Secondary Logo

Journal Logo


Teaching Father-Infant Massage during Postpartum Hospitalization

A Randomized Crossover Trial

Suchy, Carol MSN, RN, IBCLC; Morgan, Gloria MSN, RNC-LRN; Duncan, Susan MSN, RNC-OB; Villar, Susan MSN, RNC-OB; Fox, Frieda BA, IBCLC; Rutledge, Dana N. PhD, RN

Author Information
MCN, The American Journal of Maternal/Child Nursing: May/June 2020 - Volume 45 - Issue 3 - p 169-175
doi: 10.1097/NMC.0000000000000613
  • Free


Father–infant relationships, especially during the early newborn period, have not been well studied (Condon, Corkindale, Boyce, & Gamble, 2013). Shortly after birth it is important to establish early infant attachment to parents (Bowen & Miller, 1980), yet this may be a time when fathers consider themselves bystanders or invisible parents (Steen, Downe, Bamford, & Edozien, 2012). Activities such as attendance during labor and birth (Bowen & Miller), cutting the umbilical cord (Brandão & Figueiredo, 2012), and skin-to-skin contact with baby (Erlandsson, Dsilna, Fagerberg, & Christensson, 2007; Shorey, He, & Morelius, 2016) promote positive feelings in fathers about their infants and patterns of positive behavioral interactions. When asked to reflect on the hospital stay of their newborns, fathers described positive feelings when engaged in knowing (e.g., holding, consoling) as well as taking-care behaviors (e.g., bathing, cleaning umbilical area) (de Montigny & Lacharité, 2004). In general, fathers want to be involved following their newborns' birth (Poh, Koh, & He, 2014; Xue, Shorey, Wang, & He, 2018), but may perceive information for them to do so is lacking and they may feel unprepared for infant care (Deave, Johnson, & Ingram, 2008).

Several father–infant studies considered fathers in the immediate postpartum period (Scism & Cobb, 2017). Fathers exhibited mainly passive behaviors related to their infants (Tomlinson, Rothenberg, & Carver, 1991); however, positive infant outcomes can be achieved when father–infant bonding begins during this time (Scism & Cobb). In their integrative review of 28 studies evaluating father–infant bonding following interventions with infants, Scism and Cobb (2017) found fathers may feel less than appreciated postpartum, are affected by experiences during labor and birth and early postpartum, and gain positive feelings by providing physical care to their newborns.

Evidence supports promoting paternal engagement with infants soon after birth. At our hospital, parents are instructed on skin-to-skin and infant care; infant massage classes are available in the community. No instruction is given solely to fathers. There is limited evidence to support infant massage, but existing data suggest little likelihood of causing harm (Hubbard & Gattman, 2017; Scism & Cobb, 2017; Shorey et al., 2016; Underdown, Barlow, & Stewart-Brown, 2010). In one study, infant massage by fathers supported less paternal stress in fathers of infants who were taught massage through formal classes in the community (Cheng, Volk, & Marini, 2011). The purpose of this study was to evaluate acceptability of infant massage instruction and its effect on paternal perceptions and attachment behaviors in early postpartum.


This randomized crossover trial used a convenience sample of fathers of healthy infants. The crossover design was used so all participants eventually received the intervention. Based on data from Brandão and Figueiredo (2012), the desired sample of 100 fathers with an effect size of .2 (small; Cohen, 1988) and one-tailed significance level of .05 gave 64% power to detect a difference in paternal bonding. The study was approved by the hospital health system's institutional review board and was conducted in 2018.


Participants were fathers ≥18 years old of healthy infants born at ≥38 weeks gestation at a faith-based Magnet hospital in Southern California with over 4,500 births annually. Fathers of infants who required medical intervention after initial stabilization or where acute decompensation of newborn condition occurred, or were exposed to suspected prenatal substance abuse were excluded.


Demographic data measured were age, ethnicity, highest level of education, language spoken at home, preparation for parenting, and history of massage instruction. No infant demographics were collected.

Father-to-Infant Bonding Scale. This scale is a previously modified version of the 8-item Mother-to-Infant Bonding Scale (Taylor, Atkins, Kumar, Adams, & Glover, 2005) with adequate internal consistency and fair test-retest reliability (Brandão & Figueiredo, 2012). Fathers were asked to read instructions (English or Spanish) indicating their current feelings; instructions noted that the eight adjectives describe common feelings that fathers have for their infants. Next to each adjective, fathers marked the box that best described their feelings (very much, a lot, a little, not at all). Three adjectives are positive sentiments (e.g., loving); five are negative (e.g., aggressive). Items were scored on a 4-point scale from most (0) to least (4) such that higher total scale summed scores indicated more bonding.

In our study, scale internal consistency was poor (α = .364) even when coefficients were calculated separately for items with positive and negative valence: .478 (positive), .258 (negative). An exploratory factor analysis was attempted using principal components analysis; although the Barlett's test of sphericity suggested further evaluation of the data factorability, the Kaiser-Meyer-Olkin test was only .572, indicating a less than ideal sample for use.

Father–Infant Observation Scale. This scale is based upon work by Toney (1983). Trained study nurses observed fathers with their infants for 5 minutes and counted numbers of attachment behaviors they saw in five categories (verbal interaction, smiling, eye contact, fingertip touching, whole-hand touching). Scores for each category were the numbers of tally marks; category scores were summed for a total score. Before observation, fathers were instructed to sit in a comfortable chair and hold their infant. Study nurse instruction on using the observation scale was followed by viewing a video of father skin-to-skin with his newborn; nurses scored the interaction and achieved >80% agreement with a score provided by a nurse educator.

Post-Study Questionnaire. During the first week after discharge, fathers were phoned and asked to complete a short interview about their experience with infant massage. Questions were based upon those asked by Cheng et al. (2011). The caller was a trained hospital translator. Interviews were conducted in the fathers' preferred language (English or Spanish). Responses were audio recorded and if in Spanish, transcribed into English by the translator.


A convenience sample of fathers of healthy infants was recruited by study nurses on the mother–baby unit. Given typical hospital discharge procedures, recruitment and initial measures were completed within the first 12 to 48 hours postbirth. After study information was shared, written consent was obtained. Consented fathers were randomized using block randomization. Participant fathers completed demographic questions and the Father-to-Infant Bonding Scale.

  • Group 1. The videotaped instruction on infant massage was played. Within 4 hours, study nurses (blind to participant randomization) observed father–infant interactions for 5 minutes using the observation scale. After the observation, nurses asked fathers if they had seen the massage video and noted this on the data collection form.
  • Group 2. Within 4 hours of consenting, study nurses (blind to participant randomization) observed the 5-minute father–infant interaction. As with Group 1, nurses asked fathers if they had seen the video. The videotaped instruction on infant massage was played.

Within 24-72 hours after hospital discharge, infants are seen at the Mother-Baby Assessment Center. During this visit, fathers completed the Father-to-Infant Bonding Scale and were reminded of the final interview phone call; phone numbers were validated.

Intervention: Infant Massage Instruction

A video instructing fathers in newborn massage was made by study team members and the women's health clinical educator and accessed on a tablet. Participants saw the video and were reminded about the importance of touch for infant development. All were informed of community infant massage classes given at the hospital and received a coupon for these.

The video (~10 minutes) showed a certified massage instructor giving instructions to a new father in infant massage. The father provided massage to his infant during the instruction. The video included the following:

  • Recommendation for hand washing and safety (newborn behaviors indicating lack of tolerance; safe positioning—on chest, neck-airway protection, head/neck flexion; protection from cold)
  • Normal newborn behaviors (quiet sleep, active sleep, drowsy, quiet alert, active alert, cry)
  • Massage basics: infant clothing, average time (10-15 minutes), recommendation for no oil/lotion
  • Getting ready: dim, quiet, warm setting; extra blanket as needed; chair for father; decision about skin-to-skin
  • What to expect afterward
  • Massage demonstration (covering Head-Back-Feet-Hands-Head)

Data Analysis

Descriptive statistics were used to describe the sample and responses to items on the Father-to-Infant Bonding Scale. Chi-squared and phi-coefficients were used to determine differences between groups. Frequencies and percentages described results of observation scores by group and by total sample; one-way ANOVA was used to determine differences between groups. A significance level of .05 was used. Analyses were done using SPSS Version 25. Responses to the open-ended interview questions were loaded into an Excel file; they were analyzed and categorized (basic content analysis; Weber, 1985) by authors (CS & DNR).


Participants were 98 well-educated fathers from 18 to 44 years of age (M 31.0 ± 5.8) with healthy term infants (Table 1). Over half identified as Hispanic. Most (83%) spoke English at home and very few (3%) had prior massage instruction. Prenatal education varied widely. Groups had baseline equivalence on all demographic variables.

Table 1
Table 1:
Demographic Data of Father Participants

Because of poor psychometric performance, no analysis was done using total scores for the Father Infant Bonding scale. Individual item scores indicated positive paternal responses to infants (79% to 97% at both times).

Fathers differed significantly in numbers of observed interactions with their infants depending upon timing of massage instruction (Table 2). Total interactions were significantly greater in Group 1 (observed immediately after video). There were significantly fewer verbal interactions and more fingertip touchings in Group 1 than in Group 2. The effect sizes (indexed by partial eta squared) were small judged by Cohen's d interpretation (.2 = small effect size) (Cohen, 1988). Thus, viewing the massage video had a small, albeit significant for two items and total score, impact on father-infant interaction behaviors.

Table 2
Table 2:
Mean Father–Infant Observation Scores between Groups

Ninety fathers seen at the Mother-Baby Assessment Center after hospital discharge completed the Father-to-Infant Bonding Scale. Final phone interview feedback from 80 fathers reached within 1 week of birth discharge was largely positive. Almost half reported they had tried massage with their infant in the time since discharge; 54% supported teaching massage to other fathers. Table 3 displays categorized responses to interview questions. In general, comments indicate enhanced self-efficacy related to interactions with infants following the instruction. Almost all fathers voiced pleasure at having been asked to be study participants and enjoyed being singled out for the instruction and special time with their infant.

Table 3
Table 3:
Phone Interview Results


We aimed to test the effect of brief father–infant video massage instruction on paternal bonding and observed father–infant interactions during postpartum as well as acceptability of this instruction to fathers. Given the ceiling effect (high baseline scores) and poor psychometric performance of the Father-Infant Bonding scale, no conclusions can be drawn related to paternal bonding perceptions.

Father–Infant Interactions. The significant changes in observed father–infant interactions according to timing of instruction (before/after the video) are a potential clinically important finding as there is “some evidence of benefits [of massage] on mother–infant interaction, sleeping and crying, and on hormones influencing stress levels” (Underdown et al., 2010, p. 11). Skin-to-skin contact can benefit fathers by increasing initial bonding and involvement with infants (Hubbard & Gattman, 2017). Most notable in our findings is the increased fingertip touching in fathers who viewed the video before the observation. The hospital-developed video shows a massage teacher giving instructions to an anxious new father and this father reacting to instructions. Given that he is being filmed and his infant shows reactivity, this father may not represent the “average” father; however, he follows instructions carefully and his actions show increases in touching behaviors (especially with fingertips) as the video evolves. Fathers in our study showed increased fingertip touching mimicking responses of the father in the video. In hindsight, we believe that a video with three parts might be more effective in letting fathers respond to their own situation (level of anxiety and responses of their infant): (1) instruction by the massage teacher on a very calm baby or doll (with discussion of infant reactivity) followed by (2) a father giving his own infant a massage and (3) debriefing by the teacher. Future studies might account for time lapse between video observation and father–infant behavior observation.

We found significantly more tactile interactions between fathers and infants when the father had been given infant massage instruction. Toney's (1983) earlier study found no difference in behaviors in fathers who spent 10 minutes with their infants during the first hour after birth and fathers who did not; 10-minute observations involved changing the infant's shirt and diaper, and occurred 12 to 36 hours after birth. Given the publication year (1983) and contrast with current practices (e.g., fathers' attendance at births, instruction about skin-to-skin care), behaviors between infants and their parents may now differ. In any case, increasing positive interactions between fathers and infants should enhance early paternal and infant outcomes as shown by a review of studies with skin-to-skin contact between fathers and infants (Shorey et al., 2016). Longer-term outcomes are unknown as are outcomes where fathers are observed and coached doing massage.

Instruction Acceptability. Reactions of fathers to study participation were predominantly positive. Study recruitment and data collection nurses reported fathers being excited about being asked to be involved at each point in the study (on mother–baby unit and assessment center postdischarge). This is similar to findings from prior studies (Condon et al., 2013; Poh et al., 2014; Scism & Cobb, 2017). Fathers made specific comments related to positive feelings and enhanced bonding between them and their newborns (Table 3).

Many of the fathers indicated that they felt more comfortable with their infants having had the massage instruction, which included tips on holding the infant and responding to cues. This may reflect an enhanced sense of security with their role, an important factor in both paternal and maternal security (Werner-Bierwisch, Pinkert, Niessen, Metzing, & Hellmers, 2018). Healthcare professionals are important contributors to paternal security through their support of fathers' participation in birth and postpartum activities (Werner-Bierwisch et al.). Study nurse interaction with the fathers indicated that they sought involvement with their infants and wanted to feel included in their care.

Study Limitations. Our sample was fathers of healthy infants. We did not limit the sample to first-time fathers or biologic fathers, and only recruited fathers at the hospital who visited mothers and infants. Thus, we were unable to draw any conclusions about the effects of having a prior child, and not being a biologic parent. We believe that infant gender and responsivity or temperament could be measured in future studies prior to observing father–infant behaviors. Several tools measure newborn responsivity although none is commonly used in standard practice; most evaluate motor activity, organization/arousal regulation, and attention/interactive capacity (Lean, Smyser, & Rogers, 2017). If these factors were measured, they then could be considered when changes in father–infant behaviors were evaluated.

We did not measure paternal sensitivity. Sensitivity is the father's ability to “perceive and to interpret accurately the signals and communications implicit in the infant's behavior, and given this understanding, to respond to them appropriately and promptly” (Lucassen et al., 2011, pp. 986-987). Observation responses were likely influenced by paternal sensitivity. Research with mothers indicates that enhanced maternal sensitivity leads to better mother–infant attachment (Lucassen et al.). Future studies evaluating massage instruction could include paternal sensitivity.

Clinical Implications

There is very little information about perinatal programs aimed at enhancing father–infant interactions (Lee, Knauer, Lee, MacEachern, & Garfield, 2018). This study is important in showing that a brief infant massage video instruction offered during postpartum hospitalization was well accepted by fathers and increased specific father–infant interactions. Fathers welcomed the chance to learn more about their infants and enjoyed watching the video. The brief instruction led to increased infant touching by fathers, which could be encouraged by several types of instruction on postpartum units. Nurses should be aware that fathers want to be involved in infant care and look for opportunities to include them.


This project was partially supported by the St. Joseph Hospital Nursing Excellence Fund. The authors acknowledge the assistance of Alison Navarro, Gabriela Valenzuela, and Cindy Collins in gathering data for this study.

Clinical Implications

  • Fathers want to be involved in their newborns' care.
  • Video instruction about infant care may be well received by fathers during pregnancy and postpartum.
  • Video instruction about infant care (e.g., massage) may have a positive impact on father–infant interactions, specifically encouraging fathers to respond to and touch their infants in appropriate ways.
  • Using a brief video in-hospital instruction about infant massage led to a high uptake (39% of fathers reported having used the technique at home within a week).
  • An infant massage videotape with three parts may be effective in letting fathers respond to their own situation: (1) instruction by the massage teacher on a calm baby or doll (with discussion of infant reactivity) followed by (2) a father giving his own newborn a massage and (3) debriefing by the teacher.
  • Nurses should look for opportunities to instruct and engage fathers in infant care.


Bowen S. M., Miller B. C. (1980). Paternal attachment behavior as related to presence at delivery and preparenthood classes: A pilot study. Nursing Research, 29(5), 307–311.
Brandão S., Figueiredo B. (2012). Fathers' emotional involvement with the neonate: Impact of the umbilical cord cutting experience. Journal of Advanced Nursing, 68(12), 2730–2739. doi:10.1111/j.1365-2648.2012.05978.x
Cheng C. D., Volk A. A., Marini Z. A. (2011). Supporting fathering through infant massage. The Journal of Perinatal Education, 20(4), 200–209. doi:10.1891/1058-1243.20.4.200
Cohen J. (1988). Statistical power for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.
Condon J., Corkindale C., Boyce P., Gamble E. (2013). A longitudinal study of father-to-infant attachment: Antecedents and correlates. Journal of Reproductive & Infant Psychology, 31(1), 15–30. doi:10.1080/02646838.2012.757694
Deave T., Johnson D., Ingram J. (2008). Transition to parenthood: The needs of parents in pregnancy and early parenthood. BMC Pregnancy and Childbirth, 8, 30. doi:10.1186/1471-2393-8-30
de Montigny F., Lacharité C. (2004). Fathers' perceptions of the immediate postpartal period. Journal of Obstetric, Gynecologic & Neonatal Nursing, 33(3), 328–339. doi:10.1177/0884217504266012
Erlandsson K., Dsilna A., Fagerberg I., Christensson K. (2007). Skin-to-skin care with the father after cesarean birth and its effect on newborn crying and prefeeding behavior. Birth: Issues in Perinatal Care, 34(2), 105–114.
Hubbard J. M., Gattman K. R. (2017). Parent-infant skin-to-skin contact following birth: History, benefits, and challenges. Neonatal Network, 36(2), 89–97. doi:10.1891/0730-0832.36.2.89
Lean R. E., Smyser C. D., Rogers C. E. (2017). Assessment: The newborn. Child & Adolescent Psychiatric Clinic of North America, 26(3), 427–440. doi:10.1016/j.chc.2017.02.002
Lee J. Y., Knauer H. A., Lee S. J., MacEachern M. P., Garfield C. F. (2018). Father-inclusive perinatal parent education programs: A systematic review. Pediatrics, 142(1). doi:10.1542/peds.2018-0437
Lucassen N., Tharner A., Van Ijzendoorn M. H., Bakermans-Kranenburg M. J., Volling B. L., Verhulst F. C., ..., Tiemeier H. (2011). The association between paternal sensitivity and infant-father attachment security: A meta-analysis of three decades of research. Journal of Family Psychology, 25(6), 986–992. doi:10.1037/a0025855
Poh H. L., Koh S. S., He H. G. (2014). An integrative review of fathers' experiences during pregnancy and childbirth. International Nursing Reviews, 61(4), 543–554. doi:10.1111/inr.12137
Scism A. R., Cobb R. L. (2017). Integrative review of factors and interventions that influence early father-infant bonding. Journal of Obstetric, Gynecologic & Neonatal Nursing, 46(2), 163–170. doi:10.1016/j.jogn.2016.09.004
Shorey S., He H. G., Morelius E. (2016). Skin-to-skin contact by fathers and the impact on infant and paternal outcomes: An integrative review. Midwifery, 40, 207–217. doi:10.1016/j.midw.2016.07.007
Steen M., Downe S., Bamford N., Edozien L. (2012). Not-patient and not-visitor: A metasynthesis fathers' encounters with pregnancy, birth and maternity care. Midwifery, 28(4), 362–371. doi:10.1016/j.midw.2011.06.009
Taylor A., Atkins R., Kumar R., Adams D., Glover V. (2005). A new Mother-to-Infant Bonding Scale: Links with early maternal mood. Archives of Women's Mental Health, 8(1), 45–51.
Tomlinson P. S., Rothenberg M. A., Carver L. D. (1991). Behavioral interaction of fathers with infants and mothers in the immediate postpartum period. Journal of Nurse-Midwifery, 36(4), 232–239.
Toney L. (1983). The effects of holding the newborn at delivery on paternal bonding. Nursing Research, 32(1), 16–19.
Underdown A., Barlow J., Stewart-Brown S. (2010). Tactile stimulation in physically healthy infants: Results of a systematic review. Journal of Reproductive & Infant Psychology, 28(1), 11–29. doi:10.1080/02646830903247209
Weber R. P. (1985). Basic content analysis. Beverly Hills, CA: Sage Publications.
Werner-Bierwisch T., Pinkert C., Niessen K., Metzing S., Hellmers C. (2018). Mothers' and fathers' sense of security in the context of pregnancy, childbirth and the postnatal period: An integrative literature review. BMC Pregnancy and Childbirth, 18(1), 473. doi:10.1186/s12884-018-2096-3
Xue W. L., Shorey S., Wang W., He H. G. (2018). Fathers' involvement during pregnancy and childbirth: An integrative literature review. Midwifery, 62, 135–145. doi:10.1016/j.midw.2018.04.013

Attachment; Bonding; Fathers; Infant massage; Newborns

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved