Background: Balochistan is the largest province of Pakistan by area, and the least developed. It suffers from several political, tribal and border conflicts. The distances to health facilities for the catchment population are long with limited accessibility. Immunization is one of the most cost-effective interventions to prevent deaths from vaccine preventable diseases (VPDs), especially in children. While Pakistan has an overall routine immunization coverage of 66 per cent for fully immunized children (FIC), coverage in the province of Balochistan is much lower at 29 per cent. This study aimed to assess the feasibility of introducing mHealth intervention using an artificial intelligence (AI) platform based on SMS (short-message service) and Interactive Voice Response (IVRs) to remind and persuade parents to get their children vaccinated. Methods: We employed a mixed study design using both quantitative and qualitative approaches. Baseline data were collected from 1,600 eligible mothers/parents within the catchment areas of 75 basic health units (BHUs) in Quetta (provincial capital of Balochistan province), and the automated platform was instituted with SMS and IVRs from EPI (Expanded Programme on Immunization) Quetta. Daily reminders and IVRs were sent to the cell numbers in the database/records. Responses were noted on the AI platform. After a period of about two months, an end line survey of 1,203 participants was undertaken with a loss of 397 (25%) participants due to seasonal migration to warmer cities in Sindh province. For the qualitative part, three key informant interviews (KII) were conducted; two KIIs with Lady Health Supervisors and one KII with the WHO staff responsible for routine immunization. There were also three focus group discussions (FGDs). Results: We found that the intervention was feasible as well as acceptable both at the community and programme management levels. The baseline indicators of immunization coverage improved significantly on end line survey (95% confidence interval, CI = 0.208-0.269; P<0.001). Conclusions: Our study demonstrates the potential for mHealth and AI to improve childhood immunization and addresses equity in the least developed areas of this country. The replication of the strategy in subnational immunization programmes could decrease morbidity and mortality due to VPDs.

Background: Routine immunization programs in Pakistan rely heavily on estimated population denominators, limiting accurate identification and follow-up of zero-dose and under-immunized children, particularly in high-risk urban settings such as Quetta, Balochistan. Methods: A quasi-experimental, pre–post implementation study was conducted from June 2024 to June 2025 across three low-performing union councils. A CHW-based household tracking tool was integrated within existing PEI–EPI systems. Data were derived from CHW Books, Rapid Convenience Assessments (RCA), and routine MIS reports. Descriptive statistical analysis was employed to assess trends in immunization coverage and program performance; no inferential statistical tests were applied due to the use of complete programmatic (census-based) data rather than sampled observations. Results: Antigen-specific coverage improved substantially, with Penta-3 coverage increasing from 22% to 95%, zero-dose conversion from 45% to 65%, and defaulter follow-up from 14% to 79%. All three union councils transitioned to Category 1 operational status. Administrative coverage exceeding 100%, reflecting population-level coverage and denominator correction rather than true population-level coverage. Conclusions: Integrating CHW-based tracking with dynamic denominator verification enhances routine immunization, microplanning, equity, and operational performance in high-risk LMIC urban settings.