Skip to main content

Currently Skimming:

5 Innovations and Special Considerations in Assessing Dietary Intake During Pregnancy and in Children 2 to 11 Years of Age
Pages 71-94

The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.


From page 71...
... • Studies addressing dietary intakes of pregnant women appear to underuse digital technology, but such methods could com plement those that are more widely used (i.e., self-administered and web-based questionnaires)
From page 72...
... It also included a question-and-answer session for the workshop sponsors and planning committee members to ask questions of the ­speakers, as well as a moderated panel discussion that included nine panelists, most of whom were speakers at one of the four workshops in the workshop series. Presentation topics included key considerations when developing and implementing dietary assessment methods, emerging tech­nologies as opportunities to measure dietary intake, potential for eco logical m ­ omentary assessment in dietary intake research, harmonization of methods for dietary assessment, and matching a research question to the appropriate dietary assessment method and statistical analyses.
From page 73...
... Pérez-Rodrigo started by briefly restating key points from the workshop series' prior workshops, highlighting that that the ideal level of accuracy to be attained in dietary intake research depends on the research purpose, context, target population, and resources. For surveillance purposes, for example, she pointed out that although the target population might be quite diverse, researchers must collect the same amount of dietary intake data from its various subgroups to ensure that the data are of uniform quality.
From page 74...
... Sixth, she noted other considerations that can influence dietary intake reporting, such as the child and parent's weight status, body image concerns, and presence of restrictive eating behaviors. Lastly, she emphasized motivation as an overarching consideration, explain ing that children's attention to and active engagement in the reporting activity is critical.
From page 75...
... . EMERGING TECHNOLOGIES AS OPPORTUNITIES TO MEASURE DIETARY INTAKE Carol Boushey, associate research professor in the epidemiology program at the University of Hawaii Cancer Center, shared examples of emerging technologies to measure dietary intake and shared results of studies that used such approaches with pregnant women and children.
From page 76...
... After these data were sent to the study's website for analysis, she said, a structured follow-up phone call was conducted the following day to clarify the contents of the Nutricam dietary record and to probe for forgotten foods. Researchers compared intakes derived from the image-based assessment to intakes calculated from three 24-hour dietary recalls taken once weekly on random days following the image-based record, Boushey explained, and observed significant correlations between the two methods for energy, macronutrients, micronutrients, and fiber (Ashman et al., 2016; Rollo et al., 2015)
From page 77...
... According to Boushey, studies addressing dietary intakes of pregnant women appear to underuse digital technology. It seems that self-­ administered web-based questionnaires are more widely used, she observed, and suggested that this method would complement an image-based method (either passive or active)
From page 78...
... POTENTIAL FOR ECOLOGICAL MOMENTARY ASSESSMENT IN DIETARY INTAKE RESEARCH Katie Loth, assistant professor in the Department of Family Medicine and Community Health at the University of Minnesota, explored the potential for ecological momentary assessment (EMA) to be used in dietary intake research.
From page 79...
... It is unknown if EMA reduces recall bias or is affected by social desirability bias. The research question is of primary importance, she maintained, and
From page 80...
... Dietary measures that might be considered for harmonization include intake of energy, micronutrients, or specific foods or food groups, she said. The need for harmonization is propelled by the recognition that significant stakeholder investments in research warrant or even obligate data sharing and pooling across studies, she said, with the goal of expediting the speed of health-promoting scientific discoveries.
From page 81...
... Although ECHO deals primarily with extant data, she noted, there could be an opportunity to have cohorts collect both FFQ and ASA24 data from a subsample of participants. That would enable the use of regression calibration methods, she said, to correct for some of the measurement error.
From page 82...
... She centered her presentation around two primary themes: first, the concept of multidimensionality, which she described as the presence of multiple foods, nutrients, and other constituents represented in a diet, each of which could be considered its own dimension; and second, the integration of measurement error into study design and into the analysis of observational epidemiologic studies. Liese addressed multidimensionality by providing three examples of research that each jointly considered food groups and nutrients in epidemiologic analyses.
From page 83...
... She reminded attendees that dietary intakes estimated from self-report instruments often have sub stantial error owing to factors such as day-to-day dietary variation, reliance on memory, and biases such as social desirability (Freedman et al., 2011)
From page 84...
... 84 FIGURE 5-2  Shape analysis-derived constellations representing high-quality dietary patterns illustrated using radar graphs. NOTE: Add sug = added sugars; fat acid = fatty acids; grns & bns = greens and beans; ref grain = refined grains; sat fat = saturated fats; seaf & pl prot = seafood and plant proteins; t fruit = total fruits; t prot f = total protein foods; t veg = total vegetables; wh fruit = whole fruits; wh grain = whole grains.
From page 85...
... The topics covered were motivating participants to complete dietary assessments, the burdens and benefits of EMA, multidimensionality in dietary research, open-source data sharing, and the challenges associated with dietary supplement intake data. Motivating Participants to Complete Dietary Assessments Anna Maria Siega-Riz, dean and professor in the School of Public Health and Health Sciences at the University of Massachusetts Amherst, asked the speakers how to motivate participants to complete dietary assessments and wondered about gamification as a strategy.
From page 86...
... If onsite help is unavailable, Loth said that researchers can purchase apps or pay a fee to have an app's developers help with the EMA setup. Returning to Dabelea's question about how EMA can complement other dietary intake methods such as 24-hour recalls and perhaps reduce the bias associated with that method, Loth responded that EMA can provide contextual information for the eating occasions reported in a 24-hour recall.
From page 87...
... It has been challenging to collect all of the ECHO cohorts' data on one platform, she admitted, and shared her belief that it would be beneficial to stakeholders to coalesce around the best way to collect intake data, particularly for supplement use. With regard to data harmonization, she referenced an effort within the ECHO study to provide guidance for how to crosswalk source data on various dietary measures that flow into diet quality indices such as the Healthy Eating Index.
From page 88...
... Nonetheless, Liese thought this could potentially be accomplished by conceptualizing the dietary supplement as another layer of information to include in the analysis. FINAL WORKSHOP PANEL DISCUSSION: CHALLENGES AND OPPORTUNITIES TO STRENGTHEN DIETARY INTAKE DATA FROM PREGNANT WOMEN AND CHILDREN The final session in the fourth workshop concluded the workshop series with a panel discussion on the challenges and opportunities to strengthen dietary intake data from pregnant women and children.
From page 89...
... It does not have to be a lot of data, she clarified, but repeated measurements across different studies would allow statisticians to fit more specific models. Dabelea asked if groups of studies such as ECHO are available as a source of dietary intake data for researchers to use in developing the kinds of models that Catellier and Sotres-Alvarez suggested.
From page 90...
... The difference in the before-and-after food images may not necessarily translate to the exact amount the child consumed, she inferred, giving the hypothetical example of a child taking a before image and then throwing away some of the food or giving it to a friend before taking the after image. Loth submitted that a prevailing belief has been that children cannot serve as informants of their own dietary intakes, particularly for eating occasions that do not occur in their parents' presence (and thus cannot be supplemented with data from parents' proxy reports)
From page 91...
... On that note, he mentioned that improvements are expected in artificial intelligence algorithms that classify passively collected images, meaning that only those most relevant to the study would be retained. From the perspective of big data, De Vito said that because missing data is a concern when assessing dietary intakes, she proposed more routine collection of intake data using multiple methods and data integration.
From page 92...
... HIGHLIGHTS4 Cheryl Anderson, professor and dean of the Herbert Wertheim School of Public Health and Human Longevity Science at the University of C ­ alifornia, San Diego, and member of the workshop series planning committee, summarized a selection of key points from the workshop series' four workshops before she adjourned the series. She grouped these points into three categories: unique features of different life stages of interest, considerations for choosing assessment methods and designing tools, and considerations for analyzing and sharing data.
From page 93...
... • The constantly evolving food, beverage, and dietary supplement market landscape necessitates flexibility to improve food databases used in dietary assessment research and to provide updated tools to improve reporting accuracy. • Collecting data on children's dietary intakes from multiple infor mants or proxy informants is helpful but comes with its own set of challenges related to measurement error and missing data.
From page 94...
... Such data repositories would also ­ideally include metadata for the dietary intake measures and could also be used to fit sophisticated analytic models.


This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More information on Chapter Skim is available.