Kangaroo Care & Maternal Stress: A Randomized Controlled Trial

Arthur J. Engler, University of Connecticut School of Nursing

Document Type Article

Table available in full text .pdf version.


Purpose: To evaluate the effect of skin-to-skin contact between mother and newborn, or Kangaroo Care on maternal stress levels in the neonatal intensive care unit (NICU).

Theoretical Framework: Anderson’s Mutual Caregiving.

Methods: Design: Randomized controlled trial. Sample: The sample consisted of 25 mothers (experimental = 13, control = 12). Mean maternal age was 26.88 years (SD = 6.438), with 48% being Caucasian. Mean infant gestational age at birth was 32.779 (SD = 2.7932) weeks, mean birth weight was 1985.72 g (SD = 757.033), mean postnatal age was 17.32 days (SD = 19.326), with 64% being male, 76% having 5-minute Apgars of at least 8. Setting: Two level III NICUs, one in New England and one in the Mid-Atlantic region of the U.S. Measures: Fingertip temperature, galvanic skin resistance, skin conductance, salivary cortisol, Parental Stress Scale:NICU, Multiple Affect Adjective Checklist Revised (MAACL). Analysis: Inferential and descriptive statistics.

Results: Repeated measures ANOVA for fingertip temperature was significant, Wilks’ lambda = .621, F (2, 16) = 4.887, p = .022; results were non-significant for galvanic skin resistance, skin conductance, and salivary cortisol. Repeated measures ANOVA for Sights & Sounds, Wilks’ lambda = .778, F (1, 23) = 6.548, p = .018; Parental Role Alterations, Wilks’ lambda = .807, F (1, 23) = 5.448, p = .028; and Looks & Behavior, Wilks’ lambda = .671, F (1, 23) = 11.26, p = .003, were all significant (Table).

A Students’ t-test showed a significant difference in MAACL dysphoria T scores after treatment, t (8) = 3.790, p = .005, with experimental group mothers having significantly lower mean scores than control group mothers (M = 43.00, SD = 1.549 vs. M = 56.75, SD = 8.958).

Conclusions & Implications: Although the findings were not as strong as expected, the Mutual Caregiving Model was supported in several ways, especially regarding depression and parental stress in the NICU. Subsequent analysis showed that a variety of factors might have accounted for these findings, including sample size, technical difficulties with study equipment, and changes in the setting.