Improving Patient Experience and Efficiency with Data Visualization and Management
Data analytics is important in healthcare because it collects, analyzes, and evaluates massive amounts of data from a variety of sources, including electronic health records, patient experience data, and individual patient demographic information. Using modern statistical analysis and data visualization tools may give useful insights into everyday patient care operations and opportunities throughout the patient journey. These sorts of analytical insights may help clinical leaders and hospital operations leaders make more educated decisions that will improve clinical outcomes and, ultimately, the patient-family experience. Healthcare companies have to optimize their data sets to detect obstacles, improve procedures, and minimize costs for patients, which improves the entire patient experience.
Melissa M., senior director of patient experience and operations leader, will share with us her insights on using data to enhance efficiency and improve patient experience, as well as how she uses data for everyday operations.
The data displayed in the chart below, is the monthly trend of Jackson Health Systems “top box” response rate %.
This percentage is the percent of patients that responded “always” to the HCAHPS questions.
The “N” is the total number of surveys returned for that month.
The date range: Fiscal year 2024.
HCAHPS- the Hospital Consumer Assessment of Hospital Providers and Services
L: Hi, Melissa, first of all, thank you so much for being here today.
M:Thank you for having me. It’s my pleasure, I am Senior Director of Patient Experience, and operations lead over all of our patient experience data, which is considered what we call in the healthcare industry HCAP, which is the hospital Consumer Assessment of Healthcare Providers and services. And that is our patient experience survey for impatient patients only. That’s my primary responsibility in my primary data set that I oversee. In my role as Senior Director of Patient Experience.
L: What type of data do you primarily work with ?
M:We work with two different types of data, we work with structured and unstructured data. So structured data is the responses using a Likert scale for the HCAP survey. And those Likert scales are very good, good, fair poor, which are converted into numbers, we can apply advanced analytics to those types of numbers into those different types of scales. I also work with data that is categorical data, such as patient demographic information, their gender, socioeconomic status, insurance status, or what their primary DRG is, which is considered their diagnosis related group. And that is like if you got pneumonia coming into the hospital, etc. I also work with unstructured data, which is patient comments. So patients can also leave comments on Google, they can leave comments on Instagram, Facebook ,comments on the actual survey of how to make their experience better with that unstructured data. We run it through other types of models in order to do sentiment analysis. And then when we combine all of this data together, we use again other types of models to get operational opportunities and where our pain points and friction points are for patients throughout their journey.
L: Are there any specific software tools you use on a daily basis for data analysis and visualization?
M:First off, I’m going to talk about how we ingest all of the data. Once we talk about that we’ll talk about the different types of tools and visualization tools that we use in order to analyze the data. First off, we have different sources of data coming from all different aspects of the patient experience, we have it coming from their electronic medical record, which we use Cerner within that medical record, one patient in their experience can have multiple millions of data points and touch points throughout their journey depending on how long they stayed at the hospital. So that’s one source of entry is Cerner. And that goes into what we call our enterprise data warehouse. So all of those are clinical actions. Then we also have other types of API input data that goes into our enterprise data warehouse that more measures the patient journey and different aspects of their outpatient experience their wait times, which comes from our flow capacity management system. And all of that goes into one data warehouse, which is stored in servers outside of our health system. And then we set on top of that our data visualization platform, which at Jackson, we use Tableau. Now something important to understand about this data system is what we do is we normalize all of our data, and we clean up the data, there’s many different inputs, like imagine, again, a warehouse with multiple slides going into it and all this data is going in, but sometimes the Cerner the electronic health record, data doesn’t match or talk to the patient experience data. So we have set parameters within the data warehouse that clean the data and match it together so that when we sit the visualization tool on top of it, it’s easier for the operator, say, myself or other analysts to look at the relationships in the data they want to .
L: How do you use your data to identify opportunities for improvement?
M: This is a really great question. Healthcare professionals in general, they’re not used to using data every day in their life. They’re used to making clinical decisions based on assessing the patient looking at them. And using that real time data, I guess, of, you know, what is your blood pressure, your temperature, what are your lab results, and then making clinical decisions based on that, but we want our healthcare providers to use data on a daily basis. And so the opportunities that we find with our datasets, we have to be able to communicate that effectively to the operators, we continuously look at the opportunities in terms of length of stay patients reimbursement based on their DRG clinical documentation, and their patient experience data. And we look at that on a daily basis and say, How can we communicate the opportunities to the operator, so in a way they understand. So if we’re saying, Hey, Dr. G, we’ve noticed that your patients all have really long length of stay compared to the other physicians in your cohort. Why is that? And we can have those critical conversations with the physicians based on the data that we look at. And we look at on a daily basis, because the way we need to look at the data in the analytics and healthcare is that it is happening real time all the time every day. So we use all of those different datasets to look at the opportunities because if we can reduce length of stay, essentially, we’re going to improve patient experience and improve the patient’s clinical outcomes.
L: How do you handle situations where data privacy or security concerns arise ?
M:The good thing about data in healthcare is that it’s protected by HIPAA. And we have very, very strong parameters around that. We have probably a team of 20 people, a Jackson that constantly run security checks on whether or not our data is secure. They are testing the firewalls are testing everything all the time because you know, at Jackson, we have lots of important all of our patients are important. But we do frequently get celebrities, a lot of times people get curious and want to know what’s going on with them. So anytime a patient’s medical record is accessed anywhere in the system, our system is watching it and watching if the correct people are attached to that patient’s records
L: And what do you think are the biggest opportunities for improvement in hospital data analysis?
M: I think getting just people used to the new norm of data being part of their every day in the analytics space, we need to dedicate more resources to people that can program people that can create programs like ChatGPT that we can put on top of our datasets. Hospitals need to realize that that’s the future healthcare organizations need to realize that’s the future and they need to invest in it. Because again, we’re going to continue to be like dinosaurs if we don’t do those things. Because if you can think about, think about putting a chat GPT on top of all of our data, you could ask that chat GPT. Like, what, what does today look like in terms of our revenue, and it would be able to look at all the patients immediately immediately who’s insured who’s not insured, it will be able to tell us we’re going to lose $1 million today. Because these patients are going to be over the geometric length of stay by X amount of days, you should prioritize discharging them, the clinical documentation in this particular patient’s chart does not match what the diagnosis is go back and update that so that we get full reimbursements on that.
L: And it also saves a lot of time.
M: Oh, yeah. Like, it’s amazing. Like I said, everyone is sitting down and looking at static reports, if we can look at what’s happening real time in the world that we can’t see, right? Because we’re so big, not everybody can see everything at the same time.
L: Yeah.
M:So that’s what healthcare needs to do. We need to figure that out. I know some health systems have done it. It’s not publicized a lot yet, because I’m sure there’s like lots of controversy behind it. Because think about it. I mean, there’s controversy with all that kind of stuff, the AI telling, like they’re knowing a lot of stuff. So more to come, we’ll see. But that’s, that’s those are my thoughts on that. I const antly tell my daughters that I want them to be doctors when they grow up, because AI eventual ly will replace a lot of our jobs.
L: In any field
M: In any field. But one thing that will never be able to be replaced is a human caregiver to a person, no matter what happens, no matter what types of algorithms advanced analytics AI that we apply in healthcare, being able to make clinical decisions, a human has to be a part of that no matter what, because AI will never be able to replace a person blaming eyes on you, you could have your data could say one thing but me assessing you visually, I don’t believe it will ever be able to be replaced so that for people generations past us, I say go into healthcare and be a direct clinical operator, someone that takes care of a patient. Also AI can’t provide medication. Yeah, can’t do surgery, I mean, in some capacities, but it won’t be able to transplant multiple organs at the same time. That is my advice for that. But for people now that want to go into operations, and healthcare and advanced analytics, it would be to get a background in biostatistics to get a background in programming, you know how to code and SQL and all these different things. Because then you can write programming that will query the datasets. So you’re able to quickly grab data, and then use that data into whatever models you want to put it in. And data modelling to that would be huge. I also recommend, you know, going down the journey of getting your what we call our Greenbelt, your yellow belt, your black belt, which is in process improvement. I’m a black belt. So I used to teach people their green belts, they use the different terminologies of belts, yeah, because it originated in Japan. And for some reason, that’s what they use. But I would highly recommend getting a certification in Lean Six Sigma and process improvement because the tools in the skill set you get from that and you learn, you can apply to any industry. That would be my advice for people pursuing health care operations.
L: What advice would you give to hospitals looking to leverage their data for better outcomes ?
M:To invest in programmers that can build internal programs for your health system specifically, that will provide advanced artificial intelligence and helping you make decisions on what to change that day and in the future? No longer can we look at retrospective data, and historical data as a way of changing things. We have to look at what’s happening now. And that what’s going to happen in the future because, you know, nobody predicted COVID And we live in this world now where things are not predictable. So well not predictable to an extent. So we have to be proactive, and we have to be proactive with the money we spend with how we treat our patients. And again, you know how the world is changing so fast and what patients and their expectations of the healthcare they’re receiving. We have to look at all of that. So these are all massive opportunities that we have to stop thinking the old way. And we have to embrace change, change management, and that artificial intelligence and technology is here. And it’s going to be the way of the future. And how do we optimize that in our everyday operations, again, to provide the optimal patient experience better clinical outcomes for the patients, and essentially how to make money so that we can take care of more patients that don’t have the ability to pay for their care.
L: Honestly, I really enjoyed listening to you. Thank you so much for your time and sharing us all this important information.
M: No, thank you, and I appreciate your time and I hope the rest of your semester goes well with school.
