The reported TII data quality flag is determined using a preliminary algorithm that evidently does not capture all outliers or problematic intervals.
This situation is especially problematic for intervals that are designed to cover a large proportion of the population with a high degree of certainty.
Time between the different interviews is problematic since short interval are prone to recall bias, while long intervals risk being associated with the clinical evolution of the subject evaluated.
Even when the training data are treated as a random realization from nature's joint probability distribution, conventional statistical tests and confidence intervals are problematic for random forests and stochastic gradient boosting.
Because it was presumed that, as usually, the measured medical costs would follow a skewed distribution, a normality assumption would be problematic when estimating confidence intervals.
Small sample sizes or expected differences in benefit or risk can lead to no solution or problematic solutions for confidence intervals.
Confidence intervals are problematic when applied to incremental cost-effectiveness ratios, as costs and effects can fall into more than one quadrant for the cost-effectiveness plane (Briggs, 2001; Briggs et al., 2006).
Major problems with this approach are: a) the risk-benefit ratio is difficult to interpret and extrapolate outside the trial generating the information, and b) assessing uncertainty of a ratio is difficult because estimation of confidence intervals is problematic [ 9- 11].
The use of the confidence interval is also problematic because, as the sample size increases, more and more units will have values significantly different from the mean.
Consequently, the practice of using the ratio between number of deaths and number of years when comparing results across different study intervals is a problematic procedure (Timko et al., 2006).
However, a shorter time interval can also be problematic, because participants might remember their answers to the first questionnaire and be inclined to duplicate them [ 28].
