I am a dental student writing on what we here (Sweden) call a "D-essay" it is a part of our exam and supposed to be practice in/an introduction to research and using scientific methods. We have gotten a couple of lectures on statistics, but not enough.
I am writing about reduction in pain from a certain treatment. I have studied medical records. It is not an active study with a control group and the works. Just looking backwards in the charts for a period of 1 year from one surgeons treatments. It is a very small sample size. Only 24 patients, each treated between 1-3 times.
I get confused on how to treat the data. In some forums and literature the numerical pain scale (NPS) i said to be qualitative and ordinal. In other places it is said that the NPS should be considered a ratio scale.
In an article by Farrar et al it says this: "In addition, long standing clinical experience and experimental evidence suggest that patients tend to describe their pain as a percentage change throughout most of the scale, i.e. a change from 9/10 to 6/10 (absolute change 3, percentage change 33%) is likely to be equivalent to, rather than more than, the change from 6/10 to 4/10 (absolute change 2, percentage change 33%). Price et al. (1983) have recommended that the pain scale be used as a ratio scale."
In other places it says the opposite. In an article by Tae Kyun Kim (Practical statistics in pain research) is says:
"For the NRS, it is obvious that it is ordinal scale. It means that nonparametric statistics should be applied for the NRS scale. When it come to the VAS score, however, there is something to be discussed about the scale and statistical methods."
Can I just "ignore" the type of scale it may or may not be and just use nonparametric statistics and thus be "on the safe side"?
I have also found research indicating that a reduction in pain by 30-50 % in significant. Previously mentioned article also states: "In these situations, the t-test or linear regression methods may provide sufficient power to detect even small differences in mean changes in pain (or percentage changes) between groups. However, situations might arise in which a particular treatment produces a substantial benefit in a moderate proportion of patients (say 50%), but no change, or even a perceived worsening, in others. This would lead to a bimodal distribution of change scores, violating the statistical assumptions underlying such tests."
How should I go about analyzing my data? I am confused on what tests to use an conclusions to draw. Just looking at the data it "looks good". Out of the 24 patients 21 got a pain reduction over 30 % and many of them over 50 % reduction. But how do I know if it is statistically significant and if it is or isn't normally distributed?
Sorry that my level of knowledge in this ares is limited. It is very hard to grasp for me.
Very thankful for any help on the matter.
I am writing about reduction in pain from a certain treatment. I have studied medical records. It is not an active study with a control group and the works. Just looking backwards in the charts for a period of 1 year from one surgeons treatments. It is a very small sample size. Only 24 patients, each treated between 1-3 times.
I get confused on how to treat the data. In some forums and literature the numerical pain scale (NPS) i said to be qualitative and ordinal. In other places it is said that the NPS should be considered a ratio scale.
In an article by Farrar et al it says this: "In addition, long standing clinical experience and experimental evidence suggest that patients tend to describe their pain as a percentage change throughout most of the scale, i.e. a change from 9/10 to 6/10 (absolute change 3, percentage change 33%) is likely to be equivalent to, rather than more than, the change from 6/10 to 4/10 (absolute change 2, percentage change 33%). Price et al. (1983) have recommended that the pain scale be used as a ratio scale."
In other places it says the opposite. In an article by Tae Kyun Kim (Practical statistics in pain research) is says:
"For the NRS, it is obvious that it is ordinal scale. It means that nonparametric statistics should be applied for the NRS scale. When it come to the VAS score, however, there is something to be discussed about the scale and statistical methods."
Can I just "ignore" the type of scale it may or may not be and just use nonparametric statistics and thus be "on the safe side"?
I have also found research indicating that a reduction in pain by 30-50 % in significant. Previously mentioned article also states: "In these situations, the t-test or linear regression methods may provide sufficient power to detect even small differences in mean changes in pain (or percentage changes) between groups. However, situations might arise in which a particular treatment produces a substantial benefit in a moderate proportion of patients (say 50%), but no change, or even a perceived worsening, in others. This would lead to a bimodal distribution of change scores, violating the statistical assumptions underlying such tests."
How should I go about analyzing my data? I am confused on what tests to use an conclusions to draw. Just looking at the data it "looks good". Out of the 24 patients 21 got a pain reduction over 30 % and many of them over 50 % reduction. But how do I know if it is statistically significant and if it is or isn't normally distributed?
Sorry that my level of knowledge in this ares is limited. It is very hard to grasp for me.
Very thankful for any help on the matter.