Previous - Next
The working group recommended that future publications on DLW should contain as much methodological information and detail of the intermediary steps in data reduction as possible. A good paper should contain the following:
Methods
Explicit description of:
1) The dosing procedure - including mode of administration; timing; physiological state of the subjects; amount of dose given; equilibration period; any corrections for isotope losses during equilibration and any precautions taken to ensure completeness of isotope ingestion (particularly in studies of babies and young children).
2) Mass spectrometric procedures - including instrumentation; standards; estimates of analytical precision for both 2H and 18O within the ranges used; and an assurance that the dose enrichment had been measured (and not assumed from the manufacturer's value).
3) Calculations - including details of formula used; exact description of how plateaus, intercepts and slopes were derived; method of deriving a single pool space if this was used; calculations of precision where appropriate and details of any cut-off used to screen out unacceptable estimates; details of inspection of residual plots and extent of covariance in 2H and 18O values.
4) Fractionation corrections - including method of estimating fractionated proportion 'x'; statement of whether individual or group mean estimates were used; a statement of the mean value derived for 'x' if individual predictions were used; details of any environmental measurements pertinent to the determination of 'x' (ie ambient temperature, humidity and working conditions); listing of the exact values assumed for f1, f2 and f3.
5) RQ assumptions - including details of specific derivations from dietary assessment with or without corrections for energy imbalance during the measurement period; justification for group mean estimates of RQ if used.
Results
1) Tabulated 'raw' data - including (unless the data set is very large) the individual estimates of ND and NO, kD and kO, rH2O and rCO2; individual values for ND/NO if the separate estimates of volumes are not listed; individual estimates of precision (where calculated).
2) Tabulated final data - including individual estimates of total energy expenditure and BMR wherever possible.
Adoption of these recommendations
will ensure a clear and comprehensive presentation of DLW results. It will allow readers
to assess the quality of the data being presented, for instance by examining the values of
ND/NO and the estimates of precision. In certain circumstances it
may also permit them to recalculate the results using their preferred formula in order to
make a more direct comparison with their own data. It is appreciated that certain journals
may be resistant to the concept of publishing intermediate data from which the final
estimates of expenditure are derived. However workers in this field can exert considerable
influence on editors as reviewers. Reference to the recommendations contained in this
document may also help to persuade editors of the value of explicit and complete data
presentation.