An editorial error occurred in this article by Caroline Sanders and colleagues (31 October, pp 1191-4). Table 3 (referred to at the end of the Results section, p 1193) was omitted. The table is as follows:
Table 3.
Indicators of quality of reporting on quality of life in 67 randomised controlled trials. Values are numbers (%) of trials unless stated otherwise
| Indicator | Principal end point of trial
|
Difference (P value)* | |
|---|---|---|---|
| Quality of life (n=23) | Other (n=44) | ||
| Methods | |||
| Used established instrument | 20 (87) | 28 (64) | 0.044 |
| Patient provided quality of life information | 21 (91) | 25 (57) | 0.004 |
| Response rate given | 19 (83) | 19 (43) | 0.002 |
| Results | |||
| Complete reporting of items and scores | 16 (70) | 15 (34) | 0.006 |
| Absolute differences given | 23 (100) | 33 (75) | 0.009 |
| Confidence intervals given | 5 (22) | 6 (14) | 0.40 |
| P values given | 21 (91) | 31 (70) | 0.052 |
| Median (range) sample size | 198 (36-4736) | 92 (11-1914) | 0.034 |
| Mean (SD) Jadad score as percentage of maximum | 47.8 (23.2) | 56.8 (23.6) | 0.15 |
*
Calculated by χ2 tests, except for comparisons of sample sizes and Jadad scores calculated by Wilcoxon rank sum test.