SLIP, SLOP, SUBSIDISE: SHOW US THE EVIDENCE

By Ian Wishart

There is a familiar rhythm to public health advocacy. Identify a frightening disease. Produce a survey showing people want something done. Find an expert willing to recommend government spending. Then present the bill to taxpayers as though the scientific and economic arguments have already been settled.

The New Zealand Herald’s October 7 editorial on subsidising sunscreen follows that rhythm. New Zealand has high melanoma rates. People get sunburnt. Sunscreen costs money. Therefore government should make it cheaper, because prevention is cheaper than cure.

Missing from this apparently seamless argument is the question that should come before the spending: how strong is the evidence that the proposed intervention will prevent the disease invoked to justify it?

Preventing sunburn, preventing melanoma and preventing melanoma deaths are different outcomes. Demonstrating one does not automatically establish the others. Nor does showing that a product can work under controlled conditions establish what happens when millions of people use it imperfectly and change their behaviour because they feel protected.

I explored these distinctions in my 2012 book, Vitamin D: Is This the Miracle Vitamin?, and revisited them in Show Me the Money, Honey in 2016. The references were scientific papers and medical evidence reviews. The questions have outlived the books’ publication dates.

Consider the current health-professional review from America’s National Cancer Institute. It places sunscreen among skin-cancer prevention interventions for which evidence of benefit is inadequate. Its assessment acknowledges prevention of sunburn and actinic keratoses, but describes inconsistent trial evidence for squamous-cell carcinoma and no demonstrated trial benefit for melanoma in its overview.[1]

That is considerably more cautious than the impression created by the Herald and the Cancer Society. It also requires careful interpretation: inadequate evidence does not prove zero benefit. It means the certainty advertised to the public is stronger than this major evidence review considers justified.

The frequently cited Australian Nambour study illustrates the problem. After participants were assigned to daily or discretionary sunscreen use, subsequent follow-up found 11 melanomas in the daily-use group against 22 in the comparison group. That looks encouraging. The overall result, however, narrowly missed conventional statistical significance: p=0.051. Invasive melanomas were significantly fewer, but numbered just three versus eleven.[2]

These are signals worth investigating, not figures to dismiss. Equally, they are a slender foundation for assuming the scale of melanoma protection, the reduction in mortality, or the financial return from a nationwide sunscreen subsidy. The NCI notes uncertainty in the post-trial follow-up, including wide confidence intervals.

Now consider what the bottle actually promises.

SPF is determined by the product’s ability to delay sunburn under specified testing conditions. It is not a certificate declaring that no damaging radiation reaches your skin. Broad-spectrum products provide UVA protection too, but protection depends on the formulation, application thickness and coverage. The label cannot apply the lotion for you.[3]

My 2012 discussion emphasised the historical mismatch between UVB protection and UVA protection. UVB drives much of the burning response, while UVA penetrates more deeply. Both can damage biological structures. A lotion that suppresses redness more effectively than it suppresses every damaging process can create a dangerously reassuring experience.

You can see the absence of burning. You cannot see individual DNA lesions.

The possibility matters because consumers may treat the absence of pain as permission to stay outside. A sunscreen can reduce the radiation reaching skin per minute, yet encourage additional minutes of exposure. The net outcome depends on both quantities. A protective product and an unprotective habit can coexist.

That concern has experimental support. In a double-blind trial published in 1999, Philippe Autier and colleagues randomly assigned young French and Swiss holidaymakers to SPF10 or SPF30 sunscreen. Neither participants nor medical personnel knew which preparation they received. The higher-SPF group accumulated 72.6 hours of sun exposure, compared with 58.2 hours in the lower-SPF group: roughly a quarter more. Sunburn experience did not differ.[4]

A subsequent trial using personal radiation dosimeters likewise suggested sunscreen could extend intentional exposure below the sunburn threshold.[5] These were small studies of young European sunbathers using older products. They do not establish that every sunscreen user behaves this way, or that today’s products increase melanoma. They do establish that the behavioural concern is more than an armchair hypothesis.

There is counter-evidence that must also be acknowledged. A 2018 human study found high-SPF sunscreen significantly reduced DNA damage associated with non-melanoma skin cancers at adequate application thicknesses, even with substantial experimental UV exposure. Results at a thinner, more typical application were less consistent.[3]

The same study noted however that as of 2018 “There are no data on DNA protection by sunscreens under typical conditions of use.” That’s a stunning admission for a multi-billion dollar industry that’s been at the centre of summer public health messaging for decades.

It would therefore be wrong to claim sunscreen merely switches off the alarm while offering no biological protection. The defensible warning is that a quieter alarm does not certify an undamaged building. Redness can be prevented while some radiation still penetrates, and longer exposure or inadequate application can undermine protection. Subsidising purchases does not automatically solve either problem.

Preventing sunburn is only one entry in the safety ledger. A thick coating of blue paint could equally reduce sunburn redness while creating a different biological problem; its ability to block radiation would not settle its toxicity. Sunscreen ingredients must therefore be assessed for what they do inside the body as well as what they prevent at its surface. Regulators have identified systemic absorption, potential endocrine effects and unresolved safety questions for particular filters. None establishes that sunscreen is driving rising cancer rates among younger people. But none belongs outside the discussion when taxpayers are being asked to finance greater exposure. The public deserves evidence of net benefit, with uncertainty disclosed—not a single favourable endpoint presented as the whole balance sheet.

Vitamin D adds another complication the editorial overlooks.

The 2025 Australian Sun-D randomised trial assigned 639 adults to routine SPF50+ use or discretionary sunscreen use for approximately a year. The routine-use group had a smaller increase in blood vitamin D. At the final measurement, 45.7% met the study’s deficiency definition, compared with 36.9% in the control group. Deficiency was an exploratory outcome; the measured difference in vitamin D concentration was the primary outcome.[6]

That does not prove sunscreen causes cancer through vitamin D depletion. Nor does it justify treating unprotected sunbathing as cancer therapy. It does undermine blanket reassurance that routine sunscreen use cannot affect vitamin D status. A serious prevention policy should accommodate that finding rather than omit inconvenient parts of the balance sheet.

The Herald’s survey evidence needs similar scrutiny. Its 64% sunburn figure concerns September 2024 to March or April 2025, not the summer immediately preceding October 2026. Its matching 89% affordability figure comes from a separate survey conducted later in 2025.[7][8]

That second survey found 89.1% supported making sunscreen more affordable, including people expressing only slight support. Affordability is a broad objective. It does not, by itself, establish public endorsement of a particular subsidy mechanism, budget or tax trade-off.

Nor does agreement that sunscreen is expensive establish that price is the principal cause of inadequate protection. The earlier survey’s authors reported no strong pattern between household income and judging sunscreen too expensive. Their findings raise questions about affordability; they do not answer all the questions about behaviour.[7]

Meanwhile, the national mortality picture is more encouraging than the editorial’s framing suggests. EHINZ’s March 2026 report records 333 melanoma deaths in 2021, compared with 378 in 2015. The age-standardised mortality rate fell 33% from its 2011 peak of 5.3 per 100,000 to 3.5 in 2021. The report identifies improved treatment as a possible contributor.[9] Is another factor in that drop higher vitamin D levels as public awareness rose? It was a thesis I put forward in 2012: 

“So far, sun tans are looking a lot more protective against deadly skin cancers than anything humans have devised in the sunsmart arsenal. Maybe the natural response was the best after all.

“After all, Cancer Research UK has announced “Over the last 25 years, rates of malignant melanoma in Britain have risen faster than any of the most common cancers.” Melanoma’s drag race up the fatality charts, then, has coincided with a similar mass uptake of sunscreens over the same period.”

High international mortality rankings can coexist with falling domestic mortality. Readers deserve both facts. Falling deaths cannot simply be credited to sunscreen; high rankings cannot establish that cheaper sunscreen is the best next investment.

Before taxpayers subsidise the product, advocates should specify the outcome they expect to buy. Fewer painful burns? Fewer skin cancers of particular types? Fewer melanoma deaths? Each requires its own evidence and its own economic calculation.

A credible proposal would estimate additional correct use, account for longer exposure, compare alternatives, and publish its assumptions. Removing GST would cut a fully passed-through GST-inclusive price by about 13%, not 15%. Whether that changes behaviour enough to justify lost revenue remains a question.

There is also a distinction between cost-effective and cost-saving. Spending public money to improve health can be worthwhile even when treatment savings never recover the outlay. But that is a decision about priorities, not an automatic dividend. Every dollar committed here is unavailable elsewhere. A subsidy reaching existing buyers may chiefly transfer their private spending to the public purse, with little additional protection. A targeted scheme could perform differently from universal tax relief. None of this rules out assistance. It means the design matters, the counterfactual matters, and an opinion poll cannot calculate the health gained per dollar. Those are precisely the details the editorial leaves for somebody else to establish before purchase.

Public support is relevant to politics. It cannot substitute for clinical efficacy or cost-effectiveness. The Herald is entitled to advocate a subsidy. Taxpayers are entitled to ask it to show its working.

The scientific literature contains encouragement, uncertainty and a warning about false reassurance. An editorial asking the public to pay should contain all three. Until it does, “slip, slop, subsidise” is a campaign slogan awaiting a demonstrated public return.

Sources

1. National Cancer Institute. Skin Cancer Prevention (PDQ), Health Professional Version. https://www.cancer.gov/types/skin/hp/skin-prevention-pdq

2. Green et al. Reduced melanoma after regular sunscreen use: randomized trial follow-up (2011). https://doi.org/10.1200/JCO.2010.28.7078

3. Young et al. Sub-optimal Application of a High SPF Sunscreen Prevents Epidermal DNA Damage in Vivo (2018). https://doi.org/10.2340/00015555-2992

4. Autier et al. Sunscreen use and duration of sun exposure: a double-blind, randomized trial (1999). https://doi.org/10.1093/jnci/91.15.1304

5. Autier et al. Sunscreen use and intentional exposure to ultraviolet A and B radiation (2000). https://doi.org/10.1054/bjoc.2000.1429

6. Tran et al. Effect of daily sunscreen application on vitamin D: Sun-D Trial (2025). https://doi.org/10.1093/bjd/ljaf310

7. SunSmart in Aotearoa New Zealand: Survey Findings (2025). https://www.cancer.org.nz/assets/SunSmart-in-Aotearoa-Survey-Findings-2025.pdf

8. McNoe et al. Public support for government-led skin cancer prevention (2026). https://www.nzmj.org.nz/journal/vol-139-no-1641/public-support-for-government-led-skin-cancer-prevention-in-aotearoa-new-zealand-findings-from-a-national-survey

9. EHINZ. Melanoma mortality (March 2026). https://www.ehinz.ac.nz/assets/Surveillance-reports/Released_2026/Melanoma-mortality-2001-21-final-18032026.pdf

Background: supplied extracts from Ian Wishart, Vitamin D: Is This the Miracle Vitamin? (2012) and Show Me the Money, Honey (2016).

Editorial under review: NZ Herald, 7 October 2026, “There is a case for subsidising sunscreen in New Zealand”