Impact of a supportive text messaging program (Text4Support) for mitigating psychological problems in patients receiving formal mental health services: A randomized controlled trial

Abstract

Background

Text-based interventions are an innovative way to bridge gaps in mental health care. Programs like Text4Support provide accessible, cost-effective mental health support using daily messages based on cognitive behavioral principles, reaching a broad population.

Objective

The study aims to evaluate the effectiveness of Text4Support in improving mental health outcomes compared to usual care in patients receiving formal mental health services.

Methods

A randomized controlled trial was conducted on participants attending community mental health programs and those recently discharged from psychiatric inpatient and emergency care in Nova Scotia, Canada. Participants were assigned to either the Text4Support or control group. Mental health conditions were measured using validated scales. Patients in the Text4Support group received daily supportive text messages, whilst the control group received a single text message with a link to the Nova Scotia Health e-mental health resources.

Results

Seven hundred eighty-one eligible patients were randomized, and 307 were included in the analysis. After adjusting for baseline scores, the third and ninth questions of the PHQ-9 showed a significant difference between the Text4Support group and the control group. When results were assessed based on self-reports from control group participants as to whether they accessed or did not access the NS Health e-mental health resource link, significant differences between the three groups were noted, with improvements for suicidal and/or self-harm ideation and disturbed sleep in the Text4Support group and deterioration observed in the two control groups.

Conclusion

Text4Support addresses critical psychological symptoms like sleep disturbances and suicidal and/or self-harm ideation during vulnerable periods in patients’ mental health journeys. Text4Support is a promising adjunct to conventional mental health care.

Trial registration

ClinicalTrials.gov (NCT05411302). Registered on 6 June 2022.

Keywords

Mental health sleep disturbance suicidal and/or self-harm ideation Nova Scotia text-based intervention

Introduction

Irrespective of the availability of evidence-based treatments, many individuals encounter barriers to timely and adequate mental health care.¹ Psychiatric patients often face heightened vulnerability at key transition points in their journey, such as when they first engage with a mental health program or following discharge from an inpatient unit.^2,3 These periods are challenging while patients adjust to new environments and routines, highlighting the need for tailored support to foster stability and connection to care.⁴

The shortage of mental health professionals and the absence of personalized care exacerbate these challenges, often leading to poor treatment adherence and dissatisfaction among individuals.^5,6 Addressing these systemic barriers can reduce unnecessary reliance on high-cost services while providing patients with cost-effective, personalized interventions that support recovery and long-term well-being.

Digital interventions, particularly text-based programs, have arisen as scalable, innovative solutions to bridge these gaps. Text-based programs like Text4Hope, Text4PTSI, and Text4Mood use cognitive-behavioral therapy (CBT) principles to deliver supportive messages that focus on specific symptoms, such as sleep disturbances and suicidal and/or self-harm ideation.^7–9 Research has revealed the effectiveness of such interventions with significant reductions in various mental health symptoms.^10,11 Various studies have emphasized the scalability and accessibility of digital mental health solutions, particularly in deprived settings.¹²

The principle behind these improvements includes both behavioral and psychological factors. Individuals may internalize positive affirmations or coping strategies delivered through the messages, leading to shifts in perspective and reduced hopelessness, a key driver of suicidal and/or self-harm ideation.¹⁰ Other studies have revealed that text-based messages often include strategies that promote sleep hygiene, reduce cognitive hyperarousal, and establish routines, thereby addressing underlying contributors to poor sleep quality.^13,14

However, gaps remain in understanding the specific benefits of text messaging interventions, particularly as add-on services for psychiatric patients navigating transitional periods, such as after discharge or during their first engagement with community mental health programs. While evidence supports the effectiveness of digital interventions in addressing individual symptoms, their broader impact on mental health outcomes has not been thoroughly studied.¹⁵ This study contributes to the growing evidence supporting stepped and technology-based mental health care by addressing these gaps.

This randomized controlled trial assesses the impact of Text4Support, a CBT-based supportive text messaging program, on mental health outcomes, including sleep disturbances and suicidal and/or self-harm ideation, implemented as an add-on service for patients attending or discharged from psychiatric care in Nova Scotia (NS), Canada.

Methods

Study design, randomization, and procedures

This randomized controlled trial (RCT) adopts a multicenter, longitudinal, parallel two-arm design with a rater-blinded methodology. The study utilized permuted blocks to maintain a balanced 1:1 allocation ratio between the groups. To guarantee allocation concealment, a random number generator was utilized for simple random allocation. Participants were recruited and randomized into the intervention group (IG) and the control group (CG). The procedure for data collection is fully described in the study protocol.¹⁵ Study participants were not blinded, as it is impossible to conceal which intervention (Text4Support program or CG) they were assigned to due to the nature of the interventions. However, outcome assessors were blinded to treatment group allocation to minimize bias in the assessment. Data analysts were blinded during data analysis with respect to treatment group allocation. The Text4Support program is a province-wide initiative offered through NS Health and is not specific to any one university or academic institution.

Ethics and Institutional Review Board approval

The study was approved by the NS Health Research Ethics Board (REB File #1028174), and the protocol was registered with ClinicalTrials.gov (NCT05411302). Each study participant provided informed consent, and the study was conducted according to the Declaration of Helsinki¹⁶ and Good Clinical Practice (Canadian Guidelines).¹⁷ The results were documented using the Consolidated Standards of Reporting Trials (CONSORT) criteria for reporting clinical trials.¹⁸ During the course of the study, neither the Text4Support nor the control (e-mental health resources) groups experienced any negative or unexpected consequences. Both non-invasive, digital interventions were offered as supplements to clinical care. Participants were equipped with resources for support in case of mental or psychological distress.

Study participants

The study was conducted in NS, Canada, with participants recruited from the mental health and addiction program (MHAP), including the community MHAPs (CMHAPs), emergency departments (EDs), Rapid Access and Stabilization Program (RASP), specialty programs (SPs), addiction programs (APs), and psychiatric inpatient units (PIUs) across the province.

Sample size consideration

To detect a 20% difference in change scores for each primary outcome, we need a sample size of 131 participants per study arm. This is based on a population variance of 5.0, a two-sided significance level (α = 0.05), 90% power, and an acceptable difference of zero between the sample mean and population mean (μ−μ0 = 0).^19,20 Thus, the total sample size for the RCT is 262 participants. Considering that the maximum completion rate for online surveys via text message is about 20%, we plan to enroll at least 1500 participants in the RCT to ensure 262 completed responses at each time point.

Inclusion and exclusion criteria

To be eligible to participate in the study, patients must satisfy the inclusion criteria below.

Age 18 years and above

Participants who are receiving mental health and addiction services through the MHAP in NS.

Participant who possesses a mobile phone with text message capability and is familiar with text messaging technology (no data plan is required)

Participants must be able to read English and provide written consent.

Patients will be ineligible if they do not meet the inclusion criteria or refuse to consent to the study.

Participants flowchart

Figure 1 illustrates the study flowchart. Eight hundred and ninety-eight patients were assessed for eligibility to enter the trial, of whom 781 eligible patients were randomized, and 307 were included in the analysis. Thirty-eight indicated they accessed the NS Health link, while 105 reported they did not access the NS Health link, and 24 did not indicate whether they accessed the link or not.

Figure 1.

CONSORT flowchart.

Data collection

Data were collected through web self-administered surveys powered by the REDCap software program.²¹ We obtained participants’ social demographics (e.g. age, gender, income, education, housing, and relationship status) and clinical information. Validated scales assessed participants’ mental health symptoms at baseline, 6 weeks, 3 months, and 6 months post-enrollment. Data were collected between 11 October 2022 and 31 July 2024. An electronic consent form (e-Consent) hosted on REDCap was employed to obtain participants’ consent after they had been thoroughly informed about the study and provided with an information leaflet. Participants completed the baseline surveys on enrollment via the online link. A text message with follow-up surveys was sent to all participants at 6 weeks, 3 months, and 6 months.

Intervention

The Text4Support program delivered one-way, free, non-interactive daily supportive text messages at 9:00 a.m. Atlantic Daylight Time (ADT) to subscribers for 6 months, and each message was unique across the six-month intervention period. The messages were developed by psychiatrists, psychologists, mental health therapists, and end-users. They were reviewed, adjusted for clarity, and uploaded onto a web-based platform powered by ResilienceNHope. Text4Support delivers general and diagnostic-specific content and covers various mental health conditions. Participants could select the one message category that best aligns with their primary mental health concerns (e.g. depression, anxiety, bipolar disorder, substance use, etc.). This selection determined the type of daily supportive messages they received. No further tailoring occurred beyond the participant's initial category selection.

Examples of the text messages included the following:

Practice self-compassion today. Acknowledge the progress you are making. You might not enjoy activities as much as in the past, but some enjoyment is better than none and is an important step forward.

When anxious, our thoughts often focus on future “danger.” Shift your attention to the present. What is happening right now?

Try talking quietly back to voices. Tell them they are wrong. Using the vocal part of the brain can reduce the intensity of voices.¹⁵

Participants who were randomized into the CG received a single text message containing a URL directing participants to the NS Health MHAP website. This resource was provided in addition to their usual care. The website offers free, publicly available, evidence-based e-mental health resources designed to address various psychiatric conditions, with programs including the early psychosis program, the eating disorder program, and the recovery support.²²

Outcome measures

The primary outcome measure was the change in mean scores on clinical scales between the Text4Support group and CG from baseline to the six-month follow-up. The secondary outcome measure was the change in the mean scores on clinical scales between the Text4Support group and the two CGs (participants who accessed the NS Health Link and those who did not) from baseline to the six-month follow-up. Accessing these resources include clicking the weblink, reading the content, or utilizing any of the mental health tools provided on the website. Participants’ recovery, quality of life, resilience, likely major depressive disorder (MDD), likely generalized anxiety disorder, sleep disturbance, and suicidal and/or self-harm ideation were assessed through the Recovery Assessment Scale (RAS),²³ the World Health Organization's Five Well-Being Index (WHO-5),²⁴ the Brief Resilience Scale (BRS),²⁵ the Patient Health Questionnaire-9 (PHQ-9),²⁶ Generalized Anxiety Disorder-7 scale,²⁷ and the third and ninth questions of the PHQ-9, respectively.

Participants’ quality of life was evaluated using the WHO-5, a simple and widely used tool comprising five items rated on a 6-point Likert scale.²⁴ The total score is calculated by adding the responses, ranging from 0 to 25. A score below 13 indicates poor mental well-being and a poor quality of life. The WHO-5 scale demonstrates strong clinometric validity, making it a reliable tool for assessing treatment outcomes.²⁴ Additionally, it serves as a sensitive and specific screening instrument for depression, with broad applicability across various research domains.²⁴ The BRS was used to assess participants’ ability to recover from stress. A score ranging from 1.00 to 2.99 indicates low resilience, while a score ranging from 3.00 to 5.00 indicates high to normal resilience.^25,28 Regarding reliability and validity, literature shows that the BRS has good internal consistency, with Cronbach alphas ranging from 0.80 to 0.90, and test–retest reliability coefficients for a two-week interval were fair (0.61–0.69).²⁵

The PHQ-9 tool is a nine-item measure on a 4-point Likert scale with a score of 0–9 for normal to mild, 10–14 for moderate, 15–19 for moderately severe, and 20–27 for severe.²⁶ The reliability and validity of the tool indicate that it has sound psychometric properties, and the internal consistency of the PHQ-9 is high.²⁶ The GAD-7 scale was used to assess the likelihood of anxiety symptoms. The seven self-reported items are rated on a four-point Likert scale: 0 (not at all) to 3 (nearly every day), with a score range of 0–21.²⁹ The internal consistency and test-retest reliability of the GAD-7 were good, and it also provided good criterion, construct, factorial, and procedural validity.³⁰

Participants were asked whether they had any suicidal and/or self-harm ideation. This was achieved via the ninth question of the PHQ-9 scale: passive death wishes or thoughts of self-harm in the last 2 weeks.

Participants were asked whether they had any sleep disturbances. The third question of the PHQ-9 scale measured trouble falling or staying asleep, or sleeping too much in the last 2 weeks.

Statistical analysis

The data were analyzed per the CONSORT guidelines,¹⁸ using SPSS for Windows, version 28 (IBM Corporation, Armonk, NY, USA).³¹ An intention-to-treat (ITT) approach was adopted to ensure participants who completed at least one follow-up survey were analyzed within their initial assigned groups. Descriptive data for the two groups’ baseline socio-demographic and clinical characteristics were examined using chi-squared tests/Fisher's exact tests for the categorical variables, presented as frequencies and percentages, and the independent sample t-test for continuous variables.

Differences in the mean scores of the clinical scales were assessed using analysis of covariance (ANCOVA) analysis, comparing the changes in mean scores from baseline to 6 months follow-up on the PHQ-9, GAD-7, WHO-5 Well-Being Index, BRS, RAS, PHQ-9 ninth and third questions scales between the IG and CG, while controlling for their respective baseline scores. Models were run for each outcome scale. The independent variables consisted of the group type (CG or Text4Support group). At the same time, the scores on the clinical scales at 6 months were considered the dependent variables for each of the analyses. Baseline scores on the respective scales were used as the covariates in the analyses. We also compared the differences in the effectiveness of the intervention participants received among the actual group type; among the CG, some indicated they utilized the NS Health resource through the web link provided, while others did not utilize the resources on the website (achieving two groups within the CG). We used the actual group type (utilized the NS Health resources, did not utilize the NS Health resources, and the Text4Support group) to compare the mean effect in the clinical scales. The independent variables consisted of the actual group type and baseline scores as the covariate. In contrast, the scores on the clinical scales at 6 months were considered the dependent variables for each analysis. The effect size was computed using partial eta squared for the ANCOVA analysis.

To address the issue of multiple testing and potential inflation of Type I errors, the p-value was adjusted. A Bonferroni-corrected, two-tailed criterion (α<.007) was used to determine statistical differences in the clinical scales.

An exploratory post-hoc analysis was performed on the third and ninth questions of the PHQ-9 scale to assess the intervention's effects on sleep and suicidal ideation and identify specific group differences in clinical outcomes. This analytic approach is consistent with prior studies on digital mental health interventions, where item-level analyses of the PHQ-9, particularly items 3 and 9, have been used to explore changes in sleep disturbance and suicidal ideation beyond the total score differences.^7,32,33

Pairwise comparisons between the three study groups—Text4Support, control (NS Health Link), and control (without NS Health Link)—were conducted using a Bonferroni-adjusted post hoc test following analysis of variance (ANOVA). The assumptions of normality, linearity, homogeneity of variance, and regression slopes were validated to ensure reliable covariate measurement and adherence to statistical assumptions.

To handle missing data, we utilized imputation techniques, focusing on the last observation carried forward (LOCF) method. This approach replaces missing values with the most recent recorded observation for the same individual. By applying LOCF, we included all participants in the analysis despite incomplete data, maintaining the sample size and reducing the risk of diminished statistical power.

Results

Baseline sociodemographic and clinical characteristics were assessed across the group type, as illustrated in Table 1. The majority of participants were between 26 and 40 years old (121, 39.4%), female (196, 63.8%), and Caucasian (266, 86.9%). Most participants were employed (160, 52.6%), had attained post-secondary education (college or university) 183, 60.0%, and reported partnered/married relationship status (143, 46.6%). An income of < $29,590 was the most common, accounting for 30.3% of 90 participants; 149 (48.7%) participants reported living in rented accommodations, while depression (39, 22.8%) and anxiety (37, 21.6%) were the most common primary diagnoses among participants. The results found no statistically significant differences between groups, indicating no significant variation in baseline sociodemographic characteristics and the clinical measures between the CG and the Text4Support group.

Table 1.

Sociodemographic and clinical characteristics between the study groups at baseline.

Variable	Group type		Total	Chi²/Fisher's exact^a	p-value
Variable	Control group N = 167 (%)	Text4Support group N = 140 (%)	Total	Chi²/Fisher's exact^a	p-value
Age (years)
25	28 (16.8)	24 (17.1)	52 (16.9)	7.9	0.05
26–40	73 (43.7)	48 (34.3)	121 (39.4)
41–60	59 (35.3)	51 (36.4)	110 (35.8)
> 60	7 (4.2)	17 (12.1)	24 (7.8)
Gender
Male	50 (29.9)	45 (32.1)	95 (30.9)	1.1	0.6
Female	110 (65.9)	86 (61.4)	196 (63.8)
Other	7 (4.2)	9 (6.4)	16 (5.2)
Ethnicity
Caucasian	144 (86.2)	122 (87.8)	266 (86.9)	6.8^a	0.1
Indigenous	14 (8.4)	4 (2.9)	18 (5.9)
Black/Hispanic	5 (3.0)	7 (5.0)	12 (3.9)
Asian	2 (1.2)	5 (3.6)	7 (2.3)
Other	2 (1.2)	1 (0.7)	3 (1.0)
Employment status
Student	16 (9.6)	12 (8.7)	28 (9.2)	3.5	0.5
Employed	87 (52.4)	73 (52.9)	160 (52.6)
Unemployed	38 (22.9)	23 (16.7)	61 (20.1)
Retired	13 (7.8)	14 (10.1)	27 (8.9)
Other	12 (7.2)	16 (11.6)	28 (9.2)
Educational status
Elementary school	4 (2.4)	2 (1.4)	6 (2.0)	4.9^a	0.3
High school	46 (27.7)	41 (29.5)	87 (28.5)
Post-secondary (college/university)	105 (63.3)	78 (56.1)	183 (60.0)
Post-secondary (trade school)	10 (6.0)	14 (10.1)	24 (7.9)
Other	1 (0.6)	4 (2.9)	5 (1.6)
Relationship status
Single	66 (39.5)	57 (40.7)	123 (40.1)	1.8	0.8
Partnered/married	78 (46.7)	65 (46.4)	143 (46.6)
Separated/divorced	19 (11.4)	13 (9.3)	32 (10.4)
Widowed	3 (1.8)	2 (1.4)	5 (1.6)
Other	1 (0.6)	3 (2.1)	4 (1.3)
Income range (Stats Canada)
< $29,590	51 (32.1)	39 (28.3)	90 (30.3)	2.4	0.8
$29,592–$59,180	41 (25.8)	32 (23.2)	73 (24.6)
$59,181–$93,000	35 (22.0)	37 (26.8)	72 (24.2)
$93,001–$150,000	19 (11.9)	14 (10.1)	33 (11.1)
> $150,000	7 (4.4)	8 (5.8)	15 (5.1)
No income	6 (3.8)	8 (5.8)	14 (4.7)
Housing status
Own home	51 (30.7)	47 (33.6)	98 (32.0)	2.3	0.3
Rented accommodation	87 (52.4)	62 (44.3)	149 (48.7)
Live with family/friends	28 (16.9)	31 (22.1)	59 (19.3)
Primary diagnosis
Depression	18 (19.4)	21 (26.9)	39 (22.8)	5.7^a	0.6
Anxiety	16 (17.2)	21 (26.9)	37 (21.6)
Bipolar disorder	14 (15.1)	7 (9.0)	21 (12.3)
Substance use disorder	4 (0.3)	3 (3.8)	7 (4.1)
Alcohol use disorder	4 (4.3)	3 (3.8)	7 (4.1)
Trauma-related disorders	13 (14.0)	8 (10.3)	21 (12.3)
Psychosis	8 (8.6)	4 (5.1)	12 (7.0)
Other	16 (17.2)	11 (14.1)	27 (15.8)
WHO-5 Well-being Index	Mean score (SD)		t-Test (df)
	N = 167	N = 140
	6.80 (4.92)	6.73 (4.99)	0.12 (305)		0.8
RAS	N = 167	N = 140
RAS	76.49 (14.32)	75.74 (15.73)	0.44 (305)		0.66
BRS	N = 166	N = 140
BRS	2.43 (0.7)	2.52 (0.7)	−1.06 (304)		0.29
PHQ-9	N = 167	N = 140
PHQ-9	17.15 (6.31)	16.91 (6.47)	0.32 (305)		0.75
PHQ-9 third question	N = 167	N = 140
PHQ-9 third question	0.92 (0.28)	0.86 (0.34)	1.46 (305)		0.15
PHQ-9 ninth question	N = 167	N = 140
PHQ-9 ninth question	0.59 (0.49)	0.62 (0.48)	−0.40 (305)		0.1
GAD-7	N = 167	N = 140
GAD-7	14.59 (5.45)	14.19 (5.43)	0.64 (305)		0.52

WHO-5: World Health Organization's Five Well-Being Index; RAS: Recovery Assessment Scale; BRS: Brief Resilience Scale; PHQ-9: Patient Health Questionnaire-9; GAD-7: Generalized Anxiety Disorder-7.

We conducted an ANCOVA to evaluate the effects of the Text4Support intervention compared to a CG across several mental health measures, controlling for baseline scores (Table 2). The WHO-5, RAS, BRS, and GAD-7 showed no statistically significant difference between the Text4Support group and CG from baseline to follow-up. An analysis of the overall mean scores of the PHQ-9 showed no significant difference in overall scores between the Text4Support group and the CG, F(1) = 1.23, p = 0.27, partial η² = 0.004. However, significant differences were observed in two specific PHQ-9 items: The third question of the PHQ-9 showed a significant difference between the Text4Support group (M = 0.86, SD = 0.36) and the CG (M = 1.91, SD = 1.03), F(1) = 125.70, p < 0.001, partial η² = 0.31. The ninth question of the PHQ-9 also revealed a significant difference between the Text4Support group (M = 0.48, SD = 0.50) and the CG (M = 0.93, SD = 1.09), F(1) = 25.2, p < 0.001, partial η² = 0.08. Table 2 also reveals that participants in the CG experienced 56.6% more suicidal and/or self-harm ideation at follow-up than at baseline, while participants in the Text4Support group reported a 22.5% improvement in suicidal and/or self-harm ideation as recorded on the PHQ-9-9th question. Similarly, sleep disturbance, as measured on the PHQ-9-3rd question, showed that participants in the CG experienced a 107.6% worsening of sleep disturbance, while participants in the Text4Support group did not experience any change in sleep disturbance (0%).

Table 2.

Descriptive mean scores of outcome measures and ANCOVA test parameters between the two groups.

Measure	Descriptive		ANCOVA parameters
	Baseline, Mean (SD)		Follow-up after controlling for baseline, Mean (SD)		Mean difference (95% CI)	F value (df)	p-value	Partial eta squared
	Control group	Text4Support group	Control group	Text4Support group	Mean difference (95% CI)	F value (df)	p-value	Partial eta squared
WHO-5	6.80 (4.92)	6.73 (4.99)	9.04 (5.76)	9.24 (5.70)	−0.23 (−1.40–0.95)	0.14 (1)	0.71	0.001
RAS	76.49 (14.32)	75.74 (15.73)	78.16 (18.81)	81.18 (15.49)	−3.37 (−6.96−0.22)	3.42 (1)	0.07	0.01
BRS	2.43 (0.7)	2.52 (0.7)	2.55 (0.81)	2.67 (0.82)	−0.07 (−0.23−0.09)	1.51 (1)	0.22	0.005
PHQ-9	17.15 (6.31)	16.91 (6.47)	14.23 (7.17)	13.25 (7.0)	0.79 (−0.61–2.18)	1.23 (1)	0.27	0.004
PHQ-9 third question	0.92 (0.28)	0.86 (0.34)	1.91 (1.03)	0.86 (0.36)	1.01 (0.83–1.19)	125.70 (1)	<0.001	0.31
PHQ-9 ninth question	0.59 (0.49)	0.62 (0.48)	0.93 (1.09)	0.48 (0.50)	0.46 (0.28–0.65)	25.2 (1)	<0.001	0.08
GAD-7	14.59 (5.45)	14.19 (5.43)	11.10 (6.14)	11.24 (6.25)	−0.34 (−1.59–0.92)	0.29 (1)	0.59	0.001

ANCOVA: analysis of covariance; WHO-5: World Health Organization's Five Well-Being Index; RAS: Recovery Assessment Scale; BRS: Brief Resilience Scale; PHQ-9: Patient Health Questionnaire-9; GAD-7: Generalized Anxiety Disorder-7.

We conducted ANCOVA to evaluate the effects of the Text4Support intervention compared to two CGs (NS Health Link and without NS Health Link) across several mental health measures, controlling for baseline scores; the results are summarized in Table 3. No significant differences were observed for all clinical scales except the third, F(2) = 66.25, p < 0.001, partial η² = 0.33, and ninth, F(2) = 15.26, p < 0.001, partial η² = 0.10, questions of the PHQ-9 which showed a significant difference between the three groups, with the independent variable explaining 33% of the variance in scores for this specific question. Similarly, the group differences in the ninth question accounted for 10% of the variance in scores, representing a moderate effect size. Table 3 also reveals that participants in the CG who accessed the NS Health link experienced 70% more suicidal and/or self-harm ideation symptoms at follow-up compared to baseline, while participants in the CG who did not access the NS Health link reported 52% worse suicidal and/or self-harm ideation at follow-up from the time of enrollment. In comparison, participants in the Text4Support group reported a 24% improvement in suicidal and/or self-harm ideation at follow-up, as recorded on the PHQ-9-9th question. Similarly, sleep disturbance measured on the PHQ-9-3rd question showed that participants in the CG (who access the NS Health link) experienced 126% worsening sleep disturbance, the CG who did not access the NS Health link reported 107.6% worsening sleep disturbance, while participants in the Text4Supprot group did not record a change in mean scores for sleep disturbance at follow-up compared to baseline (0%).

Table 3.

Descriptive mean scores of outcome measures and ANCOVA test parameters for the three groups.

Measure	Descriptive						ANCOVA Parameters
	Baseline, Mean (SD)			Follow-up, Mean (SD)			F value (df)	p-value	Partial eta squared
	Text4 Support group	Control (NS Health Link) group	Control (without NS Health Link) group	Text4Support group	Control (NS Health Link) group	Control (without NS Health Link) group	F value (df)	p-value	Partial eta squared
WHO-5	6.74 (4.98)	7.47 (5.23)	6.59 (4.84)	9.24 (5.7)	8.53 (4.89)	9.33 (6.02)	0.79 (2)	0.45	0.006
RAS	75.74 (15.73)	79.79 (13.52)	74.88 (14.29)	81.18 (15.49)	76.87 (21.56)	78.49 (18.01)	2.65 (2)	0.07	0.02
BRS	2.52 (0.73)	2.41 (0.76)	2.39 (0.73)	2.68 (0.82)	2.64 (0.72)	2.53 (0.87)	0.44 (2)	0.65	0.003
PHQ-9	16.91 (6.48)	16.92 (6.29)	17.63 (6.15)	13.26 (7.0)	15.61 (6.79)	13.83 (7.26)	2.38 (2)	0.09	0.02
PHQ-9 third question	0.86 (0.34)	0.92 (0.27)	0.91 (0.28)	0.86 (0.36)	2.08 (0.94)	1.89 (1.06)	66.25(2)	<0.001	0.33
PHQ-9 ninth question	0.62 (0.49)	0.74 (0.45)	0.56 (0.49)	0.47 (0.5)	1.26 (1.11)	0.85 (1.07)	15.26 (2)	<0.001	0.1
GAD-7	14.19 (5.43)	15.24 (5.04)	14.59 (5.30)	11.24 (6.25)	12.37 (6.34)	10.61 (5.98)	1.16 (2)	0.31	0.009

Post-hoc analyses were conducted to examine the specific differences between groups for the two PHQ-9 items that showed significant overall effects in the ANCOVA, as illustrated in Table 4. For the PHQ-9 third question, significant differences were observed between the Text4Support group and both CGs. The Text4Support group showed significantly lower scores compared to the control (NS Health Link) group (mean difference = –1.17, SE = 0.13, p < 0.001, 95% CI [–1.44, −0.90]) and the control (without NS Health Link) group (mean difference = −0.98, SE = 0.09, p < 0.001, 95% CI [−1.18, −0.79]). There was no significant difference between the two CGs. Similar patterns were observed for the PHQ-9 ninth question. The Text4Support group demonstrated significantly lower scores compared to the control (NS Health Link) group (mean difference = −0.69, SE = 0.14, p < 0.001, 95% CI [-0.98, −0.42]) and the control (without NS Health Link) group (mean difference = −0.41, SE = 0.10, p < 0.001, 95% CI [−0.61, −0.21]). The difference between the two CGs did not show significance when applying the Bonferroni adjusted alpha level of 0.025.

Table 4.

Post-hoc analysis of PHQ-9 item scores across Text4Support and control groups.

Variable		Mean difference	Standard error	p-value	95% confidence interval
Variable		Mean difference	Standard error	p-value	Lower bound	Upper bound
PHQ-9 third question
I	J
Text4Support	Control (NS Health Link) group	−1.17	0.13	<0.001	−1.44	−0.90
Text4Support	Control (without NS Health Link) group	−0.98	0.09	<0.001	−1.18	−0.79
Control (NS Health Link) group	Text4Support	1.17	0.13	< 0.001	0.90	1.44
Control (NS Health Link) group	Control (without NS Health Link) group	0.19	0.14	0.19	−0.09	0.47
Control (without NS Health Link) group	Text4Support	0.98	0.09	< 0.001	0.79	1.18
Control (without NS Health Link) group	Control (NS Health Link) group	-0.19	0.14	0.19	−0.47	0.09
PHQ-9 ninth question
Text4Support	Control (NS Health Link) group	−0.69	0.14	< 0.001	−0.98	−0.42
Text4Support	Control (without NS Health Link) group	−0.41	0.10	< 0.001	−0.61	−0.21
Control (NS Health Link) group	Text4Support	0.69	0.14	< 0.001	0.42	0.98
Control (NS Health Link) group	Control (without NS Health Link) group	0.29	0.15	0.05	−0.001	0.58
Control (without NS Health Link) group	Text4Support	0.41	0.10	< 0.001	0.21	0.61
Control (without NS Health Link) group	Control (NS Health Link) group	−0.29	0.15	0.05	−0.58	0.001

PHQ-9: Patient Health Questionnaire-9; NS: Nova Scotia.

Discussion

This RCT assessed the effects of a CBT-based supportive text messaging program (Text4Support) compared to usual care on enhancing transdiagnostic mental health outcomes among patients receiving varying levels of formal mental health care in NS. The key finding in the study revealed that participants receiving daily supportive text messages on their cell phones showed improvements in two particular symptoms of depression—suicidal thoughts/self-harm and sleep disturbances, compared to those who only received a URL to the NS Health website. The overall clinical total differences were not statistically significant; however, the outcomes indicate areas of symptoms where text-based interventions can have a measurable impact. These results align with previous research suggesting that supportive text messaging may enhance outcomes in mental health. For instance, research conducted on such interventions like Text4Hope and Text4Mood has also noted improvement in depression, anxiety, and stress symptoms in varied populations.^34,35 This work adds to existing literature by defining symptom domains in which support via text messaging can have the greatest impact in transitional or vulnerable stages of treatment, such as upon discharge from inpatient or emergency mental health treatment.

The study revealed significant enhancement effects of the Text4Support program compared to usual care in two specific depressive symptom domains—suicidal and/or self-harm ideation and sleep disturbance—as measured by the PHQ-9 ninth and third questions, respectively. While other clinical symptoms also showed reduction in the Text4Support group compared to the CG, these reductions in symptoms were not statistically significant. The results showed that participants in the Text4Support group demonstrated a 22.5% improvement in suicidal and/or self-harm ideation, as measured by the PHQ-9 ninth question, while the CG experienced a 56.6% worsening in suicidal and/or self-harm ideation over the same period. In contrast, sleep disturbance showed no improvement in the Text4Support group (0% change), while participants in the CG experienced a 107.6% worsening in sleep disturbance. These findings align with other research highlighting the potential of digital mental health interventions in addressing acute and specific mental health symptoms. The approach holds considerable potential for the future of mental health care, inspiring a sense of hope and optimism.

Sleep disturbance is a common and debilitating symptom of depression, often exacerbating other depressive symptoms.¹³ It is often among the earliest symptoms to appear and one of the last to resolve, serving as a significant indicator of symptom burden and treatment efficacy.³⁶ The Text4Support intervention significantly reduced sleep disturbance scores compared to the control (accessed the NS Health Link) and control (did not access the NS Health Link) groups. The two CGs recorded no statistically significant differences, indicating that the impact of Text4Support is not solely attributable to general mental health care access but to its distinct features. The improvement highlights the potential of Text4Support to address this critical aspect of mental health. This finding also suggests that while the Text4Support intervention may help prevent further deterioration in sleep, it may not be a sufficient solution for addressing sleep-related issues.

A clinical trial conducted in Spain reported that sleep patterns improve when supportive messages focus on behavioral activation and stress management.³⁷ This improvement may be due to the unique way text-based interventions offer coping strategies that help reinforce healthy sleep habits by increasing self-assurance in addressing mental health challenges effectively.³⁷

Other studies have reported similar reductions in sleep disturbances among individuals receiving daily supportive text messages, attributing these improvements to the intervention's capacity to reinforce healthy sleep habits and provide emotional reassurance.^38–40

The Text4Support intervention demonstrated a significant reduction in suicidal and/or self-harm ideation compared to the CG. The study data show that participants in the CG experienced 56.6% more suicidal and/or self-harm ideation at the six-month follow-up compared to baseline, while participants in the Text4Support group reported a 22.5% improvement in suicidal and/or self-harm ideation. These results align with other research highlighting the potential of digital mental health interventions to mitigate suicidal and/or self-harm ideation, particularly in populations with limited access to traditional care.⁷ On the contrary, no significant differences were found between the two CGs for suicidal and/or self-harm ideation, and participants in the CG experienced worsening mental health symptoms irrespective of their engagement with the NS Health Link, suggesting that there are distinct advantages of the Text4Support program over simply having access to other digital mental health interventions for addressing these specific depressive symptom domains. While participation in the Text4Support program is effortless for the participant after enrollment, other digital mental health programs require registration, login, and regular, purposeful engagement with mobile applications or web-based systems to participate. The daily supportive messages likely provided a consistent source of encouragement and validation, creating a sense of connection and reducing feelings of isolation, which are commonly associated with suicidal and/or self-harm ideation.^7,10,41 A comparative cross-sectional study with two arms using text message intervention revealed that text interventions help to reduce suicidal and/or self-harm ideation.⁷

These results are consistent with the findings from the literature, which identified digital mental health tools as effective in reducing suicide risk factors through immediate and personalized support.¹⁰ Similarly, research has demonstrated the utility of SMS-based interventions in lowering distress and suicidal thoughts among at-risk individuals.⁴² The success of the Text4Support program can be attributed to its effortless patient participation, which contrasts with other e-mental health programs that require participants to register and log in regularly. Additionally, Text4Support offers timely, non-intrusive messages that cater to individuals’ specific needs, fostering a sense of care and support. Again, the consistency and predictability of these messages may have helped establish a routine, which is often lacking in individuals struggling with severe depressive symptoms. This routine may enhance a sense of stability and control, thereby reducing the intensity of suicidal thoughts.¹¹ The Text4Support intervention was an add-on to patients’ usual care, which proved to be more effective support. Hybrid models combining digital tools with in-person therapy produced broader and more sustained outcomes.⁴³ However, while participants in the Text4Support group were given messages tailored to their primary concerns, this study did not evaluate different outcomes based on message categories. Future research could investigate whether specific message types are more effective for particular audiences or diagnostic groups.

While suicidal and/or self-harm ideation and sleep disturbance showed significant improvement in the Text4Support group compared to the CG, differences in other mental health outcomes, such as general well-being (WHO-5), resilience (BRS), depression (PHQ-9), and anxiety (GAD-7), did not reach statistical significance, although participants in the IG experienced the greatest reduction in symptoms. Our data's non-statistically significant improvement contrasts with research that found statistical improvements in psychological well-being with interactive digital interventions.⁴⁴ An explanation for this divergence may be that Text4Support delivers static, non-interactive messages, whereas interventions with interactive components may have a more comprehensive impact on broader psychological outcomes. Notwithstanding, various studies have reported significant reductions in anxiety with supportive text message interventions,^8,45–47 and the statistically non-significant changes observed in this study may be attributed to differences in the target population or intervention focus.

Implications for mental health services and policy

This research documents the supportive text messaging as a scalable and practical adjunct to standard treatment of mental illness. The improvement in the specified depressive symptoms, primarily suicidal and/or self-harm ideation and sleep disturbance, demonstrates the therapeutic value of their extremely brief, CBT-informed messages in augmenting the treatment of individuals with mental illness, particularly at high-risk transition points such as discharge from acute or emergency psychiatric care.

Digital interventions like Text4Support offer a new way to tackle some of the most intractable issues in the provision of mental health care, including wait times, provider shortages, and geographic or logistical barriers to care, issues most starkly evident in rural and underserved populations.^12,48 By engaging in low-intensity but high-reach support, these interventions can potentially bridge gaps in service and protect against relapse or worsening in wait-listed, as well as transitioning patients, thereby facilitating access to higher-level services.^8,49

Literature has shown that text messaging programs are highly cost-effective. Although this study did not analyze detailed cost-effectiveness data, text messaging interventions, such as Text4Support, are deemed economically viable due to their low delivery costs, minimal infrastructure needs, and capacity to reach large populations with limited resources. The cost-effectiveness and scalability of text-based interventions make them particularly interesting to publicly funded health systems. Research indicates that similar interventions cost ∼ $0.08–$0.15 CAD per message and around $5.40 CAD for 6 months of daily support, resulting in decreases in healthcare utilization, such as emergency visits and hospital stays.^50–52 These results imply that supportive text messaging can be a scalable addition to traditional mental health services, especially in areas facing workforce shortages or geographic challenges.⁵³

Interventions such as Text4Mood and Text4Hope have, in real-world settings, reported broad public feasibility and acceptability, with users reporting increased well-being and ease of use.^34,54 These factors make supportive texting an appealing adjunct to the stepped-care model of mental health care, where interventions are matched to patient needs and the intensity of service.⁵⁵

At the policy level, the adoption of supportive text messaging as part of regular care pathways can augment the continuum of mental health care and maximize resource allocation. Health authorities and policymakers should aim to incorporate such programs into provincial or national mental health approaches, including those under digital health transformation agendas. In addition, the fact that the content of such a message can be tailored to reflect patient preferences or diagnostic categories (as in the current study) aligns with the tenets of person-centered care and the objectives of mental health reform and development agendas.^56,57

Future policy also needs to enable the ongoing evaluation and enhancement of digital interventions, including investment in infrastructure for monitoring user uptake, outcomes, and cost-effectiveness. Given the global trend toward online mental health, initiatives like Text4Support can put Canada and the world at the forefront of advancing equitable and universally accessible care.

Future research directions

This study illustrates the potential advantages of supportive text messaging in mental health care. However, various avenues for future research need to be explored. This study reviewed outcomes over a six-month period; however, the longevity of improvements in mental health symptoms remains uncertain. Longer follow-up assessments, lasting 12 months or more, are crucial for determining whether these benefits persist or diminish over time.⁵⁸ Furthermore, future research could explore whether periodic booster messages or ongoing low-frequency support can maintain progress and prevent relapse.

Analyzing the cost-effectiveness of these systems will be essential for guiding decisions regarding their wider implementation in publicly funded settings.^49,54 Future research should include formal cost-effectiveness analyses to provide better insights for policy and funding decisions regarding the widespread adoption of these interventions.

Ultimately, conducting implementation research in diverse healthcare settings, particularly in middle- and low-income countries, may offer valuable insights into the generalizability and scalability of these interventions. Studying varied demographics, like young people, the elderly, and culturally diverse populations, can also enhance equity and inclusivity in the provision of digital mental health services.⁵⁹

Strengths and limitations

The strength of this study is its solid design and careful analysis. The RCT adhered to the CONSORT guidelines and clinical trial practices detailed in a related publication by Agyapong et al.^11,18,60 Additionally, the study analyzed and reported clinical outcomes gathered by a research assistant blinded to the study group allocation. Another strength of our study was the involvement of service users in drafting the supportive text messages used throughout the study.

While this study offers valuable insights into the effectiveness of a text-based intervention (Text4Support), several limitations should be acknowledged. Firstly, the study assessed outcomes over a six-month period, which may not capture the long-term effectiveness of the intervention. Mental health improvements, particularly in terms of general well-being, resilience, and anxiety, may require a longer duration to fully manifest. Second, although our statistical methods accounted for dropout attrition and non-completion, there is a possibility that those who left the IG were less satisfied with the program or experienced fewer positive outcomes. Third, we relied on self-reports to determine whether those in the CG accessed or did not access the NS Health Link. Although engagement data was gathered via self-reports, objective metrics were missing. Future research should investigate the integration of platform-based tracking with self-reports. Another limitation involves the exclusion of the Columbia-Suicide Severity Rating Scale (C-SSRS) from the analysis. Although it was included in the original study protocol, the C-SSRS is a clinician-rated instrument, and proper administration requires a trained individual to guide and assess participant responses in real time.⁶¹ During follow-up, many participants completed the surveys independently, without support from a clinician or research assistant, which made it infeasible to collect valid and reliable data using this tool. As a result, suicidal and/or self-harm ideation was instead examined using item 9 of the PHQ-9, which was self-administered and consistently completed. Finally, sleep disturbance was assessed with the third PHQ-9 question, as it is a clinically relevant question that identifies differences between groups; however, it is not a comprehensive assessment of sleep quality. Future studies should employ standardized measures of sleep.

Conclusion

This study demonstrates how Text4Support, a CBT-based supportive text messaging program, can help people accessing or leaving psychiatric care with acute mental health symptoms, including sleep difficulties and suicidal and/or self-harm ideation. The study results highlight Text4Support's effectiveness as an accessible, affordable, and scalable intervention, especially during critical moments when patients are most vulnerable. By providing consistent, evidence-based support, the program helps patients strengthen their connection to the healthcare system, encourages cognitive reframing, and reinforces healthy habits. Even while the intervention significantly reduced some symptoms, more research is needed to determine how it affected anxiety, resilience, and overall well-being more broadly. To improve their efficacy, future studies should investigate long-term outcomes, engagement strategies, and the potential benefits of incorporating interactive or blended care models. This study contributes to the growing evidence supporting the integration of CBT-based mental health interventions into standard care. The results are relevant to the local context in NS, Canada, and offer opportunities for jurisdictions across Canada and globally to implement innovative solutions to enhance mental health care delivery. Text4Support represents a significant step toward improving outcomes and reducing the burden on healthcare systems by addressing key gaps in mental health services.

Footnotes

ORCID iDs

Gloria Obuobi-Donkor

Reham Shalaby

Belinda Agyapong

Samuel Obeng Nkrumah

Medard Kofi Adu

Ejemai Eboreime

Lori Wozney

Vincent Israel Opoku Agyapong

Ethical approval

The study was approved by the Nova Scotia Health Research Ethics Board (REB File #1028174), and the protocol was registered with ClinicalTrials.gov (NCT05411302). Each study participant provided informed consent, and the study was conducted according to the Declaration of Helsinki¹⁶ and Good Clinical Practice (Canadian Guidelines).¹⁷

Contributorship

VIOA conceived, designed, and supervised the study. VIOA, RS, and GO-D conducted data analysis. The initial manuscript was drafted by GO-D, VIOA, RS, BA, SON, MKA, EE, and LW reviewed the initial draft. All the authors contributed to the study's design, revised the initial draft manuscript, and approved the final draft before submission. All authors have read and agreed to the published version of the manuscript.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by the Queen Elizabeth II Foundation (Funding Commitment #10257) and supported by the Global Psychological eHealth Foundation, the Department of Psychiatry at Nova Scotia Health, and Dalhousie University. The funders had no role in the study's design, execution, data collection, analysis, interpretation, or writing. They were also not involved in reviewing, approving, or deciding to publish the results.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: VIOA is the founder and Principal Investigator of the ResilienceNHope messaging programs and volunteers as the Board Chair of the Global Psychological eHealth Foundation, a not-for-profit organization. BA volunteers as the President and Chief Executive Officer of the Global Psychological eHealth Foundation. RS volunteers as the Program Director of the Global Psychological eHealth Foundation. VIOA, BA, and RS receive no financial or other compensation from the Global Psychological eHealth Foundation and have no financial conflicts of interest about this article. GO-D, SON, MKA, EE, and LW have no commercial or financial conflict of interest.

Data availability statement

The data supporting this manuscript's conclusions will be made available upon reasonable request from the corresponding author.

References

Aguirre Velasco

Cruz

ISS

Billings

, et al. What are the barriers, facilitators and interventions targeting help-seeking behaviours for common mental health problems in adolescents? A systematic review. BMC Psychiatry 2020; 20: 293.

Wang

Murray

Toussaint

A-K

, et al. Why is the recognition rate of psychological distress under-estimated in general hospitals? A cross-sectional observational study in China. Medicine (Baltimore) 2019; 98: e16274.

Faessler

Kutz

Haubitz

, et al. Psychological distress in medical patients 30 days following an emergency department admission: results from a prospective, observational study. BMC Emerg Med 2016; 16: 33.

Kornfield

Stamatis

CA.

Bhattacharjee

, et al. A text messaging intervention to support the mental health of young adults: user engagement and feedback from a field trial of an intervention prototype. Internet Interv 2023; 34: 100667.

Andrade

Alonso

Mneimneh

, et al. Barriers to mental health treatment: results from the WHO World Mental Health surveys. Psychol Med 2014; 44: 1303–1317.

Richards

. Stepped care: a method to deliver increased access to psychological therapies. Can J Psychiatry 2012; 57: 210–215.

Agyapong

VIO

, Shalaby R, Hrabok M, et al. Mental health outreach via supportive text messages during the COVID-19 pandemic: improved mental health and reduced suicidal ideation after six weeks in subscribers of Text4Hope compared to a control population. Int J Environ Res Public Health 2021; 18: 2157–2170.

Obuobi-Donkor

, Shalaby R, Eboreime E, et al. Text4PTSI: a promising supportive text messaging program to mitigate psychological symptoms in public safety personnel. Int J Environ Res Public Health 2023; 20: 4215–4232.

Agyapong

Mrklas

Juhás

, et al. Cross-sectional survey evaluating Text4Mood: mobile health program to reduce psychological treatment gap in mental healthcare in Alberta through daily supportive text messages. BMC Psychiatry 2016; 16: 378.

10.

Franklin

Ribeiro

Fox

KR.

, et al. Risk factors for suicidal thoughts and behaviors: a meta-analysis of 50 years of research. Psychol Bull 2017; 143: 187–232.

11.

Agyapong

VIO

Juhás

Ohinmaa

, et al. Randomized controlled pilot trial of supportive text messages for patients with depression. BMC Psychiatry 2017; 17: 286.

12.

Andersson

Titov

. Advantages and limitations of Internet-based interventions for common mental disorders. World Psychiatry 2014; 13: 4–11.

13.

Baglioni

Nanovska

Regen

, et al. Sleep and mental disorders: a meta-analysis of polysomnographic research. Psychol Bull 2016; 142: 969–990.

14.

Appleman

Salazar

, et al. Cognitive behavioral therapy for insomnia comorbid with psychiatric and medical conditions: a meta-analysis. JAMA Intern Med 2015; 175: 1461–1472.

15.

Dias

RDL

Shalaby

, Agyapong B, et al. Augmenting mental health support for patients accessing different degrees of formal psychiatric care through a supportive text messaging program: protocol for a randomized controlled trial. Methods Protoc 2023; 6: 19–34.

16.

Association

. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. Jama 2013; 310: 2191–2194.

17.

Good Clinical Practices, H. Canada, Editor. 2022, Government of Canada.

18.

Schulz

Altman

Moher

. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. Br Med J 2010; 340: c332.

19.

Benchmark Six Sigma: Sample Size Calculator for 2 Sample T-Tests. 2022 [cited 2022 April 2]; Available from: https://www.benchmarksixsigma.com/calculators/sample-size-calculator-for-2-sample-t-test/.

20.

O'Keeffe

Ambler

Barber

. Sample size calculations based on a difference in medians for positively skewed outcomes in health care studies. BMC Med Res Methodol 2017; 17: 157–157.

21.

Harris

Taylor

Thielke

, et al. Research electronic data capture (REDCap) – a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009; 42: 377–381.

22.

Authority

NSH

. Mental Health and Addictions Tools. 2025 [cited 2025 January 1]; Available from: https://mha.nshealth.ca/en/tools.

23.

Giffort

, Schmook A, Woody C, et al. Construction of a scale to measure consumer recovery. Springfield, IL: Illinois Office of Mental Health, 1995, 10: p 15487760500339360.

24.

Topp

Østergaard

Søndergaard

, et al. The WHO-5 Well-Being Index: a systematic review of the literature. Psychother Psychosom 2015; 84: 167–176.

25.

Smith

Dalen

Wiggins

, et al. The brief resilience scale: assessing the ability to bounce back. Int J Behav Med 2008; 15: 194–200.

26.

Kroenke

Spitzer

Williams

. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001; 16: 606–613.

27.

Spitzer

Kroenke

Williams

JBW

, et al. A brief measure for assessing generalized anxiety disorder: the GAD-7. Archives of Internal Medicine (1960) 2006; 166: 1092–1097.

28.

Smith

, Epstein EM, Ortiz JA, et al. The foundations of resilience: what are the critical resources for bouncing back from stress? In: Resilience in children, adolescents, and adults. Springer, 2013, pp 167–187.

29.

Johnson

Ulvenes

Øktedalen

, et al. Psychometric properties of the general anxiety disorder 7-item (GAD-7) scale in a heterogeneous psychiatric sample. Front Psychol 2019; 10: 1713.

30.

Spitzer

Kroenke

Williams

JBW

, et al. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med 2006; 166: 1092–1097.

31.

Corp., I., IBM Corp. Released 2021. IBM SPSS Statistics for Windows, Version 28.0. Armonk, NY: IBM Corp, in Google Search. 2021.

32.

Osiogo

, Shalaby R, Adegboyega S, et al. COVID-19 pandemic: demographic and clinical correlates of disturbed sleep among 6,041 Canadians. Int J Psychiatry Clin Pract 2021; 25: 1–8. doi:https://doi.org/10.1080/13651501.2021.1881127

33.

Dias

RDL

Shalaby

Agyapong

, et al. The effectiveness of CBT-based daily supportive text messages in improving female mental health during COVID-19 pandemic: results from the Text4Hope program. Front Glob Womens Health 2023; 4: 1182267.

34.

Agyapong

VIO

Mrklas

Juhás

35.

Agyapong

VIO

Hrabok

Vuong

, et al. Closing the psychological treatment gap during the COVID-19 pandemic with a supportive text messaging program: protocol for implementation and evaluation. JMIR Res Protoc 2020; 9: e19292–e19292.

36.

Nutt

Wilson

Paterson

. Sleep disorders as core symptoms of depression. Dialogues Clin Neurosci 2008; 10: 329–336.

37.

Aguilera

Bruehlman-Senecal

Demasi

, et al. Automated text messaging as an adjunct to cognitive behavioral therapy for depression: a clinical trial. J Med Internet Res 2017; 19: e148.

38.

Luik

van der Zweerde

van Straten

, et al. Digital delivery of cognitive behavioral therapy for insomnia. Curr Psychiatry Rep 2019; 21: 50.

39.

Aguilera

Muñoz

. Text messaging as an adjunct to CBT in low-income populations: a usability and feasibility pilot study. Prof Psychol Res Pr 2011; 42: 472–478.

40.

Gee

Orchard

Clarke

, et al. The effect of non-pharmacological sleep interventions on depression symptoms: a meta-analysis of randomised controlled trials. Sleep Med Rev 2019; 43: 118–128.

41.

Luxton

O'Brien

Pruitt

, et al. Suicide risk management during clinical telepractice. Int J Psychiatry Med 2014; 48: 19–31.

42.

Berrouiguet

Baca-García

Brandt

, et al. Fundamentals for future mobile-health (mHealth): a systematic review of mobile phone and web-based text messaging in mental health. J Med Internet Res 2016; 18: e135.

43.

Erbe

Eichert

H-C

Riper

, et al. Blending face-to-face and internet-based interventions for the treatment of mental disorders in adults: systematic review. J Med Internet Res 2017; 19: e306.

44.

Proudfoot

Clarke

Birch

M-R

, et al. Impact of a mobile phone and web program on symptom and functional outcomes for people with mild-to-moderate depression, anxiety and stress: a randomised controlled trial. BMC Psychiatry 2013; 13: 312.

45.

Donker

Blankers

Hedman

, et al. Economic evaluations of Internet interventions for mental health: a systematic review. Psychol Med 2015; 45: 3357–3376.

46.

Obuobi-Donkor

Shalaby

Vuong

, et al. Effects of Text4Hope-addiction support program on cravings and mental health symptoms: results of a longitudinal cross-sectional study. JMIR Form Res 2023; 7: e40440.

47.

Agyapong

, Shalaby R, Vuong W, et al. Text4Hope effectiveness in reducing psychological symptoms among young adults in Canada: longitudinal and naturalistic controlled program evaluation. J Clin Med 2023; 12: 1942–1958.

48.

Torous

Wisniewski

Liu

, et al. Mental health Mobile phone app usage, concerns, and benefits among psychiatric outpatients: comparative survey study. JMIR Ment Health 2018; 5: e11715.

49.

MacDougall

Jerrott

Clark

, et al. Text message interventions in adolescent mental health and addiction services: scoping review. JMIR Ment Health 2021; 8: e16508.

50.

Mental Health Innovation Network, Text4Mood. London, UK, 2016. Available from: https://www.mhinnovation.net/innovations/text4mood.

51.

Agyapong

Mrklas

Suen

VYM

, et al. Supportive text messages to reduce mood symptoms and problem drinking in patients with primary depression or alcohol use disorder: protocol for an implementation research study. JMIR Res Protoc 2015; 4: e55.

52.

Agyapong

Ahern

McLoughlin

, et al. Supportive text messaging for depression and comorbid alcohol use disorder: single-blind randomised trial. J Affect Disord 2012; 141: 168–176.

53.

Hollis C, Morriss R, Martin J, et al. Technological innovations in mental healthcare: harnessing the digital revolution. Br J Psychiatry 2015; 206: 263–265.

54.

Agyapong

VIO

, Hrabok M, Shalaby R, et al. Text4Hope: receiving daily supportive text messages for 3 months during the COVID-19 pandemic reduces stress, anxiety, and depression. Disaster Med Public Health Prep 2021; 16: 1–5. doi:https://doi.org/10.3390/ijerph20054215

55.

van Straten

Hill

Richards

D. A.

, et al. Stepped care treatment delivery for depression: a systematic review and meta-analysis. Psychol Med 2015; 45: 231–246.

56.

Menear

Ashcroft

Dahrouge

, et al. Person-centered care for common mental disorders in Ontario's primary care patient-centered medical homes: a qualitative study of provider perspectives. BMC Prim Care 2024; 25: 278.

57.

Canada

. Improving mental health outcomes for all people in Canada. 2025 [cited 2025 May 29].

58.

Linardon

Cuijpers

Carlbring

, et al. The efficacy of app-supported smartphone interventions for mental health problems: a meta-analysis of randomized controlled trials. World Psychiatry 2019; 18: 325–336.

59.

Naslund

Aschbrenner

Araya

, et al. Digital technology for treating and preventing mental disorders in low-income and middle-income countries: a narrative review of the literature. Lancet Psychiatry 2017; 4: 486–500.

60.

Agyapong

Milnes

McLoughlin

, et al. Perception of patients with alcohol use disorder and comorbid depression about the usefulness of supportive text messages. Technol Health Care 2013; 21: 31–39.

61.

Posner

, Brent D, Lucas C, et al. Columbia-suicide severity rating scale (C-SSRS). New York, NY: Columbia University Medical Center, 2008, p.10.