Formulas | M group (126 pairs) | HS group (119 pairs) |
|---|
Bias | Limits of agreement | Bias | Limits of agreement |
|---|
1 | 25.0 | 2.6 to 47.5 | 19.5 | 5.1 to 33.8 |
2 | 6.0 | −18.2 to 30.2 | 0.8 | −12.9 to 14.5 |
3 | 16.3 | −7.5 to 40.2 | 11.3 | −2.1 to 24.8 |
4 | 17.3 | −6.2 to 40.7 | 12.6 | −0.6 to 25.7 |
- Data are presented as bias and lower to upper limits of agreement. The differences between measured and calculated serum osmolality were calculated in each data set. Bias was defined as the mean of difference between the measured and calculated values. SD of the mean bias was calculated according to the agreement between methods of measurement with multiple observations per individual. Upper and lower limits of agreement were defined as bias ± 1.96 SD
- Patients in M group or HS group received 125 ml of either 20 % mannitol or 3.1 % sodium chloride solution, respectively
- For each formula, the bias in HS group was statistically lower than in M group (p < 0.001). Within group comparison showed that, either in M group or in HS group, there was significant difference in bias between any of the two formulas (p < 0.001)
