After Hours Mashes Up Voice
After Hours Doctor’s Office, a mashup of Tell Me VxML (interactive voice response), Strike Iron Global SMS (text messaging) and Amazon Web Services (hosting), provides an example of how basic communication services can be integrated with applications to create new services with quantifiable business value.
The mashup also illustrates how easy it is for new communications-centric applications to be created by third party developers, hosted commercially by application providers and provided directly to end user enterprises and consumers without necessary involvement by network service providers.
“I wrote the application in a matter of days, and I didn’t need anything but my credit card to charge the $50.00 for the SMS messages. I didn’t need any contracts or hardware,” says Thomas Howe, owner of Thomas Howe Consulting.
It also, therefore, suggests why it is so important for service providers with platforms based on IP Multimedia Subsystem to spend significantly more time thinking about application development. Developers might not need IMS to create tomorrow’s communication apps. And if the intention is to avoid becoming a “dumb pipe,” Web 2.0 will do that, barring significant movement by “carriers” to integrate IMS networks with Web apps.
Basically, the application transcribes a doctor’s office inbound patient voice mails into text. The messages then are sent as text messages to a doctor, after a bank of nurses has performed “triage” on the calls, sorting the cases into “urgent” and “can wait until tomorrow” baskets.
Say a patient doesn’t feel well and calls his doctor to schedule an appointment for the next day. An IVR answers, asking if the call is a routine business matter or a health issue. If the caller reports it is a health issue, the IVR asks if the caller is on a mobile phone. The IVR then takes a voice message and hangs up.
As soon as the call is over, the patient gets a text message on his cell phone telling him that the call was received, and that the call is being forwarded to nurses. If the caller is calling from a landline phone, the patient gets a return call.
The message is sent to nurses who determine whether the case is urgent or not. If they think it is routine, they indicate that on their console, which results in another message being sent to the cell phone telling the patient that the matter is probably routine, and they would get a call in the morning.
If urgent, the patient would get a message like “A nurse thinks you need to speak with a doctor. We are looking for one now – stay near the phone,” says Howe.
“Urgent issues are forwarded to the doctor’s mobile, with a summary of the call prepared by the nurse. “The doctor’s message would be ‘Mr. Kraus – 40 WM - left side weakness, nasuea -508 364 9972.’”
The doctor would be able to press the send button on his phone to call the patient.
Howe wanted to “demonstrate that integration of real time communications into the business process saves money for the business, while increasing quality of service for the customer.”
He also wanted “to explore the use of artificial intelligence, by way of Amazon Turks, and its possible role in telephony applications.” The basic application was created in about 40 hours, except for the outbound IVR portion, which Howe says he isn’t enough of a programmer to pull off.
There are some issues here for conventional service providers. It didn’t cost Howe much to create the app. As a commercial venture there would be some incremental spending, based on usage, paid to a hosted IVR service. But Howe says the big points are creation of the app quickly, with no large capital investment and no involvement by a network service provider. “Traditional carriers will get their clocks cleaned” by apps such as this, he says.
“According to the National Center for Health Statistics, 55 percent of the 90 million visits to hospital emergency departments in the United States in 1996 were unnecessary,” says Howe.
“According to Blue Cross, Blue Shield, an average emergency room visit costs $1,049.00, whereas the average doctor’s office visit costs $149.00,” Howe notes. So “we have a $44 billion dollar problem with unnecessary emergency room visits.”
Getting the application to work in the real world would require creating a bank of on-call nurses. Howe says that’s where the Amazon Turk system comes in. Turk could aggregate tens of thousands of stay-at-home nurses raising small children.
The incentive is the ability to make several dollars in two minutes by listening to a voice message and prioritizing it. Insurance providers might provide the business model, paying the system to keep patients out of emergency rooms. IP
