The article reviewed here is ‘Can Wireless Text Messaging Improve Adherence to Preventive Activities? Results of a Randomised Controlled Trial’ by Mihail Cocosila and colleagues. As the title suggests, the aim of the study was to determine if text messaging could influence health behaviour. In brief, the researchers conducted a randomised controlled trial in which half of the participants received text message reminders to take vitamin C and the other half did not. The primary outcome was the number of ‘pills’ taken. From the abstract, it can be seen that there wasn’t a significant difference between the two groups in the number of ‘pills’ taken. There was however a significant correlation between the number of return texts by participants and the number of ‘pills’ they had missed. So the main questions here are whether text messaging isn’t very helpful in improving health behaviour and whether information on the returned text messages is both valid and if so – is it useful?
Before turning to these questions, I was interested on reading through the introduction to see that there is an International Expert Forum on Patient Adherence and also to hear the researcher’s views on the potential health utility of text messaging:
‘Wireless text messaging applications have qualities that make this mobile service suitable for some interventions to address patient adherence through portable devices like cell phones: ubiquity, low intrusiveness, low cost and relative simplicity‘
Returning to the questions above, the first is – can we draw the firm conclusion that there is no direct relationship between health behaviour and use of text messaging as in this study? Well there are a few reasons why these results might not exclucde a beneficial use of text messaging for health behaviour. Firstly this is a small study and there is always the possibility that a larger replication study will produce different findings. Secondly it is difficult to know how taking Vitamin C can generalise to other health behaviours. Taking Vitamin C as a general and hypothetical preventative measure differs from other preventative measures such as exercise and also from taking prescribed medication for illnesses. Thirdly this is not representative of the general population (in the UK) as the participants were on average 23 years old. Also particpants were paid to take part and this financial incentive may have influenced their behaviourand differs from the need to take prescribed medication for an established illness. Fourthly in terms of the results – the researchers actually find and report a significant difference in adherence (change from baseline consumption of Vitamin C to endpoint) between the intervention and control groups using an ANOVA. Indeed the p value is very highly significant (p=0.001). The researchers note this might not be a meaningful difference as it amounts to an extra tablet each week. However they had not identified what would be a meaningful difference. So statistically significance is reached but it’s meaning is difficult to interpret. In terms of the number of tablets missed in the final week there was no statistically significant difference. So in conclusion, while an extra tablet a week may not be considered meaningful by the researchers there is some scope for arguing either way particularly as this a small study (n=99) which may be insufficiently powered to detect a ‘meaningful’ difference (there is an additional difficulty in that prior data would have been needed to determine a satisfactory sample size but this is an innovative study).
The second question is can we draw useful conclusions about the returned information. Here the intervention group returned information to the researchers – responses to the researchers text messages. However it was a bit of an unfair comparison as the control group didn’t have any text messages to respond to! In my opinion this part of the study lacks an appropriate control.
There were many more issues in the study which were considered – there were detailed statistical analyses, inclusion and exclusion criteria, a Java script for random allocation (which however was treated as a black box as far as I could tell), an increase in adherence rate in the control group (which perhaps represents a placebo effect – making no judgements about whether this is statistical or psychological) and controlling for overestimation.
In conclusion, this study looked at an important emerging healthcare phenomenon which incorporates mobile phone technology and while having a small sample size, can be used to inform power calculations and in meta-analyses.
Mihail Cocosila, Norm Archer, Brian Haynes and Yufei Yuan. Can Wireless Text Messaging Improve Adherence to Preventive Activities? Results of a Randomised Controlled Trial. International Journal of Medical Informatics. 78. 230-238. 2009.
You can follow ‘The Amazing World of Psychiatry’ twitter by clicking on this link
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).
If you have any comments, you can leave them below or alternatively e-mail firstname.lastname@example.org
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