Combatting negative bias: a mental contrasting and implementation intentions online intervention to increase help-seeking among individuals with elevated depressive symptomatology

Abstract

Background: There are many reasons why individuals with depression may not seek help. Among those with elevated depressive symptomatology, some previous interventions aimed at increasing help-seeking have unintentionally decreased help-seeking intentions. Beck's cognitive theory of depression posits that individuals with elevated depressive symptomatology process information differently from those without depression (i.e., increased cognitive errors, negative bias); potentially explaining the iatrogenic results of previous interventions. Mental contrasting and implementation intentions (MCII; a self-regulatory strategy) interventions have successfully influenced physical and mental health behaviors. However, MCII has not been used specifically for initiating help-seeking for depression. The goal of this research was to ascertain whether an online MCII intervention could increase actual help-seeking or the intention to seek help for depression.

Method: Two online randomized pre-post experiments were conducted to measure the primary outcome measures 2 weeks post-intervention (Study 1 collected Summer 2019: information-only control ["C"], help-seeking MCII intervention ["HS"], and comparison MCII intervention ["E"]; Study 2 collected Winter 2020: "C" and "HS"). At Time 1, adults recruited from MTurk had a minimum Beck Depression Inventory (BDI-II) score of 14 (mild depressive symptoms) and were not seeking professional help.

Results: Study 1 (N = 74) indicated that the intervention was feasible, provided preliminary support, and clarified intervention components for Study 2. Study 2 (N = 224) indicated that the HS group reported greater intentions to seek help and actual help-seeking than the C group. Proportionally, actual help-seeking was more likely among individuals who received the HS intervention and either did not perceive themselves as depressed at Time 2 or had BDI-II scores indicating that their depressive symptomatology decreased from Time 1.

Limitations: Participation was limited to US residents who self-reported data.

Discussion: These studies indicate that a brief online MCII intervention to encourage help-seeking is feasible and preliminarily successful. Future studies should consider using ecological momentary assessment measurements to establish the temporal precedence of intervention effects and whether MCII is effective for encouraging help-seeking among individuals prone to experiencing cognitive errors who may not be experiencing negative bias (e.g., bipolar disorder or anxiety). Clinicians may find this method successful in encouraging ongoing treatment engagement.

Keywords: MCII; depression; help-seeking; implementation intentions; mental contrasting; negative bias; online intervention.

Figure 1

Study 1 layout. Time 2 occurred 2 weeks post intervention.

Figure 2

Study 1 Hypothesis 1: Success of help-seeking intervention. Non-significant interaction between groups on help-seeking success indicating that completing the HS MCII intervention had no significant effect on perceived help-seeking success 2 weeks post-intervention for either the ITT (A) or PP (B) analyses. Scores reported are the means for perceived success of help-seeking scores and the bars are the standard error.

Figure 3

Study 1 Hypothesis 2: Changes to intentions to seek help. (A) ITT: Illustrating the non-significant interaction between groups and SIIS HS over time; completing the HS MCII intervention had no significant effect on intentions to seek help over time for the ITT analyses; However, (B) PP: indicated significant main effect for group and interaction between group and SIIS HS over time. Those who received HS intervention had greater intentions seek help across both time points compared to the other conditions. Scores reported are the means for the SIIS HS scores T1 and T2 and the bars are the standard error.

Figure 4

Study 2 layout. Time 2 occurred 2 weeks post intervention.

Figure 5

Study 2 Hypothesis 1: Success of help-seeking intervention. Those who received the intervention were more likely to report seeking help in the past 2 weeks than those who received information alone. This was supported for both the IIT (A) and (B) PP analyses Scores reported are the means for perceived success of help-seeking scores and the bars are the standard error.

Figure 6

Study 2 Hypothesis 2: Changes to intentions to seek help. Illustrating the significant main effect for group and interaction between group and SIIS HS over time. Those who received HS intervention had greater intentions seek help across both time points compared to those who received information alone. This was supported for both the IIT (A) and PP (B) analyses. Scores reported are the means for the SIIS HS T1 and T2 and the bars are the standard error.

Figure 7

Study 2 Exploratory Analysis 1 3-way interaction: Utility x Resource x Success Ratings. For those who completed the HS MCII, there were no significant relationships between perceived utility, resources, nor interaction between utility and resources on perceived help-seeking success. Scores reported are the estimated marginal means for the interaction terms and the bars are the standard error for both the IIT (A) and PP (B) analyses.

Figure 8

Study 2 Exploratory Analysis 2a: The effect of actual depression at T2 on successful help-seeking. Those who received the HS MCII and whose actual depression scores fell below the threshold for mild symptomatology at T2 (BDI-II = 0–13), were proportionally the most likely to report seeking help. Depression scores did not significantly influence the success in the control group. This was true for both the IIT (A) and PP (B) analyses. BDI scores 0–13 indicates no to minimal symptomatology; BDI scores 14 or higher indicate mild or greater symptomatology. The scores represent number of participants who fell into each categorical classification and the bars are the standard errors.

Figure 9

Study 2 Exploratory Analysis 2b: The effect of perceived depression at T2 on successful help-seeking at T2. (A) The significant odds ratio for the ITT sample indicated that the probability of reporting success at help-seeking was 3.741 times higher for individual in the HS group who did not perceive themselves as having elevated depressive symptomatology (PD) at T2 than those who did. (B) Despite a significant chi square, the confidence interval for the odds ratio was not significant for the PP sample for the HS group. There were no significant relationships between the PD at T2 and perceived success for the C group (ITT or PP). The scores represent number of participants who fell into each categorical classification and the bars are the standard errors.