Effects of a Novel Supplement on Sleep, Metabolism, Body Composition and Performance

A randomized, double-blind, crossover design will be used to investigate the efficacy of a novel developed melatonin and magnesium supplement in pod (coffee machine capsule) format on sleep patterns and sleep quality, mood, sleepiness, biological rhythms, stress, metabolism, body composition and strength performance. Prior to the intervention, participants will be screened for inclusion and exclusion criteria, namely presenting with sleep disturbs according to Pittsburgh Sleep Quality Index (score >5), not working in shifts, not experiencing jet lag frequently (e.g., flight personnel), nor taking medication which could interfere with sleep. A total of 35 individuals suffering from sleep disturbs will engage in 4 weeks of supplementation with melatonin and magnesium or placebo, separated by 1 week of washout, after which the second experimental condition will be conducted (4 weeks placebo or supplementation with melatonin and magnesium). At baseline and 4 weeks post each, randomly assigned intervention, participants will experience measurements of resting metabolic rate, body composition and strength, as well as the collection of saliva to measure melatonin and cortisol. Validated questionnaires on sleep quality, mood, sleepiness, and biological rhythms will also be applied and nutritional intake will be quantified and assessed for frequency and timing on the three occasions. Throughout the 8 weeks of supplementation, participants will use an actigraphy device continuously to evaluate sleep and awaken motor activity, and compliance to the assigned intervention will be confirmed in each visit to the laboratories.

In order to achieve the main goal of the study, assessments will take place in the morning (starting at 7:00 a.m.), after an overnight fast. Wearing minimal clothing and no shoes, participants will be weighed on a scale and have their height measured on a stadiometer (Seca, Hamburg, Germany). Body composition, namely bone mineral content, fat mass and fat free mass, will be assessed by dual energy X-ray absorptiometry (DXA) (Horizon Wi, Hologic, Waltham, USA). Total body water and its intra and extracellular compartments will be estimated from whole body resistance (R) and reactance (Xc), measured through bioelectrical impedance analysis (BIA), using a single frequency device of 50 kHz (BIA-101, RJL/Akern Systems, Firenze, Italy). From the raw R and Xc data, the phase angle will additionally be determined, using the Akern Software.

Resting metabolic rate (RMR) will be measured by indirect calorimetry, using a canopy ventilated open-circuit system (Quark, Cosmed, Italy). Participants will arrive to the laboratory in an overnight-fasted state, having refrained from exercise, caffeine, and alcohol consumption in the 12h before assessments. The equipment will be calibrated daily according to standard procedures. Participants will rest in supine position prior to the start of the measurement. Data will be collected for ≥15min with the first 5min being dismissed, according to protocol. The average resting metabolic rate will be considered the period of 5min of the test with lower coefficient of variance.

The handgrip strength test will be used to evaluate the maximal isometric force of the muscles of the hand and forearm. Using a portable hand dynamometer (JAMAR, Sammons Preston, Bolingbrook, IL, USA), participants will be tested for both hands alternately, in a stand-up position. The maximal force generated out of 3 attempts will be considered for analysis.

For biochemical markers, saliva will be collected in salivettes. Saliva will be used to measure melatonin and cortisol using Enzyme-Linked Immunosorbent Assay (ELISA).

In an interview with a Registered Dietitian on the three visits to the labs, participants will report food and fluid intake in the 24h before. Timing of the meals, cooking/preparation methods, product specificities and estimated portion sizes will be recorded. To support this methodology, a validated food frequency questionnaire will be applied to assess for type and frequency of food and drink consumption, during the month-long interventions. Together, these data will allow to identify the pattern of ingestion (quantity, quality and timing) of key foods (i.e. kiwifruit, tart cherry, milk, valerian tea) and nutrients (i.e. total energy intake (TEI), total carbohydrate (CHO) in grams and as a %TEI, high glycaemic index CHO, low glycaemic index CHO, tryptophan, caffeine, alcohol) with potential impact on sleep parameters. Dietary intake will be quantified using a dietary analysis software (Nutritics Research Edition (v5.09), Dublin, Ireland). Participants will be instructed to maintain their eating habits throughout the 9 weeks of the investigation.

At baseline and after both interventions, subjects will be asked to fill in the Portuguese validated questionnaires: Pittsburgh Sleep Quality Index (PSQI), Profile of Mood States (POMS), Epworth Sleepiness Scale (ESS) and Horne \& Oestberg Morningness-Eveningness Questionnaire (MEQ), to evaluate subjective sleep quality, mood, sleepiness, and biological rhythms, respectively.

In order to objectively evaluate sleep parameters (i.e., sleep latency, total sleep time, sleep efficiency) during the night and motor activity during the day, as well as to quantify total energy expenditure and activity energy expenditure throughout the 8 weeks of supplementation, participants will be asked to wear an activity monitor device with a validated 3-axis accelerometer (ActiGraph GT9X Link, Florida, USA) continuously, during the day and at night.

Randomization and Supplementation: The order of the interventions (treatment or placebo) will be randomized by an external investigator. All eligible participants will be assigned to the same group initially, but will experience both conditions in a crossover manner. The investigator responsible for the sample randomization and the distribution of the supplements is not directly involved in participants' eligibility interview nor in data collection.

Sample justification and sample size: According to a recent systematic review and meta-analysis of randomized controlled trials (PMID: 33417003), melatonin levels decrease with age, while the incidence of sleep disturbances increases proportionally. However, a decline in melatonin levels is not typically observed before the age of 35, while after the age of 55, other comorbidities and/or the intake of medication may instead explain alterations to the sleep cycle. Notwithstanding, a cohort study looking at a large spectrum of ages (PMID: 21091391) has demonstrated the short- and long-term efficacy of 2mg of prolonged-release melatonin in insomnia patients between 18-80 years old, particularly those aged 55 and over. Considering the similar dosage to that of this novel formula (1.9mg of melatonin), recruiting a sample of participants between the ages of 35-55 is one of the inclusion criteria. People who work in shifts or who may experience jet lag due to their profession (e.g., flight personnel) are excluded on the basis that irregular sleep patterns may interfere with sleep quality. Considering the literature presented, the investigators have decided to establish a similar sample size and study duration (4 weeks + 4 weeks) to that of previous studies. Sample size was calculated using g power version 3.1., based on an effect size of 0.25, power of 0.8 and alpha 0.05. A sample of 28 subjects was deemed required. Assuming a dropout of 25%, the researchers have pursued a sample of 35 participants.

Statistics: All statistical analyses will be carried out using IBM SPSS Statistics (NY, IBM). Basic descriptive data will be used to characterize the participants in the study and all variables will be checked for normality, using Shapiro-Wilk test. All participants will initially be assigned to the same intervention group. Time and time-by-group interactions will be evaluated by Mixed ANOVA. The equality of the matrix of variance and sphericity will be explored with the Levene F test and Mauchly's test, respectively. Overall significance level for α will be set at p ≤ 0.05.