The physicians studied asked patients whether they were smokers in about 65 percent of ambulatory office visits ? a figure that remained relatively steady during the years of the study ? 1991 through 1995. However, they reported counseling smokers to quit at only 16 percent of office visits in 1991. Counseling rates nearly doubled to 29 percent in 1993, but the increase was only temporary; rates decreased to 21 percent in 1995. A similar pattern was seen for nicotine replacement therapy. The nicotine patch or gum were prescribed for smokers at only 0.4 percent of 1991 visits, then rose to 2.2 percent of visits in 1993 and fell to 1.3 percent of 1995 visits.

"We were surprised that the rates of counseling were so low," says Anne Thorndike, MD, the study's first author. "It's especially concerning that doctors didn't even know their patient's smoking status one-third of the time and failed to discuss smoking cessation with most of the patients they knew were smokers. Right now a lot of opportunities to help people quit are being missed."

Nancy Rigotti, MD, director of Tobacco Research and Treatment at the MGH and a coauthor of the study, adds, "Doctors should know that counseling about smoking is one of the most cost-effective ways of improving their patients' health."

Smoking cessation treatment guidelines for U.S. physicinas were issued in 1996 by the federal Agency for Health Care Policy and Research. These recommend that physicians ask patients about smoking in every office visit, routinely advise patients to quit and help them reach this goal. "When people go to their doctor, they should expect to be asked about smoking and counseled to quit just as routinely as they have their blood pressure taken," says Thorndike. "Patients who smoke need to be reminded that quitting is the most important thing they can do to stay healthy."

Although physician's practices did improve during the years of the study, the improvement was temporary, peaking in 1993. "That was the year after the nicotine patch first become available by prescription in this country," notes Rigotti. "We suspect that the increase in smoking treatment seen in 1993 is associated with the heavy marketing of the patch to both physicians and the general public."

This study examined data from the National Ambulatory Medical Care Surveys of 1991 through 1995. In the annual survey conducted by the National Center for Health Statistics, physicians complete a form after outpatient visits during a randomly selected week, answering questions about patients' diagnoses and the treatments provided. In addition to the results described above, the researchers found that primary care physicians were more likely than specialists to counsel smokers or prescribe nicotine replacement. All physicians were more likely to address smoking issues in patients with smoking-related illnesses, such as heart or lung disease, than in patients with other diagnoses.

Other coauthors of the study were Daniel Singer, MD, and Randall Stafford MD, PhD. All authors are from the General Medicine Division of the MGH.