Effect of Massage With Verbal Contact by Mothers of Premature Infants on Maternal Attachment and Infant's Motor Performance: A Randomized Controlled Trial

ABSTRACT

Background and Aims

An infant born before 37 weeks of gestation is called a preterm infant. In the Neonatal Intensive Care Unit (NICU), occupational therapists provide essential therapeutic interventions, including those for social‐emotional development, the promotion of parent‐infant attachment and interactions, and the developmental assessment of preterm infants. This research aimed to explore the effects of maternal massage training combined with verbal contact with preterm infants on maternal attachment and infant motor performance.

Methods

84 preterm infants in the NICU were assigned to intervention and control groups. The infants in the control group underwent standard interventions, while the mothers of the infants in the intervention group gave their babies a massage for 15 min daily over 5 days and talked to their babies during the massage. Maternal Postnatal Attachment Scale (MPAS) and the Test of Infant Motor Performance (TIMP) were utilized for assessment and the data were analyzed using SPSS version 27.

Results

In both groups, MPAS and TIMP post‐test scores increased compared to pre‐test scores, although the increase in scores was significantly greater in the intervention group than in the control group (95% CI [−6.96, −1.73], p = 0.001% and 95% CI [−0.93, −0.14], p = 0.009, respectively). However, no statistically significant weight gain was observed in either the intervention or control groups (95% CI [−186.14, −129.14], p = 0.720).

Conclusion

Massage along with verbal contact by mothers of premature infants enhances maternal attachment and the infants' motor performance, yet it does not significantly affect the infants' weight gain. Consequently, given the difficulties these babies and their caregivers encounter, massage and verbal contact may serve as a safe and economical approach.

1. Introduction

A preterm infant is characterized as an infant born prior to 37 weeks of pregnancy. Every year, around 15 million preterm infants are delivered globally, representing over 1 in every 10 births (10.6%) [1]. These infants, together with their caregivers, spend their early days in the Neonatal Intensive Care Unit (NICU), a complex setting where they receive maternal care and intricate medical treatments. Preterm infants face challenges related to underdeveloped body structures and functions, areas of occupational performance, skill acquisition, and interactions with their environment, all of which affect both individual and social involvement. These obstacles can serve as hindrances to participating in meaningful and purposeful activities in everyday life [2, 3].

In both advanced and emerging nations, occupational therapists work alongside other professionals in the NICU to deliver essential therapeutic interventions like encouraging socio‐emotional growth, enhancing parent‐infant attachment and interactions. They also evaluate upper limb abilities, head stability, non‐nutritive sucking, massage techniques, kangaroo care, promoting flexor positioning, and taking into account daily activities [3, 4, 5, 6].

Massage therapy consists of the organized use of physical touch. This stimulation engages sensory receptors in the skin, muscles, bones, and joints. Massage encompasses touch, strokes, rubbing, stretching, pressing, and passive joint movements [7]. Infant massage has demonstrated beneficial effects on preterm infants, such as increases in subcutaneous bilirubin levels, sleep quality, caloric intake, vagal tone, gastric function, variability in heart rate, parameters of the immune system, changes in electroencephalograms, behavior, neurodevelopment, pain control and duration of hospital stay [8]. Insufficient or unsuitable tactile stimulation in preterm infants may cause developmental delays, growth issues, and challenges in establishing bonds with their parents [9]. Maternal infant massage has shown notable advantages for mothers, such as lower anxiety and stress levels, improved adjustment to the motherhood role, enhanced maternal confidence, and quicker recovery from postpartum depression. Despite existing programs to promote maternal‐infant attachment and occupational therapy (OT) interventions in the NICU, premature birth is a factor that causes stress, anxiety, and depression in mothers of premature infants; such that these mothers are at high risk for developing mental health disorders. High levels of stress in mothers of premature babies lead to reduced self‐efficacy [10]. They also experience high feelings of guilt and hopelessness due to premature birth. Due to their high anxiety levels, it is difficult for them to participate in caring for their infant and use of appropriate coping mechanisms, and they often take a passive role in caring for their child, relying on the medical staff for this purpose. If these negative feelings and separation of the parent from the baby are prolonged, the formation of mother‐infant attachment becomes more difficult, and the mother cannot assume her maternal role towards her infant [11]. A study conducted by Shoghi et al. examined how infant massage performed by mothers influences maternal attachment behaviors, revealing that massage played a role in enhancing and sustaining mother‐infant attachment [12].

Early verbal contact is a family‐centered early intervention where the parent and infant participate in emotionally significant communication. Verbal communication triggers physiological, behavioral, and emotional changes, preserves physiological balance, decreases respiratory distress, enhances blood oxygen levels, and alleviates maternal stress. It acts as the basis for social interactions and, as a result, strengthens the attachment between mother and infant [13, 14]. Yu et al. reported that a mother's voice lowered infant's heart rate and enhanced mother‐infant attachment [15].

Attachment is an emotional condition between a child and their caregiver that starts in the early stages of life [16]. A good early attachment between parents and infants, characterized by parents being attuned and responsive to their child's emotional and physical needs, establishes the groundwork for emotional regulation, and the ability to establish intimate relationships later on [17]. Infants admitted to the NICU after their birth lose immediate maternal contact, which adversely affects both the infant and the mother. This scenario heightens the likelihood of maternal depression and stress, leading many mothers to undergo psychological and emotional shifts in the early postpartum phase. These alterations diminish maternal sensitivity and reaction to the infant's needs, which may negatively impact the child's growth and development, ultimately affecting their relationships [4, 18, 19].

Infants utilize movement to express themselves and engage with their surroundings, alter their position, react to external stimuli, and comfort themselves [20]. Preterm infants are at significant risk for motor deficits, which can continue into their teenage years [21]. For instance, transient neurological abnormalities occur in 40%–80% and severe neurosensory sequelae occur in 4%–20% of preterm infants [22].

Massage stimulates tactile receptors and causes the release of oxytocin, reduces anxiety and increases emotional bonding. The mother's voice activates the auditory system and emotional centers of the infant's brain, causing calmness, physiological balance and increasing emotional connection in the infant. Performing these two stimuli simultaneously creates a multisensory experience similar to the intrauterine environment and can enhance neural regulation and attachment between mother and infant [14, 23, 24, 25]. Therefore, it seems that combining massage and mother‐infant verbal contact can enhance the effects of both interventions.

Considering the above evidence and based on the findings mentioned and considering the vital role of maternal interaction and OT in the NICU, this randomized clinical trial was carried out to investigate the impact of infant massage along with verbal interaction by mothers of preterm infants on maternal attachment and infant motor performance.

2. Materials and Methods

2.1. Participants

This research was a randomized clinical trial conducted on 84 preterm infants who were admitted to the NICU at Shahid Akbarabadi Hospital in Tehran.

2.1.1. Inclusion Criteria

• Infants delivered prior to 37 weeks of gestation with a birth weight under 2500 grams at birth.

• Infant's gestational age between 32 weeks and up to 16 weeks postnatal age at the time of enrollment.

• Mothers proficient in Persian and present in the NICU.

• Infants who weren't intubated and maintained stable vital signs.

• No previous incidents of serious intraventricular hemorrhage (Grade 3 or 4), congenital disorders (cranial malformations, syndactyly), diabetes, lung or heart conditions, current sepsis, painful or infectious skin conditions, apnea, bradycardia, omphalocele, gastroschisis, cancers, hemophilia, or necrotizing enterocolitis.

2.1.2. Exclusion Criteria

• Mother's reluctance to keep participating.

• Development of health issues in the infant.

• Infants who were expected to be discharged soon at the time of study entry (according to their mothers).

Upon receiving authorization for the research proposal from the Rehabilitation Faculty of Iran University of Medical Sciences, an ethics approval code (IR.IUMS.REC.1402.174) and code from the Iranian Registry of Clinical Trials (IRCT) (79227) (Identifier: IRCT20240924063149N1; registration date: October 17, 2024; available at https://irct.behdasht.gov.ir/trial/79227) were obtained. Additionally, the study protocol was publicly available from October 17, 2024, until the start of recruitment on November 4, 2024.

The study received an official referral letter from the university. In NICU, the main researcher supplied mothers with comprehensive details regarding the study goals, and those who met the inclusion criteria and agreed to participate were enrolled upon signing a consent form and gathered demographic details through a questionnaire. Both the mother and the infant were evaluated using the Maternal Postnatal Attachment Scale (MPAS) and the Test of Infant Motor Performance (TIMP) prior to the intervention. The infant's weight was measured with a calibrated hospital scale.

2.2. Sample Size

The sample size was established according to the objectives of the study. The largest necessary sample size was determined for comparing maternal attachment scores before and after the intervention between the two groups. Using a 95% confidence level and 80% statistical power, and based on previous studies [19], a sample size of 35 participants for each group was determined, adjusted to 42 per group to accommodate a 20% dropout rate.

$$n=\frac{{\left(Z_{1-\frac{\alpha }{2}}+Z_{1-\beta }\right)}^2\left({\delta }_1^2+{\delta }_2^2\right)}{{({\mathrm{\mu }}_1-{\mathrm{\mu }}_2)}^2}.$$

$$n=\frac{{(1.96+0.84)}^2{15.88}^2}{{(20.52-12.92)}^2}~35.$$

2.3. Randomization

Simple randomization was used to assign participants into intervention and control groups. A random number generator was employed to distribute numbers 1 to 84 into two equal categories named A and B. Participants were recruited using convenience sampling and individuals who met the inclusion criteria were allocated to groups according to pre‐established randomization lists. A coin flip decided that group B would get the intervention, with group A acting as the control.

2.4. Blinding

This research utilized a single‐blind design. The outcome assessor was blinded to group allocation, whereas participants and intervention providers could not be blinded due to the nature of the intervention. Outcome evaluations were conducted by an independent occupational therapist with 3 years of NICU experience, who did not participate in the intervention implementation and was unaware of group assignments. The assessor was not involved in the randomization process, did not attend intervention sessions, and assessed infants solely at scheduled evaluation times. Furthermore, mothers were instructed to avoid disclosing their group assignment during evaluations.

2.5. Intervention

Members of the control group were given standard medical and nursing care along with routine OT treatments for 5 days. The standard OT treatments comprised kangaroo care, feeding interventions, positioning methods, and sensory consult for mothers [5]. Participants in the intervention group received the same standard interventions along with maternal infant massage paired with verbal contact. The main researcher conducted the massage training in a single session, one‐on‐one by following a protocol based on the Neonatal Massage Guidelines from the Iranian Ministry of Health [26]. This affective massage begins with the lower limbs, then moves to the abdomen and around the spine, and finally the upper limbs. First the therapist showcased the technique on a doll before instructing each mother via practical experience with her infant. After the mother showed her skill, she was taught to massage her infant for 15 min daily over five consecutive days, waiting at least 30 min post‐feeding, while engaging in natural verbal interaction with her infant. Regarding the content of this conversation, lullabies or singing were suggested, preferably with a gentle rhythm and a calm tone of voice. A training video on massage was also supplied for reference. Throughout the intervention phase, the researcher maintained regular contact with mothers to monitor adherence and provide feedback. From the 84 original participants, 24 left the study due to early discharge (12 from each group), resulting in a final sample of 60 infants (30 in each group) for the assessment after the intervention (Figure 1).

Figure 1.

Study flowchart: This diagram illustrates the process of participant recruitment, allocation, intervention, and analysis throughout the study.

2.6. Primary Outcomes

The MPAS is a self‐report questionnaire completed by mothers to assess maternal attachment after childbirth. It consists of 19 items categorized into three subscales: quality of attachment (9 items), absence of hostility (5 items), pleasure in interaction (5 items), with the total score representing overall attachment The internal consistency (IC) for both the overall scale and its subscales has been indicated as 0.69, 0.45, 0.56, and 0.33, respectively [27, 28].

The TIMP consists of two sections with observational and elicited items measuring posture and performance in infants. It comprises 42 items with scores ranging from 0 to 142 and is applicable for infants from 32 weeks of gestation to 16 weeks postnatal. This research employed only the observational part. The validity and reliability have been estimated to be excellent (ICC = 0.95–0.98) [29, 30].

2.7. Secondary Outcome

The weight of infants was measured with a RossMax hospital scale, which was consistently calibrated and displayed weight in grams.

2.8. Statistical Analysis

As the sample size was greater than 30 (n = 84), it was assumed that the data distribution followed a normal pattern according to the central limit theorem [31]. An intention‐to‐treat (ITT) method was applied, missing post‐intervention data for 24 participants were estimated using regression imputation. Regression imputation was conducted as an exploratory step, following consultation with the study statistician, since it was not prespecified in the protocol.

Independent t‐tests were used for quantitative variables (maternal attachment, infant's motor performance, and weight). Chi‐square or Fisher's exact tests were applied for qualitative data (demographic data). Effect sizes were calculated as Cohen's d [32] based on mean change scores between groups. All statistical tests were two‐sided, with a significance level set at p < 0.05. All statistical evaluations were performed using IBM SPSS Statistics for Windows, Version 27.0.

2.9. Artificial Intelligence Generated Content

AI‐based language assistance (ZeroGPT) was used for grammar correction and paraphrasing. The intellectual content of the manuscript was entirely developed by the authors.

3. Results

No statistically significant differences were observed among the groups regarding these demographic variables (birth weight (p = 0.987), gestational age (p = 0.522), infant gender (P = 1.000), maternal age (p = 0.440), maternal education (p = 0.685), maternal occupation (p = 0.608), paternal occupation (p = 0.557), family economic status (p = 0.351), birth rank (p = 0.931), infant feeding method (p = 0.641), and multiple births (p = 0.369)). (Table 1).

Table 1.

Demographic information of participants.

Variables		Groups				p value
		Intervention		Control
		n (%)		n (%)
		(with ITT) (N = 42)	(without ITT) (N = 30)	(with ITT) (N = 42)	(without ITT) (N = 30)	(with ITT)	(without ITT)
Infant gender	Girl	15 (35.7)	11 (36.7)	15 (35.7)	11 (36.7)	1.000	1.000
	Boy	27 (64.3)	19 (63.3)	27 (64.3)	19 (63.3)
Birth rank	First	(57.1) 24	17 (56.7)	23 (54.8)	16 (53.3)	0.931	0.916
	Second	(35.7) 15	10 (33.3)	17 (40.8)	12 (40)
	Third	2 (4.8)	2 (6.7)	2 (4.8)	2 (6.7)
	Forth	1 (2.4)	1 (3.3)	0 (0)	0 (0)
Infant's feeding type	Breast milk	32 (76.2)	22 (73.3)	28 (66.7)	19 (63.3)	0.641	0.668
	Dry milk	2 (4.8)	1 (3.3)	3 (7.1)	2 (6.7)
	Both	8 (19)	7 (23.3)	11 (26.2)	9 (30)
Multiple births	Singleton	31 (73.8)	23 (76.7)	30 (71.4)	21 (70)	0.369	0.241
	Twins	9 (21.4)	5 (16.7)	12 (28.6)	9 (30)
	Triplets	2 (8.4)	2 (6.7)	0 (0)	0 (0)
Maternal education	Under diploma	11 (26.2)	8 (26.7)	9 (21.4)	8 (26.7)	0.685	0.846
	Diploma	17 (40.5)	12 (40)	9 (45.2)	11 (36.7)
	Associate degree	3 (7.1)	2 (6.7)	19 (11.9)	4 (13.3)
	Bachelor's degree	5 (11.9)	4 (13.3)	5 (16.7)	5 (16.7)
	Master's degree	4 (9.5)	4 (13.3)	7 (4.8)	2 (6.7)
	PhD	2 (4.8)	0 (0)	0 (0)	0 (0)
Maternal occupation	Unemployed	33 (78.6)	24 (80)	31 (73.8)	21 (70)	0.608	0.371
	Employed	9 (21.4)	6 (20)	11 (26.2)	9 (30)
Paternal occupation	Unemployed	1 (2.4)	1 (3.3)	2 (4.8)	1 (3.3)	0.557	1.000
	Employed	41 (97.6)	29 (96.7)	40 (95.2)	29 (96.7)
Family economic status	Very good	3 (7.1)	3 (10)	1 (2.4)	0 (0)	0.351	0.125
	Good	7 (16.7)	6 (20)	11 (26.2)	10 (33.3)
	Moderate	28 (66.7)	18 (60)	22 (52.4)	13 (43.3)
	Bad	4 (9.5)	3 (10)	7 (16.7)	6 (20)
	Very bad	0 (0)	0 (0)	1 (2.4)	1 (3.3)
Birth weight (grams) (mean (standard deviation))		1868.6 (416.28)	1809.3 (441.00)	1870 (388.29)	1815.7 (391.01)	0.987	0.953
gestational age of the baby (weeks) (mean (standard deviation))		32.3 (2.32)	32.2 (2.34)	32.7 (2.43)	32.6 (2.56)	0.522	0.601
Infant age (days) (mean (standard deviation))		15.8 (11.90)	16.3 (13.00)	12.7 (10.84)	12.9 (10.16)	0.216	0.259
Mother's age (years) (mean (standard deviation))		32.5 (6.79)	33 (7.53)	31.3 (7.25)	3.7 (7.28)	0.440	0.248

Findings from independent t‐tests revealed statistically significant differences in maternal attachment and infant motor performance scores between the intervention and control groups after the intervention (95% CI [−6.96, −1.73], p = 0.001% and 95% CI [−0.93, −0.14], p = 0.009, respectively). However, the intervention did not lead to a statistically significant increase in weight (95% CI [−186.14, −129.14], p = 0.720) (Table 2).

Table 2.

Comparison of the two intervention and control groups before and after the intervention.

Variable		Groups				p value		Effect size
		Intervention		Control
		With ITT	Without ITT	With ITT	Without ITT	With ITT	Without ITT	With ITT	Without ITT
Maternal attachment (Mean (Standard Deviation))	Pre‐test	81.6 (6.39)	82.5 (6.31)	81.6 (6.86)	80.9 (6.51)	0.987	0.357	0.797	0.686
	Post‐test	86.9 (5.32)	86.9 (6.28)	82.5 (6.66)	82.5 (7.84)	0.001	0.020
TIMP (Mean [Standard Deviation])	Pre‐test	7.7 (1.11)	7.8 (1.13)	7.6 (1.02)	7.6 (1.25)	0.574	0.451	0.621	0.492
	Post‐test	8.2 (0.78)	8.2 (0.81)	7.7 (1.21)	7.7 (1.22)	0.009	0.043
Infant weight (Grams) (Mean [Standard Deviation])	Pre‐test	1822.9 (398.03)	1762.3 (418.62)	1851.3 (354.19)	1809.0 (362.57)	0.730	0.231	0.057	0.652
	Post‐test	1814.9 (387.41)	1800.1 (433.66)	1843.4 (337.11)	1834.4 (367.74)	0.720	0.270

The effect size for maternal attachment was 0.797 (95% CI [0.35, 1.24]), for the infant's motor performance was 0.621 (95% CI [0.18, 1.06]) and for the infant's weight was 0.057 (95% CI [−0.27, 0.16]).

During the study period, there were no observed or reported adverse events or safety issues associated with the intervention. All mothers and infants tolerated the procedures well, and no participants withdrew due to adverse effects.

4. Discussion

We interpreted clinical relevance using effect sizes, because established MCID values are not available for MPAS and TIMP. The effect size for MPAS scores was 0.797 (95% CI [0.35,1.24]), which, according to Cohen's criteria, is considered a large effect size. Additionally, the effect size for TIMP scores was 0.621 (95% CI [0.18, 1.06]), which falls within the moderate to large range and the effect size for the infant's weight was calculated to be 0.057 (95% CI [−0.27, 0.16]), which is considered a small effect size. The 95% confidence intervals [33] supported the reliability of the effects on maternal attachment and infant motor performance, because their ranges did not include zero. However, the confidence interval for infant weight crossed zero, suggesting that this change was not meaningful. Maternal massage combined with verbal contact increased attachment in mothers of premature infants and improved the infants; motor performance but did not significantly affect the infants' weight.

Mental health problems in mothers of premature infants are associated with problems in the mother‐infant relationship. These mothers experience varying levels of stress in the NICU based on their age, education level, and income. In addition, 25%–40% of them suffer from postpartum depression during their infant's NICU stay. They are more intrusive and less coordinated and more likely to overstimulate their infants. Simultaneously, premature infants give ambiguous signals, which means that the mother cannot read and respond to cues, which ultimately has a negative impact on mother‐infant attachment [34, 35]. Given the psychosocial variability in mothers' NICU experiences and the findings of the studies mentioned above, these variables may moderate the effects of the intervention in the present study.

Affective touch activates C‐tactile afferents and stimulates the insular cortex in the brain. This area is associated with emotional processing and attachment. These fibers also cause the release of oxytocin, which forms attachment and trust while reducing stress, thus leading to the formation of mother‐infant attachment. According to the “skin as a social organ” theory, affective touch engages the mother and infant in a nonverbal communication that is essential for establishing attachment. Premature infants have significantly more limited opportunities to receive touch due to the specific conditions of the NICU and separation from their mothers. On the other hand, they spend the final weeks of pregnancy outside the womb, deprived of its multisensory inputs (including tactile and auditory inputs), which causes disruption in their sensory systems. Ultimately, emotional and cognitive development, as well as the ability to regulate emotions are affected. Therefore, providing stimuli that are similar to sensory experiences in the womb, such as hearing the mother's voice while maintaining skin‐to‐skin contact with the baby, can help promote attachment [23].

Maternal massage along with verbal contact enhanced maternal attachment to the infant in the intervention group compared to the control group. Onozawa et al. [36] discovered that the interactions between mothers and infants who underwent massage were notably improved compared to infants who did not receive massage. Similarly, Gurol and Polat [16] discovered a notable enhancement in maternal attachment. Shoghi et al. [12] indicated that maternal massage for preterm infants in the NICU facilitated the mother‐infant attachment. In their study, Hirschel et al. [24] reported that maternal massage of premature infants, combined with the mother's voice, reduced cortisol secretion and increased oxytocin secretion (hormones involved in enhancing attachment) in mothers. Nonetheless, a few studies have indicated varying results. Elliott et al. [37] reported no notable enhancement in parent‐infant interaction after massage. The researchers observed that 65% of mothers did not completely follow the protocol, potentially influencing the outcomes. Additionally, their study mainly concentrated on infant crying instead of maternal attachment, indicating that more research on the direct impact of massage on mother‐infant interactions is necessary. In a comparable manner, Miles et al. [38] found that skin‐to‐skin contact (kangaroo care) did not markedly enhance maternal attachment with preterm infants. They ascribed this outcome to the possible existence of response bias and publication bias in earlier research indicating positive outcomes of kangaroo care. Moreover, mothers displayed minimal compliance in participating in skin‐to‐skin contact, potentially undermining the intervention's effectiveness.

The current study indicates that massage along with verbal contact caused enhanced motor performance in infants. Inal et al. [39] studied the impact of a massage on the mental—motor development of healthy infants and observed an enhancement in gross motor and fine motor skills in full term infants. Moreover, According to previous studies [40, 41] tactile kinesthetic stimulation (TKS) have a beneficial effects on motor behavior in low birth weight preterm infants. Nonetheless, the current study's results are at odds with a systematic review by Orton et al. [42], which determined that early developmental interventions did not notably enhance motor outcomes in preterm infants following discharge. This difference might arise from the diversity present in the studies included, differences in evaluation instruments, ways of delivering interventions, and specific areas of intervention focus. Additionally, the research featured in Orton et al.'s review mainly relied on the Bayley Scales of Infant Development, which exhibit lower sensitivity compared to TIMP. The Bayley Scales evaluate overall motor development, but they might not identify mild motor impairments as well as TIMP.

The present study revealed no notable impact of massage combined with verbal contact on infant's weight gain. Mollà‐Casanova et al. [43], determined that massage alone, without passive mobilization, does not noticeably improve weight gain in preterm infants. In their study, infants received massages for 5 days, which might have been too brief a duration to affect weight gain. This clarification is backed by Field et al. [44], who found that marked weight gain was noticeable after 5 days and persisted to grow until the tenth day, indicating that an extended duration of massage might be required for significant weight variations. Conversely, Dieter et al. [45] carried out a 5‐day massage intervention and noted a statistically significant rise in weight gain in preterm infants. The difference between their findings and those of the current study may stem from their smaller sample size (16 infants per group), which could have resulted in increased variability in weight gain outcomes. Moreover, the researchers involved in that study recognized that their results needed to be replicated in more extensive studies to verify the consistency of their findings. Since infant massage combined with verbal contact did not lead to significant weight gain, this intervention might act as an adjunct therapy in the NICU, but it should not be relied upon exclusively for promoting weight gain.

4.1. Limitations

A limitation of this study was the early discharge of infants from the NICU, which hindered the completion of the 5‐day intervention in certain instances and prevented evaluation of whether the observed improvements in the MPAS and TIMP scores were sustained. This also reduced the effective sample size, which may have decreased the statistical power of the study. Another limitation was that MPAS showed low internal consistency in certain subscales (Cronbach's α = 0.33–0.56). Consequently, the results concerning attachment must be viewed carefully, and future research is recommended to utilize alternative or supplementary tools with stronger psychometric properties. The lack of control over additional interventions provided by various occupational therapists in the hospital, which could have affected the results, was another limitation. In addressing missing data, regression imputation was applied to maintain the sample size and groups comparability. However, this method may reduce variability compared to multiple imputation, so the results should be interpreted with caution. Moreover, because regression imputation was used without adjustment for standard errors and confidence intervals, the variability may be underestimated. Also, the use of simple randomization with convenience sampling and lack of allocation concealment, which can increase the possibility of selection bias is among the limitations.

4.2. Generalizability

Since massage and verbal contact are accessible interventions, easy to train and implement, and do not require special equipment or tools, they are likely applicable in other NICU settings. Also, for the above reasons, this intervention can also be implemented at home after discharge.

4.3. Future Research

Future studies with longer intervention durations are recommended to investigate the possible impacts of massage on weight gain and other developmental results in preterm infants. Because of early discharge, the planned 10‐day follow‐up was excluded from the present study, and it is hoped that the follow‐up period will be considered in future studies.

5. Conclusion

Maternal massage along with verbal contact in preterm infants increases maternal attachment and improves infants' motor performance. However, it does not cause weight gain. Given the difficulties encountered by preterm infants and their parents in the NICU, this intervention can serve as a safe and affordable method to promote mother‐infant attachment, as well as enhance motor development.

Author Contributions

Ava Monfared: conceptualization, methodology, software, writing – original draft, data curation, project administration. Malek Amini: conceptualization, methodology, software, writing – original draft. Jamileh Abolghasemi: visualization, investigation, formal analysis. Ali Mazouri: conceptualization, methodology, software, writing – original draft. Zahra Darabi Boroujeni: data curation, project administration.

Funding

The authors received no specific funding for this work.

Conflicts of Interest

The authors declare no conflicts of interest.

Transparency Statement

The lead author Malek Amini affirms that this article is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Monfared A., Amini M., Abolghasemi J., Mazouri A., and Boroujeni Z. D., “Effect of Massage With Verbal Contact by Mothers of Premature Infants on Maternal Attachment and Infant's Motor Performance: A Randomized Controlled Trial,” Health Science Reports 9 (2026): 1‐9, 10.1002/hsr2.71697.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.