Personalised nutrition approaches typically rely on static information such as health status, lifestyle and genetic data. However, adherence often remains low because daily behaviours are influenced by psychological states, motivation and real-time barriers. There is growing interest in whether integrating behavioural signals and timely prompts could strengthen engagement. The purpose of this study is to explore whether incorporating psychological adaptation and just-in-time behavioural prompts leads to better adherence to a personalised nutrition programme compared with commonly used personalised approach (combining health, lifestyle and genetics data without psycho-behavioural adaptation). Using a six-week n-of-1 randomised trial design implemented across multiple participants, the research will also assess the feasibility of delivering this adaptive support in a free-living environment, as well as participants' engagement, acceptability and perceived usefulness. These findings will inform the development of an adaptive personalised nutrition AI agent.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
PREVENTION
Masking
NONE
Enrollment
20
Participants receive the same nutritional and lifestyle content, but daily goals are adapted based on: Stable Traits * Big Five personality * Health locus of control * Optimism bias * Delay discounting * Self-determination * Stress/anxiety baseline * Habit awareness * Values/motivations * Reward preferences Daily State * Morning energy rating (1-5) * Expected barriers * Context (home vs office vs weekend) Communication Preferences * tone (coachy, compassionate, factual, brief) * message length (low/medium/high) * mode (text, voice, image) Environmental and Social Context (e.g., Friday office snacks, partner-driven late dinners) Daily personalisation influences: * difficulty level of the daily goal * specific behaviour change technique (BCT) chosen (e.g., action planning vs implementation intention vs coping planning) * tone, structure and timing of the message * reinforcement style The nutrition content remains identical; only HOW it is delivered changes.
Participants receive: * One fixed daily behaviour goal, aligned with the weekly pillar * Personalised only according to diet, lifestyle, health status and genetics * Delivered at a pre-selected time chosen by the participant (e.g., 8:30-10:00) No adaptation is made based on: * psychological traits * daily mood/energy * anticipated or reported barriers * communication preferences (tone, length, mode) This reflects current "gold standard" personalised nutrition: personalised by biology and lifestyle, but not by psychology or dynamic states.
St Mary's University
Twickenham, Middlesex, United Kingdom
Adherence
Adherence is defined as daily completion or active attempt of the assigned lifestyle goal. Adherence is measured using a brief daily self-report item delivered via WhatsApp as part of an ecological momentary assessment. Each day, participants are asked whether they completed or actively attempted the assigned goal (Yes/No). Daily adherence is coded as a binary outcome (1 = completed, 0 = not completed). Adherence rates are calculated as the proportion of completed goal days relative to total analysable person-days and compared between personalised intervention days and static control days. An analysable person-day is defined as a day on which a goal message was successfully delivered and a corresponding adherence response was received within the response window. Days with missing responses or delivery failures are excluded from adherence analyses.
Time frame: From enrolment to the end of the 6 week intervention
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.