Table 5.

Digital interventions for community mental health services including common and severe mental disorders

Authors & Country	Disorder	Trial type	Sample	Methods	Instruments	Outcome measures	Time points	Findings
Malhotra et al. (2019); India	Mental health problems (general) in remote areas of India	Feasibility/initial clinical effectiveness study	2594 psychiatric patients in rural areas. Most common diagnoses were mood disorders, substance or alcohol use disorders, anxiety and stress-related disorders and psychotic disorders	Internet-based, computerized Clinical Decision Support System (CDSS): a digital diagnostic and management application to improve delivery of mental healthcare in rural areas. Included 3 modules for diagnosis, management and follow-up respectively, based on DSM-IV and ICD-10 criteria. The aim was to replicate a routine clinical situation. Psychologist or social worker used the app to interview patients, choose medication or psychotherapy as per CDSS guidelines. Patient was assessed at follow-up using the relevant module	MINI/MINI-KID, measures of validity, reliability, feasibility and satisfaction	Validity, reliability and feasibility of CDSS; satisfaction with intervention	Unclear. Follow-up timepoints varied depending on the nature of a participant’s illness and treatment	Sensitivity of screening tool was high for all disorders (ranging from 48–99%). Positive predictive values were low for most disorders except alcohol and substance use disorders. Online diagnostic module was acceptable/satisfactory to patients and interviewers
Sibeko et al. (2017); South Africa	Psychotic disorder (various)	Pilot RCT	77 participants (average 35.5 years; 28.6% female) with schizophrenia spectrum disorder, bipolar mood disorder, or substance use induced psychotic disorder, recruited from Valkenburg hospital	A treatment-partner and text message intervention. Incorporated CAU with addition of 1) a treatment partner contracting and psychoeducation session and 2) text message reminders of clinic appointments. The aim of the intervention was to improve adherence to mental health services for people with psychotic disorder. Control group received CAU	PANSS, CGI, GAF, CAN,, EUROQUOL, VAS, MARS	Clinic attendance, adherence, symptoms, readmission rate at 9 months	Baseline, post-discharge, 3-month follow-up	At 3-month follow-up, intervention participants understood their diagnosis better than CAU group. CAU group were more likely to miss their first clinic visit and relapse in the 9 months following discharge, and to show worsening PANSS scores. Efficacy outcomes tended to favour the intervention but did not reach statistical significance
Singh et al. (2017); India	Unspecified diagnoses (various mental illnesses)	RCT	214 patients (40.6% female)	Intervention to improve follow-up rates at a community mental health clinic. SMS group received SMS 1 day before appointment at clinic. Control group received no SMS. At second-level intervention patients from both groups who missed first appointment were given a voice call reminder	Measure of attendance	Attendance at clinic	No fixed time points	At first SMS intervention level, 62.3% of SMS group reached their appointment, compared to 45.4% of the CAU group. At second-level intervention, 66 of 88 came for follow-up. Diagnosis of alcohol dependence was significantly associated with missing appointments
Xu et al. (2019); China	Schizophrenia	RCT	278 community-dwelling villagers with schizophrenia (average 46 years; 55.4% female)	LEAN intervention: Lay health supporters, Eplatform, Award and iNtegration). This featured recruitment of lay health support and text messages for medication reminders, psycho- and health education, monitoring of early signs of relapse, and linkage to primary healthcare for people with schizophrenia in rural resource-poor China. Control group received CAU	Score of adherence to antipsychotic medication, BARS, DAI, CGI for schizophrenia, WHODAS, GASS	Medication adherence, patient symptoms, functioning, relapse, re-hospitalisation, death for any reason, violence	Baseline, 6 months	Medication adherence was significantly improved at 6 months (0.48 in control group, 0.61 in intervention; representing a 27% improvement in the LEAN group), with moderate effect size of d = 0.38 reported, as well as a significantly reduced risk of relapse and re-hospitalisation with the intervention. The two groups did not differ on other outcomes

BARS: Brief Adherence Rating Scale

CAN: Camberwell Assessment of Unmet Needs

CGI: Clinical Global Impressions

DAI: Drug Attitude Inventory

EUROQUOL: Quality of Life measurement

GAF: Global Assessment of Function

GASS: Glasgow Antipsychotic Side-Effect Scale

MARS: Medication Adherence Rating Scale

VAS: Visual Analogue Scale