Understanding cyberchondria in pregnant women: longitudinal assessment of risk factors, triggers, and outcomes

Abstract

Pregnancy often triggers anxiety and health concerns in women, leading many to search for health information online. Excessive, compulsive, and repetitive online health research, accompanied by heightened anxiety, can result in cyberchondria. This study aimed to explore the risk factors, triggers, and outcomes of cyberchondria in pregnant women. A total of 149 participants completed an online questionnaire longitudinally across three stages of pregnancy: early (14-19 weeks), mid (24-29 weeks), and late pregnancy (34-39 weeks). The findings revealed that health anxiety and the cognitive component of anxiety sensitivity are risk factors for cyberchondria during pregnancy. Pregnancy concerns related to motherhood emerged as triggers for cyberchondria. While a connection between cyberchondria and fear of birth was observed, fear of birth did not appear to be a direct outcome of cyberchondria. These results highlight the importance of addressing health anxiety, cognitive anxiety sensitivity and motherhood concerns in prenatal care and support interventions. Understanding the factors contributing to cyberchondria in pregnant women can assist healthcare professionals in providing targeted support and resources to mitigate excessive online health searching behaviors and alleviate anxiety during pregnancy.

Keywords:

• Cyberchondria in pregnancy
• health anxiety
• anxiety sensitivity
• pregnancy-specific anxiety
• fear of birth

Introduction

Pregnancy is a sensitive period characterized by numerous changes that necessitate women’s adaptation [1]. Consequently, it is common for women to experience pregnancy-related worries and anxiety [2]. In an effort to alleviate their anxiety, many women turn to the Internet, making it a primary medium for accessing pregnancy-related information [3]. When experiencing anxiety, some women fall into a maladaptive pattern of online searches, leading to a condition known as cyberchondria [4]. Cyberchondria is defined as an excessive, compulsive and repetitive online health research accompanied by increased distress [5].

According to cognitive-behavioral models of cyberchondria [6,7], previous health experiences contribute to the formation of health beliefs as a predisposition to vulnerability for cyberchondria. These trait characteristics, such as health anxiety and anxiety sensitivity, serve as risk factors for developing cyberchondria. Additionally, the decision to search for health information is influenced by current health circumstances, known as triggers for online research. For instance, the perception of potential health threats and the experience of worrisome symptoms can elicit aversive affective states, prompting individuals to turn to online research [6,7]. According to these models, online health research can lead to different cognitive, behavioral, and emotional consequences. In some cases, research may alleviate anxiety, provide reassurance, and lead to the termination of the search process. However, the habit of internet use can persist due to negative reinforcement. Alternatively, online health research can exacerbate anxiety and distress, resulting in decreased functioning, rumination, and a propensity for further online health research [6,7].

Research systematically demonstrates a strong association between health anxiety, anxiety sensitivity, and cyberchondria, indicating that these trait characteristics serve as predictors of cyberchondria [7]. Individuals with health anxiety exhibit excessive worry regarding their health, even in the absence of actual illness, which increases their likelihood of engaging in online research [5]. Pregnant women who repeatedly search for the same health information online exhibit higher levels of health anxiety [8]. Anxiety sensitivity is characterized by heightened anxiety in response to anxiety-related symptoms, accompanied by a person’s belief that these symptoms may have dire consequences [7]. This construct encompasses three distinct dimensions, each reflecting different types of concerns: cognitive (related to mental functioning), physical (concerning physical effects), and social (pertaining to worries about social rejection due to outwardly visible anxiety symptoms) [7]. Individuals experiencing anxiety sensitivity often resort to online research due to their belief that these anxiety symptoms might lead to harmful outcomes [7]. All dimensions of anxiety sensitivity are interconnected with cyberchondria, potentially serving as risk factors for its development [7]. Anxiety sensitivity has been identified as a predictor of distress and worries during pregnancy, childbirth, and postpartum [9].

Pregnancy is a period characterized by a multitude of concerns, including the health of the child, personal health, childbirth, finances, close relationships, appearance, and parenthood. Consequently, it is unsurprising that many women experience pregnancy-specific anxiety [2]. For health-anxious and anxiety-sensitive women, the significant life changes during pregnancy and heightened focus on health and physical sensations, often perceived as threatening, can trigger online research. If pregnant women’s online research becomes accompanied by heightened anxiety and becomes excessive, it may indicate the presence of cyberchondria [5]. As a result, this can lead to even greater worry and fear surrounding pregnancy and childbirth [10].

Research aim

We aimed to examine the risk factors, triggers and outcomes of cyberchondria in pregnant women. Our hypothesis was that trait markers such as health anxiety and anxiety sensitivity would be identified as risk factors, current pregnancy concerns would serve as a trigger, and fears related to childbirth would be an outcome of cyberchondria.

Materials and methods

Participants and procedure

This longitudinal study collected data at three time points: early pregnancy (14-19 weeks), mid-pregnancy (24-29 weeks), and late pregnancy (34-39 weeks). Initially, 160 pregnant women agreed to participate in our research. However, 11 provided only sociodemographic information without completing subsequent assessments. In first measurement point, we had 110 participants who completed the full questionnaire and shared their e-mail contacts for continued participation. Out of these, 100 engaged in the second measurement point, and 82 continued to the third (Table 1).

Table 1. Descriptive statistics by measurement points.

		1.early pregnancy		2. mid-pregnancy		3. late pregnancy
N		149		100		82
		M	SD	M	SD	M	SD
Age		30.1	4.7	30.2	4.5	30.7	4.4
Pregnancy weeks		16.7	1.3	27.3	1.6	37.2	1.5
Education level		n	%	n	%	n	%
	Elementary and high school	62	41.6	40	40.0	28	34.1
	University	87	58.4	60	60.0	54	65.9
Parity	Primiparous	60	40.3	39	39.0	36	43.9
	Multiparous	89	59.7	61	61.0	46	56.1

Participants were recruited online, by sharing the research link in platforms frequented by pregnant women, and offline, by invitations placed in the waiting rooms of the Department of Gynecology and Obstetrics at the University Hospital Center Zagreb and private gynecological clinics in Croatia. Medical staff at these clinics invited patients to join the study through the research link.

During the first measurement point, participants indicated their willingness to participate in future measurements and received e-mail invitations based on the provided contact. We initially recruited participants based on two criteria: being over 18 years old and within the appropriate pregnancy weeks. Subsequently, we maintained this group across the second and third measurement points, with a continued emphasis on their pregnancy weeks. Informed consent was obtained before filling out the questionnaire.

The research was approved by the ethics committee of the Faculty of Humanities and Social Sciences, Department of Psychology in Zagreb (EPOP-2021-023) and the ethics committee of the University Hospital Center Zagreb (02/21AG).

Measures

The questionnaire included the same measures at each measurement point: socio-demographic and obstetric questions, Internet use during pregnancy, and the following scales:

Health anxiety was measured by the Short Health Anxiety Inventory (SHAI) [11]. The inventory consists of two subscales: Illness Likelihood (health care and sensitivity to sensations and bodily changes; 14 items) and Negative Consequences (the expected severity of the consequences of the disease; 4 items). Each item has four offered answers that are scored from 0 to 3.

Anxiety sensitivity was measured with the Anxiety Sensitivity Index (ASI-3) [12]. This scale has 18 items and contains three subscales – Physical Concerns (the fear of physical symptoms of anxiety); Cognitive Concerns (the fear of cognitive symptoms of anxiety); and Social Concerns (the fear of public disclosure of anxiety), all answered on a scale from 0 (very little) to 4 (very much).

Pregnancy-specific anxiety was measured with the Pregnancy Concerns Scale (PCS) [13] which measures 16 specific concerns during pregnancy for which the participant must estimate how much it worried her in the last month on a scale from 0 (didn’t worry me at all) to 3 (extremely worried me). The authors validated the scale and established a four-factor structure in one time point with multiple cross loadings [13]. However, our exploratory factor analysis revealed that while the factor structure remained consistent in the corresponding trimester, considering all three time points, the structure is better described by five subscales: Health Concerns, Motherhood Concerns, Financial Concerns, Social Relations Concerns, and Concerns about Looks. The scale is available on request from the authors. It is important to note that although the PCS was included in the most recent systematic review of pregnancy-related anxiety scales worldwide and its cultural significance was recognized [14], it has not been validated in English for use with participants in low- or middle-income countries, unlike scales such as the Pregnancy-Related Anxiety Questionnaire, Cambridge Worry Scale, Tilburg Pregnancy Distress Scale, and Pregnancy-Related Anxiety Scale, which have been validated for use in other populations [14]. The authors suggest that further validation of pregnancy-related anxiety scales in different languages and cultural contexts, including the PCS, should be pursued in future research [14].

Cyberchondria was measured with the Short Cyberchondria Scale (SCS) [15]. Its four items measure negative emotional reactions to health information online and compulsion to search further, and are answered on a scale from 1 (strongly disagree) to 5 (strongly agree).

Fear of birth was assessed using the Fear of Birth Scale (FOBS) [16], consisting of two items. The task was to indicate the level of anxiety and fear about the upcoming birth on a visual analog scale ranging from 0 (calm/no fear) to 100 (worried/strong fear). The total score was calculated as the mean of the two items divided by 100.

Reliability ranged from .70 to .95 (McDonald’s ωt_ot and Cronbach α; Appendix A), except for the Social Relations Concerns subscale, which was not included in the analyses due to low reliability.

Data analysis

Total scores were calculated as means of available data for each subscale and compared across three time points using Friedman’s repeated measures analysis of variance and Durbin-Conover test post-hoc. Spearman correlations were calculated for each time point separately. These analyses were done due to deviations from normality and on all data available for a specific analysis.

To see if cyberchondria is predicted first by health anxiety/anxiety sensitivity and then by pregnancy concerns, and if it predicts fear of birth, we tested a sequential mediation. It was examined by comparing two series of nested models in a cross-lagged path framework [17], one series with health anxiety as the predictor, and the other with anxiety sensitivity. The models compared were: (a) autoregressions model including autoregression of each variable from the previous time point and covariation within the same time point, (b) model with additional lag-2 autoregressions, (c) the full mediation model with regressions from predictors in the first time point to the first and second mediators and the outcome in the second time point; from the first mediators in the first time point to the second mediator and the outcome in the second time point; and from the second mediator in the first time point to the outcome in the second time point, as well as all these regressions from the second time point to the third, and (d) reciprocal effects model with those regressions in the opposite direction (from outcome/mediator to mediator/predictor).

Full information robust maximum likelihood estimation was used due to deviations from normality and missing data. Adequate model fit was determined by CFI > .90, and RMSEA and SRMR < .08 [17]. The analyses were performed using R 4.0.2 (lavaan and psych) and jamovi 2.3.26.

Results

Mean levels, changes and correlations

Most scores were generally low to moderate, except for health-related pregnancy concerns and fear of birth, which were closer to the midpoint of the scale (Appendix A). There were no significant changes across three time points in all variables but two. The motherhood-related concerns were lower in the first trimester compared to the second (Durbin-Conover statistic = 2.04, p = .044) and third (Durbin-Conover statistic = 2.58, p = .011), with no difference between the second and third (Durbin-Conover statistic = 0.54, p = .588). No changes in fear of birth were found between the first and second trimester (Durbin-Conover statistic = 1.00, p = .319), nor second and third trimester (Durbin-Conover statistic = 1.78, p = .077). However, fear of birth was lower in the first trimester compared to the third (Durbin-Conover statistic = 2.78, p = .006). Most of the variables were correlated, and all of them positively (Appendix C).

Mediation models

Concerning the model with health anxiety as a predictor, adding lag-2 autoregressions and the mediation regressions significantly improved the fit and resulted in a well-fitting model (Table 2). The addition of reciprocal regressions did not significantly change the fit, so the mediation model was chosen as the best (Figure 1a).

Figure 1. Cross lagged models of sequential mediation: health anxiety (a)/anxiety sensitivity (b).

Only paths significant at p < .05 and standardized loadings are displayed. Thick lines represent expected regressions, and thin ones reciprocal (R). Full lines represent paths from first to second time point (1-2), and dashed lines paths from second to third (2-3). Variance explained in each variable is shown below it for the second (2)/third (3) time point. Suppressor paths are marked by s.

Read the detailed description of this figure

HA = Health Anxiety: Illness Likelihood (IL), Negative Consequences (NC); AS = Anxiety Sensitivity: Physical (PHY), Cognitive (COG), Social Concerns (SOC); PC = Pregnancy Concerns: Health (HEALTH), Motherhood (MOTHER), Financial (FIN), Concerns about Looks (LOOKS); CYB = Cyberchondria; FOB = Fear of Birth.

Table 2. Fit and differences between cross lagged models of sequential mediation in three time points (N = 149).

Predictors	Model	Model fit					Model difference
		χ^2	df	CFI	RMSEA [90% CI]	SRMR	χ^2	df
Health anxiety	Autoregressions	297.50***	176	.917	.069 [.055, .082]	.135
	Lag −2 autoregressions	250.25***	168	.945	.057 [.042, .072]	.115	37.60***	8
	Mediation	172.96***	126	.968	.05 [.029, .067]	.078	76.98***	42
	Reciprocal	134.18***	84	.968	.062 [.041, .081]	.056	41.13	42
Anxiety sensitivity	Autoregressions	463.44***	225	.883	.083 [.072, .094]	.148
	Lag −2 autoregressions	389.33***	216	.917	.071 [.06, .083]	.131	48.55***	9
	Mediation	274.10***	162	.947	.066 [.052, .079]	.109	114.72***	54
	Reciprocal	202.39***	108	.957	.072 [.057, .088]	.066	74.19*	54

*** p < .001. * p < .05.

Concerning the model with anxiety sensitivity as a predictor, including lag-2 autoregression, the mediation regressions and the reciprocal regressions significantly improved the fit (Table 2). Therefore, the reciprocal model was chosen as the best (Figure 1b).

When looking at both models, we can see that cyberchondria was consistently explained by motherhood related concerns and also by both dimensions of health anxiety and cognitive anxiety sensitivity. Cyberchondria did not predict fear of birth independently, which was explained only by health anxiety (illness likelihood). As for pregnancy specific concerns, they were mostly related to anxiety sensitivity (health, motherhood and financial concerns to physical and cognitive anxiety sensitivity), with the exception of financial concerns predicted by health anxiety (illness likelihood) and motherhood concerns predicted by fear of birth. Suppression effects, indicated by a negative regression coefficient in cases where bivariate correlations were positive or insignificant, were not interpreted due to their indeterminacy.

For most of these variables, about third to half of the variance was explained (R² = .35 − .67). However, most of the regression effects were small (|β| = .13 − .38), and a large part of the variance was explained by autoregressions from the same variable (Appendix B).

Discussion

This study aimed to investigate cyberchondria in pregnant women, exploring its risk factors, triggers, and outcomes. Research hypotheses were partially confirmed and several significant findings emerged. We found that health anxiety and anxiety sensitivity were risk factors for cyberchondria, and pregnancy-specific anxiety acted as a trigger for this phenomenon. However, we did not find evidence to support fear of birth as an outcome of cyberchondria.

Pregnancy often brings heightened anxiety, leading women to rely on the Internet for health information [3]. According to cognitive-behavioral models of cyberchondria, previous health experiences contribute to the formation of health beliefs and make some people more prone for development of cyberchondria [6,7]. Our findings confirm that health anxiety and anxiety sensitivity are stable traits that contribute to the development of cyberchondria. However, these constructs are distinct [7], they play different roles in predicting cyberchondria and fear of birth in pregnant women. Health-anxious individuals perceive their health as vulnerable and tend to misinterpret physiological symptoms in a threatening manner [18]. Our results show that pregnant women who are health-anxious will be more prone to cyberchondria. Those women probably hope that they will be able to prevent future illness by gathering information. Consequently, they engage in online searches to alleviate their concerns, which leads to a cycle of increased distress and compulsive research [7]. Notably, this relationship is not necessarily triggered by current health circumstances, such as pregnancy, suggesting that health anxiety plays a broader role in driving cyberchondria.

Additionally, our research highlighted anxiety sensitivity, specifically its cognitive component, as another risk factor for cyberchondria. This finding is consistent with previous research [19]. Anxiety-sensitive individuals have more specific worries regarding the negative consequences of experiencing anxiety symptoms [7]. Pregnant women who worry about the physical consequences of anxiety symptoms tend to have heightened concerns about their own health and the health of their baby, which supports findings that physical anxiety sensitivity contributes to health worries [19]. Interestingly, we found that women who express concerns about the cognitive consequences of anxiety, such as worries about their psychological well-being, concentration, and control over thoughts, experience increased concerns about motherhood and cyberchondria during late pregnancy. In this case, the current health circumstances, namely pregnancy, act as triggers for cyberchondria. It appears that women who are worried about their psychological functioning are also more likely to be concerned about handling childbirth and postpartum period [10]. Moreover, those concerned with motherhood will tend to search for health information online and may develop cyberchondria. Anxious and neurotic women often lack confidence in their parenting abilities and may feel ill-equipped to adequately care for their babies [20] which may result in excessive online research. Our findings shed light on the specific topics that pregnant women tend to search for online. We observed that pregnancy concerns related to motherhood, such as postpartum issues, were particularly significant for cyberchondria. This suggests that women may focus more on potential future health concerns rather than current symptoms during pregnancy.

Although fear of birth was not identified as an outcome of cyberchondria, positive correlation between cyberchondria and fear of birth was found. This suggests that women who engage in excessive online research are also more likely to have higher levels of fear of birth. Based on our models, it can be concluded that women who are more health-anxious also worry more about childbirth. Health-anxious women view their health as vulnerable and impaired, making them more prone to belief that birth is a dangerous medical event that can have terrifying effects on health [21].

Conclusion

This study represents the first comprehensive exploration of cyberchondria in pregnant women. Findings emphasize targeted interventions for birth-related fears in women with health anxiety. Interventions targeting cyberchondria should prioritize women experiencing health anxiety and cognitive anxiety sensitivity. Special attention should be given to women with heightened anxieties surrounding motherhood and postpartum, as our study reveals that these concerns act as triggers for cyberchondria. It is worth considering that these women may continue excessive online health research after giving birth, as one study suggests [22]. While the healthcare system typically provides enough information during pregnancy, it tends to overlook the postpartum period, which causes some women to seek information online [23]. Consequently, guidance on post-childbirth recovery and motherhood may be insufficient. The healthcare system primarily prioritizes the physical health of women, often overlooking the mother’s overall well-being. Healthcare professionals should provide post-childbirth guidance, to mitigate the adverse effects of online health searches.

It is important to acknowledge research limitations, including online data collection, participant self-selection, small sample size, and sample non-representativeness. These factors may impact validity and generalizability of findings.

Geolocation information

The research was conducted in Croatia, EU.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The data that support the findings of this study are available at Zenodo (doi:10.5281/zenodo.8146859) upon reasonable request.

Additional information

Funding

This work was supported by the Faculty of Humanities and Social Sciences through the Scholarship for Excellence, under Grant 643-02/15-01/79.

Appendices

Appendix A:

Mean levels of health anxiety, anxiety sensitivity, pregnancy concerns, cyberchondria and fear of birth across the three time points

Table

Variable	Time point	N	ωtot/α	M (SD)	Median	Friedman χ² (p)
HA IL	1	135	.86/ .86	0.7 (0.4)	0.6	.01 (.996)
	2	97	.88/ .88	0.7 (0.4)	0.7
	3	77	.87/ .86	0.7 (0.4)	0.7
HA NC	1	128	.80/ .79	0.7 (0.6)	0.7	1.73 (.422)
	2	95	.80/ .80	0.7 (0.5)	0.8
	3	77	.79/ .79	0.7 (0.5)	0.5
AS PHY	1	123	.89/ .88	0.8 (0.8)	0.5	4.22 (.121)
	2	95	.89/ .89	0.8 (0.7)	0.5
	3	77	.91/ .90	0.8 (0.8)	0.5
AS COG	1	123	.91/ .91	0.6 (0.8)	0.3	1.53 (.465)
	2	95	.89/ .89	0.6 (0.7)	0.3
	3	77	.93/ .93	0.6 (0.8)	0.3
AS SOC	1	123	.81/ .80	0.9 (0.8)	0.7	4.29 (.117)
	2	95	.83/ .83	1.0 (0.7)	0.8
	3	77	.88/ .88	1.1 (0.8)	0.8
PC HEALTH	1	149	.87/ .87	1.6 (0.7)	1.5	3.83 (.148)
	2	98	.88/ .88	1.5 (0.7)	1.3
	3	77	.88/ .87	1.4 (0.7)	1.3
PC MOTHER	1	149	.70/ .70	0.9 (0.7)	1.0	7.14 (.028)
	2	98	.63/ .62	1.1 (0.7)	1.0
	3	77	.73/ .73	1.2 (0.8)	1.0
PC FIN	1	149	.74/ .74	0.9 (0.7)	0.5	2.70 (.26)
	2	98	.80/ .80	1.0 (0.8)	0.8
	3	77	.82/ .82	1.0 (0.9)	1.0
PC SOC	1	148	.48/ .48	0.4 (0.5)	0.0	.22 (.894)
	2	98	.55/ .55	0.5 (0.7)	0.0
	3	77	.58/ .58	0.5 (0.7)	0.5
PC LOOKS	1	149	.82/ .82	0.6 (0.8)	0.5	.53 (.767)
	2	98	.72/ .72	0.6 (0.7)	0.5
	3	77	.84/ .84	0.6 (0.7)	0.5
CYB	1	141	.85/ .84	2.3 (1.0)	2.3	1.55 (.46)
	2	98	.84/ .83	2.2 (0.8)	2.3
	3	77	.88/ .87	2.4 (1.0)	2.3
FOB	1	119	.89/ .89	0.5 (0.3)	0.5	7.64 (.022)
	2	95	.95/ .95	0.5 (0.3)	0.5
	3	76	.93/ .93	0.6 (0.3)	0.6

Note. The theoretical range for HA and PC was 0-3, for AS 0-4, for CYB 1-5, and for FOB 0-1. HA = Short Health Anxiety Inventory dimensions: Illness Likelihood (IL) and Negative Consequences (NC); AS = Anxiety Sensitivity Index dimensions: Physical (PHY), Cognitive (COG), and Social Concerns (SOC); PC = Pregnancy Concerns Scale dimensions: Health (HEALTH), Motherhood (MOTHER), Financial (FIN), Social Relations Concerns (SOC), and Concerns about Looks (LOOKS); CYB = Short Cyberchondria Scale; FOB = Fear of Birth Scale.

 

 

Appendix B:

Standardized autoregression coefficients in the final cross lagged models

Table

Model	Variable	Time points
		2 - 1	3 - 2	3 - 1
Health anxiety as the predictor	HA IL	.74***	.55***	.30**
	HA NC	.72***	.52***	.29**
	PC HEALTH	.63***	.62***	.15
	PC MOTHER	.61***	.43***	.24*
	PC FIN	.65***	.57***	.17
	PC LOOKS	.70***	.21	.52***
	CYB	.54***	.25**	.41***
	FOB	.50***	.58***	.05
Anxiety sensitivity as the predictor	AS PHY	.71***	.39***	.36***
	AS COG	.58***	.38***	.39***
	AS SOC	.61***	.55***	.19*
	PC HEALTH	.50***	.69***	.09
	PC MOTHER	.51***	.38***	.29**
	PC FIN	.66***	.58***	.15
	PC LOOKS	.72***	.31*	.44**
	CYB	.61***	.24*	.43***
	FOB	.51***	.66***	.07

Note. HA = Short Health Anxiety Inventory dimensions: Illness Likelihood (IL) and Negative Consequences (NC); AS = Anxiety Sensitivity Index dimensions: Physical (PHY), Cognitive (COG), and Social Concerns (SOC); PC = Pregnancy Concerns Scale dimensions: Health (HEALTH), Motherhood (MOTHER), Financial (FIN), and Concerns about Looks (LOOKS); CYB = Short Cyberchondria Scale; FOB = Fear of Birth Scale.

*** p < .001. ** p < .01. * p < .05.

Appendix C:

Correlations between predictors, mediators and the outcome in the first (above the main diagonal, n = 118 - 149) and the second/third time point (below the diagonal, n = 94 - 98 for the second, and n = 76 - 77 for the third)

Table

Variables	Predictors					First mediator				Second mediator		Outcome
	HA IL	HA NC	AS PHY	AS COG	AS SOC	PC HEALTH	PC MOTHER	PC FIN	PC LOOKS	CYB	CYB +1	FOB	FOB +1
HA IL		.30***	.49***	.45***	.42***	.50***	.38***	.17*	.29***	.60***	.43***	.48***	.35***
HA NC	.34***/ .28*		.11	.27**	.33***	.21*	.22*	.14	.33***	.35***	.25*	.24**	.16
AS PHY	.44***/ .54***	.33**/ .35**		.56***	.41***	.29**	.12	.19*	.15	.34***	.18	.26**	.36***
AS COG	.49***/ .46***	.36***/ .31**	.57***/ .59***		.59***	.24**	.19*	.20*	.27**	.39***	.27**	.31***	.36***
AS SOC	.41***/ .34**	.30**/ .39***	.52***/ .51***	.57***/ .73***		.10	.30***	.20*	.44***	.31***	.28**	.22*	.19
PC HEALTH	.38***/ .43***	.08/ .13	.29**/ .45***	.26**/ .38***	.08/ .24*		.48***	.30***	.27**	.43***	.33***	.41***	.31**
PC MOTHER	.31**/ .40***	.21*/ .17	.28**/ .49***	.26*/ .47***	.16/ .29*	.50***/ .52***		.09	.52***	.26**	.34***	.59***	.34***
PC FIN	.01/ .12	.13/ .19	.05/ .31**	.12/ .32**	.09/ .30**	.42***/ .45***	.14/ .28*		.19*	.14	.04	.18	.23*
PC LOOKS	.34***/ .17	.2/ .29*	.12/ .17	.32**/ .21	.14/ .16	.30**/ .16	.35***/ .37**	.22*/ .20		.20*	.22*	.35***	.12
CYB	.51***/ .50***	.31**/ .21	.26*/ .33**	.41***/ .34**	.26*/ .37***	.31**/ .37**	.33***/ .31**	.12/ .10	.23*/ .10		.61***	.38***	.34***
CYB + 1	.51***/—	.29*/—	.28*/—	.46***/—	.44***/—	.22/—	.24*/—	.07/—	.10/—	.62***/—		.29**	.38***
FOB	.39***/ .43***	.22*/ .26*	.26*/ .51***	.37***/ .51***	.29**/ .35**	.41***/ .43***	.45***/ .60***	.20*/ .21	.33**/ .17	.38***/ .46***	.41***/—		.59***
FOB + 1	.52***/—	.26*/—	.38**/—	.45***/—	.33**/—	.34**/—	.44***/—	.16/—	.29*/—	.29*/—	.46***/—	.72***/—

Note. HA = Short Health Anxiety Inventory dimensions: Illness Likelihood (IL) and Negative Consequences (NC); AS = Anxiety Sensitivity Index dimensions: Physical (PHY), Cognitive (COG), and Social Concerns (SOC); PC = Pregnancy Concerns Scale dimensions: Health (HEALTH), Motherhood (MOTHER), Financial (FIN), and Concerns about Looks (LOOKS); CYB = Short Cyberchondria Scale; FOB = Fear of Birth Scale; +1 = variable measured in the following time point.

*** p < .001. ** p < .01. * p < .05.