50 nutrition and dietetics dissertation questions across diet quality, obesity, maternal and child health, food insecurity, chronic disease, sports nutrition and global nutrition.
By Siddharth Gupta · Statssy
Nutrition research can use large public surveys and product databases without recruiting new participants. But the strength of a dissertation depends on matching the dietary measure, population and statistical method to the question.
The 50 topics below offer suggested designs, datasets and related university theses. Some topics need more than one source or approved research access, so check the available variables and measurement periods before committing to a proposal.
Start with what people actually eat. Diet survey data is free and ready for regression.
Data check: WHO GHO typically offers aggregate country indicators, not adolescent-level records for a logistic model. Check a suitable individual survey such as GSHS.
Obesity topics are the most popular in nutrition. These ones use national data so you can test trends and risk factors.
Data check: This is not a true retrospective cohort when adolescent BMI is recalled in a cross-sectional survey; consider measurement and recall bias.
These topics test what helps people lose weight and keep it off. Most use large health record or survey data.
Data check: All of Us requires authorised researcher access and suitable medication timing, outcome measurements and confounding controls.
Data check: A specific postoperative cohort, social support measures and repeated weights must be confirmed; All of Us may not contain all required variables.
Data check: The USDA Atlas does not provide GLP-1 Medicaid policy timing or sufficiently granular obesity outcomes for this design. Additional policy and outcome data are necessary.
Data check: NHANES can inform baseline distributions, but a cost-effectiveness model also needs intervention effects, transition probabilities, programme costs and utilities.
Data check: Confirm longitudinal linkage and wearable coverage for the selected All of Us cohort.
The first 1,000 days shape lifelong health. These topics cover pregnancy weight, birth weight and breastfeeding.
Data check: Check whether NHANES contains the gestational weight-gain and WIC measures required in the chosen survey cycles.
Data check: DHS often records BMI at survey rather than pre-pregnancy. Verify timing before interpreting this as prepregnancy exposure.
Data check: DHS may not measure maternal blood glucose directly; confirm suitable diabetes indicators or use an alternative clinical dataset.
Undernutrition is still the biggest nutrition problem in India and much of Africa. These topics use free survey and global estimates.
Data check: DHS does not routinely include confirmed parasitic infection for all children. Mediation additionally requires defensible temporal and causal assumptions.
Data check: A DiD programme evaluation needs pre/post outcome measures, variation in programme exposure and a plausible comparison group; NFHS waves may be sparse.
Data check: NHANES is a household survey, not a hospital-patient cohort. Reframe for general-population biomarker outcomes or obtain clinical data.
Food insecurity topics link nutrition with economics and policy. US data on SNAP and food security is rich and free.
Data check: The listed datasets do not directly identify enrolment in a particular fruit-and-vegetable incentive programme or treatment timing.
Data check: FoodData Central contains food nutrient profiles, not participant-level pre/post changes or food-security outcomes; collect programme records for a paired study.
Policy topics test if a rule or campaign changed what people buy and what firms sell.
Data check: Open Food Facts can include changing product labels, but historic coverage and policy exposure must be sufficient to construct a panel.
Data check: FAOSTAT country food supply series do not identify individual household purchases needed for a household demand system.
Data check: The USDA Atlas does not contain store-level campaign sales before and after intervention; transaction data are needed.
Data check: The Atlas alone cannot identify specific programme start dates and treatment exposure for a credible DiD analysis.
Nutritional epidemiology links diet with disease. These topics use biomarkers, genetics and long-term follow-up.
Data check: Cross-sectional NHANES can estimate prevalence or associations, not incident cancer or heart-disease risk without outcome follow-up.
These topics suit dietetics and sport science students. They cover body composition, eating behaviour and genes.
Data check: NHANES is not a dedicated athlete cohort and may lack validated relative-energy-deficiency indicators.
Data check: Bland–Altman evaluates continuous measurement agreement; BMI-based obesity class versus body fat classification needs classification agreement analyses.
Data check: Cross-sectional activity and intake records cannot establish how much an individual ate after exercise.
Data check: All of Us genetics and outcome data may be restricted; confirm intervention assignment and longitudinal weight change before modelling interaction.
Data check: NHANES does not necessarily contain the Children’s Eating Behaviour Questionnaire. Use a dataset with the full validated item-level scale.
These topics compare countries or focus on India. They suit students in India, the UAE and global health courses.
Data check: FAOSTAT food supply does not directly measure access or food insecurity within households; supplementary indicators are required.
Data check: NFHS survey rounds and school-meal rollout timing must support a credible evaluation, not simply a cross-sectional correlation.
These sources offer public or approved-access nutrition data. Many contain different units of analysis—individuals, food products, counties and countries—so confirm that the source actually measures the outcome required by your topic.
US nationally representative nutrition, dietary recall, biomarker and health survey data; account for complex sampling.
UK food diaries, nutrient intake and related nutrition surveys; registration may be required.
Food composition, nutrient profiles and product data—not individual eating histories.
Open packaged-food ingredients and nutrition labels; data completeness varies by product and country.
US county-level food access, retailers, food environment and selected health indicators.
Household and child nutrition survey microdata; free approval and survey design checks may be required.
Child stunting, wasting and overweight estimates for cross-country trend analysis.
Country-level health and nutrition indicators, rather than individual diet recalls.
National agricultural production, food supply and selected price series—not individual purchases.
Harmonised food and nutrient intake estimates across countries, sex and age groups; data-access terms apply.
US health records, participant surveys, wearables and genomic information. Controlled research access applies.
Indian state and public-programme indicators; variable coverage and programme reporting differ.
A printable guide containing all 50 research questions, study designs, recommended statistical methods, dataset URLs and related university thesis links will be added after it is created and uploaded.
PDF download coming soonFor related dissertation ideas, see public health research topics, sports science research topics and nursing research topics.
Bring your proposed research question and data source to a free 15-minute call. We can review dietary measures, sample design and a suitable method.
Book a free 15-minute call15 minutes · No cost · No pressure
A good topic asks one clear question about diet or nutrition, uses data you can download and fits a method you can run. Diet quality, obesity and food insecurity are the easiest to start.
NHANES, FoodData Central, Open Food Facts, DHS, UNICEF and FAOSTAT are free. The UK National Diet and Nutrition Survey needs a free login. All links are on this page.
Survey-weighted regression, logistic regression, dietary pattern analysis, mixed-effects models and difference-in-differences are the most common.
Yes. Every topic on this list can use public survey or product data. You only need to clean and analyse it.
R and Stata handle survey weights well. SPSS works for simple regression. Python suits large product and label datasets.
Yes. Download the full Topic and Method list from this page. It includes every topic, design and dataset link.