How Many Americans Smoke?

Surprisingly, the answer is either 45.3 million or 52.7 million in 2010, depending on which federal agency does the counting. This wide variance underscores the discordant findings from government smoking data.

National statistics on smoking are provided by two federal surveys, the National Health Interview Survey (NHIS) and the National Survey on Drug Use and Health (NSDUH). I have conducted research on the differences between these two surveys (here) and I previously wrote a blog on the topic (here).

Federal officials released fresh data from both surveys last week, making the discrepancies even more stark.

The Office on Smoking and Health at the Centers for Disease Control and Prevention (CDC) counted 45.3 million American smokers, using the NHIS (available here). This number has not changed much in over 20 years. In 1990, the CDC reported that there were 45.8 million adult smokers. But as the U.S. population has grown, the percentage of smokers has declined. In 1990, smokers made up 25.5% of the adult population; by 2010, prevalence had declined to 19.3%.

The CDC notes that, “If current patterns continue, smoking prevalence is projected to fall to approximately 17% in 2020, and the national Healthy People objective to reduce smoking prevalence to < 12% will not be met.” That’s an understatement. In 1998, the government set 2010 as the target for 12% smoking prevalence (available here). That proved to be a pipedream, and the 2020 goal will prove equally unattainable. One is reminded of Albert Einstein’s definition of insanity: doing the same thing over and over again and expecting different results.

Last week, the Substance Abuse and Mental Health Services Administration, which administers NSDUH, released “selected” 2010 survey information (here). They estimated the number of smokers at a whopping 70 million, including teenagers, but they did not break out the number of adult smokers. The 2009 NSDUH survey reported 52.7 million adult smokers; the 2010 estimate likely will be similar.

It is unacceptable that two federal surveys differ by over 7 million in their adult smoking counts. Even worse is the way the government uses the divergent data to spin different stories about smoking. They use the lower NHIS numbers to boast about declining smoking rates, which they attribute to higher taxes and smoking bans. They use the higher NSDUH numbers to argue for even more onerous anti-tobacco measures.

My research (here) has explored why the NHIS and NSDUH estimates are so divergent. I have called for officials to investigate and resolve these problems so that national tobacco policy can be guided by accurate smoking data.