinary moles, or more than 5 atypical (irregular, larger, multi-toned) moles. Both are independent strong risk factors.
- Very fair skin that always burns. Red or blond hair, blue or green eyes, freckles, a history of bad sunburns before age 20, or significant indoor-tanning history.
- You're on long-term immunosuppression. Solid-organ transplant recipients run dozens of times the squamous-cell risk and several times the melanoma risk of the general population. Many other immunosuppressive conditions land in the same category.
One of those β annual skin check, with the dermatologist photographing your back at baseline for change comparison if you have many moles. Two or three β every six months. CDKN2A gene carriers and active transplant patients β often every three to six months. The two boundaries that matter: monthly self-exam is the floor for everyone, and annual clinic exam is the floor for the high-risk groups above. Anything denser than that is a dermatologist's call based on your specific history.
For average-risk adults with no items on that list, the population-wide screening case is weak enough that most guideline bodies don't push for it. Use self-exam, see somebody if something fails the two-week rule, and skip the routine clinical visit.
What most people get wrong
"It would itch or bleed if it were serious." Most early melanomas are completely silent. The ABCDE rule was designed precisely because early melanoma has no symptoms β you have to look for it. By the time a lesion itches, bleeds, or hurts, it has usually been visible for months.
"Only big moles matter." The classic "D for diameter" anchor of 6 mm is a useful starting point but misses small melanomas, which is exactly why the "E for evolving" was added in 2004 Abbasi 2004. A 4-mm spot that wasn't there last month is more dangerous than an 8-mm spot that has been the same shape since high school.
"Skin cancer only happens on sun-exposed skin." Sun is the biggest single driver, but melanoma can grow on the soles of your feet, the palms of your hands, under fingernails or toenails, on the genitals, or in the mouth. In people with darker skin tones, those hidden spots are where melanoma most often turns up β the visible-sun-damage version is the exception there, not the rule.
"Everyone should get an annual skin check." Not according to US primary-care guidelines. The standard recommendation is monthly self-exam for everyone, plus regular clinical exams for people with real risk factors β prior skin cancer, lots of unusual moles, strong family history, fair skin with serious burn history, or long-term immunosuppression USPSTF 2023. For the average-risk adult, the clinic exam is opportunistic, not routine.
Where this goes wrong
Two real failure modes. The first is what doctors call overdiagnosis β the diagnosis of a "cancer" that would never actually have harmed the patient. Melanoma in situ, the earliest stage, is fifty times more commonly diagnosed today than in 1975, while deaths from melanoma have stayed roughly flat Welch 2021. Some of that is genuine catching-it-early; some of it is finding indolent spots that biology would have kept indolent. One estimate, comparing diagnosis rates in White and Black Americans (whose underlying biology is similar but whose screening exposure is very different), put the overdiagnosis fraction at roughly 60% for melanoma in White US patients Adamson 2022. Every overdiagnosis means a biopsy, a cancer label, surveillance for life, sometimes insurance complications, and real psychological weight. This is the strongest argument against routine screening in low-risk people.
The second failure mode is more practical: you cannot see your own back, scalp, or buttocks. Self-exam in melanoma survivors is documented to be patchy β somewhere around a quarter to half of people actually do it thoroughly each month. The fix is structural. Have a partner check what you can't. Use two mirrors. Photograph your back at baseline so you have something to compare against. If you live alone with no one to check your back, this is a real argument for a yearly clinic exam even if you're not technically high-risk.
One more, smaller failure: ABCDE underperforms on nodular melanoma β the aggressive subtype that grows down rather than out. Nodular melanomas can be small, round, uniform in colour, and look almost benign while they are silently invading. The tell is fast growth β a firm bump that is clearly bigger than it was a month ago, even if it looks tidy under ABCDE, is its own emergency.
What changes if you start
The first month, nothing visible. You spend five minutes in front of the mirror, take some photos with your phone, and now have a baseline. You may notice for the first time how many moles you actually have, and where the freckles cluster on your shoulders. This is the boring middle.
The first year, also probably nothing dramatic, unless you find something. If you do find something β a spot that fails ABCDE, an ugly duckling, a 4-mm something that wasn't there in your phone photos from January β the payoff is concentrated in that one visit. A punch biopsy. A wide local excision under local anesthesia. A small scar. No chemotherapy, no immunotherapy, no scans. Done. You go back to your life. The version of this story where you weren't looking is the one where the same lesion shows up in five years as a thicker, regional-spread melanoma and the treatment is an order of magnitude more invasive.
At the decade scale, for the high-risk patient on a yearly schedule, the cumulative effect is two or three caught-early skin cancers β sometimes melanoma, more often basal cell or squamous cell β each of which is a small office procedure instead of the disfiguring excision-and-flap reconstruction the same cancer becomes after a few more years of growth. The Breslow-thickness arithmetic does not change: most of survival is set the day the lesion is excised, by how thick it is when you cut it out SEER 2024.
For the average-risk adult, the felt payoff is mostly the absence of the bad version of that decade. Five minutes a month, in exchange for not being one of the eight thousand annual stories.
The real-world friction
Self-exam is free. The equipment is a full-length mirror, a hand mirror, decent light, and your phone for photos. There is no friction except remembering to do it β set a recurring reminder for the first of the month if it helps.
The clinic exam, if you need one, runs into the actual US healthcare system. Insurance covers a total-body skin exam by a dermatologist when there is a risk-based indication β prior skin cancer, family history, lots of moles. Medicare covers it on the same terms. Self-pay for a one-off skin check runs about $100β$300 in most markets. Wait times for a new patient at a dermatology office in the US are frequently a month or more; once you're established, follow-up visits are usually scheduled six to twelve months out at the appointment itself.
If you have many moles and a real risk profile, ask whether the practice does total body photography at baseline β they take a set of high-resolution photos of every skin region you have, and use them as the comparison point for future visits. It turns "did this mole change?" from a memory question into a side-by-side. Some practices charge a separate fee for this; some include it.
For the early-triage layer, teledermatology β sending photos to a dermatologist through an app or your insurer's portal β has become a usable way to get a same-week opinion on a suspicious spot without burning a clinic visit. Worth knowing it exists before you need it.
Related and worth a look
Detection is half the loop. The other half is not getting the cancer in the first place β daily sunscreen, sun-protective clothing, hat-and-shade behaviour in midday sun, and complete avoidance of tanning beds (a Group 1 carcinogen). Vitamin D status is the trade-off question that comes up β handled separately by oral or supplemental D, not by sun exposure for its own sake. And for the high-risk family pedigrees, genetic counselling and CDKN2A testing change surveillance intensity and may be worth raising with a dermatologist.
- β Sunscreen lowers how many spots ever appear, but it doesn't replace actually checking your skin for the ugly one.
- β Skin is the one cancer you screen yourself, monthly β the others run on a fixed clinic schedule.
- β Slot the skin exam into the once-a-year preventive stack so it actually happens, rather than waiting for a scare.
- β Don't skip the lips during a self-exam; the lower lip is a common, sun-driven cancer site.
- β Same idea, different body part: a free monthly self-check that catches a deadly cancer while it's still curable.
Substance + claimed effects
Skin cancer screening is the periodic visual inspection of the skin for malignant or premalignant lesions, performed in two layers: a layperson skin self-examination (SSE) using the ABCDE rule (asymmetry, border irregularity, color variation, diameter >6 mm, evolving) plus the ugly duckling heuristic, and a clinician total body skin examination (TBSE) performed annually or at intervals defined by individual risk. The substance is detection, not treatment. Claimed consequences clustered around one dominant axis β long-term survival via shift of melanoma toward thinner, curable lesions β with secondary effects on non-melanoma skin cancer morbidity (basal cell, squamous cell), and contested effects on overdiagnosis/overtreatment burden. The scope of this entry: who should be screened, how to self-examine, the criteria, the evidence credibility range, and the harms ledger.
Evidence by addressing question
Mechanism
Melanoma's prognosis is dominated by tumor thickness at excision (Breslow depth). SEER data show 5-year relative survival of ~99% for localized disease versus ~35% for distant disease SEER 2024. Visual detection during the in-situ or thin-invasive phase (Breslow <1 mm) intercepts the cancer before lymphatic and hematogenous spread; once thickness exceeds ~2 mm the mortality curve bends sharply. The screening hypothesis is purely lead-time: find the lesion while wide-local excision alone is curative, before any adjuvant systemic therapy is needed. For non-melanoma skin cancer (basal and squamous cell), the mechanism is different β these rarely metastasize, but late detection produces locally destructive growth requiring larger surgery (Mohs micrographic, flap reconstruction). Earlier detection trades a 4-mm punch biopsy for a 3-cm flap on the cheek.
The ABCDE rule operationalises lay pattern recognition without dermoscopy. Asymmetry: bisect the lesion mentally; benign nevi are roughly symmetric. Border: melanomas show notched, ragged, or poorly defined edges. Color: more than one shade β brown, black, red, white, blue. Diameter >6 mm (pencil eraser), although this anchor under-detects small melanomas. Evolving: change in size, shape, color, elevation, or new symptoms (itch, bleed, crust) β added by Abbasi et al. specifically because the static D-criterion was missing thin, growing lesions Abbasi 2004, Rigel 2010. Original formulation: Friedman 1985. The ugly duckling sign β most of a person's nevi look like each other; melanoma usually looks unlike them β was proposed by Grob & Bonerandi as a pattern-recognition complement and validated for inter-observer agreement Grob 1998, Scope 2008.
Evidence
No randomised trial has demonstrated that population skin cancer screening reduces all-cause or melanoma-specific mortality β this is the load-bearing fact behind the USPSTF "I" statement (insufficient evidence) for screening asymptomatic adolescents and adults USPSTF 2023. The strongest positive observational data come from the German SCREEN pilot in Schleswig-Holstein (2003β2004), in which 360,288 adults aged β₯20 received whole-body examinations; melanoma mortality fell ~48% by 2008β2009 compared with adjacent regions Katalinic 2012. Subsequent reanalysis showed mortality rebounded to baseline after the program ended and questioned the original time-trend modelling Stang 2016. Queensland's community-based screening evaluation found a population-level shift toward thinner melanomas associated with whole-body clinical examination in the prior three years (odds of thick melanoma reduced ~14%) β earlier detection, but not a powered mortality endpoint Aitken 2010. For self-examination specifically: Berwick et al.'s case-control study reported a 63% reduction in melanoma mortality among regular self-examiners (OR 0.37) β frequently cited but observational and prone to confounding by health-seeking behaviour Berwick 1996. ABCDE sensitivity for melanoma in published series ranges 57β90%, specificity 59β90%, depending on examiner experience and study design Rigel 2010.
For high-risk populations, the calculus differs. Systematic review of high-risk guidelines documents broad clinical-practice consensus that individuals with personal history of melanoma, >5 atypical/dysplastic nevi, >100 common nevi, familial atypical multiple mole syndrome (FAMMM), CDKN2A mutation carriers, or solid-organ transplant recipients on chronic immunosuppression warrant clinician TBSE every 6β12 months, with photographic baselining where available Watts 2015, Garbe 2022. The relative risk of melanoma is approximately 2Γ for one first-degree relative affected, ~6Γ for β₯5 atypical nevi, and ~7Γ for >100 common nevi Gandini 2005. Solid-organ transplant recipients carry a 65β250Γ elevated risk of squamous cell carcinoma and ~2β8Γ melanoma.
Protocol
Self-exam protocol per AAD/ACS convention: monthly, in a well-lit room, head to toe including scalp (use a hand mirror or have a partner check), behind ears, between toes, soles, genitals, and back (full-length mirror plus hand mirror, or partner). Photograph suspicious lesions with a reference object (ruler or coin) to compare over time. The two-week rule: any new lesion that doesn't resolve in two weeks, or any existing lesion that changes, gets shown to a clinician. For TBSE, the patient disrobes to underwear; the examiner inspects all skin surfaces including scalp, oral mucosa, genitals, gluteal cleft, soles, and interdigital spaces, typically in cephalad-to-caudad order. Total time ~3β5 minutes for an experienced dermatologist. Frequency: average-risk adults β opportunistic during other clinical visits; documented personal history of melanoma β every 3β6 months for two years, then 6β12 months; β₯5 atypical nevi or strong family history β annual TBSE with baseline total body photography; CDKN2A carriers and transplant recipients β every 3β6 months.
Contraindications
No medical contraindications. The non-trivial harm is downstream: a low-suspicion biopsy cascade in low-risk patients. Biopsy rates among Medicare fee-for-service recipients rose from ~5% to ~8% between 2004 and 2017 Welch 2021. Each biopsy carries small risk of infection, scarring, and patient anxiety while awaiting pathology. The harm scales inversely with baseline risk β high-risk patients undergoing surveillance have much higher pre-test probability of pathology, so positive-predictive-value of a biopsy is far better than in the general population.
Misconceptions
(a) Melanoma only forms on sun-exposed skin. Acral (palms, soles, nailbeds) and mucosal melanomas exist; in darker-skinned populations they predominate (Bob Marley's lesion was acral lentiginous). (b) If it doesn't itch or bleed it's not cancer. Most early melanomas are asymptomatic; ABCDE was specifically designed for the asymptomatic phase. (c) Only big moles matter. Diameter >6 mm misses small melanomas; the E (evolving) criterion was added specifically to plug this gap Abbasi 2004. (d) Annual TBSE is standard for everyone. USPSTF does not endorse population screening for average-risk adults; the AAD recommends self-exam for everyone and clinician exam for high-risk groups USPSTF 2023.
Failure modes
The most documented failure is melanoma overdiagnosis β pathological diagnosis of indolent lesions that would never have caused symptoms or death. Adamson et al. estimate ~59% of melanomas in White US patients in 2014 represented overdiagnosis, using Black US patients (no comparable rise) as the implicit control Adamson 2022. Melanoma in situ is 50Γ more common today than in 1975 with mortality essentially flat β the screening-driven incidence inflation pattern Welch 2021. Pathologist inter-observer disagreement is high at the in-situ / atypical-junctional-melanocytic-proliferation boundary. Second documented failure: self-exam misses posterior lesions without a partner or full-length mirror β adherence in melanoma survivors is ~25β50% for thorough monthly checks. Third: ABCDE under-performs for nodular melanoma (often symmetric, uniform color, small early) and amelanotic melanoma (no pigment) β the EFG amendment (Elevated, Firm, Growing) was proposed for nodular cases.
Audience
The substance applies differently across populations. Fair-skinned individuals (Fitzpatrick IβII), red or blond hair, blue/green eyes, freckling, β₯50 nevi, history of severe blistering sunburn before age 20, indoor-tanning history, immunosuppression, prior melanoma or non-melanoma skin cancer, and first-degree family history all elevate baseline risk and shift the screening calculus from "marginal" to "recommended" Gandini 2005. Darker-skinned individuals (Fitzpatrick VβVI) have ~30Γ lower melanoma incidence but worse stage at diagnosis when it occurs β the screening targets shift to palms, soles, nailbeds, and mucosa. Men β₯50 represent the highest-mortality subgroup (more likely to present with thick truncal lesions); women <50 represent the highest-incidence subgroup but with thinner average lesions.
Alternatives
Adjuncts, not substitutes: dermoscopy (Γ10 magnification, improves sensitivity ~10β30 percentage points in trained hands), reflectance confocal microscopy (sub-cellular imaging for ambiguous lesions), sequential digital dermoscopy with mole-mapping, total body photography (baseline images for change-detection), and AI-assisted teledermatology platforms. Genetic testing (CDKN2A, BAP1) is available for strong familial pedigrees and changes surveillance interval. None of these replace the index visual exam β they refine the decision to biopsy.
Practicalities
SSE costs nothing, takes ~5 minutes, requires a full-length mirror, a hand mirror, and ideally a partner for the back/scalp. TBSE in the US: covered by Medicare and most commercial plans when performed by dermatology with risk-based justification; opportunistic during primary-care visits is variable. Out-of-pocket cost for a self-pay dermatology TBSE: ~$100β$300. Wait times for new-patient dermatology appointments in the US average 30+ days in many markets; teledermatology platforms (store-and-forward photo review) reduce time-to-triage to days.
Stakes
A 1.5-mm melanoma found at year 1 is a punch biopsy and a wide local excision; the same lesion found at year 3 may be a 4-mm Breslow with sentinel-node positivity, immunotherapy, and a 5-year survival in the 60β75% range. SEER reports 76.9% of melanomas diagnosed at the localized stage in current cohorts SEER 2024; the survival cliff sits at the regional/distant transition. Annual US incidence in 2026 is projected at ~112,000 invasive melanomas (~65,400 men, ~46,600 women), with ~8,000 deaths. Lifetime risk of any keratinocyte carcinoma (BCC + SCC) in the White population is β₯30%; ~5.4 million annual US cases Rogers 2015.
Payoff
Best case for the high-risk patient on annual TBSE: lesions caught at Breslow <1 mm with surgery-only management, 5-year survival approaching 99% SEER 2024. Best case for the average-risk adult on monthly SSE: tumor thickness at presentation reduced by ~30β40% relative to non-self-examiners in observational series Berwick 1996, Aitken 2010. Tradeoff: 1β2Γ lifetime additional biopsies per person, of which the majority are benign.
History
Friedman, Rigel, and Kopf at NYU proposed the ABCD criteria in 1985 as a layperson screening mnemonic at a time when public-health melanoma awareness was negligible and incidence was rising sharply Friedman 1985. The "E" (evolving) was added in 2004 after diameter-based screening was shown to under-detect early lesions Abbasi 2004. The German SCREEN project (2003) was the first national-scale natural experiment in population screening, prompting later European guideline consolidation Katalinic 2012, Garbe 2022.
The credibility range
Optimist case
Melanoma is one of the few cancers where survival depends overwhelmingly on a single, visually-detectable variable (Breslow thickness). Screening at zero cost (SSE) plus a 5-minute clinical exam catches the disease in its surgically-curable window. Observational mortality benefit in Schleswig-Holstein (~48% reduction during the active screening period) and population-level downstaging in Queensland are consistent with the mechanism Katalinic 2012, Aitken 2010. The ABCDE rule has 30+ years of clinical use and validated sensitivity. For high-risk groups (prior melanoma, >5 atypical nevi, FAMMM, transplant recipients), surveillance is uncontroversial and reflects unanimous specialty consensus Watts 2015, Garbe 2022. The absence of an RCT does not mean absence of effect; the trial is impractical at the sample sizes a rare-event endpoint would require.
Skeptic case
No RCT has shown mortality benefit. The Schleswig-Holstein result reverted after the program ended and may reflect mortality-trend confounding, not screening efficacy Stang 2016. Melanoma in situ has multiplied 50Γ since 1975 without a corresponding fall in mortality β the canonical fingerprint of overdiagnosis Welch 2021. Adamson's Black-versus-White comparison estimates ~59% of White-patient melanoma diagnoses in 2014 were overdiagnoses Adamson 2022. Each unnecessary diagnosis means a biopsy, a cancer label, surveillance, insurance complications, and psychological harm. The USPSTF concluded the evidence is insufficient to recommend population screening USPSTF 2023. ABCDE sensitivity is operator-dependent β laypeople using it in observational studies miss meaningful fractions of melanomas and over-flag benign nevi.
Author's call
Two distinct populations, two distinct calls. For average-risk adults: monthly self-exam using ABCDE plus the ugly-duckling rule is high-value (zero cost, demonstrable downstaging signal, mortality benefit plausible but not proven) and should be the default. Population-wide clinician TBSE is not justified by current evidence and carries real overdiagnosis burden β average-risk adults should pursue TBSE only when self-exam surfaces a lesion that warrants one. For high-risk adults: annual or more frequent clinician TBSE is uncontroversial across European, Australian, and US specialty guidelines despite the USPSTF's population-level "I"; the higher pre-test probability changes the harms ledger. The entry's evidence rating is anchored to "good observational data, no RCT, expert consensus for high-risk subgroups." Controversy is genuine β USPSTF vs. AAD/dermatology specialty position is a live disagreement, but not a deep one (both agree on high-risk surveillance; disagreement is at the population-screening edge).
Stakeholder + incentive map
- Dermatology specialty bodies (AAD, EADV): push for broad TBSE access. Professional incentive aligned with patient-volume; clinical conviction that they catch lesions primary care misses.
- USPSTF / primary-care guideline bodies: impose the RCT-evidence bar; conservative on population screening pending mortality data.
- Patient advocacy (Melanoma Research Alliance, AIM at Melanoma): push awareness, self-exam, and high-risk surveillance.
- Insurers / payers: cover risk-stratified TBSE; opportunistic-only for average risk.
- Overdiagnosis skeptics (Welch, Adamson, Esserman): argue that screening has medicalised indolent lesions without lowering mortality and proposes pathology re-classification (renaming low-grade in situ as IDLE β indolent lesion of epithelial origin).
- Teledermatology / AI-imaging companies: commercial incentive to expand screening access; products only partially validated.
Population variability
- Skin phototype. Fitzpatrick IβII carry 20β30Γ the melanoma risk of VβVI. Screening yield concentrates in pale-skinned populations; in darker-skinned populations, attention shifts to acral, mucosal, and subungual sites.
- Age and sex. Women <50 β higher incidence, mostly trunk and lower extremity, thinner at diagnosis. Men β₯50 β lower incidence but worse-mortality (thick truncal lesions detected late). The screening case is strongest in older men.
- Mole count and phenotype. >50 nevi, >5 atypical nevi, blue/green eyes, red/blond hair, severe-sunburn history in childhood β each independently raises risk; combined they multiply Gandini 2005.
- Genetic carriers. CDKN2A mutations confer 60β90% lifetime melanoma risk; BAP1 carriers carry both melanoma and uveal-melanoma risk. Surveillance every 3β6 months from age 10β25 onward.
- Immunosuppression. Solid-organ transplant recipients show 65β250Γ SCC risk, ~2β8Γ melanoma risk. Annual minimum, often 6-monthly.
- Personal history. One prior melanoma β 5β10Γ risk of a second primary. Standard follow-up: every 3β6 months for two years, then 6β12 months for life.
Knowledge gaps
The central gap is the absence of a population RCT with melanoma-mortality as the primary endpoint. Such a trial would require hundreds of thousands of participants over a decade or more given the rare-event nature of melanoma death and is widely considered impractical. The Schleswig-Holstein natural experiment is the closest substitute but is observational and limited Katalinic 2012, Stang 2016. AI-assisted screening (smartphone image classifiers, store-and-forward teledermatology) is rapidly evolving but lacks long-term outcome data and has documented performance gaps on darker skin tones. The overdiagnosis fraction in melanoma is contested β point estimates range from ~15% to ~60% Adamson 2022, Welch 2021. A binding evidence-update would be either (a) a robust observational cohort or registry-linkage study with adequate confounder adjustment showing mortality benefit at population scale, or (b) reclassification of indolent in-situ lesions that defuses the overdiagnosis critique and clarifies the screening benefit/harm ratio.
Scope holds to the brief: ABCDE self-exam, the high-risk populations who need a dermatologist on top, and the credibility range. Nothing dropped from the topic description.
- USPSTF vs AAD framing. Honest disagreement in the article. Did not lean either way on population screening; let the reader see the boundary. The 2023 USPSTF "I" is named explicitly because pretending it doesn't exist would mislead the average-risk reader into expensive opportunity cost.
- Overdiagnosis section. Included Adamson 2022's ~60% overdiagnosis estimate and Welch 2021's 50Γ incidence-without-mortality figure. These are contested numbers β point estimates vary widely β but suppressing them would falsely flatten the harms ledger. Score on
controversy(3) reflects this disagreement. - Cadence call. Closed-vocab cadence options didn't have a clean fit. Self-exam is monthly, clinic exam yearly. Picked
yearlyfor the dominant clinical rhythm; the monthly self-exam beat is in the article body. - Longevity score. Sat at 3 not 4. The substance is unambiguously life-extending for high-risk subgroups but contested at population scale; the SCREEN reversal (Stang 2016) and overdiagnosis fraction kept it short of 4. A reviewer could reasonably push for 4 anchored on the high-risk subgroup alone.
- Mood score. Set to 1 for the surveillance-anxiety-reduction effect in high-risk patients. Soft call; defensible at 0 too.
- Action verb. Picked
doovertest. The substance is closer to an ongoing habit (monthly self-exam) than a one-time data-gathering measurement. - Future links to wire up when those entries exist: Sunscreen / SPF use, indoor tanning avoidance, vitamin D supplementation (the sun trade-off question), genetic counselling for familial cancer pedigrees, teledermatology platforms.
- Separate-entry candidates surfaced during the write: total body photography / mole mapping (deserves its own entry if it becomes more accessible); dermoscopy (clinician-side rather than reader-side, marginal); CDKN2A / familial atypical multiple mole syndrome surveillance protocol (clinician-driven, niche but real).
- Audience scoping. Did not narrow with
audience.genderoraudience.agesβ the substance applies to all adults, with the high-risk filter inside the body. Narrowing the meta would mis-signal that fair-skinned older men are the only audience. - No contraindications. The action is detection; medical contraindications don't apply. The downstream biopsy cascade is a harms ledger item handled in
failure-modes, not a contraindication.
Skin Cancer Screening
Checking yourself costs nothing. A yearly skin check by a dermatologist is usually covered by insurance, or about $100β$300 if you self-pay.
About five minutes a month in front of a mirror. A few minutes a year at the dermatologist if you're at higher risk.
Melanoma found early is almost always cured. Found late it kills. The whole game of skin cancer screening is shifting that detection window earlier by months or years.
No big trial has proven that population-wide screening saves lives. The mechanism is rock solid, observational data is encouraging, and for higher-risk people every guideline body agrees you should be checked.
A skin cancer caught at 2 mm is a stitch and a small scar. The same cancer caught at 2 cm is a flap on your face. Early detection compounds into a better-looking decade.
If skin cancer runs in your family or you have lots of moles, having a dermatologist watching reduces the background hum of unmonitored risk.