Why It Matters
Staff at three hospitals improved monitoring of mothers’ health after giving birth by designing and testing changes adapted for their environment.
SIGN UP FOR IHI EMAILS
Processing ...

A Checklist for Care in the First Hour After Birth

By Meghan Munson | Thursday, August 22, 2019

A Checklist for Care in the First Hour After Birth

Recently, three hospitals working on quality improvement (QI) initiatives chose to improve care for postpartum hemorrhage (PPH), heavy bleeding after giving birth. The hospitals reviewed their clinical protocols for identifying and managing PPH in delivery. All independently identified the first hour after delivery as their biggest blind spot, and they began making a plan.

These hospitals were collaborating with Jacaranda Health, where I work, a nonprofit working primarily with government hospitals to improve the quality of maternal and newborn care in Nairobi, Kenya. It is not surprising that the hospitals focused on PPH, as obstetric hemorrhage causes nearly 40 percent of maternal deaths in Kenya. While the number of mothers dying in childbirth in Kenya has decreased from 687 to 510 per 100,000 live births over 25 years, this only signifies a 26 percent drop, far short of the 80 percent decrease goal officials had vowed to reach by 2015 as one of the country’s Millennium Development Goals.

QI teams at the hospitals set goals for themselves for how often and what to check in that first hour. According to the National Guidelines for Quality Obstetric and Perinatal Care in Kenya, moms should be monitored every 15 minutes for the first hour after delivery for possible signs of PPH, such as heavy bleeding or low blood pressure.

Every facility’s goals differed. For example, in some facilities there are only two nurses on overnight duty at a time. When two women are in active labor at the same time, it may not be possible for a nurse to monitor every woman in regular 15 minutes intervals postnatally. In such a scenario, the teams felt that achieving the guidelines would be out of their reach for a 3-month project, and a facility might set a goal to monitor women every 30 minutes instead. Each team’s goals represented a substantial increase in checks from their previous efforts. In setting their own aims, teams were motivated to achieve their target and felt, quite rightly, that in doing so they were improving the quality of care for their patients.

I came to the project after the teams had identified their focus. Working with Cathy Green, an expert QI consultant who flew into Kenya every other week to support the project, I helped the teams move through the steps of the Model for Improvement. I met weekly with the teams, facilitating idea development, small tests of change, and data collection.        

The first team to develop and test a change idea created a checklist of the items to be monitored at given time interval. They attached the checklists to the top of client files. This would prompt nurses to complete the checklist since it was visible and would serve as a simple data collection tool by counting the number of ticked boxes. When the two other QI teams heard about the checklist, they decided to test it and tailored it to their own goals.

The first test was carried out by one nurse on the QI team at the facility with the greatest number of patients. While using the checklist during a shift, she detected PPH in a mom and intervened right away to stop the bleeding. It was remarkable.

Once they heard about this, the rest of the QI team, along with the other two teams, were eager to test the checklist as well. In less than three months, all three teams saw significant improvement in the monitoring of women in the first hour after delivery.

The teams tested a checklist for postpartum monitoring.

One challenge we faced was that one of the facilities participating did not report PPH to the Demographic and Health Information Survey, as they were supposed to, even though they would need to measure and record incidence of PPH to understand if the checklist tool was helping. They were afraid that when they started to record PPH, the government would perceive it as an increase in cases rather than an improvement in reporting. Yet, if the QI project worked, they would see a reduction in the number of PPH cases over time, which would relieve pressure from outside scrutiny. That convinced them to begin reporting again. In addition, the team was concerned that nurses were not equipped to estimate amounts of blood loss. We spent an hour with the QI team champion, a clinical officer, practicing measurement of blood loss with nurses in the maternity ward.

We learned several key lessons from the project:

  • Let the team be the experts on their work. My background is in public health research, and I have no clinical qualifications. Because of this, it was up to the teams to drive the technical part of the work. They also owned the clinical impact. Every time a woman was impacted by the use of the checklist to identify and stop her bleeding, it was a result of the team’s knowledge.
  • Don’t jump in with answers. Occasionally every team will have a rough day and not participate much in a meeting. It would have been easy on these days to tell them what to do, saying, “Please go get the files from the maternity ward so we can score them.” It would save time, but the team wouldn’t learn much, nor would they own the work. Instead, I asked questions like, “So team, how did we do last week in testing out your idea?” This would lead into a discussion in which eventually someone asked about the number of patients involved in the test, and then someone asked a colleague to get the files so they could score them together. This way the data collection and analysis would be their idea, not mine.
  • Meet regularly. When making reminder calls to teams, I often heard that we needed to cancel the meeting because not enough people could attend. Even if only one person was available, I would meet that individual. Meeting weekly was one of the most important things the team did to reach their aim. If we failed to meet, the turnout at the next few meeting would be low. But if even one or two people came to our meeting, we maintained momentum: the project moved forward, the team members stayed involved, and the one person who showed up learned that their time and input was important to me. Those individuals often became the most involved members of the team.
  • Create a communication system. We set up group chats for the team to communicate between meetings. In our context, we used a free mobile messaging platform that is ubiquitous in the Nairobi area among nurses. I used the group chats to remind teams about meetings, to check on the progress of tasks if a team needed a boost, and to share information. The teams used the chat groups to remind and motivate each other of the work to be done during the week.

The first tests were done with just a few patients to see if the checklist was practical. Over time, the scale and scope of the tests grew to encompass night shift and weekends, and eventually, the checklists reached all patients who delivered in one facility.

Unfortunately, we were not able to measure the long-term effects of the checklist on PPH. The project was cut short when doctors (and later nurses) went on strike in all government hospitals across Kenya, and PPH is, thankfully, a rare event that therefore takes time to detect. Nonetheless, the results of the project are promising for facilities where better monitoring of moms immediately after delivery could improve care for PPH.

The chart shows checks in hospital 1. N is the sample size of patients. When n is a fraction (for example, 4/18), the numerator (4) is the number of patients for whom the checklist was used and the denominator (18) is the total number of deliveries during that time period. When n is a decimal (for example, 4.5), it represents the average number of checklists for that week when data was collected over multiple weeks.

The chart shows checks in hospital 2. N is the sample size of patients. When n is a fraction, the numerator is the number of patients for whom the checklist was used and the denominator is the total number of deliveries during that time period.

The chart shows checks in hospital 3. N is the sample size of patients. When n is a fraction, the numerator is the number of patients for whom the checklist was used and the denominator is the total number of deliveries during that time period.

Meghan Munson is a Program Manager at Jacaranda Health.

Building Improvement Capability sessions are a part of IHI’s National Forum this December.

first last

Average Content Rating
(0 user)
Please login to rate or comment on this content.
User Comments

​​