Oily skin and sebum: what the research measured, and where it runs out

If your face is shiny by lunchtime, someone has probably told you that you're not cleansing properly. Oil production is controlled by hormones and by the biology of the sebaceous gland itself, and the research on what changes it is thinner than the advice built on it. Here is what the studies measured.

Oil comes from a gland

Sebum is made by sebaceous glands, and a 2019 review in the British Journal of Dermatology sets out what is understood about their control: androgens, progestogens and oestrogens, retinoids, receptor tyrosine kinases (ErbB, FGFR2, insulin and IGF-1 signalling), PPAR-gamma, the aryl hydrocarbon receptor and Wnt signalling. Glands renew continually from progenitor cells, and the glands attached to comedones are shrunken rather than overactive (Clayton et al., 2019, PMID: 31056753). That is a narrative review, not an experiment.

Those controls sit inside the skin, while cleansing acts on oil already at the surface. No study here tested a face wash against sebum production, so that gap is reasoning, not a measured result.

The androgen evidence

Imperato-McGinley and colleagues measured forehead sebum in people with complete androgen insensitivity and found production comparable to children who have not yet been through adrenarche (Imperato-McGinley et al., 1993, PMID: 8381804). The same study cuts against the common claim that DHT is what makes skin oily. Men with inherited 5-alpha-reductase deficiency, who make very little DHT, had normal sebum for their age, and men taking a prescription medicine that suppresses DHT did not drop below their own baseline. Androgen signalling as a whole looks necessary, and a specific dependency on DHT was not demonstrated.

That medicine is prescription-only, a matter for a doctor, and outside what a salon or a cosmetics retailer supplies.

Oilier skin and how pores look

Roh and colleagues measured sebum output and facial pore size in 60 volunteers, then ran multiple linear regression. Sebum output was the factor most strongly associated with pore size, ahead of male sex and age, with reported correlations of r = 0.47 in men and r = 0.38 in women. Pore size was larger during the ovulation phase (P = 0.008), and acne severity was not significantly associated with pore size (Roh et al., 2006, PMID: 17034515). That is an association measured at a point in time; nobody lowered anyone's oil to see whether pores then looked different.

The rebound oil story

You will hear that stripping your skin makes it produce more oil to compensate. We could not find published research that has tested it. That is not the same as it having been disproved. Our searching is not exhaustive, so we don't know of a study either way. There are still reasons not to strip your skin, and they are about comfort, not glands fighting back.

How often people wash

A single-blinded randomised controlled trial in males with mild to moderate acne had participants wash once, twice or four times daily with a mild cleanser for six weeks. The trial did not detect statistically significant differences between the groups. Twice daily was associated with improvement in open comedones and non-inflammatory lesions, while the once-daily group showed increases in redness, papules and inflammatory lesions. The authors concluded that excessive face washing "may not be as culpable as previously thought" (Choi et al., 2006, PMID: 17014635).

The limits travel with that: one trial, males only, a single cleanser, and a sample small enough that a real difference could go undetected. A trial that does not detect a difference has not shown the options to be equivalent. Acne is a medical condition, and a washing routine is not a treatment for it.

What you wash with

Harsh surfactants interact strongly with skin proteins and lipids, which is linked with after-wash tightness, dryness, barrier damage, irritation and itch. Mild surfactants interact minimally with both, and high cleanser pH makes surfactant-induced dryness and irritation worse (Ananthapadmanabhan et al., 2004, PMID: 14728695). That one is an industry narrative review by cleanser formulation scientists, not an independent trial, so it describes a mechanism rather than proving an outcome. No cited study compared formula against frequency, so treating formula as the part that matters is inference.

Alcohol toners

A multicentre study ran patch tests and tandem wash tests with 60 to 100% alcohols against a 0.5% sodium lauryl sulphate positive control. It did not detect significant change in barrier function or redness with the alcohols, skin hydration dropped significantly, and it reported less irritation from alcohol than from detergent washing (Löffler et al., 2007, PMID: 17578437).

That was hand and forearm skin using hand-rub formulations containing emollients, not a leave-on toner on the face. A study that did not detect barrier change there has not cleared high-alcohol facial products, and has not condemned them either.

Niacinamide

Draelos and colleagues ran two trials of 2% niacinamide. In Japan, 100 subjects in a double-blind placebo-controlled study over four weeks showed a significantly lower sebum excretion rate at weeks two and four. In a second study, 30 subjects in a randomised split-face design over six weeks showed a significant reduction in casual sebum level, while sebum excretion rate did not fall significantly (Draelos et al., 2006, PMID: 16766489). Both arms were short, the second was small, the authorship was industry, we can find no replication, and there are no data on persistence after stopping.

Sebum excretion rate is production over time after degreasing; casual sebum level is the oil on the surface at one moment. A product can change how skin looks without anyone having shown it changes what the glands make.

Salicylic acid

An abridged Cochrane systematic review of 49 randomised trials and 3,880 participants looked at several topicals used for acne. On participants' own global assessment it found salicylic acid may make little or no difference compared with a prescription topical medicine (RR 1.00, 95% CI 0.92 to 1.09), on evidence the same review graded low to very low quality, limited by risk of bias and imprecision. That does not establish the two as equivalent. It means the pooled trials were not large or good enough to detect a difference if one exists. Only azelaic acid was graded moderate quality. The review reported adverse events across these topicals as mild and transient (Liu et al., 2020, PMID: 33034949). None of those trials measured sebum or pore size, and the comparator there is a prescription medicine, prescribed and monitored by a doctor.

Retinoids and oil

Two studies disagree here. A Lancet report found a lack of effect of a topical retinoid on sebum excretion rate in acne (Cunliffe et al., 1988, PMID: 2900418). A later study of 39 patients with forehead acne reported that a prescription topical retinoid, applied nightly under the study protocol for seven days, reduced sebum excretion rate compared with vehicle, along with wax esters and free fatty acids (Pan et al., 2017, PMID: 26872139). That second study was small and brief, and the abstract we worked from, which is all we read, does not describe it as randomised or blinded.

Both used prescription-only medicines under medical supervision, which is a doctor's territory and outside what a salon or a retailer supplies.

Where that leaves things

The research is patchy. Hormonal control of sebum is well described, and downstream of that it is associations, short trials and single studies, several industry-authored, with nothing tested long term.

If oiliness comes with persistent breakouts, painful lumps under the skin, a sudden change in how oily your skin is, or changes to hair growth or your cycle, take it to a GP or a dermatologist. Acne is a medical condition, treated medically.

This article is general information, not medical advice, and not a recommendation of any product or service. Cosmetic products are not intended to diagnose, treat, cure or prevent any condition, and what any individual notices varies.

References

1. Imperato-McGinley et al. (1993). The androgen control of sebum production. Studies of subjects with dihydrotestosterone deficiency and complete androgen insensitivity. J Clin Endocrinol Metab. PMID: 8381804
2. Clayton et al. (2019). Homeostasis of the sebaceous gland and mechanisms of acne pathogenesis. Br J Dermatol. PMID: 31056753
3. Roh et al. (2006). Sebum output as a factor contributing to the size of facial pores. Br J Dermatol. PMID: 17034515
4. Choi et al. (2006). A single-blinded, randomized, controlled clinical trial evaluating the effect of face washing on acne vulgaris. Pediatr Dermatol. PMID: 17014635
5. Ananthapadmanabhan et al. (2004). Cleansing without compromise: the impact of cleansers on the skin barrier and the technology of mild cleansing. Dermatol Ther. PMID: 14728695
6. Löffler et al. (2007). How irritant is alcohol? Br J Dermatol. PMID: 17578437
7. Draelos et al. (2006). The effect of 2% niacinamide on facial sebum production. J Cosmet Laser Ther. PMID: 16766489
8. Liu et al. (2020). Evidence-based topical treatments (azelaic acid, salicylic acid, nicotinamide, sulfur, zinc, and fruit acid) for acne: an abridged version of a Cochrane systematic review. J Evid Based Med. PMID: 33034949
9. Cunliffe et al. (1988). Lack of effect of topical retinoic acid on sebum excretion rate in acne. Lancet. PMID: 2900418
10. Pan et al. (2017). A Topical Medication of All-Trans Retinoic Acid Reduces Sebum Excretion Rate in Patients With Forehead Acne. Am J Ther. PMID: 26872139