The researchers began following people living in Framingham in 1948. Every 2 years, they interviewed and examined participants and reviewed medical records. In this study, the researchers include only participants who were alive in the early 1990s and did not have a heart attack or heart failure. The researchers obtained blood samples from these participants and measured serum IGF-I levels. They continued following the participants for 5 to 9 more years. They then examined whether people who developed heart failure had lower levels of IGF-I than did people without heart failure. To ensure that observed relationships were not due to factors other than IGF-I levels, they did analyses that “controlled for” other factors, such as high blood pressure, diabetes, and high cholesterol, which might be associated with increased risk for heart failure.