Our Medical Mobile Portal App: The product development journey to straiten the relationship with our ordering clinicians
At 2019, we shipped our first mobile solution at Genomika Diagnósticos for supporting our medical experts responsible for ordering genetic tests through our patients, when they come to our lab. The release of this product culminated into several improvements at our product development process, with highs and lows and a lot of learning. In this article, I will show our journey along the development and present you some lessons and use-cases in case of you want to explore the market of laboratory medical apps.
The problem
Doctors apps are becoming more popular and changing how doctors work. It is critical for laboratory providers to engage with the medical experts, specially, at the genetic testing services, which the interaction with medical professionals and patients is mandatory.
Genetic testing is a rapidly evolving field that is being integrated into patient care in both primary and specialty care settings. Test selection and integrating results into patient management can be challenging for the physician and all persons tasked with using genetic tests and results for patient care. Effective communication between health care and laboratory service providers is crucial to this process.
In our business scenario at Brazil, once a person decides to proceed with genetic testing, a medical geneticist or a primary care doctor must order the molecular test. Usually the doctor wants to investigate the genetic factors for the patients clinical conditions, and for an improved test interpretation quality, laboratories must consider information such as the patient's clinical findings, family history and influence of race in determining the relevant mutations. The pre and postanalytical processes might be in check, if the doctor doesn't have access to our lab staff to select the proper exam or help to interpret the test results.
Many of our clinicians want to access the report before the patient in order to interpret the test results and discuss with the patient at the next genetic consultation. Others want to request priority to get the result earlier for some genetic tests from their patients or to know which test he should request based on some genes or patient's conditions.
During the discovering step, we also discussed a lot with our marketing and sales team, who concerned about the difficulty the doctors had to get the reports on-line (we have a website with a single login-password, but they often forget the password, requesting the report by email or by our business whatsapp) and some difficulties to discuss the results with our laboratory experts.
A direct communication channel between clinicians and the laboratory staff is essential for an improved quality testing and result-reporting practices. Furthermore, it builds a relationship for a long time with the lab, resulting in a continuous recommendation to your lab and remembered as excellence on genetic testing services.
At the time, our ops and marketing teams came to our IT team to discuss those challenges and a possible solution in mind: "We really need a clinical portal focused on our ordering doctors" .
But for what reasons we need to develop a doctor portal app at our laboratory? This starts our journey using design thinking strategies to deeply understand the doctor needs and desires.
Design thinking with Diamond Structure
In several of our projects we begin the process with design thinking and gathering field research to a deeper understanding of our user needs and desires. Afterward, we combine the systems and design thinking processes for evaluating, defining and developing the product.
We must consider the big picture, by understanding all the user journey and answering questions such as : what are the core features in the product, how are they connected, and do these components really solve my persona's problems. The steps of discovering, defining, developing and delivering gave us a fast turnaround , having the feedback of the our user before releasing the final app.
Surveys and Interviews
When we started the project, we began our research by formulating questions to ask our lab executive staff about their expectations, the sales, our customer relationship and technical teams about the doctors feedbacks and common requests. We kept an open mind and considered everything about the concerns and needs from the usual ordering clinician.
We also interviewed some ordering clinicians (˜= 4 people: we decided to contact the ones who more requested exams to our lab in the last six months at the time) and listened to them and learned about their pains and needs along the genetic testing journey and what they would like to get from the testing laboratory. In the pictures below, we illustrate some of our artefacts produced
Based on the results of our interview research analysis, we started to resume the data collected and come with some insights and key takeways. The compiled information was our base input to define the problem, possible solutions and the user personas and journeys.
Key Takeaways:
- Our survey state that most of the doctors don't have a close connection with our lab, only to suggest the lab as a performing testing center or request our lab to get the result from some of their patients before their next appointment.
- The doctors when they have the access to patients reports, they often forget the login and passwords, requesting the report by email or our call center.
- The guidelines from Health National Agency (ANS) to request specific genetic exams, that are covered by the insurance providers aren't not quite clear. At Brazil, to have the genetic test covered by the insurance healthcare provider, the geneticist specialist must be the requester for the exam and sometimes they get confused about which exam to request or the criteria the patient must have to get it fully covered.
- Doctors have questions and doubts about the interpretation of the result, and talks directly to CEO Bosco (he is also a doctor) or discuss to close specialists. (Most of them don't come to our lab to discuss the results).
- Health information about the patient's clinical history or the familiar history aren't handed out to our lab. Those information is required for a further investigation through the genetic analysis by our specialists, and it may compromise the interpretation of the result. When those data is missing, our lab team have to contact the doctors to get the filled request forms, which might result in delays at the exam turnarounds.
- Our survey state that most of our doctors don't know the menu of genetic exams available, which might result in possible wrong tests requests (specially gene sequencing panels) or even not requesting at all, since they don't know that there is a genetic testing that can help in the diagnosis.
- Our marketing team suggested us the possibility to bring bullet news about our genetics services , news and scientific articles related to the clinician speciality or interests.
Problem Statement
For the genetic services laboratory is to easier the clinician access to medically actionable genetic testing. It means strength the relationship with the ordering doctors, resulting to a more closeness and assistance to the patient. For the doctors, specially who are not experts in genetics, the close relationship with laboratory improves the care for their patients by having access to test information, expert interpretive assistance on tests results, guidance to optimize test selection and assistance with the collection and review of patient medical and family information.
User Personas
We created two user personas that represents our target users: the laboratory staff and the ordering doctor.
Our roadmap
Our roadmap was split into five major releases distributed along one year. The core features were grouped and sorted by card sorting sessions with the product owner and the lab executive staff and ranked using Gusto's product prioritisation model . This framework helps make product decisions by comparing features as "Apple to Oranges", or product decisions with vastly different goals.
These core features organised by groups are presented at the figure as follows. Some of these features were implemented until the time of this post.
This is the step when the magic happens. In this step we focus on several usability and validations steps, and the development and integration between the frontend and backend. Through the product development we faced some technical challenges. To develop a mobile application that could run on both platforms: Android and Apple, we needed to select a framework that facilitates the life of design/front-end team, specially when the time and budget is a constraint.
After some POCs (proof-of-concepts demos) we came to React Native, a compelling platform for mobile app development, based on JavaScript framework that uses a similar design as React, which provided access to the corresponding native mobile device APIs. This means applications built with the technology will have access to native platform features like camera, user location, push notifications , etc.
Our productivity increased because we could design and build the entire front-end using mobile web framework, and react native handles the shared codebase cross-compiling to Android and Apple platforms. In overall, React Native enabled quick prototyping and a very high initial development velocity. At our server-side we had all our results, authentication and newsletters backend provided by RESTful APIs with Django + Django Rest Framework.
But implementing the basic features was easy, the main issues happened we needed to extend with native code and native views.
Our data architecture flow needed to be inverted as we added functionalities which required third-party integration such as the Freshchat plugin, Google firebase cloud messaging (for push notifications), etc. This resulted in less redundancy in several components that needed to be reimplemented for each new integration.
Deploying your application into Apple Store and Google Play Store isn't the last step for your delivery. It's now that you need to learn about your users and their flow at your app. For monitoring we had implemented integrations with the Google Firebase Analytics, Google Analytics and Sentry (for backend errors).
Essential features of our mobile doctor portal app
As we were understanding more about the user flows and areas that the user didn't access, we performed several refinements at our design and at our scalability. We faced the moment that one doctor with more than 500 patient records, decided to access one of their first patients records. Improvements such as caching, pagination and search strategies were performed to improve the real time experience from our app, removing any glitches.
We will present some essential features of our mobile doctor app. Let's find out more details at the animated gif below.
Pending documentation or priority request
Here, each doctor can ask the priority of one of your patients exams directly from the app. Our staff must approve the priority in order to take in action. We also have an alert for the physician to have informed that there is pending documentation such as clinical request form or the patient's history. The goal is to alert the doctor so he can provide as fast as possible any missing information to get the genetic analysis going on.
In-app messages
Our doctor portal app requires a two-sided communication. Doctors can also use built-in chat options to communicate with our laboratory experts. . We needed to pay attention to the secure of data, that's why we used a third-party CRM solution, Freshchat, a real-time chat combined with a ticket management system managed by our geneticists.
Our Genomika Journal
Our monthly feed curated with several news and articles about the medical genetics and its advances, all condensed in a beautiful designed newsletter delivered by email and at our app. Our hypothesis is to give the relevant content about our laboratory fresh exams and genetic advances in the areas of interest of the doctor.
Our Results DashBoard
One of the our neat features that we delivered to our ordering doctors interested on research. Many of them would like to have some summaries of their data (patients records) in order to know how many patients were positive, negative for a specific genetic test. Other information was the number of VUS, pathogenic or benign variants for all their patients requested at Genomika. All those insights were processed at our backend and delivered to our users.
Concluding Remarks
Our medical app for ordering doctors at the time of this post is on-line with several active doctors. Our metric for DAU/MAU is only 1.3% . But there is a reason for this small number. Our app is more reactive, it means that the moment that he decides to open the app is when the notification of a new report or journal, or an interaction with our team to ask a questions about results and tests.
DAU/MAU is a popular metric for user engagement — it’s the ratio of your daily active users over your monthly active users, expressed as a percentage. Usually apps over 20% are said to be good, and 50%+ is world class.
We are studying novel interactions with the doctor based on specific alerts about their patients results. Imagine for instance, our variants database monitoring tool, that could notify a change in the classification of an important mutation/variant (VUS to pathogenic) that impacts the result of the patient. The doctor can take the action and call to the patient to request a consultation: the inversion of values (doctors becoming pre-active).
For genetic testing laboratories is critical their engagement with the patients and with respective ordering clinicians. Solutions that tighten this gap in communication, can assist not only the doctors in their patient care, but also their lab staff in daily routines, reducing the turnaround of the exam because the lack of information or assist the doctor in the test select/ interpretation of the result.
In our study case, our med app was started as solution to to solve the problem of the common forgetting the password at our on-line portal and a direct access to the results, and became a aggregated app that offers several elements to help the physicians during all stages of the genetic testing journey.
In the perspective of design and backend team, pay attention to the selection of the framework that is compatible to your current scenario (in our case a team of only three people with limited budget). Don't overestimate the SDKs of the Apple/Android SDKs (they change a lot and you give headaches to fix to get back the app working again). Invest in POCs and A/B and usabilities tests , don't overlook the importance of the design process. They are critical for understanding your target user and his real pains.
Special Thanks
There are many people that we would like to thank along the mobile app development journey. Some people we would like to mention:
- Our CEO, Doctor PhD João Bosco Oliveira, who supported us and believed in the problem statement of relationship between lab and clinicians;
- Our COO, Diana Meira, who helped us with several insights including the creation of the Genomika Journal.
- Our Marketing team who helped us with all the visual identity and marketing and advertising stuff.
- All the stakeholders : all interviewed ordering clinicians, our lab staff (customers relationship, sales and technical).
- Our Lead Designer and Front-End Developer Dyego Carlos, who designed and leaded the design thinking process of this amazing app.
- Our IT development team represented by Rheno Formiga and Marcelo Coelho.
