Extreme ambient heat is a serious public health threat, especially for the elderly and persons with pre-existing health conditions. Although much of the excess mortality and morbidity associated with extreme heat is preventable, the adoption of effective preventive strategies is limited. The study reported here tested the predictive power of selected components of the Health Belief Model for air-conditioning (AC) use among 238 non-institutionalized middle-aged and older adults with chronic heart failure and/or chronic obstructive pulmonary disease living in Montréal, Canada. Respondents were recruited through clinics (response rate 71%) and interviews were conducted in their homes or by telephone. Results showed that 73% of participants reported having a home air conditioner. The average number of hours spent per 24-hour period in air-conditioned spaces during heat waves was 14.5 hours (SD = 9.4). Exploratory structural equation modeling showed that specific beliefs about the benefits of and drawbacks to AC as well as internal cues to action were predictive of its level of use, whereas the perceived severity of the effects of heat on health was not. The findings are discussed in light of the need to adequately support effective response to extreme heat in this vulnerable population.