Phones, Deodorants, Sweeteners, Stress: What the Cancer Evidence Shows

The large studies of mobile phones, underarm antiperspirants and deodorants, and work stress have found no consistent evidence that any of them causes cancer. Artificial sweeteners are less settled: aspartame was classed by the WHO's cancer agency in 2023 as "possibly carcinogenic", on limited evidence, while the WHO's food-additive committee kept its safe intake level unchanged. Much of the worry comes from misreading what that kind of label means.

Worries about everyday things are reasonable. People hear a headline about phones or sweeteners, they want to know whether something they or their children use every day is dangerous, and after a diagnosis many people go back over their lives looking for a reason. This piece sets out what the strongest studies found, how certain they are, and where the uncertainty really lies. "No consistent evidence of harm" is not the same as "proven safe", and we try to say precisely which applies.

Beliefs about unproven causes are common, and they are not a failing of the people who hold them

The survey behind the figures above questioned a representative sample of 1,330 adults in England between January and March 2016 (published 2018). Stress (43%), food additives (42%) and non-ionising electromagnetic frequencies (35%) were the unproven causes most often endorsed^[1]. The authors concluded that "awareness of actual and mythical cancer causes is poor in the general population", and found that only knowledge of established risk factors, not belief in mythical ones, was linked to behaviour such as not smoking^[1]. That is one survey in one country, not a picture of beliefs everywhere.

These beliefs are understandable: the products are used daily by very large numbers of people, stress is something everyone feels, and the scientific language is easy to misread. For the causes that the evidence does link to cancer, and how much of the global total they account for, see our study of preventable cancer causes.

An IARC "possibly carcinogenic" label measures how sure the evidence is, not how big the risk is

Most of the confusion starts with the classifications of the International Agency for Research on Cancer (IARC), the WHO's cancer agency. Its Monographs programme sorts agents into groups by the strength of the evidence that they can cause cancer in humans. In its own words, "IARC classifications reflect the strength of scientific evidence as to whether an agent can cause cancer in humans, but they do not reflect the risk of developing cancer at a given exposure level"^[2].

This is the difference between hazard and risk. A hazard is something that could cause cancer under some circumstances; risk is how likely that is at the levels people are actually exposed to. Risk assessment for food additives is done by a separate body, the Joint FAO/WHO Expert Committee on Food Additives (JECFA)^[2]. Two agents in the same IARC group can carry very different risks, and an agent in a "weaker" group is not necessarily less risky than one in a stronger group: the group says how confident the experts are that a link exists, not how large it is.

As of the list updated on 30 October 2026 (Volumes 1–142), IARC places 135 agents in Group 1 (carcinogenic to humans), 98 in Group 2A (probably carcinogenic), 329 in Group 2B (possibly carcinogenic) and 498 in Group 3 (not classifiable)^[3]. Group 3 does not mean safe; it means the evidence was not enough to decide. And IARC says that for agents it has not classified at all, "no determination of non-carcinogenicity or overall safety should be inferred"^[3].

Group counts from the IARC Monographs list, last updated 30 October 2026^[3]. The short descriptions are our plain-language summary of IARC's group names; the hazard-not-risk principle is IARC's own^[2].

Mobile phones: the largest studies, including a 2024 WHO-commissioned review, found no increase in brain tumour risk

In May 2011 IARC classified radiofrequency electromagnetic fields, the kind of radio waves used by mobile phones, as "possibly carcinogenic to humans", Group 2B^[4][5]. Its working group of 31 scientists from 14 countries judged the human evidence "limited" for glioma and acoustic neuroma among wireless phone users, inadequate for other cancers, and "did not quantitate the risk"^[5]. Many studies of phones and brain tumours are case-control studies, in which people with a tumour and people without one are asked to remember how much they used their phones years earlier. As the Million Women Study authors note, studies that record phone use after diagnosis "are prone to differential reporting"^[4]: people looking for an explanation may recall their use differently.

The largest, the international INTERPHONE study, reported that "overall, no increase in risk of glioma or meningioma was observed with use of mobile phones". Regular users actually had lower odds of glioma (odds ratio 0.81) and meningioma (0.79) than non-regular users, which the authors put down to participation bias or other methodological problems rather than any protective effect^[6]. In the tenth of people reporting the most cumulative call time (1,640 hours or more), the odds of glioma were higher (odds ratio 1.40, 95% CI 1.03 to 1.89). The authors flagged implausible values in that group and concluded that "biases and error prevent a causal interpretation"^[6].

Studies that record phone use before anyone is ill avoid the recall problem. The UK Million Women Study followed 776,156 women, born 1935–1950, for 14 years after they reported their phone use in 2001; 3,268 developed a brain tumour^[4]. The relative risk of glioma for ever-users compared with never-users was 0.89 (95% CI 0.80 to 0.99), and there were no statistically significant associations for daily use or for 10 or more years of use. The authors concluded that their findings support the evidence that phone use "under usual conditions does not increase brain tumor incidence"^[4]. They also note that age- and sex-specific incidence trends "do not suggest any adverse effects due to cellular telephone use"^[4].

The most complete synthesis so far is a systematic review commissioned and part-funded by the WHO and published in 2024. It included 63 studies published between 1994 and 2022 and found mobile phone use was "not associated with an increased risk of glioma" (pooled relative risk 1.01, 95% CI 0.89 to 1.13), with "no observable increase" in risk as cumulative call time rose^[7]. In the studies at lowest risk of bias, the estimate for people who had used phones for 10 years or more was 0.95 (0.85 to 1.05). The reviewers graded the evidence as "moderate certainty" that phone use "likely does not increase the risk of glioma" in adults^[7].

That is a careful conclusion, not a guarantee. Moderate certainty means further research could change the estimate; the grading applies to adults; and IARC's 2011 classification was a statement about the strength of the evidence available at the time, not a measure of risk.

Antiperspirants and deodorants: pooled case-control studies found no association with breast cancer

The worry about underarm products usually centres on breast cancer, and on ingredients such as aluminium salts in antiperspirants. A 2024 meta-analysis pooled seven case-control studies and "found no association between the use of underarm antiperspirants or deodorants and the risk of breast cancer" (odds ratio 0.96, 95% CI 0.78 to 1.17)^[8].

Two limits matter. The studies differed from one another quite a lot (a heterogeneity statistic, I², of 60%), and all seven were case-control studies, which rely on remembered use and carry the same recall problems described above for phones^[8]. The fair summary is that the human studies, taken together, show no consistent link; they do not prove that no effect could ever exist, and no study can.

Artificial sweeteners: aspartame is "possibly carcinogenic" on limited evidence, and the safe intake level was kept

Sweeteners are the one worry in this piece where the evidence is genuinely unsettled, and where two WHO bodies looked at the same question in 2023. In June 2023 an IARC working group classified aspartame as possibly carcinogenic to humans (Group 2B) "on the basis of limited evidence for cancer in humans", "specifically, for hepatocellular carcinoma, which is a type of liver cancer". It also found limited evidence in experimental animals and limited evidence on possible mechanisms^[2][9].

Weeks later JECFA, which assesses risk rather than hazard, "concluded that the data evaluated indicated no sufficient reason to change the previously established acceptable daily intake (ADI) of 0–40 mg/kg body weight"^[2]. The WHO illustrated what that means: an adult weighing 70 kg would need to drink more than 9–14 cans of a diet drink a day to exceed it, assuming 200–300 mg of aspartame per can and no other sources^[2]. The two findings are not a contradiction. IARC said the evidence of a possible hazard is limited; JECFA said that at usual intakes there was not enough reason to change the safe level.

Some of the newer evidence comes from observational cohorts. In France's NutriNet-Santé study of 102,865 adults followed for a median of 7.8 years, higher consumers of artificial sweeteners had a higher rate of cancer overall than non-consumers (hazard ratio 1.13, 95% CI 1.03 to 1.25), as did higher consumers of aspartame (1.15) and acesulfame-K (1.13)^[10]. The authors reported an association, not a cause, and listed "potential selection bias, residual confounding, and reverse causality" among the limitations: people who choose sweeteners may differ in weight, diet and health in ways that are hard to adjust for^[10].

The WHO release quoted a WHO official calling for "better studies with longer follow-up and repeated dietary questionnaires in existing cohorts", and IARC and WHO said they would "continue to monitor new evidence"^[2]. So the honest position is neither "sweeteners cause cancer" nor "sweeteners are proven safe": there is a limited signal, unresolved by the studies so far, alongside a safe intake level that the risk assessors kept.

Stress: in 116,000 European adults, job strain was not linked to cancer risk

Stress is the most widely held of these beliefs, and for many people after a diagnosis it is the most painful, because it can feel like self-blame. The strongest single analysis we found pooled individual data from 12 European cohort studies: 116,056 men and women aged 17 to 70, free of cancer when they joined, followed for a median of 12 years, during which 5,765 cancers were diagnosed^[11].

High job strain was not associated with overall cancer risk (hazard ratio 0.97, 95% CI 0.90 to 1.04). None of the four cancers studied separately showed a statistically significant association: colorectal 1.16, lung 1.17, breast 0.97 and prostate 0.86^[11]. The authors concluded that work-related stress, "measured and defined as job strain, at baseline is unlikely to be an important risk factor" for these cancers^[11].

The limits are real and worth stating. Job strain is one kind of stress, measured once, at work; the study does not cover bereavement, caring, money worries or trauma, and wide confidence intervals for some cancers mean small effects cannot be ruled out. What the evidence does not support is the idea that people brought their cancer on themselves by being stressed. The mental health of people living with cancer is covered in our study of cancer and mental health.

Why unfounded fears matter: they compete for attention with causes that are real

A worry that the evidence does not support is not harmless just because it is unfounded. The 2016 survey in England found that knowing the established risk factors was linked to behaviour such as not smoking, while believing in mythical causes was not^[1]. Misread classifications also spread easily: a "possibly carcinogenic" label sounds like a verdict when it is an invitation to further research^[2]. How false health claims circulate online is covered in our study of cancer misinformation.

What the evidence asks of us

The evidence here points less at individuals than at the people who communicate about cancer risk.

• Explain hazard and risk together. When IARC and JECFA reported on aspartame in 2023, the WHO published both findings side by side and set out what the safe intake level means in cans per day^[2]. That is a model for health agencies, journalists and charities: a classification should never be reported without saying what it does not mean.
• Say "no consistent evidence" when that is what the studies show. Overclaiming safety damages trust as much as overclaiming harm. The phone and stress evidence is strong enough to reassure; the sweetener evidence is not settled, and saying so plainly is more credible than either alarm or dismissal.
• Fund the studies that would settle open questions. The WHO asked for cohorts with longer follow-up and repeated dietary measurement for sweeteners^[2]; prospective designs such as the Million Women Study avoid the recall bias that clouds case-control evidence^[4].
• Close the awareness gap on established causes. Only about half of established causes were recognised in the 2016 survey^[1]. Public information that leads with the causes that are well supported, rather than the fears that are not, is where the population-level gain lies.
• Remove blame from the conversation. No study cited here can say why any one person developed cancer, and the evidence on stress in particular gives no grounds for the self-blame many people feel after a diagnosis.

More on the evidence across the cancer library is at our cancer research hub.

Language: this piece says "people with cancer" and "people living with and after cancer", and describes the illness and its treatment plainly, without combat metaphors. Clinical terms such as "stage", "incidence" and "net survival" are used in the sense the cited source defines them.

Why we're publishing this

People affected by cancer meet these worries constantly: in headlines, in conversations with friends, and in their own search for a reason after a diagnosis. Charities and support groups are often asked about them and need something accurate to point to. We set out what the strongest studies found, with their limits, so that the answer can be honest in both directions: reassuring where the evidence supports reassurance, and plain where a question remains open. Nobody who has held one of these worries should feel foolish for it; the language of cancer classification is genuinely easy to misread.

How to read this data

Most of the evidence here is observational. Case-control studies (INTERPHONE, the antiperspirant studies) compare people with and without cancer and rely on remembered exposure, so recall bias is a central concern; they report odds ratios. Cohort studies (the Million Women Study, NutriNet-Santé, the job-strain cohorts) record exposure before diagnosis and report relative risks or hazard ratios. A ratio of 1.0 means no difference; a confidence interval that includes 1.0 means the result is compatible with no difference. These measures come from different designs and should not be compared directly with one another.

Much of the evidence comes from Europe and other high-income countries: the survey of beliefs is from England only, the phone cohort is UK women born 1935–1950, the sweetener cohort is French, and the stress cohorts are European. IARC classifications describe the strength of evidence that something can cause cancer, not how large any risk is. Finding "no consistent evidence" of an effect is not proof that no effect exists, and a single association in one cohort is not proof that one does.

Use this data

Free to cite with attribution to Better Health For Everyone. Charities, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.

Important: This article summarises published research for general information. It is not medical advice and cannot be used to diagnose, treat or predict anything about an individual. If you have concerns about your own or someone else's health, please speak to a qualified health professional.

Sources

1. Shahab L, McGowan JA, Waller J, Smith SG. Prevalence of beliefs about actual and mythical causes of cancer and their association with socio-demographic and health-related characteristics: findings from a cross-sectional survey in England. European Journal of Cancer, 2018. europepmc.org/article/MED/29705530
2. World Health Organization. Aspartame hazard and risk assessment results released (IARC and JECFA). WHO news release, 14 July 2023. who.int/news/item/14-07-2023-aspartame-hazard-and-risk-assessment-results-released
3. International Agency for Research on Cancer. Agents classified by the IARC Monographs, Volumes 1–142 (last updated 30 October 2026). IARC, 2026. monographs.iarc.who.int/agents-classified-by-the-iarc/
4. Schüz J, et al. Cellular telephone use and the risk of brain tumors: update of the UK Million Women Study. Journal of the National Cancer Institute, 2022;114(5):704–711. europepmc.org/article/MED/35350069
5. International Agency for Research on Cancer. IARC classifies radiofrequency electromagnetic fields as possibly carcinogenic to humans. IARC Press Release No. 208, 31 May 2011. iarc.who.int/wp-content/uploads/2018/07/pr208_E.pdf
6. INTERPHONE Study Group. Brain tumour risk in relation to mobile telephone use: results of the INTERPHONE international case-control study. International Journal of Epidemiology, 2010. europepmc.org/article/MED/20483835
7. Karipidis K, Baaken D, Loney T, et al. The effect of exposure to radiofrequency fields on cancer risk in the general and working population: a systematic review of human observational studies – Part I: most researched outcomes. Environment International, 2024;191:108983. europepmc.org/article/MED/39241333
8. Trinh TTK, et al. Use of antiperspirant products and risk of breast cancer: a meta-analysis of case-control studies. Cancer Investigation, 2024. europepmc.org/article/MED/39324502
9. Riboli E, Beland FA, Lachenmeier DW, et al. Carcinogenicity of aspartame, methyleugenol, and isoeugenol. The Lancet Oncology, 2023;24(8):848–850. europepmc.org/article/MED/37454664
10. Debras C, et al. Artificial sweeteners and cancer risk: results from the NutriNet-Santé population-based cohort study. PLOS Medicine, 2022. europepmc.org/article/MED/35324894
11. Heikkilä K, et al. Work stress and risk of cancer: meta-analysis of 5700 incident cancer events in 116 000 European men and women. BMJ, 2013;346:f165. europepmc.org/article/MED/23393080