Use of Short Messaging Services in Social Services to Assess the Severity of Depressive Symptoms: Implications for Task Shifting and Mobile Health Technology in Mental Health Services of a Resource-Poor Country

Background: Current access to and use of mental health services amongst refugees and migrants in Durban, South Africa remains at low levels.  The scarcity of mental health services has necessitated task shifting within the public healthcare sector. It has been advocated as a strategy with which to address the shortage of specialized healthcare in resource-poor settings. Mobile health technology may be an effective tool when used as a component of an overall task shifting strategy to potentially change when, where, and how mental health services are provided. This pilot study (1) examined the reliability of a short messaging service to assess the severity of depressive symptoms, and (2) assessed feedback from clients of the system.

Methods: Data was collected using a longitudinal study survey design. Adults receiving primary health, paralegal and soup kitchen services at a social service agency in Durban were the target study population. Inclusion criteria were: 21 years and older; asylum seekers/ refugees/migrants. Potential participants were excluded on the following criteria: those who did not speak English and those who were unable to provide informed consent. Consecutive persons meeting the above criteria were approached to enter the study. 153 and 135 participants completed the baseline and follow-up assessment respectively. Median time between assessments was 33 days.

A 16-item self-report South African version of Quick Inventory of Depressive Symptomatology (QIDS) was used to assess the severity of depressive symptoms. QIDS, a psychometrically reliable/valid instrument, was administered at either baseline or at the follow-up assessment utilizing face-to-face interviews or the short messaging service (SMS). With regard to screening for depressive symptoms in non-face to-face assessment, a member of the research team used a laptop computer with an SMS console program to send a series of QIDS questions to the participant’s mobile phone. Study participants were asked to answer each question by sending a response code by SMS. After receiving the response code, a subsequent QIDS question was sent to the study participant's mobile phone. The total score from QIDS indicated the following category: no depression (<5), mild (6-10), moderate (11-15), severe (16-20), and very severe (>21). The reliability between face-to-face and SMS-based depression symptoms assessment was evaluated based on a test and retest method using both weighted kappa coefficient and intraclass correlation coefficient two-way random effect model ICC (2,1).

Results: The test and retest analysis (n=138) revealed that weighted Kappa coefficient was 0.25 based on depression symptoms ordinal category. The ICC was 0.45 when depression symptoms were treated as continuous outcome.  Both of these findings indicate a fair level of agreement between face-to-face and SMS-based methods. The feedback from study participants indicate that delay in receiving SMS caused by mobile network (n=66) was a major challenge.

Implications: While there are benefits to using the SMS-based depression assessment tool, it should not replace face-to-face assessments provided by social services to identify mental health issues. The use of such mobile health technology should be used as a last resort, and only in the case of extreme scarcity of mental health services.