Exploring social media platforms like Facebook, Twitter and LinkedIn, amongst others to collect Real-World Evidence (RWE) is set to take the healthcare industry by storm. Being perceived as a rapid, inexpensive way of assembling vast amounts of data reporting patients’ real experiences of treatment, including drug efficacy, safety and societal impact, is attracting pharma giants to this source. However, its seemingly unedited constructs might not seem like a logical place to put one’s trust. Dr. Dee Amanze in this blog explores the pros and cons of using Social Media Data as a RWE.
In the not too distant past, healthcare was delivered and evaluated by an elite group of doctors and other professionals as the patients quietly accepted whatever the treatment options and outcomes were. Today, the industry is rapidly evolving towards a consumer base of informed, engaged and empowered patients. Social media has also become a platform where large numbers of patients can freely express themselves, uninhibited and uncensored. Whether this new trend of patients freely expressing themselves in social media has any benefits in Real World Evidence (RWE) depends on what perspectives from which one is looking at the situation. One thing is certain though – the trend is very likely to grow exponentially in the near future.
Real World Data (RWD) has become a valuable source of decision-making insight in many aspects of RWE. The traditional sources of RWD today are patient-level claims data, pharmacy transactions, laboratory test results, payer records, Electronic Medical Records (EMR) and Electronic Health Records (EHR). Other more recent sources of RWD include data from the various groups of ‘-Omics’ (genome, proteins, metabolites etc.) and healthcare ‘Wearable’ devices. The value of these sources of data has come primarily from our ability to link de-identified patients across these multiple disparate sources of ‘Big Data’. Now comes social media as a potential source of RWD!
The patient is the primary source and an integral part of most of Healthcare Big Data. However, the reality is that Providers, Payers and the Pharma partners appear to be the primary consumers and beneficiaries of most of the insight derived from Big Data analytics. Consequently, RWE has largely been seen from the perspective of these stakeholders. In most cases, the needs and expectations of patients are considered almost as an after-thought rather that the ultimate goal of healthcare in general and RWE in particular.
Benefits of Social Media Data
Social Media, as a potential source of healthcare RWD, is unique in being the main source of data that reflects the patient’s perspective. It has become an effective platform where average patients can make their voices heard.
Challenges in Integrating Social Media Data (SMD) in RWD
Proper integration and analysis of SMD calls for assessment and understanding of the challenges and characteristics of patient sentiments. Some of the top challenges include:
- Data is in an unstructured, free-text format and typically requires machine learning capabilities or manual abstraction
- SMD cannot easily be cross-linked to other datasets in RWD. Indeed, experiences and sentiments reported in social media may be those of the patient or those of family members, friends and/or associates thus making it impossible to cross-link to a particular patient
- Currently, there are no nationally available data repositories for SMD
- Another major challenge in integrating SMD to healthcare RWD, is the lack of clarity on exactly what information needs to be collected and knowing how best to integrate such information with other datasets
- There are often restrictions and limitations in mining a number of social media sources
- There are also challenges with data validation and source verification of data
So, value or distraction
With due consideration to the pros and cons described in this article, one must ask whether integrating SMD into RWE is a value proposition or more of a distraction.
Perhaps, all that is required to effectively integrate SMD into RWE is a new approach to data gathering and ingestion. In my opinion, successful integration of SMD into RWE will require new analytics tools such as Machine Learning (ML), Natural Language Processing (NLP) and a source-agnostic analytics platform.. The ultimate goal is integration of all available data sources that enable various stakeholders gain more insight into direct patient sentiments. The journey to integrating SMD into RWE might not be an easy path but companies taking this plunge the fastest will have the most distinct and perhaps the most enduring competitive advantage.
There lies the potential for Big Data and Analytics companies like Saama to take the initiative in incorporating aspects of data from social media to obtain decision-making insight into areas such as patient sentiments towards healthcare services received and for comparative analysis of provider performance.