Suicide attempts are a significant public health issue, with a high risk of relapse within 3–12 months following an initial attempt. Telemedicine-based follow-up interventions have emerged as a promising low-cost strategy for preventing reattempts, but the optimal duration and structure of these programs remain unclear. This study aimed to evaluate the effectiveness of a 12-month telemedicine-based follow-up intervention in reducing suicide reattempts compared to treatment as usual (TAU).
MethodsThis multicenter, prospective, non-randomized controlled study was conducted in two hospitals in the Basque Country, Spain. A total of 140 patients with a recent suicide attempt were included, with 70 receiving a structured 12-month telephone follow-up in addition to TAU, and 70 receiving TAU alone. The primary outcome was the proportion of patients who reattempted suicide within 12 months. Secondary outcomes included time to first reattempt and the impact of adherence to the intervention response.
ResultsPatients in the intervention group had a significantly lower rate of reattempts (21.4 % vs. 41.4 %, p = 0.011), fewer total reattempts (p = 0.031), and a longer time to first reattempt (p = 0.032) compared to the control group. Cox regression analysis showed a 54 % reduction in the risk of reattempts in the intervention group (HR=0.46, p = 0.001).
ConclusionA 12-month telemedicine-based follow-up significantly reduces suicide reattempts and delays relapse, supporting its integration into standard mental health care for individuals at high risk of suicide.
Suicide is a significant global health issue, accounting for approximately 700,000 deaths annually.1 Despite the presence of various risk factors associated with suicide, the ability to predict individual risk remains limited.2,3 A crucial risk factor for completed suicide is a history of previous suicide attempts, with estimates suggesting that for every completed suicide, there are around 20 to 30 attempts.1,2,4 Suicide attempts and reattempts pose substantial public health challenges,5 particularly as individuals are at a heightened risk of relapse within 3 to 12 months following an attempt.6,7 Consequently, preventive interventions aimed at reducing reattempts should prioritize this critical period.
While psychopharmacological treatments can be beneficial, they are often ineffective when used as the sole intervention for preventing suicide attempts and fatalities.8,9 This underscores the need for further research to identify the optimal duration and structure of preventive interventions in suicide prevention. Previous studies have demonstrated that brief contact interventions, such as telemedicine-based follow-up programs, serve as low-cost alternatives for suicide prevention, providing continuous support to at-risk patients. .10 According to the literature, brief telemedicine contacts include modalities such as telephone, internet, and video conferencing, and they promote accessibility.11 This approach has been shown to reduce the rate of reattempts12–17 and extend the time before relapse,12,18 while also improving adherence to mental health services.18
However, there is still no agreement on the ideal number of follow-up calls or how long these interventions should last, but a minimum follow-up period of 12 months has been suggested.19,20 In a previous study conducted by our team with a six-month follow-up, we observed an increase in the time until reattempt, as well as improved adherence, but no significant reduction in the number of reattempts.18 Conversely, another intervention program with a 12-month follow-up indicated a decreased risk of reattempts among participants.12
To address the ongoing debate surrounding follow-up duration and frequency of contact, we designed a study to evaluate the effectiveness of a telemedicine-based brief intervention for individuals who have attempted suicide.
MethodsStudy designPatients treated at the Emergency Services of Araba University Hospital and Donostia University Hospital were offered participation in a 12-month telephone follow-up following their recent suicide attempt. The primary objective of this project is to assess the efficacy of this telemedicine intervention in comparison to the standard care provided, known as Treatment As Usual (TAU).
This was a multicenter, prospective, non-randomized, controlled study conducted from 2018 to 2020 in two public hospitals in the Basque Country (Spain): Araba University Hospital (HUA) and Donostia University Hospital (HUD). The study aimed to compare the efficacy of a 12-month telemedicine-based follow-up intervention with Treatment as Usual (TAU) in preventing suicide reattempts. The study followed the Transparent Reporting of Evaluations with Nonrandomised Designs (TREND) guidelines. .21 The research protocol was approved by the Clinical Research Ethics Committee of Euskadi, and all participants provided written informed consent prior to enrolment.
Participants and proceduresThe study included adults aged 18 or older who had been treated for a suicide attempt, as defined by Silverman's criteria,22 in the Emergency Departments of the two hospitals: HUA and HUD. Participants were recruited at discharge from either Emergency or Hospitalization Services, and their allocation was based on the hospital where they received care. Patients treated at HUA were invited to join the intervention group, while those treated at HUD formed the control group. Control group participants were matched with those in the intervention group based on severity level, which was determined by the requirement for hospitalization. A non-randomized convenience sampling method was used for recruitment, and all participants were followed for 12 months after enrollment.
Participants in the intervention group received Treatment as Usual (TAU), consisting of standard psychiatric care, in addition to a structured telemedicine-based follow-up program. This program included five scheduled telephone calls conducted by trained nurses from the Basque Public Health System's Health Counseling Call Center. The protocol was applied uniformly across all participating centers after appropriate training for those responsible for its development.The calls, lasting 10–15 min each, were made at 1 week, 1 month, 3 months, 6 months, and 12 months post-discharge. The objectives of these calls were to assess current suicide risk, reinforce adherence to treatment and follow-up, provide psychoeducation, and conduct crisis intervention if an immediate suicide risk was identified. Data for the intervention group were collected during these telephone calls.
The control group received only TAU, which was delivered by mental health professionals and individualized based on clinical judgment. Unlike the intervention group, the control group did not receive a structured telephone follow-up. Data for the control group were extracted from electronic medical records.
OutcomesThe primary outcome was the percentage of participants who reattempted suicide and the average number of reattempts during the 12-month follow-up period. Suicide reattempts were identified through a review of electronic medical records from both participating hospitals, which are the sole psychiatric emergency care providers in the region.
Secondary outcomes included the time from enrollment to the first suicide reattempt (survival analysis) and the impact of adherence to the intervention on the time to the first suicide reattempt. Adherence to the intervention was categorized as low (1 call), moderate (2–3 calls), or high (4–5 calls).
Data on suicide reattempts were gathered from emergency records of both HUA and HUD. Adherence to the telephone follow-up intervention was recorded through a call log developed specifically for the study.
Sample sizeNo formal sample size calculation was performed. The sample size was determined by the number of eligible patients treated at the two hospitals during the recruitment period. This decision was driven by practical constraints, and the resulting sample size may have limited the study's statistical power to detect differences between the groups.
Statistical analysisDescriptive statistics were used to summarize sociodemographic, clinical, and adherence data. Group differences were evaluated using non-parametric tests for continuous variables and chi-square tests for categorical variables. Logistic regression models, adjusted for age and sex, were used to assess the effectiveness of the intervention in reducing the percentage of suicide reattempts within 12 months.
Kaplan-Meier survival curves and Cox proportional hazards regression models were employed to estimate and compare the time to the first suicide reattempt between the intervention and control groups. Covariates in the Cox regression models included age, sex, and hospitalization at the time of the index suicide attempt. Due to the high proportion of missing data regarding the method of suicide attempt in the control group, this variable was not included in either the logistic or Cox regression models. A significance level of 0.05 was used for all analyses. Data were analyzed using IBM SPSS Statistics (version 19, Armonk, NY, USA).
ResultsDescription of the sampleA total of 140 patients were included in the study, with participants assigned to either the intervention group (n = 70) or the treatment as usual group (n = 70). The sociodemographic and clinical characteristics of the sample are presented in Table 1. The mean age of the participants was 47.6 ± 13.1 years, and 65.7 % were women.
Clinical background and baseline data of study participants.
When comparing the intervention and control groups, no significant differences were observed in terms of age, sex, or the need for hospital admission during the index episode. However, notable differences were identified in the method of suicide. Patients in the intervention group were more likely to use drug poisoning compared to those in the control group (78.6 % vs. 52.9 %, p = 0.003). It is important to note that the control group had a high proportion of missing data regarding the method used (47.1 % vs. 0 %, p < 0.001).
Reattempts during the follow-up periodForty-four patients (31.4 %) showed a suicide reattempt during the 12-months follow-up. Patients who received the intervention showed lower frequencies of reattempt during follow-up (21.4 % vs 41.4 %; p = 0.011), lower number of reattempts (0.7 [SD 1.1] vs 0.4 [SD 0.8]; p = 0.031) and higher mean time until first reattempt (286.7 ± 151.3 vs 338.3 ± 128.3; p = 0.032) than the control group (Table 2). No patients died by suicide neither in the intervention nor in the control group.
Frequency and Number of Suicide Reattempts and Time to First Reattempt in Intervention and Control Groups.
Fig. 1 shows the survival curve. When comparing the survival rates between groups, those who received the intervention showed lower proportion of suicidal reattempts than the control group at 6 months with no statistical significance (17.3 % vs 28.6 %; p = 0.111) and reached the significance at 12 moths (21.4 % vs 41.4 %; p = 0.010).
In the survival analysis, patients who received the intervention showed a suicide reattempt later than the control group (p = 0.013 by log-rank test). Furthermore, the Cox regression model for time to reattempt adjusted for age, sex and hospital admission at index episode showed that the intervention group was a protective factor for suicidal reattempt (HR=0.46; 95 %CI=0.25 to 0.87; p = 0.001). In other words, we found a 54 % decrease in the risk of suicidal reattempt in the intervention group compared to the control. (Fig. 1).
Adherence and efficacy of the interventionTo further explore the intervention group, the impact of adherence on survival to suicide reattempts was analyzed. Seventeen patients (24.3 %) showed low adherence, 18 patients (25.7 %) showed moderate adherence, and 35 patients (50 %) showed excellent adherence. In the survival analysis for reattempts, no differences were found among the three adherence groups (p = 0.884, by log-rank test).
DiscussionThis non-randomized controlled study evaluated the impact of a 12-month telemedicine-based follow-up intervention on suicidal relapse among patients after a suicide attempt. The findings demonstrate that participants in the intervention group had a significantly lower rate of reattempts and a longer average time to their first reattempt compared to the control group. Furthermore, survival until the first reattempt was higher in the intervention group, indicating a 54 % reduction in the risk of suicide reattempt. Interestingly, adherence to the intervention did not significantly influence survival outcomes. These results contribute to the growing evidence supporting the integration of structured follow-up interventions into standard mental health care protocols to reduce suicide reattempts and improve patient outcomes. Building and developing the therapeutic alliance during the different phases of the therapeutic relationship can contribute to better clinical outcome.23
In contrast to our previous study,18 which found no significant reduction in reattempts with a shorter follow-up period this 12-month intervention demonstrated clear benefits in reducing both the frequency and timing of reattempts. These findings are consistent with evidence from a recent systematic review, which suggests that extending brief contact interventions beyond 12 months provides greater reductions in suicide reattempts (RR = 0.46; 95 % CI 0.10–0.82) compared to interventions lasting 12 months or less (RR = 0.67; 95 % CI 0.54–0.80).20 Although the extended duration likely contributed to the intervention's effectiveness, other factors may also have influenced the results. Compared to our previous 6-month study,18 the current sample included older patients, a higher proportion of hospitalizations after the index suicide attempt, and a lower use of drug poisoning as a method—all of which may indicate greater clinical severity. These differences could partially explain the improved outcomes. However, due to the non-randomized design and limited clinical data, it is not possible to isolate the effect of duration from other variables.
In our study, adherence to the telemedicine protocol was moderate to high, with half of the sample completing 4 or 5 calls over the 12-month period. The number of calls during follow-up is a key factor influencing both adherence and effectiveness. Previous studies employing similar telephone follow-up designs have used between 117 and 1123 calls, with 6 calls being the most common approach.11,24 Our protocol consisted of 5 calls distributed over 12 months, and although adherence did not significantly affect survival outcomes, this unexpected finding may be due to several factors. Minimal engagement early after discharge may have been sufficient to offer some benefit. Also, small group sizes could have limited statistical power, and our adherence measure—based only on number of calls—may not fully capture patient engagement. Determining the optimal number of calls to maximize both effectiveness and adherence remains a critical area for future research. There is a growing interest in this topic,25,26 but the scarcity of robust comparative studies underscores the need for further research and strategies to apply these findings in clinical practice.
These findings are consistent with previous studies endorsing telemedicine as an effective strategy to enhance treatment adherence, reduce suicide reattempts,19 and mitigate suicidal ideation.27 Telemedicine has proven particularly useful for improving access to care for high-risk populations and individuals in geographically isolated areas.28,29 Telehealth has become more interactive, affordable, and widely available for healthcare professionals addressing chronic conditions.30 This approach has also been validated during the COVID-19 pandemic, when in-person healthcare access was significantly restricted.25 Telephone-based interventions, in particular, have demonstrated a 5 % absolute reduction in reattempt rates and a 20 % relative risk reduction over 12 months in patients treated for suicide attempts in emergency settings.23Other studies on telecare with people at risk of suicide have found that follow-up interventions, and specifically telephone follow-up programs, have provided promising results and could be considered a useful new tool in the care of people at risk.10 Other studies on the impact of non-pharmacological preventive strategies addressing risk management focused on a telehealth monitoring protocol. When participants reported suicidal plans or intentions, the nurse contacted a physician, who provided a rapid response. Furthermore, staff immediately contacted participants when they responded affirmatively to a question about suicidal behavior.31 Support tools linked to the use of artificial intelligence also contribute to decision-making and could also allow for a more accurate assessment of suicidal risk in situations where the patient denies suicidal ideation.32 On the other hand, the collection of information through smartphones using explicit (active) and implicit (passive) methods to construct a digital phenotype, which should be integrated with genetic and epigenetic data to develop tailored therapeutic and preventive approach.33 A sample of 249 adults completed a baseline assessment of their suicidal ideation characteristics during psychiatric hospitalization, five daily ecological momentary assessments (EMAs) for 21 days after discharge, and follow-up assessments of suicide-related outcomes at 3 weeks and 6 months follow-up. The results implicated low controllability of suicidal ideation as a potentially useful marker of short-term risk for suicide attempts. This finding is of high clinical relevance, particularly because perceived controllability is a less frequently assessed characteristic of suicidal ideation in clinical settings than other characteristics of ideation commonly thought to indicate the severity of suicidal ideation (i.e., frequency, duration, intensity), and to date, almost no useful indicators of short-term suicide attempt risk have been identified that can be assessed through a single item.34
Our control group showed a suicide reattempt rate of 41.4 %, which is notably higher than the 20 % reported in a recent meta-analysis.7 Variability in reattempt rates across studies has been attributed to factors such as female sex, self-poisoning or self-cutting methods, and mental health disorders. Although we did not collect data on participants’ psychiatric diagnoses, our sample primarily consisted of women, with drug poisoning being the most common method—factors that may help explain the elevated reattempt rate in our study.
Suicide prevention remains a challenging priority, particularly among individuals with a history of suicide attempts. Hospitals provide a unique opportunity to implement preventive interventions, as many patients seek emergency care following suicide attempts and can be enrolled in post-discharge follow-up programs. In our study, two-thirds of participants were recruited following hospital admission, reflecting the severity of their attempts—a finding consistent with prior research.35 Additionally, the sample predominantly consisted of women, a common pattern observed in suicide attempt studies, though this is not typically considered a significant factor.
Limitations and strengthsThis study has several limitations. First, the non-randomized design and reliance on convenience sampling may have introduced selection bias, potentially affecting the generalizability of the findings. Second, the relatively small sample size limits the ability to draw broader conclusions. Third, important variables such as prior psychiatric diagnoses, psychometric assessments, and specific characteristics of suicide attempts were not collected, which could have provided deeper insights. In particular, the absence of diagnostic information limits our ability to adjust for key clinical predictors of suicide risk. Fourth, data for the control group were extracted from electronic medical records, limiting the level of detail available—for example, the method of suicide was not recorded for 47.1 % of the patients in the control group. This imbalance may have influenced group comparability, and residual confounding related to suicide method cannot be ruled out. Fifth, while age and sex were balanced between groups and adjusted for in the analyses, the lack of matching on these variables may still introduce some residual confounding. Sixth, the study focused exclusively on adults, excluding adolescents and children, a particularly vulnerable population that warrants further research. Seventh, we did not assess barriers such as digital literacy, access to technology, or motivation, which may have influenced adherence and limited the intervention's accessibility. Future studies should explore these factors to improve reach and equity. Finally, it has not been possible to determine the impact of users' socioeconomic status on access to the devices required for telemedicine.
Despite these limitations, the study has notable strengths. The extended 12-month follow-up period allowed for a thorough evaluation of the intervention's long-term impact, while the use of uniformly trained nursing professionals ensured consistent follow-up assessments. Adherence, often underreported in similar studies, was systematically measured, providing valuable insights into implementation. Additionally, the use of an integrated clinical history system at the region's sole hospitals providing psychiatric emergency services, covering a population of 350,000 residents, ensured the inclusion of a representative sample, thereby enhancing the validity and reliability of the findings.
ConclusionsThis study demonstrates that a 12-month telemedicine-based follow-up intervention significantly reduces both the frequency of suicide reattempts and the time to reattempt when compared to standard care alone. The findings emphasize the importance of sustained, structured contact in supporting high-risk patients during the critical post-discharge period, showing a 54 % reduction in the risk of reattempts in the intervention group. These results highlight the potential of telemedicine as an effective and cost-efficient tool for suicide prevention, supporting its integration into standard mental health care protocols, particularly for vulnerable populations at high risk of relapse. Its applicability is comprehensive, with the potential to transfer this knowledge to routine healthcare practice.
Ethical considerationsThe research protocol was approved by the Clinical Research Ethics Committee of Euskadi, and all participants provided written informed consent prior to enrolment.
FundingThis study had been funded by Instituto de Salud Carlos III through the projects PI18/01055 PI15/00789 (Co-funded by European Regional Development Fund/European Social Fund. Investing in your future) PI21/00713 and PI23/01367;the Basque Foundation for Health Innovation and Research (BIOEF); Networking Center for Biomedical Research in Mental Health (CIBERSAM) and the University of the Basque Country (GIC18/107).
Dr. Gonzalez-Pinto has received grants and served as consultant, advisor or CME speaker for the following entities: Janssen-Cilag, Lundbeck, Otsuka, Alter, Angelini, Novartis, Rovi, Takeda, the Spanish Ministry of Science and Innovation (CIBERSAM), the Ministry of Science (Carlos III Institute), the Basque Government, and the European Framework Program of Research. The other authors declare no conflicts of interest.




