This is a three arm cluster randomized controlled trial where the unit of randomization is the immunization camp. Immunization camps are the sole provider of vaccinations in our study area and have catchment areas of several villages. The 96 immunization camps will be randomized evenly to one of three arms:

1. control arm; where near field communication (NFC) stickers will be placed on existing immunization booklets. These stickers can be scanned using a mobile phone with an installed application whereby information can be entered onto the phone and then scanned onto the sticker
2. pendant only; where the same NFC technology will be placed into a pendant that is worn by the infant, and
3. pendant as described above with voice reminders sent 15 days, 1 day prior, and the day of the scheduled immunization camp dates.

Routine immunization (RI) is an important tool to achieve the UN Millennium Development Goal (MDG) 4 of reducing child mortality by ensuring childhood protection from infectious disease. Recent 2013 estimates found that of approximately 6.3 million children died before their fifth birthday, with 51.8% (3.3 million) of these deaths attributed to infectious diseases (Liu et al., 2014). In regards to vaccine preventable diseases (VPDs), pneumonia (935,000 deaths) and diarrhea (578,000 deaths) were responsible for the most deaths. Measles, pertussis, tetanus, and meningitis, four other vaccine preventable diseases, were responsible for 362,000 deaths in children under five years old. (Liu et al., 2014)

Vaccination is one of the most cost-effective interventions for increasing childhood survival. (WHO, 2009). Global estimates find that current immunization programs save over 2.5 million lives a year (WHO, 2009). Despite the lifesaving potential of vaccines, in 2012 approximately 17% of children, or about 23 million infants, did not receive all the scheduled vaccines (CDC, 2013) and if they were immunized, they often received the vaccinations late (Clark & Sanderson, 2009).

In 1985, the Government of India launched the Universal Immunization Programme (UIP) to protect all infants (0-12 months) against six preventable diseases: tuberculosis, diphtheria, pertussis, tetanus, poliomyelitis, and measles. The goal of UIP was to ensure 100% coverage of all eligible children with one dose of BCG, three doses of DTP and OPV, and one dose of the measles vaccine. According to the Coverage Evaluation Survey (CES) 2009, a nationwide survey covering all States and Union Territories of India conducted during November 2009 to January 2010 by UNICEF, the national fully immunized coverage (FIC) against the six vaccines included in UIP in the age-group of 12-23 month old children was just 61% (CES, 2009). Coverage with third dose of diphtheria, tetanus and pertussis vaccine (DTP3) is a widely accepted global indicator for RI program performance (Lim SS, et al. 2008). India national estimates found DTP3 coverage was 72% (WHO India 2011), against a global target of 90%. The District Level Health Survey 3 (DLHS 3) findings suggest that the retention rate of childhood immunization was in maximum decline between the second round of DT and polio - 70 % and the third round - 56 % (Shekhar & Yadav, 2013).

The proposed study site, the Udaipur district of Rajasthan, has traditionally had lower immunization estimates than national levels while also performing poorly in other key developmental indicators: infant mortality rate (IMR) is 47 per 1000 live births (SRS Bulletin, 2014, GoI) and maternal mortality rate (MMR) is 244 per 100,000 women of reproductive age (MMR Bulletin, 2013).

To counter the long-standing deficiencies in immunization coverage and other health outcomes, a non-governmental organization (NGO) named Seva Mandir began operations in the area starting in 2004. Seva Mandir, has had a long standing relationship with over 700 villages in Rajasthan, with 105 villages participating in their immunization program. Immunization camps take place at regular, expected intervals throughout the month so caregivers are aware of when the scheduled camps will take place. This study area and its partners have conducted a seminal randomized controlled trial on non-monetary incentives to improve immunization coverage (Banerjee A, et al., 2010).

Although there has been a renewed global focus on improving immunization coverage levels (Alonso PL, et al., 2013; Berkley S, et al., 2013), Rajasthan district still performs poorly in routine immunization delivery and reporting. In efforts to improve the delivery and reporting of immunizations, with subsequent gains in proportions of infants fully vaccinated, we will conduct a cluster RCT that employs novel mHealth technologies to improve the reporting of immunization records and to increase the demand for immunizations through the use of voice message reminders.

One of these novel mHealth technologies is the use of a wearable electronic immunization record. This technology couples a mobile application (app) for health workers with a wearable digital pendant (necklace) for individuals that store their updated medical history. The pendant contains a read/writable NFC chip which allows health workers to digitally update a participant's vaccination records with the tap of a phone, and without the need to be connected to a central database. This technology can also be placed into a sticker that could be attached to an immunization booklet which would simplify record keeping. Importantly, this technology can be distinguished from other mHealth solutions as it does not require a centralized database of patients to look up patient specific records. Rather, the solution is decentralized allowing patients to wear their vaccine history on a necklace that can be accessed digitally in the last mile at the point of care. This work extends the use NFC trackers to simply identify patients and count disease occurrence (Marcus, et al., 2009) to medical record storage at the point of care. The potential for the necklace to play a symbolic role for the promotion of health benefits is an additional reason for the chosen form factor.

The pendant with NFC technology was developed by students at Yale as part of a class project and is called "Khushi Baby" (KB). Yale investigators visited the proposed study site to informally solicit local opinions on the proposed pendant and to receive feedback from Seva Mandir immunization teams to make the KB app user friendly. Importantly, this pendant is contextually relevant as many children in this study area wear similar talismans with the local thought being that they promote good health and can prevent coughs

Others have found that the color and shape of insecticide bed nets was a relevant factor in influencing people's willingness-to-buy the ITNs and thereby adopt the intervention in Southern Ethiopia (Gebresilassie & Mariam, 2011). The KB necklace, a unique feature of the system, also merits a deeper investigation into the characteristics that influence its adoption and whether it can generate a peer effect throughout the community which impacts vaccination adherence rates.

In addition to potentially being able to increase demand for vaccination, the NFC chip pendant could potentially improve the quantity and quality of immunization services available through a more efficient monitoring and outreach system. The current paper system is prone to loss of records due to wear and tear over the 9 to 12 month process to complete immunizations. Paper systems also have the disadvantages of being difficult to query, not individually specific as data is aggregated and transferred, and out-dated due to the time lag in data entry. The system to digitize vaccine records makes individually specific data available to remote monitors in real time so that they can better manage: vaccine stock, resource mobilization for camps, and camp reminder messages via automated voice reminders to child's household mobile phones. It also holds promise in improving workflow in the field. Simply scanning the chip quickly brings a child's records and their due vaccines for the visit without guesswork based on otherwise incomplete records. Finally, new information such as events of vaccine denial and reasons for vaccine denials can be automatically recorded to reveal greater insights on the true vaccine demand for a given village.

Data do not only have the potential to inform the service delivery NGO, but in this system, to also connect back with the community being served via voice call reminders. Many mHealth pilots on "push" based reminders systems have shown promise in improving health outcomes in developing countries (e.g. MedicMobile, M-Power, Frontline SMS, Mobile Alliance for Maternal Action); however further evidence must be collected to determine whether personalized voice calls can specifically reduce immunization drop out. An automated backend allows our push system to reach mothers/primary caregivers in villages before their baby is due for his/her next vaccine.

Our approach stands to be successful for several reasons:

• mHealth solutions have been successfully used to improve the quality and efficiency of data collection across a range of health outcomes, even in developing world settings.
• SMS reminders have shown some success to improve immunization coverage and timeliness in other developing areas (Bangure et al., 2015); we look to seek enhancement of this effect through personalized voice reminders in local language.
• Mobile phone ownership and access levels are at an all-time global high. Globally, the number of people owning mobile phones grew from 1 billion in 2000 to 6 billion in 2012 (World Bank, 2012). Reaching the last mile may be facilitated through remote reminder systems that alleviate the burden of local workers hired to go door-to-door, walking tens of miles the day before an immunization camp
• Monthly immunization camps are the predominant avenue for infants to be vaccinated. This immunization delivery system benefits the proposed study as study staff will not have to travel to many different clinics and/or households to track enrolled infant's immunization status.

Inclusion Criteria:

• Mother/ Primary Caregiver has an infant between the age of 42 days (minimum age for DTP1 administration per WHO guidelines) and 6 months of age; there is no age restriction for inclusion. In the very rare instance that a male is the primary caregiver, they will also be invited to participate in the study. Survey questions are not gender or age specific, but wil require the caregiver to answer questions about his/her children, socioeconomic and health status, and vaccine awareness.
• Self-identified resident within one of the villages associated with the immunization camp
• Willing to give informed consent (sign or thumbprint)

Exclusion Criteria:

• Infant has received at least 1 dose of DTP vaccine
• Caregiver and infant intends to move before the next 6 camps have been completed (in the next 6-9 months)