In depth
Lupus and environment
Three separate reviews reach almost the identical smoking figure, which is what makes it the firmest thing on this page. Silica at work is the other exposure with real numbers behind it. The rest of this subject is a good deal thinner than the number of papers on it suggests.
What the research found.
One review pooled 13 studies on smoking and lupus, published between 1990 and 2015. Current smokers had higher risk than nonsmokers, at an odds ratio of 1.56 from 1.26 to 1.95. Former smokers did not, at 1.23 from 0.93 to 1.63, on an interval that crosses 1.
One Bayesian meta-analysis covered 12 studies, 3234 people who developed lupus, and 288336 controls between them. Current smoking came out at an odds ratio of 1.54, with a credible interval from 1.06 to 2.25. Former smoking came out at 1.39, from 0.95 to 2.08.
One Boston study compared 95 women with lupus against 191 matched controls. Silica exposure at work lasting more than a year came out at an odds ratio of 4.3, from 1.7 to 11.2, and longer exposure came with higher risk, at a trend P of 0.01. Organic solvents came out at 1.04, from 0.34 to 3.2.
One Canadian study compared 258 people with lupus against 263 matched controls. Odds of lupus rose with the number of breathable silica sources reported, running 1.0, 1.4, and 2.1 for zero, one, and two or more sources, at a trend P under 0.01. Outdoor work in the year before diagnosis came out at 2.0, from 1.1 to 3.8.
One review pooled 10 studies and 6984 people with lupus, looking at heart risk. Current smokers had higher cardiovascular risk than never smokers, at 1.42 from 1.21 to 1.66. Smoking was more common among those who went on to develop heart disease, at 39.28 against 31.36 percent.
One systematic review searched 1354 papers on air pollution and lupus disease activity, and it found three usable studies covering 652 people between them. They differed so much in their methods that no pooled figure was possible. That absence of a figure is itself the finding worth reporting here.
Gilcrease and colleagues, European Journal of Rheumatology, 2020
Two nurse cohorts covered 185824 women and 222 confirmed lupus cases between them. A birth weight of 10 pounds or more came with a rate ratio of 2.7, from 1.2 to 5.9, and being born two or more weeks preterm came with 1.9, from 1.2 to 3.0. Having been breastfed came with 0.8, from 0.6 to 1.1.
The exposure with three reviews behind it
Smoking has the strongest evidence of any exposure in this disease, and what makes it strong isn't the size of the effect. It's that three separate teams asked this question independently, with different methods, different pools of studies, and different statistical machinery. All three came out within a few hundredths of each other.
The first review pooled 13 studies published between 1990 and 2015, and it put current smokers at an odds ratio of 1.56, with the true value somewhere from 1.26 to 1.95. Former smokers came out at 1.23, on an interval ranging from 0.93 to 1.63. An odds ratio of 1.56 means current smokers in those studies had about half again the odds of developing lupus that nonsmokers did.
The second was a Bayesian meta-analysis covering 12 studies, which between them held 3234 people who developed lupus and 288336 controls. Current smoking came out at 1.54, with a credible interval from 1.06 to 2.25. That's a different statistical approach to the first review, built on a different set of assumptions, and it produced a figure two hundredths away from it.
The third pooled 9 case-control studies and put current smokers at 1.49, from 1.06 to 2.08. Former smokers in that review showed no raised risk at all. Three teams, three methods, and three answers between 1.49 and 1.56 is the reason to trust the figure, and it's also the reason not to quote any one of them as though it stood alone.
What quitting does, and where the evidence ends
All three reviews reported former smokers separately and not one found a raised risk, which sounds like a clean answer. It isn't one, and the reason is in the intervals rather than in the numbers people quote from them. One of them ranges from 0.93 to 1.63, another from 0.95 to 2.08, and both of those intervals cross 1 along the way.
An interval crossing 1 means the study can't rule out a real excess in former smokers, and it equally can't rule out a real drop. So these two results can't separate no change from a genuine fall in risk. What they can say is that the excess sits with current smoking rather than with having ever smoked.
The authors of the Bayesian review make that point themselves, calling their former smoker result a trend rather than a finding. That's a careful word and it's the right one here. What to do about a habit is a conversation for your own team, who know your history and your treatment.
Smoking after a diagnosis
Two findings here are about people who already have lupus rather than about who develops it. Neither is a reason to panic and both are worth knowing. They're also the reason smoking is worth raising with your team even if the diagnosis is already made.
One review looked at how treatment performed in smokers, and it found hydroxychloroquine worked worse on skin lesions, at a pooled odds ratio of 0.53. The true value ranges from 0.305 to 0.927, which is an interval that stays below 1 along its whole length. That's a finding about a drug's performance rather than a recommendation about taking it.
Another review pooled 10 studies and 6984 people with lupus to look at heart risk. Current smokers had 1.42 times the cardiovascular risk of never smokers, from 1.21 to 1.66. Smoking was also more common among the people who went on to develop heart disease, at 39.28 percent against 31.36 percent.
Hold that second finding loosely, because the authors rate the included studies as low to moderate quality themselves. They also note that several of the individual results inside the pooled figure were negative. A pooled number built out of a mixed set of studies is a starting point and not a conclusion, which is how the authors themselves read it.
Silica at work
This is the exposure with the second strongest case behind it, and two case-control studies found it on different continents. Both of them found risk rising as exposure rose, which is one of the better signs that a link is real. A dose gradient is harder to produce by accident.
One Boston study compared 95 women with lupus against 191 matched controls, and silica exposure at work lasting more than a year came out at an odds ratio of 4.3. The true value ranges from 1.7 to 11.2, and longer exposure came with higher risk, at a trend P of 0.01. One design feature sets that study apart from most of the literature on this question.
Exposure was rated by two reviewers who didn't know which of the women had lupus and which didn't. That removes a bias almost every study on this subject has to live with, because a person who already has a diagnosis remembers their working life differently. Blinding the rating is the closest thing to a fix available in a case-control design.
One Canadian study compared 258 people with lupus against 263 matched controls, counting how many sources of breathable silica each person reported. The odds ran 1.0, 1.4, and 2.1 for zero sources, one source, and two or more, at a trend P under 0.01. That study also found pottery or ceramics as a hobby at 2.1, from 1.1 to 3.9, counting anybody who'd done it for 26 days or more in total.
So this isn't only a question for heavy industry, and 26 days across a lifetime is a low bar to clear. Anybody who has thrown pots, cut stone, worked with sandblasting, or spent years in construction has a reason to mention it. Your team can't ask about an exposure they don't know you had.
Solvents, separately
Solvents get named alongside silica in almost every summary of this subject, and they're usually named in one breath. One of the studies on this page pulled the two exposures apart and asked about each of them on its own. The answer it got was different enough to deserve a section here.
Occupational solvent exposure came out at an odds ratio of 1.04, on an interval ranging from 0.34 to 3.2. That interval falls squarely across 1, and it's wide enough to be consistent with almost any effect you care to name. Silica in that study came out at 4.3 instead.
Read that as one study, not as a settled answer, because it rests on 95 women living in four Boston neighborhoods. It's still the only study on this page that asked the two questions apart from each other, which is why it's here at all. That makes it worth reporting and not worth leaning on.
Sunlight and outdoor work
One Canadian study asked about outdoor work in the 12 months before diagnosis, and it came out at an odds ratio of 2.0, from 1.1 to 3.8. That interval clears 1 along its whole length, so it's a real finding. The study went on to look at people who blister or rash in midday sun, and among them outdoor work came out at 7.9.
That figure must never be quoted as an eightfold risk. The interval on it ranges from 0.97 to 64.7, which tells you that very few people produced it. It doesn't clear the usual bar either, because the lower end of the interval is below 1.
There's a further problem under this whole section, and it applies to the pollution work too. Exposures were recalled by telephone after diagnosis, which is the weakest way to measure anything. People with a new diagnosis search their own history for a cause, and that's recall bias in its purest form, falling right in the middle of the method.
Air pollution, and how thin this is
The number of papers published on this subject suggests a settled question, and the evidence inside them doesn't. This section is mostly about what hasn't been established yet. That's an unsatisfying thing to read and it's the accurate thing to write.
One systematic review searched 1354 papers on air pollution and lupus disease activity, and it found three usable studies. Between them those three covered 652 people, and they differed so much in method that no pooled figure was possible. The absence of a number is itself the finding.
The studies published since that review are all small ones. One followed nine children with childhood-onset lupus across 108 visits, and the tight intervals inside it reflect repeat measurement on those nine rather than a large sample. Two more from that group measured pollution on the child rather than from a home address, which is a genuine strength.
Neither of those two says how many children produced its 108 measurements. Both tested a lot of outcomes across several lag days without stating a correction for it. When you test enough combinations without correcting for it, some of them come back significant by chance alone, which is why stating the correction is part of the method.
One Brazilian study of 32 women found links to wildfires and to carbon monoxide while finding no link to particulates. Wildfires produce particulates, so that combination is hard to explain in any straightforward way. One Iranian study of 50 people reported that pollution was associated with higher disease activity, and it gives no effect sizes, no confidence intervals, and no P values at all.
So the honest position is that this question is open rather than answered. It isn't that the studies looked and found nothing there. It's that there aren't enough of them, and the ones that exist don't agree with each other.
Your own birth
This section is here because the evidence includes it, and nothing in it can be changed by anybody. That's worth saying at the start of the section, because a reader can otherwise spend three paragraphs looking for something to act on. What follows isn't something anybody can act on, and it's still worth having in front of you.
Two nurse cohorts followed 185824 women who didn't have lupus at the point they enrolled, and over 26 years and 14 years respectively they confirmed 222 cases against formal criteria. The researchers then looked at what those women reported about their own births, decades after the fact. Two limits follow from that and both are worth stating before the numbers.
A birth weight of 10 pounds or more came with a rate ratio of 2.7, against a normal weight of 7 to 8.5 pounds, on an interval ranging from 1.2 to 5.9. Being born two or more weeks early came with 1.9, from 1.2 to 3.0. Having been breastfed came with 0.8, from 0.6 to 1.1, and that interval crosses 1, so the study found nothing there.
The women reported their own births from what their families had told them decades later, which is a long chain for a fact to travel down. The participants are also nurses, and mostly White ones, so this doesn't describe the groups at highest lupus risk. Both of those limits belong with the figures rather than after them.
Measured exposures, not assigned ones
Every study on this page measured an exposure that the people taking part already had in their lives. Not one of them assigned an exposure to anybody, and not one removed an exposure to see what happened next. That difference decides what these figures can mean.
A measured exposure is tangled up with everything else that comes with it. Smoking is tangled up with income, with other exposures, with stress, and with how often somebody gets to see a doctor. The studies adjust for some of that and not one of them can adjust for all of it.
So a link on this page is a reason to look further into something, and it isn't proof that changing the exposure changes the disease. Those are two different claims, and the studies collected here only support the first of them. The trials that would answer the second question haven't been done in this illness yet.
Common misconceptions.
Myth. Quitting takes the risk back down.
Reality. That may well be true and these studies can't show it. All three reviews report former smokers separately, one at 1.23 from 0.93 to 1.63, one at 1.39 from 0.95 to 2.08, and one finding no raised risk at all. Every one of those intervals crosses 1, and they're wide enough to contain a real drop and no change alike, so the honest reading is that the excess sits with current smoking.
Myth. Smoking only affects whether you get lupus.
Reality. It was linked to two other things here as well. One review found hydroxychloroquine worked worse on skin lesions in smokers, at a pooled odds ratio of 0.53 from 0.305 to 0.927. Another pooled 10 studies and 6984 people with lupus, where current smokers had 1.42 times the cardiovascular risk of never smokers, from 1.21 to 1.66, and neither of those is about getting the disease.
Myth. Air pollution is a proven trigger for flares.
Reality. It isn't, and the reason is worth knowing. One review searched 1354 papers and found three usable studies covering 652 people, too unalike to pool. One Brazilian study found links to wildfires and to carbon monoxide while finding no link to particulates, which is hard to explain given that wildfires produce particulates, and one Iranian study of 50 people reported no effect sizes at all.
Myth. Nail salon work raises lupus risk tenfold.
Reality. One study did report an odds ratio of 10.2 for that work, on an interval ranging from 1.3 to 81.5. An interval that wide means very few exposed people produced it, so the point estimate is close to meaningless resting on its own. Read the interval, not the number in front of it. That study's silica gradient is the part worth keeping, and this figure isn't.
Myth. Silica is an old industrial problem.
Reality. One Boston study rated exposure with two reviewers who didn't know who had lupus, where silica exposure longer than a year came out at 4.3, from 1.7 to 11.2. One Canadian study counted reported sources of breathable silica, and the odds ran 1.0, 1.4, and 2.1 for zero, one, and two or more. Pottery or ceramics as a hobby for at least 26 days came out at 2.1, from 1.1 to 3.9.
Cautions specific to this condition.
- Smoking has the strongest evidence of any exposure here, and it's the only one that costs nothing to cut. Raise it with your team if it applies to you.
- Tell your team what you do or did for work. Two separate case-control studies put breathable silica ahead of everything else they measured.
- Read a very wide confidence interval as very few people, not as a large effect. An interval from 1.3 to 81.5 isn't a tenfold risk.
- The birth findings are about your own birth and they can't be changed. They're here because they're real, not because they're a lever.
- One genetic analysis put nitrogen oxides at 7.26, from 2.25 to 23.40. An interval spanning a tenfold range is a question rather than an answer.
- No study here assigned an exposure to anybody. Every one of them measured an exposure people already had, which is a weaker kind of evidence.
Discuss any change with the rheumatologist who manages your care. Nothing here replaces that conversation.
Questions patients ask.
Does smoking cause lupus?
These studies can't answer that particular word. What they show is a link that three reviews agree on, with current smokers coming out at 1.56, at 1.54, and at 1.49 across the three. Every one of those intervals is above 1, and all the underlying studies measured smoking rather than assigning it, so a link is what's established here and cause isn't.
What about people who used to smoke?
All three reviews looked at former smokers separately and not one found a raised risk. One had them at 1.23, from 0.93 to 1.63, and one had 1.39, from 0.95 to 2.08. Those intervals cross 1, so the studies can't tell a real drop from no change, and what they do suggest is that the excess sits with current smoking.
Which workplace exposure has the most behind it?
Breathable silica, by a clear distance from anything else. One Boston study of 95 women found exposure longer than a year at an odds ratio of 4.3, from 1.7 to 11.2, with exposure rated by reviewers who didn't know who had lupus. One Canadian study found odds rising with the number of silica sources reported, at 1.0, 1.4, and 2.1.
Do solvents raise the risk?
The one study here that separated them says no. Occupational solvent exposure came out at an odds ratio of 1.04, from 0.34 to 3.2, on an interval falling squarely across 1. Solvents get named alongside silica a great deal, and this study pulled them apart and found a difference, though it's one study in 95 women and settles less than it sounds.
Does sunlight at work count?
One Canadian study found outdoor work in the year before diagnosis at an odds ratio of 2.0, from 1.1 to 3.8. Among people who blister or rash in midday sun it reported 7.9, on an interval ranging from 0.97 to 64.7, so it doesn't clear the usual bar. That figure must never be quoted as an eightfold risk.
Does air pollution make lupus worse?
It's an open question, and the literature is thinner than it looks from the outside. One review searched 1354 papers and found three usable studies covering 652 people, too unalike to pool. The studies since are small, and one of the most quoted followed nine children, where the tight intervals reflect repeat visits rather than a big sample.
What about how I was born?
Two nurse cohorts followed 185824 women and confirmed 222 lupus cases between them. A birth weight of 10 pounds or more came with a rate ratio of 2.7, from 1.2 to 5.9, and being born two or more weeks early came with 1.9, from 1.2 to 3.0. Having been breastfed came with 0.8, from 0.6 to 1.1, which crosses 1.
Is there anything here that has been tested by changing it?
No, because every study on this page measured an exposure rather than removing one. So not one of them can say what happens if you cut something out of your life. That's different from saying the environment doesn't count, and it means the studies that would answer your question haven't been done yet.
References.
- Jiang F; Li S; Jia C. Smoking and the risk of systemic lupus erythematosus: an updated systematic review and cumulative meta-analysis. Clin Rheumatol. 2015;34:1885-92. 10.1007/s10067-015-3008-9Systematic review and cumulative meta-analysis of 12 published articles reporting 13 case-control or cohort studies from 1990 to 2015 on smoking and systemic lupus erythematosus
- Chua MHY; Ng IAT; W L-Cheung M et al. Association Between Cigarette Smoking and Systemic Lupus Erythematosus: An Updated Multivariate Bayesian Metaanalysis. J Rheumatol. 2020;47:1514-1521. 10.3899/jrheum.190733Updated multivariate Bayesian meta-analysis of 12 eligible studies comprising 3
- Parisis D; Bernier C; Chasset F et al. Impact of tobacco smoking upon disease risk, activity and therapeutic response in systemic lupus erythematosus: A systematic review and meta-analysis. Autoimmun Rev. 2019;18:102393. 10.1016/j.autrev.2019.102393Systematic review and meta-analysis of smoking and systemic lupus erythematosus across disease risk
- Zhang WT; Liu Z; Zhu BC et al. Effects of tobacco smoking on cardiovascular disease in patients with systemic lupus erythematosus: A systematic review and meta-analysis. Front Immunol. 2022;13:967506. 10.3389/fimmu.2022.967506Systematic review and meta-analysis of 10 studies comprising 6
- Finckh A; Cooper GS; Chibnik LB et al. Occupational silica and solvent exposures and risk of systemic lupus erythematosus in urban women. Arthritis Rheum. 2006;54:3648-54. 10.1002/art.22210Case-control study in four predominantly African American neighborhoods in Boston
- Cooper GS; Wither J; Bernatsky S et al. Occupational and environmental exposures and risk of systemic lupus erythematosus: silica, sunlight, solvents. Rheumatology (Oxford). 2010;49:2172-80. 10.1093/rheumatology/keq214Case-control study through the Canadian Network for Improved Outcomes in SLE
- Simard JF; Karlson EW; Costenbader KH et al. Perinatal factors and adult-onset lupus. Arthritis Rheum. 2008;59:1155-61. 10.1002/art.23930Prospective cohort analysis of 87
- Gilcrease GW; Padovan D; Heffler E et al. Is air pollution affecting the disease activity in patients with systemic lupus erythematosus? State of the art and a systematic literature review. Eur J Rheumatol. 2020;7:31-34. 10.5152/eurjrheum.2019.19141Systematic literature review of air pollution and systemic lupus erythematosus disease activity
- Zhang M; Wang Y; Hu S et al. Causal relationships between air pollution and common autoimmune diseases: a two-sample Mendelian randomization study. Sci Rep. 2025;15:135. 10.1038/s41598-024-83880-9Two-sample Mendelian randomisation study using published genome-wide association data
- Alves AGF; de Azevedo Giacomin MF; Braga ALF et al. Influence of air pollution on airway inflammation and disease activity in childhood-systemic lupus erythematosus. Clin Rheumatol. 2018;37:683-690. 10.1007/s10067-017-3893-1Longitudinal panel study across 108 consecutive appointments with childhood-onset systemic lupus erythematosus patients who had no respiratory disease
- Goulart MFG; Alves AGF; Farhat J et al. Influence of air pollution on renal activity in patients with childhood-onset systemic lupus erythematosus. Pediatr Nephrol. 2020;35:1247-1255. 10.1007/s00467-020-04517-3Longitudinal panel study of 108 repeated measures from 9 children with childhood-onset systemic lupus erythematosus
- Farhat SCL; Ejnisman C; Alves AGF et al. Air pollution influence on serum inflammatory interleukins: A prospective study in childhood-onset systemic lupus erythematous patients. Lupus. 2021;30:2268-2275. 10.1177/09612033211061479Longitudinal observational study with 12 repeated measures of serum samples and clinical evaluations
- Blaskievicz PH; Silva AMC; Fernandes V et al. Atmospheric Pollution Exposure Increases Disease Activity of Systemic Lupus Erythematosus. Int J Environ Res Public Health. 2020;17. 10.3390/ijerph17061984Retrospective longitudinal cohort study of lupus patients at the General Hospital in Cuiaba
- Sahebari M; Rajabi E; Esmaili H et al. Effects of Air Pollution on Disease Activity and Health-Related Quality of Life of Systemic Lupus Erythematous Patients: An Iranian Observational Longitudinal Study. Curr Rheumatol Rev. 2023;19:222-229. 10.2174/1573397118666221011121308Observational longitudinal study of 50 lupus patients assessed every two months over six months in Mashhad
- Chen P; Huang J; Li S et al. Nitrogen dioxide and hospital length of stay and cost for systemic lupus erythematosus in Hunan, China. Sci Total Environ. 2023;856:159013. 10.1016/j.scitotenv.2022.159013Study of 11
This page gathers the published research on this subject into one place. The studies behind it were published between 1989 and 2026, and every figure links to the paper it came from. Those studies were peer reviewed. This summary of them was not. Dr. Sarah Luebker is reviewing these pages one at a time and has not reached this one yet, so it carries no medical review date and nothing here is her opinion or her advice to you. Each page gets updated as she reaches it. It is here in the meantime because the science is worth having in one organized place that is easy to find and easy to read. Talk to your own clinician before acting on any of it.
Across the site