Evaluating the health impact of a public-private partnership

Abstract

The purpose of this article is to describe the RotaTeq^® Nicaragua Partnership and the evaluation of the public health impact of the vaccine conducted by the partners, including the creation of a rotavirus surveillance program and a vaccine effectiveness assessment. The three main objectives of the partnership were to demonstrate that a new rotavirus vaccine could (1) be introduced rapidly in a developing country, (2) be successfully integrated into the existing vaccine delivery infrastructure, and (3) have a significant and measurable public health impact at the end of the 3-y program. The vaccine impact assessment required collaboration among partners with different areas of expertise, including the Nicaraguan Ministry of Health, Merck, local hospitals, government health clinics, laboratories, and a Technical Advisory Group. Through the partnership, RotaTeq^® became available in a GAVI-eligible developing country, Nicaragua, in the same year it was approved in the United States. Vaccine coverage rapidly reached over > 90% of eligible Nicaraguan children. The impact assessment evaluated over 10,000 subjects and leveraged and enhanced the existing diarrheal surveillance infrastructure, ultimately providing the scientific community with some of the first real-world rotavirus vaccine effectiveness data from a developing country. The successful public-private partnership (PPP) was internationally recognized as a model for the rapid adoption of a new vaccine in a developing world setting. The model could be adapted to benefit other PPPs interested in demonstrating the impact of their own programs.

Keywords: :

• collaboration
• partnership
• provision
• RotaTeq
• rotavirus vaccine

Introduction

Public-private partnerships (PPP's) are an important mechanism to address public health challenges seen in the developing world. Merck, a United States-based pharmaceutical company, has had a long-standing commitment to improve access to new medicines and vaccines in the developing world. One example of how Merck has fulfilled this commitment is the Mectizan^® Donation Program. Mectizan (ivermectin, Merck and Co., Inc.) is indicated for the treatment of onchocerciasis (river blindness). The donation program for this drug was established in 1987 as a collaboration among the World Health Organization, the World Bank, Ministries of Health, non-governmental development organizations and local communities, and Merck. Its goal was to provide Mectizan to treat onchocerciasis for as long as it was needed, wherever it was needed. Through the program, Mectizan has been provided to patients in 33 endemic countries for over 20 y. In 1998, the program was expanded to include the elimination of lymphatic filariasis in African countries where the two diseases co-exist. This ongoing program has been identified as one of the most successful and sustained PPPs in public health.¹

In collaboration with the Ministry of Health (MoH) of Nicaragua, Merck launched another PPP, the RotaTeq^® Nicaragua Partnership, on October 26, 2006. The core component of the partnership was the provision of a new vaccine, rotavirus vaccine, which recently had been shown to be highly efficacious in preventing severe rotavirus disease in industrialized countries.² Prior to vaccine introduction, rotavirus gastroenteritis (RGE) was a major global public health burden causing 527,000 deaths and 2.4 million hospitalizations per year in children less than 5 y of age.³

RGE is the third most common cause of death in developing countries, including countries such as Nicaragua, causing up to 17% of deaths in children less than 5 y of age where access to appropriate health care is often not optimal.⁴ The 2006 licensure of two new live, oral attenuated vaccines, RotaTeq^® (RV5) (Merck and Co., Inc., Whitehouse Station, NJ) and Rotarix^® (GlaxoSmithKline, Rixensart, Belgium) provided hope for the global reduction of rotavirus-associated deaths and hospitalizations.

The RotaTeq^® Nicaragua Partnership provided RV5 free of charge for three years to the MoH of Nicaragua, a country with high mortality and morbidity associated with rotavirus.³ Merck partnered with the MoH to develop, design, and conduct this partnership. The 3 main objectives of the partnership were to demonstrate that a new rotavirus vaccine could 1) be introduced rapidly in a developing country, 2) be successfully integrated into the existing vaccine delivery infrastructure, and 3) have a significant and measurable public health impact at the end of the 3-y program.⁵ Although clinical trial data supporting the efficacy of rotavirus vaccines in developed countries were widely available at the time of product licensure, there were no data demonstrating the effectiveness or efficacy of any rotavirus vaccine in a real world, low income, developing country setting. Such data were critical to the Strategic Advisory Group of Experts (SAGE) of the World Health Organization (WHO) in determining whether global recommendations for the vaccine should be made.^6,7 This was of particular importance in view of the fact that other oral vaccines had been found to be less efficacious in developing world settings.⁸ Recognizing that the vaccine donation would cover only the first three years of the program, the partners also worked closely to address the long-term sustainability of the vaccine program.

The purpose of this article is to highlight the key elements, successes, and challenges of this PPP, with emphasis on the critical elements of the vaccine impact assessment and lessons learned. This partnership demonstrated the feasibility of rapidly making a new vaccine available in a developing country, an objective that had taken decades to accomplish in the past.⁹ Thus, the project became both a regional and global model for the successful early introduction and assessment of a new vaccine.

Results

Vaccine Delivery and Coverage

Merck provided sufficient vaccine to allow the MoH to vaccinate every eligible infant born in Nicaragua during the three-year partnership (2006–2009). Over 1.3 million doses of RV5 were shipped from Merck to Nicaragua using standard procedures to maintain the cold chain. Routine vaccination with RV5 was initiated by the MoH in October 2006. As a result of Nicaragua’s efficient vaccine delivery system and the dedication of those administering the vaccine, RV5 coverage reached > 90% by 2009.⁵

Vaccine Impact

The impact assessment activities, including disease surveillance and evaluation of vaccine effectiveness, were initiated in February 2007. From then until October 2009, over 10,000 children were enrolled into the study as surveillance subjects, possible rotavirus cases, or controls. Of the enrolled subjects, 6,174 were surveillance subjects, 2,015 were community controls, and 1,977 were hospital controls. The stool sample collection rate among the subjects in the surveillance group was 99% (n = 6108). Overall, study enrollment procedures and follow-up were very efficient; only 1% of subjects were discontinued or excluded from the study.

Three hundred (300) rotavirus cases were identified from the surveillance subjects and were matched with 851 community controls and 792 hospital controls. Vaccine effectiveness against severe disease was 87% (95% CI: 0.93, 0.74) in community controls, 64% (95% CI: 0.78, 0.44) in hospital controls, and 76% (95% CI: 0.84, 0.63) for the combined groups.⁵

Discussion

The RotaTeq^® Nicaragua Partnership was highly successful in both its execution and assessment of public health impact. Specific successes of the program included:

• Quickly introducing RV5 into the routine vaccination schedule in Nicaragua. Historically, new vaccines have taken as many as 15 y or more to reach populations in developing countries following their introduction in industrialized countries.⁹ Through the partnership, RotaTeq^® became available in a GAVI-eligible developing country, Nicaragua, in the same year it was approved in the United States. RV5 was approved in the US on February 3, 2006 and was available in Nicaragua by October of the same year. By working together, the partnership achieved an unprecedented acceleration of access to an important vaccine resulting in both early and substantial reductions in rotavirus disease.

• Working with the existing infrastructure whenever possible. By taking advantage of the existing diarrheal surveillance infrastructure, the partnership was able to rapidly implement a rotavirus-specific surveillance system. This enabled the first subject to be enrolled in the impact assessment evaluation in February of 2007, only five months after the vaccine was introduced in Nicaragua. Efficiency was gained by leveraging the experience and previous training of the local study staff. This expedited the identification and enrollment of subjects with AGE into the case-control effectiveness study.

• Extensive communication of the burden of rotavirus and prevention through vaccination. The partners created many effective means of educating parents and physicians about the morbidity and mortality associated with rotavirus and the benefits of vaccination. Multiple media (e.g., radio, television, videos, and billboards) were used to insure that this information was widely disseminated to the general public in Nicaragua.¹⁷ Extensive training of the local physicians and health care professionals was provided to ensure a good understanding of the safety and efficacy profile of the vaccine.

• High vaccination coverage with RV5. Integration of RV5 into the country’s existing and traditionally efficient routine vaccine program was an important element in achieving high vaccination coverage. With > 90% vaccine coverage, Nicaragua achieved one of the highest rotavirus vaccination rates in the world in less than three years.¹⁸ This high coverage rate was achieved as a result of several key activities. First, extensive education of parents about the burden of disease and the value of the vaccine was conducted. Many immunization clinics were established when the vaccine was first available and vaccination continued during national immunization weeks to promote rapid uptake and provide easy access. Extensive efforts were made by the local health providers to ensure that the vaccine reached families in hard to reach areas. In addition, health clinic workers were encouraged to administer the vaccine at the same time as other age-appropriate childhood vaccines so there were no missed opportunities for vaccination.

• High enrollment and stool sample collection rates in the assessment of vaccine impact. The assessment of vaccine impact was optimized by the ability to effectively enroll subjects in a short period of time and to obtain a high proportion (99%) of stool samples from the enrolled subjects, allowing for the rapid identification of possible rotavirus cases. This high collection rate can, in part, be attributed to the involvement of the MoH in the selection of sites and staff. This demonstration of commitment by the government provided motivation for the local health administrators and study staff. In addition, the involvement of physicians, hospital staff, and community workers affiliated with the MoH resulted in excellent rapport with subjects and subsequent high study compliance. Finally, no blood samples were collected and study procedures were simplified in order to enhance successful recruitment.

• International recognition and support of the partnership. General support and interest in the design and outcome of the project was garnered from the scientific and public health communities through presentations given from 2006–2009 at local and international scientific meetings.^19-21 The partnership also was recognized for its innovative commitment to public health by the Clinton Global Initiative in 2008, resulting in the publication of a best practices case study, and by the Bill and Melinda Gates Foundation in 2009 as a Living Proof Project highlighting progress in global health.^22,23

• Research capacity building. Hospital surveillance sites were provided with laboratory equipment and technical assistance in performing the assays used to detect rotavirus. All sites were trained on the conduct of a case-control study, sample collection methods, Good Laboratory Practices, and standard international shipping procedures (including International Air Transport Association training). This training helped to accomplish the immediate study objectives, but also provided training that study staff could use in future research and/or surveillance efforts after this study was completed.

• The fulfillment of the project's commitments to key stakeholders including the global scientific community, Nicaragua, the MoH, and Merck. The assessment of the vaccine's impact provided real world vaccine effectiveness data for RV5 from a developing country. This information was critical for both the scientific community as well as policy makers involved with investments in vaccination and public health programs. These data, along with data from clinical trials in Africa and Asia, supported the 2009 change made by the SAGE of WHO to move from a regional recommendation for the inclusion of rotavirus vaccines in national immunization programs in the Americas and Europe only to a global vaccine recommendation.^6,7 In turn, the revised recommendation enabled GAVI to widen its potential rotavirus support to Africa and Asia, a key step toward lowering childhood mortality and reaching the Millennium Development Goal.²⁴

• The transition to a sustainable vaccine program. In October 2009, Nicaragua transitioned to a self-sustaining program with co-financing for RV5 provided by GAVI, which allowed for long-term availability of RV5 in Nicaragua. The MoH and Merck worked very closely to insure sustainability of the program from the onset of the PPP. Merck pursued WHO prequalification of the vaccine, a prerequisite for GAVI supporting any vaccine and United Nations agencies (e.g., PAHO, UNICEF) procurement, shortly after the donation program began. Prequalification was obtained in 2008. During the same time period, the MoH began the process of applying for GAVI support, which was approved in 2008. The first shipment of RV5 to Nicaragua through GAVI occurred in March 2010, ensuring uninterrupted vaccine supply. Since that time, Nicaragua has continued to vaccinate all children against rotavirus with RV5 purchased with GAVI support. Per the co-financing terms of the GAVI agreement, Nicaragua began paying 15 cents for each dose received.

• The opportunity for other research scientists to evaluate, publish, and present data on vaccine impact. Since this study was started, other evaluations of the reduction in disease burden and the effectiveness of the vaccine in Nicaragua have been conducted and published by investigators in Nicaragua, CDC, and PAHO.^3,25,26 For example, although the partnership did not directly measure the impact of rotavirus vaccination on mortality, other groups have used RV5 effectiveness data from Nicaragua and estimated that the effectiveness of RV5 on rotavirus mortality in children younger than 5 y of age was 74% (95% CI, 35–90).²⁵

This collaboration confirms that it is possible, through a PPP, to conduct a successful vaccine delivery program and impact assessment for a new vaccine introduction. Through this experience, a few valuable lessons were learned, including:

• The need to gather pre-vaccine baseline surveillance data. Because of the project timelines, there was no opportunity to assess the burden of rotavirus in Nicaragua prior to vaccine introduction. Although the Nicaraguan MoH routinely conducted diarrheal disease surveillance, little rotavirus-specific burden of disease data were available at the time the vaccine was introduced. Disease specific data could have allowed for the comparison of pre-vaccine rotavirus prevalence rates to rotavirus rates following vaccine introduction, thus demonstrating the broader impact of the vaccine. For some vaccine policy makers, this type of data (pre vs. post) may have been more intuitively informative than vaccine effectiveness estimates based on the case-control method used in this project.

• Consideration of whether interim reporting of results is feasible. Since the partnership was designed as a 3-y program, an interim analysis was not planned in the statistical analysis. However, interim impact assessment activities could have provided real-time surveillance and vaccine effectiveness results much earlier to the MoH, public health officials, and other developing world countries interested in rotavirus vaccine introduction. The availability and application of appropriate statistical methods associated with interim effectiveness analyses should be carefully weighed in future studies.

Methods

Country Selection

There were several factors that led to the decision to select Nicaragua as a partner. Nicaragua is a country in Central America with a birth cohort of approximately 150,000.³ The country ranks among the poorest in Latin America. In 2005, an outbreak of RGE in Nicaragua led to 47,470 consultations and 52 childhood deaths.¹⁰ A subsequent case-control study suggested that half of the severe RGE cases and more than two-thirds of diarrheal deaths could have been prevented with a rotavirus vaccine. These data highlighted the significant burden of rotavirus for the public policy makers in Nicaragua and demonstrated the potential impact of a national rotavirus vaccination program. The outbreak emphasized the need for future rotavirus gastroenteritis surveillance, and an existing MoH diarrheal disease surveillance system provided the opportunity for the establishment of a multi-site rotavirus surveillance network. Thus, the governmental awareness of rotavirus disease burden combined with existing surveillance networks created was one ideal characteristic for establishing a partnership.

Nicaragua also had a strong track record for the implementation of nationwide immunization programs with high vaccine uptake for vaccine-preventable diseases such as measles, polio, tuberculosis, diphtheria, pertussis and tetanus.¹¹ Through Nicaragua’s Expanded Programme on Immunization (EPI), vaccination coverage rates for routine childhood vaccines ranged from 79–92%, equal to or better than the rates in some developed countries.¹² Since the three-dose RV5 was developed to be administered concomitantly and on the same 2, 4, 6-mo schedule as most EPI vaccines, the previous success of the Nicaraguan program suggested that RV5 could be smoothly integrated into the existing EPI schedule.

Finally, the GAVI Alliance (formerly known as the Global Alliance for Vaccination and Immunization), whose mission is to accelerate access to vaccines in developing countries through the purchase of new vaccines, expanded its funding support in 2006 to include rotavirus and pneumococcal vaccines. Nicaragua was identified as one of the 72 countries eligible for this funding.¹³ The strong awareness of rotavirus disease, demonstrated commitment to vaccination, and GAVI eligibility meant that Nicaragua would be an optimal partner for the introduction of a new rotavirus vaccine.

Vaccine

RV5 is an oral, liquid vaccine that provides protection against five serotypes of rotavirus, including serotypes G1, G2, G3, G4 and P1. RV5 is given in 3 doses at 2 mo, 4 mo, and 6 mo of age as part of the routine childhood immunization schedule in the United States and in Nicaragua. RV5 was approved in the United States, Nicaragua, and many other countries in 2006.

Building Partnerships

Merck first approached the Nicaraguan MoH in 2005 to propose using RV5 to help address its significant burden of rotavirus disease. After numerous discussions between the MoH and Merck, the RotaTeq^® Nicaragua Partnership was announced at the Clinton Global Initiative in October, 2006.¹⁴ The partnership included the involvement of several non-governmental organizations (NGO) and multi-lateral organizations, including the Pan American Health Organization (PAHO), the US Centers for Disease Control and Prevention (CDC), Program for Appropriate Technology in Health (PATH), and NicaSalud, all of whom became involved with the program. PAHO, CDC and PATH provided technical assistance to the MoH with respect to vaccine implementation, delivery, and additional impact assessment activities, while the local NGOs helped to facilitate collaboration with local health teams and helped the partnership identify and reach underserved populations.

An important component of the partnership was the inclusion of a vaccine impact assessment to understand the public health benefit of RV5 introduction. The initiation and conduct of the impact assessment required a broad alliance of partners with different areas of expertise. Merck partnered with the MoH to establish a study infrastructure consisting of six local hospitals; SILAIS (Sistemas de Atención Integral de Salud), a district-level network of MoH access health clinics in the same districts as the hospitals; and a shipping and coordinating center located in the MoH Managua location. In addition, the project included a central laboratory in the United States and an external and independent Technical Advisory Group (TAG). The TAG consisted of 14 subject matter experts in rotavirus disease, infectious diseases, and epidemiological methods. Each group had defined roles and expectations that were identified and communicated at the project's inception. The MoH and TAG participated in the design and review of the surveillance and impact assessment protocol and statistical analysis plan, while the sites, laboratories and SILAIS were responsible for the conduct of the study protocol. These partners each added their own perspective, insight, and knowledge. Collaboration between these groups was critical in carrying out the assessment activities that were designed to understand the public health impact of the partnership.

Vaccine Impact Assessment

Assessment of the impact of the vaccine in Nicaragua consisted of two components: 1) an observational hospital-based prospective surveillance program to estimate the proportion of AGE that was attributable to rotavirus among children under 5 y of age and; 2) a case-control study to estimate the vaccine effectiveness of RV5 against severe rotavirus AGE among children who completed the 3-dose regimen. The details of the case-control study have been described previously.⁵

Rotavirus surveillance was initiated at the six hospital sites located in the Western part of Nicaragua. Each site had five to eight staff members including the principal investigator, at least one sub-investigator, study coordinators, and laboratory technicians. Staff responsibilities included recruitment, administration of informed consents, confirmation of vaccination status, completion of case report forms, evaluation of any serious adverse events, data cleaning, stool collection, laboratory testing, and sample shipment. Stool samples were first tested locally at several laboratories in Nicaragua using a rapid test and then a separate aliquot was tested in a laboratory in the United States for confirmation via immunoassay techniques.

Once the rotavirus surveillance program was operational, the case-control vaccine effectiveness study was initiated. The study was conducted from 2007 through 2009. The case-control design was selected by Merck and the MoH because it was a widely accepted method, allowed for comparison to similar studies, and could be performed relatively quickly and inexpensively compared with other methodologies.^15,16 Subjects in the surveillance program provided the rotavirus cases needed for the vaccine effectiveness study. The effectiveness study compared the vaccination status of children with rotavirus disease (cases) and an age-matched group of children without disease (controls) in order to make statistical statements about how well the vaccine prevented disease.¹⁵ The primary objective of the study was to evaluate the effectiveness of the vaccine in reducing the risk for severe, wild-type RGE resulting in hospitalizations and emergency department visits among children who had completed the recommended 3-dose regimen as part of the routine national vaccine program.⁵

Conclusion

Nicaragua was the first GAVI eligible country to introduce rotavirus vaccine into its national EPI schedule. The RotaTeq^® Nicaragua Partnership played a key role in the introduction of RV5, confirming that a PPP could successfully implement the rapid and widespread uptake of a novel vaccine in a developing country. This resulted in a dramatic reduction in rotavirus disease burden in a country with significant rotavirus-associated morbidity and mortality prior to vaccine introduction. The partnership serves as a comprehensive model of a PPP that allowed for the provision of vaccine, the establishment of surveillance and vaccine effectiveness protocols, the delivery of some of the first data on the performance of a new vaccine in a developed country, and the successful transition from a donation program to a sustainable program for vaccine accessibility and funding.

The addition of a scientifically rigorous disease impact assessment, as was done in this partnership, provided valuable data for decision makers about the impact of the vaccine when used routinely in a developing country. The data are particularly important in that they were obtained by administering a new vaccine concomitantly with other routine EPI vaccines, and all vaccines were delivered through Nicaragua's already existing infrastructure for vaccine delivery. Such data are extremely important in helping other countries recognize the potential impact of early vaccine adoption following implementation of this new vaccine. Further, the project laid the groundwork for individuals or groups, both locally in Nicaragua or globally, to independently examine vaccine impact and contribute to understanding the effectiveness of RV5 and the broader knowledge base of vaccine effectiveness studies.

This partnership brought a new vaccine to a country with great need and documented a mechanism by which critically needed scientific data were generated, helping to accelerate access to a new vaccine, a process which has traditionally taken decades after licensure to become available to the developing world.⁹ This PPP model has already been used as the prototype for the introduction of human papillomavirus vaccine in Bhutan and Rwanda.^27,28 Other similar PPP's may be developed using this model to speed new vaccine introduction efforts and evaluate their impact.

Abbreviations:
AGE	=	acute gastroenteritis
CDC	=	Centers for Disease Control and Prevention
EPI	=	Expanded Programme on Immunization
MoH	=	Ministry of Health
NGO	=	Non-Governmental Organization
PAHO	=	Pan American Health Organization
PATH	=	Program for Appropriate Technology in Health
PP	=	Public-Private Partnership
RGE	=	Rotavirus gastroenteritis
RV5	=	RotaTeq®
SAGE	=	Strategic Advisory Group of Experts
SILAIS	=	Sistemas de Atención Integral de Salud
TAG	=	Technical Advisory Group
UNICEF	=	United Nations Children's Fund

Acknowledgments

Merck Sharp and Dohme Corp., a subsidiary of Merck and Co., Inc., provided financial support for this study. We are also indebted to Monet Saunders for attentive study monitoring and local coordination.

Disclosure of Potential Conflicts of Interest

All authors are employees of Merck Sharp and Dohme Corp. and were involved in at least one of the following: conception, design, acquisition, statistical analysis, interpretation of data and drafting the manuscript and/or revising the manuscript for important intellectual content. All authors provided final approval of the version to be published. The manuscript was approved by Merck Sharp and Dohme Corp. Merck, Sharp and Dohme Corp., a subsidiary of Merck and Co., Inc., provided financial support to the conduct of the study. All Merck authors are employees of Merck, Sharp and Dohme Corp., a subsidiary of Merck and Co., Inc., and may potentially own stock and/or hold stock options in the Company. Merck manufactures and distributes RotaTeq^® (live, oral, pentavalent rotavirus vaccine).