EPA-0404 – Correlation and Causality Between the use of Online Health Forums and Suicidality: Evidence from a German Representative Survey and a Large-scale Online Panel Survey

Abstract

Introduction:

Suicide is a leading cause of death worldwide and a top leading cause of death among youths. Since younger age groups spend more time with internet-based applications, chances and risks of health forums have been scrutinized. It has been assumed that the use of health forums increases suicidality.

Objectives/Aims:

This study aims to disentangle the presumed causality between the use of online health forums and suicidality.

Methods:

A German representative telephone survey and a two-wave online panel survey with same question wording have been concurrently conducted in 2013. Both surveys included questions on suicidality (Pöldinger, 1968) and internet-based health forums use.

Results:

The German representative cross-sectional data show significant positive correlations between suicidality and online health forum use, especially among younger age groups (Table 1). Using a large-size online access panel, we estimated an autoregressive model that reveals a positive cross-lagged effect of suicidality on internet-based health forum use one month later (Figure 1).

Conclusions:

Despite the wide-spread notion that online health forums can increase suicidality our data shows an inverse effect. These results point to the preventive potential of accessible and helpful information in online health forums.

Table 1

Bivariate correlations between suicidality and internet-based health forum use for different age groups in a German representative survey.

	internet-based health forum use
	18–29y	30–39y	40–49y	50–59y	60–69y	70y+
suicidality	.12*	.15*	.11*	.16**	.05	.03
n	257	265	358	332	213	144

Note. For all scales, higher scores are indicative of more extreme responding in the direction of the construct assessed. Suicidality was measured by the suicidality scale suggested by Poldinger (1968) and consisted of 16 yes/no questions. Internet-based health forum or support group use was measured on a 5-point Likert-like scale ranging from 1 “(almost) never“to 5 “(almost) every day“.

^*

p< .05.

^**

p< .01.

Figure. 1

Higher suicidality (t₀) influences the use of internet-based health forums or support groups one month later (t₁).

Participants (n = 1319) were recruited from an online access panel (soSci Panel).

Suicidality was measured by the suicidality scale suggested by Poldinger (1968) and consisted of 16 yes/no questions. Internet-based health forum or support group use was measured on a 5-point Likert-like scale ranging from 1 “(almost) never” to 5 “(almost) every day”. For all scales, higher scores are indicative of more extreme responding in the direction of the construct assessed.

Coefficients are standardized coefficients.

p<.O5. **p<.01. ***p<.001 = 8.50

X²= 1, p <.01 ; = .075; SRMR = . 015; CFI= . 99; 71/=.96