Dec 162011


One of life’s great mysteries is finally answered: “living in polluted City XYZ is equivalent to smoking how many cigarettes a day?” OK, it’s not on everyone’s top ten, but I’ve been asked that question many times by patients and by the media — and now I know what to tell them: a day in Beijing is like smoking one sixth of a cigarette. More specifically, on an average day in Beijing an average adult inhales a total of 1.8mg of PM2.5 particles from air pollution, which is 1/6 of the average 12mg of PM2.5 particles inhaled from an average cigarette.  Yes, that’s a very strange number, but if I’ve done the math correctly, it is indeed true. And if it is true…well, it’s surprisingly low, isn’t it? First, let’s walk through the steps so we can all agree on the facts.

Step 1: Read Two Crucial Papers

The last two months have been amazing for anyone interested in public health in China. There has literally been an explosion of air pollution information in all forms of media, not only promoting the US Embassy’s PM2.5 monitor but actively discussing and criticizing many official air pollution stats. In other words, regular Chinese people have finally discovered what many expats in Beijing have already known for three years; the official PM10 daily pollution numbers were not nearly as helpful or as realistic as the US Embassy’s PM2.5 hourly feed.

This wonderful data explosion has culminated with two extraordinarily useful and detailed articles from two of my heroes of public health: Steven Andrews and Doctor C. Arden Pope III. Andrews wrote a dramatic research paper in 2008 discussing official pollution data in China, and last week on Chinadialogue he posted what is easily the most researched analysis to date comparing the US Embassy’s data with official data. It’s a must read.

The second is an even more scholarly article from Dr Pope, who has authored or co-authored most of the most important research on air pollution over the last 20 years. In his latest spectacularly useful research from Environmental Health Perspectives, he analyzes death rates from heart disease and lung cancers caused by smoking and air pollution. The detail is stunningly useful, and it also provided me with the data I needed (finally!) to actually measure mortality risk of air pollution, as well as to answer my original question above.

Step 2: Do The Math

First, just what is an average day in Beijing? We get rightfully scared about these occasional 500′s, but those are rare peaks. From Andrew’s paper, we get the data that the official average PM10 level from 2010 is 121 µg/m³, which converts to an AQI of 84, which is “Moderate” on the US EPA’s guideline. Andrew also did us all an enormous benefit and calculated that the US Embassy’s average daily PM2.5 over the last two years is 100 µg/m³ (which converts to an AQI of 171, “Unhealthy” for the US EPA).

Now we know the daily data. So how much of these particles do we breathe in a day? First, we need to know how much air an average person breathes. This can vary in the research literature, but our esteemed Dr Pope mentions that most studies use an average adult inhalation of 18m³/day of air. Now we can do our calculations. If we use:

  • PM10 data: 121 µg/m³ x 18m³/day (÷1,000 to convert µg to mg) = 2.2mg/day of inhaled PM10
  • PM2.5 data:  100 µg/m³ x 18m³/day (÷1,000 to convert µg to mg) = 1.8mg/day of inhaled PM2.5

Since Pope’s studies mostly use PM2.5 data for comparisons, we will focus mostly on PM2.5 and not the “official” PM10. As you can see, it’s not a stretch to use the US Embassy’s “unofficial” data, as 2.2mg and 1.8mg are relatively close numbers, in my opinion. Also, since it’s estimated that Beijing’s PM2.5 concentration is roughly 85% of PM10, then the numbers mostly are equal.

Step 3: Compare To A Cigarette

Pope also uses the research standard that an average cigarette contains 12mg of inhaled PM2.5. Yes, there are differences in smoking habits (some of us even claim “I didn’t inhale”…) but this seems to be the standard in most research literature.
So if we compare the above numbers:

  • PM2.5 data: 1.8 mg/d air pollution ÷ 12 mg cigarette = 15% of one cigarette

That means that a daily dose of Beijing’s average air pollution is equivalent to 1/6 of one cigarette. Do we agree? If so, let’s continue analyzing this shocking finding.

What About The Really Bad Days?

So what about the really bad days when the US Embassy’s 24-hour data hits “Very Unhealthy” 200; “Hazardous” 300 or “Crazy Bad/Beyond Index” 500? Let’s find out. First, you must convert the AQI to PM2.5 concentration, which is easy to do with the online calculator:

  • AQI200 = 150µg/m³ x 18m³/day (÷1,000) = 2.7mg/d ÷12 = 23% of one cigarette a day
  • AQI300 = 250µg/m³ x 18m³/day (÷1,000) = 4.5mg/d ÷1238% of one cigarette a day
  • AQI500 = 500µg/m³ x 18m³/day (÷1,000) = 9.0mg/d ÷12 = 75% of one cigarette a day

This means that even with an extremely rare average of 500, that’s still equivalent to smoking less than one cigarette a day!

What About Your City?

What about Shanghai, or Mexico City, or LA? Just find out the average daily PM2.5 concentration or AQI and plug in the numbers! For example:

  • City X: PM2.5 concentration =  __ µg/m³ x 18m³/day (÷1,000) = 0.27mg/d ÷12 = __ % of one cigarette a day

So What Does This Mean?

Honestly, I was quite surprised by these numbers (assuming I did the math correctly). I had grown up hearing apocalyptic quotes such as “living in Mexico City is like smoking a pack a day”. Well, that just seems preposterous right now, in any city. According to Pope, a pack a day equals 240mg of PM2.5, which would be equivalent to a daily PM2.5 concentration of 13,333µg/m³, which is essentially impossible! So how on earth did such misleading urban legends start? And how can no scientist or reporter have answered this common question when these stats have been in Pope’s research papers for years?

My next reaction, as I’m sure many of you feel right now, was this: does this mean that air pollution isn’t as serious as we thought? I say no, it doesn’t mean that, but this data certainly puts things in a new perspective.

My Bottom Line

Once the shock wore off, a few major points lingered with me:

  • Yes, this number is a lot lower than I would have assumed. But this absolutely does not mean that air pollution is now less dangerous. Hundreds of studies, many by Dr Pope, have documented the very real health risks of air pollution. And many studies have documented improved community health conditions after their air was cleaned up.
  • But my #1 take-home message is that smoking is an astonishingly toxic and destructive addiction, and is far, far worse than the worst air pollution in the world. Even one cigarette a day is more toxic than a day in any city in the world! And China, with limited public health budgets, would get magnitudes more national health benefit if they focused on eliminating tobacco use when compared to eliminating high air pollution.

There is, of course, a much more important question than the parlor trick above: how much does air pollution increase mortality risk? Find out here in my follow-up article. Here’s a sneak preview of the crucial data from Pope’s article, discussing mortality risks of lung cancer (on the left) and heart disease (on the right) from PM2.5:
mortality air pollution smoking C Arden Pope myhealthbeijing

  29 Responses to “A Day In Beijing Is Like Smoking Only One Sixth Of A Cigarette? It’s Almost…Disappointing!”

  1. An interesting follow-up question would be, “How much PM2.5 is in the 2nd-hand smoke from 1 cigarette?”

    • That IS an interesting followup — and the answer is included in Pope’s study! He details second-hand smoke’s effects in detail. That will make a nice blog piece in the future…but in general, secondhand smoke relative risks are roughly similar to air pollution risks…

  2. I will be living in Shanghai for up to 2 years. From what I have seen, quick research suggests that the pollution may affect young children and older people; however once foreigners leave china, there are not long term adverse health effects from being in China.

    From Atlantic Magazine:How I survived China:

    “The anecdotal evidence that health experts offered to me suggested that most people bounce back if they return to healthier settings. (Longtime expats: watch out!) There is no folklore of a Chinese burden comparable to Gulf War syndrome or Agent Orange, sickening waves of foreigners long after their in-country exposure”

    Is there any clinical research that suggests otherwise?

    Also, I was considering buying an air cleaning device. Do expats buy these devices mostly for personal comfort and not to control how much harmful pollution enters their space?

    I ask, because the US EPA website that discusses air cleaners states the following:

    ” While air cleaning devices may help to control the levels of airborne
    allergens, particles, or, in some cases, gaseous pollutants in a home, they may not decrease adverse health effects from
    indoor air pollutants.”

    So I am wondering why expats always discuss air cleaners. Are they mostly for the alleviation of symptoms and personal comfort? Can they really help with any long term alleviation?

    • Dr. St. Cyr, thanks for another interesting read. Down here in Shanghai, I also get this question a lot but never ran the numbers. Although the PM level here is not as bad as Beijing’s, the PM2.5 is on average 2-4x the EPA 24-hr standard (so typically 70-120 ug/m3). So, we’re probably somewhere around 1/8 of a cigarette!

      Couple other considerations:
      * I’m assuming 18 m3 is the total air breathed in. Our 24-hr average, however, is not the outdoor average. We sometimes are in a bar, when secondhand smoke reaches 200-300ug/m3. On the average, however, we are indoors for up to 80-90% of our time, where the PM is significantly lower. Usually, when I test a home without air filtration devices, the PM levels are typically about 50-70% of outdoors.
      * On the flip side, kids are more susceptible to PM and intake of other inhaled pollutants due to being lower to the ground, where settled particulates are easily stirred up, less developed lungs, and a faster breathing rate. So, for kids, 1/6th probably understates the risk.
      * Although useful as a theoretical calculation, I suspect that suddenly elevated levels of PM have a non-linear impact on our health and well-being. This is based on the times (like a few weekends ago), when PM2.5 levels shot to 400+ for several days. Although this would theoretically only be a half cigarette based on your calculations, many people (myself included), felt a significant impact on our health not usually associated with just cigarette smoke. I’ve inhaled more than my share of cigarettes, and when PM levels are “crazy high”, I feel much worse.
      * Aside from PM, cigarettes have a host of toxic chemicals — benzene, formaldehyde, ammonia, carbon monoxide, among others. Many of these are also in secondhand smoke.

      Incidentally, the second most frequent question I get related to PM is why air quality reporting sites vary so much. I gave some of my thoughts here:

  3. Hi Scott,

    I wouldn’t differentiate between personal comfort and alleviating long-term health impact. The damage that pollutants have on individuals is based on dosage and response. Air filtration reduces dosage (assuming you have the right machine for the right problem) and the response is your own individual sensitivity to that pollutant (ie why some people can live around mold with no problems, while a whiff of the same air may trigger asthma attacks with someone else). So, if you feel better, you are also reducing your total intake, which contributes to long-term health.

    Legitimate air filters work, period. I’m referring to particulate filters. Assuming they have a HEPA filter and are properly sealed, and have a decent fan to circulate air, they reduce particulate matter. A home or office that does not have air filtration and keeps their windows generally closed typically has about 50-70% of the outdoor levels. When one has filtration, however, the inside level stays constant — usually between 5-25 ug/m3 (24-hr standard for healthy air is 35).

    The EPA is more concerned with some of the hokey, gimmicky claims, like UV-light scanners, ozone-generators (which are actually very damaging to lung tissue), and ionizers, that are often more marketing than real science.

  4. This nasal filter company advertises a nasal filter that can respond to PM2.5 particulates.

    Dr, Please comment.

    General facts about SANISPIRA®

    It’s 10mm high and weighs less than one gram. It is simple, practical and well tolerated.

    It provides the ability to breathe and speak freely, not currently permitted for example by fabric masks.

    It’s designed to be used by persons under normal breathing conditions including those undergoing minor muscle strain, as well as cyclists and joggers in the city.

    It’s a non-toxic medical device disposable and can be used for max 8 hours even if it is removed and replaced.

    It protects from PM 10, but also captures the PM 2.5 and PM 1.0 particles, which can reach the finest coronary and bronchial branches. It also protects from seasonal pollen.

    • Interesting concept — but no data provided! Let’s see independent testing first. And again, what about mouth breathing?

      • Unless I am doing strenuous activity, i normally don’t breathe through my mouth.

        For a casual walk to work, I imagine that a nasal hepa filter would be ideal.

  5. I’m almost sure that Tehran is now the most polluted city in the world . I use n95 mask here, but it’s not enough by no way.

  6. Thanks for your analysis. I have been wondering about this issue for awhile as well. I would like to comment in more detail, but my initial question concerns the amount of PM smoke emitted from cigarettes that is actually inhaled. The language in the Pope study: “the estimated sales-weighted average of PM from cigarettes sold in the United States in the 1980s and 1990s was approximately 12–14 mg per cigarette (National Cancer Institute 2001). Because of uncertainty regarding estimates of PM2.5 exposure per cigarette smoked …” seems to be different and perhaps indicate a different analysis than the language that you use in your calculations: “average 12mg of PM2.5 particles inhaled from an average cigarette.”

    I am busy at the moment but a cursory search indicates that the 12mg may be the smoke emitted by the cigarette and not inhaled. [not sure of accuracy of source] “Although a single cigarette is small in size and typically weighs less than 1 gram, a cigarette typically emits between 7 and 23 milligrams (mg) of PM2.5 when it is smoked,”

    Therefore, if only a fraction of the PM emitted from cigarettes is inhaled it would have a very significant impact on your calculations. The Pope study you cite also doesn’t mention filters on cigarettes. Possibly a large effect from filters as well?

    • Pope’s related article in Circulation makes it clear that the 12 mg is the inhaled does and that filters may have no effect. “changes in cigarette composition such as reduced tar or filtered cigarettes may change machine-measured yields but not necessarily the dose because of compensatory changes in human smoking patterns or behavior. The average dose of PM2.5
      from cigarette smoking has been estimated to be from 7 to 17.5 mg per
      cigarette. On the basis of approximate sales-weighted average
      delivery of particulate matter for cigarettes sold in the United States
      from the late 1970s to the early 1990s, the baseline estimates of
      average daily dose in this analysis assume an inhaled dose of 12 mg
      per cigarette.”

      • If on average a cigarette reduces your life by 11 minutes (See ) then 1/6 of a cigarette per day would shorten your life by less than two minutes a day, and even over 70 years that would only amount to about a month of life expectancy reduced for breathing air pollution at Beijing’s current levels on a daily basis. Something doesn’t seem quite right.

        • Yes, it doesn’t really add up, but which data would be incorrect? I think we’re just playing with statistics a bit here. It’s still a fact that the PM2.5 from air pollution comes to less than a cigarette a day, even in the worst cities in the world. Those are numbers from Pope’s own data.

          But even if you look at the risk ratios of pollution and lung cancer and cardiac deaths, most risk ratios are all quite low, under 2.0. I did some calculations for a future post, but here’s some data:
          Beijing’s average PM2.5 AQI of 174 = 1.8mg = lung cancer deaths 1.49 RR (relative risk) and heart disease deaths 1.32 RR
          AQI200 = lung cancer death 1.67 RR, heart disease death 1.35 RR
          AQI300 = lung cancer death 1.98 RR, heart disease death 1.40 RR
          AQI500 = lung cancer death 2.64 RR, heart disease death 1.49 RR

          As you see, the long-term individual risk per person truly is quite low (in my opinion), but the overall population risk is high.

      • Thanks for clearing that up!

  7. Shoot, this blows the theory, then of my friend’s ex-pat friend, who is a cigarette smoker in Beijing. He figured he was better off than non-smoking Beijingers, having the benefit of filtering Beijing air through his cigarettes.

  8. The stuff I cough up after 3 days of 400+ AQI doesn’t even compare to smoking one cigarette. Nor the constant feeling of nausea on those days. I just can’t buy into that assumption.

  9. One thing that is not addressed here is how a lower dose spaced out over a whole day of breathing compares to a high dose over a smoke. Ie how much of that dust and chemical combo actually remains in your lungs given that you have far more air passage over a full day of continuous exposure versus a short sharp one followed by clean air to help eliminate the noxious fumes you have inhaled with your cigarette…

  10. it sounds small to say 1/6th of a cigarette!

    New study out today. I cut and paste the whole thing here

    Children and young adults from areas with highly polluted air in Mexico had physical and genetic changes in their brains akin to those found in adults with Alzheimer’s disease.

    The changes seen are surprising because they are not supposed to occur in younger brains. Alzheimer’s afflicts older adults and those in middle age with specific family genetics. Experts are not sure what causes the onset in older age but think environment might play a role.

    This study builds on a growing body of research suggesting that air pollution exposure can affect the brain. Previous studies have found links between air pollution exposure and signs of inflammation – a basic body response that indicates injury – in dog and mice brains.

    The team of researchers from North America examined 43 brains from children and young adults who died in accidents. More than half were younger than 17 years old and the oldest was 40. Thirty-five cases came from urban, highly polluted Mexico City, while the eight controls were from the rural, unpolluted areas of Tlaxcala and Veracruz. They looked for changes in gene expression, immune markers and physical indicators similar to those associated with Alzheimer’s disease. Results from pollution-exposed city dwellers were compared with those from the rural areas.

    The gene expression analysis showed there were differences between locations in how the genes work. More than 100 genes were changed in the brains from subjects who lived in urban areas when compared to brains from those who lived in the country.

    Over half of the brains from the urban areas showed signs of amyloid-B plaques and 40 percent had pretangle material. In contrast, none of the brains from the rural areas had either condition. Amyloid-B plaques are protein deposits commonly found in the brains of person with Alzheimer’s. Pretangle material is often a precursor and is also associated with the disease.

    Since the APO-E gene is linked with early-onset Alzheimer’s disease, the authors also examined whether the risk version of this gene was linked with similar Alzheimer’s pathology. They found that persons with the APO-E risk genotype were more likely to have the plaques commonly linked with Alzheimer’s in the brain.

    While air pollution was not directly measured, monitoring data show Mexico City’s air quality consistently ranks as some of the worst in the world. Residents, such as those in the study, would be exposed to traffic and industrial pollution throughout their lives.

    Many questions remain unanswered, say the authors, who suggest long-term and comprehensive studies “into the association between air pollution exposures and CNS damage in children is of pressing importance for public health.”

  11. Someone thinks this story is fantastic…

    This story was submitted to Hao Hao Report – a collection of China’s best stories and blog posts. If you like this story, be sure to go vote for it….

  12. [...] and Xi’an) in 2012 alone. (Yet, one expatriate doctor living in Beijing provocatively contends that even at an API of 500, the PM2.5 particulates that one would inhale would be roughly [...]

  13. [...] up with ways to cope is a cottage industry here. One Beijing based doctor, a foreigner, has even calculated just how a PM2.5 500 day (normal in Beijing) compares with smoking. Cutting to the chase, a day spent in those conditions equals smoking 1/6th of a cigarette in [...]

  14. Do you disregard the other pollutants in the air or a cigarette because they’re comparable, and so focus on PM2.5 for simplicity’s sake?

  15. [...] air pollution in any city in China. The average air pollution in Beijing is equivalent to only one sixth of one cigarette a [...]

  16. […] near a pack a day. Still after more than 9 years here, that makes it  over 450 cigarettes for me! (Read the full article here.) As the doc points out, however, knowing how awful the air is here, this makes it clear just how […]

  17. Not all PM are equals, your demonstration is just as silly as comparing drinking gazoline, baiju, water and milk because you cand rink them.

 Leave a Reply



You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>