Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2024 Oct 10;14:23683. doi: 10.1038/s41598-024-74829-z

Developing a cyberchondria severity scale to promote self-care among university students during COVID-19

Wan-Chen Hsu 1,✉
PMCID: PMC11466950  PMID: 39390121

Abstract

Cyberchondria is a hybrid term signifying a state in which individuals compulsively search for health-related information online because of health distress or anxiety, further aggravating their worries. This study develops a Cyberchondria Severity Scale (CSS) to assess the current situation of university students. Exploratory factor analysis (EFA) was conducted using 145 valid questionnaires. Subsequently, a nationwide survey was conducted at various universities in Taiwan, and 802 questionnaires were used for confirmatory factor analysis (CFA). The EFA led to the development of a CSS for college students with four constructs: increased anxiety (α = 0.91), obsessive-compulsive hypochondria (α = 0.87), perceived controllability (α = 0.88), and online physician-patient interaction (α = 0.86), with a Cronbach’s alpha of 0.92 and variance of 66.81%. The CFA indicated that item reliability ranged from 0.50 to 0.86, factor loadings ranged from 0.71 to 0.93, and the composite reliability for latent variables ranged from 0.83 to 0.90 (p < .001). The extracted average variance ranged from 0.46 to 0.60. There were significant differences in various dimensions: perceived controllability had the highest score and obsessive-compulsive hypochondria had the lowest (F3,2403=256.26, p < .001). Therefore, CSS has emerged as a reliable and valid measure. Future studies should explore the factors that influence cyberchondria, promote online health information searches, and enhance reading comprehension skills.

Keywords: Cyberchondria severity, Exploratory factor analysis, Confirmatory factor analysis, Measurement invariance, Undergraduate students

Subject terms: Psychology, Health care

Introduction

The coronavirus disease 2019 (COVID-19) pandemic, caused primarily by novel strains of severe acute respiratory syndrome coronavirus, began in Wuhan, China, in December 2019 and quickly spread globally1. The World Health Organization declared it a Public Health Emergency of International Concern on January 30, 2020, and a pandemic on March 11, 2020. The pandemic’s progression has shown fluctuating trends with periods of decline and resurgence. Although there has been a reported decline in the number of cases since March 2022, there was a notable increase in weekly cases from early June to mid-July 2022. By August 2022, the number of confirmed cases exceeded 600 million globally, with over 6.4 million deaths. Despite widespread vaccinations and precautions, the pandemic’s future impact remains uncertain, with significant socio-economic and psychological repercussions, including stress and mental health burdens exacerbated by misinformation, vaccine inequity, and disease-associated stigma2. Although the psychological impact of COVID-19 has been evaluated in the literature, further research is needed, particularly on the Cyberchondria Severity Scale (CSS) and related psychological outcomes.

Electronic health information has emerged as an essential tool in public health management. Various health-based websites have been developed to address the increasing demand for self-testing. However, the accuracy and quality of the information available on some of these websites is questionable, and facts can be misrepresented because the content is based on readers’ personal experiences. Although reputable organizations also provide information on their websites, it may not be edited or validated, which may directly or indirectly influence readers’ understanding, sense of judgment, and decision-making abilities. Furthermore, some self-diagnosis websites exploit people economically by marketing drugs or other treatments2. Readers are exposed to a large amount of easily accessible online information; however, the close association between technological risks and individual lifestyles is often overlooked. People with electronic health literacy can gather high-quality health information through online searches with fewer technological risks3. By contrast, individuals with fewer technology skills may develop cyberchondria because they are unable to evaluate the veracity of the online health information available to them.

The COVID-19 pandemic has significantly impacted the behaviors of individuals, social groups, and organizations, affecting emotions, cognition, behavior, overall mental health, and related psychosocial factors4–6. Quarantine and lockdown measures have triggered various psychological and behavioral responses, including depression, anxiety, stress, sleep disorders, increased healthcare demand, and suicidal ideation, resulting in a notable increase in the monthly suicide rate. These widespread effects have dramatically influenced the public’s general mental well-being. Aljaberi et al.4 examined COVID-19’s role in escalating anxiety and its disruptive impact on daily life due to quarantine measures. The strong associations between the pandemic and the symptoms of depression, anxiety, and insomnia highlight the urgent need for comprehensive psychological and public health interventions to mitigate these effects.

Cyberchondria is a neologism that combines cyber and hypochondriasis. It is a state in which an individual engages in compulsive and excessive online searches for health-related information because of distress or anxiety, resulting in the onset of related symptoms7. During the pandemic, people’s daily lives were disrupted because they were restricted from leaving home. Consequently, patients’ physical activity decreased. On May 19, 2021, Taiwan’s Ministry of Education announced the suspension of classes at all levels. This, along with other significant social changes, resulted in many challenges for higher education institutions, bringing a renewed focus on student health, a topic that has already elicited concern8.

Taiwan’s internet penetration rate is higher than the global average. In recent years, the number of people using the Internet in Taiwan has increased. According to the 2022 Taiwan Broadband Internet Usage Survey Report, 81.47% of the total population used mobile broadband, and 84.3% used mobile broadband9. The popularity of the Internet has contributed to the phenomenon of individuals using it for longer periods. Taiwanese people use the Internet for up to eight hours and seven minutes every day, higher than the global average of six hours and 58 min. The primary purpose of Internet browsing is to find information and follow current events10. College students are among the main users of the Internet as a primary channel for obtaining information9.

The COVID-19 pandemic has also affected students’ Internet use. Abiddine et al.11 highlighted the fact that problematic social media use is significantly associated with increased insomnia and decreased overall well-being among university students. Social media addiction has also been linked to poor sleep quality, short sleep duration, and increased fatigue. These issues worsened during the COVID-19 pandemic because of increased online activity and the fear of missing out. Individuals usually search the Internet for health-related information to ensure peace of mind. However, search results may make one feel more uncertain, stressed, and anxious because online health information is not always accurate or reliable. This can lead to incorrect self-diagnosis12. Chen and Lee13 noted that people often use their limited skills to search for and evaluate information from various online sources of varying quality. Although readers may benefit from self-diagnosis and gain a better understanding of their conditions through information gathered from health websites, they may experience excessive worries about their physical health, which can lead to anxiety and life-threatening situations14. The risks associated with exposure to Internet technology and excessive online information highlight the importance of using health information judiciously. Kim et al.15 indicated that the existing research has explored variables such as technology use, perceptions, and intentions in relation to discomfort with technology. However, further investigations are warranted to identify the factors that predict technology-related discomfort. This is essential for identifying effective strategies to alleviate the anxiety and hesitation associated with technology among adults. The use of CSSs has also been analyzed in the literature. As university students frequently search for health information online, the aim of this study was to create a CSS for surveying Taiwanese university students. The survey results may inspire further studies and help to formulate public health information communication strategies. Previous studies have shown the relationship between problematic social media use and negative psychological outcomes, like depression, anxiety4,5, and poor sleep quality11. However, the impact of self-diagnosis through Internet use on university students must be further explored. The aim of this study was to develop a highly valid self-reporting CSS and understand the current status of university students. The goal was to foster self-care behaviors among university students and address the unique challenges posed by the pandemic. Cyberchondria is a new concept that has emerged with the rise of Internet use and is characterized by excessive online health information-seeking, which leads to increased health anxiety. Although the existing CSSs have provided some evidence of reliability and validity, a significant gap remains in understanding the multi-dimensional nature of this phenomenon, especially in varying cultural contexts. Early studies primarily used single-item measures16, which did not capture the full complexity of cyberchondria.

Furthermore, localized scales specifically designed for Asian populations are lacking. Cultural differences can significantly influence the manifestation and perception of cyberchondria, making it crucial to develop a scale that accurately reflects individuals’ experiences in Asian contexts. The aim of this study was to construct a multi-dimensional and multi-item CSS tailored to the Asian context, thereby enriching our understanding and providing a more comprehensive tool for future research and clinical assessments.

Literature review

Online health information-seeking behavior and cyberchondria

Online information seeking can be unintentional, passive, or active17. It typically targets individuals seeking information to satisfy their personal needs or objectives. A review was conducted on human information behavior (HIB) in the context of the COVID-19 health crisis, assuming that HIB and information practices serve as tools for human adaptation to evolving life circumstances. Montesi18 found it imperative to further investigate information-seeking behaviors within particular contexts, focusing on the needs of vulnerable and marginalized groups. Additional areas for exploration include the intricate dynamics among emotions, knowledge, and behaviors in the information-seeking process, a more nuanced comprehension of local and experiential knowledge, and recognition of the limitations associated with information and communication technologies (ICTs). Health information-seeking is a behavioral pattern of accessing online resources to facilitate decision-making regarding one’s health19. Browsing the Internet for health information is termed online health information-seeking behavior and is widely prevalent today. The positive effects of online health information-seeking behaviors, which include cost savings and improvements in health literacy, are mostly determined by readers’ ability to critically evaluate online information. This indicates that online health information may have either a positive or negative impact on individuals’ health20.

Diagnoses obtained from online searches can only provide temporary answers to health problems and are usually the reason for unease and anxiety. Readers often self-diagnose based on their symptoms by browsing the Internet. However, search results often increase anxiety because it is difficult to determine the authenticity of online health information21. Most people searching for health-related online content have not received medical training. Thus, their inability to critically examine online health information and understand technical jargon may have negative results, leading to anxiety and distress, which may develop into severe cyberchondria20.

CSSs are still in the developmental phase, with initial studies employing a single question to gauge the extent of cyberchondriasis. Subsequent research has progressively formulated multi-dimensional and multi-question severity scales. Within these scales, “compulsion” denotes the excessive pursuit of online health information, “distress” pertains to the negative emotions and physiological responses stemming from health information searches, “reassurance” involves seeking confirmation from medical professionals, and “mistrust of medical professionals” entails trust, or lack thereof, in online health information search results, as well as the trustworthiness of doctors, leading to contradictions and conflicts arising from diverse information sources16. Along with the development of reliable assessment tools, it is essential to include inquiries tailored to different populations and sample characteristics when investigating cyberchondria, a psychological phenomenon described by Starcevic22.

Despite the increasing trend in health-related online browsing, post-search health behaviors have not been comprehensively examined. Myrick23 conducted a naturalistic experiment on the post-search emotions of 380 Americans who searched for influenza and examined how to construct cognitive and behavioral theoretical models. She found that the participants had difficulty searching for additional information related to the verification of dubious or questionable sources. Researchers have further examined ways to improve search techniques and found that individuals experience many emotions (fear, hope, contentment, interest, and inspiration) after searching for information and that social cognitive factors mediate and affect subsequent attitudes and behaviors. Emotions such as enthusiasm, interest, and hope during online searches positively affect confidence and behavioral intentions.

Online health information-seeking behavior outcomes

Social cognitive theory can be used to explain post-search results. For example, individuals’ confidence in finding quality health information, known as self-efficacy for searching, along with post-search outcome experiences, referred to as self-efficacy and outcome experiences, is a strong predictor of online health information-seeking behavior23.

In Bandura’s24 social cognitive theory, self-awareness pertains to examining one’s ability to determine competency, which is the central focus of self-efficacy theory. Bandura24 considers self-efficacy to be the belief that individuals can organize and execute a series of actions to achieve certain results. This signifies their ability to participate in certain tasks or present certain behaviors and provides a subjective assessment of their work or behavior. This assessment may differ from actual competency but molds an individual’s internal thoughts, thereby affecting their effort to seek success or their level of perseverance when faced with difficulties. In other words, self-efficacy is an important factor driving behavioral motivation. Bandura25 proposes that efficacy and outcome experience are the two major axes of self-efficacy. Efficacy experience refers to an individual’s ability to determine whether they can successfully complete certain tasks. Outcome experience refers to one’s focus on evaluating the results arising from certain behaviors; that is, an individual believes that certain behaviors lead to certain outcomes. A positive self-evaluation results in contentment, confidence, and self-validation, whereas a negative self-evaluation leads to self-deprecation and blame.

One may believe that certain behaviors lead to certain outcomes (high-outcome experiences) but that an action may not be performed if there are doubts about one’s ability to perform certain behaviors (low-outcome experiences). For example, most people believe that a mix of true and false information is available online; however, they still browse the internet to find solutions or dispel doubts about their health problems. Therefore, an individual’s knowledge may not be equivalent to their behavior, and the efficacy experience may not be consistent with the outcome experiences.

Bandura26 found that different self-efficacy scales yielded varying psychological results when used to evaluate behavioral outcomes. When efficacy and outcome experiences are high, an individual will have the confidence to act. When efficacy experience is low, a person tends to become stuck when facing challenges, ultimately leading to failure. However, when efficacy experience is high and outcome experience is low, people exert more effort to overcome adversity, leading them to not give up easily. Thus, a person’s behaviors in any given situation are affected by efficacy and outcome experiences. Furthermore, the interaction between these two factors has a significant impact on self-efficacy. In addition, individuals may encounter two types of futility: efficacy- and outcome-based. Individuals must enhance their capabilities and self-efficacy to manage efficacy-based futility. Regarding outcome-based futility, individuals must change their environments and identify suitable tasks. Overcoming both forms of futility empowers them to take more affirmative action.

Starcevic and Berle’s7 findings are consistent with the theoretical explanation of self-efficacy. According to these studies, when individuals with high anxiety search for health information online, they feel relieved after receiving assurance, particularly from their physicians. This reinforces their online search behaviors. In contrast, people who search for online information to handle their negative emotions are more likely to continue to rely on the internet for health information, which may lead them to experience repeated disappointments, thereby aggravating their emotional burdens and exacerbating the severity of their cyberchondria. Consequently, they might refrain from conducting online searches.

When self-efficacy theory is applied to university students’ online health information-seeking behavior, the axis of efficacy experience indicates that such behavior is a form of evaluation, whereas outcome experience indicates that the outcome influences their online information-seeking behavior. When university students’ efficacy and outcome experience were high, they had the confidence to overcome the limitations of their online behavior. They rated themselves highly in post-search assessments, which resulted in a positive application of online health information. When their efficacy experience was low, they tended to become stuck and unable to solve their problems. This ultimately led to failure or lack of results from online health information-seeking behaviors. However, when their efficacy experience was high and outcome experience was low, they were likely to increase their online presence or employ other methods to overcome challenges, reflecting the fact that difficult challenges cause greater concern and result in excessive critical thinking and resistance27.

Methods

Research design

This study adopted a survey research method using stratified cluster sampling. The samples were drawn based on the proportion of university students in the northern, central, southern, and eastern regions, with classes as the sampling units. The quality of the CSS was tested, and the current status of university students was analyzed.

Participants and sample size

An official cyberchondria severity questionnaire was constructed using convenience sampling to recruit participants from four universities, one each representing northern, central, southern, and eastern Taiwan. The class formed a unit for block sampling in these four universities, with one class each from general education centers recruited as a pilot sample. Gorsuch28 recommends that the sample size should be at least five times the number of questions. With more than 100 questionnaires for factor analysis, a total of 145 valid pilot questionnaires were administered. Furthermore, data from the Department of Statistics of the Ministry of Education29 indicate that the four universities had 1,035,218 students in 2017. Stratified block sampling was used to obtain more representative samples, in which the region was used for stratification. The proportion of university students from Taiwan’s four regions was used as the basis for the sample size. Of the 802 final samples, 367 (46%) were from female participants and 435 (54%) were from male participants. Hair et al.30 suggested that for covariance based on structural equation modeling (SEM; e.g., AMOS), the analysis requires a sample size greater than 10030,31. In this study, a sample size of 802 was considered adequate for employing SEM to address the research objectives.

Ethical approval and consent to participate

Ethical approval was granted by the Ethics Research Committee of National Cheng Kung University (No. 110-573-2), and informed consent was obtained from all participants. Participants had the right to withdraw from the study at any time. Principles of anonymity and confidentiality were strictly followed. All methods were carried out in accordance with relevant ethical standards, guidelines, and regulations.

Instruments

The CSSs used in this study were modified from those proposed by White and Horvitz32 and Batigün et al.33. The items were rated on a five-point Likert scale ranging from 1 (never) to 2 (rarely), 3 (sometimes), 4 (often), and 5 (always). After completing the initial scale, the researchers invited two experts in the field of psychological counseling and two university students to review the items. After reviewing and revising the questionnaire, a preliminary scale was developed.

Procedures and data analysis

Two experts reviewed the content validity of the CSS. Next, exploratory factor analysis (EFA) was performed on the preliminary test samples, followed by confirmatory factor analysis (CFA) of the final samples using SEM. AMOS 7.0 was used for statistical processing and testing of the composite reliability of this univariate model. EFA was conducted using 145 valid questionnaires. Subsequently, a nationwide survey was conducted at various universities in Taiwan; 802 questionnaires were used for CFA, and the current status of cyberchondria among university students was analyzed. SPSS 20 was used to calculate the mean and standard deviation. Repeated-measures analysis of variance was used to compare the differences in the current status and various aspects of cyberchondria severity among the participants. The scale construction process in this study included two main steps: EFA and CFA.

Exploratory factor analysis

Using the orthogonal rotation method, EFA was conducted to determine the number of underlying factors. The extracted factors needed to meet several criteria, including a theoretical justification, scree plot, Kaiser’s eigenvalue ≥ 1, and total variance explained ≥ 0.6034. In addition, the direction and proportion of factor loading needed to be ≥ 0.55, and the corrected item-total correlation needed to be at least 0.30 and positive30.

Confirmatory factor analysis

CFA was conducted to confirm the hypothesized measurement model. CFA is meant for measurement models within SEM analysis, whereas full-fledged SEM comprises a measurement and a structural model. In other words, CFA deals with the relationships between observed measures or indicators (e.g., test items, test scores, and behavioral observation ratings) and latent variables or factors but not the relationship between exogenous latent variables and endogenous latent variables31. This study evaluates the quality of the CSS measurement tool based on Preliminary Fit Criteria, Overall Model Fit Criteria, and a Fit of Internal Structural Model35,36, which are presented in the findings.

Results

Construction of the CSS

This study aimed to construct questions for the CSS and test its reliability and validity. Data from 145 university students were used for item analysis. Four constructs accounting for 66.81% of the total variation were obtained from the factor analysis: increased anxiety (six items), obsessive-compulsive hypochondria (six items), perceived controllability (five items), and online physician-patient interaction (six items). Increased anxiety refers to the anxiety experienced by an individual due to apprehensions about online content seeking to understand a disease or its symptoms. Obsessive-compulsive hypochondria refers to excessive worry or excessive Internet surfing to understand a disease or its symptoms. Perceived controllability refers to the decreased doubts or psychological burdens caused by an online search for health information. Online physician-patient interaction refers to an individual’s online information search, which affects their judgment of their symptoms.

This study used 802 final scales for the CFA. Additionally, we followed the recommendations of Bagozzi and Yi35 and Yu36. Compatibility between the model and the actual observational data was tested using the Preliminary Fit Criteria, Overall Model Fit Criteria, and Fit of Internal Structural Model. Additionally, each factor needed to have at least one indicator with a standardized factor load above 0.70, and the construct reliability of each factor needed to be above 0.50. The results of the initial model analysis satisfied these criteria (Table 1). The official scale (Table 2) and model used in this study (Fig. 1) are shown below.

Table 1.

Preliminary fit criteria, overall Model Fit Criteria, and fit of Internal Structural Model.

Assessment indicator Fit value Initial model Evaluation results
χ2 Significant (p > .05) 1109 (df = 224, p < .01) Non-compliant χ2 is suitable for a sample size of 100–200 and tends to be affected by sample size
χ2/df < 5 (Acceptable) 4.95 Compliant
GFI > 0.90 0.889 Acceptable
AGFI > 0.90 0.863 Acceptable
RMR < 0.05 (The smaller the better) 0.041 Compliant
SRMR < 0.05 0.0459 Compliant
RMSEA (90%CI) < 0.08 0.070 Compliant
TLI > 0.90 0.936 Compliant
NFI > 0.90 0.930 Compliant
CFI > 0.90 0.943 Compliant
IFI > 0.90 0.943
RFI > 0.90 0.921
CN > 200 200 (α = 0.01) Compliant The CN value reflects sample size suitability. If the CN is more than 200, the sample size is sufficient.
PNFI > 0.50 0.823 Compliant
PGFI > 0.50 0.722 Compliant
PCFI > 0.50 0.835 Compliant
Open in a new tab

Table 2.

Exploratory factor analysis and confirmatory factor analysis of the cyberchondria severity scale for university students.

Construct 1: Increased anxiety 1. To understand a certain disease or symptom, I went online to search for information. The large amount of information was too much for me to handle and increased my anxiety. Average variance extracted 0.60 Composite reliability 0.90
2. To understand a certain disease or symptom, I went online to search for information. The related information (such as content title, table, or figure description) was too much for me to handle and increased my anxiety.
3. To understand a certain disease or symptom, I went online to search for information. Judging the reliability of information sources was too much for to me handle and increased my anxiety.
4. To understand a certain disease or symptom, I went online to search for information. The information was overly graphic, and the stringent description was too much for me to handle and increased my anxiety.
5. To understand a certain disease or symptom, I went online to search for information. The terms used (e.g., critical, lethal, life-threatening) made me feel apprehensive, which was too much for me to handle and increased my anxiety.
6. To understand a certain disease or symptom, I went online to search for information. The web pages often used complex medical jargon, which was too much for me to handle and increased my anxiety.
Construct 2: Obsessive-compulsive hypochondria 7. I often spent several weeks or months searching for information online to understand a certain disease or symptom. Average variance extracted 0.46 Composite reliability 0.83
8. When seeking to better understand a certain disease or symptom, the search process caused me to ignore other things I can do online (e.g., chatting and playing games).
9. When seeking to better understand a certain disease or symptom, the search process affects my daily activities.
10. When seeking to understand a certain disease or symptom, I went online to search for information, which increased my anxiety.
11. I feel that I am someone who worries excessively about health.
12. My friends, family members, and medical professionals feel that I worry excessively about health.
Construct 3: Perceived controllability 13. Searching for health-related information online allows me to understand my medical condition and decreases my anxiety. Average variance extracted 0.59 Composite reliability 0.87
14. Searching for health-related information online lets me find information regarding my medical condition from reputable sources (e.g., hospital websites), which decreases my anxiety.
15. When I search for health-related information online, I integrate all information and views on my medical condition from various websites, which decreases my anxiety.
16. When I search for health-related information online, I read personal cases of medical diagnosis, which decreases my anxiety.
17. I search for health-related information online, which provides me with knowledge of my current medical condition, thereby affecting my behavior.
Construct 4: Online physician-patient interaction 18. I previously searched for one or more symptoms on the Internet to determine possible medical conditions. Average variance extracted 0.53 Composite reliability 0.87
19. I previously used the Internet as a medical professional knowledge system to search for symptoms of potential conditions.
20. When I search the Internet, I focus on the symptoms, which causes me to adopt a stringent attitude when viewing the content.
21. When my search includes medical symptoms, the sequence of the search results may be a ranking of the likelihood of a disease occurring (e.g., the disease with the highest likelihood of occurring appears at the top of the search results).
22. After seeing a physician, I would use the Internet to obtain detailed information on the diagnosis.
23. The Internet is useful for preliminary/initial disease diagnosis.
Open in a new tab

Figure 1.

Figure 1

Open in a new tab

Cyberchondria severity scale confirmatory factor analysis measurement model.

Analysis of the current status of university students’ cyberchondria severity

The mean individual item score of the cyberchondria severity constructs was between 2.19 and 2.75 (2 = rarely and 3 = sometimes), and the mean scores of the individual items were all below the median. This finding indicated that college students do not often experience anxiety or hypochondria during their online searches for health-related information. They do not feel that anxiety is controllable, nor do they adopt online physician-patient interaction behavior (Table 3). As the students’ item scores for the four constructs differed, the mean individual scores were used for the univariate variance analysis. The results showed significant differences in the cyberchondria severity constructs (F [3, 2403]= 256.26, p < .001, η2 = 0.16).

Table 3.

Analysis of differences between various cyberchondria severity scale constructs.

Minimum Maximum Mean Standard deviation F Post hoc comparisona
Mean increased anxiety individual question score 1 5 2.39 0.82 256.26*** 3 > 4 > 1 > 2
Mean obsessive-compulsive hypochondria individual question score 1 5 2.19 0.77
Mean perceived controllability individual question score 1 5 2.75 0.87
Mean online physician-patient interaction individual question score 1 5 2.70 0.80
Open in a new tab

N = 802 ***p < .001.

a1: increased anxiety; 2: obsessive-compulsive hypochondriasis; 3: perceived controllability; 4: obsessive-compulsive hypochondriasis.

Top left: increased anxiety, top right: obsessive-compulsive hypochondria, bottom left: perceived controllability, bottom right: online physician-patient interaction.

The least significant difference test was used as a post hoc test. The results showed that perceived controllability had the highest score, with item 14, “Searching for health-related information online lets me find information regarding my medical condition from reputable sources (e.g., hospital websites), which decreases my anxiety,” obtaining the highest score (M = 2.79). The obsessive-compulsive hypochondria construct had the highest score, in which item 11, “I feel that I am someone who worries excessively about health,” had the highest score (2.29). This finding indicates that university students do not worry excessively about their health.

The various constructs showed a significant positive correlation between perceived controllability and online physician-patient interaction and increased anxiety and obsessive-compulsive hypochondria. This indicates that students with controlled anxiety and frequent online physician-patient interactions during their Internet searches also have increased anxiety and obsessive-compulsive hypochondria. Furthermore, perceived controllability was significantly positively correlated with online physician-patient interactions.

Discussion

This study demonstrated that the CSS developed for university students has good validity and reliability. The scale was divided into four constructs: increased anxiety (α = 0.908), obsessive-compulsive hypochondria (α = 0.865), perceived controllability (α = 0.884), and online physician-patient interaction (α = 0.858). The Cronbach’s alpha ranged from 0.858 to 0.908. A CFA of 802 questionnaires showed that item reliability ranged from 0.50 to 0.86, factor loadings ranged from 0.71 to 0.93, and the composite reliability for latent variables ranged from 0.83 to 0.90 (p < .001). The extracted average variance ranged from 0.46 to 0.60. Cyberchondria severity differed among university students and showed stratification. Significant differences were found in various dimensions of the CSS, of which perceived controllability had the highest score and obsessive-compulsive hypochondria had the lowest. This study contributes to the theoretical understanding of cyberchondria among university students, particularly during the COVID-19 pandemic. The development and validation of the CSS provides a robust tool for future research in this area. Practically, the CSS can be used by university health services to identify students at risk of cyberchondria and develop targeted interventions to foster self-care behaviors. This is particularly relevant given the increased reliance on online health information during the pandemic.

The various constructs were significantly correlated with each other; increased anxiety and obsessive-compulsive hypochondria showed a significant moderate positive correlation with perceived controllability and online physician-patient interaction. When individuals with high health anxiety search for health information online, their anxiety is alleviated after receiving assurance from others (particularly doctors). This, in turn, reinforces online health information search behaviors7. By contrast, if individuals experience more uneasiness or fear, they may continue to search for health information online to cope with their negative emotions. However, they may experience repeated disappointment, which may aggravate their emotional burden and increase cyberchondria severity20. These findings are consistent with Bandura’s 1982 study26, which stated that individuals with higher self-efficacy have more confidence or determination to perform corresponding actions. Uncertainty and doubt may increase when interpreting online information.

The COVID-19 pandemic, which confined people to their homes and upended routines, has intensified digital scouring. Although readers may benefit from self-diagnosis and gain a better understanding of their condition through information gathered from health websites, self-diagnosis may cause excessive worries about physical health, which can lead to anxiety and life-threatening situations37. The risks associated with exposure to Internet technology and excessive online information highlight the importance of using health information judiciously38. Regarding online search techniques affecting cyberchondria among students, Hsu37 found that 59% of the participants used jargon as a keyword, and only 27% used Boolean logic in their search strategy. Of the participants, 81% were concerned about the date on which the information was obtained, whereas 85% did not consider the author’s profession. This indicates that there is scope for improvement in the students’ functional and critical literacy. Moreover, the motivation from self-awareness to action is driven by critical awareness, and improving critical skills teaching is key to guiding students in acquiring health literacy. In particular, the information available on the Internet is not fully reliable, and attention should be paid to students’ ability to search for and evaluate it. Lee et al.39 indicated that older adults’ acquisition and utilization of ICT skills contributed to enhancements in their overall eHealth literacy and the perceived usefulness of the Internet. Individuals who initially harbored technophobia experienced a reduction in anxiety and boosted their confidence when using computer technology over the course of the study. Thus, the desirability and feasibility of developing targeted educational programs for university students with limited access to ICT and minimal digital literacy are important.

Conclusions and recommendations

The CSS for university students showed that various markers had a good fit and that the scale had good validity and reliability. However, this scale is still in the initial stages of development. Various constructs were significantly correlated with each other; increased anxiety and obsessive-compulsive hypochondria showed a significantly moderate positive correlation with perceived controllability and online physician-patient interaction. As this study included only healthy university students, subsequent studies should recruit different adult populations for a more holistic health perspective. Additionally, the participants’ innermost thoughts on specific questions could not be obtained as the study employed a self-rated scale. Therefore, we recommend broad-based qualitative and quantitative designs for future studies to obtain a deeper understanding of online information searches and cyberchondria. As studies on CSSs are in their infancy, further research should focus on examining the factors affecting cyberchondria, promoting online health information searches, and enhancing reading comprehension. This study constructed a CSS scale, but the single-measurement model has limitations in its application. One limitation was its cross-sectional design, which prevented the establishment of causality. Future longitudinal studies are needed to examine the causal relationships among Internet use, cyberchondria, and psychological outcomes. Scale development was limited to the scale itself and lacked concurrent validity testing. It is suggested that future research should investigate related variables such as anxiety and stress5 to further validate the cross-validation of the scale. In addition, the cross-sectional design prevented us from establishing causality. Future longitudinal studies are needed to examine the causal relationships among Internet use, cyberchondria, and psychological outcomes.

Acknowledgements

This work was supported by the National Science and Technology Council, R. O. C., grant no. 111-2410-H-992 -037 -MY2.

Author contributions

Single authorConceptualization, Methodology, Software, Investigation, Formal Analysis, Writing.

Data availability

The datasets generated and/or analysed during the current study are not publicly available because they contain information about the original materials used and can be used to copy works of art but are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Sabah, A. et al. Family communication as a mediator between family resilience and family functioning under the quarantine and COVID-19 pandemic in Arabic countries. Children10(11), 1742. 10.3390/children10111742 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Ryan, A. & Wilson, S. Internet healthcare: do self-diagnosis sites do more harm than good?. Expert Opin. Drug Saf.7(3), 227–229. 10.1517/14740338.7.3.227 (2008). [DOI] [PubMed] [Google Scholar]
  • 3.Quinn, S., Bond, R. & Nugent, C. Quantifying health literacy and ehealth literacy using existing instruments and browser-based software for tracking online health information seeking behavior. Comput. Human Behav.69, 256–267. 10.1016/j.chb.2016.12.032 (2017). [Google Scholar]
  • 4.Aljaberi, M. A. et al. A cross-sectional study on the impact of the COVID-19 pandemic on psychological outcomes: multiple indicators and multiple causes modeling. PLOS ONE17(11), e0277368. 10.1371/journal.pone.0277368 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Aljaberi, M. A. et al. Psychological toll of the COVID-19 pandemic: an in-depth exploration of anxiety, depression, and insomnia and the influence of quarantine measures on daily life. Healthcare11, 2418. 10.3390/healthcare11172418 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Mohammed, L. A. et al. Exploring factors affecting graduate students’ satisfaction toward e-learning in the era of the COVID-19 crisis. Eur. J. Investig. Health Psychol. Educ.12(8), 1121–1142. 10.3390/ejihpe12080079 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Starcevic, V. & Berle, D. Cyberchondria: towards a better understanding of excessive health-related Internet use. Expert Rev. Neurother.13(2), 205–213. 10.1586/ern.12.162 (2013). [DOI] [PubMed] [Google Scholar]
  • 8.Son, C., Hegde, S., Smith, A., Wang, X. & Sasangohar, F. Effects of COVID-19 on college students’ mental health in the United States: interview survey study. J. Med. Internet Res.22(9), e21279. 10.2196/21279 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Taiwan Internet Information Center. Taiwan internet report. https://report.twnic.tw/2022/assets/download/TWNIC_TaiwanInternetReport_2022_CH.pdf (2022).
  • 10.Kemp, S. Digital 2022: global overview report. https://wearesocial.com/cn/wp-content/uploads/sites/8/2022/01/DataReportal-GDR002-20220126-Digital-2022-Global-Overview-Report-Essentials-v02.pdf (2022).
  • 11.Abiddine, F. Z. E., Aljaberi, M. A., Gadelrab, H. F., Lin, C.-Y. & Muhammed, A. Mediated effects of insomnia in the association between problematic social media use and subjective well-being among university students during COVID-19 pandemic. Sleep Epidemiol.2, 100030 10.1016/j.sleepe.2022.100030 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Xu, D. Y., Ding, J. L. & Yang, Z. H. Pathogenesis of cyberchondria and its impact on individuals: a qualitative study. Chin. Gen. Pract.21(22), 2755–2759. 10.12114/j.issn.1007-9572.2018.22.022 (2018). [Google Scholar]
  • 13.Chen, W. & Lee, K. H. More than search? Informational and participatory ehealth behaviors.. Comput. Human Behav.30, 103–109. 10.1016/j.chb.2013.07.028 (2014). [Google Scholar]
  • 14.Smith, P. K., Fox, A. T., Davies, P. & Hamidi-Manesh, L. Cyberchondriacs. Int. J. Adolesc. Med. Health18(2), 209–213 (2006). [PubMed] [Google Scholar]
  • 15.Kim, H. N., Freddolino, P. P. & Greenhow, C. Older adults’ technology anxiety as a barrier to digital inclusion: a scoping review. Educ. Gerontol.49(12), 1021–1038. 10.1080/03601277.2023.2202080 (2023). [Google Scholar]
  • 16.McElroy, E. & Shevlin, M. The development and initial validation of the cyberchondria severity scale (CSS). J. Anxiety Disord.28(2), 259–265. 10.1016/j.janxdis.2013.12.007 (2014). [DOI] [PubMed] [Google Scholar]
  • 17.Case, D. Looking for information—a survey of research on information seeking, needs, and behavior(Academic Press, 2003).
  • 18.Montesi, M. Human information behavior during the Covid-19 health crisis. A literature review. Libr. Inf. Sci. Res.43(4), 10.1016/j.lisr.2021.101122 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Moorman, C. & Matulich, E. A model of consumers’ preventive health behaviors: the role of health motivation and health ability. J. Consum. Res.20(2), 208–228. 10.1086/209344 (1993). [Google Scholar]
  • 20.Jokić-Begić, N., Mikac, U., Čuržik, D. & Jokić, C. S. The development and validation of the short cyberchondria scale (SCS). J. Psychopathol. Behav. Assess.41, 662–676. 10.1007/s10862-019-09744-z (2019). [Google Scholar]
  • 21.Gray, N. J., Klein, J. D., Noyce, P. R., Sesselberg, T. S. & Cantrill, J. A. Health information-seeking behavior in adolescence: the place of the internet. Soc. Sci. Med.60(7), 1467–1478. 10.1016/j.socscimed.2004.08.010 (2005). [DOI] [PubMed] [Google Scholar]
  • 22.Starcevic, V. Cyberchondria: challenges of problematic online searches for health-related information. Psychother. Psychosom.86(3), 129–133. 10.1159/000465525 (2017). [DOI] [PubMed] [Google Scholar]
  • 23.Myrick, J. G. The role of emotions and social cognitive variables in online health information seeking processes and effects. Comput. Human Behav.68, 422–433. 10.1016/j.chb.2016.11.071 (2017). [Google Scholar]
  • 24.Bandura, A. Self-efficacy: toward a unifying theory of behavioral change. Psychol. Rev.84(2), 191–215. 10.1037//0033-295x.84.2.191 (1977). [DOI] [PubMed] [Google Scholar]
  • 25.Bandura, A. Self-efficacy: the exercise of control (W. H. Freeman and Company, 1997).
  • 26.Bandura, A. Self-efficacy mechanism in human agency. Am. Psychol.37(2), 122–147. 10.1037/0003-066X.37.2.122 (1982). [Google Scholar]
  • 27.Dugas, M. J. et al. Intolerance of uncertainty and information processing: evidence of biased recall and interpretations. Cognit. Ther. Res.29(1), 57–70. 10.1007/s10608-005-1648-9 (2005). [Google Scholar]
  • 28.Gorsuch, R. L. Factor analysis (Lawrence Erlbaum, 1983) 10.4324/9780203781098.
  • 29.Department of Statistics, Ministry of Education. Number of students by tertiary institution. https://data.gov.tw/dataset/6231#r3 (2019).
  • 30.Hair, J., Black, W., Babin, B. & Anderson, R. Multivariate data analysis (Prentice Hall, New Jersey, 2019). [Google Scholar]
  • 31.Alareqe, N. A. et al. Validity of adult psychopathology model using psychiatric patient sample from a developing country: confirmatory factor analysis. Ment. Illn.2022, 9594914. 10.1155/2022/9594914 (2022). [Google Scholar]
  • 32.White, R. & Horvitz, E. Cyberchondria: studies of the escalation of medical concerns in web search. ACM Trans. Inf. Syst.27(4), 23. 10.1145/1629096.1629101 (2009). [Google Scholar]
  • 33.Batıgün, A. D., Gor, N., Komurcu, B. & Erturk, I. S. Cyberchondria scale (CS): development, validity and reliability study. J. Psychiatry Neurol. Sci.31, 148–162. 10.5350/DAJPN2018310203 (2018). [Google Scholar]
  • 34.Clark, L. A. & Watson, D. Constructing validity: new developments in creating objective measuring instruments. Psychol. Assess.31(12), 1412–1427. 10.1037/pas0000626 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Bagozzi, R. P. & Yi, Y. On the evaluation of structural equation models. J. Acad. Mark. Sci.16(1), 74–94. 10.1007/BF02723327 (1988). [Google Scholar]
  • 36.Yu, M. N. Latent variable models: applications of SIMPLIS (Higher Education Press Company, 2006).
  • 37.Hsu, W. C. Undergraduate students’ online health information-seeking behavior during the COVID-19 pandemic. Int. J. Environ. Res. Public Health18(24), 13250. 10.3390/ijerph182413250 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Hsu, W. C., Chen, S. F. & Ho, C. J. Information among college students: an analysis from the health literacy perspective. J Health Promot Health Educ35, 1–22 (2011). [Google Scholar]
  • 39.Lee, O. E., Kim, D. H., Lee, H. & Beum, K. A. Information communication technology use to improve ehealth literacy, technophobia, and social connection among community dwelling older adults. Educ. Gerontol.48(10), 445–457. 10.1080/03601277.2022.2045727 (2022). [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The datasets generated and/or analysed during the current study are not publicly available because they contain information about the original materials used and can be used to copy works of art but are available from the corresponding author on reasonable request.

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES