A 12 week intervention with MEL-S01.

Ruminative Response Scale (RRS), short version, is an eight item self-report questionnaire that uses a four point Likert-scale. This leads to a total score between 8 and 64 where higher values indicate higher depressive brooding.

For each of the three study arms, the change in total RRS scores between before the intervention compared and after the 12 week intervention are measured. Then, the average change scores in each of the three groups are compared with each other.

The general anxiety disorder scale (GAD-7) is a seven-item self-report questionnaire that uses a 4-point scale (between 0 and 3) to evaluate the degree of anxiety symptoms. The questionnaire provides a total between 0 and 21: the higher the total score, the more severe the anxiety.

For each of the three study arms, the change in total GAD-7 scores between before the intervention and after the 12 week intervention is measured. Then, the average change scores in each of the three groups are compared with each other.

The Sheehan disability scale (SDS) is a self-administered three item questionnaire using a 11-point scale (between 0 and 10). The scale is used to measure functional impairment. The scale provides a total score between 0 and 30: the higher the score, the higher the functional impairment. In addition, the scale measures "the number of days lost to symptoms" and "the number of days unproductive due to the symptoms".

For each of the three study arms, the change in total SDS scores between before the intervention and after the 12 week intervention is measured. Then, the average change scores in each of the three groups are compared with each other.

The positive valence systems scale (PVSS) is a 21-item self-report questionnaire using a scale from 1 to 9. This leads to a total score between 21 and 189 where a lower total score indicates higher anhedonia.

For each of the three study arms, the change in total PVSS scores between before the intervention and after the 12 week intervention is measured. Then, the average change scores in each of the three groups are compared with each other.

The well-being index (WHO-5) is a five item self-report questionnaire using a scale from 0 to 5. This leads to a total score is between 0 and 25: higher scores indicate higher subjective well-being.

For each of the three study arms, the change in total WHO-5 scores between before the intervention and after the 12 week intervention is measured. Then, the average change scores in each of the three groups are compared with each other.

The patient health questionnaire (PHQ-9, Kroenke et al., 2001) is a self-report measure of major depressive disorder. The PHQ-9 scores each of the nine DSM-IV criteria for MDD from "0" (not at all) to "3" (nearly every day), providing a 0-27 severity score of depression. The higher the total score the more severe the depression.

For the treatment MEL-T01 group, the change scores in the total PHQ-9 score between before the intervention and after the 12 week intervention is measured. Then, the correlation between the change score and the minutes played during the intervention is calculated. This leads to a correlation score that evaluates whether a higher symptom change is associated with higher playing time.

The patient health questionnaire (PHQ-9, Kroenke et al., 2001) is a self-report measure of major depressive disorder. The PHQ-9 scores each of the nine DSM-IV criteria for MDD from "0" (not at all) to "3" (nearly every day), providing a 0-27 severity score of depression. The higher the total score the more severe the depression.

For the active comparator MEL-S01 group, the change scores in the total PHQ-9 score between before the intervention and after the 12 week intervention is measured. Then, the correlation between the change score and the minutes played during the intervention is calculated. This leads to a correlation score that evaluates whether a higher symptom change is associated with higher playing time.

The immersive experiences questionnaire (IEQ, Jennett et al., 2008) is a self-report questionnaire that measures the subjective experience of being immersed while playing a video game. It has 31 questions on a scale of 1-7. This creates a total score between 31 and 217: the higher totals indicate higher immersion. In addition, there is one question - "How immersed did you feel?" - on a scale of 1-10.

For the treatment MEL-T01 group, we measure the correlation between 1) the change in the total PHQ-9 score between before the intervention and after the 12 week intervention and 2) the total IEQ score (excluding the one additional questionnaire) as measured after the 12 week intervention.

This leads to a correlation score that evaluates whether a higher symptom change is associated with higher immersion in the game.

The immersive experiences questionnaire (IEQ, Jennett et al., 2008) is a self-report questionnaire that measures the subjective experience of being immersed while playing a video game. It has 31 questions on a scale of 1-7. This creates a total score between 31 and 217: the higher totals indicate higher immersion. In addition, there is one question - "How immersed did you feel?" - on a scale of 1-10.

For the treatment MEL-S01 group, we measure the correlation between 1) the change in the total PHQ-9 score between before the intervention and after the 12 week intervention and 2) the total IEQ score (excluding the one additional questionnaire) as measured after the 12 week intervention.

This leads to a correlation score that evaluates whether a higher symptom change is associated with higher immersion in the game.

Treatment-as-usual in healthcare.

Total of all reporting groups

Race and Ethnicity were not collected from any participant.

The patient health questionnaire (PHQ-9, Kroenke et al., 2001) is a self-report measure of major depressive disorder. The PHQ-9 scores each of the nine DSM-IV criteria for MDD from "0" (not at all) to "3" (nearly every day), providing a 0-27 severity score of depression. The higher the total score the more severe the depression.

Treatment As Usual in healthcare for 12 weeks. This group combines both "First TAU, then MEL-T01" and "First TAU, then MEL-S01".

The patient health questionnaire (PHQ-9, Kroenke et al., 2001) is a self-report measure of major depressive disorder. The PHQ-9 scores each of the nine DSM-IV criteria for MDD from "0" (not at all) to "3" (nearly every day), providing a 0-27 severity score of depression. The higher the total score the more severe the depression.

The change in the PHQ-9 total score between before the 12 week intervention and after the intervention is measured. The change score of MEL-S01 group is compared with the change score of the treatment-as-usual (TAU) group.

Per protocol completer cohort. Analyses were restricted to weeks 0-12; TAU participants were pooled as all received identical treatment during this period, regardless of later allocation.

The patient health questionnaire (PHQ-9, Kroenke et al., 2001) is a self-report measure of major depressive disorder. The PHQ-9 scores each of the nine DSM-IV criteria for MDD from "0" (not at all) to "3" (nearly every day), providing a 0-27 severity score of depression. The higher the total score the more severe the depression.

The change in the PHQ-9 total score between before the 12 week intervention and after the intervention is measured. The change score of MEL-T01 group is compared with the change score in the MEL-S01 group.

Per protocol completer cohort. Analyses were restricted to weeks 0-12.

The quick inventory of depressive symptomatology (QIDS-SR16) is a self-report measure comprising of 16 questions on a scale of 0-3 which assess the severity of depressive symptoms. The inventory provides a total score between 0 and 48: the higher the total score the more severe the depression.

For each of the three study arms, the change in total QIDS-SR16 scores between before the intervention and after the 12 week intervention is measured. Then, the average change scores in each of the three groups are compared with each other.

Per protocol completer cohort. Analyses were restricted to weeks 0-12; TAU participants were pooled as all received identical treatment during this period, regardless of later allocation.

Ruminative Response Scale (RRS), short version, is an eight item self-report questionnaire that uses a four point Likert-scale. This leads to a total score between 8 and 64 where higher values indicate higher depressive brooding.

For each of the three study arms, the change in total RRS scores between before the intervention compared and after the 12 week intervention are measured. Then, the average change scores in each of the three groups are compared with each other.

Per protocol completer cohort. Analyses were restricted to weeks 0-12; TAU participants were pooled as all received identical treatment during this period, regardless of later allocation.

The general anxiety disorder scale (GAD-7) is a seven-item self-report questionnaire that uses a 4-point scale (between 0 and 3) to evaluate the degree of anxiety symptoms. The questionnaire provides a total between 0 and 21: the higher the total score, the more severe the anxiety.

For each of the three study arms, the change in total GAD-7 scores between before the intervention and after the 12 week intervention is measured. Then, the average change scores in each of the three groups are compared with each other.

Per protocol completer cohort. Analyses were restricted to weeks 0-12; TAU participants were pooled as all received identical treatment during this period, regardless of later allocation.

The Sheehan disability scale (SDS) is a self-administered three item questionnaire using a 11-point scale (between 0 and 10). The scale is used to measure functional impairment. The scale provides a total score between 0 and 30: the higher the score, the higher the functional impairment. In addition, the scale measures "the number of days lost to symptoms" and "the number of days unproductive due to the symptoms".

For each of the three study arms, the change in total SDS scores between before the intervention and after the 12 week intervention is measured. Then, the average change scores in each of the three groups are compared with each other.

Per protocol completer cohort. Analyses were restricted to weeks 0-12; TAU participants were pooled as all received identical treatment during this period, regardless of later allocation.

The positive valence systems scale (PVSS) is a 21-item self-report questionnaire using a scale from 1 to 9. This leads to a total score between 21 and 189 where a lower total score indicates higher anhedonia.

For each of the three study arms, the change in total PVSS scores between before the intervention and after the 12 week intervention is measured. Then, the average change scores in each of the three groups are compared with each other.

Per protocol completer cohort. Analyses were restricted to weeks 0-12; TAU participants were pooled as all received identical treatment during this period, regardless of later allocation.

The well-being index (WHO-5) is a five item self-report questionnaire using a scale from 0 to 5. This leads to a total score is between 0 and 25: higher scores indicate higher subjective well-being.

For each of the three study arms, the change in total WHO-5 scores between before the intervention and after the 12 week intervention is measured. Then, the average change scores in each of the three groups are compared with each other.

Per protocol completer cohort. Analyses were restricted to weeks 0-12; TAU participants were pooled as all received identical treatment during this period, regardless of later allocation.

The patient health questionnaire (PHQ-9, Kroenke et al., 2001) is a self-report measure of major depressive disorder. The PHQ-9 scores each of the nine DSM-IV criteria for MDD from "0" (not at all) to "3" (nearly every day), providing a 0-27 severity score of depression. The higher the total score the more severe the depression.

For the treatment MEL-T01 group, the change scores in the total PHQ-9 score between before the intervention and after the 12 week intervention is measured. Then, the correlation between the change score and the minutes played during the intervention is calculated. This leads to a correlation score that evaluates whether a higher symptom change is associated with higher playing time.

Intention-to-treat cohort.

The patient health questionnaire (PHQ-9, Kroenke et al., 2001) is a self-report measure of major depressive disorder. The PHQ-9 scores each of the nine DSM-IV criteria for MDD from "0" (not at all) to "3" (nearly every day), providing a 0-27 severity score of depression. The higher the total score the more severe the depression.

For the active comparator MEL-S01 group, the change scores in the total PHQ-9 score between before the intervention and after the 12 week intervention is measured. Then, the correlation between the change score and the minutes played during the intervention is calculated. This leads to a correlation score that evaluates whether a higher symptom change is associated with higher playing time.

Intention-to-treat cohort.

The immersive experiences questionnaire (IEQ, Jennett et al., 2008) is a self-report questionnaire that measures the subjective experience of being immersed while playing a video game. It has 31 questions on a scale of 1-7. This creates a total score between 31 and 217: the higher totals indicate higher immersion. In addition, there is one question - "How immersed did you feel?" - on a scale of 1-10.

For the treatment MEL-T01 group, we measure the correlation between 1) the change in the total PHQ-9 score between before the intervention and after the 12 week intervention and 2) the total IEQ score (excluding the one additional questionnaire) as measured after the 12 week intervention.

This leads to a correlation score that evaluates whether a higher symptom change is associated with higher immersion in the game.

Intention-to-treat cohort.

The immersive experiences questionnaire (IEQ, Jennett et al., 2008) is a self-report questionnaire that measures the subjective experience of being immersed while playing a video game. It has 31 questions on a scale of 1-7. This creates a total score between 31 and 217: the higher totals indicate higher immersion. In addition, there is one question - "How immersed did you feel?" - on a scale of 1-10.

For the treatment MEL-S01 group, we measure the correlation between 1) the change in the total PHQ-9 score between before the intervention and after the 12 week intervention and 2) the total IEQ score (excluding the one additional questionnaire) as measured after the 12 week intervention.

This leads to a correlation score that evaluates whether a higher symptom change is associated with higher immersion in the game.

Intention-to-treat cohort.

The game addiction scale (GAS, Lemmens et al., 2009) is a self-report 7-item questionnaire using a 1-5 scale which measures game addiction. This provides a total between 7 and 35: the higher the scores indicate higher addiction.

For each of the three study arms, the change in total GAS scores between before the intervention and after the 12 week intervention is measured. Then, the average change scores in each of the three groups are compared with each other.

Per protocol completer cohort. Analyses were restricted to weeks 0-12; TAU participants were pooled as all received identical treatment during this period, regardless of later allocation.

MEL-S01 is an active comparator similar to MEL-T01 but without personalized cognitive training elements. MEL-S01: A 12 week intervention with MEL-S01.

Treatment-as-usual in healthcare for 12 weeks.

Then, the participant is given acess to MEL-T01 for 12 weeks.

Treatment-as-usual in healthcare for 12 weeks.

Then, the participant is given acess to MEL-S01 for 12 weeks.

Robust linear mixture model (rlmer). DoF with Satterthweite. Robust Wald-type tests for p-values.

Alpha level 0.05. Holm-Bonferroni correction applied across comparisons.

Robust linear mixture model (rlmer). DoF with Satterthweite. Robust Wald-type tests for p-values.

Alpha level 0.05. Holm-Bonferroni correction applied across comparisons.

Robust linear mixture model (rlmer). DoF with Satterthweite. Robust Wald-type tests for p-values.

Alpha level 0.05. Holm-Bonferroni correction applied across comparisons.

Robust linear mixture model (rlmer). DoF with Satterthweite. Robust Wald-type tests for p-values.

Alpha level 0.05. Holm-Bonferroni correction applied across comparisons.

Robust linear mixture model (rlmer). DoF with Satterthweite. Robust Wald-type tests for p-values.

Alpha level 0.05. Holm-Bonferroni correction applied across comparisons.

Robust linear mixture model (rlmer). DoF with Satterthweite. Robust Wald-type tests for p-values.

Alpha level 0.05. Holm-Bonferroni correction applied across comparisons.

Robust linear mixture model (rlmer). DoF with Satterthweite. Robust Wald-type tests for p-values.

Two-tailed, alpha level 0.025. Holm-Bonferroni correction applied across comparisons.

Testing whether the GAS change in the two groups is equal.