Ariadne Labs Spark Grants support early-stage projects to design new interventions or new capacities for solving problems in health care delivery. Some of our Spark Grants are supported through a gift from the Paul G. Allen Family Foundation. Other projects are supported through Ariadne Labs.
Digital Phenotyping*, led by Drs. Alex Haynes and JP Onnela: This pilot uses a research platform and smartphone data to provide information to clinical teams on a patient’s physical and social functioning before and after surgery in order improve recovery.
The surgical community is good at measuring when a patient dies or experiences complications due to a procedure. This is important, said Haynes, a surgeon at Massachusetts General Hospital and director of Ariadne Labs Safe Surgery Program, but it’s not always the most important question for patients.
“Patients often ask when they can start exercising or playing golf, when they can get back to work, when they are going to feel like themselves again,” said Haynes. “It’s really important to understand these questions when we think about the burden of surgery on patients.”
To address these questions, surgeons need better measures of patient social and physical functioning before and after surgery. Social and physical functioning include how frequently a patient calls or texts their friends and family, how physically active they are and how well they are sleeping, among other variables. These are all indicators of how a patient’s recovery is going and if they are getting back to their pre-surgery lives.
The challenge is that post-surgical recovery increasingly happens at home rather than in the hospital, making it difficult to identify problems. That’s where digital phenotyping and a new research platform called Beiwe comes in.
The term digital phenotyping is a play on the idea from the field of genetics. A person’s phenotype is the trait we can see that is a result of their genes, like their height or eye color. Digital phenotyping is the expression of a person’s social and physical functioning as captured through different measures collected on a device like a smartphone. For example, the number of steps a person takes, which can be measured via accelerometers that are part of a smartphone’s standard sensors, is an indication of their ability to move around.
The Beiwe (pronounced BEE-wee) research platform was developed by Dr. JP Onnela, co-investigator on the project, and his team in the Department of Biostatistics at the Harvard T.H. Chan School of Public Health. Beiwe utilizes a smartphone app to collect information on a patient’s physical and social functioning before and after surgery. Some of this information is collected passively, like the number of steps a person takes during the day. Other data is collected actively through a brief survey that is periodically sent to the individual through the app.
In the future, Beiwe could feed the information to the clinical team on the back end, providing an understanding of quality of life a patient had before surgery and then during recovery. This data, or digital phenotype, could be used to intervene if the recovery were not progressing as expected. It could even be used to counsel patients prior to surgery on their different treatment options.
“This platform could be used as a research tool to measure effectiveness of interventions or to track patient outcomes over time,” he said. “It could be applied to a wide breadth of interventions in the future, meaning it could have broad-reaching impact. That’s pretty exciting.”
Better Evidence*, led by Dr. Rebecca Weintraub: This project seeks to understand the barriers and facilitators to using an evidence-based clinical resource, such as UpToDate, in low-income settings when subscriptions to these resources have been donated.
Diagnostic and treatment errors cause a significant amount of patient harm in settings around the world. While some are the result of inadequate supplies and equipment, a portion is due to gaps in life-saving knowledge and skills. Evidence-based clinical resources, like the UpToDate clinical decision support tool, were created to help address this gap.
Studies in high-resource countries have shown that clinicians who use UpToDate are able to make better, more efficient decisions that result in better patient outcomes. UpToDate can be accessed from a computer, tablet or mobile device. It allows users to search conditions, symptoms and treatments, pulling up information related to the search terms. Information is updated as new research or guidance is released, so clinicians have better evidence to make decisions.
Despite the benefits, getting UpToDate to providers in low-resource settings remains a challenge for a variety of reasons: a lack of reliable internet connections, low awareness of available tools and prohibitive subscription costs.
Harvard University’s Global Health Delivery project, led by Weintraub, seeks to remove financial barriers by providing free UpToDate subscriptions to clinicians who can’t afford them. Nearly 10,000 clinicians in 120 countries have received subscriptions through the program over the last six years. The majority of these clinicians access UpToDate more than once a week and read a wide spectrum of topics.
“This is a clear indication that there is a demand for evidence-based clinical resources in these settings,” Weintraub said. “But usage and, therefore, impact could be much greater.” There are still challenges in spreading the tool.
Through the Spark Grant, Weintraub and her team will assess the impact the free subscriptions from the Global Health Delivery project have on the providers’ access to UpToDate. The team will collect data on providers’ motivation to integrate the tool into their practice, their sense of self-efficacy in diagnosis and clinical management and other factors. This information will provide a better understanding of the non-financial barriers providers face and what helps them use the tool in their practice.
“Global health providers are already working beyond their capacity given the global physician shortage,” she says. “The Spark Grant will help us figure out how to get the latest practical, life-saving information into these providers’ hands so they can work more efficiently and effectively.”
Spreading Team Training, led by Dr. Alex Hannenberg: This project tests assumptions as to why team training is not more widely used by surgical teams to practice clinical care, and develops options for a training model that addresses barriers and increases access to this approach.
A clinician’s knowledge and skills are important to the care patients receive, but studies have shown that how a clinical team communicates with one another also plays a big role. Team training, in which clinicians run through different clinical scenarios without real patients, is a powerful and effective tool to practice clinical care and improve communication within a healthcare team. However, this approach is thought to be largely underutilized.
This Spark Grant will test the assumption that team training is not adequately used and investigate the reasons why. Then, the project team will develop options for a training model designed to address barriers and increase use of this approach.
“The evidence for team training’s value has become incontrovertible, and it is broadly distributed across different domains of health care,” said Hannenberg, an anesthesiologist at Newton-Wellesley Hospital. “People get to see and feel what it is to do something different and new. They can feel themselves performing better than they did the time before. It’s really a potent tool for introducing a new practice and honing one’s skills.”
Though examples of team training use can be seen in various areas of health, including mental health, Ariadne’s experience suggests that relatively few health-care professionals participate in team training. Those who have spend little time in these exercises, perhaps only once every few years. Through research conducted during the development of the OR Crisis Checklists Implementation Toolkit, the project team learned that even among those surveyed providers who had downloaded the OR Crisis Checklists, only 30 percent used emergency drills, a type of team training, in their facility.
“This is a highly-selected sample with a demonstrated interest in improving surgical safety through checklist use, and even they weren’t close to 100 percent use of team training,” Hannenberg notes. “The gap among unselected clinicians is presumed to be significantly higher – but this assumption merits exploration to validate a key premise of the project.”
Recognizing that teamwork is important in many fields outside of healthcare, the Ariadne team will begin with a search of the literature for approaches used in industry, sports, military settings and others. These models will be assessed to see if they fit the needs of healthcare teams. This research will also uncover barriers, facilitators and different approaches to the uptake of team training. The second phase of this work will be to develop options for widely disseminating team training and increasing its use among clinical teams. In the future, the project team will test and spread these solutions.
Hannenberg envisions a long-term goal in which team training is used broadly and consistently beyond just the operating room. Still, the impact on surgical care alone would be significant. “There are more than 40 million surgical procedures a year in just the United States,” he said. “The potential number of lives saved and amount of suffering avoided by reducing errors caused by poor communication is massive.”
Measuring Management*, led by Dr. Raffaella Sadun: In this project, Ariadne will use an adaptation of the World Management Survey to provide the first Census-level measurement of management across hospitals in the U.S.
How management processes are adopted and how decision rights are distributed in private sector businesses have been studied for a long time. That is not the case for hospitals.
“We still have a pretty naive understanding of what happens in a hospital from an organizational perspective compared to what we know about other industries. It’s quite shocking,” said Sadun, associate professor of business administration at Harvard Business School.
Through this Spark Grant, Sadun leads a team at Ariadne to assess how basic managerial capabilities vary across acute care hospitals in the United States using the World Management Survey. This first of its kind census-level measurement of management across hospitals is made possible by partnering with the U.S. Census Bureau and its Center for Economic Studies. The data collected will allow the project team to study the relationship between management and clinical care in this large sample and identify possible areas for improvement across organizations.
The Measuring Management work grows out of more than 10 years of research on measuring management at scale across industries. Sadun began looking specifically at management practices within hospitals in 2009. She points to the increasing number of hospital consolidations and mergers as a reason why this work is so relevant now. The goal during mergers is often to deliver more efficient and overall better care, but this is grounded in the assumption that it is easy to manage a hospital well and easy to combine different facilities. Sadun says this assumption hasn’t truly been tested yet.
“Hospitals are complex organizations with many people and moving parts, so it makes sense to study how they work,” she said. “We need to understand what’s happening before we can expect that a merger or other similar initiative can be successful.”
She discovered that it is possible to measure the intangible aspects of what happens in the hospital using the World Management Survey. This survey is based on phone interviews and has been used to build the first cross-country, cross-industry dataset measuring the differential adoption of management practices across organizations. About five years ago, the Census Bureau started working on an adaptation of the World Management Survey for the manufacturing sector (the Management and Organizational Practices Survey, or MOPS). The adaptation translated the interview-based tool into a paper-based one, eliminating the need to train interviewers and coordinate with interviewees’ busy schedules. Sadun has already begun a similar translation project geared towards hospitals, so the collaboration with the Census Bureau comes at an opportune time.
If successful, this approach could become one instrument to measure hospital management at scale over time, and potentially across countries. That means it could potentially influence policy decisions within hospitals and even government bodies. Sadun points to the manufacturing survey as an example of the kind of spread that could happen with the right tool. Several countries, including the United Kingdom, Japan, Pakistan and Australia, have adopted the manufacturing survey created by the Census Bureau. “The beauty of this type of approach is that it is scalable,” Sadun added.
Ready. Aim. Implement, led by Dr. Natalie Henrich : This project aims to determine how to reliably and conveniently assess a site’s “readiness” to implement a new intervention informing strategies to improve the likelihood of successful implementation.
Even the most promising interventions can fail when it comes to real-world implementation. A facility may be understaffed. Sometimes the right medications aren’t available, or a lack of leadership creates low morale. A number of factors will either improve or reduce the chances of success.Through the work of the Ready. Aim. Implement. grant, Ariadne will determine how to reliably and conveniently assess these factors to determine a site’s “readiness,” their willingness and ability to implement a new intervention. This understanding will inform how sites may introduce a new intervention and improve the likelihood they will meet their goals.
Led by Dr. Natalie Henrich, Ariadne Labs’ associate director of Science and Technology, the Spark Grant has two goals. The first is to select or adapt an existing readiness assessment tool or create a new tool that can be used by all of Ariadne’s programs. The second is to create a list of recommendations for Ariadne’s approach to working with sites once their readiness has been determined.
“Readiness for implementation is not simply a question of whether sites are ‘ready’ or ‘not ready.’ It’s a spectrum,” Henrich explained. Sites have varying strengths and weaknesses across a range of factors that may impact their ability to implement any given program. Sites that are more ready are stronger in more of these factors, and sites that are less ready are weaker in more of these factors.
Readiness can also be thought of as the context in which an intervention is being implemented, so it may change depending on the intervention that is considered. A site may be more ready to adopt an intervention related to administering a medication because the drug is reliably in stock, but may be less ready for an intervention related to a procedure because the staff have not received training on how to do it.
Having a reliable way to assess readiness is important for Ariadne as more of its solutions, like the Serious Illness Care Program, the OR Crisis Checklists and the Safe Surgery Checklist, begin to spread more widely. Ariadne has an increasing need to be more effective in supporting implementation.
“If we can better understand the context in which our solutions are being implemented, then we can hopefully increase the rate of successful implementation,” Henrich said. “Additionally, what we learn from implementing our current solutions will help prepare us for when we have new solutions that are ready to spread.“
*Funded by the Paul G. Allen Family Foundation project