care, health and development Child: Original Article bs_bs_banner
doi:10.1111/cch.12166
A randomized controlled trial of burping for the prevention of colic and regurgitation in healthy infants R. Kaur*, B. Bharti† and S. K. Saini† *National Institute of Nursing Education, Postgraduate Institute of Medical Education and Research, Chandigarh, India, and †Department of Pediatrics, Advanced Pediatrics Center, Postgraduate Institute of Medical Education and Research, Chandigarh, India Accepted for publication 11 May 2014
Abstract
Keywords breastfeeding, burping, infant, infantile colic, regurgitation Correspondence: Bhavneet Bharti, Department of Pediatrics, Advanced Pediatrics Center, Postgraduate Institute of Medical Education and Research, Chandigarh 160012, India. E-mail: [email protected]
Background Efficacy of burping in lowering colic and regurgitation episodes in healthy term babies lacks evidence in literature. Methods We conducted a randomized controlled trial to compare efficacy of burping versus no-burping in 71 mother–baby dyads in community setting. Primary outcome was reduction in event rates of colic and regurgitation episodes over 3 months. Results Baseline characteristics were similar in two groups. Difference in incidence rates of colic between the control and burping group was 1.57 episodes/infant/100 weeks [95% confidence interval (CI): −0.63 to 3.76].There was statistically no significant reduction in colic episodes between burping and non-burping study subjects during 3 months of follow-up (adjusted relative risk 0.64; 95% CI: 0.22–1.86, P-value 0.41). Incidence rate difference of regurgitation episodes/infant/week between burping and control group was 4.36 (95% CI: 4.04 to 4.69) and there was statistically significant increase in burping group (adjusted relative risk 2.05; 95% CI: 1.92–2.18, P-value < 0.0001). Conclusions Although burping is a rite of passage, our study showed that burping did not significantly lower colic events and there was significant increase in regurgitation episodes in healthy term infants up to 3 months of follow-up.
Introduction Infantile colic and regurgitation are common problems of infants in first 3 months of life (Miller-Loncar et al. 2004; Savino 2007) Infantile colic, characterized by episodes of inconsolable crying, affects around one in six families and was reported to cost UK National Health Service (NHS) in excess of £65 million per year (Morris et al. 2001). It is generally considered to be a benign condition yet symptoms cause significant distress to caregivers such that physicians often feel the need to intervene (Savino & Tarasco et al. 2010). Burping after feeding is commonly advised by paediatricians, nurses and parenting websites to promote expulsion of gases that accumulate during
52
feeding with aim of decreasing discomfort and crying episodes. Scientific evidence for the efficacy of burping is lacking (Gupta 2007). We, therefore, sought to determine the impact of advising mothers on burping technique on colic and regurgitation episodes in a cohort of term newborn babies prospectively followed for 3 months using a randomized controlled design in a community setting.
Methods We conducted this trial from July to September 2011 in a suburb colony of a north Indian city, Chandigarh. The area included 3003 households with a total population of 18 000. Antenatal
© 2014 John Wiley & Sons Ltd
Role of burping in colic and regurgitation 53
80 babies assessed for eligibility
74 were eligible
6 were not eligible • • •
3 LSCS including one twin delivery (prolonged hospital stay beyond 2 weeks) 1 LBW baby hospitalized 2 mothers stayed at maternal houses
Parents declined participation in 3 babies
71 babies randomized
35 babies received burping intervention
36 babies in non-burping control arm
Lost to follow up
Lost to follow up 5 subjects lost to follow up on day 3, 5, 9, 12 and 15 2 subjects followed up telephonically after 45 days
2 subjects lost to follow up on day 9 & 11 1 mother separated from father on day 13 1 baby died of sepsis on day 30
Figure 1. Enrolment, randomization and
follow-up of the study cohort. LSCS, lower segment caesarean section; LBW, low birthweight.
30 babies had complete follow-up for 3 months
data records documented in 18 community centres (known as Anganwadis in the Indian context) were scrutinized to develop a sampling frame for pregnant women in third trimester in the target study area. Auxiliary nurse midwives enrolled every pregnant mother by home visiting with the help of Anganwadis. From that sampling frame, home visits were made by one of the investigators (R.K.) and mother–infant dyads were enrolled within 15 days of birth. Eighty mothers were assessed for eligibility and six were excluded for their non-availability because of prolonged hospitalization due to caesarean section (three mothers), preterm delivery (one mother) and migration to native villages (two mothers). Three families with eligible mother baby dyads refused to participate in the trial (see Fig. 1). Once trial started, attrition rate for next 3 months was 11% and 14% in the control and intervention group, respectively. The study protocol was approved by institutional ethics committee
32 babies had complete follow-up for 3 months
at Postgraduate Institute of Medical Education and Research, and written informed consent was obtained from caregiver of each enrolled baby. Baseline data were collected for first 3 days of enrolment and subsequently babies were randomized into control and intervention groups using computerized random generator to select blocks of four and six. Allocation concealment was achieved by using numbered, opaque, sealed envelopes containing intervention. Blinding was not possible as the intervention was obvious to caregivers. Control group was offered standard care which included advice about exclusive breastfeeding, temperature maintenance, hygiene and immunization. Control strategy deliberately excluded any counselling on burping. Daily documentation of outcome episodes by mothers occurred throughout the trial period. However, collection of data by researchers was carried out by daily home visits for first 6 weeks
© 2014 John Wiley & Sons Ltd, Child: care, health and development, 41, 1, 52–56
54
R. Kaur et al.
(except Sundays) and then weekly home visits for next 6 weeks of trial. Caregivers in the intervention group received standardized instructions regarding burping using flash cards. Burping involved holding an infant against shoulder or putting in prone position over thigh and then lightly patting on lower back so that given infant burps. Mothers could opt for any position according to their comfort. Recording of colic and regurgitation episodes for illiterate mothers involved use of tally marks. Frequencies of regurgitation and colic episodes were also totalled over a 24-h period. Regurgitation was defined as any expulsion of milk with or without other material from mouth. The operational definition of colic was characterized by symptoms, including (1) excessive crying; (2) paroxysmal onset; (3) qualitative difference in cry; (4) physical discomfort signals; and (5) inconsolability. Daily log recording was not feasible for nine families (five in intervention group and four in control group) because of parental time constraints and unscheduled movement of the families (Fig. 1). The primary outcome measure was count data about colic and regurgitation episodes. This trial was registered in ‘Clinical Trial Registry India’ with number CTRI/2011/11/002174.
Table 1. Baseline characteristics of mother–infant dyads Variable Age at enrolment (in days) (mean ± SD) Maternal age (in years) ( mean ± SD) Mother educational status ≤8 class Up to 12th class Graduate and above Type of family Nuclear family Joint family Monthly income (mean ± SD) Gender Male/females (n) Birthweight (in kg) (mean ± SD) Birth order 1 ≥2 Gestational age (in weeks) (mean ± SD) Feeding patterns Ever bottle fed Ever spoon fed Socioeconomic status† Upper (I) Upper middle (II) Lower middle (III) Upper lower (IV)
Intervention group (n = 35)
Control group (n = 36)
7.11 ± 4.3 25.03 ± 3.4
7.42 ± 3.5 23.81 ± 3.1
18 (51.4%) 9 (25.7%) 8 (22.9%)
22 (66.1%) 12 (33.3%) 2 (5.6%)
6 (17.1%) 29 (82.9%) 9057 ± 7843
9 (25.0%) 27 (75.0%) 8652 ± 6113
15/20 2.68 ± 0.4
17/19 2.77 ± 0.4
16 (45.7%) 19 (54.3%) 36.83 ± 2.2
17 (47.2%) 19 (52.8%) 36.81 ± 2.2
13 (38.2%) 11 (32.4%)
11 (30.6%) 4 (11.1%)
1 (2.9%) 8 (22.9%) 9 (25.7%) 17 (48.6%)
– 3 (8.3%) 14 (38.9%) 19 (52.8%)
†Socioeconomic status according to modified Kuppuswamy scale.
Statistical analysis We estimated that a sample size of 78 babies with baseline event rate of colic episodes in control group of 5 per 100 infant weeks and event rate ratio of 2 (control vs. intervention group) would have 90% power rate to detect statistically significant difference of colic episodes at an alpha level of 5%. The prevalence rates of colic and regurgitation were compared between two study groups by estimating incidence-rate ratios (IRR). Zero inflated Poisson models were also used to allow for ‘excess zeros’ in count outcome data under the assumption that target population is characterized by two binary groups, one with zero counts, and other with positive counts.
Results A total of 80 babies were screened; 74 were eligible for the clinical trial and 71 babies were finally enrolled (Fig. 1). The baseline characteristics of the two groups are shown in Table 1. The median (interquartile range) daily burp count was 6.58 (2.43) in intervention group. The proportion of infants in the study cohort reporting any colic at 3 months was 19.7% [95% confidence interval (CI): 10.2–29.2%] and the mean incidence rate was 2.70 episodes
© 2014 John Wiley & Sons Ltd, Child: care, health and development, 41, 1, 52–56
per 100 infant weeks. Out of total study sample (intervention and control groups included), 14 out of 71 (19.7%) infants reported at least one colic episode during study period. The proportion of study infants reporting colic episodes was statistically not different between intervention (14.3%) and control groups (25%) (P = 0.26). Caregivers of infants in the intervention group reported 1.90 episodes per 100 infant weeks of colic against 3.47 episodes per 100 infant weeks in control group (crude IRR 0.55; 95% CI: 0.20–1.38. P = 0.17) (Table 2). In view of high number of zeroes in outcome of total colic episodes leading to over dispersion of data, we used zero inflated Poisson regression to estimate the relative risk after adjusting for the frequency of breast, spoon and bottle feeding. The Vuong test (P = 0.02) indicated better fit of zero-inflated model over routine Poisson regression model. The adjusted relative risk of colic episodes was 0.64 (95% CI: 0.22–1.86) (Table 2). Yet, this effect size failed to reach statistical significance (P-value 0.41). Compared with breastfed babies, there was no statistically difference in the relative risk of colic episodes in newborns with bottle feeding (RR 1.88; 95% CI: 0.67–5.32) or spoon
Role of burping in colic and regurgitation 55
Table 2. Effect of burping on incidence of regurgitation and colic episodes on the study subjects
Outcome Episodes of regurgitation Intervention (n = 35) Control (n = 36) Episodes of colic Intervention (n = 35) Control (n = 36)
Infant weeks observed
Total
420 432
3393 1605
420 432
8 15
Incidence Episodes/infant/week 8.08 3.72 Episodes/infant/100 weeks 1.90 3.47
Crude IRR (95% CI)
Adjusted IRR† (95% CI)
2.17 (2.05–2.31) 1
2.05 (1.92–2.18) 1
0.55 (0.20–1.38) 1
0.64 (0.22–1.86) 1
IRR, incidence rate ratio; CI, confidence interval. †Adjusted for spoon feed, bottle feed and breastfeed frequencies.
400
Discussion
Episodes of regurgitation/week
350 300 250 200
Control
150
Intervention
100 50 0 1
2
3
4
5
6
7
8
9 10 11 12 13
Weeks of follow-up Figure 2. Mean frequency of regurgitation episodes/week in the control (non-burping) and intervention (burping) group.
feeding (RR 2.01; 95% CI: 0.52–7.81) irrespective of allocation. In the whole cohort, 93% reported regurgitation episodes and the mean incidence was 5.87 episodes per infant week. The proportion of infants reporting episodes of regurgitation was statistically not different between intervention (91.4%) and control groups (94.4%) (P = 0.62). The incidence rates of regurgitation episodes were 8.08 episodes per infant week in intervention group as compared with 3.72 episodes per infant week in control group [absolute risk difference 4.36 episodes per infant week (95% CI: 4.04–4.69)] (Fig. 2). The unadjusted relative risk of regurgitation in intervention versus control study groups was 2.18 (95% CI 2.05–2.31). After adjusting for spoon feeding, bottle feeding and total breastfeeding episodes using the Poisson regression model the relative risk of regurgitation was twice as high in the intervention as control group (RR 2.05; 95% CI: 1.92–2.18). This adjusted relative risk was statistically significant (P-value < 0.0001).
The prevalence of infantile colic (19%) in our cohort was within the wide reported range of 2–40% (Lucassen et al. 2001) The differences in criteria used for definition of colic and varying proportions of known high risk factors (first birth order, high socioeconomic and educational status, maternal smoking and hospital-based cohorts) could underpin this wide variability of reported colic prevalence (Wessel et al. 1954). We found no statistically significant reduction either in the prevalence of colic or in the number of episodes reported between intervention and control groups (adjusted RR: 0.64; 95% CI: 0.22–1.86). On the other hand, prevalence of regurgitation episodes was estimated to be 93% (95% CI: 89–99) in our study cohort which closely mirrored the prevalence (86.9%) reported in literature (Osatakul et al. 2002) and it was significantly higher in the intervention group (adjusted RR 2.05; 95% CI: 1.92–2.18). The present study has few limitations. Chief among these is relatively small sample size which reduced the precision of our effect size estimates. The technique and frequency of burping was not supervised, although procedure was explained rigorously to enrolled mothers in intervention group. We did not record duration of crying – maintaining log of crying time would have been a labour- and time-intensive operation for Indian mothers who have low literacy rates and are invariably pitch-forked into multiplicity of responsibilities at home besides infant care. To conclude, our results suggest that the widely used empirical practice of burping may actually worsen regurgitation frequency and provide no evidence that it significantly lowers episodes of colic in infants up to 3 months. Given the small sample size further research is needed to confirm these findings.
© 2014 John Wiley & Sons Ltd, Child: care, health and development, 41, 1, 52–56
56
R. Kaur et al.
Key messages • Infantile colic characterized by episodes of inconsolable crying affects around one in six families in UK. • Impact of empirical practice of burping on colic and regurgitation episodes has not been rigorously tested in literature. • A randomized controlled trial for burping efficacy in healthy term infants failed to demonstrate statistically significant decrease in colic episodes as compared with nonburping group. • Risk of episodes of regurgitation, on the contrary, was more than two times higher in burping group as compared with non-burping group.
Funding Nil.
Conflict of interests None declared.
Contribution statement Dr B. Bharti and Dr S.K. Saini conceived the study and contributed in the design of study, drafted the article and critically
© 2014 John Wiley & Sons Ltd, Child: care, health and development, 41, 1, 52–56
revised it. R. Kaur assisted in the acquisition of data, analysis and interpretation of data besides writing of the manuscript.
References Gupta, S. K. (2007) Update on infantile colic and management options. Current Opinion in Investigational Drugs (London, England: 2000), 8, 921–926. Lucassen, P. L., Assendelft, W. J., van Eijk, J. T., Gubbels, J. W., Douwes, A. C. & van Geldrop, W. J. (2001) Systematic review of the occurrence of infantile colic in the community. Archives of Disease in Childhood, 84, 398–403. Miller-Loncar, C., Bigsby, R., High, P., Wallach, M. & Lester, B. (2004) Infant colic and feeding difficulties. Archives of Disease in Childhood, 89, 908–912. Morris, S., James-Roberts, I. S., Sleep, J. & Gillham, P. (2001) Economic evaluation of strategies for managing crying and sleeping problems. Archives of Disease in Childhood, 84, 15–19. Osatakul, S., Sriplung, H., Puetpaiboon, A., Junjana, C. O. & Chamnongpakdi, S. (2002) Prevalence and natural course of gastroesophageal reflux symptoms: a 1-year cohort study in Thai infants. Journal of Pediatric Gastroenterology and Nutrition, 34, 63–67. Savino, F. (2007) Focus on infantile colic. Acta Paediatrica, 96, 1259–1264. Savino, F. & Tarasco, V. (2010) New treatments for infant colic. Current Opinion in Pediatrics, 22, 791–797. Wessel, M. A., Cobb, J. C., Jackson, E. B., Harris, G. S. Jr & Detwiler, A. C. (1954) Paroxysmal fussing in infancy, sometimes called colic. Pediatrics, 14, 421–435.
doi:10.1111/cch.12166
A randomized controlled trial of burping for the prevention of colic and regurgitation in healthy infants R. Kaur*, B. Bharti† and S. K. Saini† *National Institute of Nursing Education, Postgraduate Institute of Medical Education and Research, Chandigarh, India, and †Department of Pediatrics, Advanced Pediatrics Center, Postgraduate Institute of Medical Education and Research, Chandigarh, India Accepted for publication 11 May 2014
Abstract
Keywords breastfeeding, burping, infant, infantile colic, regurgitation Correspondence: Bhavneet Bharti, Department of Pediatrics, Advanced Pediatrics Center, Postgraduate Institute of Medical Education and Research, Chandigarh 160012, India. E-mail: [email protected]
Background Efficacy of burping in lowering colic and regurgitation episodes in healthy term babies lacks evidence in literature. Methods We conducted a randomized controlled trial to compare efficacy of burping versus no-burping in 71 mother–baby dyads in community setting. Primary outcome was reduction in event rates of colic and regurgitation episodes over 3 months. Results Baseline characteristics were similar in two groups. Difference in incidence rates of colic between the control and burping group was 1.57 episodes/infant/100 weeks [95% confidence interval (CI): −0.63 to 3.76].There was statistically no significant reduction in colic episodes between burping and non-burping study subjects during 3 months of follow-up (adjusted relative risk 0.64; 95% CI: 0.22–1.86, P-value 0.41). Incidence rate difference of regurgitation episodes/infant/week between burping and control group was 4.36 (95% CI: 4.04 to 4.69) and there was statistically significant increase in burping group (adjusted relative risk 2.05; 95% CI: 1.92–2.18, P-value < 0.0001). Conclusions Although burping is a rite of passage, our study showed that burping did not significantly lower colic events and there was significant increase in regurgitation episodes in healthy term infants up to 3 months of follow-up.
Introduction Infantile colic and regurgitation are common problems of infants in first 3 months of life (Miller-Loncar et al. 2004; Savino 2007) Infantile colic, characterized by episodes of inconsolable crying, affects around one in six families and was reported to cost UK National Health Service (NHS) in excess of £65 million per year (Morris et al. 2001). It is generally considered to be a benign condition yet symptoms cause significant distress to caregivers such that physicians often feel the need to intervene (Savino & Tarasco et al. 2010). Burping after feeding is commonly advised by paediatricians, nurses and parenting websites to promote expulsion of gases that accumulate during
52
feeding with aim of decreasing discomfort and crying episodes. Scientific evidence for the efficacy of burping is lacking (Gupta 2007). We, therefore, sought to determine the impact of advising mothers on burping technique on colic and regurgitation episodes in a cohort of term newborn babies prospectively followed for 3 months using a randomized controlled design in a community setting.
Methods We conducted this trial from July to September 2011 in a suburb colony of a north Indian city, Chandigarh. The area included 3003 households with a total population of 18 000. Antenatal
© 2014 John Wiley & Sons Ltd
Role of burping in colic and regurgitation 53
80 babies assessed for eligibility
74 were eligible
6 were not eligible • • •
3 LSCS including one twin delivery (prolonged hospital stay beyond 2 weeks) 1 LBW baby hospitalized 2 mothers stayed at maternal houses
Parents declined participation in 3 babies
71 babies randomized
35 babies received burping intervention
36 babies in non-burping control arm
Lost to follow up
Lost to follow up 5 subjects lost to follow up on day 3, 5, 9, 12 and 15 2 subjects followed up telephonically after 45 days
2 subjects lost to follow up on day 9 & 11 1 mother separated from father on day 13 1 baby died of sepsis on day 30
Figure 1. Enrolment, randomization and
follow-up of the study cohort. LSCS, lower segment caesarean section; LBW, low birthweight.
30 babies had complete follow-up for 3 months
data records documented in 18 community centres (known as Anganwadis in the Indian context) were scrutinized to develop a sampling frame for pregnant women in third trimester in the target study area. Auxiliary nurse midwives enrolled every pregnant mother by home visiting with the help of Anganwadis. From that sampling frame, home visits were made by one of the investigators (R.K.) and mother–infant dyads were enrolled within 15 days of birth. Eighty mothers were assessed for eligibility and six were excluded for their non-availability because of prolonged hospitalization due to caesarean section (three mothers), preterm delivery (one mother) and migration to native villages (two mothers). Three families with eligible mother baby dyads refused to participate in the trial (see Fig. 1). Once trial started, attrition rate for next 3 months was 11% and 14% in the control and intervention group, respectively. The study protocol was approved by institutional ethics committee
32 babies had complete follow-up for 3 months
at Postgraduate Institute of Medical Education and Research, and written informed consent was obtained from caregiver of each enrolled baby. Baseline data were collected for first 3 days of enrolment and subsequently babies were randomized into control and intervention groups using computerized random generator to select blocks of four and six. Allocation concealment was achieved by using numbered, opaque, sealed envelopes containing intervention. Blinding was not possible as the intervention was obvious to caregivers. Control group was offered standard care which included advice about exclusive breastfeeding, temperature maintenance, hygiene and immunization. Control strategy deliberately excluded any counselling on burping. Daily documentation of outcome episodes by mothers occurred throughout the trial period. However, collection of data by researchers was carried out by daily home visits for first 6 weeks
© 2014 John Wiley & Sons Ltd, Child: care, health and development, 41, 1, 52–56
54
R. Kaur et al.
(except Sundays) and then weekly home visits for next 6 weeks of trial. Caregivers in the intervention group received standardized instructions regarding burping using flash cards. Burping involved holding an infant against shoulder or putting in prone position over thigh and then lightly patting on lower back so that given infant burps. Mothers could opt for any position according to their comfort. Recording of colic and regurgitation episodes for illiterate mothers involved use of tally marks. Frequencies of regurgitation and colic episodes were also totalled over a 24-h period. Regurgitation was defined as any expulsion of milk with or without other material from mouth. The operational definition of colic was characterized by symptoms, including (1) excessive crying; (2) paroxysmal onset; (3) qualitative difference in cry; (4) physical discomfort signals; and (5) inconsolability. Daily log recording was not feasible for nine families (five in intervention group and four in control group) because of parental time constraints and unscheduled movement of the families (Fig. 1). The primary outcome measure was count data about colic and regurgitation episodes. This trial was registered in ‘Clinical Trial Registry India’ with number CTRI/2011/11/002174.
Table 1. Baseline characteristics of mother–infant dyads Variable Age at enrolment (in days) (mean ± SD) Maternal age (in years) ( mean ± SD) Mother educational status ≤8 class Up to 12th class Graduate and above Type of family Nuclear family Joint family Monthly income (mean ± SD) Gender Male/females (n) Birthweight (in kg) (mean ± SD) Birth order 1 ≥2 Gestational age (in weeks) (mean ± SD) Feeding patterns Ever bottle fed Ever spoon fed Socioeconomic status† Upper (I) Upper middle (II) Lower middle (III) Upper lower (IV)
Intervention group (n = 35)
Control group (n = 36)
7.11 ± 4.3 25.03 ± 3.4
7.42 ± 3.5 23.81 ± 3.1
18 (51.4%) 9 (25.7%) 8 (22.9%)
22 (66.1%) 12 (33.3%) 2 (5.6%)
6 (17.1%) 29 (82.9%) 9057 ± 7843
9 (25.0%) 27 (75.0%) 8652 ± 6113
15/20 2.68 ± 0.4
17/19 2.77 ± 0.4
16 (45.7%) 19 (54.3%) 36.83 ± 2.2
17 (47.2%) 19 (52.8%) 36.81 ± 2.2
13 (38.2%) 11 (32.4%)
11 (30.6%) 4 (11.1%)
1 (2.9%) 8 (22.9%) 9 (25.7%) 17 (48.6%)
– 3 (8.3%) 14 (38.9%) 19 (52.8%)
†Socioeconomic status according to modified Kuppuswamy scale.
Statistical analysis We estimated that a sample size of 78 babies with baseline event rate of colic episodes in control group of 5 per 100 infant weeks and event rate ratio of 2 (control vs. intervention group) would have 90% power rate to detect statistically significant difference of colic episodes at an alpha level of 5%. The prevalence rates of colic and regurgitation were compared between two study groups by estimating incidence-rate ratios (IRR). Zero inflated Poisson models were also used to allow for ‘excess zeros’ in count outcome data under the assumption that target population is characterized by two binary groups, one with zero counts, and other with positive counts.
Results A total of 80 babies were screened; 74 were eligible for the clinical trial and 71 babies were finally enrolled (Fig. 1). The baseline characteristics of the two groups are shown in Table 1. The median (interquartile range) daily burp count was 6.58 (2.43) in intervention group. The proportion of infants in the study cohort reporting any colic at 3 months was 19.7% [95% confidence interval (CI): 10.2–29.2%] and the mean incidence rate was 2.70 episodes
© 2014 John Wiley & Sons Ltd, Child: care, health and development, 41, 1, 52–56
per 100 infant weeks. Out of total study sample (intervention and control groups included), 14 out of 71 (19.7%) infants reported at least one colic episode during study period. The proportion of study infants reporting colic episodes was statistically not different between intervention (14.3%) and control groups (25%) (P = 0.26). Caregivers of infants in the intervention group reported 1.90 episodes per 100 infant weeks of colic against 3.47 episodes per 100 infant weeks in control group (crude IRR 0.55; 95% CI: 0.20–1.38. P = 0.17) (Table 2). In view of high number of zeroes in outcome of total colic episodes leading to over dispersion of data, we used zero inflated Poisson regression to estimate the relative risk after adjusting for the frequency of breast, spoon and bottle feeding. The Vuong test (P = 0.02) indicated better fit of zero-inflated model over routine Poisson regression model. The adjusted relative risk of colic episodes was 0.64 (95% CI: 0.22–1.86) (Table 2). Yet, this effect size failed to reach statistical significance (P-value 0.41). Compared with breastfed babies, there was no statistically difference in the relative risk of colic episodes in newborns with bottle feeding (RR 1.88; 95% CI: 0.67–5.32) or spoon
Role of burping in colic and regurgitation 55
Table 2. Effect of burping on incidence of regurgitation and colic episodes on the study subjects
Outcome Episodes of regurgitation Intervention (n = 35) Control (n = 36) Episodes of colic Intervention (n = 35) Control (n = 36)
Infant weeks observed
Total
420 432
3393 1605
420 432
8 15
Incidence Episodes/infant/week 8.08 3.72 Episodes/infant/100 weeks 1.90 3.47
Crude IRR (95% CI)
Adjusted IRR† (95% CI)
2.17 (2.05–2.31) 1
2.05 (1.92–2.18) 1
0.55 (0.20–1.38) 1
0.64 (0.22–1.86) 1
IRR, incidence rate ratio; CI, confidence interval. †Adjusted for spoon feed, bottle feed and breastfeed frequencies.
400
Discussion
Episodes of regurgitation/week
350 300 250 200
Control
150
Intervention
100 50 0 1
2
3
4
5
6
7
8
9 10 11 12 13
Weeks of follow-up Figure 2. Mean frequency of regurgitation episodes/week in the control (non-burping) and intervention (burping) group.
feeding (RR 2.01; 95% CI: 0.52–7.81) irrespective of allocation. In the whole cohort, 93% reported regurgitation episodes and the mean incidence was 5.87 episodes per infant week. The proportion of infants reporting episodes of regurgitation was statistically not different between intervention (91.4%) and control groups (94.4%) (P = 0.62). The incidence rates of regurgitation episodes were 8.08 episodes per infant week in intervention group as compared with 3.72 episodes per infant week in control group [absolute risk difference 4.36 episodes per infant week (95% CI: 4.04–4.69)] (Fig. 2). The unadjusted relative risk of regurgitation in intervention versus control study groups was 2.18 (95% CI 2.05–2.31). After adjusting for spoon feeding, bottle feeding and total breastfeeding episodes using the Poisson regression model the relative risk of regurgitation was twice as high in the intervention as control group (RR 2.05; 95% CI: 1.92–2.18). This adjusted relative risk was statistically significant (P-value < 0.0001).
The prevalence of infantile colic (19%) in our cohort was within the wide reported range of 2–40% (Lucassen et al. 2001) The differences in criteria used for definition of colic and varying proportions of known high risk factors (first birth order, high socioeconomic and educational status, maternal smoking and hospital-based cohorts) could underpin this wide variability of reported colic prevalence (Wessel et al. 1954). We found no statistically significant reduction either in the prevalence of colic or in the number of episodes reported between intervention and control groups (adjusted RR: 0.64; 95% CI: 0.22–1.86). On the other hand, prevalence of regurgitation episodes was estimated to be 93% (95% CI: 89–99) in our study cohort which closely mirrored the prevalence (86.9%) reported in literature (Osatakul et al. 2002) and it was significantly higher in the intervention group (adjusted RR 2.05; 95% CI: 1.92–2.18). The present study has few limitations. Chief among these is relatively small sample size which reduced the precision of our effect size estimates. The technique and frequency of burping was not supervised, although procedure was explained rigorously to enrolled mothers in intervention group. We did not record duration of crying – maintaining log of crying time would have been a labour- and time-intensive operation for Indian mothers who have low literacy rates and are invariably pitch-forked into multiplicity of responsibilities at home besides infant care. To conclude, our results suggest that the widely used empirical practice of burping may actually worsen regurgitation frequency and provide no evidence that it significantly lowers episodes of colic in infants up to 3 months. Given the small sample size further research is needed to confirm these findings.
© 2014 John Wiley & Sons Ltd, Child: care, health and development, 41, 1, 52–56
56
R. Kaur et al.
Key messages • Infantile colic characterized by episodes of inconsolable crying affects around one in six families in UK. • Impact of empirical practice of burping on colic and regurgitation episodes has not been rigorously tested in literature. • A randomized controlled trial for burping efficacy in healthy term infants failed to demonstrate statistically significant decrease in colic episodes as compared with nonburping group. • Risk of episodes of regurgitation, on the contrary, was more than two times higher in burping group as compared with non-burping group.
Funding Nil.
Conflict of interests None declared.
Contribution statement Dr B. Bharti and Dr S.K. Saini conceived the study and contributed in the design of study, drafted the article and critically
© 2014 John Wiley & Sons Ltd, Child: care, health and development, 41, 1, 52–56
revised it. R. Kaur assisted in the acquisition of data, analysis and interpretation of data besides writing of the manuscript.
References Gupta, S. K. (2007) Update on infantile colic and management options. Current Opinion in Investigational Drugs (London, England: 2000), 8, 921–926. Lucassen, P. L., Assendelft, W. J., van Eijk, J. T., Gubbels, J. W., Douwes, A. C. & van Geldrop, W. J. (2001) Systematic review of the occurrence of infantile colic in the community. Archives of Disease in Childhood, 84, 398–403. Miller-Loncar, C., Bigsby, R., High, P., Wallach, M. & Lester, B. (2004) Infant colic and feeding difficulties. Archives of Disease in Childhood, 89, 908–912. Morris, S., James-Roberts, I. S., Sleep, J. & Gillham, P. (2001) Economic evaluation of strategies for managing crying and sleeping problems. Archives of Disease in Childhood, 84, 15–19. Osatakul, S., Sriplung, H., Puetpaiboon, A., Junjana, C. O. & Chamnongpakdi, S. (2002) Prevalence and natural course of gastroesophageal reflux symptoms: a 1-year cohort study in Thai infants. Journal of Pediatric Gastroenterology and Nutrition, 34, 63–67. Savino, F. (2007) Focus on infantile colic. Acta Paediatrica, 96, 1259–1264. Savino, F. & Tarasco, V. (2010) New treatments for infant colic. Current Opinion in Pediatrics, 22, 791–797. Wessel, M. A., Cobb, J. C., Jackson, E. B., Harris, G. S. Jr & Detwiler, A. C. (1954) Paroxysmal fussing in infancy, sometimes called colic. Pediatrics, 14, 421–435.
