Elsevier

Journal of Anxiety Disorders

Volume 43, October 2016, Pages 32-40
Journal of Anxiety Disorders

The curious case of cyberchondria: A longitudinal study on the reciprocal relationship between health anxiety and online health information seeking

https://doi.org/10.1016/j.janxdis.2016.07.009Get rights and content

Highlights

  • 4-wave longitudinal study on cyberchondria among 5,322 participants.

  • Examining reciprocal relationship between health anxiety and online health information seeking.

  • Individuals who are more health anxious than others, search more frequently for health information online.

  • Reciprocal relationship in respondents with non-clinical levels of health anxiety at the start of the study.

  • For clinically health anxious individuals online health information seeking might serve as maintaining rather than exacerbating factor.

Abstract

The current study is the first to longitudinally investigate the reciprocal relationship between online health information seeking and health anxiety, i.e., cyberchondria. Expectations were that health anxious individuals who go online to find health information, experience an increase in health anxiety, which in turn will reinforce online seeking. A 4-wave longitudinal survey study among 5322 respondents aged 16–93 was conducted. Our results showed that individuals who are more health anxious than others, search online for health information more. Moreover, the results provided initial evidence for the expected reciprocal relationship between health anxiety and online health information seeking in respondents with non-clinical levels of health anxiety at the start of the study. However, this reciprocal relationship could not be found in a subsample of clinically health anxious individuals. Although for these individuals online health information seeking did not seem to exacerbate health anxiety levels, it might still serve as a maintaining factor of clinical health anxiety.

Introduction

The Internet has become an important source of health information and provides the general public with access to a great amount of medical information (Chung, 2013, Cline and Haynes, 2001, Fox and Duggan, 2013, Fox and Jones, 2009, Gallagher and Doherty, 2009; Koch-Weser, Bradshaw, Gualtieri, & Gallagher, 2010; Lee, 2008, Lee and Hawkins, 2010, Morahan-Martin, 2004). Online health information is widely used by Internet users (European Commission, 2013, Fox and Duggan, 2013; Higgins, Sixsmith, Barry, & Domegan, 2011). However, this information is often disorganized, of poor quality and contains technical language (Chung, 2013, Cline and Haynes, 2001, Korp, 2006). Despite its overall usefulness (Cotten and Gupta, 2004, Dickerson et al., 2004, Gallagher and Doherty, 2009, Koch-Weser et al., 2010, Ybarra and Suman, 2008), online health information may thus also distress certain users. More specifically, it has been repeatedly suggested that seeking online health information may further reinforce the anxiety of those who are already overly anxious about their health (Baumgartner & Hartmann, 2011; Muse, McManus, Leung, Meghreblian, & Williams, 2012; Singh and Brown, 2014, Starcevic and Berle, 2013, Starcevic and Berle, 2015).

Health anxiety reflects the – often unfounded – distress or anxiety that a person feels regarding his or her personal health and, because of the misinterpretation of bodily sensations, extremely health anxious people often believe that they have a serious illness or medical condition (Abramowitz, Olatunji, & Deacon, 2007; Ferguson, 2009; Salkovskis, Rimes, Warwick, & Clark, 2002). The level of health anxiety varies among individuals, whereby severe health anxiety may manifest as hypochondriasis (Abramowitz and Moore, 2007, Abramowitz et al., 2007, Ferguson, 2009, Salkovskis et al., 2002). In the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders [DSM-V], hypochondriasis is replaced by illness anxiety disorder and somatic symptom disorder (American Psychiatric Association, 2013). Severe health anxiety is known to co-occur with depressive disorders and anxiety disorders such as panic disorder, generalized anxiety disorder or obsessive compulsive disorder (American Psychological Association, 2013).

The phenomenon of increased health anxiety due to online health information seeking has frequently been referred to as ‘cyberchondria’. Cyberchondria is generally defined as online health-related information seeking that is fueled by anxiety about one’s health (i.e., health anxiety) and that also amplifies this particular anxiety (Starcevic and Berle, 2013, Starcevic and Berle, 2015). Cyberchondria thus implies a reinforcing spiral in which anxiety about one’s health drives online health information seeking, which in turn increases health-related fears. However, although many past studies have referred to the term cyberchondria (e.g., Fergus, 2014, Loos, 2013, McElroy and Shevlin, 2014, Muse et al., 2012, Starcevic and Berle, 2013, White and Horvitz, 2009a), these studies have only provided cross-sectional evidence for a potential relationship between online health information seeking and health anxiety. Accordingly, what may be regarded as the very core of cyberchondria, namely a reciprocal relationship between online health information seeking and health anxiety that develops over time, has not yet been sufficiently examined. Furthermore, although the definition of cyberchondria as proposed by Starcevic and Berle (2013), suggests that health anxiety precedes an increase in online health-related information seeking, cross-sectional data do not inform us about the causal primacy of the reciprocal relationship, and it has yet to be examined whether online seeking may also precede an increase in health anxiety (Aiken & Kirwan, 2014; Harding, Skritskaya, Doherty, & Fallon, 2008; Starcevic & Aboujaoude, 2015; Starecvic & Berle, 2015).

The present article, therefore, substantially extends the existing literature by examining how health anxiety and online health information seeking are related longitudinally in a large nation-wide sample. Adding to the innovativeness of the present approach, we applied an advanced data-analytical procedure to examine the obtained longitudinal data (Hamaker, Kuiper, & Grasman, 2015). This procedure enabled us to examine the proposed relationship within individuals as well as across people.

From a cognitive-behavioral perspective, health anxiety is maintained by several factors: increases in physiological arousal as a response to feeling anxious (physiological factor, e.g., increased heart rate or numb fingers); a bias in the way health information is processed (cognitive factor, e.g., confirmatory attentional bias or sensitivity towards bodily sensations); and safety seeking behaviors (behavioral factor, e.g., checking bodily state; Abramowitz, Schwartz, & Whiteside, 2002; Salkovskis and Warwick, 1986, Warwick, 1989). Reassurance seeking is the most noticeable safety seeking behavior and people with health anxiety feel a constant need to seek reassurance to reduce anxiety and uncertainty about their health (Abramowitz and Moore, 2007, Abramowitz et al., 2007, Abramowitz et al., 2002).

Searching for health information on the Internet is one way to achieve this reassurance (Salkovskis et al., 2002). In light of the cognitive-behavioral model of health anxiety we would thus expect that, as a form of reassurance seeking, health anxious individuals are more likely to search for health information online. Previous studies have indeed shown that health anxious people go online more often to find health information. For example, Muse et al. (2012) revealed that people with high levels of health anxiety go online more frequently and for longer periods of time than people with low levels of health anxiety. Similarly, Singh and Brown (2014) found positive correlations between health anxiety and the frequency of online health information seeking. In addition to increased online searching, health anxious people are also more likely to post health-related questions on online forums (Baumgartner & Hartmann, 2011). Health anxious people thus seem to exhibit more online health information seeking behavior.

The definition of cyberchondria further refers to the proposition that seeking health information online increases health anxiety. Previous studies have shown that health anxious people experience more worries and distress after online health information seeking (Baumgartner and Hartmann, 2011, Muse et al., 2012, Singh and Brown, 2014). More specifically, health anxious people indicated feeling more frightened and anxious based on the health information that they found online (Baumgartner and Hartmann, 2011, Muse et al., 2012). Two mechanisms may account for this effect. First, if health anxious people go online to find reassuring information, they may become overwhelmed by the amount or complexity of the information that they find online (Baumgartner & Hartmann, 2011). The negative information regarding symptoms and illnesses that an individual is likely to encounter online may fuel levels of health anxiety (White & Horvitz, 2009a). Furthermore, although online information may provide some initial reassurance, the effects are often short-lived for health anxious individuals. Indeed, frequent reassurance seeking may increase awareness of bodily symptoms or sensations and thereby reinforce health anxiety (Abramowitz & Moore, 2007; Asmundson, Abramowitz, Richter, & Whedon, 2010; Rachman, 2012). In addition, a recent study by Singh and Brown (2015) shows that health anxious people are likely to engage in query escalation (i.e., an escalation of the seriousness of search terms based on previous search findings). For health anxious people, online health information may thus increase rather than decrease anxiety.

The second mechanism that may account for the effect of online health information seeking on health anxiety can be found in the selective perception of external stimuli. Health anxious people are known to selectively attend to information that confirms their worries about being ill, and they ignore information that counters their existing belief of being ill: this is referred to as the illness-related attentional bias (Hadjistavropoulos, Craig, & Hadjistavropoulos, 1998; Owens, Asmundson, Hadjistavropoulos, & Owens, 2004; Warwick & Salkovskis, 1990). For example, previous studies have shown that health anxiety is positively associated with a bias towards threatening health-related images (Jasper & Witthöft, 2011) and that health anxious people pay more attention to threatening health information compared to less health anxious people (Owens et al., 2004). Accordingly, health anxious people may be prone to attend to more negative online health information that fuels their already existing worries about health.

In sum, health anxious individuals search for health information online more frequent, but the online health information that they find may increase already existing levels of anxiety, which implies a mutually influencing process. Until now, however, this reciprocal relationship has not been studied longitudinally. It is thus unclear whether health anxiety is influenced by online health information seeking over time and vice versa.

The rationale for cyberchondria as described previously, implicitly assumes that this phenomenon pertains to individuals with high or clinical levels of health anxiety. However, the downsides of online health information such as technical language or lack of quality and an abundance of negative information may also increase health anxiety in individuals with non-clinical levels of health anxiety (White and Horvitz, 2009a, White and Horvitz, 2009b). Thus, another interpretation of cyberchondria may be that online health information seeking may lead to health anxious beliefs even in individuals who were not clinically health anxious before (Aiken and Kirwan, 2014, Starcevic and Aboujaoude, 2015, Starcevic and Berle, 2015). Since studies until now have only focused on cross-sectional associations between health anxiety and online health information seeking, and not on causal effects and primacy of causes, the question remains to what extent cyberchondria constitutes a phenomenon associated with clinical health anxiety or an impairing feature of the Internet that may affect everyone.

In the present study we aim to examine the reciprocal relationship between health anxiety and online health information seeking in a longitudinal design. We firstly hypothesize that higher levels of health anxiety are associated with higher levels of online health information seeking (H1). Secondly, we expect that an increase in health anxiety relative to one’s own average level of health anxiety will predict increases in one’s online health information seeking (H2). Thirdly, we hypothesize that an increase in online health information seeking relative to a person’s own average will predict increases in a person’s health anxiety (H3). We examined the proposed hypotheses in a general population sample. However, because cyberchondria may be a phenomenon not only present in clinically health anxious individuals, but also in the general population, and because the processes underlying this phenomenon may operate differently among clinically health anxious individuals versus people with normal levels of health anxiety, we examined our hypotheses separately in both groups.

Choosing the right time lags in longitudinal studies is an important issue. However, only few studies exist that provide clear recommendations for specific time lags (e.g., Cole and Maxwell, 2009, Collins and Graham, 2002, Dormann and Griffin, 2015, Selig and Little, 2012). Furthermore, determining which time lag is optimal is more difficult if no prior studies exist that examined the proposed relationships longitudinally. In longitudinal designs such as the current, it is of importance that time lags are not too long to prevent effects from declining over time. At the same time, time lags should also not be too short, in order to make it possible to examine sustainability of effects (Dormann & Griffin, 2015). In the present study, two-month time lags were chosen partly for practical reasons as the study was part of a larger panel. However, the time lag in the current study is assumed to be long enough to reveal potential lasting effects of online health information seeking on health anxiety.

Section snippets

Procedure and sample

The data for the present study were collected in the longitudinal Internet studies for the social sciences (LISS) panel that is administered by CentERdata (Tilburg University, The Netherlands).1 The LISS panel is a representative sample of Dutch individuals (aged 16 years and

Correlations

Correlational analysis in the clinical subsample revealed weak associations between health anxiety and online health information seeking across the four waves (see Table 1). The high correlations between the waves with regard to health anxiety (0.57, 0.65, 0.72) indicate a substantial stability of the rank order of individuals with regard to the level of health anxiety across waves. The intra-class coefficient (ICC; a method used for multi-level analysis; computed in SPSS version 23; Hox, 2010)

Discussion

In the current study, we sought to further our understanding of the relationship between online health information seeking and health anxiety. More specifically, we investigated the possible reciprocal relationship between these two constructs. This relationship has often been labeled cyberchondria. Cyberchondria entails individuals who are health anxious going online to find health information, e.g., to reassure themselves in order to decrease anxiety. Going online, however, is expected to

Acknowledgements

The authors thank Ellen Hamaker and Loes Keijsers for their valuable advice on the application of the random intercept cross-lagged panel model. This paper makes use of data from the Longitudinal Internet Studies for the Social Sciences (LISS) panel administered by CentERdata (Tilburg University, The Netherlands). The LISS panel data were collected by CentERdata through its MESS project funded by the Netherlands Organization for Scientific Research. More information about the LISS panel can be

References (68)

  • F. Jasper et al.

    Health anxiety and attentional bias: the time course of vigilance and avoidance in light of pictorial illness information

    Journal of Anxiety Disorders

    (2011)
  • E. McElroy et al.

    The development and initial validation of the cyberchondria severity scale (CSS)

    Journal of Anxiety Disorders

    (2014)
  • K. Muse et al.

    Cyberchondriasis: Fact or fiction? A preliminary examination of the relationship between health anxiety and searching for health information on the Internet

    Journal of Anxiety Disorders

    (2012)
  • S. Rachman

    Health anxiety disorders: A cognitive construal

    Behaviour Research and Therapy

    (2012)
  • P.M. Salkovskis et al.

    Morbid preoccupations, health anxiety and reassurance: A cognitive-behavioural approach to hypochondriasis

    Behaviour Research and Therapy

    (1986)
  • H.M.C. Warwick et al.

    Hypochondriasis

    Behaviour Research and Therapy

    (1990)
  • H.M.C. Warwick

    A cognitive-behavioural approach to hypochondriasis and health anxiety

    Journal of Psychosomatic Research

    (1989)
  • M. Aiken et al.

    The psychology of cyberchondria and cyberchondria by proxy

  • American Psychological Association

    Diagnostic and statistical manual of mental disorders

    (2013)
  • G.J. Asmundson et al.

    Health anxiety: Current perspectives and future directions

    Current Psychiatry Reports

    (2010)
  • A.J. Barsky et al.

    Overview: Hypochondriasis, bodily complaints, and somatic styles

    American Journal of Psychiatry

    (1983)
  • A.J. Barsky et al.

    Resource utilization of patients with hypochondriacal health anxiety and somatization

    Medical Care

    (2001)
  • S.E. Baumgartner et al.

    The role of health anxiety in online health information search Cyberpsychology

    Behavior and Social Networking

    (2011)
  • J.M. Bland et al.

    Transformations, means, and confidence intervals

    BMJ: British Medical Journal

    (1996)
  • T.A. Brown et al.

    Confirmatory factor analysis

  • T.A. Brown

    Confirmatory factor analysis for applied research

    (2015)
  • J.E. Chung

    Patient-provider discussion of online health information: Results from the 2007 Health Information National Trends Survey (HINTS)

    Journal of Health Communication

    (2013)
  • R.J. Cline et al.

    Consumer health information seeking on the Internet: The state of the art

    Health Education Research

    (2001)
  • D.A. Cole et al.

    Statistical methods for risk-outcome research: Being sensitive to longitudinal structure

    Annual Review of Clinical Psychology

    (2009)
  • P.J. Curran et al.

    The disaggregation of within-person and between-person effects in longitudinal models of change

    Annual Review of Psychology

    (2011)
  • A. Diamantopoulos et al.

    Formative versus reflective indicators in organizational measure development: A comparison and empirical illustration

    British Journal of Management

    (2006)
  • C. Dormann et al.

    Optimal time lags in panel studies

    Psychological Methods

    (2015)
  • European Commission

    ICT for societal challenges

    (2013)
  • E. Ferguson

    A taxometric analysis of health anxiety

    Psychological Medicine

    (2009)
  • Cited by (108)

    View all citing articles on Scopus
    View full text