This study is a single-center, single-blind (data analysts), 2-arm pilot randomized clinical trial comparing 2 conditions: the intervention group, which receives a CBT-based intervention targeting sleep quality, and the control group, which receives no intervention. The randomization procedure was performed using a computer-generated random number sequence (Sealed Envelope software, version 1.21.0) and was assigned to participants’ academic record numbers to randomly designate participants to 1 of the 2 groups (intervention or control group). Random permuted blocks were also used to reduce the predictability of the random sequence and guarantee a 1:1 allocation ratio. This sequence was stored in a password-protected table and was hidden from other researchers during the study.

The study is registered at ClinicalTrials.gov with reference number NCT05273086, and the study protocol has been published [22]. This study was performed according to the CONSORT (Consolidated Standards of Reporting Trials) statement (Checklist 1).

The study was conducted at a private university located in Madrid (Spain) during the 2022 to 2023 academic year. Participants were recruited in October 2022, and the study lasted from October 2022 to March 2023, when the postmeasure tests were carried out.

A convenience sampling method was used. Eligible students were informed about the study through an in-class presentation delivered by the research team, during which the study aims, procedures, duration, eligibility criteria, voluntary nature of participation, and confidentiality safeguards were explained. Students who expressed interest in participating were screened according to the inclusion and exclusion criteria. Those who met the eligibility criteria and agreed to participate provided written informed consent before baseline assessment and randomization.

The eligibility criteria of this study were as follows: the inclusion criteria were students aged between 18 and 25 years who were enrolled in the first year of a nursing degree during the 2022 to 2023 academic year. Participants were excluded if they were diagnosed with a prior mental pathology, a sleep disorder, or both, and were treated with or without medication (hypnotics, sedatives, or melatonin); were taking medications to aid sleep; or were combining work and study.

Participants were not restricted to individuals with clinically diagnosed sleep disorders because the intervention was conceived as a universal sleep health promotion program for first-year nursing students. The objective was to promote healthier sleep-related habits and behaviors in a population exposed to academic demands and future shift-work schedules, rather than to diagnose or treat clinical or subclinical sleep disorders. This approach is consistent with current strategies aimed at promoting healthy sleep at the population level [23].

This study was designed as a pilot randomized controlled trial. Pilot studies are intended to assess feasibility and estimate key parameters for the design of a definitive trial, such as outcome variability and effect size estimates. Therefore, a formal a priori sample size calculation based on statistical power was not performed.

The methodological literature [24,25] on pilot studies suggests sample sizes ranging from 12 to 35 participants per group for this type of design. Although the protocol planned to recruit 60 participants, only 40 eligible students were enrolled during the recruitment period. This sample size also allows for the estimation of preliminary parameters with reasonable precision, in line with approaches focused on estimation rather than hypothesis testing. The findings of this study may inform the sample size calculation of future fully powered randomized controlled trials.

Independent variables were selected to characterize factors known to influence sleep quality and circadian regulation in young adults and university populations, including sociodemographic characteristics, lifestyle habits, physical activity, and contextual factors that may act as potential confounders or moderators of sleep-related outcomes. Satisfaction was additionally assessed to explore the acceptability of the intervention in this pilot study. Questionnaires on sociodemographic data and lifestyle factors were prepared ad hoc by the researchers.

First, sociodemographic data and anthropometric variables included date of birth, age, sex, and sleep habits during the study, including whether the participant slept alone or with someone. Participants reported their weight (kg) and height (cm) for BMI calculation. Second, lifestyle factors included tobacco consumption (cigarettes per day), alcohol consumption (daily, weekly, and sporadic), drug consumption (yes/no and type), consumption of stimulating drinks (eg, coffee, sugary drinks, and energy drinks), and consumption of nonstimulating sweetened beverages. Third, physical activity was assessed using the simplified version of the International Physical Activity Questionnaire (IPAQ) [33]. The questionnaire consists of 7 questions assessing frequency, duration, and intensity of moderate and intense physical activity during the previous 7 days, as well as walking and sitting time during a working day. The questionnaire is classified into low, moderate, and high physical activity based on metabolic equivalent task minutes per week. The higher the score, the more physically active the profile. The questionnaire has been validated, with a mean reliability of 0.80 [33]. Fourth, satisfaction with the program was assessed using a questionnaire adapted from the proposal by Azpeleta et al [34], consisting of 12 questions with responses based on a 5-point Likert scale. Higher scores indicated greater satisfaction.

Data collection was conducted at 2 time points: baseline (preintervention) and immediately after completion of the intervention (postintervention). At baseline, eligible participants completed self-administered questionnaires assessing sociodemographic characteristics, lifestyle factors, physical activity, and sleep quality through the PSQI. Following questionnaire completion, participants were fitted with the Kronowise 3.0 wearable device and received standardized instructions for its use during the 7-day ambulatory monitoring period.

Participants wore the device continuously for 7 consecutive days, including weekdays and weekends, during their usual daily routines in their home environment. At the end of each monitoring period, devices were returned to the research team for data download and processing using the Kronowizard platform.

After baseline data collection, participants were randomized to either the intervention or control group. Following completion of the intervention period, both groups underwent a second 7-day monitoring period under the same conditions and completed the postintervention PSQI. Participants allocated to the intervention group additionally completed the satisfaction questionnaire immediately after the intervention. All assessments and device management procedures were coordinated by an independent investigator trained in study procedures and not involved in outcome analysis.

This study was approved by the Research Ethics Committee of the Camilo José Cela University (code: 06‐22-UCJC-Sleep) on April 1, 2022. All participants were informed of the duration and characteristics of the study, as well as its voluntary nature. No financial or nonfinancial compensation was provided to participants for study participation.

Participants were free to leave the study at any time, without further consequences. The utmost professionalism and absolute confidentiality were always ensured, in accordance with the European Regulation (EU) 2016/679 on the protection of natural persons and the processing and free movement of data, the Framework Law 3/2018 on the protection of personal data and guarantee of digital rights, and the Law 14/2007 on biomedical research. Only the main project researcher had access to the encrypted database, data processing, and free movement, in accordance with the Framework Law 3/2018 on the protection of personal data and guarantee of digital rights and Law 14/2007 on biomedical research. Only the main project researcher has access to the encrypted database.

A descriptive analysis was made of all the variables included in the study. Categorical variables were analyzed using absolute and relative frequencies, whereas continuous variables were described using mean (SD) if the variable followed a normal distribution or median (IQR) if the variable followed a nonnormal distribution. Normality was checked using graphic methods (histogram) and statistical methods (Kolmogorov-Smirnov test or Shapiro-Wilk test).

A bivariate analysis was performed to compare the control group and the intervention group, using the Student t test (1-tailed) or the Mann-Whitney U test for continuous variables according to the distribution of the variable and the chi-square test or Fisher exact test for categorical variables when necessary. For pre-post comparison, the Student t test was used for paired samples, the Wilcoxon test was used for continuous variables, and the McNemar test was used for categorical variables. Effect size for parametric comparisons was estimated using Cohen d [35]. For nonparametric analyses, effect size was calculated as r based on the standardized Z statistic (r = Z / √N) proposed by Rosenthal and Rubin [36], where N represents the total sample size. Effect sizes were interpreted according to Cohen conventional thresholds for d (0‐0.3 for small, >0.3‐0.8 for moderate, and >0.8 for large effect) and r (0‐0.3 for small, >0.3‐0.5 for moderate, and >0.5 for large effect).

To account for baseline imbalances between groups, adjusted analyses were additionally performed using analysis of covariance. For each continuous sleep and circadian outcome, the postintervention value was entered as the dependent variable, treatment group (intervention vs control) as the fixed factor, and the corresponding baseline value of the outcome as a covariate. Smoking status, school schedule, and coffee consumption were additionally included as covariates because these variables differed between groups at baseline. Adjusted treatment effects (regression coefficients [β]), 95% CI, P values, and partial eta-squared (η²) were estimated. Because baseline differences were identified between the intervention and control groups, the adjusted analysis of covariance models were considered the primary analytical approach for estimating treatment effects, whereas the unadjusted bivariate analyses were retained as complementary analyses.

In accordance with the trial registration, the prespecified primary outcomes were objective sleep quality, assessed using the Kronowise system, and subjective sleep quality, assessed using the PSQI. The continuous circadian rhythm and sleep parameters generated by the Kronowise system were analyzed to provide a detailed characterization of the objective sleep outcome. Given the pilot nature of this trial, analyses of these individual parameters were considered exploratory and hypothesis-generating. Therefore, no formal adjustment for multiple comparisons was applied, and findings from these analyses should be interpreted as exploratory rather than confirmatory.

The present pilot randomized controlled trial explored the preliminary effects of a brief group-based CBT-I–informed intervention delivered as a sleep health promotion program for first-year nursing students. Overall, the intervention showed more favorable effects on subjective sleep outcomes, particularly self-reported sleep latency, whereas objective circadian rhythm and sleep parameters assessed through ACM showed small and imprecise treatment effects. Participants also reported high satisfaction with the intervention, supporting the feasibility and acceptability of implementing this type of program within a university setting.

The pattern of findings observed in this study may be explained by the characteristics of both the intervention and the study population. Unlike most CBT-I trials, which recruit participants with clinically significant insomnia symptoms, this intervention was delivered as a universal health promotion strategy to first-year nursing students regardless of their baseline sleep status [10,12,13]. Consequently, participants may have had relatively limited objective circadian impairment at baseline, reducing the opportunity to detect measurable physiological improvements over the short intervention period. At the same time, increasing participants’ awareness of sleep-related behaviors and promoting healthier sleep habits may have influenced their perception of sleep before producing detectable changes in objective circadian regulation [37].

The characteristics of the intervention itself should also be considered when interpreting these findings. Standard CBT-I programs are typically delivered over multiple sessions and incorporate individualized behavioral components, including stimulus control, sleep restriction, cognitive therapy, and the systematic review of sleep diaries, which are considered core mechanisms of treatment effectiveness [38,39]. By contrast, our intervention consisted of 2 face-to-face group sessions designed as a pragmatic university-based health promotion program rather than as an intensive therapeutic intervention. This lower treatment intensity may have been sufficient to increase participants’ awareness of healthy sleep practices and influence their perception of sleep while being insufficient to induce sustained behavioral changes that could be reflected in objectively measured circadian outcomes during the relatively short follow-up period [40,41]. Furthermore, participant engagement with the behavioral recommendations outside the classroom was not formally assessed. Consequently, it remains unclear to what extent the absence of objective effects reflects the limited intensity of the intervention, suboptimal adherence to the recommended behavioral changes, or a combination of these factors, all of which are recognized determinants of CBT-I effectiveness [39,42].

The discrepancy between subjective and objective outcomes deserves particular attention. Participants in the intervention group reported improvements in sleep latency, with trends toward better perceived sleep duration and sleep efficiency, whereas objective monitoring did not demonstrate corresponding improvements. Similar discrepancies between subjective and objective sleep measures have been described previously in young adults and may reflect the fact that these instruments capture different aspects of sleep [37]. The discrepancy between subjective and objective outcomes should also be interpreted in light of the study design. Because participants were aware of their group allocation and the control group received no active intervention, subjective outcomes were more susceptible to expectancy effects, increased awareness of sleep behaviors, facilitator attention, and group interaction. Consequently, the observed improvements in subjective sleep outcomes may partly reflect nonspecific effects associated with participation in the intervention, in addition to the specific therapeutic components of CBT-I [43]. In contrast, objectively measured circadian outcomes are less directly influenced by participants’ expectations and therefore provide a complementary perspective on intervention effects.

Our findings are broadly consistent with previous studies evaluating CBT-I and behavioral sleep interventions in university students. Randomized trials and systematic reviews have generally reported greater improvements in subjective sleep outcomes than in objective sleep measures, particularly when interventions are implemented in nonclinical student populations [13,40,44]. However, direct comparisons should be interpreted cautiously because studies differ substantially in intervention intensity, treatment duration, participant characteristics, and outcome assessment. Moreover, most previous studies have relied primarily on self-reported questionnaires or conventional actigraphy [13]. By incorporating multidimensional ACM alongside validated subjective measures, the present study provides a more comprehensive evaluation of behavioral intervention effects under real-world conditions and highlights the value of assessing both perceived and physiological dimensions of sleep [45].

Although the objective circadian outcomes did not demonstrate consistent intervention effects, these findings should not be interpreted as evidence that behavioral sleep interventions are ineffective in university populations. Rather, they suggest that a brief group-based health promotion program may be insufficient to modify physiological circadian regulation over a short period in generally healthy young adults. Future trials should evaluate whether greater intervention intensity, individualized behavioral support, reinforcement sessions, longer follow-up, and systematic assessment of participant adherence improve both subjective and objective sleep outcomes [44]. Such research may also help clarify the temporal relationship between changes in perceived sleep and objectively measured circadian outcomes.

One of the main strengths of this study is the combined assessment of subjective sleep quality and multidimensional objective circadian monitoring using a validated ambulatory wearable device under free-living conditions. This approach enabled continuous assessment of sleep and circadian function beyond self-reported outcomes—an aspect that has been infrequently incorporated into behavioral sleep intervention studies in university populations. In addition, the randomized controlled design and the high levels of participant satisfaction support the feasibility and acceptability of implementing sleep health promotion programs among first-year nursing students.

Overall, this pilot trial suggests that a brief, group-based, CBT-I–informed intervention is feasible, well-accepted, and may improve selected subjective aspects of sleep among first-year nursing students. However, measurable changes in objective circadian outcomes were not observed under the conditions evaluated. These findings support the need for adequately powered randomized controlled trials incorporating active control groups, longer follow-up periods, adherence assessment, and more intensive or individualized behavioral interventions to better establish the effectiveness of sleep health promotion programs in university populations.

One of the main limitations of this study was the lack of blinding for participants and facilitators, which is common in behavioral interventions and may introduce bias. Additionally, the absence of an active control group may affect internal validity and potentially lead to an overestimation of the intervention effects. Although the use of a usual-care control enhances ecological validity in a university context, it does not account for nonspecific effects inherent to group-based interventions, such as therapist attention, peer interaction, and participant expectations. Furthermore, the face-to-face group format itself may have contributed to the observed outcomes through contextual or placebo effects, which cannot be ruled out.

The inclusion of an active control condition (eg, group-based education sessions without CBT-I components) would have allowed for more rigorous control over these nonspecific effects. Therefore, the findings should be interpreted with caution. In addition, the group-based format of the intervention did not allow for individualized tailoring based on the specific sleep characteristics identified during preintervention assessment. Although key aspects of sleep behavior were addressed, more personalized approaches may yield stronger effects.

Furthermore, although mediation analysis was prespecified in the study protocol, it was not performed due to the limited sample size and the absence of multiple follow-up assessments required to establish temporal relationships between potential mediators and outcomes. As a result, the mechanisms through which the intervention may have influenced sleep- and circadian-related outcomes could not be explored. Future, adequately powered trials that incorporate repeated measurements over time should investigate potential mediating pathways underlying intervention effects.

In addition, the study may be subject to selection bias. Participants were recruited from a single cohort of first-year nursing students at one university, and participation was voluntary. Consequently, students with greater interest in sleep-related issues or health-promoting behaviors may have been more likely to participate. This may limit the representativeness of the sample and reduce the generalizability of the findings to other student populations, academic disciplines, or educational settings.

Although participants were randomly allocated, some baseline differences were observed between groups in variables such as smoking status, school schedules, and caffeine consumption. While additional adjusted analyses were performed and yielded results consistent with the primary findings, residual confounding cannot be completely excluded given the small sample size characteristic of this pilot study.

Future research should aim to incorporate active control conditions, larger sample sizes, and more individualized interventions. Further studies are also needed to determine the clinical relevance of changes observed in circadian parameters and to validate multidisciplinary approaches based on individual sleep and circadian profiles. Finally, this pilot trial was not intended to address implementation at the institutional level. Future implementation studies are needed to evaluate scalability, resource requirements, and cost-effectiveness before broader adoption can be considered.

This pilot randomized controlled trial provides preliminary evidence that a brief group-based CBT-I intervention may improve selected subjective sleep outcomes among first-year nursing students. The largest treatment effect was observed for self-reported sleep latency, whereas objective circadian rhythm and sleep outcomes generally showed small treatment effects with wide CIs, reflecting considerable uncertainty around the estimated effects. Participants reported high levels of satisfaction with the intervention, supporting the feasibility and acceptability of this approach within a university setting. Given the exploratory nature and limited sample size of this pilot study, adequately powered randomized controlled trials with longer follow-up are needed to confirm these findings and to better define the magnitude and durability of the intervention effects.