Health anxiety by proxy among parents: Measuring health anxiety for self,child and partner and its associations with family functioning

Abstract

Health-related worries can be either self-directed or occur in close relationships by proxy (e.g., among parents, couples, or pet owners). Our aim was to test the feasibility of assessing health anxiety by proxy in parents and to examine the relationship of health anxiety with family well-being. Health anxiety by proxy was measured between parent-child and parent-partner relationships with adapted versions of the modified Short Health Anxiety Inventory. Parents reported significantly higher levels of health anxiety by proxy regarding their child than for themselves, t (585) = −3.61, p < .001. Exploratory factor analysis showed that each health anxiety questionnaire formed a distinct latent factor. Parenting stress and poorer family functioning had small associations with health anxiety towards one’s self, child, and partner (r = 0.15 to r = 0.20). Further research is needed to determine clinical relevance and its impact on health behaviors and functioning within the family unit.

Keywords

Health anxiety parents parent-child couples family functioning

Introduction

Health anxiety is characterized by disproportionate concerns about one’s own health and a preoccupation with having or acquiring a serious disease accompanied by high levels of stress or misinterpretation of bodily symptoms as threatful (Asmundson et al., 2010; Rask et al., 2020). Severe health anxiety is associated with increased health care utilization and negative well-being outcomes. Epidemiological studies suggest that health anxiety concerns approximately 2.1 to 13.1% of the general population (Weck et al., 2014). Affected individuals exhibit dysfunctional health behaviors, ranging from maladaptive coping strategies like avoidance of health consultations to excessive contact with health services, which has an impact on health care system costs (Fink et al., 2010). Elevated health anxiety is associated with more functional impairment, disability, lower quality of life and higher risk for psychiatric treatment (Bleichhard and Hiller, 2007; Sunderland et al., 2013; Horenstein and Heimberg, 2020).

Recent research has expanded the study of health anxiety beyond one’s own health. Health anxiety that relates to another individual has been described as health anxiety by proxy and has been mainly examined among parent-child dyads (Aziz et al., 2018; Lockhart, 2016; Thorgaard et al., 2017a) and among expectant mothers (Harpel, 2008; Reiser and Wright, 2019). Affected parents show dysfunctional cognitive, behavioral, and emotional outcomes such as negative illness perceptions, frequent presentation of the child to the doctor and elevated worrying about their child’s health (Aziz et al., 2018; Lockhart, 2016; Thorgaard et al., 2017a). A qualitative study by Ingeman and colleagues (2024) explored parents’ perceptions and found three core themes: 1. Feelings of distress, which can include fear, shame and guilt. 2. Mistrust in others, including doubts about the child’s ability to report symptoms accurately, concerns about their partner’s decision-making ability regarding the child’s health, and uncertainties about healthcare professionals' opinions. 3. Making sense of their worries with ambivalence in relation to worrying and rationalization of those sorrows. Health anxiety by proxy has already been discussed for pet owners (Nakano et al., 2019; Oxley and Kogan, 2018), but to our knowledge has not been considered for any other relationship types than the ones described above (parents or expectant mothers).

Health anxiety by proxy is not part of any diagnostic system although it has been considered for potential relevance (Rask et al., 2020). One current gap is that validated measures of health anxiety by proxy are lacking. Ingeman et al. (2021) newly introduced the Health Anxiety by Proxy Scale (HAPYS) with 26 items on parental worries on children’s health. So far, adapted versions of existing anxiety inventories have been used for assessing health anxiety by proxy in parent-child constellations, including the Illness Attitude Scale (Kellner, 1987) in the study by Thorgaard et al. (2017a) and the Short Health Anxiety Inventory (SHAI, Salkovskis et al., 2002) in Reiser and Wright’s study (2019). Other assessment approaches have been used for subtypes such as pregnancy-related anxiety (Bayrampour et al., 2016; Hadfield et al., 2022) or parental distress caused by web searches related to a child’s medical condition (Barke and Doering, 2020). To the best of our knowledge, there was no study to date that has assessed different relationship constellations of health anxiety by proxy simultaneously. It is unclear whether certain relationship types have elevated levels compared to other types and whether elevated proxy anxiety co-occurs across different types of relationships within the family.

Another important consideration is how health anxiety by proxy relates to family functioning and processes within the family unit (e.g., couple satisfaction, parental stress, and family functioning). Family Systems Theory describes the link between family well-being and general anxiety in the family (Kerr and Bowen, 1988). We extend this approach to examine indicators of family relationships (couple satisfaction, parental stress, and family functioning) and their association with health anxiety by proxy among parents. Since anxiety disorders tend to aggregate in families (Lawrence et al., 2019; Shimada-Sugimoto et al., 2015; Telman et al., 2018), we expect a degree of co-occurrence in health anxiety across ties (i.e. parent-child, parent-parent) along with impairment in family well-being.

Our aim was first to compare the health anxiety and health anxiety by proxy inventories – a partner-to-partner version and a parent-to-child version by examining mean and item level differences in anxiety. We hypothesized the highest level of health anxiety in relation to one’s own child when compared to self or partner. A second aim was to verify the factor structure of the health anxiety by proxy inventories. We assumed that each of the three questionnaires would form distinct factors. The third aim was to examine the relationship between health anxiety, health anxiety by proxy and family well-being. We expected strong positive correlations among the different health anxiety types. Furthermore, small to medium positive associations of health anxiety by proxy with couple dissatisfaction and poor family functioning were hypothesized based on past literature which documents associations between general anxiety and family constructs.

Method

Procedure

An online survey was conducted after ethical approval was obtained by the Ethics Council of the University on June 21^st, 2020 (approval number: 2018-116). Recruitment lasted during a 6-week period between June 29^th to August 9^th, 2020. The study took place during the SARS-CoV-2 pandemic, a time characterized by globally heightened prevalence of anxiety symptoms (da Silva et al., 2021). At the time of data collection, Germany and Austria experienced relative stability, with regulations being relaxed and low infection numbers (Austrian Agency for Health and Food Safety, 2024; Robert Koch Institute, 2020).

Facebook groups and message boards with topics related to parenting, leisure activities or lifestyle were targeted for recruitment. In addition, we advertised on Instagram and Facebook for German-speaking parents between the age of 18 to 40 years as the target. Analysis of the HTTP referer revealed that 52.04% (n = 305) of the participants in the final sample reached the survey via Facebook. We also emailed more than 4000 kindergartens in Germany, Austria and Switzerland, asking them to share our call for participation.

Informed consent was obtained before data were collected. For eligibility, participants had to be at least 18 years old, parent of at least one child between the age of 0 and 6 years and had to be in an intimate relationship. The inclusion of parents with children between 0 and 6 years is based on the understanding that caregivers’ level of health anxiety may vary as children reach the developmental stage where they can effectively communicate their health complaints in a more detailed manner. To ensure that pre-existing illnesses do not influence the health anxiety levels under investigation, we implemented exclusion criteria that involved individuals, their partners, and their children having either chronic or acute diseases. By employing these criteria, we aimed to eliminate any potential confounding factors associated with existing health conditions. As an incentive, we offered the chance to receive one of 10 gift vouchers.

Participants

The online survey was accessed a total of 1526 times. After checking for eligibility criteria and data cleaning, 586 eligible parents were included in the analysis (see data cleaning). Demographic characteristics of participants are provided in Table 1.

Table 1.

Sociodemographic characteristics of participants.

	Full sample (N = 586)
	n	%
Sex
Women	527	89.9
Man	58	9.9
Other	1	0.2
Nationality
Austria	194	33.1
Germany	365	62.3
Switzerland	7	1.2
Other	20	3.6
Current health status
Very good	147	25.1
Good	322	54.8
Moderate	107	18.3
Bad	10	1.7
Marital status
Unmarried	146	24.9
Married	422	72.0
Divorced	18	3.1
Number of children
1	236	40.3
2	268	45.7
3	66	11.3
4	11	1.9
5 or more	5	0.8
Current pregnancy	24	4.1
Age of the youngest child
Below 1 year	135	23.0
1 year	83	14.2
2 years	99	16.9
3 years	102	17.4
4 years	76	13.0
5 years	62	10.6
6 years	29	4.9
Health status of the youngest child
Very good	440	75.1
Good	134	22.9
Moderate	10	1.7
Bad	2	0.4
Health status of the partner
Very good	167	28.5
Good	290	49.5
Moderate	108	18.4
Bad	21	3.6
Educational level
Without a degree	1	0.2
Lower secondary	36	6.1
Higher secondary	79	13.5
High school	124	21.1
University	346	59.0
Employment
Employed	324	55.3
Working in household	211	36.0
Other	51	8.7

Note. Percentages are rounded. Participants were on average 35.08 years old (SD = 5.37).

Data cleaning

The data set of 1526 entries was cleaned in four waves. First, 718 participants were excluded because they provided only demographic data, but did not continue the survey. Second, 128 participants dropped out because the information on at least one of the three anxiety questionnaires was missing. Third, 92 participants were excluded because they did not have a child between 0 and 6. Lastly, two participants were excluded because the response pattern did not seem plausible (e.g., all items endorsed at the extreme end or no items endorsed). This resulted in a sample of 586 eligible parents.

Measures

Demographics

The demographic characteristics of the participants that were collected included gender, age, nationality, occupation, marital status, health status and education. Further, we assessed length of relationship in years, age of the youngest child (categorized into below 1 year, 1 year, 2 years, 3 years, 4 years, 5 years, 6 years), sex of the youngest child, health status of the child, household size, total number of children, current pregnancy and health status of the partner.

The modified Short Health Anxiety Inventory

The German translation of the Modified Short Health Anxiety Inventory (mSHAI) by Bailer and colleagues (2013) is a self-rating scale of health anxiety and includes 14 items, each rated on 0 to 4. The inventory was originally published by Salkovskis et al. (2002) as Short Health Anxiety Inventory (SHAI) with a more elaborated response format and an additional subscale on perceived negative consequences. Individuals rate their level of trait health anxiety according to the past 6 months. Total scores range between 0 and 56. A higher score indicates higher levels of health anxiety. The scale is intended to measure health anxiety as a one-factor and continuous trait (Bailer and Witthöft, 2006). A meta-analysis has shown that the SHAI is a psychometrically strong instrument for assessing health anxiety (Alberts et al., 2013). In this sample, Cronbach’s α of the mSHAI was excellent (α = 0.94).

The modified Short Health Anxiety Inventory for Child

The modified Short Health Anxiety by Proxy Inventory for Child (mSHAI-Child) was designed to assess health anxiety by proxy regarding the own child. Each of the 14 original items of the mSHAI by Bailer et al. (2013) was reformulated to comply with the parent-child situation. For example, “I spend much of my time worrying about my health” was reformulated to “I spend much of my time worrying about my child’s health”. Answers were scored from 0 to four resulting in a score range of 0 to 56. Higher scores indicate higher levels of health anxiety by proxy towards the child. The mSHAI-Child showed excellent internal consistency in our sample (α = 0.93).

The modified Short Health Anxiety by Proxy Inventory for Partner

The modified Short Health Anxiety by Proxy Inventory for Partner (mSHAI-Partner) is derived from the 14 items of mSHAI by Bailer et al. (2013) to measure health anxiety by proxy concerning the intimate partner. Items were adapted to relate to partner’s health instead, for example “My family/friends would say I worry too much about my health” was reformulated to “My family/friends would say I worry too much about my partner’s health”. Possible scoring within 0 to 56 points, with higher scores reflecting higher levels of health anxiety by proxy towards the intimate partner. The internal consistency in the current sample was excellent (α = 0.94).

Couple Satisfaction Inventory

The Couple Satisfaction Inventory (CSI) by Funk and Rogge (2007) is a self-report measure for assessing relationship satisfaction in couples. The original version consists of 32 items, however, a shorter version with 16 items (CSI-16) demonstrated similar psychometric properties (Funk and Rogge, 2007). We used the German translation by Treffner and Foran (2017). Participants can score between the range 0 to 81. Higher scores indicated higher couple satisfaction. Scoring below 51.5 indicates substantial relationship dissatisfaction. The inventory was found to be highly reliable in our sample (α = .97).

Parenting Stress Scale

The Parental Stress Scale (PSS) by Berry and Jones (1995) is an assessment for measuring stress that results from having a child. Parents rate on 18 items different aspects of the parent-child relationship (e.g. satisfaction of being a parent). Scores can range from 18 to 90. A higher score indicates a higher level of parental stress. Cronbach’s coefficient of the PSS in this study was good (α = 0.86).

General Functioning Scale

The General Functioning Scale (GFS, Byles et al., 1988) is an extracted 12-item subscale of the McMaster Family Assessment Device (FAD, Epstein et al., 1983) to assess overall functioning in families. Each item is evaluated on a four-point Likert scale with scoring from one to 4. The sum of scores is divided by the number of items for calculating the average score. A score of one indicates healthy family functioning, while four represents extremely poor family functioning. In our sample, the GFS demonstrated good internal consistency (α = 0.87).

Data analytic strategy

IBM SPSS version 25 was used for data analysis. For hypothesis 1, nonparametric Friedman tests were carried out to compare the three health anxiety conditions at the overall and individual item level. Additionally, an independent sample t test was conducted for comparison between health anxiety towards self and health anxiety towards child. Regarding hypothesis 2, the factor structure of the mSHAI, mSHAI-Child and mSHAI-Partner was analyzed with an exploratory factor analysis (EFA). For hypothesis 3, correlations between health anxiety inventories and family variables (i.e. family functioning, couple satisfaction, parental stress) were examined.

Results

Preliminary analysis

Descriptive statistics for all health anxiety and family functioning scales can be found in Table 2. A nonparametric Friedman test of differences among repeated measures was conducted for testing each condition. There was a significant difference among the distributions of the three scores of health anxiety, χ² (2) = 85.30, p = < .001. Post hoc analysis with Dunn’s test revealed significant differences between all conditions after Bonferroni correction, p = < .001. The mean rank was 1.74 for health anxiety towards partner, 2.00 for own health anxiety and 2.26 for health anxiety towards child. Participants scored significantly higher on the mSHAI-Child than for mSHAI, t (585) = −3.61, p < .001.

Table 2.

Descriptives for health anxiety inventories and family functioning scales.

Scale	Mean	SD	Score range	Skewness	Kurtosis
1. mSHAI	14.05	10.69	0-56	1.14	1.38
2. mSHAI-Child	15.56	10.36	0-56	.75	.37
3. mSHAI-Partner	12.79	10.38	0-47	.67	.02
4. CSI-16	58.22	17.49	3-81	−1.06	.52
5. PSS	39.02	8.61	18-73	.22	.14
6. GFS	1.72	.53	1-3.75	1.00	.83

Note. N = 586. mSHAI = modified Short Health Anxiety Inventory; mSHAI-Child = modified Short Health Anxiety Inventory For Child; mSHAI-Partner = modified Short Health Anxiety Inventory For Partner; CSI-16 = Couple Satisfaction Index; PSS = Parenting Stress Scale; GFS = General Functioning Scale.

In addition, Friedman’s tests were conducted for each of the corresponding items between mSHAI, mSHAI-Child and mSHAI-Partner (Table 3). 12 out of 14 items showed significant differences across the three conditions (p < .05). Only item 5 (Afraid of a serious illness) and item 6 (Having mental images of illness) did not differ significantly (p > .05). The highest scoring for health anxiety on individual item level were all achieved in the proxy condition regarding child’s health, in particular, worry about health (item 1), attention to bodily sensations (item 3), wonder what bodily changes mean (item 10) and unexplained bodily sensations (item 13).

Table 3.

Individual item comparison for each health anxiety scale.

	mSHAI	mSHAI-C	mSHAI-P
Item	Mean (SD)	Mean (SD)	Mean (SD)	χ²	p*
1. Worry about health	1.33 (1.03)	1.59 (1.06)	1.23 (1.03)	58.47	<.001
2. Noticing aches/pain	1.17 (1.01)	1.24 (.98)	1.01 (.93)	30.51	<.001
3. Attention to bodily sensations	1.68 (1.06)	1.98 (1.09)	1.15 (1.05)	244.42	<.001
4. Resist to thoughts of illness	1.12 (1.04)	.99 (.98)	.79 (.90)	63.89	<.001
5. Afraid of a serious illness	1.14 (1.08)	1.16 (1.10)	1.08 (1.08)	5.76	.056
6. Having mental images of illness	.69 (.93)	.75 (.91)	.71 (.89)	3.63	.163
7. Distracting from thoughts about health	.78 (.95)	.79 (.91)	.70 (.84)	7.43	.024
8. Worries return after doctor visit	1.00 (1.10)	1.08 (1.06)	.90 (.97)	21.33	<.001
9. Considering of illness after hearing it	.67 (.88)	.62 (.80)	.53 (.71)	7.93	.019
10. Wonder what bodily changes mean	1.06 (1.02)	1.61 (1.19)	1.25 (1.13)	137.66	<.001
11. Risk for developing a serious disease	.82 (1.00)	.56 (.75)	.92 (1.04)	60.02	<.001
12. Thinking about having a serious disease	.75 (.96)	.56 (.78)	.75 (.91)	27.98	<.001
13. Unexplained bodily sensations	1.13 (1.07)	1.68 (1.16)	1.05 (.99)	201.09	<.001
14. Perception of family/friends	.71 (.96)	.94 (.99)	.72 (.84)	55.56	<.001

Note. N = 586. * Friedman test. mSHAI = modified Short Health Anxiety Inventory; mSHAI-C = modified Short Health Anxiety Inventory For Child; mSHAI-P = modified Short Health Anxiety Inventory For Partner.

Mothers scored significantly higher on health anxiety (mSHAI) than fathers (Welch’s t-test, t (85.21) = 2.70, p = .008). There were no statistically significant differences between mothers and fathers for health anxiety regarding the child, (t (583) = 1.00, p = .31) or for health anxiety towards the partner, (t (583) = 0.74, p = .45). Further, we asked parents with more than one child (n = 350) whether they would have answered the questions about the mSHAI-Child differently if it were not related to their youngest only but to all children. Only 12.1% (41/340) stated that they would have answered differently.

Factor structure of the mSHAI, mSHAI-Child and mSHAI-Partner

Exploratory factor analysis was applied on all 42 items of the health anxiety inventories (see Table 4). The Kaiser-Meyer-Olkin measure of sampling adequacy was good (0.95). Bartlett test of sphericity was significant (p < .001). Varimax (orthogonal) rotation was applied. The Kaiser criteria and scree plot retained a 5-component solution with initial eigenvalues of 17.21, 4.13, 3.41, 2.05, and 1.17, which accounted for 66.63% of the total variance.

Table 4.

Exploratory factor analysis of health anxiety items.

	F1	F2	F3	F4	F5	Uniqueness
mSHAI_1		0.640				0.481
mSHAI_2		0.596			0.408	0.464
mSHAI_3		0.541			0.409	0.506
mSHAI_4		0.695				0.405
mSHAI_5		0.771				0.327
mSHAI_6		0.700				0.383
mSHAI_7		0.764				0.331
mSHAI_8		0.688				0.416
mSHAI_9		0.691				0.441
mSHAI_10		0.643				0.416
mSHAI_11		0.677				0.435
mSHAI_12		0.812				0.242
mSHAI_13		0.632		0.410		0.365
mSHAI_14		0.745				0.361
mSHAI-Child_1			0.630			0.408
mSHAI-Child_2			0.569		0.467	0.381
mSHAI-Child_3			0.439		0.482	0.475
mSHAI-Child_4			0.762			0.278
mSHAI-Child_5			0.755			0.331
mSHAI-Child_6			0.784			0.257
mSHAI-Child_7			0.757			0.303
mSHAI-Child_8			0.575			0.519
mSHAI-Child_9			0.707			0.368
mSHAI-Child_10			0.492	0.510		0.417
mSHAI-Child_11			0.670			0.399
mSHAI-Child_12			0.739			0.287
mSHAI-Child_13			0.492	0.519		0.415
mSHAI-Child_14			0.629			0.401
mSHAI-Partner_1	0.749					0.359
mSHAI-Partner_2	0.665					0.410
mSHAI-Partner_3	0.651					0.426
mSHAI-Partner_4	0.792					0.267
mSHAI-Partner_5	0.758					0.314
mSHAI-Partner_6	0.813					0.243
mSHAI-Partner_7	0.782					0.284
mSHAI-Partner_8	0.587					0.468
mSHAI-Partner_9	0.639					0.411
mSHAI-Partner_10	0.535			0.555		0.340
mSHAI-Partner_11	0.738					0.399
mSHAI-Partner_12	0.768					0.315
mSHAI-Partner_13	0.570			0.450		0.406
mSHAI-Partner_14	0.629					0.448

Note. Applied rotation method is varimax. Factor loadings below 0.4 are omitted. mSHAI = modified Short Health Anxiety Inventory; mSHAI-Child = modified Short Health Anxiety Inventory For Child; mSHAI-Partner = modified Short Health Anxiety Inventory For Partner.

Each item of the respective questionnaires (mSHAI, mSHAI-C, mSHAI-P) loaded sufficiently on a separate factor without cross loadings (>0.40), resulting in three definable factors. However, all items in relation to bodily sensations had cross-loadings on factor 4 or factor 5. This includes noticing aches/pain (item two in mSHAI and mSHAI-C), attention to bodily sensations (item three in mSHAI and mSHAI-C), wonder what bodily changes mean (item 10 in mSHAI-C and mSHAI-P) and unexplained bodily sensations (item 13 in mSHAI, mSHAI-C and mSHAI-P).

Correlations of health anxiety with family variables

Table 5 presents the correlations among study variables. The mSHAI showed significant correlations with mSHAI-Child (r = 0.54) and mSHAI-Partner (r = 0.50). There was also a strong positive correlation between both health anxiety by proxy inventories, mSHAI-Child and mSHAI-Partner, r = 0.58. Parental age (r = −0.15) and age of the youngest child (r = −0.13) were negatively correlated with the mSHAI-Child, suggesting higher health anxiety levels by proxy for younger parents and for those with younger children. No association was found with the number of children in the household.

Table 5.

Correlations for health anxiety measures, family variables and stress inventories.

Variable	1	2	3	4	5	6	7	8	9	10
1. mSHAI (n=586)	—
2. mSHAI-Child (n=586)	.54**	—
3. mSHAI-Partner (n=586)	.50**	.58**	—
4. Parental age (n=586)	−.09*	−.15**	−.08*	—
5. Relationship duration (n=583)	−.12**	−.16**	−.12**	.45**	—
6. Number of children (n=586)	−.02	−.07	−.03	.21**	.24**	—
7. Age of the youngest child (n=586)	.04	−.13**	.06	.36**	.23**	−.06	—
8. CSI-16 (n=530)	−.14**	−.13**	−.08	−.09*	.00	.01	−.14**	—
9. PSS (n=558)	.18**	.20**	.15**	−.00	−.09*	.04	−.00	−.31**	—
10. GFS (n=515)	.17**	.17**	.15**	.04	−.02	−.01	.06	−.81**	.35**	—

Note. Pearson-Correlation for all variables, except age of youngest child with Spearman’s ρ. mSHAI = modified Short Health Anxiety Inventory; mSHAI-Child = modified Short Health Anxiety Inventory For Child; mSHAI-Partner = modified Short Health Anxiety Inventory For Partner; CSI-16 = Couple Satisfaction Index; PSS = Parenting Stress Scale; GFS = General Functioning Scale.

*p < .05,

**p < .01.

Further, small negative associations with relationship duration were found for all health anxiety inventories (r = −.12 to r = −0.15), indicating lower levels of health anxiety in longer existing relationships. Parenting stress and family functioning had small positive correlations with all health anxiety inventories (r = 0.15 to r = 0.20). Couple satisfaction was negatively correlated with mSHAI (r = −0.14, p < .001) and mSHAI-Child (r = −0.13, p = .003) but did not reach significance towards mSHAI-Partner (r = −0.08, p = .05).

Discussion

Based on our findings, we found support that health anxiety and health anxiety by proxy should be assessed independently and evaluated further to understand health anxiety in relation to family processes. There were significant mean differences between the inventories. Despite women having significantly higher health anxiety scores for themselves, there were no gender differences in the health anxiety by proxy conditions.

Health anxiety, its relation to health anxiety by proxy and its measurement

Our results confirm previous findings on the strong relationship between health anxiety and health anxiety by proxy (Thorgaard et al., 2017a; Reiser and Wright, 2019), suggesting closeness between both constructs. However, we found noteworthy differences between health anxiety for self and by proxy. We observed distinct latent factors in the exploratory factor analysis for health anxiety regarding self, child and partner. Established screening instruments like the Health Anxiety Inventory (HAI) or Whiteley-Index (WI) may not be suitable to presort or screen those who may be affected by health anxiety by proxy. Further questions of conceptualization of the measurement approach arise between the general conditions (health anxiety vs health anxiety by proxy) as well as within the several health anxiety questionnaires (mSHAI vs mSHAI-C vs mSHAI-P). Especially the factor structure of both introduced health anxiety by proxy measures should be investigated in future studies due to inconsistent results in past research with the original SHAI/mSHAI. While the German mSHAI is intended as one-dimensional (Bailer et al., 2013), the rarely used 14 item version of the SHAI was found to be two-dimensional (Alberts et al., 2011). Of note, the comparable validation study of the health anxiety by proxy measure for pregnant women by Reiser and Wright (2019) proposed their adaption of the SHAI as a unitary scale; however, their EFA resulted in a two-factor solution with “illness likelihood” and “body vigilance”. Therefore, the present study may add further evidence that health anxiety by proxy cannot be measured unidimensional with the adapted mSHAI questionnaires and that further validation studies are necessary to elucidate the factor structure.

In addition, the level of health anxiety regarding the child was significantly higher than parental health anxiety indicating the importance of health anxiety for the child as an area for further study. Research is also needed to understand first the clinical relevance of elevated health anxiety and its persistence through different conditions (i.e. towards self, child and partner) and second, the mutual impact in individuals with one elevated condition and the relevance of increase or onset of further health anxiety conditions.

Associations of health anxiety by proxy with family variables

A recent review by Thorgaard and colleagues (2018) found that little research has been done on the influence of family factors on health anxiety. In our sample, parenting stress, couple dissatisfaction and poor family functioning had small correlations with health anxiety and proxy conditions. This is similar with findings in the literature on anxiety symptoms and family quality measures (Hughes et al., 2008; Rapee, 2012; Whisman et al., 2004) and indicates that worse family functioning covaries with anxiety. Further research is needed to evaluate the dynamics of family functioning and health anxiety across family forms.

Surprisingly, the number of children was not correlated with the mSHAI-Child, but lower age of the youngest child correlated with higher health anxiety by proxy as assessed with the mSHAI-Child. It may support the fact, that uncertainty plays a major role in maintaining health-related worries towards the child (Carleton 2012; Fergus and Bardeen, 2013), while gained experience has less impact (e.g., experiences with various health conditions of older siblings). On the contrary, only 12.1% of parents with more than one child stated that they would have answered differently in mSHAI-Child items if they would be related to another (i.e. older) child. Thus, there may be greater variability in parental health anxiety by proxy, but this could be examined further in subsequent studies.

Health anxiety by proxy in the family system

Considerable research has shown that anxiety disorders tend to aggregate in families (Lawrence et al., 2019; Shimada-Sugimoto et al., 2015; Telman et al., 2018). We recommend aligning parental health anxiety by proxy as a process within the family system rather than to individual’s impairment for the following reasons. First, health anxiety by proxy is based on relational interaction between at least two subjects and includes inevitably a significant other. According to social network theory (Smith and Christakis, 2008; Valente 2010), the health between two or more socially strongly tied individuals is interconnected and characterized by mutual impact on health-related variables. In contrast, health anxiety is defined as disproportionate concerns about one’s own health (American Psychiatric Association, 2013) without attention to health concerns for others specifically. Since several family members can be significantly involved in this dysfunctional health behavior, couple or family interventions should be considered. Second, health anxiety by proxy can have an impact on health and well-being outcomes of others (i.e. family members). Children of mothers with severe health anxiety were presented more often to the doctor (Thorgaard et al., 2017a, 2017b) and confronted with more negative illness perceptions (Thorgaard et al., 2017a), and thus at potential risk for developing unrealistic health beliefs.

In contrast to previous research, we extend the consideration of health anxiety by proxy towards the intimate partner. Elevated health anxiety by proxy was hypothesized to impair relationship quality and partnership based on past studies on anxiety disorders rather than health anxiety specifically (Kasalova et al., 2017; McLeod, 1994; Pankiewicz et al., 2012). In the current study, levels of health anxiety towards the partner were significantly lower than towards the child. We attribute the significant difference in our sample mainly to higher uncertainty tolerance (Strout et al., 2018) and less perceived responsibility for the partner’s health situation than for the child’s. Health anxiety towards the partner may play a much bigger role in later life when the partner’s health deteriorates or in a caring situation. High rates of anxiety were found among family caregivers across a variety of diseases or impairments (Denno et al., 2013; Mahoney et al., 2005; Sklenarova et al., 2015) and where, among others, fear of deterioration of partner’s health status was reported. This work on health anxiety by proxy adds to related work that borders in conceptualizations such as on caregiver anxiety or burden (Novack et al., 1991; Mahoney et al., 2005; Nabors et al., 2013), death anxiety (Adelbratt and Strang, 2000), interdependent anxiety (Segrin et al., 2007) and general anxiety. Consistent use of a common term could help in the future to better examine health anxiety by proxy.

Accordingly, we encourage researchers to investigate the dyadic interaction of parental health anxiety on the child, respectively the triad with mutual repercussions on health anxiety by proxy between mother, father, and child. Connecting to family systems theory (Kerr and Bowen, 1988), perceived anxiety can worsen emotional processes within the family unit, including relationship quality, communication, and stress. Consequently, health anxiety should not be treated as an isolated individual construct but relevant for models of families and health. Further work could also investigate the role of various forms of anxiety (health anxiety, health anxiety by proxy, generalized anxiety, etc.) and how they interact in combination on family functioning. This research could shed light on which types and severities of anxiety disorders are most detrimental to family climate and children’s development.

Implications for clinical practice and public health

Health-related anxiety disorders are often associated with intensive healthcare utilization, and thus, at high-cost for health care systems (Barsky et al., 2001; Fink et al., 2010). In the case of health anxiety by proxy, similar dysfunctional behavior patterns may occur, and result in increased need for health care utilization. Clinicians should take into general consideration that health anxiety by proxy can be a relevant issue for some groups such as parents and caregivers that are seen in clinical practice and health psychology settings and that high levels of health anxiety by proxy can occur regardless of one’s own health anxiety.

In the long term, treating parental anxiety by proxy could possibly prevent later developed anxiety disorder in the offspring, but this should be examined in future studies (Eley et al., 2015; Ramchandani et al., 2011; Thorgaard et al., 2017). Parents may play a major role in transgenerational transmission of health anxiety through maladaptive health beliefs and excessive reinforcement of health-related behaviors (Marshall et al., 2007; Köteles et al., 2015; Wright et al., 2017; Thorgaard et al., 2018). Therefore, additional assessment of not only general health anxiety but anxiety by proxy could be informative.

Implications for researchers

The findings on health anxiety by proxy are based on only a few empirical studies. To enhance the evidence, researchers should further investigate the connection between health anxiety and health anxiety by proxy. This can be achieved by cross-validating with other anxiety-related questionnaires, for example, general anxiety disorder, nosophobia, cyberchondria, or depression. Additionally, using questionnaires to assess maladaptive health-related behaviors would be beneficial in understanding the transition from typical parental concerns to harmful health anxiety by proxy. Moreover, it is plausible that essential core elements for assessing health anxiety by proxy have been inadvertently neglected during the adaptation of the mSHAI. To address this, researchers could consider conducting a comparative analysis with the recently published Health Anxiety by Proxy Scale developed by Ingeman et al. (2021). Future studies should also consider other potential vulnerable relationships for health anxiety by proxy, especially non-parent relationships, like pet owners, informal caregivers, or close friends.

Limitations

There are several limitations to mention. Our study was cross-sectional and thus, we cannot make any statements about causality. The questionnaires were tested and validated in a German-speaking sample. Cultural differences on socially accepted health-related behavior regarding child or partner could impact on total scoring. Although the survey took place during a steady phase of the COVID-19 pandemic in central Europe, influences on the level of health anxiety cannot be excluded. As a result, it is plausible that the average levels of health anxiety inventories in the sample may exhibit higher values than would typically be expected. Another limitation of our study relates to the potential presence of selection bias resulting from our recruitment strategy. Participants were primarily recruited from online parenting groups and may differ from individuals that are not active online; although the majority of parents now fall into the group of active online users. Additionally, the majority of the participants were high-educated mothers; therefore, further validation work with fathers and educationally disadvantaged individuals is necessary.

Conclusion

Our results suggest that health anxiety by proxy is worthy of further investigation and may be relevant for working with families. The mSHAI-Child and mSHAI-Partner seemed to be promising inventories for assessing health anxiety by proxy, and thus should be considered in future work. Studies are needed to evaluate the prevalence in the general population among different relationship types and further examine clinical relevance. In this study, health anxiety by proxy was associated with family well-being, which may be relevant for family interventions and psychoeducation addressing anxiety in families and health problems. These relationships should be explored further in prospective studies.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Office of Academic Affairs, University of Klagenfurt.

Availability of questionnaires

The questionnaires utilized in this study are available upon request from the corresponding author.

ORCID iD

Christian Kubb

References

Adelbratt

Strang

(2000) Death anxiety in brain tumour patients and their spouses. Palliative medicine 14(6): 499–507.

Alberts

Sharpe

Kehler

, et al. (2011) Health anxiety: comparison of the latent structure in medical and non-medical samples. Journal of Anxiety Disorders 25(4): 612–614.

Alberts

Hadjistavropoulos

Jones

, et al. (2013) The short health anxiety inventory: a systematic review and meta-analysis. Journal of Anxiety Disorders 27(1): 68–78.

American Psychiatric Association (2013) Diagnostic and Statistical Manual of Mental Disorders (DSM-5®). Washington, DC: American Psychiatric Pub.

Asmundson

Abramowitz

Richter

, et al. (2010) Health anxiety: current perspectives and future directions. Current Psychiatry Reports 12(4): 306–312.

Austrian Agency for Health and Food Safety (2024) Corona virus: 7-day incidence over the entire course of the pandemic. Available at: https://www.ages.at/en/human/disease/pathogens-from-a-to-z/coronavirus (accessed 01 March 2024).

Aziz

El-Gabry

Al-Hosani

, et al. (2018) Illness Anxiety (Hypochondriasis) by Proxy in Parents towards Their Children: A Survey of Doctors Views. Birmingham, United Kingdom: Royal College of Psychiatrists, International Congress 2018. [Conference Presentation].

Bailer

Witthöft

(2006) Deutsches modifiziertes health anxiety inventory (MK-HAI). In: Zusammenstellung sozialwissenschaftlicher Items und Skalen (ZIS).

Bailer

Rist

Müller

, et al. (2013) Erfassung von Krankheitsangst mit dem short health anxiety inventory (SHAI). Verhaltenstherapie and Verhaltensmedizin 34(4): 378–398.

10.

Barke

Doering

(2020) Development of an instrument to assess parents’ excessive web-based searches for information pertaining to their children’s health: the “children’s health internet research, parental inventory”(CHIRPI). Journal of Medical Internet Research 22(4): e16148.

11.

Barsky

Ettner

Horsky

, et al. (2001) Resource utilization of patients with hypochondriacal health anxiety and somatization. Medical Care 39(7): 705–715.

12.

Bayrampour

Ali

McNeil

, et al. (2016) Pregnancy-related anxiety: a concept analysis. International Journal of Nursing Studies 55: 115–130.

13.

Berry

Jones

(1995) The parental stress scale: initial psychometric evidence. Journal of Social and Personal Relationships 12(3): 463–472.

14.

Bleichhardt

Hiller

(2007) Hypochondriasis and health anxiety in the German population. British Journal of Health Psychology 12(4): 511–523.

15.

Byles

Byrne

Boyle

, et al. (1988) Ontario child health study: reliability and validity of the general functioning subscale of the McMaster family assessment device. Family Process 27(1): 97–104.

16.

Carleton

(2012) The intolerance of uncertainty construct in the context of anxiety disorders: theoretical and practical perspectives. Expert Review of Neurotherapeutics 12(8): 937–947.

17.

da Silva

Rocha

RSB

Buheji

, et al. (2021) A systematic review of the prevalence of anxiety symptoms during coronavirus epidemics. Journal of Health Psychology 26(1): 115–125.

18.

Denno

Gillard

Graham

, et al. (2013) Anxiety and depression associated with caregiver burden in caregivers of stroke survivors with spasticity. Archives of physical medicine and rehabilitation 94(9): 1731–1736.

19.

Eley

McAdams

Rijsdijk

, et al. (2015) The intergenerational transmission of anxiety: a children-of-twins study. American Journal of Psychiatry 172(7): 630–637.

20.

Epstein

Baldwin

Bishop

(1983) The McMaster family assessment device. Journal of Marital and Family Therapy 9(2): 171–180.

21.

Fergus

Bardeen

(2013) Anxiety sensitivity and intolerance of uncertainty: evidence of incremental specificity in relation to health anxiety. Personality and Individual Differences 55(6): 640–644.

22.

Fink

Ørnbøl

Christensen

(2010) The outcome of health anxiety in primary care. A two-year follow-up study on health care costs and self-rated health. PLoS One 5(3): e9873.

23.

Funk

Rogge

(2007) Testing the ruler with item response theory: increasing precision of measurement for relationship satisfaction with the Couples Satisfaction Index. Journal of Family Psychology 21(4): 572–583.

24.

Hadfield

Akyirem

Sartori

, et al. (2022) Measurement of pregnancy-related anxiety worldwide: a systematic review. BMC Pregnancy and Childbirth 22(1): 1–10.

25.

Harpel

(2008) Fear of the unknown: ultrasound and anxiety about fetal health. Health: An Interdisciplinary Journal for the Social Study of Health, Illness and Medicine 12: 295–312.

26.

Horenstein

Heimberg

(2020) Anxiety disorders and healthcare utilization: a systematic review. Clinical Psychology Review 81: 101894.

27.

Hughes

Hedtke

Kendall

(2008) Family functioning in families of children with anxiety disorders. Journal of Family Psychology 22(2): 325–328.

28.

Ingeman

Frostholm

Frydendal

, et al. (2021) A new measure of excessive parental worries about children’s health: development of the Health Anxiety by Proxy Scale (HAPYS). Nordic Journal of Psychiatry 75(7): 523–531.

29.

Ingeman

Hulgaard

Rask

(2024) Health anxiety by proxy – through the eyes of the parents. Journal of Child Health Care 28(1): 22–36.

30.

Kasalova

Prasko

Holubová

, et al. (2017) Anxiety disorders and marital satisfaction. Neuroendocrinology Letters 38(8): 555–564.

31.

Kellner

(1987) Abridged manual of the illness attitude scale. Albuquerque, NM: Department of Psychiatry, University of New Mexico.

32.

Kerr

Bowen

(1988) Family Evaluation. New York: W. W. Norton.

33.

Köteles

Freyler

Kökönyei

, et al. (2015) Family background of modern health worries, somatosensory amplification, and health anxiety: a questionnaire study. Journal of Health Psychology 20(12): 1549–1557.

34.

Lawrence

Murayama

Creswell

(2019) Systematic review and meta-analysis: anxiety and depressive disorders in offspring of parents with anxiety disorders. Journal of the American Academy of Child & Adolescent Psychiatry 58(1): 46–60.

35.

Lockhart

(2016) Health anxiety in children and parents. Bmj 353: i2891.

36.

Mahoney

Regan

Katona

, et al. (2005) Anxiety and depression in family caregivers of people with Alzheimer disease: the LASER-AD study. American Journal of Geriatric Psychiatry 13(9): 795–801.

37.

Marshall

Jones

Ramchandani

(2007) Intergenerational transmission of health beliefs in somatoform disorders: exploratory study. The British Journal of Psychiatry 191(5): 449–450.

38.

McLeod

(1994) Anxiety disorders and marital quality. Journal of Abnormal Psychology 103(4): 767–776.

39.

Nabors

Kichler

Brassell

, et al. (2013) Factors related to caregiver state anxiety and coping with a child’s chronic illness. Families, Systems & Health 31(2): 171–180.

40.

Nakano

Matsushima

Nakamori

, et al. (2019) Depression and anxiety in pet owners after a diagnosis of cancer in their pets: a cross-sectional study in Japan. BMJ Open 9(2): bmjopen–2018.

41.

Novack

Bergquist

Bennett

, et al. (1991) Primary caregiver distress following severe head injury. The Journal of Head Trauma Rehabilitation 6(4): 69–76.

42.

Oxley

Kogan

(2018) Health anxiety by proxy and pet owners. JAVMA-Journal of the American Veterinary Medical Association 252(5): 530.

43.

Pankiewicz

Majkowicz

Krzykowski

(2012) Anxiety disorders in intimate partners and the quality of their relationship. Journal of Affective Disorders 140(2): 176–180.

44.

Ramchandani

Murray

Romano

, et al. (2011) An investigation of health anxiety in families where children have recurrent abdominal pain. Journal of Pediatric Psychology 36(4): 409–419.

45.

Rapee

(2012) Family factors in the development and management of anxiety disorders. Clinical Child and Family Psychology Review 15(1): 69–80.

46.

Rask

Gehrt

Rimvall

, et al. (2020) Health anxiety: conceptualization and future directions. Zeitschrift für Psychologie 228(2): 141–144.

47.

Reiser

Wright

(2019) Fetal health anxiety: development and psychometric properties of the fetal health anxiety inventory. Journal of Psychosomatic Obstetrics and Gynecology 40(4): 264–273.

48.

Robert Koch Institute (2020) Coronavirus disease 2019 (COVID-19) daily situation report of the robert Koch institute. Available at: https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_Coronavirus/Situationsberichte/2020-07-28-en.pdf?__blob=publicationFile (accessed 01 March 2024).

49.

Salkovskis

Rimes

Warwick

, et al. (2002) The health anxiety inventory: development and validation of scales for the measurement of health anxiety and hypochondriasis. Psychological Medicine 32(5): 843–853.

50.

Segrin

Badger

Dorros

, et al. (2007) Interdependent anxiety and psychological distress in women with breast cancer and their partners. Psycho-Oncology: Journal of the Psychological, Social and Behavioral Dimensions of Cancer 16(7): 634–643.

51.

Shimada Sugimoto

Otowa

Hettema

(2015) Genetics of anxiety disorders: genetic epidemiological and molecular studies in humans. Psychiatry and Clinical Neurosciences 69(7): 388–401.

52.

Sklenarova

Krümpelmann

Haun

, et al. (2015) When do we need to care about the caregiver? Supportive care needs, anxiety, and depression among informal caregivers of patients with cancer and cancer survivors. Cancer 121(9): 1513–1519.

53.

Smith

Christakis

(2008) Social networks and health. Annual Review of Sociology 34: 405–429.

54.

Strout

Hillen

Gutheil

, et al. (2018) Tolerance of uncertainty: a systematic review of health and healthcare-related outcomes. Patient Education and Counseling 101(9): 1518–1537.

55.

Sunderland

Newby

Andrews

(2013) Health anxiety in Australia: prevalence, comorbidity, disability and service use. The British Journal of Psychiatry 202(1): 56–61.

56.

Telman

van Steensel

Maric

, et al. (2018) What are the odds of anxiety disorders running in families? A family study of anxiety disorders in mothers, fathers, and siblings of children with anxiety disorders. European Child & Adolescent Psychiatry 27(5): 615–624.

57.

Thorgaard

Frostholm

Walker

, et al. (2017a) Health anxiety by proxy in women with severe health anxiety: a case control study. Journal of Anxiety Disorders 52: 8–14.

58.

Thorgaard

Frostholm

Walker

, et al. (2017b) Effects of maternal health anxiety on children’s health complaints, emotional symptoms, and quality of life. European Child & Adolescent Psychiatry 26(5): 591–601.

59.

Thorgaard

Frostholm

Rask

(2018) Childhood and family factors in the development of health anxiety: a systematic review. Children's Health Care 47(2): 198–238.

60.

Treffner

Foran

(2017) German Version of the Couple Satisfaction Inventory. Unpublished manuscript: Klagenfurt University.

61.

Valente

(2010) Social Networks and Health: Models, Methods, and Applications. New York: Oxford University Press.

62.

Weck

Richtberg

Neng

(2014) Epidemiology of hypochondriasis and health anxiety: comparison of different diagnostic criteria. Current Psychiatry Reviews 10(1): 14–23.

63.

Whisman

Uebelacker

Weinstock

(2004) Psychopathology and marital satisfaction: the importance of evaluating both partners. Journal of Consulting and Clinical Psychology 72(5): 830.

64.

Wright

Reiser

Delparte

(2017) The relationship between childhood health anxiety, parent health anxiety, and associated constructs. Journal of Health Psychology 22(5): 617–626.