a n = number of subjects with endpoint b See Table 5 for further definition of endpoint c Where possible exposure is classified as "current" or "former" with exposure given as "use" only where the source paper did not clearly distinguish how former users were considered.
However, the information on study endpoints given to subjects was limited to minimise potential conscious influence by participants on cough and urge to cough.
The corresponding non-exposure is never or non-use to the same type of ST, except where indicated b Number of exposed subjects with endpoint (where available) c Tests are unadjusted for any potential confounding variable, except where stated.
The corresponding unexposed group is never for ever, and non-current for current, except where indicated b Number of exposed subjects with endpoint (where available) c Tests are unadjusted for any potential confounding variable, except where stated.
To test the aim-point assumption, we examined, for each subject, the endpoint distribution at each target location and tested it against the hypothesis that the mean end points fell on the center of the target (Hotelling's <img src="http://journals.plos.org/plosone/article/asset?id=info?doi/10.1371/journal.pone.0008228.e180.PNG" class= inline-graphic"/> test).
For the survival endpoint, subjects alive at 5 years were compared to subjects with death up to 5 years, and censored subjects (n = 12) were ignored from testing analysis.
In the CF, subjects produced the endpoint forces which were almost identical both with and without vision.
This was tested by quantifying the error with which subjects tracked the endpoint target, then comparing this error across the tested postures.
Therefore, we quantified the influence of pronation/supination in two subjects by estimating endpoint stiffness with the forearm supinated to 45°, pronated to 45°, and in the neutral position for a single arm posture (elbow flexed to 90°; shoulder elevated to 70°).
The selective increase in x-stiffness was seen under both visual feedback conditions which can be interpreted to mean that subjects adapted their endpoint impedance to attain stability in the DF in a similar manner whether or not online visual feedback was available.
