Study design and participants
This double blind randomized clinical trial investigated the effect of behavioral therapy counseling on colic in infants of 46 anxious mothers of 2–6-weeks-old colicky infants admitted to the pediatric clinics of Al-Zahra, Taleghani and Children hospitals of Tabriz, Iran. The inclusion criteria were mothers of 2–6-weeks-old colicky infants with high PSAS scores (≥ 112) and at least middle school certificates, healthy term exclusively breastfed infants weighed at least 2500 grams, gained normal weight, and were diagnosed with infantile colic based on Wessel’s criteria and a pediatrician’s diagnosis. The exclusion criteria included giving birth to twins, using any chemical or herbal medicines (e.g. anti-flatulence) to treat infantile colic, suffering from congenital autoimmune diseases or nutritional problems (the infant), having a history of mental illnesses requiring medication (e.g. postpartum depression) in previous deliveries (based on the participant’s statements), having extremely high levels of anxiety, abusing drugs during pregnancy and in the postpartum period (based on the participant’s statements), receiving similar training in the past, attending formal/organized training courses to reduce anxiety in the past, being unsure about their ability to attend all counseling sessions, experiencing traumatic life events (e.g. death of loved ones, divorce, etc.) in the past 2 months, and using antibiotics.
The Sample size was determined in G-power based on the study of Aktas et al.(15) It was calculated as 19 by considering the mean infant colic score (m1 = 72.44), a default 15% decrease in the post-intervention colic score (m2 = 61.574), α = 0.05, and Power = 95%. The final sample size was then determined as 23 for each group by assuming a loss to follow-up of 20%.
Sampling And Randomization
The researcher initiated the sampling process after obtaining the approval of Ethics Committee of Tabriz University of Medical Sciences (IR.TBZMED.REC.1398.1083) and registering the study at Iranian Registry of Clinical Trials (IRCT20111219008459N14). Then, she visited the pediatric clinics of Al-Zahra, Taleghani and Children Hospitals of Tabriz and used convenience sampling to select mothers of infants diagnosed with infantile colic (based on Wessel’s criteria and a pediatrician’s diagnosis). After providing explanations about the research objectives and methods, mothers who were willing to participate in the study completed PSAS and those with PSAS scores ≥ 112 who also met other inclusion criteria were enrolled after signing informed consent forms. The researcher completed the socio-demographic questionnaire, Infant Colic Scale (ICS), and Mother-Infant Attachment Questionnaire (MIAQ) by interviewing the participants.
By considering a 1:1 allocation ratio and using randomized block design (4 and 6-individual blocks), the participants were randomly assigned to the intervention (counseling) and control groups. The type of each intervention was written on papers and placed in numbered opaque envelopes in order to conceal the allocation sequence. The participants were then provided with sealed and numbered envelopes based on their enrollment time.
Intervention
After completing the initial questionnaires (including the socio-demographic questionnaire, PSAS, ICS, and MIAQ), the participants received training about the nature and causes of colic and learned about existing methods used to reduce the severity of colic pain such as massaging, proper breastfeeding, hugging, burping, keeping calm during infant crying, etc. As prescribed by a pediatrician, colic drops were used equally for infants of mothers in both groups to control colic pain. Women in the intervention group attended 8 systematic desensitization counseling sessions (2–3 sessions per week) for 4 consecutive weeks in a cozy environment as described below.
Session 1: Teaching proper abdominal breathing; Session 2 and Session 3: Teaching and practicing progressive muscle relaxation technique with proper breathing; Session 4: Listing stressful stimuli of mothers, sequencing and generalizing their mediators, and selecting the main stimulus by mothers; Session 5: Starting the desensitization process from the stimulus with the lowest anxiety score; Sessions 6 to 8: Continuing the desensitization process based on maternal anxiety hierarchy.
Two weeks after the end of the counseling sessions, the researcher again completed the PSAS, ICS, and MIAQ by interviewing the participants. The control group received routine care and 6 weeks after completing the initial questionnaires, the post-test questionnaires were completed for them too.
Data Collection Tools
The data were collected using the socio-demographic questionnaire, PSAS, ICS, and MIAQ. The socio-demographic questionnaire included questions about parents’ age, educational qualifications, and job, family income level, place of residence, and life satisfaction. The content validity of this researcher-made questionnaire was confirmed by 10 faculty members of Tabriz University of Medical Sciences.
Infant Colic Scale (ICS): This 22-item questionnaire was developed by Ellett et al (44) to assess infantile colic. It includes subscales of cow’s milk/soy protein allergy, immaturity of the gastrointestinal system and CNS, parent-infant interaction, and infant temperament. The items are scored on a six-point Likert scale including strongly agree (score 6), almost agree (score 5), slightly agree (score 4), strongly disagree (score 1), almost disagree (score 2), and slightly disagree (score 3) (Total score range: 22–132). Higher scores indicate greater pain due to increased gas formation. Ellett et al. (44) (2002) confirmed the validity and reliability of this tool. In the present study, a Cronbach’s alpha value of 0.73 was obtained for all items. The psychometric properties of the scale have been confirmed for the Iranian population in a study which is currently under review.
Postpartum Specific Anxiety Scale (PSAS)
This 51-item questionnaire measures maternal anxiety using subscales of maternal competence and attachment anxiety, infant safety and welfare anxiety, infant care anxiety, and psychosocial adjustment to motherhood. The items are scored on a four-point Likert scale including highly relevant (score 3), very relevant (score 2), somewhat relevant (score 1), and not relevant (score 0). Streiner et al. (45)(2015) calculated the content validity ratio (CVR) of the items to provide a quantitative expression of content validity. The mean CVR for all the items was 0.76 indicating desirable content validity of the whole scale. In addition, the test-retest reliability of 0.88 (p < 0.001) revealed excellent stability of PSAS. Hassanzadeh et al. (46) standardized this tool in Iran in 2021.
Mother-Infant Attachment Questionnaire (MIAQ)
This 19-item scale assessed the attachment of mothers to infants in the 0–36 months-old age group. It is completed by the mother or any person who spends the most time with the infant. This questionnaire consists of the following three subscales
A. Attachment quality (items 3, 4, 5, 6, 7, 10, 14, 18, and 19)
B. Absence of hostility (items 1, 2, 15, 16, and 17)
C. Satisfaction with interaction (items 8, 9, 11, 12, and 13)
The summation of scores given to the three subscales determines the total attachment score. The total score ranges from 19 to 95, and higher scores indicate strong mother-infant attachment (47). In Iran, Hassanpour et al. obtained a Cronbach’s alpha of 0.73 for this scale (48).
In this study, the reliability of the questionnaires was assessed using internal consistency (Cronbach’s alpha) and test-retest reliability. The test-retest reliability was assessed by calculating intra-class correlation coefficients (ICC) for 20 mothers who completed the questionnaires twice at a two-week interval. The calculated ICC and Cronbach’s alpha values for ICS, PSAS, and MIAQ included “0.94, 0.91, and 0.93” and “0.70, 0.93, and 0.85”, respectively.
Statistical analysis
The data were analyzed in SPSS 21. The Kolmogorov-Smirnov (K-S) test results indicated the normality of the quantitative data distribution. The chi-square, Fisher’s exact, and independent t tests were used to examine the homogeneity of the groups in terms of socio-demographic characteristics. The independent t-test and ANCOVA (with controlled baseline values) were used to compare the mean infant colic, postpartum anxiety, and mother-infant attachment scores of the two groups before and two weeks after the intervention, respectively.