Toward a patient-centric, networked and holistic healthcare ecosystem

Rising healthcare needs and exploding spending

The health care system is facing major challenges to reduce spiraling costs while a demographic shift to an aging population increases the need for medical treatment. In the recent 10 year period, health care spending in the developed world has increased approximately 70%, outpacing a GDP growth of around 40%. Non-infectious chronic diseases influenced by lifestyle factors such as diet and lack of physical activity are on the rise. Obesity is considered a global epidemic by WHO and linked to the top causes of mortality: heart disease, stroke, diabetes, and certain types of cancers.

From disease treatment to outcome based healthcare

The traditional model of healthcare has focused on managing diseases. A patient with a condition visits a doctor when the burden of disease starts to cause discomfort or impact her daily life. The doctor, as a sole decision-maker, relies on incomplete data for diagnosis: mainly one-time clinical readings of typical physiological parameters, heart rate, temperature, blood pressure together with observable symptoms and subjective recall of the patient. A partial patient history, missing information about the patient’s environment and lifestyle and no consideration of an individual’s genomics lead the doctor to conjecture a treatment. Once the patient is sent home with medication or instructions for therapy or lifestyle changes there is very little or no follow-up. The doctor does not know if the patient has taken her medication regularly as prescribed or has given up early as a result of side effects or neglect. Treatments based on an average patient profile with little opportunity learning from real world patient feedback lead to sub-optimal outcomes. Payers under pressure from rising costs are reluctant to pay for ineffective treatments. Pay for outcome reimbursement models are replacing fee for service, with an associated need for more evidence to justify costs.

Consumers have lost faith in a health care system which is seen as high cost, inefficient, and inconvenient and seek non-traditional approaches to health such as lifestyle changes, diet, nutrition, mindfulness, and alternative medicines.

A patient-centric, networked and holistic model

As other industries have adapted to personalize the consumer experience through digitalization and use of multi-channel / continuous reach, the health care system must adapt from the traditional, sporadic point-to-point, doctor-to-patient relationship to a patient-centric, networked, holistic model. Collaboration among different stakeholders in the system creates a connected ecosystem of healthcare connecting patients, practitioners, payers, medical researchers, community, fitness and lifestyle coaches. New models to reduce costs and improve outcomes emphasize disease prevention and overall health management rather than simply treatment of disease.

Patients increasingly want to be involved in treatment choice use online resources to research and understand their condition and choose treatment options. In the new healthcare ecosystem the patient can connect to a variety of people and /or information sources. The doctor-patient relationship is evolving. A patient’s first line of contact with the healthcare system may even be a chatbot. Ask NHS is a free app available on iOS and android where a patient can talk through symptoms in with Olivia, a virtual health assistant. If needed, Olivia will arrange for a call back from a nurse to discuss symptoms further. Patients can also search for healthcare advice, lookup doctors and clinics, check opening times and schedule appointments.

Empowered by advanced technologies

Technologies to enable more patients to manage their own health at home, 24 hours a day, while still having access to professionals when needed, will reduce the cost of health care and improve convenience and the quality of life. New technologies provide an opportunity to close the loop, not only by continuous monitoring of patients for adherence but also engaging with patients to provide real time advice, support and to drive lifestyle changes through techniques such as gamification and cognitive behavior therapy to improve health.

 

Consumerization of health care with mobile apps, and wearables is blurring the lines between health and fitness. There has been an explosion of mobile health apps and wearable devices in recent years. In 2017 there were 325,000 mobile health apps available with over 3 billion downloads. While most of these are health and fitness apps, a large number provide novel method of diagnosing and monitoring medical conditions.

 

Traditional clinical measures of health were heart rate, blood pressure and temperature which required specific medical devices such as blood pressure cuff and thermometer. Driven by advances in technology, particularly low-cost sensors, together with advanced analytics and machine learning, a number of new digital biomarkers are emerging in research, but also launched as medical applications. Some require specific wearable sensors, others are enabled only through a mobile phones built-in accelerometer, camera or microphone.  At their special event on Sept 12, Apple announced an FDA Approved medical-grade EKG feature is integrated in its latest Apple Watch Series 4, which can detect atrial fibrillation, a condition with potentially fatal consequences. The EKG was previous available on Alivecor’s KardiaBand accessory for the watch. The Riva Digital app developed using CSEM’s optical blood pressure monitoring technology uses a mobile phones camera to measure heart beat and blood pulse wave at the fingertip and calculates blood pressure by analysis.  A number of apps are in development that use a mobile phone’s microphone to detect tuberculosis, COPD, and other respiratory diseases via cough analysis. Voice analysis can be used to detect depression, dementia, psychosis, Parkinson’s, coronary artery disease and more; saccadic or other irregular eye movements (nystagmus, tremor) can be used to detect neurodegenerative diseases such as Alzheimer’s, Parkinson’s, Huntington disease, Multiple Sclerosis. Artificial intelligence is used with the app Autism and Beyond, which analyses a child’s facial response to videos to detect early signs of autism.

Data, the new ointment for a better Healthcare

Sensors, wearables and a connected ecosystem provide opportunities for collecting vast amounts of data. Genomics is another important source of big data. Genetic testing services such as 23andMe allow users to submit saliva samples via mail for FDA approved screening of breast, ovarian and prostate cancer, Parkinson’s, Alzheimer’s, Celiac disease, lung & liver disease, dystonia, blood organ and tissue disorders. The availability of such low-cost genetic profiling services, enables extensive sets of genomic data to be collected in national programs. In the US, the NIH’s “All of US” research program has a goal of collecting data on lifestyle, environment and biology from over 1 million people. Biobank in the UK is another program with the goal of collecting data on genetics, biomarkers, demographics, environment, lifestyle activity, imaging from over 500,000 people and the Million European Genomes Alliance (MEGA) has a goal of collecting 1 million genomes by 2022.

Analytics and AI also play a major role in supporting decision making with life-saving consequences due to the improved accuracy and reduced time compared to methods relying only human analysis. Artificial Intelligence is used for multifactor screening of individual risk factors for cancers including genetics, age and family history with better prediction compared to screening of single genes, as risk is related not only to genetics, but also behavior and environment. When AI is used for rapid diagnosis of stroke through CT image analysis, it significantly improves outcomes as time to treatment is critical.

Cardiologs, a cloud-based AI system analyzes large EEG data to detect life-threatening arrhythmia Google together with Biobank performed in a study to use AI to predict cardiovascular disease by analyzing images of the back of a patient’s eye. In Copenhagen, AI voice analysis of calls to an emergency dispatching service detected heat attacks with 95% accuracy compared to 73% accuracy for dispatchers.

The availability of information through a connected ecosystem combining research data and patient’s own historical and real-word data can enable personalized medicine using advanced analytic tools. The traditional model of healthcare based on disease treatment can then change to a more holistic model of health management that relies increasingly on prevention and overall management of health and well-being to improve satisfaction, outcomes and reduce costs.