The article reviewed here is ‘Patterns and Changes in Prescriber Attitudes Toward PDA Prescription-Assistive Technology’ by Arun Vishwanath and colleagues. The authors look at the characteristics of ‘early’ and ‘late’ adopters of PDA prescription-assistive technology. Their interest in attitudes is explained in the introduction when they cite research suggesting that attitudes towards new technology is correlated with adoption of this same technology.
They selected 244 clinicians from an American academic tertiary care children’s hospital representing different professions but accounting for 90% of prescribers in the inpatient service. This high percentage suggests that the sample is likely to be representative of the inpatient prescribers. The prescribers had recently undergone training in the use of the PDA software. Commercial software was used and supplied by the company producing the software. Hospital and staff-owned PDA’s were used with the software and all PDA’s used in the course of the study were inspected to ensure that they met specific hardware standards. These standards in turn would ensure that the software was appropriately ‘responsive. Details of the training were given.
Assessments of clinician’s attitudes towards the technology before and after implementation were assessed using a Likert-scale questionnaire. The authors then state that the questionnaire design and administration was conducted by an external company that had previous experience with this type of research. Regarding the various tools that were used pre and post-implementation, the authors write that
‘All the measures used in the study were valid measures drawn from prior technology acceptance research and modified to suit the clinical context‘
I wasn’t sure of the type of validity that was being referred to and how the previous research tools were modified. They also wrote that
‘All multi-item measures were reliable and achieved an acceptable alpha level‘ (alpha greater than 0.85).
A number of questionnaires were used and most used closed-questions but one gave the option for open-ended responses. Although I might have missed this I couldn’t find a clear definition of early and late adopters but the authors state that these groups can be distinguished on the basis of the results of a ‘five-item measure of the clinician’s attitude twoards PDAs, and a measure of the clinician’s likelihood to adopt the PDA within the next year’. Under the statistical analysis the authors write that
‘The data were analy(s)ed using the combination of multivariate techniques…..Segments were derived by applying a two-step clustering algorithm. The resulting segments were validated statistically using t-tests (Boneferroni adjustments) and theoretically compared to the EA and LA profiles suggested by diffusion theory‘
This excerpt is quite information dense and contains lots of different and complex statistical processes which would involve judgments as to their suitability. I would be interested to learn more about the analysis that took place at this stage. The use of the Boneferroni adjustments reflects that multiple comparisons were taking place on the dataset and that relationships were being explored rather than primary hypotheses tested. Another question I asked on reading this section is what was diffusion theory and why was it being used in this analysis?
The difficulty that I had with this analysis was that it led to the stratification of the sample into early and late adopters of the technology. I would argue therefore that it is difficult to know how to apply this term in an intuitive way. For instance if I were to ask if a person were a late or early adopter I would need to refer to their exploratory statistical analysis of this group – I would argue that it becomes difficult to translate into other populations. If this is the case, then I would add that it means I don’t really know how meaningful this is to me. While intuitively we might know what early and late adopters of the technology might be this statistical meaning is very specific and defined by several layers of abstraction. If as a result I just say that early adopters are people that adopt the technology soon after its advent and late adopters otherwise, I will perhaps be able to apply this knowledge to other scenarios but as a result of the above arguments my conclusions may be markedly inaccurate.
Taking into account the previous arguments there are a number of observations that the authors have made about their groups of LA and EA’s. They identified the EA group as significantly younger than the LA group and tending to be residents rather than attending. On a number of measures the researchers found that after the intervention there were still significant differences between the EA’s and LA’s in their attitudes towards PDA’s. Essentially even after training early adopters generally had more positive attitudes towards PDA’s than late adopters.
In conclusion, I found it difficult to conceptualise the constructs of EA and LA’s. Even so, I could see that here was a method to create potentially useful categories on the basis of the statistical analysis and to identify significant associations which could then be used qualitatively. What I found interesting and to some extent predictable is that with interventions, people within one category tended to remain in that category. Perhaps this suggests that to have any impact particularly with attitudes there should be a sustained intervention, that such impact should be intended and that the systems have some advantage over contemporary systems. It is difficult for me to extrapolate further other than to say that attitudes do not necessarily have a direct translation into behaviours and triangulation can be helpful. As for the clinical utility of the PDA’s, this is a topic in itself and will be dependent on the hardware and software as well as the setting. Rapid changes in technology can also have an impact on the application of such studies.
Vishwanath A, Brodsky L, Shaha S et al. Patterns and changes in prescriber attitudes toward PDA prescription-assistive technology. International Journal of Medical Informatics. 78. 2009. 330-339.
You can listen to this post on Odiogo by clicking on this link (there may be a small delay between publishing of the blog article and the availability of the podcast).
You can follow the TAWOP Channel on YouTube by clicking on this link
If you have any comments, you can leave them below or alternatively e-mail email@example.com
The comments made here represent the opinions of the author and do not represent the profession or any body/organisation. The comments made here are not meant as a source of medical advice and those seeking medical advice are advised to consult with their own doctor. The author is not responsible for the contents of any external sites that are linked to in this blog.