more_vert
DESPITE the zeal with which electronic systems are being rolled out in health care, there’s still scant evidence of their benefit in the real world of our busy and chaotic wards.
That’s not to say we don’t believe technology has a valuable contribution to make. In fact, I was involved in recent research that showed clinicians using electronic information technology (IT) systems in emergency departments find they enable faster and better-informed decision making, hopefully leading to improved patient outcomes.
But the problem remains that electronic systems are being introduced without a clear picture of how they impact on often entrenched work practices and staff interactions.
At issue is not the technology itself, or its undeniable potential to improve health care processes — it’s the social issue of how people work and, in turn, how that affects patient care.
One of the main concerns voiced by doctors when a new IT system is introduced is that it just means more work for them. Whose job is it to enter data into the system, which is essentially a clerical role? Will this data entry place more demands on doctors’ valuable time, and will it detract from the time they spend with their patients? And who will benefit from electronic clinical documentation?
Even if clinicians can appreciate their potential benefits, many find these systems hard to use. They worry that their introduction will lead to the duplication of tasks, requiring them to enter data into both a computer and an existing paper record system. They are also worried that the technology is changing familiar workflow patterns.
For example, a study conducted in the emergency department (ED) of a community hospital in the US found that using computerised decision support to evaluate suspected pulmonary embolism resulted in a 4.4% increase in the rate of computed tomography (CT) angiography. However, doctors did not like the system — they said it meant they spent too much time on the computer — and eventually the decision-support system was removed.
Our research serves to allay at least some of clinicians’ most common fears. In what we believe to be the first systematic review of the English-language literature presenting quantitative data on the effect of computerised provider order entry in EDs, we have shown that the electronic systems did not decrease time spent with patients.
While nurses spent up to 16.2% and physicians 11.3% more time on computers, this was time taken from non-patient activities.
The review also confirmed that using electronic systems substantially improved laboratory turnaround time. Where decision-support systems were included — for example, proactive alerts around test follow-up — compliance with guidelines was consistently improved with significant decreases in prescribing errors, potential adverse drug events and the prescribing of excessive dosages.
Surprisingly, of the few studies that have been conducted into health IT in emergency departments — just 22 worldwide, all of them since 2006 — none have considered the systems’ effect on clinical work processes.
IT systems potentially disrupt workflows and practices across the entire hospital. These systems need to be evaluated in terms of their usability and their unintended consequences to people’s work patterns.
In addition, there has been little work to address the cost-effectiveness of these systems, despite the millions of dollars being spent by health departments internationally to install them.
One of the problems with evaluating IT is that people tend to measure performance in the same way as they have always measured it in the past. We need better, more patient-centred ways of judging its impact.
For a correct evaluation of the impact of electronic systems, we need to take a step back and understand how IT affects the complex interactions between all the players within the ward and across the hospital, and whether the net gain is positive.
If clinicians perceive that an electronic system is making more work for them without delivering quantifiable benefits, then they will be slow to engage with it and the system is not likely to succeed.
A system’s usability and functionality for clinicians is as vital to its chances of success as its technological prowess.
Associate Professor Andrew Georgiou is with the Centre for Health Systems and Safety Research (CHSSR), Australian Institute of Health Innovation, University of NSW.
Dr Sue Ieraci makes some valuable points. They reinforce many of the concerns expressed in the MJA Insight piece (Andrew Georgiou: Can you work with IT? 6 May 2013). The evidence outlined in this piece was drawn from our systematic review in the Annals of Emergency Medicine (see abstract hyperlink provided in the MJA Insight). This review was of quantitative literature (1990 – May 2011) regarding the effect of Computerised Provider Order Entry systems on clinical care and work processes. The review assessed the evidence within the framework of the US Agency for Healthcare Research and Quality-supported consensus conference which proposed a research agenda to guide the consideration of health IT (see Handel et al. 2011, Acad Emerg Med 18:e45-e51). Among the factors identified in this research agenda were patient flow and work integration, decision support systems, safety-critical computing and handoffs and interoperability. The findings from the systematic review are discussed with reference to this research agenda. The review also highlights important gaps in the evidence base, and identifies areas where it needs to be supplemented by studies dealing with some of the clinical, technical and organisational dimensions of the emergency department.
Prof. Georgiou says “Surprisingly, of the few studies that have been conducted into health IT in emergency departments — just 22 worldwide, all of them since 2006 — none have considered the systems’ effect on clinical work processes.” and yet there was one such study in a recent MJA. Even Prof Georgiou’s own work has found “there was difficulty incorporating the use of the ED information system with clinicians’ work, particularly in relation to increased task complexity; duplicate documentation, and computer issues related to system usability, hardware, and individuals’ computer skills and knowledge.” There is reason to believe that the current information systems used in EDs, in a time-critical, intense and rapid-turnover environment, do take clinicians away from patient care. Writing and speaking are still faster than interacting with a standardised information recording system. The time-limiting step is not typing speed but the requirement to log on and off and negotiate multiple screens and steps. Clinicians need to spend minimal time on documentation and order entry, with some more efficient process (be it scribes, dictaphone or voice recognition) transferring the information into the electronic record.
More worrying is the $millions spent on non-delivery of clinically useful IT tools after so much investment by the public eg HealthSmart. The disconnect between government, IT, clinicians and research is tragic. Decision assist tools will improve take up of evidence based practice, and when the software is sensibly written on stable reliable databses, the tools can be used as useful audit tools essential to identify and rectify gaps in practice. The profession and our patients need the IT industry to recognise there is a crisis in lack of effective and efficient collection of clincal data “on the run” and this would have a tremendous impact on idetifying and correcting gaps in care.
The industry’s obsession with the production of “the perfect product” needs to be replaced by a willingness of qualified software developers to the clinical issues and being prepared to work on a continuously improving product by liasing on a regular basis with “coalface” clinical users
Electronic medication systems need to provide prescribers with a clinically useful point of care description of drug interactions, immediate access to evidence based therapeutics, and software developers need to listen to prescribers, not just pharmacists. Electronic methods of requesting investigations in hospital need to be linked with efficient systems to deliver results and verify they have been received. Electronically recording relevant clinical features and differential diagnoses will facilitate clinical audit, communication within the hospital, link to discharge summaries and facilitate communication with community providers. The time to break down clinical-IT-government barriers was 10 years ago