This post was originally published on Project Syndicate here.
To improve quality of care before, during, and after childbirth, the global health community must develop new, evidence-backed interventions that address the underlying – often hidden – reasons why health-care providers fail to take the necessary steps. The first step is to identify what those reasons are.
Wash hands? Check. Monitor heart rate? Check. Prepare essential supplies? Check. These might seem like obvious steps for medical professionals to take while delivering a child. But lapses in care remain one of the leading causes of preventable patient deaths in low- and middle-income countries, and initiatives aimed at addressing the problem are not working.
One such initiative focuses on creating a checklist for birth attendants to consult. But in a multiyear, multimillion-dollar randomized controlled trial conducted in northern India in 2017, the use of the World Health Organization’s Safe Childbirth Checklist, together with coaching on its implementation, did not improve outcomes for babies or their mothers.
Even if a childbirth checklist has some potential benefits, it amounts to an insufficient basis for efforts to address the scourge of high infant and maternal mortality. And yet, as a report in the British medical journal The Lancet shows, such micro-level interventions – including direct mentoring of providers – constitute 72% of all strategies for improving the quality of primary care globally. Although such micro-level interventions can help to improve local commitment to quality, “people tend to revert to entrenched ways of doing things, especially when surrounding systems do not support transformation.” The focus on micro-level interventions alone can even be detrimental, as such measures consume limited time and resources.
To improve quality of care before, during, and after childbirth, the global health community must develop new, evidence-backed interventions that address the underlying – often hidden – reasons why health-care providers fail to take the necessary steps. The first step is to identify what those reasons are.
One answer that can immediately be ruled out is that more training is all providers need. Existing survey data indicate that, in general, transferring relevant knowledge and skills to nurses and other health-care providers is not enough. Even when more nurses become aware that they must monitor blood pressure or refer a complicated case to a better equipped hospital, for example, they don’t always do it, or the effects do not last.
The imperative, then, is to explain the gap between providers’ knowledge and behavior. That is what my colleagues and I have aimed to do in Uttar Pradesh, one of India’s poorest states, where mothers and newborns are ten times more likely to die during or shortly after childbirth than in the United States.
After observing more than 20 clinics and conducting in-depth interviews with dozens of nurses and other staff, we developed several hypotheses to explain providers’ failure to take the necessary actions. We then tested those hypotheses using a series of novel decision-making games designed to elucidate the factors driving nurses’ choices.
The factors we identified fall into two categories: perceptual drivers (health workers’ fears, beliefs, motivations, biases, and perceptions) and contextual drivers (demands from patients and their families, the attitudes of doctors, hospital infrastructure, and processes). The two categories are closely interconnected.
Nurses in Uttar Pradesh, our research showed, have little support from the doctors with whom they work, but tend to be blamed – and punished – when something goes wrong. Moreover, families do not always respect nurses, and often will resist a nurse’s recommendation to refer a woman to a bigger hospital.
As a result, nurses are under constant stress and live in fear of risks to themselves, which end up taking precedence over the risks faced by patients. Given this, many nurses focus on those tasks for which they are solely responsible, such as the delivery itself, while letting less acute tasks related to that process fall by the wayside.
I witnessed such a dynamic firsthand during our field research. A nurse was sitting at her desk when I heard voices coming from the room behind her. A young woman had just delivered a baby and was lying on the floor, very thin and obviously exhausted. In that moment, someone should have been encouraging the mother to breastfeed; doing so immediately after delivery is vital to build the child’s immunity. But when I pointed this out, the nurse told me that it was not her job.
In truth, it was the nurse’s job. But once the baby had been safely delivered, she had disengaged – not least, in that particular case, to avoid dealing with the baby’s grandmother. As a result, both mother and baby missed out on critical care, a lapse that could have devastating consequences.
The point is not to cast more blame onto nurses; on the contrary, our research makes clear that nurses need better working conditions to do their jobs well. To this end, hospitals should offer supportive, rather than punitive, supervision. A culture of collaboration and team problem solving should be fostered. Hospital managers should be held accountable for health outcomes. Efforts to manage community expectations, so that patients and their families know what to expect and how to interact with providers, would also help. What will not help is another checklist.
Of course, the main factors that drive health-care providers’ decision-making are not exactly the same everywhere. More localized research is needed to enable us to tailor solutions to each context. But by asking why providers behave as they do, global health programs can save millions of dollars – and many more lives.
Wash hands? Check. Monitor heart rate? Check. Prepare essential supplies? Check. These might seem like obvious steps for medical professionals to take while delivering a child. But lapses in care remain one of the leading causes of preventable patient deaths in low- and middle-income countries, and initiatives aimed at addressing the problem are not working.
One such initiative focuses on creating a checklist for birth attendants to consult. But in a multiyear, multimillion-dollar randomized controlled trial conducted in northern India in 2017, the use of the World Health Organization’s Safe Childbirth Checklist, together with coaching on its implementation, did not improve outcomes for babies or their mothers.
Even if a childbirth checklist has some potential benefits, it amounts to an insufficient basis for efforts to address the scourge of high infant and maternal mortality. And yet, as a report in the British medical journal The Lancet shows, such micro-level interventions – including direct mentoring of providers – constitute 72% of all strategies for improving the quality of primary care globally. Although such micro-level interventions can help to improve local commitment to quality, “people tend to revert to entrenched ways of doing things, especially when surrounding systems do not support transformation.” The focus on micro-level interventions alone can even be detrimental, as such measures consume limited time and resources.
To improve quality of care before, during, and after childbirth, the global health community must develop new, evidence-backed interventions that address the underlying – often hidden – reasons why health-care providers fail to take the necessary steps. The first step is to identify what those reasons are.
One answer that can immediately be ruled out is that more training is all providers need. Existing survey data indicate that, in general, transferring relevant knowledge and skills to nurses and other health-care providers is not enough. Even when more nurses become aware that they must monitor blood pressure or refer a complicated case to a better equipped hospital, for example, they don’t always do it, or the effects do not last.
The imperative, then, is to explain the gap between providers’ knowledge and behavior. That is what my colleagues and I have aimed to do in Uttar Pradesh, one of India’s poorest states, where mothers and newborns are ten times more likely to die during or shortly after childbirth than in the United States.
After observing more than 20 clinics and conducting in-depth interviews with dozens of nurses and other staff, we developed several hypotheses to explain providers’ failure to take the necessary actions. We then tested those hypotheses using a series of novel decision-making games designed to elucidate the factors driving nurses’ choices.
The factors we identified fall into two categories: perceptual drivers (health workers’ fears, beliefs, motivations, biases, and perceptions) and contextual drivers (demands from patients and their families, the attitudes of doctors, hospital infrastructure, and processes). The two categories are closely interconnected.
Nurses in Uttar Pradesh, our research showed, have little support from the doctors with whom they work, but tend to be blamed – and punished – when something goes wrong. Moreover, families do not always respect nurses, and often will resist a nurse’s recommendation to refer a woman to a bigger hospital.
As a result, nurses are under constant stress and live in fear of risks to themselves, which end up taking precedence over the risks faced by patients. Given this, many nurses focus on those tasks for which they are solely responsible, such as the delivery itself, while letting less acute tasks related to that process fall by the wayside.
I witnessed such a dynamic firsthand during our field research. A nurse was sitting at her desk when I heard voices coming from the room behind her. A young woman had just delivered a baby and was lying on the floor, very thin and obviously exhausted. In that moment, someone should have been encouraging the mother to breastfeed; doing so immediately after delivery is vital to build the child’s immunity. But when I pointed this out, the nurse told me that it was not her job.
In truth, it was the nurse’s job. But once the baby had been safely delivered, she had disengaged – not least, in that particular case, to avoid dealing with the baby’s grandmother. As a result, both mother and baby missed out on critical care, a lapse that could have devastating consequences.
The point is not to cast more blame onto nurses; on the contrary, our research makes clear that nurses need better working conditions to do their jobs well. To this end, hospitals should offer supportive, rather than punitive, supervision. A culture of collaboration and team problem solving should be fostered. Hospital managers should be held accountable for health outcomes. Efforts to manage community expectations, so that patients and their families know what to expect and how to interact with providers, would also help. What will not help is another checklist.
Of course, the main factors that drive health-care providers’ decision-making are not exactly the same everywhere. More localized research is needed to enable us to tailor solutions to each context. But by asking why providers behave as they do, global health programs can save millions of dollars – and many more lives.