Whole nuts, butter and measured energy
In this investigation
Almonds and the balance method
Metabolizable energy is estimated from food energy minus energy lost in stool and urine. It is different from burning a food sample in a calorimeter, measuring daily expenditure or observing weight change.
In a controlled crossover study, 18 adults aged 32–67 completed three 18-day periods. Twelve received each of 0, 42 and 84 g/day of whole almonds; six repeated a condition. Foods were scaled using calculated calories, and the final nine days included complete stool and urine collection. The measured average was 129 kcal per 28 g, compared with 168–170 kcal calculated using Atwater factors. Individual measurements varied substantially. This was a balance experiment, not a weight-loss trial. USDA and the Almond Board of California supported it. Novotny et al., 2012.
A separate walnut experiment
All 18 adults, averaging 53 years, completed two three-week controlled-diet periods, with or without 42 g/day of walnut halves and pieces. Researchers reduced other foods proportionately to match calculated calories; this did not establish equal absorbed energy. After adaptation, participants collected stool and urine for about a week.
Measured walnut energy averaged 146 kcal per 28 g, compared with 185 kcal calculated from Atwater factors. Metabolizable energy was the registered primary outcome. USDA was the registry sponsor and the California Walnut Board a collaborator. Later microbiome reports used this same cohort, rather than independently repeating the energy experiment. Baer et al., 2016, trial registration.
Whole, chopped and butter
See the existing whole-versus-butter measurements. Chopped and whole roasted almonds did not differ significantly. Pure almond butter contained no added ingredients. USDA and the Almond Board supported the trial. Gebauer et al., 2016.
Chewing is another source of variation
A short crossover experiment started with 20 people and analyzed 13, excluding seven for incomplete attendance, collections or data, or personal withdrawal. Participants ate 55 g of almonds under assigned chewing conditions; all food was provided during subsequent four-day collection periods. More fecal energy was lost with 10 or 25 chews than with 40. However, appetite responses were not a simple progression: the 25-chew condition produced less fullness than both extremes at two hours.
This supports a role for oral processing, not an optimal chewing prescription or a prediction of long-term weight change. The Almond Board funded the study and provided almonds. Registration occurred after the recorded study completion, so it does not establish prospective outcome specification. Cassady et al., 2009, registration dates.
What changes in my food log
I enjoyed the whole almonds and walnuts. The recent quantities were rough counts and an approximate equal amount, rather than weighed portions.
These experiments cannot supply my personal absorbed-energy correction. The existing estimated total remains unchanged.
Nut butter remains an optional food choice if I enjoy it and can use the package. Its ingredients, serving amount and label would need their own entry; I should not carry a whole-nut entry over unchanged. The next useful improvement is clearer portion recording, alongside energy calibration over time. See nut choices for the separate question of food preferences and longer-term outcomes.