Health information and health behaviours: Does new information on hypertension status matter?

Does the receipt of (negative) health information affect one’s subsequent health behaviours? Using data from clinical health assessments carried out as part of a nationally representative longitudinal study on ageing, a regression discontinuity design is used to identify the effect of health information (i.e., information about hypertension status) on subsequent health behaviours among the older population in Ireland. The results show that while there is no significant impact of new information about hypertension status on smoking, alcohol consumption, physical activity and diet two years later, a significant impact on body mass index (BMI) is identified. New information on hypertension status leads to a 0.7 decrease in BMI and a 7 percentage points decrease in the probability of being overweight or obese. The impact is mainly driven by individuals with higher levels of conscientiousness and without free access to primary health care services. Moreover, compared to females, males are more responsive to their own health information and to a lesser extent, to their spouses’ health information. The findings provide important insights for policymakers tasked with designing public health interventions to enhance diagnosis and management of chronic diseases and promote population health.

Introduction

Health behaviours have important consequences for health and well-being (Grossman, 1972, Cawley and Ruhm, 2011), yet they are difficult to change (Kelly and Barker, 2016). Besides public health campaigns and interventions, do personalized policies work? Utilizing novel Irish longitudinal data, we examine whether the receipt of new personal health information about hypertension status changes individuals’ and their spouses’ subsequent health behaviours, and how personality and access to healthcare influences those changes.

Hypertension, or high blood pressure, is one of the leading risk factors for death and disability. Hypertension and its associated complications have profound impact on individuals, families, health systems and national economies (World Health Organisation, 2023).1 Like many other cardiovascular diseases, although hypertension is preventable and the risk factors are known and modifiable, it is known for its latency. Variations in access to healthcare can influence levels of hypertension awareness in particular (Zhou, Perel, et al., 2021).2 Awareness levels can be low in the absence of population screening or regular opportunistic testing (Williams et al., 2018). People need to make persistent health investments to prevent the development of hypertension (or other chronic cardiovascular diseases) yet often find themselves imperfectly informed of their own health status. The lack of information on own health status does not only hinder disease prevention, but also creates barriers for proper disease management.

There is increasing emphasis on lifestyle modification as an essential component of disease management (National Institute for Health and Care Excellence, 2019, Williams et al., 2018, Unger et al., 2020). While the heritability of high blood pressure is estimated to range between 35 and 50 per cent (Unger et al., 2020), there is considerable uncertainty over the genetic loci associated with hypertension risk (Evangelou et al., 2018). In addition, research has shown that adherence to a healthy lifestyle is associated with lower blood pressure regardless of the underlying genetic risk (Pazoki et al., 2018). Therefore, clinical guidelines for the management of hypertension highlight the importance of changes in health behaviours (i.e., smoking, alcohol consumption, physical activity, diet) in addition to pharmaceutical interventions (National Institute for Health and Care Excellence, 2019, Williams et al., 2018, Unger et al., 2020, Zhou et al., 2021).

In this paper, using data from the Irish Longitudinal Study on Ageing (TILDA), we investigate the impact of the receipt of information about hypertension status on subsequent health behaviours among the older population in Ireland. The Irish context provides a useful setting in which routine opportunistic screening of hypertension is not provided to the majority of the population, and in which the majority must pay the full out-of-pocket cost of general practitioner (GP) care.3 The prevalence level of hypertension among the older population in Ireland is relatively high, yet the awareness level is relatively low. Using data from 2010, nearly two-thirds of the population aged 50 + had hypertension (Murphy et al., 2016). Yet of those with hypertension, only about half of them were aware of their hypertension status. The awareness rates were considerably lower than in other high-income countries (such as Australia, the UK and US) (Zhou, Carrillo-Larco, et al., 2021).4

The biomarkers data collected in objective health assessments in TILDA enables us to employ a regression discontinuity design that exploits the exogeneous cut-off of both systolic and diastolic blood pressure in the diagnosis of hypertension. We find that, while there is no significant impact of new information about hypertension status on smoking, alcohol consumption, physical activity and diet two years later, a significant impact on body mass index (BMI) is identified. New information on hypertension status leads to a 0.7 decrease in BMI and a 7 percentage points decrease in the probability of being overweight or obese. And the impact is mainly driven by individuals with higher levels of conscientiousness and without free access to primary health care services. Moreover, compared to females, males are more responsive to their own health information and to a lesser extent, to their spouses’ health information.

We contribute to the literature in the following ways. First, to the best of our knowledge, this is the first study to examine the impact of personalised health information revealed by biomarkers on both individuals’ and their spouses’ health behaviours in Ireland, a high-income setting without universal access to free primary care and national programme for hypertension screening. Second, given that the blood pressure test results from a nurse-led health examination and the exogenous cut-offs of blood pressure in the diagnosis of hypertension are exploited, the exogeneity of the health information shock is enhanced. Third, with the availability of comprehensive information on health behaviours, we derive an extensive set of behavioural outcomes, including major behavioural risk factors for hypertension and other cardiovascular diseases. Fourth, thanks to the richness of our data, we are able to explore the moderating effects of personality and access to healthcare.

The paper is organised as follows: Section 2 provides an overview of relevant literature, Section 3 discusses the institutional setting, Section 4 introduces the data and methods, Section 5 presents the empirical results and robustness checks. Section 6 discusses the spousal spillovers, while Section 7 discusses the findings and concludes.