This will be a behavioural intervention with no investigational medicinal product.

The intervention will be a mobile messaging service that sends short messaging service (SMS) as texts or graphics to people with epilepsy to remind them to take their medication and to refill their prescription and educational messages to share important messages tackling stigma and tips to improve quality of life.

The investigators will also engage peripheral health facilities where people with epilepsy (PWE) participating in the study go for ASM refills, in collaboration with the respective county departments of health, to maintain adequate supply of anti-seizure medications through:

i. ongoing capacity building studies in Kilifi such as the mental health Gap Action Programme-Intervention Guide (mhGAP-IG) training which is empowering primary healthcare providers at peripheral health facilities to identify and manage epilepsy and other mental health disorders. ii. supporting healthcare providers at peripheral facilities through in person visits, if the COVID-19 situation, permits or by telephone or standard message reminders to restock their ASM supply.

The participants in the no-intervention group will receive "placebo" health messages not related to epilepsy such as use of bednets. The SMS reminders will be sent at a frequency that will be agreed upon during pre-study engagements with potential participants, whether daily, weekly, or monthly. The participants will be able to respond to these texts to report on their health status and any adverse events.

To evaluate whether SMS reminders improve adherence, we will use: i. Self-reporting adherence scales- the Morisky Medication Adherence Scale (MMAS-8) ii. Measurement of ASM plasma levels at 12 months from baseline.

Improved outcomes for epilepsy treatment depend on a continuous supply and daily adherence to anti-seizure medications (ASMs). In Kilifi County, the treatment gap which includes both the diagnostic and adherence gap, is greater than 70% and investigators have found interruption of supply of ASMs to peripheral clinics, distance from clinic and lack of availability of ASMs, to be barriers to adherence. In Nairobi County, factors such as environment hazards, lack of social amenities and correlates of poverty are preponderant in slums, but the prevalence of epilepsy has not been studied in such settings and consequently, the adherence gap remains unknown. Innovative mobile Health (mHealth) strategies including messages delivered by mobile phones have been used to ensure an adequate supply of drugs in health centres, and daily mobile messages have improved adherence to medication in Human Immunodeficiency Virus (HIV) programs, for example. Text messaging requires that the patient has access to a mobile phone and can understand the text message. Multimedia messaging, such as simple pictures, may improve understanding of the necessity to take medication, particularly in people who are illiterate, and we will explore this as an avenue to improve uptake.

The investigators plan to randomize 1200 people with epilepsy at each site, from a defined area in Kilifi and Nairobi County, Kenya. They will be divided into four groups of 300 each, to receive either text SMS, graphic SMS, both text and graphic or SMS on public health promotion not related to epilepsy e.g. use of bednets (for comparison). The investigators aim is to i) describe the perceptions and perspectives of people with epilepsy and their caregivers on the use of text and graphic SMS reminders to improve adherence to ASMs, ii) compare the effectiveness of text versus graphic messaging service in improving adherence in people with epilepsy and to engage the County Departments of Health through current ongoing training and capacity building studies to maintain supply of ASMs in peripheral clinics iii) identify the factors associated with improvement in adherence, improved QoL and reduction in stigma among people with epilepsy and family members and iv) conduct cost-effectiveness assessment for the roll out of the intervention. Besides medication-related messages, there will be other messages, from previous community-based feedback, selected to address stigmatization and improve quality of life. Blood-level monitoring and adherence questionnaires at baseline and during subsequent follow up visits will be used to assess as measures of medication adherence. If found useful, this intervention may be applicable for self-managing other chronic conditions in under-resourced settings.

Inclusion Criteria:

• Children or adults with a diagnosis of epilepsy ascertained by a clinician at the epilepsy clinic
• Taking anti-seizure medications at the time of the study
• Living within an area defined as the Kilifi Health Demographic Surveillance System, or the Nairobi Urban Health and Demographic Surveillance System at the time of the study or attending the epilepsy clinic in Kilifi or a KAWE-led clinic in Nairobi
• Able to give written informed consent or assent in addition to parental consent (if aged between 13 and 17 years old) to participate in the study either by themselves or in the presence of an independent witness

Exclusion Criteria:

• Have intellectual disability
• Do not have access to basic mobile phones
• Are currently enrolled in ongoing interventions aimed at improving their health care

Individual-level anonymized data will be shared with the Sponsor. Summary-level statistical analyses will be shared with wider stakeholder engagement and the medical community.

Information collected or generated during this study may be anonymised for use to support new research on epilepsy. Any future research using information from this study must first be approved by a local or national expert committee to make sure that the interests of participants and their communities are protected. Data will be managed by the data governance committee of KEMRI CGMRC.

From 2-5 years after the study completion.

Approved proposal and analysis plan and permission to access the data from the data governance committee.

The participants can tell if they are receiving adherence messages or general public health messages. The clinicians will interact with participants and may therefore get to know what kind of messages they receive. However, the investigators, who will conduct the analysis and the assessors will be blinded.