Table 2.

PhenX^a Toolkit recommendations^b

Measure	Features	Validity information	Limitations
5-Factor  Screenerc	Frequency of intake for specific  food items for past month Age- and sex-specific median  portion sizes are assigned	Indirect measurement error model  versus multiple 24 HRs for 260  men and 222 women in the  OPENsd study	■Estimation equations based on US National Health Interview Survey ■“…estimates may be less useful for populations with more ethnic diets, including Asian and possibly Latino populations”c
Calcium  intake  (mg/d)	Frequencies of 20 items used in  weighted regression equations	Deattenuated r = 0.59 for men  and 0.44 for women	■ Regression weights and selected foods likely to differ in different populations
Dairy food  (servings/d)	Frequency of milk and cheese  intake summed and used in  regression equations	Deattenuated r = 0.64 for men  and women	■ Does not include yogurt ■ Does not capture milk in coffee or tea
Fiber intake  (g/d)	Frequencies of 20 items used in  weighted regression equations	Deattenuated r = 0.52 for men  and 0.54 for women
Fruits and  vegetables  (servings/d)	Pyramid portions of 9 fruit and  vegetable items calculated and  then estimated in regression  equations	Deattenuated r = 0.58 for men  and 0.73 for women
Added sugars  (tsp/d)	Median tsp of sugar summed  from 4 items and then estimated  with regression equations	Deattenuated r = 0.68 for men  and 0.66 for women
%Energy  from fat	Questionnaire with frequencies  for 15 food items + 2 use  questions. Amount from added  solid fats is calculated and then  included with other frequencies  in weighted regression equations	r = 0.52 to 0.77 among men and  0.36 to 0.59 among women in the  BCCe study. Positive predictive  value for classification <30%  energy from fat ranged from 0.15  for women in Georgia to 0.71  for women in Rhode Island	■ Does not include oils, snack chips, most meats, fried chicken, pizza, etc. ■ Regression equations likely to be highly population specific ■ “If accurate assessment of diet… is essential, a more detailed instrument… may be warranted”e
Caffeine  intake	Frequencies of 13 items,  including coffee, tea, and soft  drinks; with small, medium, and  large portion options
Total dietary  intake	24HR using the ASA24f		■ Misclassification from day-to-day variability likely limits use in complex interactions

^aPhenX, Phenotypes and eXposures; collaborative funded by the National Human Genome Research Institute.

^bRecommendations for food intake only.

^cPhenX Toolkit; https://www.phenxtoolkit.org/index.php?pageLink%20=%20browse.protocoldetails&id%20=%2050202.

^dOPEN, Observing Protein and Energy Nutrition study.

^eBCC, Behavioral Change Consortium (105).

^fASA24, Automated Self-Administered 24-Hour Recall; http://riskfactor.cancer.gov/tools/instruments/asa24/.