David L. Doggett, Ph.D.
Senior Medical Research Analyst
Health
Technology Assessment and Information Services
ECRI, a
nonprofit health services research organization
5200 Butler Pike
Plymouth Meeting, PA 19462, USA
Phone: (610) 825-6000 x5509
FAX: (610) 834-1275
e-mail: [log in to unmask]
-----Original Message-----
From: Susan Stacpoole Shea [mailto:[log in to unmask]]
Sent: Thursday, September 12, 2002 3:06 AM
To: [log in to unmask]
Subject: Two feet or one personHello,This message was originally put to the Podiatry mail base. The author and I would appreciate any advice that is forth coming from the members concerned with Evidence based health.Thank you in anticipation,Susan Stacpoole-SheaBallarat, Victoria, AustraliaDear all,I'd like to pose a thorny question for the researchers on the mailbase. Thisquestion was initially bought to my attention by Lloyd Reed from QUT acouple of years ago and has stimulated much debate among my colleagues atUWS. My apologies for the length of this posting (the text is taken from apaper I've started writing).In many fields of biomedical research, information is collected on multiplejoints or organs from the same subject. For example, many ophthalmologystudies record data from both eyes, and in the case of foot and ankleresearch, data is often collected from both feet. This raises a significant,yet largely overlooked problem when it comes to statistical analysis. One ofthe fundamental requirements of statistics is that each data point mustrepresent an independent observation to justify being considered a "unit".In most cases, the unit of measurement is the subject, so if, for example,50 subjects are enrolled in the study, each observation recorded from eachsubject counts as a single unit, ie: n=50. However, if data is recorded fromboth feet, a major problem arises. What is the unit of measurement – asubject, or a foot? Do we have a sample of n=50 people, or a sample of n=100feet? A cursory examination of the foot and ankle literature reveals dozensof examples of statements like "We recruited thirty subjects (sixty feet)".From a conceptual viewpoint, it does seem a little odd to conduct researchinto individual feet rather than people, as clearly the way an individualfoot functions is dependent on the person attached to it. For example, thehealing rate of a surgical wound is strongly dependent on its blood supply,the pressure distribution under a foot is strongly dependent on the gaitpattern of the individual, and the pain experienced following localanaesthetic injection strongly dependent on the individual’s pain threshold.In each of these examples, it is likely that the degree of associationbetween right and left feet in the same subject would be far greater thanthe association between different subjects. Therefore, if both right andleft feet were counted as single independent observations, the researcher isessentially "double-dipping" their data, ie: counting each subject twice.Doing so will increase sample size and decrease variability in the data,thereby increasing the power of the study and increasing the likelihood ofdetecting statistical differences. But are these "significant" differencesreal?In order to demonstrate how the decision to pool or not pool right and leftfoot data can influence results, I have developed a dataset of "dummy" datafor 30 subjects (see below). For the purpose of discussion, the data can beconsidered to represent rearfoot motion values (in degrees) for 30 subjectswith and without foot orthoses for both right and left feet. Paired t-testswere then used to compare the "without orthosis" and "with orthosis"conditions for the right foot only, the left foot only, the average of theright and left feet, and with right and left foot data combined.Key to table:WOOR - without orthosis right foot, WOOL - without orthosis left foot, WOR -with orthosis right foot, WOL - without orthosis left footWOOR WOOL WOR WOL1 2 1 3 22 4 4 2 23 6 6 4 34 8 7 2 25 4 4 1 16 5 5 4 37 6 6 5 48 3 3 7 79 2 1 5 510 4 3 3 311 5 5 2 212 7 7 4 213 4 3 3 314 2 1 1 115 2 2 1 116 2 1 3 217 4 4 6 218 6 6 4 319 8 7 2 220 4 4 6 521 5 5 4 322 6 6 5 423 3 3 7 724 2 1 5 525 4 3 3 326 5 5 2 227 7 7 4 228 4 3 3 329 2 1 1 530 2 2 1 1For right foot data, there was no difference in rearfoot motion with orwithout foot orthoses (t29=1.83, p=0.077). Similarly, for left foot data,there was no difference in rearfoot motion with or without foot orthoses(t29=1.68, p=0.104). For the averaged data, there was no difference inrearfoot motion with or without foot orthoses (t29=1.82, p=0.079). Forpooled right and left foot data (thereby increasing the sample size from 30to 60), the paired t-test revealed a significant reduction in rearfootmotion when wearing foot orthoses compared to the without orthosis condition(t59=2.44, p=0.018).The results of this simple example clearly highlight the problems inherentin analysing pooled data from paired limbs. In the example provided, footorthoses had no effect on rearfoot motion on the right foot when analysed inisolation, no effect on the left foot when analysed in isolation, and noeffect when the two feet were averaged. However, when the right and leftdata was pooled, a significant reduction in rearfoot motion was apparent inthe orthosis condition. Although the difference was small in absolute terms,there is little doubt that such a difference would be reported as a"significant" finding. Thus, depending on whether data is pooled or not, itcould be concluded that foot orthoses either do influence rearfoot motionwhen walking or they do not.So my questions are as follows:What is the best approach for the statistical analysis of paired data?If we decide to analyse one foot only, which foot do we pick (and why)?Are there any situations in which analysing paired data is justifiable?Kind regards,HyltonHylton B. Menz