Analytics & Social media impact on Healthcare

CIO — BOSTON—The increasing digitization of healthcare data means that organizations often add terabytes' worth of patient records to data centers annually.

At the moment, much of that unstructured data sits unused, having been retained largely (if not solely) for regulatory purposes. However, as speakers at the inaugural Medical Informatics World conference suggest, a little bit of data analytics know-how can go a long way.

 

It isn't easy, namely because the demand for healthcare IT skills far outpaces the supply of workers able to fill job openings, but a better grasp of that data means knowing more about individual patients as well as large groups of them and knowing how to use that information to provide better, more efficient and less expensive care.

Feature: 13 Healthcare IT Trends and Predictions for 2013

Here are six real-world examples of how healthcare can use big data analytics.

1. Ditch the Cookbook, Move to Evidence-Based Medicine

Cookbook medicine refers to the practice of applying the same battery of tests to all patients who come into the emergency department with similar symptoms. This is efficient, but it's rarely effective. As Dr. Leana Wan, an ED physician and co-author of When Doctors Don't Listen, puts it, "Having our patient be 'ruled out' for a heart attack while he has gallstone pain doesn't help anyone."

Dr. John Halamka, CIO at Boston's Beth Israel Deaconess Medical Center, says access to patient data—even from competing institutions—helps caregivers take an evidence-based approach to medicine. To that end, Beth Israel is rolling out a smartphone app that uses a Web-based- drag-and-drop UI to give caregivers self-service access to 200 million data points about 2 million patients.

Analysis: Is Healthcare IT Interoperability (Almost) Here?

Admittedly, the health information exchange process necessary for getting that patient data isn't easy, Halamka says. Even when data's in hand, analytics can be complicated; what one electronic health record (EHR) system calls "high blood pressure" a second may call "elevated blood pressure" and a third "hypertension." To combat this, Beth Israel is encoding physician notes using the SNOMED CT standard. In addition to the benefit of standardization, using SNOMED CT makes data more searchable, which aids the research query process.

 

Continue Reading

The Bipartisan Policy Center, Heritage Provider Network, and the Advisory Board Company have announced a new national competition to develop ways to use big data to transform care delivery and solve some of the most pressing problems facing healthcare providers today. 

“Hospitals, health plans, physician practices, and post-acute care providers are being asked to provide higher quality care while lowering costs,” says the Care Transformation Prize website. “Successfully doing so requires access to and analysis of large data sets to predict, identify interventions for, and assess cost and quality outcomes for patient populations. Most health care organizations, however, have little knowledge or expertise on how to leverage and analyze the clinical data sets now being developed as the result of an increasingly digitized U.S. health care system.”The program will offer at least three quarterly prizes of $100,000 to the teams that develop the best solutions to selected challenges, including data analytics and data use.  The three as yet unannounced questions will be posed and answered over the next sixteen months.  Interested teams can register for the challenge here.

Mining the records of routine interactions between patients and their care providers can detect drug side effects a couple of years before an official alert from the U.S. Food and Drug Administration, a Stanford University School of Medicine study has found. 

The study, led by Nigam Shah, MBBS, PhD, assistant professor of medicine, was published online April 10 in Nature Clinical Pharmacology and Therapeutics.

This approach is a step forward in mining patient-based information, as opposed to coded insurance reports or drug-specific databases, to improve health-care strategies, said engineering research associate Paea LePendu, PhD, the lead author of the paper. The technique is intended to complement the FDA’s Adverse Event Reporting System, which has compiled reports of medication side effects from patients, physicians and pharmaceutical manufacturers since 1968.

Clinical notes include the information a caregiver dictates into a patient’s record, such as the patient’s symptoms or medical issues. It would also include what a doctor advises or prescribes for the patient.

“If you ask any audience related to health care how much of the clinical knowledge is bundled up in text, you won’t get an answer below 70 percent,” said Shah. “If 70 to 80 percent of the data is locked up in text notes, we asked ourselves, ‘What would be a good way to unlock it?’” Their approach builds on recently published work that developed a gold standard for assessing the performance of data-mining methods.

RELATED NEWS» Mining consumers’ web searches can reveal unreported side effects of drugs, researchers say» Dangerous side effect of common drug combination discovered by data mining

The information gleaned is intended to support current protocols or clinical experience. Shah and LePendu see their work as a move toward a learning health system, in which we learn from the day-to-day experience and the collective wisdom of the decisions that doctors make when treating patients. They believe health-care providers can apply such data mining to clinical data warehouses to create a new source of evidence — practice-based evidence — for patient care. 

Although their application is new, their information-gathering methods are based on well-established text processing techniques. It’s also simpler and faster than current strategies used in the same arena, said LePendu. Content is first grouped via “ontologies,” which are information graphs organized by associative relationships instead of a rigid linear structure. For example, melanoma is a kind of skin cancer, and so is Kaposi’s sarcoma; by knowing “skin cancer” encompasses both kinds of cancer, the search process picks up this medical knowledge. The system also de-identifies patient information in the process, so sensitive data, such as names and addresses, doesn’t get revealed. With these methods, LePendu said, the technique allows them to process 11 million clinical notes in about seven hours on hardware no different from a laptop computer — a pace that other programs can’t match. 

The information is also current: It’s generated from what is observed and recorded in the hospital or doctor’s office. That’s an advantage over the FDA's AERS reports, which rely on patients and health providers to make the additional effort to report adverse events.

The researchers developed the computerized method to sift through the contents of clinical notes in electronic medical records and used it to examine how often specific drugs and diseases were mentioned in roughly 10 million notes for about 1.8 million patients over 15 years. The goal was to organize these notes into a data-mining substrate they refer to as a patient-feature matrix. “Everyone is excited about the prospect of ‘big data’ mining on electronic health record data,” Shah said. “We demonstrate it in practice.”

Although clinical notes provide an excellent source of untapped information, this mining technique does have limitations. It requires a big database to extract accurate trends, and the volume of information the system sorts through makes it more useful for looking at common events, such as heart attacks, said Shah. He added that the FDA reporting system is probably still superior for looking at rare problems, which wouldn’t occur in high enough volume at any single institution. Also, the system can’t evaluate adverse drug reactions that are dose dependent. 

But, the research team is working on refinements that will cull even more useful information from clinical notes, such as reports of reactions caused by drug combinations, the use of medications typically prescribed for one condition but found effective for treatment of a different health problem, or finding medical profiles of patients that fit a certain scenario.

“This method is exciting, and it raises the possibility that mining clinical notes can augment traditional pharmacovigilance monitoring,” said Steve Goodman, MD, PhD, associate dean of clinical and translation research for the medical school who also co-chaired the 2012 Institute of Medicine committee that studied the safety of approved drugs. Goodman was not involved in the research. “It also takes advantage of electronic health records which are already there.”

One downside is that most electronic health record systems are set up for patient care, not patient research, Goodman noted. In this study, the researchers mined a data system created for this kind of research, which isn’t widely available. The researchers used the Stanford Translational Research Integrated Database Environment, known as STRIDE.

Other Stanford co-authors were associate chief information officer Todd Ferris, MD; engineering research associate Rave Harpaz, PhD; postdoctoral scholar Anna Bauer-Mehren, PhD; and STRIDE clinical data warehouse curator Tanya Podchiyska.

LePendu, Iyer and Shah are inventors of technology, owned by Stanford, for generating the patient-feature matrix.

The study was funded by grants from the National Institutes of Health (grant HG004028) for theNational Center for Biomedical Ontology. 

Information about Stanford’s Department of Medicine, which also supported the research, is available at http://medicine.stanford.edu.

The push by the government to reduce healthcare costs and the increased liability providers have is forcing them to more easily identify and help chronic care patients to better manage their conditions. At the same time healthcare IT vendors are expanding their big data armories to help providers, particularly accountable care organizations mine claims and clinical data to get a better sense of patient outcomes, performance and how and where they can reduce costs. As more providers convert from paper to electronic records they are working with health IT vendors that can help them produce more accurate assessment of their patient populations to mine patient data to help predict outcomes.

At HIMSS earlier this month Dr. Anil Jain, the CMIO of Explorys, a spinout from the Cleveland Clinic, highlighted some of the different analytics approaches it is offering clients as they get more involved in population health.

Descriptive Analytics This accounts for the biggest chunk of big data across industries and it tends to focus on what went wrong or assessing why outcomes are more or less than what was expected. “Most people are pretty well covered when you think of descriptive analytics,” says Jain. One example of descriptive analytics is giving hospitals a better understanding of current assessments, like how many of its patients should have received a pneumococcal vaccine or how many diabetes patients in an endocrinology department have their blood sugar under control?

Predictive AnalyticsBig data is chiefly being used to identify patterns, predict how to predict future outcomes, and avoid preventable events as a way to reduce healthcare costs. Jain says the most frequently asked question, particularly from accountable care organizations is, “‘What percent of our patients will be re-admitted?’ They are also looking at how many patients will use the emergency room.” Explorys’ big data platform includes a tool that can score patients based on their risk profile, such as whether they have chronic conditions, so providers can develop more effective approaches to care.

Prescriptive Analytics One of the most noticeable trends at HIMSS this year was the increasing interest in prescriptive analytics. A recent report from Gartner looking across business intelligence said that only 13 percent of organizations are using predictive analytic but even fewer — 3 percent — are using prescriptive analytics, so there is plenty of opportunity for growth and the demand is increasing. Prescriptive analytics involves helping a provider measure and manage a patient population. For example, one tool from Explorys’ big data platform allows users to focus on patients with obesity, add a morbidty like diabetes and assess their LDL levels or other measurement to determine where they need to focus attention.

“When you have an ACO that is trying to change the cost curve it is about good data but once the data is in, most providers look at the computer screen and try to figure out what the focus should be.”

Jain likens shifting from a descriptive to prescriptive data analytics platform to the equivalent of going from a broad, fluorescent light to a laser beam focus. “We don’t bring customers on if they are not ready to address population health as a solution. Provider groups have the same end-goal in mind: How do we stay relevant as pay-for-performance models change?”

Comparative Analytics One of the most interesting ways providers can use big data is to compare their performance to other healthcare facilities. Explorys expanded into the comparative analytics market this year with its National Benchmarks platform. The platform uses comparative metrics throughout more than 92 billion clinical, financial, and operational data sets across a continuum of care. By combining clinical data with claims and administrative data, it gives insights into patterns and trends. Providers can compare their performance with a particular patient population compared with the aggregate network, made up of providers such as the Cleveland Clinic, St Joseph Health System and Legacy Health. Patient information is de-identified and made HIPAA compliant while keeping participating providers private. For example, providers can see how the LDL levels of their patients compares with that of the network and can use different sets of criteria across age, race, geography and gender. Providers can use the information to develop insights to improve performance.

 

The separation, or fragmentation, of data among labs, hospital systems, financial IT systems and EHRs, is another significant obstacle to leveraging big data in health care. Each entity serves as a single repository, or silo, for information whose purpose is to provide clinical care, scheduling or billing information, or operational information. This continues to be problematic for organizations seeking to get individual systems to communicate with each other easily. It remains especially challenging in smaller organizations with multiple systems and taxonomies that make extracting useful information difficult for data mapping.

If big data is to have a big impact on the healthcare industry, namely in improvements to care delivery and health outcomes for individuals and entire populations of patients, then the industry needs to first address how it can implement, adopt, and use EHR and other health IT systems more effectively to support healthcare big data.

The rate of EHR adoption and usability of data contained in current EHR and health IT systems contribute to the healthcare industry’s lack of preparedness to benefit from the use of big data and healthcare analytics, according to new research by the Institute for Health Technology Transformation (iHT2). “Although EHR use has grown from about 20 percent of providers to some 60 percent in 2012, U.S. health care remains predominantly a paper-based system. It will take significant effort to shift attitudes and educate providers about available and emerging technologies,” note the authors of “Transforming Health Care Through Big Data.”

 

Even for those organizations that have successfully adopting EHR and health IT systems, the ability to manage large amounts of data lags significantly behind their ability to capture health information. “Health care organizations are accumulating 85 percent more data than they did two years ago, but 77 percent of health care executives give their organizations a ‘C’ or below for managing their data,” explain the authors, citing a Oracle survey of American and Canadian C-level executives. “Of health care executives interviewed, none gave their organizations an ‘A’ for data “preparedness.’”And even methods for capturing patient data showed weaknesses, the authors observe:

Moreover, despite the high priority they place on implementing EHRs, health care leaders report their organizations are struggling to leverage them: while 34 percent reported being able to capture data from EHRs to help patients, 43 percent said they were unable to collect sufficient data to improve care.

Two obstacles to leveraging these captured data and the big data tools for analyzing them are the structure of the data and how this information is stored.Of the first, the authors draw attention to the unstructured content comprising an important part of any EHR. “Most clinical data is stored in ‘unstructured’ form, especially within EHRs, making it difficult to access for effective analytics,” they continue. “For example, while individual physicians can read narrative text within an EHR, most current analytics applications cannot effectively utilize this unstructured data.”Being largely unable to access this kind of information, the healthcare industry has come to rely on claims or administrative data, which are too far removed from the point of care to influence patient and population outcomes. As the researchers emphasize and our recent interview with Health Fidelity CEO Dan Riskin, MD, confirms, new health IT tools are emerging to tap into value trapped within unstructured data such as free text.Of the second, the authors indicate the need to normalize health data into a central repository where big data solutions can be applied and prove valuable:

 

IBM, WellPoint and Memorial Sloan-Kettering Cancer Center are collaborating on a project using Watson as a means to develop an oncology treatment tool. Once imagined ideas regarding the progress of healthcare are quickly becoming a part of the real world. A majority of the value from Watson comes from its ability to aggregate massive amounts of medical journals. Declared a ‘clinical decision support system’, it will operate in collaboration with physicians and specialists.

Although it is not taking over the entire doctor patient relationship, Watson is offering valuable data driven insight for medical professionals.
Big Data analytics have been brought to the forefront of popular culture thanks to IBM’s Watson. The advanced computer is a symbol of Big Data’s proliferation and the future of data driven decision making in healthcare, business and government. For instance, President Obama’s 2012 reelection team boasted a 100 person analytics department to analyze voters, with its own internal 12 member team to conduct analysis regarding their operations.

The beauty of Watson, rests in its ability to analyze millions of pages of journals containing unstructured natural language data in a rapid fashion. Analytics for unstructured data is the most challenging aspect of data analysis, Big Data analytics however are beginning to solve this challenge. With the progression toward electronic medical and health records, analytics will become more practical and have added use cases.

Hospitals in control of their patient data can conduct powerful analysis through technology similar to Watson, to improve future decision making. By creating a data friendly environment within an enterprise like a hospital, data will become less of a burden and more of a resource. Having the ability to assess the past 20 years of treatment in the context of real-time patient data, will undoubtedly influence future scenarios and treatment.

One instance of practical analytics usage at University of Pittsburgh Medical Center (UPMC) occurred when the new IT team used analytics to highlight patients with propensity to staph infection. Staph or MRSA, is a contagious bacterium that can wreak havoc on a patient’s fragile immune system and even cause death due to infection.

By conducting analytics to create a list of likely staph patients, the provider was able to specially cater to those patients in a way that prevented infection. Also applying analytics to medication procedures has the potential to drastically reduce mistakes in drug administration.

Watson is a great step forward for Big Data analytics. As the implementation of Watson within Sloan Kettering progresses, it will be exciting to see the improvements in treatment and the rate at which quality of care increases. It is highly likely that commodity versions of Watson will be made available to general practitioners within their individual practice (and eventually for consumers on their smartphone).

IBM stresses that their brain-child is much more than an advanced search engine. Its ability to process such massive amounts of unstructured data and provide contextual insight for decision makers puts it squarely in the territory of Big Data.

Healthcare organizations and providers are maturing in their ability to use clinical intelligence as a means to improve the care of patients, the business of providing care, and the process of reporting clinically-relevant medical information to public health agencies and other organizations charged with managing the health of whole populations. “The promise of meaningful use is that this data is going to be available for them to manage care, improve quality, and reduce cost,” says John McInally, former CIO and current Partner of Healthcare Big Dataand Analytics Group CSC, in an interview prior to HIMSS13.

Harnessing the power of big data and healthcare analytics, however, first requires that clinical and patient information is documented and captured appropriately. And only recently have healthcare organizations and providers begun to understand the value of big data and analytics to their clinical and business initiatives.“The marketplace in healthcare really wasn’t ready last year,” recalls McInally. “People were just struggling to get their data warehouses up and be able to do meaningful use reporting — that was the thing that was all-consuming last year. Then this new thing came along which was accountable care and readmission risk, and suddenly the answers that we wanted to come from our traditional data warehouses weren’t there.”The new level of accountability for providers requires that they have a comprehensive and accurate picture of a patient’s care comprising both structured (e.g., demographics) and unstructured data (e.g., physician notes), which is only possible through a new approach to data mining.“In a traditional data warehouse, it’s entirely structured data, and you have to have a report writer and some pretty sophisticated thinking behind your questions,” explains McInally. “What we’re doing in big data is applying analytics tools that come from other segments in the industry and allowing the merge of text and unstructured data with structured data and not only answer the question but also expose other questions you may not have thought of.”In short, big data is the synthesis of data irrespective of the form it takes. That being said, its strength is based on its ability to incorporate components of the patient experience that fit neatly into drop-down menus, checklists, or form fields, with the most important of these being physician notes in free text.Coupled with that is the capacity to help both the individual patient and patient population. “We begin to stitch together a community of data, structured and unstructured, and we’re able to very quickly using these non-SQL tools answer big questions across large population and we’re able to personalize it down to the individual patient in a bed at the same time,” adds McInally.Although the healthcare industry has been slow to adopt the concept of big data, nearly all healthcare reform via health IT is looking to this source of information for insight, from accountable care to clinical decision support. In order for a provider to have access to this kind of support, his organization must have the necessary infrastructure in place (i.e., data warehouse) to support the merging and parsing of patient data.“A lot of the answers that you’re looking for today don’t come out of a traditional relational database management system or if they’re possible, they just don’t come out fast enough,” argues McInally. “The notion here is to be able to mine the data so that the data itself exposes a new hypothesis for you, and that’s the horizon we’re taking people out on.”With the close of Stage 1 Meaningful Use for early adopters and demonstrated users of certified EHR technology, an ample amount of data is now residing in data warehouses. Next in line is the need to realize the use of this valuable information in the form of primary, secondary, and tertiary applications.Related White Papers:White Paper: Achieving Stage 2 Meaningful Use in Private PracticeStage 1 Meaningful Use ChecklistList of Health Information Exchanges (HIEs)Complete directory of Accountable Care Organizations (ACOs)Webcast: Moving from Stage 1 to Stage 2 Meaningful UseBrowse all White Papers

Big Data analytics can take healthcare to a new level by enhancing the overall quality of patient care, enabling faster and more holistic decision-making, and reducing the cost and the time to market for new drugs. Let’s take a closer look how Big Data analytics can transform the sector

A patient hospitalized with congestive heart failure is more likely to be readmitted if there are signs of a distended jugular vein. This is what a Texas hospital found by applying Big Data analytics. In terms of the underlying mechanics, the process is not dissimilar to the collaborative filtering that many e-commerce websites employ – use patterns emerging from aggregated information to predict the predispositions of individual customers. But in the context of healthcare, the implications are significantly more profound. It's the difference between selling books and saving lives. And it's why the value of data in the healthcare sector can never be adequately emphasized or even overstated.  

The reality is that the sector is awash in data but is still somewhat data indifferent. A 2011 estimate placed the sectoral data volume at about 150 exabytes, increasing by 1.2 to 2.4 exabytes every year. And healthcare providers routinely discard up to 90 percent of generated data. Almost 80 percent of the data, comprising doctor's notes, scans and surgery feeds, is unstructured and therefore beyond the purview of conventional data analytic tools and technologies. To make things worse, data is scattered around in proprietary silos belonging to healthcare providers, insurers, pharmaceutical companies, ancillary service providers and other medical institutions. 

Given the implications, Big Data in healthcare is not a choice but a compulsion; more obligation than option. At the level of individual patients, it will enable faster and more holistic decision-making, blending personal information with aggregated trends. At the macro level, data aggregation across regions and geographies will deliver larger samples for more statistically accurate clinical studies, health trending and such like. And it will enhance the overall quality of patient care while simultaneously reducing costs associated with under or over treatment.

But even as healthcare institutions gear up to extract value from existing data sources, data volumes are set to increase exponentially. Though still a micro-trend, data from personal health monitoring devices or even smartphone apps is rapidly accumulating in private systems. Data from social media is also growing and has tremendous potential in helping healthcare practitioners construct holistic patient profiles by combining clinical data with lifestyle and behavioral patterns.

New trends in genome sequencing are opening up brand new possibilities for the sector. It took 13 years and USD 3 billion to sequence the first human genome. Today the same process takes 15 minutes and costs USD 1000. As sequencing becomes more affordable and commonplace, it will generate massive volumes of data that, when correlated with other datasets like life and treatment histories, will lead to the hyper-personalization of medicine.  

Today, a significant proportion of the cost and time spent in the drug development process is attributable to unsuccessful formulations. By enabling researchers to identify compounds with a higher likelihood of success, Big Data can help reduce the cost and the time to market for new drugs. Also, by integrating learning from clinical data into the early stages of development, researchers will now be able to customize drugs to suit aggregated patient profiles.    

Currently, information privacy concerns are the single biggest obstacle to Big Data adoption in healthcare. Another is the absence of an analytics solution powerful enough to gather massive volumes of largely unstructured health data, perform complex analyses quickly, and trigger meaningful – read life-saving – action in real-time; a solution that could, for instance, gather all the data from ICU monitors which today goes un-stored, put it on the Cloud, decipher significant medical patterns that are yet undiscovered, and trigger a medical action instead of merely an alarm. 

This hypothetical example could soon be reality, when a specialist Big Data analytics platform currently under development, comes to fruition. 

And the way I see it, it could be just what the doctor ordered.

It's an exciting era for healthcare. Aside from the propagation of new devices and technologies that one can spot in any hospital corridor, the very way healthcare will be delivered is under reconstruction. Accountable care organizations (ACOs) link practices under a common standard of payment and care. While they are becoming the norm for how the business and oversight of healthcare is practiced, there are still a lot of wrinkles to iron out. Mike Detjan, vice president of service lines and Greg Chittim, director of analytics and performance, at Arcadia Solutions spoke discuss three ways that analytics will play a major role in making ACOs functional and profitable.

1. ID heterogeneous data and make it an effective asset. The core idea behind an ACO is that many small practices and organizations, banded together, will be able to provide a wider range of services to a higher standard of care. What happens when each one of those practices has a different EHR system? It's like trying to get a room full of people that all speak different languages to talk to each other. "When you're trying to look at an organization as a whole, how do you normalize that information," asked Detjen. "Having a platform that can normalize that and integrate it is essential." He said having a system that can pull in the diverse streams of information and churn out reliable analytics is necessary because it can build trust across an ACO. "If you don't have the trust in the data, then you don't have a chance to drive change," he explained.

2. Understand early wins in a program and establish momentum. With any new change or development, people are going to want to see results and are going to want a reason to continue. It's just good business to be able to show progress and rally troops for the next initiative. This is where analytics can be a powerful tool. Chittim said the first "big win" an analytics-touting ACO can look for is the success and quality of the data being captured itself. He noted that an ACO feeding good data into its analytics engines will be the source for all other potential changes. "Making sure that you're improving data quality upfront is a quick win," he commented. "The quality of data pushing up to the analytics should accurately reflect the quality of care." When it can be demonstrated that the data driving analytics is sound, it can be leveraged to promote and track any number of "performance improvement sprints," said Chittim, who added that one can look at analytics to solve two questions: "What's our problem?" and "Where are we now?" as changes are made.

3. Drive engagement with the provider network. There is no element of healthcare that can be described as "set it and forget it." Analytics is an ongoing task, said Detjen and Chittim, and recognizing it as such gives its users the power to drive substantial change over time. One important trick is knowing how to use the data that analytics provides. Detjen cautioned against using it to bully. "The physicians out in the field are independent thinkers," he explained. "They have their own opinions...they're the ones out there touching the patients every day." To bring the full force of analytics to bear, he said, "You need a very systematic program that touches the doctors on a regular basis and gives feedback." This way, a constant back-and-forth is kept going, where the input from practitioners informs the analytics that track the quality of care they provide – and vice versa. Chittim described this as "collaboratively designing where you are, when you want to go, and collaboratively designing how we're going to get there." It all boils down to quality data, open and continuous communications and a desire to leverage new technology to improve care. "Having a trusted set of data," observed Chittim, "and teams working together within a healthcare ecosystem – that is what we need to do in order to change healthcare in this country."

ACO study reflects cost savings and reduced readmissionsACOs and mHealth aligned within care-improvement initiativesACOs may stay afloat even if health reform law goes under

This past year our NHS foundation trust has been beginning to get to grips with the opportunities as well as the challenges of using social media at a corporate as well as clinical level.

The incentives, to us, are clear – the ability to enter into conversations with different stakeholders, and getting authentic feedback to create meaningful change and improve services.

At the same time, we hope social media can help us become more sociable as an organisation.

A corporate approach

From the outset we wanted to embed social media at a corporate as well as clinic level. We've been experimenting with different kinds of social interactions such as live-tweeting and inviting feedback. Both through our individual accounts as well as official accounts that represent the trust and its initiatives.

We're setting the tone as friendly, helpful and informative. This approach also gives individual staff the freedom to showcase their personalities.

We have also begun producing original multimedia content to promote our campaigns that tackle mental health and learning disability stigma.

In clinical services

Some of the most significant insights have come from trying to engage with people who use our clinical services online. This includes initiatives such as the Leeds Club Drug Clinic which uses a blog, Facebook and Twitter to access a specific demographic.

Along the way we have realised that the trust needs to have a wider digital footprint to reach greater numbers. As a result, we use now use our many channels including Pinterest and YouTube.

We have also developed an understanding of how to bridge the gap between our online and offline activities. Sometimes an initiative that is launched online needs to have physical activities and events to support it.

For example, when the trust backed the development of the Leeds wellbeing web – a blog where individuals can contribute information, stories, pictures or films about places and activities in Leeds which help maintain their wellbeing. For this, we provided training as well as engagement through social media workshops.

We also have monthly social media surgeries planned for people using our services and our staff.

Many staff continue to be nervous about social media or don't see the value. This is quite understandable given the constant hyperbole that surrounds it and the frequent negative press it receives.

But our approach is empathetic. We encourage peer-to-peer support, finding clinicians who are using the medium to share their experiences with others. We have also invited external practitioners to deliver training that encourages storytelling as a means to get people online.

The way forward

From an operational perspective we have ditched automated tweets, broadcasting and self-congratulation. We have realised the importance of getting our extended network online and focus on building capacity.

And while we continue to make every effort to ensure the right policies, processes and safeguards are in place, we realise the need to continually experiment and improvise our activities online to find the best fit.

The longer-term vision is to enable practitioners to support people's recovery journeys in both their offline and online lives. And our hope is that as we continue our journey of using social media, our stakeholders will play a significant role in collaborating with us.

Victoria Betton is a deputy director of Leeds and York partnership NHS foundation trust with responsibility for strategy, innovation, partnerships and communications. Abhay Adhikari is a digital engagement specialist who has run a number of digital identity workshops in the trust.

This article is published by Guardian Professional. Join the Healthcare Professionals Network to receive regular emails and exclusive offers.

More than simply an IT catchphrase, "big data" in health care will lead doctors to become more accountable for positive patient outcomes, an EMC executive said.

Big data is going through a transformation, and health care IT will be a major beneficiary of its analytics capabilities, according to an executive at EMC, a major provider of cloud computing, data backup and big data infrastructure. The health care industry can use big data analytics to better detect diseases and aid medical research. "There's a fair amount of hype in big data in general," Dave Dimond, chief strategist for industry solutions at EMC, told eWEEK. "In health care it's coming together." Although the concept of big data may not have been as well-understood a year ago, rather than just being aligned with research, it's getting more focused, said Dimond.

 

"Our position on health care is that big data is real," he said.In fact, EMC predicted a threefold increase in health care data between the beginning of 2013 and the beginning of 2016, according to Dimond. The amount of health care data will eventually be 15 zettabytes worth of information. One zettabyte is equal to about 15 million iPads of data, said Dimond. (A zettabyte is equal to 1,000 exabytes, and one exabyte is the equivalent of 1 million terabytes.) As the health care industry shifts from pay for service—or pay per pill—to Medicare incentives for outcomes under the Affordable Care Act, big data analytics will play a role in helping doctors predict outcomes for patients they're responsible for monitoring. Predictive analytics can also help a doctor determine if a patient would need to be readmitted into a hospital. Big data will allow hospital systems to centralize information from their multiple facilities, Dimond noted. This will enable the health organizations to better keep track of data for a large patient population and monitor health outcomes.

 

To win in this business model [of accountable care] and thrive, they need to be able to do analytics," said Dimond. "They need to be able to access data." Some health organizations may have big data applications but haven't been able to put all the data together because they're all in incompatible, unstructured formats, Dimond suggested. Enterprise data warehousing needs to mature for some health providers, though others are using this technology well, he said. In addition, when researchers can analyze entire human genomes, they will be able to compare populations of individuals. Then scientists will examine analytics models to see how data from electronic health records correlates with the genome data to help determine the cause of various illnesses. Analytic models can also enable medical researchers to evaluate whether certain treatments will be effective, said Dimond. Analytics applications also enable health care organizations to analyze financial data and billing, he said. "The data scientist can look at accumulating all the data possible from a number of sources—clinical systems, financial systems, outside providers, skilled nursing homes and putting it together in a way where you can look at it as a sandbox point of view," Dimond explained. Analytics software such as EMC's Greenplum Chorus can help by pulling disparate health care data all into one place. Chorus is a collaborative data science platform that allows for file sharing, versioning, change tracking and archiving.

 

The Greenplum analytics platform also incorporates Hadoop, and the open-source software framework will enable health organizations to reduce unstructured data and associate it with structured data, said Dimond. EMC acquired Greenplum in July 2010

This week, Big Data Week brings together experts in the fields of science, technology, community development, business and government from across the world to discuss the impact of big data. If used to its full potential, the analysis of large data sets can open previously unthought-of opportunities, and it is time to consider the ways in which the NHS can take advantage.

The NHS is a relentless producer of big data but the underuse and misuse of this information has started to cost lives and is not sustainable.

Back in February, the Francis Report exposed the systemic deficiencies of the Mid Staffordshire trust and highlighted the need for more accurate, useful and relevant information, compliance measured by evidence-based methods (as opposed to gut instinct and out-of-date information) and improvements to core information systems.

While access to data is important, having the tools to analyse all the data, rather than a sample, is vital to identify trends and build a complete picture of what's happening.

One of the reasons years of mismanagement at Mid Staffordshire went unnoticed is because it was possible to misclassify and misrepresent patient statistics.

Tim Kelsey, the NHS national director for patients and information, told delegates at the recent HC2013 Conference that care.data – a programme designed to link patient data from different care sectors for the first time – was essential, because the NHS had almost no information on what it did, never mind about the outcomes of its work.

It is not just about the ability to analyse all the relevant information, but having access to it at the right time so early intervention can prevent adverse developments affecting patients' health, and therefore avoiding retrospective treatment. This could deliver huge efficiency savings by reducing the cost of treatment and freeing time for hospital staff. According to the CEBR report Data Equity – Unlocking the Value of Big Data, the use of big data analytics across the healthcare sector could deliver additional revenues of £14bn from 2012 to 2017.

If the government is to achieve its aim of digitising all health data and making it available to staff at the touch of a button by April 2018, it is important that processes are standardised and appropriate analytical tools are used to enable effective information sharing across thehealthcare network.

The Francis report highlighted the fact that quality is about more than improving outcomes and hitting targets: it is about developing a culture of patient-centred, compassionate and responsive care. Health secretary Jeremy Hunt said he would put compassion back into the NHS by introducing a statutory duty of candour that would punish health providers for concealing mistakes.

That means, however, that staff must be supported to deliver the best care to every patient and be able to make decisions based on accurate information – not just experience or the balance of probabilities. At the same time, trusts must be able to continuously monitor and analyse the performance of every ward and department to ensure patients are receiving the right quality of care. Hunt's planned chief inspector of hospitals and chief inspector of social care will only be able to avoid future mismanagement and make objective and balanced decisions if they have access to up-to-the-minute accurate information.

The NHS must perform a difficult balancing act between providing high-quality care and saving £20bn a year by 2015. Big data analytics offers a way of achieving this and, if approached correctly, it does not necessarily require major new investment. Many foundations have already been laid and now need to be joined together. For example,NHS Blood and Transplant has already been using analytics for some time to improve patient survival rates.

To achieve a sustainable NHS for future generations, co-operation and cross-industry knowledge exchange is essential. This is why events such as the Big Data Week are so important. By sharing best practice, providing support and exchanging ideas, it is possible to improve civil services across the globe and even tame our Leviathan.

David Downing is director of health at SAS UK.

What makes a social network valuable ? Facebook (FB), with more than 1 billion active monthly users posting photos, sending messages, and updating their status, has an impressive market capitalization of $65 billion, or about $65 per user. But Wall Street has assigned a valuation of almost $18.5 billion, or $92.50 per user, to LinkedIn (LNKD), the professional networking site that offers its 200 million members arguably more crucial services, such as help finding jobs. Now a cadre of social platforms aims to disrupt the way consumers share information about personal health, physicians, and treatments. Despite a proliferation of apps that let people monitor every movement and morsel they eat, information technology has yet to revolutionize health care the way it has upended, say, shopping. What the upstarts lack in scale (for now), they more than make up for in utility. Imagine joining an online global community of people with the same rare disorder, or finding a doctor on the basis of detailed patient reviews. Facebook may provide its fans with tools they love, but this new wave of social networks offers tools that its users can't live without -- in some cases literally.

The patient-to-patient network

When brothers Ben and Jamie Heywood, both engineers at MIT, learned that their other brother, Stephen, had ALS (Lou Gehrig's disease), they were frustrated by the lack of reliable information and support online. In 2004 they launched PatientsLikeMeas a destination for visitors to share personal stories, medical histories, and responses to online questionnaires. Today the site has 200,000 users covering about 1,800 diseases.

Patients aren't the only ones finding value in the content on PatientsLikeMe. The company makes money selling its users' data to drugmakers, such as Merck (MRK, Fortune 500)and Novartis (NVS), and other research institutions, like universities. Even with all the privacy laws that regulate patient data, PatientsLikeMe, based in Cambridge, Mass., is able to bundle and release its network's information because, as Ben Heywood says, "we're radically open about it. We tell our members exactly what we do with their data, where it's going, and for what purpose." And the purpose, they argue, is for the greater good: The data can be used to make better, more targeted drugs and more efficient devices. Paul Wicks, a neuropsychologist and research director at PatientsLikeMe, says the company is expanding its patient-driven, standardized questionnaires, and envisions a day when patients can transfer data from health monitors and other devices, such as Google's(GOOG, Fortune 500) augmented-reality Glass product, to create a "learning health care system."

The doctor-to-patient network

Practice Fusion does not at first seem like a social network. The company provides a cloud-based electronic medical records system for doctors, then sells ads for this platform that subsidize the free service. CEO Ryan Howard knew that doctors would never switch to such a system -- even a free one -- unless it offered them more convenience. To win over physicians, Practice Fusion threw in a bunch of tools. Most crucially, it allowed MDs to easily transfer medical records to one another. Nearly 150,000 medical professionals are on Practice Fusion, and the service touches nearly 60 million patients. The reason doctors have been quick to adopt the service, which was founded in 2005 and has about $64 million in venture capital funding, is the intra-network information sharing. "That's the sell," says Howard.

Practice Fusion has just launched a new service, Patient Fusion, that allows patients to post doctor reviews and check their schedules for an opening before booking an appointment; it's like TripAdvisor meets OpenTable for health care. New York-basedZocDoc already offers those services free to patients, but doctors must pay the company $300 a month, an amount CEO Cyrus Massoumi says they're happy to shell out because the service "cuts out paperwork, adds convenience, and can open up their practice to new people."

MORE: Rethinking health care with PatientsLikeMe

Indeed, the way we find doctors -- and our access to them -- has always revolved around networks; these new, online platforms simply upend all tradition. As Jamie Heywood puts it, "Social networks have existed in health care for 100 years -- as guilds, mailing lists, and simply who you knew."

One of the newest networks is HealthTap, an online hub of 1.2 million doctors worldwide who field questions from anyone, anywhere. (The homepage provocatively keeps a real-time ticker of "Answers served," now approaching 670 million.) What works as information delivery to patients is reputation building for doctors, and -- just as at PatientsLikeMe and Practice Fusion -- the ecosystem offers a trove of data to mine or use to build applications.

The risk with any of the new networks, of course, is that a big social network could decide to leverage its massive scale to enter the health care business. Jamie Heywood estimates that of the 400,000 Americans with multiple sclerosis, 300,000 of them are probably on Facebook, while 30,000 are on PatientsLikeMe. Facebook theoretically could track its users' behavior to identify those with MS and exploit its position as the largest registry of MS patients in the world. But managing privacy issues and monitoring the quality of the customer experience for hundreds of diseases is incredibly complex. It's a barrier to entry that health care disrupters are counting on -- and what they hope will make their social networks especially valuable.

This story is from the April 29, 2013 issue of Fortune.

The best thing about the Big Data hype in pharma is how effectively it's shed light on all of the Small Data problems the industry is facing. The roots of the Big Data movement in pharma were innocent enough: challenges in storage, data access, and data analytics that organizations started seeing with shifts toward high-throughput screening and massive genomics data sets. But as Big Data became more and more mainstream, the range of business challenges that got slapped with the "Big Data" label started ranging further and further afield. Industry analysts noticed this quickly, redefining Big Data in terms of the three (or four) Vs--not just volume but also variety, velocity, and variability. Others have been quick to follow. At a recent conference on data-driven drug development, speaker after speaker stood up to talk about their approach to Big Data, and each speaker immediately qualified that they were speaking about the variety of data, rather than the volumeof data.ets, about relevant manufacturing or reimbursement concerns, etc. This is a typical "small data" problem. The information needed to form a complete understanding of the drug-development landscape is scattered across journal articles, grant and IP databases, regulatory filings, clinical trial results, and research presentations. Requirements also vary from one licensing opportunity to the next, meaning that there's no possibility to build a one-size-fits-all solution. The total data involved in this sort of competitive intelligence analysis may be relatively small--certainly no more than a few GB of data--but both the diversity of data and the value of this Small Data problem are enormous.

The most important data-related challenge facing pharma is to use data--any data--to make more and more critical business decisions. Most of these decisions don't need Big Data: they need the right data--whether Big or Small--and they need it at the right time.

 

There's a good reason for this. While it's true that voluminous Big Data problems are sexy and grab headlines easily with exotic talk of petabytes and exabytes, the number of people across a pharma company who actually deal with these volumes of information as part of their day-to-day job is vanishingly small. Put another way, while Big Data is a real problem, it's not a Big Problem. What is a Big Problem, on the other hand, is the challenge of dealing with the diverse variety of (small) data that's needed for decision-making throughout the drug discovery, development, and commercialization life cycles.

You might see analysts refer to this as the variety axis of Big Data, but the challenge is really around getting unified information access.

One aspect of this challenge that every pharma organization faces is in harmonizing data as it is aggregated. For example, any references to ALS, Lou Gehrig's disease, or amyotrophic lateral sclerosis need to be known as the same disease so that data about the disease from one source (e.g., pathway data) can then be integrated against other information from another source (e.g., affected population data).

Another aspect of this challenge is the extent of data diversity that faces pharma today. Any unified approach to data must take as broad an interpretation of relevant information as possible. That means information needs to include traditional structured data (e.g., pathway, target, and genomics databases, CDRs and CTMSs, or manufacturing, finance, and CRM systems), completely unstructured text content (e.g., trial protocol documents, in vivo assay write-ups, clinical case reports, or product perception in social media sites), and all sorts of semistructured sources in between (e.g., CRO-generated spreadsheet data or public NCBI XML data).

 

That kind of broad and deep view of data grants scientists, business analysts, safety officers, managers, directors, and executives access to the critical data that informs their decision-making, wherever the data may be. The process of harmonizing data may be internal, but the data itself may come from just about anywhere--CROs and CMOs, content vendors, even public data--and access needs to be timely. Decision makers can't afford to wait three months for an IT project to gain access to data needed for a decision due this week.

By allowing business users to get immediate and integrated access to all data relevant to critical business decisions, regardless of its location and format, pharma companies can gain a significant competitive advantage. For example, to maintain robust pipelines, Big Pharma continues to look for earlier- and earlier-stage drug candidates to license. But the earlier in development a compound is, the riskier a licensing deal can be. Mitigating this risk requires knowing as much about the candidate drug as possible: about its indication, about its mechanism of action, about competing products and development programs, about the IP landscape, about leading researchers in the area, about expected safety and efficacy targ

Read more: Big Data sheds light on pharma's 'Small Data' problems - FierceBiotechIT http://www.fiercebiotechit.com/story/big-data-sheds-light-pharmas-small-data-problems/2013-03-27#ixzz2PrlRnEyI
Subscribe at Fierce Pharma

With all the doom and gloom we've been reading about the PC market recently, many of you may be wondering where, exactly, the growth opportunities are in IT.

Well, wonder no more. The answer is healthcare IT. I've been talking about the growing need for electronic health records for quite some time now, and certainly EHR is a growth market.

But what can you do with all that information? How can we move from gathering information for more effective billing to gathering and processing information for more effective healthcare itself?

Assuming we get past the privacy issues (by anonymizing data appropriately), if we can can gather and process enormous streams of health data, and if we can possibly do that in real-time or even near real-time, we might have the opportunity to get ahead of disease vectors and outbreaks.

If you want to learn more about how big data works in real-time, our own David Gewirtz did a webcast on the subject last week. Go ahead and watch Data go vroom! How to keep up with the volume, velocity, and variety of big data in real-time. It's free and definitely worth a view.

But what about the opportunity for jobs and IT industry growth? Well, that's where it gets interesting. Industry analysis firm Markets and Markets recently published a report detailing the growth of healthcare analytics.

The bottom line is pretty amazing. The global healthcare analytics market is growing at a Compound Annual Growth Rate of 23.7% from 2012 to 2017. That puts just the analytics segment of the healthcare market at $10.8 billion within four years.

So, if I were you, I'd make sure I spent some time learning about big data, real-time analytics, and take a few statistics classes. Odds are, it will be worth it.

Crisis management has long been an important issue for the healthcare industry, but the continued rise in popularity of social media has made the situation even more complex.

Looking beyond pharma, the healthcare industry as a whole has been slow to adopt social media. Often there seems a cultural gap between the industry and technology, and this makes it even harder for organisations to deal with crises that develops in the social sphere.

No industry seems immune these days from a social media crisis, with large and small organisations suffering and healthcare is no exception, with Israeli hospitals among those facing problems that began on Facebook. 

One was a complaint about a derogatory remark an anesthesiologist made to a woman during labour last year that prompted the woman to write about it on the hospital's Facebook timeline. The hospital emailed her saying they would look into it, but a few minutes later she was copied in, by mistake, to an internal mail saying they want to “bulls**t” her. Naturally this made her very angry and the web, as is often the way, shared and amplified her anger. The story was subsequently picked up by the traditional media, at which point the hospital began taking the issue seriously, but the damage had already been done.

This story is one of many that show how the healthcare industry does not understand the media. This has some basic reasons – Healthcare is slow and conservative by nature, it is highly regulated, evidence-based driven, speaks in professional jargon, is complicated, maintains privacy to the highest standard and is very authoritative.

While, these attributes are necessary, to a certain degree at least, they are also the opposite of social media, which is fast moving, transparent, short (with messages that can be 140 characters or less) and uncontrolled, or uncontrollable. 

These gaps have to be bridged in the near future, and when that happens it will change the face of healthcare.

What has changed with Social Media?

Social media has fundamentally changed the relations between health and pharmaceutical organisations and their patients and customers. What has changed and how? Here are a few changes we need to bear in mind from now on:

The customer/patient has become media: Customers and patients were always satisfied or dissatisfied, and usually shared their experiences with friends and family. Nothing new there. What has changed now is that their friends are now on Facebook, and what they write can easily be published and reach thousands of people and spread like wildfire in a matter of minutes. Data published on socialnomics.net shows that 34 per cent of bloggers voice their opinion on brands, and that 25 per cent of search results for the world's top 20 brands are for user generated content.

Content production and distribution is easy: This has particularly been driven by the growing use of smartphones, and ability to quickly self-publish in this way makes every conversation between a healthcare professional and a patient a potentially public interaction. This needs to be addressed.

No control over messaging: Organisations, especially in healthcare, are eager to control their public or brand message. The truth may be that they never really had that control, due to interference and interpretation in the message disseminated by traditional media. But now social media exposes the lack of control and lets the consumer drive the message, evening out the playing field. This becomes a major issue, especially when it comes to dealing with communications crises.

Timetables: In the 'old' days when crises became public through traditional media, organisations had time to prepare. Today firms are often left chasing after a crisis, which has often gone public before they were even aware internally it had occurred. Consequently companies can still be checking out the facts of an event when a movie of what happened has already had a million views online, in the process becoming a major news story far away in a different continent. Timetables are now measured in seconds for a global event.

No more 'top secret': Our privacy has gone up in flames due to social media, and in the Wikileaks era this is true for people as well as for organisations and brands. This loss of privacy has to be remembered in every email or document produced and in every conversation. This mode of action will prevent us from embarrassment in the future. Every business has its secrets, but we need to be prepared, should they become public and act accordingly.

Social media drives traditional media: Can you remember the last time a news show was aired without a movie taken from YouTube? Social media has become a major driver of news and media coverage. Stories break on the back of tweets and public outrage from social networks and blogs uncovers major stories. Therefore, every tweet should be treated as a potential headline. Every customer should be treated as a journalist, and managers should be trained in how to answer questions online.

The tools of the trade

There are many ways of dealing with crises that begin with social media or the web and they apply in healthcare as much as in any other industry. But there are some similarities between the rules of treating a media crises and a disease – prevention is best way! So here are a few rules I deem important.

Listen in: “If a tree falls in a forest and no one is around to hear it, does it make a sound?” is one of the best thought experiments, and a valid question about social media. The simple fact that an organisation did not hear someone criticising it doesn't mean that it didn't happen. That is why you should keep listening, with the aid of technology, to the conversation about you out there. This could help you treat a small fire before it becomes a wild fire. The second advantage is that you might learn a whole lot about perceptions of your organisation or brand.

Strengthen your social immune system: I personally admire vaccines; I think they are one of the most wonderful and important health advancements of modern times. Vaccines strengthen our immune system and in the same way firms should look to strengthen their immune systems. The way to do this in social media is through active participation, creating a positive atmosphere, building a community around the brand. All these actions will help during a crisis, when a loyal community could help you weather out a storm.

Golden hour: One of the first rules learned in treating trauma patients in the field, is that survival rates increase if you get them to a medical facility within one hour of injury (the so-called 'golden hour'). The same applies to crises in social media. Respond, and respond quickly. Even if the response is – “we still haven't got an answer but are checking into it”, it is a good one. Do not ignore the issue, because then the story becomes about a lack of response. If a public response is impossible for regulatory reasons, then reply to the customer privately. If he or she is satisfied, they will probably share it with the community, and the matter will be resolved.

Treat the cause, not the symptoms: Remember that a social media crisis arises from a real problem, so treating and solving it will turn off the flames, whereas handling only the coverage in traditional or social media but not treating the actual cause is insufficient.

Honesty, transparency, empathy: When dealing with a crisis in social media these three traits are crucial. Never lie or tell half-baked answers - they will be revealed very quickly. Every attempt to cover-up is doomed to fail. Empathy is important; treat your customers with respect. As I said before, if you fix the problem then the whole scene is changed.

Standard Operation Procedures (SOP): Every organisation has them. Health and pharmaceutical organisations even excel in this area, but all these SOPs have to be adjusted and written with social media in mind, especially the crisis management SOP. If these SOPs haven't been visited in a while, now is the time to check them out.

Digital assets: The organisation probably has a website, sometime a blog or other social media channels. All of these are digital assets, and need to be tended and kept current, so that in times of trouble they can be updated and assist the brand or firm to communicate with their customers. Furthermore, when trouble hits the organisation, all the digital assets need to be lined up with the messages relating to the crisis. That having been said, a crisis can happen in every arena on the web, and needs to be dealt with where it happens.

Educate your employees: Last, but certainly not least, is the importance of internal education. Employees can be a firm's greatest ambassadors and its eyes and ears in the social sphere. So educate employees on how to respond on the web, what is right or wrong and let them be involved, don't block them out.

To sum up, social media crises may occur on any day and can be devastating to an organisation, but they might also be an opportunity for betterment. Keeping your eyes and ears open, and being prepared will help in times of need.

Uri Goren is general manager at the Israeli digital health company e-Pochonder.com

 

In the world of Big Pharma, big data is a looming giant that can be a tough beast to tackle, but it can be a rewarding one too. With a vast amount of data in myriad formats, the Ecuadorian arm of Pfizer knew that using Excel spreadsheets to record and analyze marketing, sales, and distribution data was a Stone Age solution for their Space Age needs. They turned to Noux, an Ecuador-based company that works with SAP to implement a more nimble data-driven system to help with their business intelligence.

Esteban Burbano de Lara, commercial manager from Noux, and Eduardo Saenz, business technology director, Bolivia Ecuador and Perú from Pfizer speak about some of the ways transitioning to embrace Big Data has helped Pfizer and can help other organizations conduct powerful business intelligence and analytics of their markets. 

Deal with data overload

Pfizer collects massive amounts of data on everything from how much of its product is sold at any particular pharmacy to whether a doctor prescribes its medication versus a competing one. Burbano de Lara's team helped Pfizer implement a system that could analyze data as it was coming in. He outlined the complexities of the task, saying "you'll get data from 50 distributors, from five to six marketing forms, data from your in-house teams." Eduardo Saenz worried about the data's density being a factor to overcome, saying that "big data sometimes overwhelms people because it's too big... there is too much data to analyze." He says the system they put in place allows individual users to scale down and segregate the appropriate data for their research. A system that is able to retrieve the appropriate data and ignore the rest is a crucial tool in harnessing big data, Saenz says.

Wrangle multiple streams of information

Most people picture big data as a series of cleanly organized files full of numbers and easily-digestible information. If only, laments Burbano de Lara. "These data sources are very heterogeneous, very dispersed," he says. "Sometimes it's a file, sometimes it's a database, sometimes it's encrypted." Pfizer needed a streamlined way to stay on top of this wide amount of information and make informed decisions on where to head next. 

Keeping all of that information straight can be a challenge. Does a doctor prescribe the competing pharmaceutical more? Should more of an effort be made to reach that doctor, or should resources be focused elsewhere? Saenz says that it was "a big challenge to design a process on how to present all of the information in a friendly manner, using one interface to display all of this information." With the data warehousing solution, he says, "we can analyze each information source individually, but at the same time we can mix all of the information sources. That allows us to see things that other companies cannot see." 

"What in the past took [Pfizer] weeks if not months... now happens instantly," says Burbano de Lara about how Pfizer harnesses data. Sales representatives visit doctors to tout products, but how effective are their pitches? While Pfizer grabs lots of information, Burbano de Lara notes: "This data is difficult to analyze. There's so much information, but I don't really know where my products are going to, whether doctors are prescribing my products or not." 

Enter the big data. "If you don't measure something, you can't improve it," says Saenz, who goes on to say that before Pfizer had implemented its big data strategy, it couldn't reliably measure how Pzifer was  performing in markets – certainly not at the level tit does now. "To have this data warehouse allows us to do things in a very efficient way," he says. Saenz says that the warehousing solution provides "street-precision inventory (of our own products) by combining sell-in (what we sell to our wholesalers) and sell-out (what our wholesalers place on the market." Pfizer can measure the efficiency of its representatives and track competition.  Burbano de Lara says that by consolidating all the information Pfizer collects and analyzing it effectively, the company builds "very laser-targeted information, so they can make decisions really really fast." He likens the transformation to "going from riding a horse to going on a plane."

Hospitals can take advantage of social media, just like any other company can. It’s an excellent way to market to their geographical audience and keep the hospital name in front of a wide range of people on a regular basis.  It can also benefit those in the community who are connected to the hospital. For many reasons, social media is the perfect way for a person to feel connected to their caregiver in between visits. How can such a big, regulated organization get in on all the sharing and liking? Here are just a few ideas.

1. Offer pre- and post-care information to patients

Social media can be an excellent way to remind patients about the processes of checking in before an appointment. It can also be used as a general reminder to those who have had surgery to remember to attend any appointments or to get in touch with their doctor if they experience complications. Unfortunately, with HIPPA and privacy laws, specific patients cannot be singled out or contacted over social media, but general reminders can help patients keep up with their healthcare.

2. Share stories

Social media is a good way to share, with permission, patients’ success stories. Hospitals and medical facilities can highlight the success of patients after specific surgeries, share when they are recognized for an award or any other feel-good stories appropriate to share. They can also allow others to use their social media marketing outlets as a way to share their personal stories and sing their praises. Again, personal information cannot be used without permission, so be sure you get signed releases from your patients or find a way to share without revealing personal information.

3. Answer questions

Social media can make it easy for hospitals and medical facilities to interact with patients. It can be used to help people learn the best way to get in touch with their doctor or find out what certain procedures are. Patients’ minds can be put at ease when they need a quick response about what they should expect during the healing process.

4. Share general information

Hospitals and medical facilities can use social media as a way to share general information with those in their community. They can use social media to spread important information about the flu and colds or any outbreaks or other health hazards in the community. It can also be a way to share information about allergies, chicken pox and other common ailments. The Center for Disease Control even offers a Social Media Toolkit for Health Communicators.

5. Introduce new technologies or doctors

It’s important to keep patients and the community in the know about what is new at the hospital or medical facility. The community might be interested in knowing about the newest doctor in town and what she specializes in. Likewise, they might want to know about new medical technology and how it helps patients receive better care. With social media, it’s easy to share new information with fans, followers and visitors to the profile.

Social media can be a great way for a hospital or medical facility to keep in touch with current or prospective patients and develop relationships with them. There are many ways to engage people from the community and keep them interested in what is going on.  A marketing firm can be an excellent resource for a hospital or medical facility when they begin their social media outreach.

Read more at http://www.business2community.com/social-media/using-social-media-for-good-healthcare-0411025#uQcTvzXtJ84iGxj9.99

Using artificial intelligence and simulation modeling, researchers from Indiana University have found that machine learning can improve cost and quality of healthcare in the U.S., according to an announcement from IU.

The artificial intelligence framework, which combines Markov Decision Processes and Dynamic Decision Networks, used by IU researchers, shows how simulation modeling that "understands and predicts" the outcomes of health treatments could reduce healthcare costs and improve patient incomes by about 50 percent. Their work has been published in the journal Artificial Intelligence in Medicine.

The research expounds previous work by the study's authors, which showed how machine learning could decide a patient's best treatment. The approach is not disease-specific and could work for any diagnosis or disorder.

The Markov Decision Processes and Dynamic Decision Networks enable the system to deliberate about the future, considering all the different possible sequences of actions and effects in advance, even in cases where we are unsure of the effects," researcher Casey Bennett said in the announcement.

The study's authors conclude that an artificial intelligence simulation framework "can approximate optimal decisions, even in complex and uncertain environments.

"Future work is described that outlines potential lines of research and integration of machine learning algorithms for personalized medicine," they said.

FierceHealthIT's recent interview with Tina Buop, CIO of La Clinica de la Raza in Oakland, Calif., highlighted the importance of predictive analytics in healthcare. "You can bankrupt an organization very quickly if you don't understand your patient population," Buop said. Predictive analytics are used to know where to accept risk and how best to help a patient in advance. For example, Buop said, you can't fix your balance sheet when getting paid a fixed amount for a patient per month when they're your highest utilization.

Buop will join a panel at HIMSS13 in New Orleans hosted by FierceHealthIT and the College of Healthcare Information Management Executives, that will focus on using predictive analytics to improve care and efficiencies, including how to use data, artificial intelligence and clinical decision support to identify and create customized interventions for patients who are most at risk for adverse events and readmissions.

A study published in the International Journal of Medical Informatics in January highlighted how predictive modeling could reduce unnecessary lab tests for intensive-care patients with gastrointestinal bleeding.

To learn more:
- read the announcement from IU
- here's the the study's abstract