If you have ever had to fill out a client charge-out sheet you will know just what a pain it can be to have to record what you’re doing every 10 minutes or so…and if you’ve EVER tried to get a patient to fill out a daily activity diary, daily pain diary – or any type of intensive monitoring record throughout the day, you will KNOW it just doesn’t fill anyone with joy. I was glad, then, to read Joan Broderick and colleagues paper today about using ‘End of Day’ reporting instead of moment-by-moment reporting of pain and fatigue.
Firstly, I’d like to quickly go over why it’s helpful to use moment-by-moment reporting instead of simply asking people to remember – and the basic reason is our rather erratic and biased recall of pain. As humans, we simply don’t have a terribly good memory for things like this, and instead we’ll draw on the most dramatic incident, or the most recent incident – or will try to come up with a estimated ‘average’.
The very best approach would be to have some sort of vacuum suction that grabs the information out of our brains as we go through the day, but in reality we don’t have this, so the next best thing is to randomly sample our ratings. The problem with this is the amount of interference with ‘real life’, the cost of programming some automated system to do it, or using humans as an alternative which is even more intrusive and expensive.
There are also some conceptual arguments suggesting that moment-by-moment sampling doesn’t account for the human process of ‘making meaning’ of a situation, but as the authors in this paper say ‘…while this is plausible, to date there is little empirical evidence to support this’ – and you know my views on empirical data!
Anyway, in this paper Broderick, Schwartz, Schneider and Stone did some of the hard work out of determining whether there is any significant difference between using moment-by-moment data (Ecological momentary assessment, or EMA), or asking people to give an estimate of their pain and fatigue at the end of the day (EOD).
They recruited 117 people attending a rheumatology practice and asked them to complete the following:
1. An electronic diary that randomly scheduled calls for the participants to report their pain and fatigue during the person’s waking hours (more details of the questions shortly)
2. An electronic diary that also asked the person to rate their experience at the ‘end of the day’.
Several pain measures were used: SF36, a measure of health used often in this type of survey, the Brief Pain Inventory, the Brief Fatigue Inventory, and the McGill Pain Inventory. They also completed a Visual Analogue Scale.
Basically, the electronic diary beeped randomly through the day for an average of 7 times, and asked the person to respond to several questions about their context, then complete the questionnaires (or items drawn from these). At the end of each day, they were also asked to complete the same ratings. The ratings were carried out over a month, but the hypotheses of this study were tested using the first seven days of data.
91% of the random ratings were completed, which is pretty good really. The researchers took care over the quality of the data they analysed in terms of how many responses were completed for the day, and found that participants completed 5.6 of the 7 assessments each day. I’ll review the rest of the results by describing the hypotheses being tested.
Hypothesis one was that there would be a high correlation between the person’s average EMA over a day, and their own EOD assessment that same day – and yes, there is a high correlation for most of the measures (between .92 and .90), the lowest was .71 for ‘energetic’ on the BFI.
Hypothesis two was that there would be a high degree of correlation between the average of 7 days EOD ratings with the average of the week’s EMA ratings. Once again, yes there is a high degree of correlation from between .95 for the pain ratings to .88 for the fatigue ratings.
The third hypothesis was whether EOD reports would accurately capture daily variations in symptoms over the week. Yes, the correlations, while lower than the previous ones, ranged from .71 – .80 for pain ratings, and .46 – .72 for fatigue. So not quite so accurate for fatigue.
What does this all mean for us?
Well, it seems that if we ask a person to record their assessment of their pain at the end of the day, it’s pretty well correlated with their moment-by-moment reporting of pain. This is great news! It means we can save people the hassle of having to write down, or use an electronic diary, their pain experience throughout the day if we’re wanting to know about changes over a week or so.
An interesting twist to the study is that the authors asked the participants to retrospectively assess their pain ratings (on the last day) over the past 7 days. The average within-subject correlation was only .29 for this, as compared with the correlation of .76 for the EOD ratings.
Collecting real-life information about people’s experience is a challenging task. To balance accuracy, intrusion and cost requires a bit of a juggling act, so it’s a really good thing to know that we can ask people about their pain experience once, at the end of the day, and end up with what we can view as an accurate report. We know that recalling information isn’t terribly accurate – and so, if we’re going to use recall (eg ‘what was your pain like over the last week?’) we need to make sure we use the same timescale at the beginning of the intervention, and again at the end, or alternatively, use an end of day report.
Interestingly, fatigue didn’t work out quite so accurately – so some caution is needed if you’re working with people for whom monitoring fatigue is necessary. It may also be less accurate for other subjective experiences like emotions or thoughts – but in the case of pain, if you need to ask about it – like an apple, once a day keeps the doctor at bay (or the clinician happy!).
J BRODERICK, J SCHWARTZ, S SCHNEIDER, A STONE (2008). Can End-of-Day Reports Replace Momentary Assessment of Pain and Fatigue? The Journal of Pain DOI: 10.1016/j.jpain.2008.09.003
Daut RL, Cleeland CS: The prevalence and severity of pain
in cancer. Cancer 50:1913-1918, 1982
Melzack R: The short-form McGill Pain Questionnaire.
Pain 30:191-197, 1987
Mendoza TR, Wang XS, Cleeland CS, Morrissey M,
Johnson BA, Wendt JK, Huber SL: The rapid assessment of
fatigue severity in cancer patients: Use of the Brief Fatigue
Inventory. Cancer 85:1186-1196, 1999
Ware JE Jr., Kosinski M, Turner-Bowker D, Gandek B:
How to score version 2 of the SF-12 Health Survey, Lincoln,
RI, Quality Metric Incorporated, 2002