Cell and gene therapies are becoming more and more popular because of encouraging clinical results worldwide. Major pharma manufacturing companies have invested in the concept’s commercialization worldwide. Recently, we read about Takeda’s license for commercialization of Aloficel (developed by TiGenix), Celgene’s acquisition of Juno Therapeutics or Gilead’s acquisition of Kite Pharma.Continue reading “Data Analytics for cell and gene therapy by Dr. Ruchi Dass, @drruchibhatt”
DIGITAL TRANSFORMATION AND THE PLACE FOR DATA
For the sake of clarity, when I say data, I will be referring to digital data throughout this piece. Data is a by-product of any activity. Therefore, creating data is as natural as breathing. So we have data. A lot of data. So what? Accumulating data, structuring it, storing it, analyzing it are a natural progression from that point onwards. How and what we do with the data is more important. Software.
The magic that is software, to me, is more transfixing. Consider the prospect of a language written in a semantic that is alien to our natural human language. A cryptic command, logic, condition, trigger – anything at all – that is magically read, understood and acted upon by silicon. Hardware that contains baked-in code that can parse and carry out complex instructions at blazing speeds. Pieces of such chips soldered on a board and communicating through ‘roadways’ of circuits laid out on a board. The miracle of hardware coupled with the magic that is software is what gets my adrenalin pumping. How can such a marvel not be exciting?
Even the awesomeness of hardware pales in comparison to software. Hardware is more or less static. It is confined to physical and functional dimensions. Software, however, is supreme. It can use the same hardware (with some limitations of course) and carry out simple tasks, entertain with games, or perform wildly complex calculations at very very high rates of speed, accurately all the time. And it can do this million million times with alacrity. This is just the beginning of what software can do. But wait, there’s more!
Consider intelligence in software. It suddenly becomes a living, breathing, dynamic being. Almost. Software can learn and teach itself. Crunching data and spitting out patterns and actionable analysis suddenly becomes mundane, banal almost pedestrian. No. I am not against data or big data. By itself, big data is just that. A monstrosity. Sometimes, big data actually gets in the way. Misleads us in making decisions quickly. Software breathes life into data.
Take any software language or tool. Examine it. Study its flow, the eloquence, the nuance and its brilliance. Brevity in software coding is revered by programming perfectionists. There is elegance in a well-written piece of code that executes beautifully, perfectly, every time. Anyone that can find literary melody in Shakespeare or Milton can certainly begin to enjoy the harmony in a beautifully crafted software application code. So, my appeal goes out to all those who are worshipping big data to take a moment to reflect upon the joy that software brings to our daily lives. After all, the future is software!
Q1. Nishita Mehta: What is data’s role in healthcare & how do you see it influencing future health sector growth in India?
A. Inder Davalur:
Q2. Nishita Mehta: What are the unique challenges of working with clinical data?
A. Inder Davalur:
Q3. Nishita Mehta: Healthcare seems to be moving from the use of structured data to unstructured data. What is the difference between them when it comes to clinical utility & improving patient outcomes?
A. Inder Davalur:
Q4. Nishita Mehta:What do you think does a hospital need to implement Big Data solutions, i.e. Big Data Analytics Capabilities in terms of infrastructure and personnel requirement?
A. Inder Davalur:
One of the major challenges a CIO or an IT head faces in a hospital is the lack of budget allocation for anything beyond the basic networking, computing and storage needs. Hospitals do not see the value in the data they currently possess most likely because they are more empirically driven rather than evidence driven. What this means is that hospitals and by nature the doctors who hold a sway over management decisions are more comfortable with their own decades of experience over some hotshot CIO trying to promote the idea of data mining and predictive value of patient outcomes based on past data. There is also a severe shortage of technology-rich personnel in hospitals due to the dull routine of maintenance of existing hardware and software rather than experimenting with new technology. The pay structure for IT personnel in hospitals is also woefully poor in comparison to the technology companies. All these factors combine to deter anyone who is driven to create in hospitals a digital core
Q5. Nishita Mehta: While Big Data can generate a plethora of interesting patterns or hypotheses, there is still a need of experts to analyze the results to confirm whether they make sense or merit further inquiry. Would you like to comment on this?
A. Inder Davalur:
Absolutely. Right now, there is a paucity of people with skills to interpret and recommend action plans once an organization implements any sort of analytics software. Unlike other verticals, healthcare is lagging far behind in its focus on data interpretation and application in its business model. It might be a whole decade before hospitals wake up to the reality of meaningful interpretation of data and building an action plan around it
Q11. Nishita Mehta: Do most doctors now have a checklist for what they should be doing with patients with certain conditions? How does Big Data solution change what they are doing currently?
A framework developed by Robert Kaplan and David Norton that suggests four perspectives of performance measurement to provide a comprehensive view of an organisation. These are service user perspective, internal management perspective, continuous improvement perspective and financial perspective.
A point of reference or standard by which something can be measured
Casemix is an internationally recognised system of measuring clinical activity incorporating the age, gender and health status of the population served by an organisation with a view to objective determination of hospital reimbursement.
Data are numbers, symbols, words, images, graphics that have yet to be organised or analysed
A descriptive list of names (also called representations or displays), definitions, and attributes of data elements to be collected in an information system or database.
A unit of data for which the definition, identification, representation, and permissible values are
DOMAINS OF QUALITY:
Are those definable, preferably measurable and actionable, attributes of the system that are related to its functioning to maintain, restore or improve health
As per gartner, Descriptive Analytics is the examination of data or content, usually manually performed, to answer the question “What happened?” (or What is happening?), characterized by traditional business intelligence (BI) and visualizations such as pie charts, bar charts, line graphs, tables, or generated narratives.
As per gartner, Predictive analytics describes any approach to data mining with four attributes:
1. An emphasis on prediction (rather than description, classification or clustering)
2. Rapid analysis measured in hours or days (rather than the stereotypical months of traditional data mining)
3. An emphasis on the business relevance of the resulting insights (no ivory tower analyses)
4. (increasingly) An emphasis on ease of use, thus making the tools accessible to business users.
As per Gartner, Prescriptive Analytics is a form of advanced analytics which examines data or content to answer the question “What should be done?” or “What can we do to make _______ happen?”, and is characterized by techniques such as graph analysis, simulation, complex event processing, neural networks, recommendation engines, heuristics, and machine learning.
Health Information is defined as information, recorded in any form or medium, which is created or communicated by an organisation or individual relating to the past, present or future, physical or mental health or social care of an individual or cohort. It also includes information relating to the management of the health and social care system
KPI SELECTION CRITERIA
Data that defines and describes other data
MINIMUM DATA SET:
The minimum set of data elements that are required to be collected for a specific purpose
The specifications that define the subset of data items in the denominator that meet the indicator criteria.
KEY PERFORMANCE INDICATORS:
Performance Indicators are specific and measurable elements of practice that can be used to assess quality of care. Indicators are quantitative measures of structures, processes or outcomes that may be correlated with the quality of care delivered by the healthcare system.
Performance indicators that monitor the activities carried out in the assessment/diagnosis and treatment of service users.
Performance indicators that monitor the desired states resulting from care processes, which may include reduction in morbidity and mortality, and improvement in the quality of life.
Reliability is the consistency of your measurement, or the degree to which an instrument measures the same way each time it is used under the same condition with the same subjects.
Performance indicators that monitor the attributes of the health system that contribute to its ability to meet the healthcare needs of the population.
The Delphi Technique:
The Delphi technique is a facilitated structured process whereby a panel of experts complete questionnaires (see Appendix 3 for example) remotely and, through feedback and scoring over a number of rounds where some KPIs are discarded, a consensus is achieved on a final set of KPIs. The panel need not ever meet face-to-face and each individual’s feedback is provided anonymously to the panel, which eliminates the possibility of undue influence by dominant personalities within the panel.
The RAND appropriateness method:
The RAND appropriateness method combines scientific evidence with expert
opinion by facilitating experts to rate, discuss and re-rate KPIs. Unlike the Delphi technique the expert panel meet face-to-face to discuss possible KPIs and are given a copy of the scientific literature in support of the KPIs so that they can ground their opinion on evidence-based literature
Validity of indicators refers to whether performance indicators are measuring what they are supposed to measure. e are constantly looking for Healthcare Informatics & Digital Health Experts to share their experiences by writing articles on Technology benefiting in the delivery of Healthcare Services.
My takeaway from that conversation were the two keywords ‘facts‘ and ‘fingertips‘! For running a successful organization, you do always need to have near real-time relevant and critical (may be up to ten, one for each fingertip!) facts on what is happening within the company. However, just the facts (measures) may not always be sufficient to arrive at a decision unless those are benchmarked against the desired performance and/or trends over different periods for those measures. Deployment of analytics enables the stakeholders to have that additional edge over the decision-making, by making the exercise based more on validated data than just a gut feeling.
Meeting the patient expectations
... and timely …
… and fairly …
… and thus earning patients’ trust!
Managing the hospital operations efficiently
necessary and sufficient?
Monitoring the financial health of the hospital
Where does the hospital stand when compared with its peers?
Note: A version of this article also appears in my blog gyaan-alytics and more…