Preliminary forecasts predict a close presidential election in 2016, making voter turnout a key factor, especially in battleground states. Voter turnout has been studied for years, but only recently have researchers begun to study how people’s health affects their involvement in politics. Unsurprisingly, healthy people are more likely to go to the polls, even after taking account of factors known to be associated with turnout, such as people’s level of formal education or the strength of their partisan loyalties.

However, our new research considers something different — not just how healthy people are overall, but whether they suffer from specific chronic illnesses. The biggest surprise is that the illness many consider to be the most serious — cancer — is associated with higher turnout.

In our research, published in the Journal of Health Politics, Policy, and Law, we examine the link between voter turnout in the 2008 election and five widespread chronic illnesses: cancer, heart disease, diabetes, asthma, and arthritis. Our data come from a Centers for Disease Control and Prevention survey that was fielded in eight states (Alabama, Alaska, Hawaii, Delaware, Illinois, Louisiana, South Carolina, and Wisconsin) in 2009.

This survey allowed us to take account of many factors related to both voting and health, such as people’s socioeconomic status, whether they had health insurance, how much social and emotional support they received, whether they were experiencing any physical limitations, and how they evaluated their health overall.

We found that three of these diseases — diabetes, asthma, and arthritis — were not related to voter turnout. Heart disease was associated with a lower chance of voting. But cancer was associated with a higher change of voting. Compared with those who did not have cancer, those with cancer were almost 3 points more likely to vote in 2008 — again, even after taking into account many other factors. The graph below shows these findings.

We also found that different groups appear to respond differently to the experience of these chronic diseases. For blacks and those with lower education — two groups who bear a greater burden of disease — a cancer diagnosis was associated with even higher turnout, almost 4 percentage points. This is larger than the cancer-related boost in turnout among whites and those with more education.

Notably, this pattern didn’t occur for every disease. Asthma was linked to higher turnout among the higher educated but lower turnout among the less educated.

Why would cancer be associated with higher turnout? Perhaps the experience of having cancer gives patients and cancer survivors more social resources and richer social networks, in contrast to the experience of heart disease. These factors may be especially consequential for groups — like ethnic minorities or those with less formal education — who typically have less access to these participation-boosting assets.

Here is one possible explanation, although we have no direct evidence for it: in 2008, the American Cancer Society (ACS) worked to mobilize people around health-care reform. The chief fundraising event for the Society is its annual Relay for Life, which involves millions of participants and volunteers in thousands of local communities across the nation.

In 2008, people who attended the Relay for Life also might have had the opportunity to board the “Fight Back Express,” a cross-country bus tour organized by ACS-CAN, the American Cancer Society’s political arm to register voters and recruit activists for the ACS’s policy agenda.

According to ACS-CAN, the “Fight Back Express” went to 512 events in 480 different communities. A similar effort, Cancer Votes, was directed by ACS-CAN in 2012. Most other disease-related nonprofits, in contrast, have neither the long history of grass-roots involvement nor the local visibility of the ACS and other cancer-related advocacy or support groups.

Of course, our study has identified correlations, not necessarily evidence that chronic diseases, including cancer, cause people to vote at a higher or lower rate. That is an important task for future research.

Nevertheless, as we look forward to the 2016 election, our findings suggest the importance of examining the inequalities inherent in both voting and in health. Some chronic illnesses, even cancer, may help to mobilize sufferers and heighten their political power, while other illnesses may have the opposite impact.

Sarah Gollust is Assistant Professor of Health Policy & Management and Wendy Rahn is Professor of Political Science at the University of Minnesota.