Improving Sleep and Daytime Function with Tryptophan, Magnesium, Melissa and Lactuca Formulation: An Exploratory Study in Adults with Sleep Disturbances

Introduction

Insufficient sleep has emerged as a significant public health issue in modern society, often under-recognized and under-reported.¹ Sleep disorders significantly impair physical and mental health, increasing the risk of cardiovascular disease, cognitive impairment, and metabolic dysfunction. Additionally, poor sleep correlates with a heightened risk of accidents, decreased quality of life, and greater health-care utilization.^2–4

Alongside sleep disturbances, stress and mood disorders are escalating due to the pressures of contemporary lifestyles. The interconnection between sleep and stress is mediated by the hypothalamic-pituitary-adrenal (HPA) axis, where sleep deprivation enhances HPA axis activity, leading to neuroendocrine dysregulation.⁵

Data from the German Health Interview and Examination Survey for Adults (DEGS1) reveal that approximately 25% of 8152 subjects reported poor sleep quality, with 5.7% meeting the criteria for insomnia syndrome.⁶ These prevalence rates vary across gender and age groups⁶ and are on the rise.⁷ Insomnia is a sleep disorder characterized by persistent difficulty initiating or maintaining sleep, which often leads to significant distress and preoccupation with sleep-related issues. These challenges can substantially impair daytime functioning, manifesting as excessive sleepiness, reduced cognitive performance, and diminished quality of life. According to the ICD-10/11 (International Classification of Diseases) diagnostic criteria (F51.0/7A00), insomnia is diagnosed when symptoms occur at least three times per week for a duration of one month or longer.⁸

With the growing prevalence of sleep disturbances and stress, many individuals are exploring special diets and nutritional supplements as an alternative to conventional medication, which is often associated with adverse side effects.⁹ As sleep-related disorders continue to rise and impose significant personal and societal costs, easily available over-the-counter products have become increasingly popular. Nutritional supplements present a promising, well-tolerated, and safe option for improving sleep quality and supporting overall well-being.

This study evaluated the potential of Sleep Well Granulate™, a novel formulation combining herbal extracts of fresh Lactuca sativa and Melissa officinalis with L-Tryptophan and Magnesium, over a 14-day intervention. Designed as a direct granulate for convenience, the study primarily assessed its impact on sleep quality with secondary evaluations of sleep quantity, its effects on psychological well-being and cognition in healthy adults reporting sleep complaints.

Materials and Methods

Study Design

This exploratory, open-label, one-armed, prospective, interventional trial was carried out in accordance with the study protocol at one study centre (daacro in Trier, Germany) from March to July 2023 strictly adhering to ICH-GCP principles, the declaration of Helsinki (2013), and local regulations. The trial was prospectively registered at clinicaltrials.gov with the identifier: NCT05748574. The study protocol was approved by the Institutional Review Board of the International Medical & Dental Ethics Commission (IMDEC) GmbH Freiburg, Germany (IRB Ref. No. 2023/103).

Subjects were recruited via the study center’s database and local flyers in Trier, Germany. Interested individuals underwent telephone pre-screening and were included at their first visit (V1) upon providing written informed consent and meeting eligibility criteria. The 17-day study included a 3-day run-in phase as baseline without intervention, followed by a 14-day therapy phase starting on day 4 (Figure 1). Subjects visited the study center three times: upon inclusion to record baseline measures (day 1, V1), to receive the study product and record additional baseline and acute measures (day 4, V2), and for final close-out to assess post-treatment and acute measures (day 17, V3). At V2 and V3, subjects took one sachet under the supervision of the study staff after the completion of the pre-acute intervention assessments. The sachet was administered one hour prior to conducting the post-acute intervention assessments. Acute assessments included computerized tests to evaluate cognitive performance and the State-Trait Anxiety Inventory State questionnaire (STAI-X1) to measure acute changes in state anxiety.

Figure 1 Study Scheme.

A sub-collective of 10 volunteering subjects had two additional study centre visits to be equipped with the portable PSG devices (day 3 at visit V2-1, day 16 at V3-1). Portable PSG devices were utilized for home recording sleep on the nights prior V2 (baseline) and V3 (post treatment). Additionally, saliva samples for assessment of cortisol and melatonin were collected by each subject at home at baseline on days 2 and 3 and treatment days 15 and 16.

The primary objective of this exploratory study was to evaluate the effect of the intervention on subjective sleep quality, assessed via the SF-B/R questionnaire. Secondary objectives included assessments of sleep quantity (via diary, PSG, and wearables), psychological well-being (questionnaires), endocrine markers (salivary cortisol and melatonin), cognitive performance (COMPASS battery), and safety. As no formal a priori hypotheses were statistically tested, expected improvements were based on known properties of the individual ingredients.

Subjects

Generally, healthy subjects aged 18–65 years with self-reported impaired sleep as a precursor of the S3 guideline criteria for non-organic insomnia (F 51.0), including difficulties initiating or maintaining sleep or poor sleep quality occurring regularly over the past three months. These criteria were adapted in accordance with the S3 guideline on non-restorative sleep to reflect preclinical complaints relevant to OTC interventions. Individuals with the following criteria were excluded from participation: not having habitual bedtime between 9pm-12am, BMI <18 or >30 kg/m², high systolic/diastolic blood pressure (≥159 mmHg / 99 mmHg), pregnant/lactating, any known history of organic/non-organic disorder affecting sleep (including diagnosed insomnia), serious acute/chronic disease, intake of medication influencing sleep patterns, drug misuse, excessive alcohol consumption, heavy smokers (>10 cigarettes/day), learning/behavioral difficulties, visual impairments, history of time zone travel within one month of V1, participation in another trial within the last 30 days, shift worker, not fluent in local language or having known allergies to the components of the study product. Other supplements were not permitted, and caffeine was limited to ≤10 cups/day. A complete list of applied inclusion/exclusion criteria can be found in the Supplementary Materials, Section A.

Formulation and Dosing

The study product (Sleep Well Direct Granulate^TM) contains per single dose sachet 190 mg of a standardized, aqueous Melissa officinalis leaf extract ≥6% rosmarinic acid, 75 mg of pressed juice concentrate from freshly-harvested Lactuca sativa herb (10 x concentration factor from pressed juice), 410 mg Magnesium dicitrate (60 mg Magnesium, 15% NRV), 120 mg L-Tryptophan, natural orange/vanilla flavors and excipients. GMP-compliant manufacturing was performed by A. Vogel AG (Roggwil, Switzerland). Each subject was supposed to take one sachet daily 0.5–1 hour prior going to bed. The applied dosage regimen reflects common OTC intake levels and safety standards for the contained ingredients. Subjects were instructed to take the content of one sachet directly into the mouth, followed by at least 10 seconds dissolution before swallowing for 14 consecutive days. A 14-day intervention period was selected to evaluate short-term tolerability and preliminary efficacy. Subject boxes containing sufficient sachets for 14 days of intervention were identically labelled with batch number (1097005/12012024) and subject ID according to Annex 13 (ICH-GCP). Compliance with respect to study product intake was confirmed via online diaries and by counting upon return at final visit V3.

Assessments

Sleep Recordings

At V2 and V3, subjective sleep parameters were evaluated using the SF-B/R questionnaire, a validated tool for assessing sleep quality and difficulties (Schlaffragebogen B/Sleep questionnaire B, revised).¹⁰ The SF-B/R comprises 31 questions across 12 scales, including: sleep quality (SQ, primary outcome), difficulty falling asleep, difficulty staying asleep, early waking, fragmented sleep, total sleep duration, feeling refreshed, pre-sleep well-being, pre-sleep exhaustion, psychosomatic symptoms, dream recall, and sleep-wake patterns in retrospective of the past 2 weeks. Items were rated on a 5-point Likert scale (1 = “never” to 5 = “very often”).

Daily sleep diaries recorded the following sleep parameters continuously from days 1–17: sleep onset latency (SOL), total sleep time (TST), number of awakenings (NWAK), time in bed (TIB), and sleep efficiency. The diary also assessed physical/psychological tension, sleep enjoyment, and restedness on numeric rating scales from 0 = “very low ratings” to 10 = “very high ratings”; completed electronically by subjects each morning after waking up.

Wearable devices (Fitbit Charge 5, Fitbit, Inc., USA) recorded TIB, TST, SOL, and sleep stages (deep, REM, light, awake) continuously from days 1–17, using proprietary algorithms.

Full-night polysomnography (PSG) was conducted in a subgroup of N = 10 with subjects accepting the increased effort using SOMNO HD eco (Somnomedics AG, Germany) for one night at baseline (night of day 3) and for one night post-intervention (night of day 16). PSG measured TIB, TST, SE, SOL, wake after sleep onset (WASO), sleep stages (N1, N2, N3+N4 combined, N4 exclusively, REM), apnea-hypopnea index (AHI), and oxygen desaturation (DESAT). Acknowledging AASM guidelines that combine N3+N4 into slow-wave sleep (SWS), a separate exploratory analysis of N4 was conducted to investigate potential distinctions within SWS. Data was analyzed by a blinded sleep specialist using Domino software (version 3.0.0.6, Somnomedics AG, Germany).

Psychometric Assessments

Baseline assessments (V1) included the Trier Inventory for Chronic Stress (TICS)¹¹ and STAI Trait (STAI X2 scale).¹² At V2 and V3, subjects completed questionnaires assessing daytime sleepiness (ESS scale);¹³ depression, anxiety and stress (DASS-21 scale);¹⁴ general well-being (WHOQOL-BREF scale);¹⁵ happiness (OHQ scale),¹⁶ and state anxiety (STAI X1 scale).¹² State anxiety was assessed pre- and post-interventional during acute evaluations at V2 and V3.

Cognitive Assessments

Cognitive performance was evaluated using the Computerized Mental Performance Assessment System (COMPASS, Northumbria University, UK). The 14-days intervention effects were assessed by comparing baseline (V2-pre) to post intervention (V3-pre) performance, while acute effects were calculated by comparing baseline (V2/V3-pre) to post dosage measurements (V2-/V3-post). Tasks included memory, executive function, and attention assessments (eg, Immediate Word Recall, Reaction Time, Serial Subtraction), alongside a 30-minute Cognitive Demand Battery (CDB) and visual analogue mood scales (VAMS) for alertness, stress, and mood. The COMPASS battery has been validated in prior studies.^17–19 Scoring and domain-specific details are provided in Supplementary Materials Section B and Figure S1.

Laboratory Procedures

Saliva samples were self-collected at home using Salivette^® tubes (Sarstedt, Germany) following staff instructions. Samples were collected at baseline (days 2–3) and post-treatment (days 15–16). For cortisol, three morning samples were taken (at awakening, +30 min, +45 min) to assess the cortisol awakening response (CAR),²⁰ along with one evening sample at 8 pm. Melatonin was measured from an additional single evening sample at 8 pm.

Subjects were instructed to avoid eating, drinking, or brushing teeth throughout the entire sampling period (despite drinking water until 10 minutes before each collection). Samples were refrigerated at home and stored at −20°C at the study site until analysis. Cortisol and melatonin levels were measured using ELISA kits (Salimetrics, LLC, USA), adhering to the manufacturer’s protocols. Hormonal changes from baseline to post-intervention were analyzed for associations with sleep and psychometric variables.

Safety

Safety assessments included adverse event (AE) monitoring and vital sign measurements (blood pressure and pulse). AEs were reported via daily electronic diaries and evaluated for causality, severity, and seriousness during each visit. Vital signs were assessed at each site visit.

Sample Size Estimate

Sample size was calculated using G*Power 3.1.9.7 for within-subject repeated measures ANOVA with two time points. Assuming alpha = 0.05, power = 0.80, and a medium effect size (f = 0.3), a minimum of 39 subjects was required. To account for the potential dropouts, 40 subjects were enrolled in the non-PSG group. An additional 10 subjects were included for PSG analysis, resulting in a total of 50 subjects.

Statistical Analysis

Analyses were performed using SPSS version 27 (IBM Corp., 2020). The intent-to-treat (ITT) population, which was identical to the safety (SAF) and per-protocol (PP) populations, included all subjects who received at least one dose of the intervention and completed all study procedures. The primary outcome was the change in sleep quality (SF-B/R sub score: SQ), while secondary outcomes included assessments of sleep quantity (PSG, diaries, wearables), daytime sleepiness, psychological well-being, endocrinological markers, and cognitive functioning.

Within-subject changes between two timepoints were analyzed using paired t-tests or Wilcoxon signed-rank tests. Repeated measures ANOVA were applied for outcomes assessed across multiple timepoints or for detecting interaction effects in subgroups (eg, time × subgroup based on anxiety level as per the STAI trait score upon inclusion). When assumptions for parametric tests were not met, equivalent non-parametric tests (eg, Mann–Whitney U) were used. All tests were two-tailed unless otherwise stated. For exploratory PSG variables, one-tailed p-values with a threshold of 0.15 were used to identify potential directional trends in this reduced sized subgroup, as is common in early-phase research. Given the exploratory nature of the study, formal corrections for multiple testing (eg, Bonferroni or FDR) were not applied to maintain sensitivity in detecting potential effects in this early phase study, while increased risk for type 1 errors is acknowledged. Data normality and variance homogeneity were assessed with Shapiro–Wilk and Levene’s tests, with transformations applied as needed.

Results

Baseline Characteristics

Out of 138 screened individuals, 50 subjects (25 females and 25 males) were eligible, enrolled, and completed the study. There were no dropouts and only minor protocol deviations occurred; thus, the ITT collective was identical to the safety collective. Table 1 shows demographic and physiological baseline characteristics of the ITT population (N = 50).

Table 1 Demography and Baseline Characteristics of the ITT Collective

The BMI, systolic and diastolic BP, and heart rate of the subjects were within normal ranges (Table 1). Normal ranges were defined as systolic/diastolic BP below 120/80 mmHg,²¹ HR between 60 and 100 bpm,²² and BMI between 18.5 and 24.9 kg/m².²³ Trait anxiety and chronic stress levels were moderate,^24,25 indicating that the study cohort was generally healthy with moderate psychological complaints in addition to their self-reported sleep issues (ICD-10, F 51.0 criteria). Subjects were highly compliant at 102 ± 12%, with two individuals over 120% and none below 80%.

Efficacy Analysis

The primary outcome was the change in sleep quality, assessed by the SF-B/R Sleep Quality (SQ) score. Secondary outcomes included sleep quantity (online sleep diaries, PSG, wearable), daytime sleepiness, psychological state (validated questionnaires), endocrinological markers, and cognitive function.

Sleep Quality Assessments

Figure 2A shows SQ scores from the SF-B/R questionnaire. The supervised completion during study visits ensured high data quality. Baseline scores averaged 4.05 ± 0.20. After the 14-day intervention, scores increased significantly to 4.61 ± 0.19 (±SEM, N = 50), representing a 13.8% improvement (p = 0.003).

Figure 2 Sleep quality scores during the 2-week intervention. (A) Sleep Quality (SQ) scores measured by the SF-B/R index (primary outcome) in retrospective of the past 2 weeks; (B) Sleep quality ratings recorded per e-sleep diaries in retrospective of the last night.**p < 0.01, ***p < 0.001 vs baseline. Error bars denote standard error of the mean (±SEM). Number of subjects contributing to the measure N = 50.

About 32% of all subjects (N = 16) experienced a ≥30% improvement in their SQ scores, while 26% (N = 13) showed no change. For trait anxiety, a median split categorized subjects into two groups: low baseline trait anxiety (0–49^th percentile; N = 25) and high baseline trait anxiety (50–100^th percentile; N = 25). A significant interaction effect for time × subgroup indicates that the temporal trajectory differs between individuals with high and low anxiety levels (F_{1, 44} = 9.53, p = 0.003): Those with high baseline trait anxiety showed a significantly greater improvement by 37.2% (N = 25, ±10.9%SEM), compared to those with low baseline trait anxiety 13.1% (N = 25, ±9.7%SEM as depicted in Supplementary Section A and Table S3).

The majority of SF-B/R subscales improved significantly, except for “total sleep time”, “feeling exhausted before sleep”, “dream recall”, and “sleep-wake pattern” (see Supplementary Section A and Table S1).

E-diary entries (see Figure 2B and Supplementary Section A, Table S2) confirmed these observations with positive effects on sleep quality measures. Physical tension scores decreased on average from 4.12 ± 0.20 to 3.2 ± 0.18 score points (±SEM, −22.3%, p < 0.001) after 1 week, and to 2.82 ± 0.18 (±SEM, −31.6% p < 0.001) after 2 weeks, and psychological tension decreased from 4.15 ± 0.24 to 3.28 ± 0.19 after 1 week (±SEM, −21.0, p < 0.001), and to 2.94 ± 0.2 after 2 weeks (±SEM, −29.2%, p < 0.001). Sleep enjoyment scores increased from 4.69 ± 0.2 to 5.52 ± 0.17 (±SEM, +17.7%, p < 0.001) resp. to 5.75 ± 0.19 (±SEM, +22.6%, p < 0.001), and feeling of restoration from 4.32 ± 0.19 to 5.26 ± 0.19 (±SEM, +21.8%, p < 0.001), resp. to 5.55 ± 0.19 (±SEM, +28.5%, p < 0.001) after 1 and 2 weeks.

Sleep quality improvements occurred within ≤1-week intervention, with further gains during the subsequent period.

Sleep Quantity Assessments

Data from electronic sleep diaries revealed modest effects in sleep quantity over the intervention period. After two weeks, time in bed (TIB) increased from 8.06 ± 0.14 hours to 8.36 ± 0.10 hours (±SEM, N = 50, +3.4%, p = 0.035), total sleep time (TST) increased from 6.92 ± 0.14 hours to 7.35 ± 0.09 hours (±SEM, N = 50, +6.2%, p = 0.005) and the number of awakenings (NWAK) decreased from 2.57 ± 0.22 events to 1.78 ± 0.14 events (±SEM, N = 50, −30.7%, p < 0.001). Sleep onset latency (SOL) and sleep efficiency did not change significantly (p > 0.05).

In the PSG subgroup (N = 10), no significant pre-post changes were observed for PSG measurements of SOL, TIB, TST, WASO, SE, light sleep stages (N1, N2) or Desaturation index (DESAT). However, directional trends were noted using one-tailed analysis (p < 0.15), including a reduction in REM sleep from 55.9 ± 22.5 min to 44.1 ± 25.1 min (−18% ratio change, p = 0.111) and an increase in deep sleep (N3+N4) from 62.3 ± 44.9 min to 73.3 ± 44.3 min, (+27.8% ratio change, p = 0.116). Notably, N4 sleep significantly increased from 15.7 ± 22.4 min to 24.6 ± 24.1 min (+70.3% ratio change, p = 0.042). The Apnea-hypopnea index (AHI) declined from 2.22 ± 2.08 to 1.65 ± 1.53 (−25.7% ratio change, p = 0.110). No significant changes were found in wearable-derived metrics.

Psychological and Endocrinological Assessments

Significantly improved psychological well-being was observed following the 14 days’ intervention. Daytime sleepiness (Epworth Sleepiness Scale) decreased from 8.84 ± 3.58 to 7.72 ± 3.67 (−12.7%, p = 0.006). State anxiety (STAI-X1) dropped from 36.52 ± 8.38 to 33.18 ± 7.39 (−9.2%, p < 0.001) after two weeks. Depression (DASS) scores fell from 3.08 ± 2.78 to 2.38 ± 2.66 (−22.7%, p = 0.014), anxiety (DASS) from 2.46 ± 2.42 to 1.84 ± 2.22 (−25.2%, p = 0.018), and stress (DASS) from 6.36 ± 3.64 to 5.22 ± 3.36 (−17.9%, p = 0.014). Quality of life (WHOQOL-BREF) improved slightly, from 74.25 ± 14.60 to 77.75 ± 13.19 (+13.8%, p = 0.008), while social happiness (OHQ) increased modestly from 4.48 ± 0.58 to 4.58 ± 0.59 (+2.2%, p = 0.033).

Endocrine outcomes showed no overall changes; exploratory subgroup trends suggest anxiety-related variation, warranting further investigation (see Supplementary Materials Section C, Tables S8–S10).

Cognitive Function Assessments

Tasks addressing executive function improved modestly over 14 days (+13.2% on average, p < 0.001), particularly in reaction time and calculation accuracy. Memory performance also improved post-intervention. While acute assessments showed transient declines in certain memory tasks (eg, delayed recall, picture/word recognition), these were paralleled by improvements in executive function tasks and attention (eg, serial subtraction and peg-and-ball tasks) and reduced subjective stress (VAS scale), suggesting a more refined acute response (see Supplementary Materials Section B, Tables S4–S7).

Clinical Safety Outcomes

During the 14-day intervention, 39 adverse events (AEs) were reported by 25 subjects via daily diaries and reviewed by the principal investigator. Of these, 69% were mild and 31% moderate in severity. Most AEs (79%) were deemed unrelated or unlikely related to the intervention, while 21% (8 events in 5 subjects) were classified as possibly related, including headaches, stomach aches, and circulatory complaints. Of these, mild cases accounted for 13% and moderate for 8% of events, of which all were self-limited and resolved spontaneously.

No severe or serious adverse events (SAEs) occurred nor were any follow-ups required. Vital signs remained stable throughout the intervention, indicating good overall tolerability of the formulation.

Discussion

Lactuca sativa (common lettuce) has long been a calming agent, with sesquiterpene lactones (eg, lactucin, lactucopicrin) shown to be bioavailable and active for up to six hours after ingestion.^26,27 Mature, freshly harvested plants offer significantly higher concentrations of these compounds.^28,29 Clinical studies report improvements in insomnia symptoms, including reduced PSQI scores.^30,31 Melissa officinalis (lemon balm) has demonstrated sleep-enhancing and anxiolytic effects.^32–35 L-Tryptophan, a precursor of serotonin and melatonin, has shown benefits for sleep and mood,^36–38 while magnesium supports sleep regulation via NMDA and GABAergic modulation.^39,40 The Sleep Well granulate combines these ingredients into a multi-modal formulation intended to support sleep and mental performance. The combination of Melissa (potentially GABAergic), magnesium (with anti-inflammatory and neuromodulatory effects), and L-Tryptophan (a precursor to both serotonin and melatonin) is hypothesized to act synergistically via complementary antioxidant, neurotransmitter, and circadian-regulating pathways—supporting broader sleep—wake stabilization than single agents.

This exploratory study suggests that 14 days of supplementation improved sleep quality, especially in anxious subjects, with supporting effects on well-being and cognitive function.

The primary outcome sleep quality, assessed via the SF-B/R questionnaire, improved by 13.8% overall and by +37.2% in anxious individuals—comparable to interventions like Valerian or Oxazepam,⁴¹ reporting 23.6% and 29.2% improvements in sleep quality (reported with the identical SF-B/R questionnaire) over 2 weeks. Unlike single-agent preparations such as melatonin or Valerian, which primarily target circadian rhythm or GABAergic pathways, respectively,^36,37,42 the Sleep Well formulation combines ingredients acting via complementary mechanisms. Notably, L-Tryptophan serves as a precursor to both serotonin and melatonin—two neurotransmitters essential to sleep regulation—thereby offering broader neurochemical support than, eg, melatonin alone. E-diary ratings in our study further showed up to a 29% increase in restoration and 32% reduction in tension, aligning, for example, with findings from saffron trials requiring a longer duration of one month, where saffron supplementation improved sleep quality by 21.5% compared to placebo (reported with the PSQI questionnaire).⁴³

Sleep diary data showed modest increases in sleep duration (TIB and TST) and fewer awakenings, while wearable and PSG data showed no significant changes in duration. However, PSG results suggested trends to increased deep sleep, which could explain improvements in perceived subjective sleep quality. Subjective-Objective Discrepancies (SOSD) between subjective and objective data are well documented and may reflect perceptual and psychological shifts, especially in anxious individuals.^44–46 Technical limitations, such as limitation in the wearable algorithm sensitivity and small PSG sample size may also contributed.^47,48 Nevertheless, interpretations are supported by theories emphasizing sleep quality over quantity in restorative outcomes.^49–52

Psychological outcomes improved consistently over the 14-days period and across measures, with reductions in daytime sleepiness (ESS), stress and anxiety (DASS), and increases in quality of life (WHOQOL-BREF) and happiness (OHQ)—consistent with literature linking better sleep to enhanced psychological states.^53,54 Though no group-level endocrine changes were observed, a higher melatonin-to-cortisol ratio was found in anxious individuals’ post-treatment (see Supplementary Materials Section C), suggesting possible subgroup-specific neuroendocrine effects, warranting further investigation.

Cognitive performance improved modestly, particularly in executive function and memory recall. These effects may relate to increased deep sleep, which has been linked to mental and physical recovery.^42,55–57 Minor fluctuations after acute dosing were transient and likely reflect task fatigue. Since study site visits were conducted depending on scheduling availability, but not at a uniform time for all participants, variability in mood or cognitive performance is acknowledged, but also supports generalizability to everyday settings.

The intervention demonstrated a favourable safety profile, with no serious adverse events and high tolerability. Most of the daily assessed AEs were mild and unrelated to the intervention. Importantly, the observed AE rate is consistent with expectations in exploratory studies where proactive daily monitoring tends to increase reporting frequency.^58,59 The absence of SAEs and no significant changes in vital signs further underscore the intervention’s safety. Long-term safety should be monitored in future studies to confirm these preliminary findings.

Limitations of the Study

The open-label, single-arm design limits causal inference and necessitates cautious interpretation. The absence of a control group and the small PSG sample reduce generalizability and objective power. No light-controlled conditions were applied during salivary melatonin sampling, and neither the timing nor caffeine or food intake were controlled for the administration of the cognitive test battery. Future trials should adopt randomized, placebo-controlled designs with larger samples and longer follow-up to validate and extend these findings.

Conclusion

This exploratory study suggests potential for the Sleep Well granulate formulation to improve sleep quality, psychological well-being, and cognitive performance—particularly among individuals with elevated anxiety; however, confirmatory evidence from placebo-controlled trials is warranted.