Quasi‐Experimental Evidence on the Effects of Health Information on Preventive Behaviour in Europe
| Published date | 01 December 2016 |
| Date | 01 December 2016 |
| DOI | http://doi.org/10.1111/obes.12134 |
| Author | Ansgar Wuebker,Vincenzo Carrieri |
765
©2016 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd.
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 78, 6 (2016) 0305–9049
doi: 10.1111/obes.12134
Quasi-Experimental Evidence on the Effects of Health
Information on Preventive Behaviour in Europe*
Vincenzo Carrieri†,‡ and Ansgar Wuebker‡,§
†Department of Economics and Statistics, Via Giovanni Paolo II, 84084 Fisciano, (SA),
Italy, (e-mail: vcarrieri@unisa.it)
‡Health Econometrics Data Group, University of York, Heslington, York, YO10 5DD, UK
§Rheinisch Wesff¨alisches Institut f¨ur Wirtschaftsforschung, Department of “Health
Economics”, Hohenzollernstraße 1-3, D-45128, Essen, Germany (e-mail:
ansgar.wuebker@rwi-essen.de)
Abstract
We investigate the effect of information on preventive decisions in a quasi-experimental
setting arising from the implementation of local breast cancer screening programmes in
Europe. To identify the causal effect of invitation on preventive uptake, welink administra-
tive public data on regional screening policies to individualdata from the Sur veyof Health
Ageing and Retirement in Europe (SHARE) and exploit regional variation in the availabil-
ity of screening policies and in age eligibility criteria. We find home invitation increases
mammography uptakes by around 24%. Significant effects are found when at least 50% of
the population is reached by the invitation letter. The stock of health information and the
ability to process the information received seem to play a significant role, as the effects of
invitation are higher among loweducated and lower among cognitivelyimpaired women.
I. Introduction
Since Grossman’s model of health investment (Grossman, 1972) and Cropper’s(1977) and
Phelps’s (1978) extensions accounting for preventive care as a specific input, economic
studies of healthcare demand consider the marginal benefits of consuming healthcare as the
key determinant of the decision to invest in health. Health information plays an important
role in such decisions by affecting consumers’ perceived marginal benefits of healthcare
(Kenkel, 1990). More generally, the provision of information around health risks seems to
induce precautionary behaviours in individuals, in a manner consistent with the predictions
JEL Classification numbers: C10, I11, I14, I18
*The authors are grateful toAndrew Jones, Hendrik J ¨urges, Owen O’Donnell, Nicola Persico and Richard Cookson
for useful comments. Wealso thank all participants at HEDG seminar held in York (October 2014) and all participants
at Italian Health Economics Annual Meeting held in Trento (October 2013), the participants at Centre of health
Economics Research (COHERE) seminar held in Odense (December 2013), the participants at the 6. Jahrestagung
der Deutschen Gesellschaft f¨ur Gesundheits¨okonomie held in Munich (March 2014), the participants at the Augsburg
Workshop on the Economics of Health Behaviorsand Prevention (November 2013), and the participants at the PhD
Seminar on Health Economics and Policy in Grindelwald(January 2014) for helpful comments and suggestions. We
thank Maryna Ivets, Ieva Scriubaite and Jan Thom for proofreading the paper. All remaining errors are our own.
766 Bulletin
of rational economic actions (Viscusi, Magat and Huber, 1986). However, the identifica-
tion of the impact of health information on risk-taking decisions is complicated because
individual characteristics are likely to affectboth the final decision and the individual effort
to acquire information. For instance, learning and acquiring new information is costly and
may be a time-intensiveprocess, thus the optimal stock of infor mation is likelyto vary with
the individual’s expected costs and benefits of acquiring such information (Kenkel, 1990).
Moreover, the value that individuals place on health, which influences the willingness to
spend money or time to acquire information (Ippolito and Mathios, 1991), varies a lot
among individuals (Hersch and Viscusi, 1990). Previous papers relying on observational
data (e.g. Hsieh and Lin, 1997; Parente, Salkever and DaVanzo, 2005; Maurer, 2009; Car-
man and Kooreman, 2014; Nuscheler and Roeder, 2014) or field experiments(Viscusi et al.,
1986) found a positive effect of information on risk-taking behaviour. While these studies
advance our understanding of the role of information on such behaviour,with the exception
of Viscusi et al. (1986), which is based on a controlled field experiment, they do not allow
establishing a clear causal link between health information and risk-taking decisions.
In this paper, we contribute to this literature by investigating the importance of health
information on preventive decisions in a quasi-experimental setting arising from the im-
plementation of breast cancer screening programmes in Europe. Moreover, we assess the
role of the individual stock of health information and the ability to process the information
received on the link between health information and preventive decisions. Breast cancer
screening programmes represent an interesting case study because empirical evidence sug-
gests that women tend to havefalse perceptions of the risks and seriousness of breast cancer
(Carman and Kooreman, 2014; Wuebker, 2014). Car man and Kooreman (2014) also show
that individuals misperceive the effectiveness of mammography in reducing the risk of
death. Some of them actually think that preventive interventions will increase their risks.
In this situation, the delivery of both accurate information about breast cancer risks and the
effectiveness of mammography in reducing the risk of death might encourage preventive
care actions. This paper will try to assess whether breast cancer programmes are effective
in this respect.
Organized screening programmes (hereafter OSPs) have been provided by local au-
thorities (typically within NUTS-2-regions) since the 1980s in Europe.1Their goal is to
increase mammography use, the best tool available for the early diagnosis of breast can-
cer. Breast cancer is one of the most important health concerns in Europe because of its
high incidence and high mortality risk (OECD, 2009; Von Karsa et al., 2008; Ferlay et al.,
2013). Fortunately, mortality risks and health deterioration caused by breast cancer can be
substantially reduced if cancer is detected early and treated appropriately (WHO, 2015).
Screening programmes are generally effective in this respect because they detect disease
in asymptomatic people. This is especially relevant to breast cancer which typically takes
years to develop and has no early symptoms. Earlydiagnosis leads to a significant increase
in survival probabilities, as documented by substantial literature. The American Cancer
Society (2015) reports a five-year survival rate of 96% among women with earlydiagnosis
1The NUTS classification (Nomenclature of territorial units for statistics) is a hierarchical system for dividing
up the economic territory of the EU. NUTS 2 are the basic regions for the application of regional policies. The
current NUTS classification lists 276 regions at NUTS 2 level. More details can be found on EUROSTAT website
(http://ec.europa.eu/eurostat/web/nuts/overview).
©2016 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
To continue reading
Request your trial