Lawrence Garfinkel, an epidemiologist with the American Cancer Society who helped design landmark studies that linked smoking to lung cancer, died last Thursday in Seattle. He was 88. The cause was cardiovascular disease, his son Martin said.

Mr. Garfinkel became a leader in cancer epidemiology despite having no formal education in the field. His college degrees were in statistics: a bachelor’s from the City College of New York and a master’s from Columbia.

“He started as a statistical clerk at the cancer society,” his son said. “He was a recent veteran, just looking for work. It was a temporary job, and he stayed for 43 years.”

Hired in 1947, Mr. Garfinkel learned epidemiology on the job. His mentor was Dr. E. Cuyler Hammond, an epidemiologist and the director of the statistical research section. Scientists had begun to suspect that smoking might cause lung cancer, but large studies were needed to find out for sure. Mr. Garfinkel helped Dr. Hammond and Dr. Daniel Horn conduct a study in the 1950s that tracked nearly 188,000 men for 44 months. Its conclusion became a milestone in epidemiology: Smokers had a marked increase in lung cancer risk.

The grim evidence began to turn the medical profession against tobacco and inspired public health campaigns against smoking. The tobacco industry fought back, picking over studies for flaws and questioning the researchers’ objectivity.

But the epidemiologists had just begun to work. Mr. Garfinkel and Dr. Hammond started an even bigger project in 1959, the Cancer Prevention Study, which enrolled a million men and women.

A study begun in 1982, by Mr. Garfinkel and Dr. Steven D. Stellman, had 1.2 million participants. This research confirmed the earlier indictment of tobacco, and captured the steep rise in lung cancer among women who had taken up smoking. “Those studies have been extraordinarily valuable in that they were a major impetus for tobacco control in the United States,” said Dr. Michael Thun, the vice president of epidemiology and surveillance research at the cancer society.

The studies also gathered data on alcohol, medications, hormones, occupational exposures, reproductive issues, other cancers and chronic diseases. Researchers are still using that information, said Dr. Stellman, a clinical professor of epidemiology at the Mailman School of Public Health at Columbia University.

Mr. Garfinkel and Dr. Hammond also worked with Dr. Oscar Auerbach, a pathologist who studied tissue from deceased smokers and showed that the degree of precancerous change depended on how much and how long they had smoked. The pathology findings bolstered the statistical correlations from the earlier studies by providing physical evidence of what smoke could do to the lungs.

Mr. Garfinkel became head of epidemiology at the cancer society in 1979, retired in 1990 and worked as a volunteer until 2003.

When Mr. Garfinkel retired, Dr. Richard D. Klausner, who was then the director of the National Cancer Institute, said, “Few individuals have contributed as much to our present-day knowledge about the disease consequences of smoking.”

Although Mr. Garfinkel contributed to 147 articles in scientific journals, colleagues say he never flaunted his achievements. “He carried with him the fact that he was the product of two poor immigrants, one of whom never learned to read,” his son said.