Loved the presentation. Nigel shows how real-world QI demands very adaptive and contextually based learning and testing of solutions. Given our DNA of central planning and control, how do we evolve to promote diffusion of decentralized / localized hubs of know-how in these processes?

ANSWER

I hope we learn how to do this in this network ! My current thoughts on this are :

• Get started at the patient level. Support a few selected facilities or districts to use adaptive approaches to fix problems.
  
  • Start with specific improvement projects based on 4 R’s
    • Results – look for projects that are going to get results
      • Under the control of the team working on them
      • Doesn’t obviously require new resources
      • Is a routine aspect of care not a relatively rare event
      • Is easy to measure
    • Rapid – look for things that you can fix within weeks rather than months so focus on every day rather than rare events
    • Relevant – the projects and their outcome should matter to the team doing the work
    • Recognition – the outcomes should be something people outside of the team will recognize and value (e.g. communities, patients, supervisors)
  • Usually you can’t get all 4 but do what you can
• Show the results as soon as possible and as widely as possible
• When talking about the work, focus on the results rather than the method until people ask about the method
• Look for examples of problems that QI teams that could not have been solved by central planning or command and control
• Look for organizations that may be interested in promoting these ideas institutionally and help them. Don’t just focus on the skills of QI but support them to build structures and systems and develop a culture (not so easy)

How long do you have to implement a change idea in the cycle before adding another one in the quality improvement process?

ANSWER

• It depends on the change – some can be tested very quickly. For example, staff in an NICU were trying to improve aseptic technique while putting in IV lines. They decided the way the cleaning tray was organized was a problem so they tested 3 different options for setting up the kit. They did this with a mock patient initially – so it took about 10 minutes to test all 3 options. When they picked the one that worked best they tested it on 3 patients to see if it was feasible. They then tested for 3 days to see if it was feasible for lots of people and if it helped.
• In general, tests to see if a change is feasible or to learn how to adapt it to your setting can happen very fast. Test to see if a change is effective often need to take a bit longer.

Hi, my question is whether or not any approaches you have used or recommend are now implementing social and behavior change methodologies, given that half the WHO QoC framework focuses on experience of care, which is often negatively influenced by provider biases and attitudes

ANSWER

• I think a lot of this approach is based on behavior change literature and thinking. In general, people will change their behavior when they have information that the change is useful, motivation to change and the skills to change. Self-efficacy (the confidence that you can change) is another important predictor of people’s ability to change. We focus more on the skills, motivation and self-efficacy elements since health workers often have the information about what they should be doing.
• We try to build motivation and self-efficacy by: helping teams pick problems that matter to them, that are going to lead to quick wins and will be recognized by others. We also focus on their successes and are very diplomatic about the challenges they face (which occur whenever people learn new skills so are universal). We also promote sharing of results and progress between teams so they can learn from each other and add to motivation.
• The key skills we work on are: seeing that behaviour change is a decision that people make so they should be involved in deciding how to change their behaviour rather than having change forced on them (this is why team work is important: staff should not be told what to do to fix a problem they should decide for themselves actively participate in deciding what to change and how), identify the critical barriers to good performance so you are not doing work that doesn’t lead to benefits, using different change management approaches to support change.

How can the driver of national quality strategy be utilized maximally for this work?

ANSWER

Hopefully we can learn more about this in the network. There are a lot of national strategies that stay in paper form. We need to learn how to translate them into something that changes the patient/provider interaction. National strategy can be useful for communicating the government’s commitment and for emphasizing that this work is a priority, for outlining required organizational changes (for example new committees and so on) and describing new activities that should take place (e.g. setting up methods to support learning communities). But too often, national quality strategies don’t really use all the available approaches to improve care (they focus a lot on measurement and inspection rather than focusing on supporting change) and are not appropriately resourced. So using strategy maximally would involve that that they include a wider range of strategies to improve care and that they are resourced.

I’ve often felt that one often needs a champion to continue driving a project. Can you comment? Is it needed or how to motivate w/o an internal driver(s)?

ANSWER

I think this is very useful and important at the beginning of improvement work. Over time though, an organization should work to build structures and systems so that it is easier for a wider range of people to do more projects. Leadership can also help by modeling and encouraging the attitudes that are important for improving care in complex situations.

The 4 steps you spoke about: I miss a step in which teams learn from the impact of their changes as well as continuous data collection to monitor the impact of the change. Is this included in the 4 steps?

ANSWER

The 2^nd step involves measuring what you are doing. I agree with you that this is critical. People should continuously measure what they are working on so they can learn what works and what doesn’t. The webinar on this step will describe some ways of doing this.

How do we measure QI?

ANSWER

The first and most important thing to measure is if it is working – if mortality is going down and patient experience is going up. We should also measure some things along the journey of deciding to do QI work and getting the desired outcomes. This would include the QI teams measuring how well they are delivering the processes are intended to lead to the outcomes. In addition, the QI program would like to measure how many QI teams have been trained, how many are getting regular onsite support, how many teams are meeting in peer learning communities, how many teams are making progress on improving processes and outcomes.

How can implementation inform the refinement of a national quality strategy to get it off the shelf??

ANSWER

Great question. Hopefully, as governments and other stakeholders implement their QoC plans in this network they will identify weaknesses in existing strategies (‘Everyone has a plan, until they get punched in the mouth’) and then address those weaknesses. I don’t think that a perfect QoC strategy exists anywhere in the world so there is room for improvement in all of them. This network provides an opportunity for 9 countries to focus on supporting a few facilities to improve care using all the available methods. if we can carefully document what is happening, accept that we are going to make mistakes and look for those mistakes, share among ourselves and use this learning to write and implement better strategies then in five years we are going to know so much more than we currently do about how to make sure people get the care that they deserve.

Do u think Nigel we need political, bureaucracy and professionals on board additional to local champion for QOC success ?

ANSWER

Very much so. Champions are good for getting started. Garnering political, bureaucratic and professional support is essential for large-scale, long-term improvement. We need both – committed individuals and strong systems.

Increasing intervention coverage does not directly reduce maternal mortality – WHO Multicountry survey showed higher coverage of life saving interventions in settings with higher MMR. Also the large trial on use of Safe Childbirth Checklist showed improved in processes but not in outcomes. Any comment on how this impacts on the 50% reduction envisaged through this process?

ANSWER

I am not sure how these things were measured in the studies you mentioned so can’t comment directly. We do quite often see that, initially, different teams get good results on process measures but not on outcome measures. This is usually because outcomes are often the result or more than one process. For example, some teams we worked with a few years ago started by working on the process of drying babies immediately. They did this well but hypothermia rates didn’t change. This was because labour rooms were often cold, babies were not placed skin to skin etc. so the teams added the new processes (keeping the room warm, doing earlier and longer skin to skin). This is why focusing on the outcomes is so important. If you just measure processes and hope that this will lead to the right outcomes you are often disappointed. Instead, you should routinely measure outcomes, start work on the process that you think is important and easy to fix and move on to new processes if you aren’t getting the outcomes you desire. Humility is key, humans are bad at predicting whether our efforts are going to work (and way too overconfident). Using outcome measures to keep ourselves honest and keep us asking what else we need to do is useful.

Do you have any suggestions to begin a QI project in terms of convincing the wider hospital/clinic team of the need to change?

ANSWER

Yes, get results. Start with work that is going to quickly lead to results (See first question) and then show people the results. Try to show them the results in an interesting way as well. If you do this, often a few more people will be interested and join in and you can use this approach to get broader buy-in.

The results you showed in the presentation are very impressive. How do you sustain performance beyond partner support?

ANSWER

Great question. Our main interest is helping governments and other large institutions build the systems to continue to improve care. Some of that is happening as part of this network so we hope to contribute some learning about this question (and to learn from others).

In your experience, what’s the best way to engage patients and their families in QI?

ANSWER

This can be challenging. The separation between the healthcare system and the people it is meant to serve is usually distressingly large and it can be hard to get healthcare workers to really engage patients and families in QI. Over time, this can change but it is tough at the beginning. I think it can be easier to work with communities directly to help different groups use QI approaches and then have those groups engage the health care system. I haven’t done tons of this work but I think it would be a good topic for a webinar and there are a lot of people with a lot of experience in this.

How can leadership support if resources are really not available (no money to buy equipment, drugs, train staff).

ANSWER

Changes in processes typically don’t require new resources so they are good places to start.

Mistreatment in health facilities is a reflection of the extent that communities tolerate abuse. How is the community involved to changing their expectations and value on women’s health?

ANSWER

Increasing the communication between the two groups can help. We worked in a district in India were there was a major trust problem between the health system and community. Institutional deliveries were about 10% because the community thought that if they went to facilities, they would be attended by a male doctor and would be treated badly by staff. To address this, the QI team at the facility organized an orientation visit for pregnant women and their families to come at the same time to meet the staff. Because they were going in a group, they felt comfortable. When they got there they met the female doctors who delivered babies and staff were nice to them. They also did some antenatal care and found some health issues that they were able to take care of. This improved the relationship and more women delivered in these facilities. I don’t think there are quick solutions but more communication between communities and health services in situations where the power differential is minimized should be part of the solution.

Do we have QI advocacy tools for the engagement of key stakeholders?

ANSWER

I don’t know of any tools that would be universally useful. The tools I use are results, stories, and simple explanations of QI approaches. Results is the most important. Working with a limited number of facilities to get good results gives you something that you can use to talk to stakeholders. This is of more interest than simply talking about another set of approaches for improving health care. Telling specific stories about how staff worked in teams to get these results is also useful. When people do ask about methods it is then good to have simple explanations and tools for teaching QI skills.

I think this network is a great opportunity to get good results and to build a large set of stories about how teams fixed problems and got the good results. I think this is going to provide a lot of good advocacy materials but the key thing is to get started supporting facilities to improve care. until that happens we won’t make much progress.

What are the risk and returns on QI ? Any study on patients perspective of effective QI?

ANSWER

I think the returns are that this is an approach to allows healthworkers to solve process type problems. The risk is that, like anything, QI can be done badly. Some mistakes include:

• Trying to do everything at once and not prioritizing
• Trying to apply standard solutions to fix problems (solutions typically need customization at each facility even if the basic logic of the solutions is the same)
• Using inspection too much so people hide problems
• Focusing too much on what activities are happening (such as how many people trained, how many coaching visits), and not enough on outcomes – what is happening to mortality and patient experience

There are quite a few studies on patient perspective on QI in pubmed and others but I don’t know how many are from LMIC. In general the emphasis on patient centred care and QI go hand in hand. By involving patients in designing services that work for them (a crucial concept in QI) they get better outcomes and better experiences.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4778631/

Can you discuss the process of how to get started, how to build the capabilities in facilities to start the quality improvement? How can the guides developed in South East Asia region help?

ANSWER

Start quickly. You learn much than from getting started that you will from a long planning period. Getting tens of facilities to start QI work fast is relatively easy. You can organize a quick introduction to QI skills, help them pick a simple problem and set up a system for regular onsite support of these facilities. The SEARO guides can help with this. They provide a format for a 2 days training that runs through a set of case studies about how a hypothetical team used QI approaches to improve care for mothers and babies. On the second day, they then plan a project for their facility.

Do make sure you have a plan to provide regular support to the new facilities to help them apply these skills. The key thing for the people providing this support are that they are nice to the health workers. They are not there as police, they are there to help them learn the QI skills. So during these visits they should find out how the team is going with analyzing and measuring the problem and how well they are developing and testing changes. The visits should mainly be about what they are going to do and helping them do it better.

How do you deal with big problems-water and electricity?

ANSWER

We talk to the people who control the budgets for these items and see if they can do anything.

Role of managerial skills in addition to the clinical skills? Because multitasking by the health care provider team is often the case

ANSWER
Yes, managerial skills are very important. Many of the same steps apply to managers as well. Prioritizing what you are going to work on is particularly important and it is important to try to get managers to be clear about what are their priorities.

I think the outcome measures change or improvement come late after the input and process improvement that we measure like mortality data and how long shall we wait for the change ideas are linked with improvement? I mean with outcome specially mortality unlike PPH and other outcome data.

ANSWER

Yes, outcomes will come later. Changes in processes will add up to changes in outcomes but you typically need to change quite a few processes to get the outcome you want. I don’t think there is a single answer to how long to wait but we tend to encourage teams at facilities to measure their process indicators in days or weeks so that they can see the effect of their changes quickly. Conversely we typically suggest they measure outcome data in months since these do take longer.

Probably the best advocacy to politicians and bureaucrats are those “survivors” testimonials.

ANSWER

I agree, these can be very powerful.

Why is that QI projects are not decentralized? Only a few centres are involved. Can we involve more centres?

ANSWER

Any center can get involved. There are a lot of materials freely available on line if someone wants to start a QI project. Hopefully, governments and other institutions will work more on building systems to support the spread of QI work to make it easier for centres to use QI approaches.

How do you select a champion, what is their benefit to the programme?

ANSWER

I think they usually select themselves and different people emerge as champions at different times and for different purposes. Some of the benefits they can have are: getting a QI project started, persevering with difficult problems to get great results, advocating with peers and leaders, spreading new ideas to new groups, modeling the key attitudes that are critical for QI, being good teachers, having special skills in different QI methods, building networks, building systems to support QI. I don’t think any one individual needs to be good at all of these but having individuals with these strengths definitely helps an organization.

How can QI be made sustainable in the long run especially in public health facilities (in an Indian context)?

ANSWER

I think in the long run, governments will need to put in systems to support the use of QI. My guess is that this should include:

1. staff positions in districts and states for people with expertise in QI methods,
2. resources for training and (especially important) ongoing support for health care providers and managers to learn about and apply QI skills to solve their problems,
3. Mechanisms to support peer-to-peer learning communities
4. Methods to integrate QI approaches and work with the rest of the health care system
5. Changes in HR and data systems to make it easier to use improvement methods
6. Efforts by leaders to promote a culture shift in the health system to one that focuses on solving problems, promotes team work, looks for system problems when things go wrong rather than blaming individuals

Hopefully, by working in this network, we will all learn more about different ways of building a sustainable system to support health workers to use QI approaches.

When you come to the QI meeting what should happen in the meeting?

ANSWER

Great question. A typical first QI meeting would focus on picking a specific improvement goal and identifying a team and measures that the team is going to use. It may end with some members volunteering to collect some baseline data and doing some analysis of the problem (using tools that will be discussed 2 webinars from now). The next meeting would review the data and the analysis. The group may further refine the analysis. They may then come up with some ideas of changes that they think will work and develop a plan to test those changes (using tools that will be discussed 2 webinars from now). Volunteers then go and test these changes and make refinements until the next meeting. At the next meeting the team would discuss the changes and what they learned, discuss the data to see how they are performing. They would then make plans for what to do next – what changes to test, how and who will test them. So most meetings involve looking at what they have learned in the past (about what changes were feasible or not, how changes were adapted to make them more feasible and if the changes led to better performance) and what they plan to do in the future. Signs of good meetings include:
• More people talking – not just the leader lecturing
• More facts (from data and from testing) and less opinion
• Planning and next steps involve more questions (“Is this feasible? Will this work?”) and fewer commands (“X, go and do this tomorrow)

Did health workers always come up with their own solutions; or did they require outside help at times? If yes, what were the common types of the outside support?

ANSWER

In general yes. People on the QI teams came up with ideas that they thought would be solutions and they then tested those to see if they are feasible and to refine the feasible ones until they solved the problems. Sometimes, teams are not sure what to do. In that case the ‘coach’ (a person visiting to help them learn the application of QI skills) may suggest some ideas to test. Alternatively, they may hear about something that worked in another facility and try that. However, they come up with ideas that may solve the problem, they still need to test and refine them so that they work in their setting.

COMMENT/RESPONSE

Prioritization matrix discussion among Team – identifies the problem to fix while solutions again are decided by frontline workers and tested as small PDSAs. This is in the POCQI materials shown on the website.

I have the same question especially in how to get started in quality improvement. I am implementing a WASH in Healthcare facility and would want to do QI in Infection Prevention and Control. So how do i monitor for instance the six clans within the labour ward?

ANSWER

You should pick a problem that is likely to get results, that can allow rapid changes and learning, doesn’t require too many outside resources and will be recognized by others. You should also be careful with the team. If someone is going to have to do their work differently then they need to be an important part of the team. This usually means for IPC work that they cleaners need to be involved. You should not have clinicians making decisions about how cleaners will work and then telling the cleaners what to do. The cleaners should understand how their work will contribute to patient outcomes and come up with ideas about how they can help. Overcoming power differences and hierarchy issues needs some work for these types of problems.

I tend to stay away from handwashing as a first QI project since the measurement usually requires inspection – someone keeping an eye on whether people are washing their hands properly. It can be hard to teach the principles of team work, encouraging people to think in terms of systems, trusting health workers to solve problems when your project is using inspection. So I usually start with things that can be measured without inspection and which lets you know how the whole system is working rather than how an individual is performing.

On training, did you also assess clinical skills and how or did you address any clinical skill gaps?

ANSWER

Yes. When trying to look for the cause of poor care, the team sometimes identifies a gap in clinical knowledge or skills. Some of these they solve themselves with in-service training or setting up practice stations. In other cases, they need outside support. In the work I presented, if we found skill gaps needing external support we would connect facilities to government resources for training and we also sometimes contracted with local specialists to provide on-site training and mentoring in the specific skills.

Post-training follow up is essential, as Nigel said. It is a collaborative learning within the team in a health facility and between teams at neighbouring health facilities in a geographic area.

Will the SEARO manuals you recommended have additional case studies and reports added… eg through a blog system?

ANSWER

There are a number of case studies in them already and there are plans to develop more. The idea of a blog system so people could add their own stories is great. I’ll pass this suggestion on to SEARO.

Since QI is based on data how do we ensure that the data being used is valid?

ANSWER

During coaching visits, the coach often does some verification work with the team to look for data issues. We have also commissioned external data validation checks. In general, the self-collected data that the teams use is pretty good. It often overestimates performance by a bit but not much. It is typically better than the data in reports submitted to those higher in the organization.

How large are the networks that exist in Asia today and are they at the tertiary hospital level or secondary/primary as well?

ANSWER

The SEA network currently has over 200 facilities providing care to over 400 000 deliveries per year. It is a mix of primary, secondary and tertiary

How do you institutionalize QI process at health facilities?

ANSWER

We have a lot to learn about this. I hope we all learn more through this network. Training people in using QI skills and setting up a QI team is not enough. In addition, the leadership of the facility needs to be interested and needs to provide support to the QI team. This support may include: permission to use time and hospital resources to apply improvement methods to deliver better care, encouragement to staff to make problems with quality of care visible and to fix them, looking for system solutions rather than blaming individuals.

It is on the website. The audio that accompanies the ppt should be available soon.

How best we can address involvement of the nurses for a problem solving approach, without being depended on the doctors? In some settings, they lack confidence to do it on their own. How to simplify the process further for them?

ANSWER

In my experience, nurses are just as capable of using QI approaches as doctors and in many countries the bulk of QI practitioners and leaders are nurses. Unfortunately, many health systems discourage nurses from applying their full potential to fixing problems. I think there are many approaches for training in QI that are appropriate for nurses. If nurses in a specific setting are lacking in confidence, that may be because of how they are being treated rather than their ability.

It is important for the trained and practicing healthcare teams to embed the QI within the work culture of the health facility so that it sustains.

ANSWER

Yes, agree.

Is training still can be cost effective approach for skill gap in terms of quality dimension as compared to mentorship?

ANSWER

I am not sure of that. I would guess that for simple things classroom training can be cost-effective. For complex skills there is a lot of evidence that people don’t learn complex skills well in classrooms. They learn them much better from coaching and review approaches to build skills. I am sure others in the group will have a better answer.