Category: Methods/Tools

Knowing Nutrition

NCC director Lisa Harnack meets with her team.
NCC director Lisa Harnack meets with the center’s team.

Nutrition studies raise the quality of our diets and increase our understanding of how our bodies use and respond to food.

The modern age of nutrition research began in the early 1900s with the discovery of essential nutrients, like vitamins, and their importance in preventing crippling diseases. In recent decades, nutrition studies have continued to be fundamentally important in protecting health by showing how we can avoid cancers, heart attacks, and other dangerous conditions by eating well.

Since 1974, the School of Public Health’s Nutrition Coordinating Center (NCC) has been making much of this research possible. The NCC’s flagship product is its Nutrition Data System for Research (NDSR), a software program researchers use to analyze the composition of foods found in recipes and menus, or eaten by study participants. It’s a popular tool used by more than 100 institutions ranging from Johns Hopkins University to NASA.

Stanford University Professor of Medicine Christopher Gardner calls the NDSR “the ultimate tool for nutrition researchers.”

“So many researchers use NDSR because it’s linked to a comprehensive food database — and the program is really easy to use too,” says NCC director and SPH Professor Lisa Harnack.

NCC's Mercedes Taneja conducts a phone
NCC’s Mercedes Taneja conducts a dietary recall phone interview using the NDSR.

Promptly Remembered

In a lot of nutrition studies, a study participant describes to a researcher what they ate through a process known as “food recall.” Researchers then use this data to study everything from calorie intake to nutrient consumption.

But food recall research can be tricky — often people forget what and how much they ate. So NDSR intelligently guides interviewers while they ask dietary recall questions, prompting them to gather more information based on a study participant’s answers. For example, a participant may be asked what they ate for breakfast. If they respond that they had toast and peanut butter, the program will prompt the interviewer to further investigate the exact type of bread and peanut butter (regular or reduced-fat? 1 tablespoon or two?).

“People think that they can describe what they eat accurately, but the prompts factor in the forgetfulness of human nature,” says Gardner. “It can even anticipate the condiments people would typically eat with certain foods.”

Gardner used the NDSR for a recent NIH weight loss study of 609 participants and says it helps his research be more accurate and credible.

“It raises the validity of my research,” says Gardner. “Any reviewer familiar with the challenges of diet assessment knows I’m choosing the gold standard if I use NDSR.”

Based on Accurate Data

Any software is only as good as the data that drives it and the strength of the NDSR comes from its Food & Nutrient Database. This database feeds the NDSR software with information for approximately 18,000 foods. The database is also often licensed to software developers and researchers for variety of purposes, such as supporting diet and nutrition smartphone apps.

The strength of the database is found in how completely it accounts for the nutrients contained in foods. Each food in the system has up to 165 nutrients listed, compare that to the 10 nutrients found on an average food label.

The database is so comprehensive and accurate because it’s kept current through weekly updates by a team of scientists who carefully gather and scrutinize nutrition data.

“The food market is always changing. We check product manufacturer  websites or actually go to the supermarket and pull the info right from the box or label,” says Harnack. To gather complete nutrition information, Harnack and her team even pull detailed information from ingredient lists. “That’s where our database is different,” says Harnack. “Other databases just take information from the nutrition label and the rest of the nutrients are missing. We take all of the ingredients listed and figure out all 165 nutrients in the product.”

Good Service

Sometimes, researchers want not only the benefits of using the NDSR and its database, but also the NCC’s expertise in conducting nutrition studies. In those instances, the center offers a service to conduct the food recall research on behalf of institutions.

Eric Rimm, a professor of epidemiology and nutrition from Harvard, recently used this service to validate a 25-year-old food frequency questionnaire produced by his school. Rimm was interested in comparing the quality of the questionnaire’s data to that gathered by the NCC using NDSR.

“We knew the NCC would give very good quality data and to do the study [at Harvard] would’ve taken a lot longer and have been much more expensive,” says Rimm. “Minnesota’s Nutrition Coordinating Center was the first place we went to because of its reputation — it was our number one choice.”

Kuntz Leads Addition of Computer Modeling for Cost-Effectiveness Analysis Guidelines

Professor Karen Kuntz
Professor Karen Kuntz

A paper published in JAMA offers new guidelines in health care cost-effectiveness analysis (CEA), replacing recommendations published in 1996. Professor Karen Kuntz was a member of the panel charged with drafting the new guidelines and led the writing of an additional section on developing computer models for conducting CEAs.

In the two decades since the first guidelines were written, computer models have become a common tool used in completing a CEA.

“Enough time has passed, and there have been a lot of changes in the cost-effectiveness analysis field — like the increase in the use of decision-analytic models — so we knew an update was needed,” says Kuntz.

A CEA is a decision-making tool in which the costs and effects of tests, therapies, and prevention techniques are calculated. The results of CEAs are used by governments in making health care coverage decisions. Guidelines help ensure that the results from any one analysis can be compared to another.

The computer models used in CEAs, such as decision-analytic models, can take evidence from multiple sources and use it to extend results to different populations or to make projections beyond the time horizon of collected data.

Kuntz sought to provide recommendations for modeling that produce relevant, reliable, and useful projections.

“We emphasized transparency and the importance of users being clear in describing the assumptions they make in creating a model’s structure,” says Kuntz. “We recommended incorporating all the available data, and being clear about why you did or did not include certain sources. We also discussed the importance of adopting a lifetime horizon to capture all of the costs and effectiveness that may be relevant to decision makers.”

Kuntz said the new CEA guidelines will hopefully be used as the method of analysis by researchers and advance best-practice discussions within the field.

Wolfson Named Reproducibility Editor for Leading Statistics Journal

WolfsonSchool of Public Health Assistant Professor Julian Wolfson was named an associate editor for reproducibility for the Journal of the American Statistical Association (JASA). The appointment is in support of the journal’s new requirement for authors to submit scientific code and data for review along with their papers.

The journal said it’s adding the editors and requirement to ensure the reproducibility of scientific results reported in its studies. The move follows similar action recently taken by many medical journals.   

“Reproducibility is a hot topic and there’s been a lot of back-and-forth about it among clinical trials and open-science researchers,” says Wolfson, who has extensive experience in statistical programming.

JASA says the code and data from published studies will be offered mainly through its website, and the aim of the review work by Wolfson and the editorial team is to assess the information’s availability, quality, and potential usability for people wanting to reproduce research.

Stemming Global Climate Change

Scientists currently have no accurate way to estimate the current total coverage of plants across the planet or project what coverage will be like in the future. PhD biostatistics student Abhirup Datta has changed that.

Continue reading

SPH Researchers Create Tool to Provide More Accurate Representation of Family Satisfaction with Nursing Home Care

Making the decision to move a loved one into a nursing home is a big decision, which often leaves family members with questions and concerns regarding their satisfaction with their loved one’s care and quality of life (QOL) while in the facility.

To measure family satisfaction in nursing homes, most states rely upon measures developed by the nursing home industry that have not undergone rigorous testing. There is now, however, a new tool developed by researchers at the University of Minnesota’s School of Public Health, in collaboration with Minnesota Department of Human Services, that provides validated measures of family satisfaction. These measures have been used in all nursing homes in Minnesota and show strong performance among consumers.

The tool consists of 32 questions. Family members are asked to reflect on their experiences with the nursing facility and the care given there. They grade each item on a scale from A-F, where A=excellent; B=very good; C=average; D=below average; and F = failing.

In a new study published in Research on Aging, researchers explored the attributes of this tool including what domains it measures as well as whether family satisfaction is associated with measures of resident QOL.

Tetyana Shippee
Lead author Tetyana Shippee

“We have been asked why family satisfaction is an important measure. The literature has many criticisms of ‘satisfaction’ as a quality indicator—generally speaking, that it’s too general and suffers from social acceptability bias,” said Tetyana Shippee, Ph.D., lead author of the study and assistant professor in the School of Public Health. “In the case of nursing homes, however, family members are key stakeholders and their voice matters both for consumer choices and for policy efforts in improving the quality of nursing home care.”

The study asks the question about how closely family satisfaction is aligned with resident QOL scores on the facility level. Shippee found a weak association between family satisfaction reports and resident QOL reports, confirming that residents and families are distinct audiences and their voices need to be captured separately (in other words, family satisfaction surveys should be used in addition to, not in place of, resident surveys).

Furthermore, Shippee and her colleagues studied facility characteristics associated with family satisfaction. They found:

  • Chain affiliation, higher resident acuity, more deficiencies, and large size were all associated with worse family satisfaction across different areas of satisfaction. Of these, higher resident acuity (residents with more health limitations in the facility) deserves particular attention because of its larger effect size.
  • Findings about resident acuity in the facility are also most consistent in predicting lower family satisfaction. These findings may represent stigma attached to higher health needs, which may create an environment less favored by families and family members refusing to accept the fact that their loved ones also need higher care.
  • Facilities with higher quality of care scores and non-profit ownership were associated with better family member satisfaction in all four domains.

“As long-term care becomes more consumer-focused, understanding patterns of consumer satisfaction and the factors that are correlated with it will take on increasing importance. Family members are important consumers and stakeholders,” said Shippee. “For persons with significant cognitive deficiencies, they are the spokespersons. For other residents, they are an important group, whose opinions count in daily decisions and facility reputation overall. In a growing climate of quality improvement, consumer feedback will be a critical source of information.”  –This post originally appeared on the Academic Health Center’s HealthTalk blog

© 2015 Regents of the University of Minnesota. All rights reserved. The University of Minnesota is an equal opportunity educator and employer. Privacy Statement