Use of SMS and Interactive Reminders to Improve Timely Immunization Coverage

Hypothesis:

The coverage rate of Penta-3, Measles-1 and Measles-2 in infants in Karachi will improve with:

1. Short text Messaging Service (SMS) reminders
2. Interactive reminders

Aims:

1. To determine if SMS reminders increases timely vaccine coverage rate in infants from Karachi.
2. To determine if Interactive reminders increases the timely vaccine coverage rate in infants from Karachi.

Background and Rationale:

The World Health Organization's (WHO) goal is to protect all people at risk against vaccine-preventable diseases. The availability of vaccines, however, does not mean that all eligible children are vaccinated. Despite several decades of global efforts in program implementation, vaccine-preventable diseases still contribute 25% of the 10 million deaths/year in under-5 children occurring worldwide. The Expanded Program on Immunization (EPI) does provide wide-access to vaccines in developing countries, but low vaccine uptake and delayed immunizations make infants and children more vulnerable to vaccine preventable diseases. The result of suboptimal immunization rates is the persistent existence of several vaccine preventable diseases. Many different strategies are being tested in these settings to increase immunization coverage, including out-reach services, health education, information dissemination, vaccination requirements for schools, conditional cash transfers and enhancing access to vaccination centers. This study proposes to investigate the use of SMS and interactive text reminders to improve timely immunization coverage. SMS reminders have the potential to increase immunization coverage and could prove to be a simple low cost intervention to improve timely immunization coverage.

SMS reminders Although there is limited data, SMS has been successfully used in various health programs including raising preventive and clinical treatment compliance, increasing HIV testing,4 reducing missed primary care and dental visits etc. Kaewkungwal's study designed a maternal and child care module based on SMS reminders and increased EPI on-time vaccinations to 44% from 35% (p<0.001). Vilella's study showed 47% (95% Confidence Interval (CI): 41-54%) of the participants in SMS reminder intervention arm completed third dose of hepatitis A + B vaccine as compared to 27% (95% CI: 23-32%) in control arm (RR 1.75; 95% CI: 1.41-2.17). SMS reminders have shown a potential simple cost-effective solution to remember the next vaccination date. With the growing use of mobile phones, SMS reminders can be quite useful in health communications. The cost of SMS in Pakistan is very low and use of SMS has broadened beyond the conventional use of peer-peer chat to include social networking, SMS banking, delivery tracking, advertising, business, disaster management, and identity verification etc. With such a low cost and varied use of SMS, it has a potential to be a frequent mode of health communications.

Interactive reminders Interactive reminders are a variant of SMS reminders where recipients are asked to respond back to SMS with a specific/non-specific response. Our experience of pilot with interactive reminders to tuberculosis patients to comply with drug adherence has shown positive acceptability. The system can be potentially used for compliance to vaccine schedule. The pilot study will be carried out only in 2 EPI centers at Korangi. The pilot will determine whether automated text messages produce increased immunization coverage in the reminder groups compared to the control group over the 8 weeks follow-up. The results of proposed pilot will provide the basis for a large scale-up study in urban and rural settings in Pakistan.

Study site \& Participants:

The study will be conducted in Korangi town, Karachi. Healthy infants 6-12 weeks of age visiting the Indus Hospital EPI center for first dose of pentavalent vaccine will be recruited for the study. Study will require parents/caregivers of infants to own a cell phone or have regular access to one. Parents/caregivers planning to go to a different center for next immunization will be excluded.

Sample size:

A sample of 522 participants would give 90% power (alpha= 0.05, 2-sided) to detect a difference of 15% over the baseline vaccine coverage rate of 70%. The final sample size of 656 infants (164 in each arm) will allow for a potential dropout rate of 20% during the follow-up period.

Randomization Enrolled subjects will be randomized to receive the SMS, interactive SMS or to serve as a control. The randomization sequence was created using SAS, version 9.3 (SAS Institute, Cary, NC), with a 1:1:1 allocation, using random block sizes of 3, 6, and 9.

Vaccines:

The vaccines will be administered as per routine EPI program in Pakistan which include Bacille Calmette-Guérin (BCG) at 0-6 weeks of age, pentavalent (DPT + Hepatitis B (HepB) + Hib) vaccine and oral polio vaccine at 6, 10 and 14 weeks of age, measles vaccine at 9 and 15 months of age.

Interventions:

The two interventions include SMS and interactive reminders. Infants will be randomized to either of the interventions or control groups. All participants will receive the routine EPI immunizations.

Study procedures \& Recruitment:

Children will be recruited at the EPI center when they arrived for the Penta-1 vaccine. Field workers will screen and obtain consent from the child's caregiver. Once the caregiver agrees to be a part of the study the infant will be randomised to one of the study arms and the name and age of the child will be recorded along with the immunization received on the visit. Each infant will be provided with a unique EPI number by the EPI staff and the same identifier (ID) will be used throughout the study to update and maintain the immunization record for all follow up visits. The child will be followed up to the measles 2 vaccine after which their immunization status will be marked as complete.

The parents of infants in SMS and interactive reminder groups will received 3-5 reminders for all scheduled follow-up immunization visits until the measles 2 vaccine. The infants in the control group will carry out routine follow up without any added intervention.

After the completion of follow-up duration a phone interview will be conducted to collect data on experiences of study participants with the SMS and interactive reminders. This will be a brief 3-5 minute interview to confirm if they received the messages and if there were any suggestions to improve the reminder system.

Statistical Analysis An intention to treat analysis will be conducted using STATA'13. Summary statistics and percentages will be calculated to establish coverage and timeliness rates within the three study arms. Fisher's Exact test with an alpha of 0.05 will be conducted to test differences in proportions for coverage and timelines

Additionally to determine factors associated with measles-2 completion at 24 months of age, a bivariate analysis will be conducted.Variables previously examined for association with measles 2 completion at 24 months via a bivariate analysis will then be selected for determining final multivariate model. All the variables will first be used in the model and the coefficients and the vif will be examined. Next, forward selection and step wise selection will be carried out in STATA to determine final model for a logistic regression.