Personalization Strategies for Increasing Engagement With Digital Mental Health Resources: Sequential Multiple Assignment Randomized Trial
Notice bibliographique
Résumé
Background: Although web-based mental health resources have the potential to assist millions, particularly those who face barriers to treatment, most mental health website visitors disengage before accessing resources that can help improve their mental health. Objective: We used a sequential multiple assignment randomized trial to test whether personalized tailoring improved engagement on a self-guided mental health website. Methods: Data were collected via voluntary response sampling on the Mental Health America website. Inclusion criteria included residing in the United States and viewing a postscreening survey after completing the Patient Health Questionnaire-9 (PHQ-9). Participants were randomized to 1 of 2 postscreening survey conditions: the demographics survey or the Next Steps survey, which included additional tailoring questions assessing perceived need and participants' intended next steps. Participants who viewed the following screening results page were subsequently randomized to 1 of 5 conditions that displayed nontailored or tailored messages and featured resources, as well as persistent general resources that did not vary by condition. Data were analyzed using logistic regressions predicting disengagement and clicks on featured resources (versus persistent general resources) by condition. Results: Adding questions to inform tailoring significantly increased the odds of disengaging by 14% (demographics survey: 25%; Next Steps survey: 27.5%; odds ratio [OR] 1.14, 95% CI 1.11-1.16; P<.001). Among participants who viewed a postscreening survey (n=169,647), 87,712 participants were randomized to the demographics survey condition, and 81,935 participants were randomized to the Next Steps survey condition. Among participants who submitted the demographics survey (n=38,490), tailoring resources to demographics reduced the odds of disengaging by 10% (OR 0.90, 95% CI 0.87-0.94; P<.001) and, among those who engaged, increased the odds of clicking a featured resource versus a persistent general resource by 90% (OR 1.90, 95% CI 1.79-2.01; P<.001). Among participants who submitted the Next Steps survey (n=34,204), tailoring messages to perceived need (P=.33), tailoring resources to intended next steps (P=.51), and a combination of both (P=.52) did not significantly reduce the odds of disengaging compared with the nontailored condition. However, tailoring resources to intended next steps and combining a tailored message to perceived need with tailored resources to intended next steps increased the odds of clicking a featured resource by 25% (OR 1.25, 95% CI 1.14-1.37; P<.001) and 34% (OR 1.34, 95% CI 1.23-1.47; P<.001), respectively. Tailoring resources to demographics was significantly more effective in improving engagement than tailoring to perceived need or intended next steps (P≤.004). Conclusions: There was a small but statistically significant cost to engagement from adding tailoring questions assessing perceived need and intended next steps. Among the strategies tested in this study, tailoring resources to demographics was the most effective strategy for increasing engagement among visitors who viewed their screening results. This study demonstrates how personalization may increase engagement with mental health websites and provides design implications for future research.
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction machine sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Le volet Gemma est une étiquette directe du modèle pour chaque travail de la base, lue sur la notice réduite au titre. Le volet Codex est un classifieur appris des 10 348 étiquettes directes de Codex et calibré sur les taux pondérés de l'échantillon; les champs sans appui suffisant ne portent aucun appel Codex. Le mode candidate est l'union des deux volets; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont pas des étiquettes humaines.
Scores du classifieur distillé par catégorie (deux têtes)
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,008 | 0,016 |
| Méta-épidémiologie (sens strict) | 0,003 | 0,002 |
| Méta-épidémiologie (sens large) | 0,006 | 0,004 |
| Bibliométrie | 0,001 | 0,001 |
| Études des sciences et des technologies | 0,001 | 0,002 |
| Communication savante | 0,002 | 0,003 |
| Science ouverte | 0,002 | 0,002 |
| Intégrité de la recherche | 0,005 | 0,004 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,016 | 0,001 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule source (Gemma direct ou Codex distillé), pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».