There has been a dramatic shift in healthcare technology in recent years, thanks to a rapidly growing population, evolving technology trends and demand by providers and consumers for real-time information and support. This has led to a mass introduction of electronic health records (EHRs) and other electronic systems by healthcare delivery organizations (HDOs).
Unfortunately, according to Gartner, most HDOs lack the technical infrastructure to accommodate the demands of EHR implementation. In fact, Gartner states that “the typical ‘as is’ state of infrastructure in healthcare is characterized by spotty wireless coverage, inadequate network bandwidth, limited-capacity data centers and an overall shortage of redundancy.”1
That’s bad news for HDOs. When IT is integrated into critical business systems, like patient records, scheduling and billing, downtime can have negative consequences that extend far across the organization. For instance, if a provider can’t access EHRs, they may be forced to limit care available to patients, staff might not be able to process bills or schedule future appointments, and frustration will likely mount for everyone who touches the business.
This pain point is best described by Tom Shurer, Vice President Information Resources at Devereux, one of the U.S.’s largest nonprofit behavioral healthcare providers serving individuals with intellectual and developmental disabilities. “Communication among those involved in the care of our patients is essential for positive treatment outcomes,” he said. “We’re accountable for the well-being of each person and their family. We need to exceed expectations and avoid risk.”
To this end, most HDOs have begun, or are at least considering, the process of upgrading their infrastructures to handle the demands of EHRs and other electronic systems. This is a costly process for most providers that requires the participation of key stakeholders.
As an HDO upgrades its infrastructure, it’s critical that they ensure visibility and performance in both their existing and next generation platforms, and prove the return on their investment. Infrastructure performance management solutions can help achieve these goals by offering end-to-end visibility, and making it possible to recognize issues quicker, allocate resources correctly and avoid costly downtime.
Here are four challenges that HDOs face when supporting an upgraded infrastructure, along with solutions facilitated by performance management.
Challenge #1: HDOs are expected to support an always-on and highly complex infrastructure
Most operations at an HDO can continue in a paper-based or even hybrid record-keeping environment when electronic systems are unavailable. However, when a fully digital EHR system is introduced, operations can grind to halt during downtime, resulting in costly consequences.
For instance, within one North American healthcare delivery organization, it’s estimated that a one-hour outage of an application would cost $24,093.00. If the outage lasted 24 hours, it could cost the hospital upward of $500,000. These costs include:
- Inability to service patients: HDOs may have to delay seeing patients or divert them to another organization if records and systems aren’t available, resulting in lost revenue.
- Lost productivity: Outages impact many business processes throughout the organization, including accounting, claims, scheduling, purchasing, transfers and discharges.
- Recovery costs: Downtime has immediate and long-term consequences. When considering the cost of an outage, the time it takes for any business process to recover and return to normal should be taken into account.
- Reputational damage: Continual and repeated downtime can cause patients, providers, suppliers and partners to lose confidence in the HDO.
Solution #1: Implement a digital infrastructure management platform that gives real-time information on system health
Clearly, downtime or even slowness isn’t an option. Implementing a system that provides a single source of truth within a complex infrastructure and gives real-time information is essential. Items to consider include:
- Network bandwidth utilization: It’s important to understand where resources are being consumed, where they’re tipping the threshold, and where they’re over-allocated. NetFlow records provide insight into who the top talkers are in any HDO. And many companies discover that automated backups of system data, when not scheduled at optimum times, can bring network operations to a crawl.
- Baseline deviations: Understanding the baseline of how users interact with the application is important. Even more essential is instant alerting for when the baseline deviates above or below a threshold that is based on use patterns. This can provide real-time insight into what’s happening in the network. For example, teams could receive an alert when log data indicates that the number of failed user logins for the EHR system spikes above normal activity for that time of day. They could then proactively resolve the issue, rather than waiting for complaints to start rolling in.
- High frequency polling: Effective performance management relies on having access to the right information on a timely basis. It may seem like it’s enough to poll the infrastructure every five minutes, but that frequency is woefully inadequate for a critical link that carries latency-sensitive data. Often, down-to-the-second polling is required to gain proper visibility of infrastructure performance issues.
- Integrated metrics, flows and logs: Teams need access to multiple types of data to determine the source of an issue and reduce Mean Time to Repair (MTTR). Being able to pivot from one type of data to another within the same interface allows them to get to the bottom of an issue quickly and move on to solving it. For example, SNMP and NetFlow records may reveal that a certain class of traffic is no longer reaching its destination, but only the log data would reveal that a recent configuration change was the culprit.
Challenge #2: HDOs require end-to-end visibility into their hybrid cloud infrastructure to deliver a consistent experience to their users
The reality is that almost every enterprise today is using a variety of public or private, internal or external cloud services -- typically referred to as the hybrid cloud -- to support their infrastructure. But this presents a dilemma for those charged with managing an increasingly complex and robust architecture.
The question becomes: How do network teams manage what they don’t own? Moving resources to the cloud can create significant visibility gaps over infrastructure performance, creating vulnerability for errors.
Patient confidentiality is also of huge concern for HDOs when dealing with EHRs, even more so if they don’t own the infrastructure where that information is stored. According to Reuters, a patient’s medical information is worth 10 times more than their credit card number on the black market.2
Solution #2: Ensure cloud and on-premises infrastructures can be monitored from a single platform
It’s important to be able to see information about on-premises infrastructure metrics collected via SNMP, IP SLA or other standard protocols, along with cloud metrics collected from services like Amazon Web Services (AWS)™, Microsoft Azure™ and IBM Bluemix™.
It’s also critical that all of this information can be normalized so it can be monitored, baselined, alerted and reported upon, regardless of the source. It makes problem correlation much easier. When the data is uniform across whatever objects and devices are being monitored – whether they’re in the datacenter or in the cloud – there is complete visibility into the network and applications.
Monitoring can also detect potential security issues in real time, such as the beginnings of a DDoS attack. Catching an issue before it becomes full blown can be the difference between releasing highly sensitive information or not. It’s important to understand what’s happening quickly so the network team can act to mitigate a potentially costly situation.
Challenge #3: HDOs are responsible for maintaining all elements of their infrastructure, whether or not they own it
Another challenge that HDO IT teams face is monitoring performance of applications that run in a Platform as a Service (PaaS) environment. Operations teams need to be able to monitor privately managed applications and data with the same reporting consoles and workflows they use to manage provider-managed network, storage and servers.
Many specialized vendor’s tools don’t monitor what’s happening inside virtual machines (VMs) at the network traffic level, leaving teams at a disadvantage. HDOs that want full visibility into their infrastructure need tools that use APIs to dynamically capture KPIs from both physical and virtual elements -- no matter who manages them -- and feed them into a single screen.
The advent of software-defined networking (SDN) brings even more complexity to this issue. SDN allows operators to spin up additional virtual resources on demand, vastly expanding their capacity to handle information, and adding elasticity to an ever-changing infrastructure. HDOs will want to prepare themselves for this quickly approaching trend.
Solution #3: Ensure the entire infrastructure can be monitored with an integrated platform
Teams will want to use an infrastructure performance management tool that can scale to monitor millions of objects across the entire network and server, and virtual infrastructure. This is because PaaS environments create much more performance data than static physical environments. New VMs are spun up all the time -- fast -- and teams will need insight.
No matter the data collection source, they’ll also want to be able to build a dynamic baseline of normal behavior for all performance indicators, and set threshold-based alerts for when performance levels deviate from historical norms. This enables IT operations staff to monitor and proactively manage applications and data using the same reporting tools used to manage their network, storage and servers.
In addition, teams will want to ensure any user, whether they’re a network operations center (NOC) engineer or a specialized VM administrator, can log into the infrastructure management platform and get a consistent dashboard and reporting interface. This all-in-one solution approach facilitates a rapid troubleshooting workflow and efficient communication among the NOC and virtualization teams to ensure service levels are met proactively.
An infrastructure management platform will also help teams prepare for SDN. When they begin using SDN frameworks such as OpenStack to dynamically add and remove elements of their infrastructure, they’ll need the ability to automatically update and understand rapidly changing topologies. An infrastructure management platform will help them achieve this goal and assure the continued health of their system.
Challenge #4: HDOs need to support future IT trends, such as the Internet of Things
EHRs are today’s reality. However, the amount of data in the healthcare industry is poised to explode in the coming years. The Internet of Things (IoT) is expected to add close to 30 billion connected things by 2020, according to Gartner3 . These things will be used across a wide range of industries, especially healthcare.
According to MarketResearch.com, the healthcare IoT market segment will hit $117 billion by 20204 . Goldman Sachs sees the opportunity for $305 billion in savings for the U.S. from digital healthcare in the near future, focusing on measures like behavior modification, remote patient monitoring and drug management5 .
Behavior modification helps patients change their habits, lowering the healthcare costs associated with obesity, smoking cessation and overall lifestyle. It could be accomplished with wearables, incentives and other things that would provide information for aggregation, normalization and monitoring.
Remote patient monitoring would be a game changer for the healthcare industry. The average daily cost for a single inpatient stay was over $1,700 in 2013, according to the Kaiser Family Foundation6 . However, remote monitoring would give patients the option of going home while still being monitored by their healthcare provider. The healthcare provider could then receive up-to-date information on vitals, pain levels and other important factors in real-time to ensure the patient receives proper care.
Finally, the process of creating and managing drugs is one of the biggest issues facing the healthcare industry. According to Forbes, the average cost to develop an approved drug is roughly $4 billion7 . Integrating the IoT into drug creation and usage could help mitigate those costs. One example is an edible “smart” pill that stays in the body to help monitor medication regimens and health issues.
Solution #4: HDOs need to update their infrastructures and monitoring platform today in order to remain relevant tomorrow
HDOs will be able to use IoT sensors and applications to monitor medical devices, personnel, and patients. But they’ll need to prepare now for how they’ll collect, normalize and understand the immense amount of data the IoT will provide.
The best way to do this is to leverage their infrastructure management system as a smart data framework that can collect, enrich and integrate IoT data at scale and with context. This framework allows users to perform three functions:
- Real-time service assurance performance management: A smart data framework provides real-time insight into performance metrics to ensure service is delivered efficiently. Not only will it monitor sensor data, it will give visibility across all layers of the infrastructure. It will also allow teams to enrich, contextualize and baseline this data, helping assure not only the integrity of the IoT data, but of the service, as well.
- IoT performance insight: Collecting and analyzing IoT data – in context – can provide insight that aids in both top- and bottom-line decision-making. Questions like “How is the service itself performing?” and “Can the underlying infrastructure handle the projected growth or do I need more capacity?” can be answered with an infrastructure management platform.
- Data delivery to analyzer tools: Teams will also want to ensure their smart data framework can collect data from devices and their supporting networks, compute resources and applications, then normalize it so it can be easily analyzed, visualized and integrated. It can then feed that information via API to a tool that can provide context for strategic business decisions.
Many HDOs understand the necessity of upgrading their infrastructures to meet increasing consumer and provider demand, and are taking steps to make a more robust architecture a reality. This is transforming HDOs as they become more like software companies at their core -- driven and differentiated by data, analytics and applications.
When considering upgrades, it would be remiss to not factor in an infrastructure management solution that provides end-to-end visibility into increasingly intricate architecture. Factors like the need for always-on availability, managing a hybrid cloud environment as well as physical and virtual infrastructures, and growing trends like the IoT bring added complexities that administrators need to understand and use to their advantage.
Having a single source of truth into what’s happening in the infrastructure in real-time can help network teams make informed decisions about how to manage their system effectively. This allows the HDO to focus on the primary matter at hand: providing first-class care to patients.1Gartner, Oct. 2013, Health Delivery Organization CIOs Must Deliver a Robust Technical Infrastructure for Electronic Health Records Success
2Reuters, Sept. 2014, http://www.reuters.com/article/us-cybersecurity-hospitals-idUSKCN0HJ21I20140924
3Gartner, April 2015, Agenda Overview for the Internet of Things, 2015
4MarketResearch.com, April 2015, http://www.marketresearch.com/land/product.asp?productid=8926222&progid=87380>
5Goldman Sachs, June 2015
6Kaiser Family Foundation, 2014, http://kff.org/other/state-indicator/expenses-per-inpatient-day/
7Forbes, Feb. 2012, http://www.forbes.com/sites/matthewherper/2012/02/10/the-truly-staggering-cost-of-inventing-new-drugs/#1bc4f73f4477