Driving economic efficiencies with speech recognition

If trend watchers are correct, German healthcare is facing dark days ahead. With rising costs for clinical stays and the number of state-subsidies decreasing, a third of the country’s hospitals are currently in the red. But there are is a ray of hope. According to Dr. Martin Specht, head of the Application Management Department at University Clinic Jena, speech recognition can increase quality of care as well as a hospital’s economic efficiency. At a recent industry roundtable, Dr. Specht explained why they believe speech recognition can be the right therapy for any modern-day hospital suffering from financial challenges.

The general problems of the healthcare sector are indeed extensive.  The necessity to increase the number of patients in order to get out of the red is no longer the only challenge hospitals face, there is also the need to improve internal processes.  On the one hand, documentation processes must be mapped qualitatively; on the other hand, we must relieve our doctors and, from a technical side, do everything to implement a type of documentation process that produces maximum quality but requires minimum effort.  This is where speech technology comes in, and the reason why we at University Clinic Jena have been using voice recognition for some time now.

What strategy do you use when aligning various IT systems with medical specialty groups?
Large hospitals very rarely work with one consistent, monolithic IT solution from a single source.  At University Clinic Jena, we focus on the individual sub-systems and their potential for specialization; we try to adapt them as much as possible to the users’ needs, so that they optimally support the process at hand.

How are clinicians at Jena using speech recognition?
We have been leveraging speech recognition in the Radiology department for ten years now, although exclusively for diagnostics.  In other medical areas things are different and  speech recognition is predominantly used for writing discharge letters, however, the fundamental challenge is motivating physicians to try to use speech recognition.  This works quite well, as long as the physicians received good initial training, but are also constantly supported and motivated.

How did you achieve physician adoption?
At the Pediatric Clinic, we introduced the use of voice recognition for discharge letters, which can be time consuming for physicians to create for each patient.  We use cloud-based speech recognition technology, which has made this process much simpler and has contributed to widespread adoption. Additionally, the commitment and success of the pilot physicians was like a germ cell, and colleagues actually spread their zeal for adoption and it became contagious.

Were there unique needs of each medical specialty?
In the Neurology department, we recently launched a project that focuses on medical progress documentation.  Neurology is the field where we have almost all processes electronically mapped – although via highly heterogeneous IT systems. Therefore, the clinical team was especially keen to use speech recognition for their progress notes.  The Nuance solutions helped us speed this process by offering physicians the option to document at a specific point in the record using their voice, which directly simulates keyboard input. This enabled us to support many colleagues in their respective systems, and acceptance was good. Another exciting aspect is that all the information is brought together in the private and secure cloud and the vocabulary is automatically improved each time a physician dictates– no matter whether it is performed using front-end dictation or  via backend technology or.  According to the feedback we get from our neurologists, this allows them to quickly reach a high recognition rate and accuracy that is vital to their workflow.  Currently, we are also testing medical speech recognition using smartphones, so that we will soon be able to maintain our medical progress documentation via an app.

Can speech be the treatment for a “sick” hospital?
We are noticing that we achieve an increasing number of positive results with speech recognition.  Sure, it is an investment and must be used effective.  With older solutions, the individual frequency of use was usually not that high because physicians were only able to document 10 percent of their daily tasks via speech. Today, we can support standard processes much better and the recent technology advances have been adopted by physicians and increased their efficiency so we can justify the investment.  It is exciting that we are now able to support our physicians in so many different situations.

What do you recommend for improving the satisfaction of patients and physicians?
For us, economic efficiency and the satisfaction of physicians are directly linked.  If the physician is happy with the process and the technology supports his or her workflow, this results in a more efficient processes and improved economics.  Patients will be satisfied if we allow them to benefit from our improvements: for example, if we optimize the turnaround time for our discharge letters. We have an internet patient portal where we provide discharge letters to referring doctors, but this only beneficial if we do it in a timely manner, so that a patient who sees a referring physician in the afternoon has a discharge letter already available.  Speech recognition helps us speed up these processes.

Does speech recognition also optimize the billing process?
It does, but less so because of speeding things up, since adequate billing requires the patient record to be complete. However, we experience an improvement of quality.  In the Neurology department, for example, we observed this when looking at the medical progress documentation; it’s done in a more comprehensive way simply because the process has become so efficient – and will even work via smartphone.  Improved quality also means more additional documentation which later helps clarify expenses and support appropriate funding were taken.  For practice-oriented surgery in particular, which has little time for documenting inpatient processes, these technologies contribute significantly to an overall improvement.


5 reasons to speech-enable electronic health records (EHRs)

Download this whitepaper to learn more about the value of speech recognition


Tags: , , ,

Dr. Martin Specht

About Dr. Martin Specht

This was a contributed post by Dr. Martin Specht, head of the Application Management Department at University Clinic Jena. If you’re interested in more content like this, visit the Healthcare section of the blog.