Kalyadina S.A., Ivanova M.O., Uspenskaya O.S., Manikchas G.M., Chikrizov S.I., Sokolova L.A., Wang. Sh, Ionova T.I., Cleeland C., Novik A.A.

Validation of the Russian versions of questionnaires for symptom assessment in cancer patients: Brief Pain Inventory (BPI-R), Brief Fatigue Inventory (BFI-R) and M.D. Anderson Symptom Inventory (MDASI-R).

The aim of the study was to develop Russian versions of the Brief Pain Inventory (BPI-R), the Brief Fatigue Inventory (BFI-R) and the M.D. Anderson Symptom Inventory (MDASI-R).
The validation sample consisted of 226 cancer patients. Cultural and linguistic adaptation was performed using standard forward/back-translation procedure. Validation analysis plan involved establishing construct, convergent, and known-group validity. To establish reliability Cronbach alpha coefficient was assessed. Construct validity was demonstrated by confirmatory factor analysis.
We established convergent validity by comparing items of tested versions with subscales of the Russian version of SF-36. Known-group validity was demonstrated by correlating the mean of the severity items of Russian versions with ECOG performance status.

The MDASI-R factor analysis found 3 underlying constructs for symptom items with Cronbach alpha coefficients of 0,86, 0,68 and 0,90, respectively. BPI-R and BFI-R factor analysis resulted in a two-factor solution.
These results satisfied the criteria of reproducibility, interpretability and fit in confirmatory setting. Cronbach alpha coefficients for the BPI-R and BFI-R items were higher than 0,9 and indicated reliability of self-report instruments. Symptoms severity was significantly correlated with the SF-36 subscales for all three tools.
These results allow to consider convergent validation as sufficient. Significant differences in symptom severity and interference levels by ECOG performance status were found supporting known-group validity of the questionnaires. Thus, the Russian versions of self-reported tools for assessment severity and interference of the leading symptoms were developed.
These measures are feasible, reliable, and valid assessment tools for use among the Russian cancer population. Questionnaires are appropriate instruments for clinical practice as well as for clinical studies.