Validity and reproducibility of a short food frequency questionnaire among patients with chronic kidney disease

BMC Nephrology

on behalf of CKD-REIN study investigators¹  Denis Fouque²  Guy Fagherazzi³  Sandra Wagner³  Françoise Clavel-Chapelon³  Bénédicte Stengel³  Douae El Fatouhi³  Aurélie Affret³  Julie De Chefdebien³  Courtney Dow³  Maryvonne Niravong³  Marie-Christine Boutron-Ruault³  Emmanuelle Correia³ 

[1] ;FCRIN INI-CRCT;Inserm U1018, Center for Research in Epidemiology and Population Health (CESP);

关键词: Chronic kidney disease;    Short food frequency questionnaire;    Validity;    Reproducibility;    Diet;    Dietary assessment;   

DOI  :  10.1186/s12882-017-0695-2

来源: DOAJ

【 摘 要 】

Abstract Background A balanced diet is essential to slowing the progression of chronic kidney disease (CKD) and managing the symptoms. Currently, no tool is available to easily and quickly assess energy and macronutrient intake in patients with non end-stage CKD. We aimed to develop and evaluate the validity and reproducibility of a new short 49-item food frequency questionnaire (SFFQ) adapted to patients with CKD. Methods The CKD-REIN study is a prospective cohort that enrolled 3033 patients with moderate or advanced CKD from a national sample of nephrology clinics. A sub-sample of 201 patients completed the SFFQ twice, at a one-year interval and were included in the reproducibility study. During this interval, 127 patients also completed six 24-h recalls and were included in the validity study. Main nutrient and dietary intakes were computed. Validity was evaluated by calculating crude, energy-adjusted and de-attenuated correlation coefficients (CC) between FFQ and the mean of the 24-h recall results. Bland-Altman plots were performed and cross-classification into quintiles of consumption of each nutrient and food group was computed. Reproducibility between the two SFFQs was evaluated by intraclass CC (ICC). Results Regarding validity, CC ranged from 0.05 to 0.79 (unadjusted CC, median: 0.40) and 0.10 to 0.59 (de-attenuated CC, median: 0.35) for food group and nutrient intakes, respectively. Five of the most important nutrients of interest in CKD, i.e. protein, calcium, phosphorus, potassium, and sodium had de-attenuated CC of 0.46, 0.43, 0.39, 0.32, and 0.12, respectively. The median of classification into the same or adjacent quintiles was 68% and 65% for food and nutrient intakes, respectively, and ranged from 63% to 69% for the five nutrients mentioned before. Bland-Altman plots showed good agreement across the range of intakes. ICC ranged from 0.18 to 0.66 (median: 0.46). Conclusions The CKD-REIN SFFQ showed acceptable validity and reproducibility in a sample of patients with CKD, notably for CKD nutrients of importance. It can now be used in large-scale epidemiological studies to easily assess the relations between diet and CKD outcomes as well as in clinical routine. It may also serve as a basis for the development of FFQs in international CKD cohort networks.

【 授权许可】

Unknown