The accuracy of cloud speech recognition for healthcare dictators has come a long way in the last 10 years. Not only is the initial accuracy impressive with newer speech products, but the continued learning and speed of text display are dramatic improvements compared with early desktop speech products. Those considering a move to cloud healthcare speech may wonder, “In addition to the correct display of my words, how else will the software assist me? What features are really going to save my time and energy to make this transition a worthwhile investment?”
Below are the most widely used speech-to-text techniques that answer the above. Users apply these built-in “extras” to streamline their note making, giving them more time for patients and family. Continuous text is great, but busy providers require more such as assistance with abbreviations, formatting and the appearance of the note, in addition to note contents. Plus, the added benefits of shorter reimbursement cycles, more accurate coding and richer notes strengthens the value proposition for a cloud speech product.
1. Shortcuts
Perhaps the most significant practice is Shortcuts, the rapid insertion of blocks of text with just a short phrase, which is ideal for frequently narrated diagnoses, paragraphs and summaries. Essential note content such as findings, care plans and discharge instructions often contains repetitive text, perfect for Shortcut insertion. Quickly adding the routine part of a note via Shortcuts, sometimes called Normals, enables providers to concentrate on each patient’s unique encounter details and results in more descriptive narrative. Arrendale installation teams teach new users how to add their personal Shortcuts during software trials and initial training. Arrendale will even pre-load your existing shortcuts for those transitioning from a desktop speech product to our simpler cloud solution. Users quickly add new Shortcuts to their personal user profile as they dictate in the initial weeks. Even better, personal shortcuts are available from secondary locations and home offices due to cloud technology.
2. Substitutions
Particularly in healthcare speech recognition, the Substitution feature of a speech engine is hugely beneficial. Abbreviations for diagnoses, symptoms and medications are frequently dictated by clinicians, but confusing to patients and their family members. A prime example is patient comprehension with discharge instructions. Our speech engine is quickly trained to spell out abbreviations and acronyms with the full text, correctly capitalized, per facility standards. Think how much faster it is to say “Presby Main” instead of “Presbyterian Main Campus Medical Center.” Likewise, speech engines can be programmed to substitute abbreviations for full words in circumstances where the abbreviation is preferred, i.e. “DBP” rather than “diastolic blood pressure.”
3. Navigational Voice Commands
For savvy providers, Navigational Voice Commands save click time and effort before they even get to their EHR text box. Arrendale speech products perform more than just continuous display of narrative. Navigational voice commands are useful for opening websites and moving about within EHRs. Repetitive clicks to get to the correct EHR screen can be accomplished faster with a single voice command. Examples include “Open WebMD,” “Open Progress Note Template” and “Open Microsoft Word”. The frequency of multi-click routines is a huge annoyance according to providers who are grateful for the quick response when a voice command replaces repetitive, sequential clicks.
4. Formatting Voice Commands
Dictated Formatting Voice Commands save significant time compared with straight typing. Typical examples include verbal commands of “Number Next” to detail diagnoses and medications into understandable numbered lists. Accurately listing medications and dosages is critical to the prevention of medical errors and successful communication with patients, thus appearances matter greatly with lists. Likewise, users swiftly invoke formatting voice commands for spacing and capitalization such as “New Paragraph,” “Caps On,” and “Open Parens”. Incorporating commands for punctuation becomes second nature for dictators with Arrendale speech programs and is recommended, but not necessary. Speech users are grateful to reduce the tabbing, indents, shift keys and retyping so common when transcribing directly into an EHR text box.
5. Pre-loaded vocabulary
It is well known that the US and Canada are cultural melting pots with ethnic names for cities, streets and residents coloring our vocabulary. Providers dictate proper names of clients, streets, facilities and nearby towns on a daily basis and these are often the most challenging for speech engines to get right the first time. Just think of the Indian “Homosassa Springs” (River of Fishes) and the French “Gervais” (Spear) Street. One way to reduce correcting frequent non-English based names is to pre-load them into the speech engine. A short list of nearby unique, geographic names proactively added to the speech engine is well worth the effort during the transition to cloud speech.
The clinical documentation migration to speech technology has matured in acute care, ambulatory and behavioral health settings in recent years. This short video link, provides a concise summary of Arrendale’s speech capabilities. Additionally, technology savvy providers are gratified when a small upfront time investment yields hours of time savings with their daily note making. In the words of user John Dorsky, MD, FACS, CPHIMS, CHTS-CP “Imagine stealing back some of the time you spend in your EHR every day. Integrating voice recognition commands into your workflow with SayIt will improve your efficiency.”
Whether you are considering a replacement of your older desktop speech recognition software or you’re a first time dictator, Arrendale is available for a conversation about your specific workflow goals and suggests a no cost trial. Call 800.344.1323 to learn more.