Experts are improving care in developing nations, and rapidly spreading the improvements, by adapting quality improvement techniques used in some of the world’s most advanced and resource-rich health systems. Demonstrating this ability is University Research Co., LLC, an organization based in Bethesda, Maryland, that helps clients make changes that lead to improved programs and outcomes. University Research Co. (URC) has adapted IHI’s Breakthrough Series methodology to make widespread improvements in the care that pregnant women and newborn babies receive in several countries in Latin America and Africa.
“Essential obstetric and newborn care have been weak areas for many health systems,” says Lani Marquez, MHSc, Knowledge Management Director at URC. Under the auspices of the Quality Assurance Project (QAP), a program designed to strengthen the quality of care in developing countries and funded by the US Agency for International Development (USAID), URC has been working since 2003 to help more than six countries in Latin America and Africa expand their use of quality improvement methodology to improve maternal and newborn care.
Marquez says URC’s Quality Assurance Project Director David Nicholas learned about the Breakthrough Series methodology when he attended IHI’s Breakthrough Series College in October 2002, and became excited by the possibility of adapting it to URC’s work in developing countries. “Applying quality improvement methods and team-based problem-solving facility by facility limits our reach,” she says. “The ‘Improvement Collaborative’ methodology helps move results to a new, broader level.”
Ecuador
According to URC, there are three underlying causes of preventable obstetric-related deaths in Latin American and Caribbean countries: women don’t have access to the obstetric care or underutilize it; the care is of poor quality; and the danger signs of an obstetrical emergency go unnoticed. URC says the key to reducing maternal mortality involves increasing access to skilled care at birth, effectively managing complications, and increasing understanding among family and community members about signs that could indicate complications or problems with the pregnancy.
To tackle these challenges, URC, in partnership with the national Ministries of Health, launched a regional Collaborative focused on improving maternal care in Ecuador, Honduras, and Nicaragua. Resource constraints made it difficult to bring people from the three countries together in a central location, Marquez says, so much of the improvement work was organized as three national Collaboratives with some degree of regional sharing of results and best practices. Information was shared among the three nations on a website designed specifically for the project, through periodic site visits from local QAP staff members, and at one large multinational meeting.
Ecuador was the first to get started. Launched in July 2003 in Tungurahua Province, one of 22 provinces in Ecuador, the Essential Obstetric Care (EOC) Improvement Collaborative sought to address the causes of maternal mortality by targeting those patients in need of more than basic care, and thus in need of a referral to a regional facility, and focusing on the basics of the care all expecting and delivering mothers need. Improvers also identified the need to reduce economic, geographic, or cultural barriers to care.
“Many indigenous women didn’t feel accepted or welcomed at health facilities,” says Marquez. “They felt their traditional practices were sometimes viewed as ‘backward.’” For example, health facilities did not typically allow women to deliver in a squatting position, which is common among indigenous women, nor did they allow spouses or companions in the delivery room. “These things were important to many women,” says Marquez.
So the Collaborative leaders in Ecuador initiated a series of workshops, bringing together health workers, mothers and mothers-to-be, and women who assist in home deliveries, and began to educate the groups about each other’s needs, goals, and desires. “This was a way of building bridges between the traditional practitioners and the modern care providers,” says Marquez.
Eight improvement teams were formed in Tungurahua Province: one in the main referral hospital, and seven in health facilities in surrounding areas. During the first Learning Session, the teams reviewed the change package and the measurement strategy, then returned to their facilities to begin taking baseline measurements of the key indicators that would be used to measure results. These included process measures (such as the percentage of deliveries in the facility in which a partograph — a tool to assess the progress of labor — is used), outcome measures (such as maternal deaths), and patient satisfaction measures for both prenatal care and care during delivery. When they came together three months later for the second Learning Session, they were ready to begin implementing improvement strategies through rapid tests of change.
Clinical training of staff was done at the start in partnership with the Medical School of the University of Cuenca, through a two-day training course on Evidence-Based Medicine Applied to Essential Obstetrics and Neonatal Care. Clinical training on life-saving procedures and basic obstetric and neonatal care was done through a modular course designed to develop both knowledge and skills through practice on anatomical models. A cadre of specialists from the provincial hospitals served as instructors, after they themselves were trained as trainers. Twenty-eight staff participated, including obstetricians, nurse-midwives, and nursing staff from labor and delivery, who then returned to their facilities to train other staff. At the end of the EOC training program, more than 100 professionals had been trained in EOC, reaching nearly every health unit that provided delivery care.
In addition to learning techniques for appropriate prenatal care and care during labor, as well as important basics such as proper hand hygiene, staff were taught a sequence of clinical steps called Active Management of the Third Stage of Labor (AMTSL), which has been proven to reduce the incidence of post-partum hemorrhage, a leading cause of maternal death. With AMTSL, a drug called oxytocin is administered to the mother within one minute of childbirth, accompanied by delivery of the placenta by gentle controlled traction on the umbilical cord and counter-pressure to the uterus, plus massaging the uterus through the abdomen after delivery of the placenta.
“AMTSL is a good example of a very specific best practice that is simple, low-cost, and supported by a lot of evidence that reduces the frequency of hemorrhage and perhaps maternal death,” says Jorge Hermida, MD, Regional Director for Latin America for the USAID Health Care Improvement Project (HCI), the follow-on to the Quality Assurance Project. When the EOC Collaborative began, AMTSL was being practiced in only 17 percent of deliveries in Ecuador, concentrated in a few facilities. Despite the evidence base for this practice, it was not officially endorsed by the country’s Ministry of Health.
Hermida says that training sessions were essential to help maternity care providers grow more comfortable with this new way of managing a woman’s delivery. “Some providers were reluctant to administer oxytocin,” says Hermida. “They thought it would incarcerate the placenta,” making it more difficult to expel. “We gave them evidence-based literature addressing the fact that oxytocin does not produce that effect, plus explaining the benefits of AMTSL.”
Many clinicians were also unsure of what constitutes “gentle controlled traction” on the umbilical cord plus counter-pressure to the uterus. “They were afraid to pull too hard, or not hard enough, or they forgot to apply counter-pressure to the uterus,” says Hermida. So using obstetric anatomical models, and even a homemade device made from tire inner tubes in a few cases where models were not available, “We taught young interns and residents how to pull the cord — ‘this is too hard, this is too gentle’ — and use counter-pressure to the uterus, until they felt comfortable.”
The EOC Collaborative also introduced a number of other required action steps for clinicians, including steps considered essential in the immediate care of the newborn (placing the baby into skin-to-skin contact with the mother after delivery, immediately drying the infant with a clean cloth, managing body temperature control, promoting early and exclusive breastfeeding, checking respiration, providing reanimation/resuscitation if needed, performing hygienic care of the umbilical cord, and administering prophylaxis for eye infection) and steps for proper management of three obstetric complications: preeclampsia, sepsis, and hemorrhage. “The Collaborative was about the entire package,” says Hermida.
By June 2007, the Collaborative had spread beyond Tungurahua Province to include 12 additional provinces in Ecuador and more than 90 facility-based teams, including not only AMTSL but the entire package of evidence-based EOC for mothers and newborns. In coordination with the Ministry of Health (MOH), the practice of AMTSL was scaled up to the entire country. “Based on what was learned in the initial improvement phase of the Collaborative, we would bring together two or three obstetricians from each hospital in three large regional workshops, and train them in the techniques and in the obstacles they might face back home,” says Hermida. “They went back to their hospitals to replicate the training, and AMTSL began spreading to the rest of the country.” In June 2007, AMTSL was used for 89 percent of the deliveries in the participating facilities throughout Ecuador.
In addition, by June 2007 the percentage of deliveries in which the partograph was used correctly rose from 44 percent to 73 percent. The percentage of mothers receiving appropriate postpartum care rose from 21 percent to 87 percent; and the percentage of newborns receiving appropriate care rose to 86 percent, up from 25 percent initially. Meanwhile, in parallel efforts underway in Nicaragua and Honduras, improvers were seeing similarly encouraging results.
“The Collaborative model works because it involves working with facility-based teams of practitioners who are directly in charge of the processes that they themselves are intending to improve,” reflects Hermida. He also says that focusing on specific clinical processes is key for the involvement of hospital-based staff. “This is very appealing to clinical practitioners,” he says. “They feel they are learning and truly improving the work they are doing every day. It is practical, not just theoretical.”
Hermida says the Ministry of Health in Ecuador, which has now approved AMTSL as part of the national norms, is ready to spread the Collaborative methodology and its lessons learned on how to improve maternal and newborn care to yet more provinces, as part of a National Plan for Accelerated Reduction of Maternal and Newborn Mortality, which the Ministry of Health is preparing with support from HCI and other international agencies.
Niger
Post-partum hemorrhage is the leading cause of maternal death in the African nation of Niger, a country with one of the highest maternal mortality risks in the world. In 2006, QAP launched an Essential Obstetric and Newborn Care Collaborative in Niger. “Having had good results in Ecuador with EOC, we felt we should apply it there,” says Lani Marquez. URC had been active in improving care in Niger since the early 1990s, so there already existed a foundation of experience with quality improvement methods that facilitated a new collaborative on EONC.
Phase one of the EONC Collaborative in Niger has introduced improved coverage and quality of AMTSL, Essential Newborn Care and basic infection prevention. “Our first Learning Session was in January of 2006, and it included 80 percent of Niger’s public reference maternity hospitals,” says Marquez, including four of the nation’s five regional maternity hospitals and all three of the national maternity hospitals.
High maternal mortality rates in developing nations are often due to a common set of factors, including a high incidence of home births (about 80 percent in Niger), compounded by poor health of the mothers due to poverty and/or malnutrition and lack of access to quality skilled maternal health care. Marquez says that given these factors, when complications arise in a home-assisted birth, women often arrive at the hospital already in the late life-threatening stages of common obstetric complications such as obstructed labor, eclampsia, infection, and hemorrhage.
“In Niger, a number of internationally accepted evidence-based maternal and newborn best practices were not being routinely applied,” she says, including AMTSL. In a QAP simulated case study of 26 providers from four regional maternity hospitals working with a pelvic mannequin, 33 percent of providers appropriately administered post-partum oxytocin, but only 25 percent to 33 percent were able to demonstrate the elements of controlled cord traction, stabilization of the uterus, and appropriate removal of the placenta. In addition, fewer than 15 percent of providers demonstrated appropriate hand washing.
“In many countries, as inadequate as the normal care for delivering mothers is, there is even less attention paid to the new baby,” says Marquez. This was the case in Niger, where a QAP case study among 18 doctors and 35 midwives, randomly sampled, showed that none of them demonstrated all five elements of evidence-based essential immediate care for newborns, and only 50 percent demonstrated all essential post-partum newborn care elements (eye care, administration of Vitamin K, examination and weighing of newborn, temperature verification, support for breastfeeding, and appropriate vaccination). So the Collaborative also addressed Essential Newborn Care.
On-site training sessions were conducted for all maternal health providers by regional trainers, and at quarterly regional Learning Sessions local midwives and doctors were able to share their local innovations and successes. For example, to improve the routine use of AMTSL, sites instituted 24-hour call schedules to ensure that a skilled birth attendant is present at all births; the purchase of coolers in delivery areas made it possible to maintain oxytocin at the required cold temperatures in a setting of frequent power outages.
Kathleen Hill, MD, Senior QA Advisor for Maternal Newborn Child Health for the USAID Health Care Improvement Project, says the resulting improvements have generated a lot of excitement. “We have been able to take the Collaborative model and adapt it to very resource-constrained settings in Africa and Asia,” she says. “In Niger we have been able to leverage the Collaborative model to achieve significant improvements at scale in a relatively short time frame.” The chart below shows the dramatic decrease in measured post-partum hemorrhage rates as AMTSL was introduced into 28 public maternities in Niger, reaching 45,760 births in 2007, or 32 percent of country-wide public facility births just 18 months after the launch of the Collaborative.
Niger EONC Collaborative AMTSL Results
Hill says the impressive success of this effort must be understood in the context of the extreme challenges of working in developing settings. “To list just some of the constraints, there is a lack of updated standards for evidence-based high-impact care; a weak health system and infrastructure; and a lack of basic supplies, such as medical implements, blood pressure cuffs, and urine dipsticks to measure protein. Another challenge is that of human resources, including an insufficient number of skilled providers and frequent staff turnover and shortages of skilled providers in facilities. Much of maternal newborn care in Niger is provided by a nurse or nurse midwife with two years of professional training after the equivalent of a middle-school education in the US. Another huge challenge is the lack of medical records. In fact, when we started the Collaborative, a majority of the hospitals were out of medical record forms. There simply was no paper.”
Add to this the lack of email and teleconferencing, and general unfamiliarity with data collecting and analysis, and you gain a clearer picture of how and why the Collaborative methodology had to be adapted, says Hill.
“We depend on bringing representatives from the sites together at Learning Sessions; that’s where the real sharing occurs and the cross-fertilization of ideas,” says Hill. “But the cost of bringing people together is high, so we are looking at other ways to communicate, like regular newsletters that get circulated to all teams.”
Hill is enthusiastic about the potential improvements still to come in subsequent phases of the Niger EONC Collaborative and mindful of the unique challenges as well.
“We are looking at ways to apply the principles of quality improvement at the community level in settings like Niger,” she says. “Counseling alone rarely changes behavior; you have to engage people and communities in processes that change behavior. A woman may recognize the danger signs when she develops complications, but it doesn’t mean she will be able to access care. She may need transport, money, a driver, a vehicle. It is more complex than just teaching her what to do.”
Hill says that one of the hard things for improvement leaders is accepting that in countries such as Niger, improvements must be phased in. “We can’t try to do everything at once. It sometimes feels hard to justify given how great the need is, but we have found that we are most successful in applying QI methods to maternal newborn care when we phase in several focused pieces at a time. We have made significant improvements in post-partum care and reducing the incidence of hemorrhage, and our next steps are to use this momentum to continue improving in other priority areas, such as treating complications like preeclampsia and hemorrhage.”
“The Collaborative model has really given us a vehicle to produce rapid changes that can be scaled up. It would be hard to do what we’ve done without this model,” says Hill.
06/03/2008
