People with type 1 diabetes may face a much higher risk of bladder cancer than previously believed. A new systematic review and meta-analysis from the Keck School of Medicine of USC shows that these individuals are 4.29 times more likely to develop the disease. This is the first analysis to properly separate the effects of smoking—a major cause of bladder cancer—from the effects of type 1 diabetes itself.
Bladder cancer affects about 2% of people in the United States. Cigarette smoking causes roughly half of all cases in the U.S. and Europe, but the causes of the remaining cases remain unclear. Previous research has suggested that type 2 diabetes may raise bladder cancer risk, but studies over many years found no such association for type 1 diabetes.
The new study explains why. Earlier research did not control for smoking, even though smoking has a strong influence on bladder cancer risk. Without controlling for this factor, it was difficult to identify other contributors.
Victoria K. Cortessis, Ph.D., the study’s senior author and a clinical professor of population and public health sciences at USC, said the team reviewed nine earlier studies and one meta-analysis. “None had adequately accounted for smoking,” she explained. “Once we addressed that issue, we saw a clear pattern showing a much higher bladder cancer risk among people with type 1 diabetes.”
These results offer important insight into possible causes of bladder cancer and may help guide medical decisions for people living with type 1 diabetes. For example, avoiding or quitting smoking may be especially important for this group, Cortessis noted.
For decades, researchers tracked patients with type 1 diabetes to understand how the condition affected long-term health. These early datasets focused on medical complications and hospitalizations. They rarely included complete smoking data. Sometimes smoking habits were recorded only once and never updated.
Years later, scientists used the same datasets to study whether type 1 diabetes increased the risk of diseases such as bladder cancer. They compared cancer rates in people with type 1 diabetes to rates in the general population. Those studies found no higher risk—but the comparisons were incomplete.
Because smoking is a key cause of bladder cancer, data on smoking are essential. Without accurate smoking information from both groups, researchers could not tell whether differences in smoking behavior were influencing the results.
The USC team suspected that people with type 1 diabetes might smoke less than the general population. Living with a serious lifelong condition could cause them to make more health-conscious choices. If true, their lower smoking rates could hide any increased risk caused by diabetes itself.
To test this idea, first author Helena Oskoui Bennett, MPH, used World Health Organization and other authoritative data sources to estimate smoking rates in the populations included in the original studies. Using a technique called meta-regression, the team discovered that earlier results were indeed affected by differences in smoking patterns.
After adjusting for smoking, the researchers found a strong link between type 1 diabetes and bladder cancer, estimating a 4.29-fold increased risk.
Researchers still do not know exactly why type 1 diabetes may raise bladder cancer risk. One theory is that the disease may create biological changes that contribute to cancer development. Managing type 1 diabetes carefully—including maintaining healthy blood sugar levels—may help reduce this risk.
Smoking could also interact with diabetes-related changes to increase risk even further. “There may be a biological synergy between diabetes and smoking,” Cortessis said. “More work is needed to understand how these factors combine.”
This study is part of a broader effort by the team to understand how both type 1 and type 2 diabetes influence bladder cancer risk. Researchers are now reviewing nearly 100 studies on type 2 diabetes and its treatments to uncover what drives the increased risk observed in those patients.
Related topics
