Study participants
Our sample consists of 88% mothers (n=673), the mean parental age is 35.7 years (SD 4.3), 76% have a tertiary level of education (n=577), 72% are of Swiss nationality (n= 451) and 42% of the sample have a monthly net household income of over CHF 9000 (≈ € 8400, ≈ $ 9300, n=305). Children’s mean age is about 16 months (SD 7.1), 49% of the children are female and in 6% parents report their child to have a disability. Study sample characteristics do not significantly differ by disability status, with exception of birth weight showing a borderline significance (p=0.086).
Frequency of use of information resources
The study sample shows a high use of all information resources (Table 1). Examining the frequency of use of information resources by the child’s disability status, we find significant differences between parents with a child with and without disability in use of print media. The former resort to print media significantly less frequently than the latter (p<0.05), both when seeking information on general health and development and in the case of an acute child’s illness.
********************
Table 1 about here
********************
Use of information resources for different health information seeking targets
We find that parents use digital media significantly more often to search for information about general health and development questions (30.4, IQR 21.7 - 43.5) than for an acute child’s illness (26.1, IQR 17.4 - 39.1, p<0.001) (Figure 1a). Similarly, the median standardized frequency for print media use is higher when parents inform themselves about general health and development questions (20.0, IQR 13.3 - 33.3) than about an acute child’s illness (13.3, IQR 6.7 – 20.0, p<0.001) (Figure 1b). No difference by health information seeking targets is seen with respect to “personal contacts”: The median standardized frequency is 39.1 for both, general health and development questions (IQR 30.4 - 47.8) and an acute child’s illness (IQR 26.1 - 47.8, p=0.67) (Figure 1c).
Fig 1: Use of information resources for different health information seeking targets
(a) Search for information in digital media
(b) Search for information in print media
(c) Search for information among “personal contacts”
Note: Respective N by information source: (a) n = 585. (b) n = 418. (c) n = 744. Difference across frequency scores:paired-samples t-test (a) p<0.001, (b) p<0.001, (c) p=0.67.
Use of individual items of information resources
Examining the individual information sources and their single items by type of health information seeking targets in Table 2, we receive a more differentiated picture of the results presented in Figures 1a-1c, even though numbers for single items were too small to compare by disability status. Whereas Figure 1a indicates a significantly higher use of digital media for general health and development questions than for an acute child’s illness, results presented in Table 2 show that the difference is mainly driven by more frequent use of social media (p<0.1) and of websites for parents (p<0.05).
********************
Table 2 about here
********************
For print media, Table 2 indicates significant differences between the health information seeking targets for all items: books (p<0.001), journals (p<0.001), newspaper (p<0.01) and other print media (p<0.001). For all items, the frequency is higher for general health and development questions than for an acute child’s illness.
We also find significant differences between most items of “personal contacts”: paediatrician, telephone consultation of a children’s emergency service, telephone consultation of a health insurance, family members and friends, acquaintances or neighbours (all p<0.001) by the health information seeking target. The paediatrician, telephone consultation of a children’s emergency service and telephone consultation of a health insurance are used more frequently in the case of an acute child’s illness. In contrast, family members and friends, acquaintances or neighbours are contacted more frequently for questions about general health and development.
Use of digital media before and after a paediatrician visit
Parents were asked whether they used digital media to access health information before or after their last visit to the paediatrician. Almost half of the parents (45.5%, n=195) did so before the visit while 27.1% did so after the visit (n=117). A chi2-test reveals that differences between parents with a disabled and a non-disabled child are not statistically significant.
Among parents searching for information before the visit inform themselves about the health issue in general (90%), as well as about alternative medical (43%) and academic medical treatment options (47%). When asked for the reasons why they use digital media after the visit to the paediatrician, slightly more than 54% of the parents had the need to exchange with others or to search for experiences and tips, about 39% were looking for other therapies, about 38% needed to check information from the paediatrician, about 28% indicate to have received too little information and, for about 17%, the information was incomprehensible or contradictory. Differences between parents with a disabled and a non-disabled child are not statistically significant.
Health information seeking behaviour by disability status
In Table 3, we present the results from the ordinary least square (OLS) regression analyses. The main analyses (models 1 and 2) do not yield significant association between the frequency of parents’ use of digital media and disability, neither in case of general health and development questions nor in case of an acute child’s illness. With respect to model covariates, we find significant effects for the child’s age (β-Coef.: -0.21, CI: -0.41-0.02) in the general health and development model (1). In both models we find an effect for parents’ education: parents with compulsory education, only, use digital media much more frequently than parents with upper secondary education in the general health and development model (1) (β-Coef: 10.58, CI: 1.24-19.92) and in the acute child’s illness model (2) (β-Coef: 12.61, CI: 0.80-24.43).
********************
Table 3 about here
********************
With respect to parents’ use of print media we observe a statistically significant association for the child’s disability both for general health and development questions (model 3) (β-Coef: -8.32, CI: -13.90 - -2.73) and an acute child’s illness (model 4) (β-Coef: -6.30, CI:-10.16 - -2.44) while we do not find significant results for use of “personal contacts” (models 5 and 6). However, the model covariate parental education is significantly associated with information seeking: parents with no or compulsory education address “personal contacts” much more frequently for information about general health (model 5) (β-Coef: 5.57, CI: 0.35-10.78) and development and acute child’s illness (model 6) (β-Coef: 5.31, CI: -0.57-11.19) than parents with an upper secondary level of education. In addition, in model (6) we find an effect for respondent’s role: mothers have a significantly less frequent use of personal contacts in case of an acute child’s illness than fathers (β-Coef: -4.31, CI: -8.01 - 0.61).
Sensitivity analyses excluding the covariate parents’ age yielded consistent results for the primary outcome variable. In sensitivity analyses including net household income, the income variable did not prove to be a significant covariate, however, its inclusion lead to a loss of the statistical significance of education. Moreover, it contained a large number of missing values. Sensitivity analyses including only those observations for which information was reported for both general health and development and an acute child’s illness, lead to a lower number of observations for each analysis with no relevance for the effect estimates. We therefore present the results of the initial models.