Risk Perceptions, Optimism Bias and Information Response: Evidence from the Nueva Ejica Cardiovascular Risk Experiment
SeriesPhD Lunch Seminars
SpeakerJenny Kudymowa (Erasmus University Rotterdam)
LocationErasmus University Rotterdam, Tinbergen Building, Room H09-02
Date and time
December 04, 2018
13:00 - 14:00
Background: Around 80% of premature mortality due to cardiovascular disease (CVD) – the leading cause of death worldwide – could be avoided through changes in lifestyle. However, many are unaware of their exposure to CVD risks. In low- and middle-income countries, less than 45% of individuals with raised blood pressure are aware that they are hypertensive. Provision of information on CVD risks might therefore be expected to reap major health benefits. But not if perceptions of those risks are optimistically biased.
Objectives: (1) To measure accuracy of perceptions of CVD risk in a low-income setting. (2) To determine whether these beliefs are optimistically biased. (3) To evaluate their responsiveness to the receipt of personalized information on CVD risk.
Methods: We collect data on CVD risk factors (blood pressure, body mass index, smoking, age & sex) and elicit perceptions of CVD risk (own risk and average risk) from a representative sample of 3795 individuals aged 40-70 in Nueva Ecija. A (clustered) random sub-sample of respondents is given three types of information: 1) average CVD risk for someone of the same age and sex, 2) CVD risk for someone with the same risk factor profile, and 3) CVD risk for someone with the optimal risk factor profile. CVD risk is the probability of having a heart attack or stroke within 10 years. We assess the accuracy of perceptions by comparing them with individual-specific predictions from measured risk factors. We test for optimism bias using the mean and median difference (and ratio) between perception of own risk and the average risk. We examine the extent risk perceptions are updated using the information provided.
Results: Subjective estimates on personal risk of experiencing a heart attack or a stroke are overestimated by 11 percentage points (ppt) on average and 2 ppt at the median. This is due to overestimation of the base rate, not pessimism bias. In fact, there is consistent evidence of optimism bias. On average, respondents assess their own risk as 14 ppt below the average. There is partial updating of risk perceptions in response to the personalized information. Further, the provision of information on the optimum risk achievable increases the extent to which respondents believe that changes in health behaviour are effective in reducing CVD risk.
Conclusions: Optimism bias can weaken the effectiveness of health campaigns that inform of the average risk. Our study demonstrates that by circumventing this bias, the provision of personalized information can potentially be more effective.