Driving Data-driven Medicine to Advance Personalized Care
Clinical trials are often designed with the participant’s everyday life in mind – trying to gather as much clinically-relevant data in a controlled manner as possible while allowing individuals to live their regular lives. However, it is true that trials are artificial environments, and participants are actively monitored to ensure their safety and to determine the efficacy of a drug in development. Study participants are expected to remain engaged throughout the trial’s duration. And in the end, the sponsor and FDA have the data needed to decide whether a drug is approved.
After all that, following commercial launch, the complexity of real-world applications takes over. Unlike the physical contract a trial participant has with a site, a patient has an implied contract with their provider – a general expectation that they’ll follow the clinical guidance. Very often, a patient’s dosing is a proxy for how engaged they are in their own care. Ultimately, so much more goes into understanding patients’ behaviors in real-life settings, and ensuring they are supported and remain engaged in their long-term care. By focusing on identifying and applying the right data, healthcare providers can aim to reduce the variability in patients’ treatments to drive more precise, personalized care.
Precise Dosing for Maximum Outcomes
There are many factors that impact the effectiveness of a treatment, such as when a patient takes their medication (before or after eating, the time of day, etc.) or how their body responds. Consider Parkinson’s Disease, for example. The timing of when someone takes their medication can fundamentally impact the drug’s efficacy. While we know this fact, determining the best possible time for a patient to medicate is more complicated. However, by integrating a smartphone-based application that leverages computer vision and artificial intelligence (AI) to remotely assess patient dosing behavior, clinicians could monitor when a patient doses, assess their response, and help a patient identify more precise dosing timing to maximize the impact of their treatment. This approach would enable more personalized treatment for patients with minimal interruption in their day-to-day lives.
Understanding Patient Patterns to Personalize Treatment Plans
Some individuals and particular patient populations experience significant challenges when it comes to taking their medication and maintaining a healthy regimen. For example, clinical trials using computer-vision confirmed dosing support have shown that people with schizophrenia often fall into three categories:
- Those who take their medications in line with protocol expectations
- Those who need frequent reminders and outreach from the site to remain on track
- Those who struggle to take their medication despite being provided reminders and assistance
As we think about how this applies to real-world populations, we may find a similar distribution of varying needs for assistance – or general resistance to adherence. By implementing monitoring for a period of time as a patient is prescribed a medication, clinicians can simultaneously deploy support programs to help engage patients who may need reminders or more frequent check-ins. And for patients who remain disengaged, clinicians have greater visibility to allow for more timely intervention. For example, prescribing a different route of administration, such as an injection, could eliminate some of the difficulties associated with more frequent self-administration. By integrating these interactive technologies, clinicians can monitor and connect with patients much more easily, allowing for increasingly personalized care.
Aspiring to the Quality of In-person Care for Chronic Condition Management
When clinicians are caring for patients in inpatient settings, they see the individual daily and can adjust medications to optimize their treatment. Care – along with any side effect or outcomes – is happening in real-time. However, when providers see patients in an outpatient setting, they may only see them once every six months. In such situations, they need to make their best judgment based on the information at hand and under the assumption that the patient will adhere to any medications prescribed.
As we seek to better manage chronic conditions, technology has the power to give visibility into a patient’s well-being between visits and understand how their adherence plays into the stability of their condition. Patients with congestive heart failure (CHF), for example, are usually elderly and have a complex medication regimen. While many will be able to manage their condition for long periods of time taking their particular combination of diuretics and heart medication, etc., having tools to monitor a provider’s overall CHF population allows them to know who is stable and when someone may have a change in their condition. By ensuring this data includes adherence information, providers are better equipped to identify a medication issue versus another underlying factor. Instead of a clinician trying a different dose because they believe the medication is not effective, they may actually already know that the patient missed several doses due to a different issue and can make more informed, intentional decisions in the patient’s care.
Shining Light Using Data-driven Technology
When it comes to understanding how people behave, data shines light into corners that would otherwise be left to assumptions – that an individual patient is taking their medication, that it’s working for them, that they’d go back to their provider if they were having a side effect. These assumptions – whether well-intentioned or not – are not driven by data. In our increasingly virtual ecosystem, patients’ health and well-being are dependent on our ability to understand behavior and deliver precise, personalized care. Ultimately, by learning from how tools are deployed in clinical trials, we can create better support systems for patients to drive improved outcomes.
To receive updated news from AiCure emailed to you, opt-in here.