Digital pen project draws interesting conclusions
A research project looking at the use of digital pen and paper technology by nurses and allied health teams in a Victorian hospital has revealed some interesting findings, not the least of which is that clinicians and computer programmers speak very different languages.
The digital pen and paper project, led by Virginia Plummer of Monash University's School of Nursing and Midwifery, was designed to see if the use of digital technology could improve patient care and patient flow by providing real-time information and streamlined clinical decision-making.
Carried out at Ward 5GN at Frankston Hospital on the Mornington Peninsula, the project studied patients with a fractured neck or head of femur, the average age of whom was 91. As Dr Plummer told the Health Informatics Conference (HIC 2013) in Adelaide last week, with the first patient studied aged 103, “this is an indication of things to come”.
The digital pen and paper technology, provided by the Print Media Group, is considered a "bridging" technology to a full electronic medical record. As Dr Plummer told Pulse+IT last year, nurses and allied health professionals in public hospitals often don't have access to computers, and while tablets and similar devices are becoming increasingly common, nurses in particular don't often get their hands on them.
“In large public hospitals, access to computers is not easy," she said. "The reality is that nurses don't have PDAs, they don't have iPads, they don't have access to the computers when they need to and with the digital pens, they can carry them with them from the beginning of the shift to the end.”
The digital pens were used as part of the nurses' role in recording data for the fractured neck of femur clinical pathway. Dr Plummer said Peninsula Health used over 300 pathways, most of which are still paper-based.
“We chose [fractured NOF] because it is a routine pathway and it is relatively easy to observe any variance in the pathway,” she said. “The aim was to study real-time data capture and retrieval of clinical pathway information at the point of care, and to analyse how it could be used for clinical decision-making and resource management.
“The objective was to see how patient care could be improved using new point of care technology that wasn't previously used in the ward, and to understand how bed management could be improved by using electronic clinical pathways.”
In association with the Print Media Group and NEC, the team designed a back-end, browser-based proprietary database and integrated it with the front-end digital pen and paper technology to build what the team called a digital clinical pathway management system (DCPMS).
The digital pen and paper allowed nurses to simultaneously record clinical data on paper-based forms while capturing and authenticating that data digitally, which was then uploaded using wireless technology.
“The front-end database is essentially an interpretation engine that recognises handwriting and converts it into digital letters, numbers and words,” Dr Plummer said.
“The software and subsequent data was stored in a pre-defined directory on a secure computer. We put in our own server which was quite separate from the server that the rest of the hospital had their data on.
“Digital pen and paper technology working in real time can be integrated with the existing mobile data management environment and this technology seeks to address the issue of time spent entering paper-based data into computer systems.”
NEC also built a dashboard that could be viewed in the ward or on other devices. When the data is captured and displayed in real time, any variations from the clinical pathway can be seen and interpreted immediately, and interventions made.
When clinical pathways are done in just paper-based form, the information is often not collated for analysis for months, meaning it can't be used for actual interventions. In addition, the data from clinical pathways is useful for bed management, as it can help predict length of stay.
There were some interesting findings from the project, and missing and unreadable data was one, as expected, Dr Plummer said. “At project finalisation, 18 per cent of the UR numbers were missing but traceable. 34,000 fields were interpreted but 1880 had unreadable fields. Thirty-one were written by a regular pen in the digital field – so the nurse had picked up a regular pen and didn’t pick up the digital pen.”
There were also some pen stroke issues, she said. When they asked the patients what their pain was on the 10-point pain scale, the nurses needed to write the numeral, for example ‘5’, in the box. However, they commonly wrote 5/10, which they do on other medical records and this showed up digitally as a pain score of 5110.
"Similarly with oxygen saturations – users needed to put in the percentage as a numeral, for example 94, but they put in 94%, so it showed up as 94010. Likewise this was the difference between what they were required to write for the project and what they usually write on a medical record."
The study also shed light on patients' reporting of pain and their experiences with care. Older people in particular are often reluctant to report pain, for a number of reasons.
“One example is of a patient who told the nurses that he had no pain, but when the nurse unit manager could see changes in vital signs on the dashboard, he attended the patient and enquired 'are you pain-free?'
"The patient explained that he under-reported his pain because he was receiving pain relieving medication 'but it wasn’t making any difference'."
The nurse then offered stronger pain relief.
Dr Plummer told the conference that the project was a successful proof of concept, as the researchers achieved what they wanted to in trialling the use of digital technology. However, it was also a typical information technology project where clinicians and programmers, although keen collaborators, still tended to think, work and speak in their own languages, she said.
“Even though we were learning a lot from each other – for example, the clinicians would learn about 'GUIs' and the programmers would learn about the implications of wound 'ooze' in the design phase – both groups misunderstood some very important points. At project end I was given a memory stick with only technical data for statistical analysis and when I asked for all of the clinical data, after three years, working very closely together, I was asked 'did you want all that?'
"I did. I needed it to inform our practice as clinicians, and preferably not in code."
The clinical team also found out that it was never truly real-time data displayed on the dashboard, and the programmers didn’t understand the critical need for that. They tended to only work business hours, she said.
“We found out after the study that there was up to a 12-hour lag, and there was a 48-hour lag between Friday and Monday, over the weekend.”
There was, however, some valuable information that could be used in bed and resource management. “Length of stay dominated the whole project, and it was informed by pain and vital signs. Wound breakdown and mobility were not explored for impact as planned because the pain and vital signs were the main indicators of outcome.
“Now with the NEAT targets, greater precision in bed management for emergency department admissions is required. With the use of the DCPMS, the status of beds of patients that discharge early or deteriorate and are not discharged as expected will be readily identifiable on the dashboard and potentially by the unit manager, and the bed manager remotely.”
Posted in Allied Health