Patient care is changing in the United States right before our eyes and technology innovation is driving that transformation. What kinds of technology?
Some familiar examples include engaging patients in prevention and ongoing care (the "consumerization" of healthcare), leveraging analytics from large data sets ("Big Data") and using external service providers (going to the Cloud and SaaS). While consumerization, Big Data and Cloud are important components of any health system information technology strategy, don't forget one of the most critical ingredients in your strategy: speed. This article describes how to increase the pace of your health care IT development efforts while staying aligned with your operational capabilities. By increasing the speed of your software development, you will realize greater benefit for your organization, with less hassle and frustration.
Move farther with faster DevOps
There are many definitions of DevOps from vendors selling software that include aspects of continuous deployment, automation and agile software development methods using sprints and scrums, Kanban, etc. At Trace3 we define DevOps as the ongoing process of building the right software, faster. Take notice that this simple definition doesn't prescribe specific tools or methods, but it does emphasize speed, velocity and momentum.
Build the right software, faster
How do you build the right software? The traditional (snail-pace) software development process has three steps. First, business analysts carry out numerous interviews and to collect user requirements (preferred features, functions, expectations, security requirements, and compliance and audit/control concerns) in excruciating detail to create a "requirements document." Then, these carefully constructed documents are delivered to the development group, who code based on their interpretations of the requirements, rather than actually speaking with the people who will ultimately use the software. And finally, once written, the software is handed off to the Operations Group to install the code for use in the specific environment.
At Trace3, we condense those three steps into a seamless and more efficient process. The Trace3 system includes:
- Compressing the distance between Developers, Consumers and Operations
- Reducing the time budget for each step in the phase via automation, while maintaining a laser-focus on quality
- Using Cloud and SaaS Services
Compression
Successful DevOps programs require close integration of all system components, a process Trace3 calls "compression." The goal is to consolidate the separate Development and Operations function cycles into a single, integrated unit, known as DevOps. In this context compression is both physical and abstract and includes the people involved in both areas. For example, we suggest physically co-locating teams based on the desired outcomes of the project. This can mean moving developers, sysadmins and analysts into pods or office-groupings based on the applications they are building, rather than leaving them near their functional peers who might be working on different, unrelated projects. By putting the developer next to the analyst who is next to the DBA and who all are working on the same deliverable, we compress the time and effort required for collaboration. Sometimes, an old-fashioned conversation is the most efficient and quickest: it is much faster to ask someone a question and get an immediate answer without waiting for a reply from email or Slack. Closer office quarters also result in better alignment around goals, objectives, and roadblocks, both perceived and real. Traditional software development and IT operations groups rarely operate to their potential, and reorganizing the team into DevOps will deliver better software, faster.
Reduce the Time Budget with Automation
Most organizations can also significantly improve the velocity of software development with automation, without sacrificing quality. In fact, the two usually go hand-in-hand and quality improves with automation while software development speeds up rapid advances in infrastructure tools (such as Chef, Puppet and Ansible) and technologies (containerization, software defined networking, cloud/SaaS services, database virtualization) open the possibilities for automation to help increase velocity and improve quality. How is that possible? First, remember the current work model supporting traditional software development and infrastructure installation. In most organizations, the process is closer to "craftwork" than a modern assembly line. Current methods involve tickets, queues, deployment gates, configuration management and considerable time. A developer request for a VM requires a work order in the queue system, which is then routed to a sysadmin who provisions the VM, who then creates a ticket for the network group to provision an IP address and create a DNS entry. Lots of steps and lots of involvement from people who can also delay the progress by being gone on the weekend, taking holidays, getting sick, or going on vacation.
Instead, large scale organizations such as Google, Facebook and Twitter automate their infrastructures with very few people. Increasing the velocity of IT development to match the speed of business requires adopting different ways of working and automation is a key element toward achieving that goal.
Cloud and SaaS
Cloud and other external services (infrastructure as a Service, Software as a Service, etc.) are becoming increasingly relevant in the healthcare market. While traditionally viewed as appropriate for only non-regulated industries with more lenient security concerns, Cloud services are now being adopted in healthcare and financial services industries and are meeting demands for higher levels of protection around data security, information security, data governance, and threat detection. But beware! Building Cloud and SaaS services is different from building your own infrastructure, and requires different skill sets for secure, successful adoption. Many assume that internally-hosted systems are inherently more secure and better protected than external cloud services. However, we have observed too many examples of unsuccessful home-grown Cloud services. Many organizations that are not in the hosting business overestimate their expertise, resources, people, staff and ability to deliver and maintain highly secure environments. The sheer volume and dynamic nature of today's information and data security threats are immense and quite frequently internal IT Security departments are reactive; leaving them the most vulnerable to data breach and loss. If you decide to build your own Cloud, it isn't impossible, but please be careful.
Summary
Organizations that win market share in the next decade will get there by developing the right software, faster. DevOps, automation and Cloud services are the critical tools that winners will embrace early – even if those companies have imperfect data but a clear path forward and enough data and appetite for risk to start the journey. Taking the leap and making some big changes to the process are worth it. The race to "consumerize" healthcare will require significant change in the traditional delivery of IT services, and will lead to faster delivery with better quality.
The views, opinions and positions expressed within these guest posts are those of the author alone and do not represent those of Becker's Hospital Review/Becker's Healthcare. The accuracy, completeness and validity of any statements made within this article are not guaranteed. We accept no liability for any errors, omissions or representations. The copyright of this content belongs to the author and any liability with regards to infringement of intellectual property rights remains with them.