Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
December 2002

Sequelae of Infant ColicEvidence of Transient Infant Distress and Absence of Lasting Effects on Maternal Mental Health

Author Affiliations

From the Departments of Epidemiology and Biostatistics (Drs Clifford, Campbell, and Speechley), Pediatrics (Drs Campbell, Speechley, and Gorodzinsky), and Obstetrics and Gynecology (Dr Campbell); the Child Health Research Institute (Drs Campbell, Speechley, and Gorodzinsky), and the Lawson Health Research Institute (Drs Campbell, Speechley, and Gorodzinsky), University of Western Ontario, London. Dr Clifford is now affiliated with the Chalmers Research Group at the Children's Hospital of Eastern Ontario Research Institute, Ottawa.


Copyright 2002 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2002

Arch Pediatr Adolesc Med. 2002;156(12):1183-1188. doi:10.1001/archpedi.156.12.1183

Background  Colic is widely believed to remit by 3 months of age, with little lasting effect on the infant or the family.

Objectives  To determine the prevalence of colic at 3 months and the proportion of cases of colic (identified at 6 weeks) that remitted by 3 months; to identify the factors predictive of colic's remission; and to explore the potential lasting effects of colic on maternal mental health.

Design  Prospective cohort study of 856 mother-infant dyads. Self-administered questionnaires were mailed to mothers at 1 and 6 weeks and 3 and 6 months post partum. Standardized instruments were incorporated into the first and last questionnaires to assess maternal anxiety, postnatal depression, and social support. At 6 weeks and at 3 months, mothers completed the Barr diary and/or the Ames Cry Score.

Results  Data from 547 dyads were available for analysis. The prevalence of colic at 3 months was 6.4%. More than 85% of cases of colic had remitted by 3 months of age. These infants were more likely to be female, whereas the mothers of these infants were more likely to have received pain relief during labor/delivery and to have been employed during pregnancy. Reductions in scores for trait anxiety and postnatal depression, although smaller for mothers whose infants were colicky at 6 weeks of age, were not significantly different from those of mothers whose infants were never colicky.

Conclusion  This study provides support for the belief that, in most cases, colic is self-limiting and does not result in lasting effects to maternal mental health.

DESPITE UNCERTAINTY as to its etiology, the prevalence of colic has been estimated to range from 5% to 28% when assessed in prospective studies of relatively high quality.1 Colic is characterized by excessive and inconsolable crying, hypertonicity, and wakefulness that cluster in the evening hours. The onset usually occurs during the second to the sixth weeks of life. The most widely cited criterion for colic, known as the "rule of threes," defines a fussy infant as one who, "otherwise healthy and well fed, [has] paroxysms of irritability, fussing or crying lasting for a total of 3 hours per day and occurring more than 3 days in any 1 week."2 The term 3-month colic was coined by Illingworth3 after noting that 85% of cases of colic remitted by 3 months of age.

Although colic is believed to be self-limiting, the absence of a fail-safe cure4 can translate into a number of weeks that can be overwhelming for parents.5,6 Concomitant changes to the parent-child relationship7,8 may result in long-term deficits to the infant's mental and social development,9,10 whereas an episode of colic may adversely affect family life, even after it has resolved.1117

The potential sequelae of colic underline the importance of continued research. Although previous works have examined the natural history of colic and its effects on the family, methodological issues such as nonstandard or inconsistent definitions of colic6,11,14,17 and the use of clinically referred populations6,12,16 and retrospective assessments7,9,11,12,17 have restricted the interpretation of study findings. Thus, prospective methods are needed to determine whether colic is truly self-limiting and benign.

Building on an earlier examination of colic,18 this work used prospective methods in a population-based sample to determine the prevalence of colic at 3 months of age and the proportion of cases of colic (identified at 6 weeks of age) that remitted by 3 months and to identify factors predictive of colic's remission. In addition, we explored the potential for an episode of colic to result in lasting maternal distress by examining levels of maternal trait anxiety and depression at 6 months post partum, after controlling for earlier (ie, precolic) levels of maternal distress.


Approval for this study was granted by the Review Board for Health Sciences Research Involving Human Subjects of the University of Western Ontario, London. The prospective cohort included 856 mothers who were identified after giving birth in London. Eligible mothers were approached between January 15 and September 16, 1999, before their discharge from the hospital. Those who wished to participate provided written consent and an address to which surveys could be mailed. A detailed description of the study design is presented elsewhere.18

Briefly, baseline data were collected from eligible consenting women by means of a self-administered questionnaire that was mailed to their homes approximately 1 week after their discharge from the postpartum ward. This baseline questionnaire requested information on the current source of infant nutrition, prenatal and postnatal health behaviors, and demographic and biological factors. Standardized instruments assessing maternal anxiety (the State-Trait Anxiety Inventory),19 postnatal depression (the Edinburgh Postnatal Depression Scale),20 and social support (the Social Behaviours Inventory)21 were incorporated into this questionnaire. At 6 weeks and 3 months post partum, mothers were asked to complete a shortened version of the postpartum questionnaire (ie, identical to the original postpartum questionnaire except the standardized measures of anxiety, depression, and social support were excluded) and, by completing the Barr Baby's Day Diary22 and/or the Ames Cry Score,15,23 to provide details regarding their infants' crying behaviors. The strengths and limitations of each cry/fuss instrument are discussed elsewhere (T.J.C., M.K.C., K.N.S., and F.G., unpublished data, March 2000).

The final stage of data collection, at 6 months post partum, asked mothers to complete a questionnaire that included the standardized measures of anxiety19 and depression.20 All data collection was performed by mail. Reminder postcards were sent to all participants, and follow-up mailings were used, if necessary, as per the Total Design Survey Method of Dillman.24

Data were analyzed by means of the statistical program SPSS 9.0 for Windows (SPSS Inc, Chicago, Ill). All variables were categorical, with the exceptions of labor support, social support, maternal anxiety, and postnatal depression, which were measured on continuous scales. The prevalence of colic at 3 months was estimated on the basis of the application of a modified rule of threes to data recorded in the Barr diary. For this study, a colicky infant was defined as one who was "otherwise healthy and well fed but who had paroxysms of irritability, fussing, or crying lasting for 3 or more hours in any 1 day and occurring on 3 or more days in any 1 week." If a diary was not completed, analysis of responses on the Ames Cry Score, whereby a total cry score of 3 or greater was interpreted to indicate colic, provided estimates of colic's prevalence. Similar approaches were used to determine the proportions of cases of colic that remitted by 3 months, persistent cases of colic, and infants in whom latent distress developed (ie, infants who were not colicky at 6 weeks but whose behaviors rendered them colicky at 3 months of age).

To identify predictors of colic's remission, analyses were restricted to those with colic at 6 weeks of age. For univariable analyses, we used the 2-sample t test, Mantel-Haenszel χ2 test, and Fisher exact test, where appropriate, to analyze group differences. A multivariable model predictive of colic's remission or persistence at 3 months of age could not be developed because a limited number of cases of colic persisted that long.

To determine whether changes in maternal trait anxiety and depression scores were associated with the presence or absence of colic at 6 weeks of age, a 1-way analysis of variance was performed. Change scores, calculated separately for trait anxiety and depression, were derived by subtracting scores reported on the first questionnaire (1 week) from those reported on the final questionnaire (6 months).


A total of 1019 eligible mothers were approached from January 15 through September 16, 1999. Consent was received from 856 mothers (84.0%). Response rates were above 80% at each stage of data collection (Figure 1), with 65.4% of those who consented to participate returning a completed study package at 6 months post partum.

Figure 1.
Image not available

Study hierarchy. Q indicates questionnaire.

The 617 participants who completed the second (6-week) questionnaire were eligible to participate at the third (3-month) stage of data collection. Of these, 588 (95.3%) were married or living with a common-law partner; 517 (83.8%) delivered vaginally; and 511 (82.8%) were aged 20 to 34 years. The 547 mothers who participated at 3 months post partum were more likely to be married, to have completed more formal education, and to have higher household incomes than those who did not (Table 1). Participants were also more likely to be nonsmokers, to reside in a nonsmoking household, to breastfeed, and to have higher levels of social support and lower levels of anxiety and depression (as assessed at 1 week post partum) than those who did not complete the 3-month questionnaire.

Table 1. 
Image not available
Characteristics of Those Eligible for Participation at 3 Months*

Of the 547 mother-infant dyads who participated at 3 months, 35 infants (6.4%) had colic at 3 months. When examined in closer detail, 131 had colic at 6 weeks. Of these, 18 (13.7%) still had colic at 3 months. This means that 113 (86.3%) of the 131 cases of colic that had been identified at 6 weeks of age had remitted by 3 months. However, 17 infants had crying behaviors at 3 months of age that fulfilled our modified rule of threes and thus, were classified as colicky, although they had not been identified as colicky at 6 weeks of age. In other words, of the 35 cases of colic identified at 3 months of age, an almost equitable split was found between persistent (n = 18) and latent (n = 17) cases.

The cry/fuss distribution (Figure 2) was derived from detailed examination of the 320 diaries that were returned. From this analysis, the mean amount of crying/fussing recorded during an infant's 12th week of life was less than 6.7 hours (SD, 4.7 hours). Although this is an average of less than 1 hr/d, the cry/fuss behavior of some infants was truly excessive, with 1 mother recording that her infant cried/fussed for 21 hours during that week.

Figure 2.
Image not available

Cry/fuss distribution at 3 months of age for 320 infants. Minutes per week represent total minutes of crying and fussing during the 12th week of life (mean, 280.6 minutes; SD, 400.5 minutes).

Table 2 presents factors identified, in univariable analyses, to be significantly associated with the remission/persistence of colic at 3 months of age. Infants with colic that remitted by 3 months of age were more likely to be female, whereas the mothers of these infants were more likely to have received obstetric anesthesia or analgesia during labor and/or delivery, to have been employed outside the home during pregnancy, and to have consumed alcohol or caffeine in the postpartum period. Although household income also reached statistical significance in univariable analyses, interpretation of this association is complicated by small-cell counts in certain income categories. Since only 18 cases of colic persisted to 3 months of age, it was not possible to develop a multivariable model in which the dependent variable was defined according to the persistence or remission of colic at 3 months.

Table 2. 
Image not available
Factors Associated With Remission of Colic at 3 Months*

Results of the 1-way analysis of variance examining changes in maternal trait anxiety and depression are presented in Table 3. Between the first (1 week post partum) and final (6 months post partum) stages of data collection, scores fell by an average of 2.4 and 1.7 points for trait anxiety and depression, respectively. Reductions in scores for trait anxiety and postnatal depression were smaller for mothers whose infants were colicky at 6 weeks of age than for those whose infants were never colicky, but the difference was not statistically significant.

Table 3. 
Image not available
Change in Maternal Trait Anxiety and Depression Scores, Relative to Infant's Colic Status*

This study used prospective methods in a relatively large population-based sample to provide evidence that most cases of colic remit by 3 months of age with little lasting impact on levels of maternal anxiety or depression. We ascertained the outcome of interest, colic, objectively by applying a modified rule of threes to data collected by means of instruments specifically designed for this purpose. Unlike earlier studies, potential etiologic agents, including the source of infant nutrition and levels of maternal distress, were assessed before the development of colic, allowing for the exploration of temporal relationships. Moreover, we examined the potential for an episode of colic to adversely affect maternal mental health after controlling for earlier (ie, precolic) levels of anxiety and depression.

This study's finding that more than 85% of colic cases remit by 3 months of age is in keeping with the observations of others2,2529 and should provide some relief to parents who find themselves caring for a colicky infant. Additional relief can be derived from the fact that, by the twelfth week of life, the total duration of crying and fussing had fallen to an average of 1 h/d, which is about half that seen at 6 weeks of age. This finding is in concordance with the patterns depicted by the normal crying curve, which peaks around 6 weeks of age and decreases to a baseline level around the third month of life.2,26,2937 That the normal crying curve has been found to be robust across time and place and to persist despite optimal caregiving practices38,39 suggests that early crying patterns reflect a process of normal maturation.33,40

Excessive crying persisted beyond 3 months of age in almost 14% of cases. This phenomenon, too, appears to be robust across time, place, and study design.2,32,35,36,39,4144 This observation suggests that a number of meaningful subgroups may exist within the constellation of colic complaints44: one subgroup exhibits typical 3-month colic, whereas in another, colic persists beyond 3 months, perhaps stemming from a persistent mother-infant distress syndrome.16 As in the study by Wurmser et al,43 we found no significant relationships between persistent cases of colic and birth order, source of infant nutrition, parental education, or maternal age. A number of factors, including the infant's sex, administration of obstetric anesthesia or analgesia during labor and/or delivery, maternal employment status during pregnancy, household income, and postpartum maternal health behaviors, were identified in univariable analyses to differentiate between colic that remitted and colic that persisted. Because of the potential development of persistent mother-infant distress syndrome,16 mother-infant dyads who fall into this latter category deserve special attention by future research.

A third subgroup of infants experienced latent distress. These infants did not demonstrate colicky behavior at 6 weeks of age, but did at 3 months of age. Therefore, future studies should look beyond rigid definitions of colic to provide parents with an accurate description of age-appropriate crying behaviors.

Although previous studies that relied on subjective assessments found mothers of formerly colicky infants to be more bothered by their child's behavior,12,16,17 no such differences were seen in studies that used objective measures.11,12,16 Along with the findings of this study, those findings suggest that once colic resolves, the residual effects on levels of maternal distress are negligible. Perhaps, once colic resolves, parents then turn their attention to the positive aspects of their child's development. This finding should provide additional reassurance to parents with indications that, for most infants, colic is self-limiting and benign.

Limitations of this study include the possibility that the experiences of this study population may not generalize to postpartum populations with different access to resources that may ease the burden of caring for a fussy infant. Mothers were not asked about their use of interventions and/or health care services in response to their infant's crying or if they were able to seek respite when they felt overwhelmed. The observation that most mothers in this study were married and financially secure, along with the relative universality of health care in Canada, suggests that the families in this study would have adequate resources to buffer the effects of their infants' colic.6,15 Moreover, the attrition seen in this study's 4 stages of data collection was not uniform; ie, the manner in which participants differed from nonparticipants translated to a study population that was not necessarily representative of the entire postpartum population. Although such differential participation in postal surveys of this population is not unique,45,46 the absence of noted effects on levels of maternal anxiety and depression may be due to the relative uniformity of the study sample. Replication of this study in populations that are more heterogeneous could address this issue. In addition, this study did not assess the possibility that parental perceptions of their infants (ie, his or her vulnerability and/or temperament) were altered by the colicky behavior. This will be addressed in a follow-up study of this cohort.


This study represents an important contribution to furthering society's understanding of infant colic. Parents can be reassured that, for most infants, colic is confined to the first 3 months of life, with little lasting effect on levels of maternal distress. These observations should not justify the abandonment of research. Rather, the absence of definitive conclusions regarding colic's etiology, along with its relatively frequent occurrence, the feelings of despair it can evoke, and, in a few cases, its persistence, underlines the importance of continued research that examines the many subgroups of excessive infant crying. In the meantime, we should provide parents with education and support that enable them to cope during this potentially difficult time.

Back to top
Article Information

Accepted for publication June 25, 2002.

Corresponding author and reprints: Tammy J. Clifford, PhD, Children's Hospital of Eastern Ontario Research Institute, 401 Smyth Rd, Ottawa, Ontario, Canada K1H 8L1 (e-mail:

What This Study Adds

Colic is a relatively common condition of early infancy, characterized by excessive and inconsolable crying. Although the absence of a fail-safe cure for colic can translate into a number of weeks that can be overwhelming for parents, colic is widely believed to be self-limiting and benign. Methodological shortcomings of existing studies, however, have limited our ability to make definitive conclusions.

This study used prospective methods in a community sample of infants to provide evidence that, for most infants, colic remits by 3 months of age with no evidence of deleterious lasting effects on levels of maternal anxiety or depression, as measured by standardized assessments. Because this population has universal access to health care and because most mothers were married, well-educated, and financially secure, future research may wish to examine these findings in populations that differ in their socioeconomic profile and/or access to care.

Lucassen  PLBAssendelft  WJJvan Eijk  JTMGubbels  JWDouwes  ACvan Geldrop  WJ Systematic review of the occurrence of infantile colic in the community. Arch Dis Child. 2001;84398- 403Article
Wessel  MACobb  JCJackson  EBHarris  GSDetwiler  AC Paroxysmal fussing in infancy, sometimes called "colic." Pediatrics. 1954;14421- 433
Illingworth  RS Three month's colic. Arch Dis Child. 1954;29165- 174Article
Garrison  MMChristakis  DA A systematic review of treatments for infant colic. Pediatrics. 2000;106184- 190
Wilkie  CFAmes  EW The relationship of infant crying to parental stress in the transition to parenthood. J Marriage Fam. 1986;48545- 550Article
Thompson  PEHarris  CCBitowski  BE Effects of infant colic on the family: implications for practice. Issues Compr Pediatr Nurs. 1986;9273- 285Article
Bell  SMAinsworth  MDS Infant crying and maternal responsiveness. Child Dev. 1972;431171- 1190Article
Campbell  SB Mother-infant interaction as a function of maternal ratings of temperament. Child Psychiatry Hum Dev. 1979;1067- 76Article
Sloman  JBellinger  DCKrentzcel  CP Infantile colic and transient developmental lag in the first year of life. Child Psychiatry Hum Dev. 1990;2125- 35Article
Stifter  CABono  MA The effect of infant colic on maternal self-perceptions and mother-infant attachment. Child Care Health Dev. 1998;24339- 351Article
Forsyth  BWCCanny  PF Perception of vulnerability 3½ years after problems of feeding and crying behavior in early infancy. Pediatrics. 1991;88757- 763
Jacobson  DMelvin  N A comparison of temperament and maternal bother in infants with and without colic. J Pediatr Nurs. 1995;10181- 187Article
Rautava  PLehtonen  LHelenius  HSillanpaa  M Infantile colic: child and family three years later. Pediatrics. 1995;9643- 47
Raiha  HLehtonen  LKorhonen  TKorvenranta  H Family life 1 year after infantile colic. Arch Pediatr Adolesc Med. 1996;1501032- 1036Article
Elliott  MRPedersen  ELMorgan  J Early infant crying: child and family follow-up at three years. Can J Nurs Res. 1997;2947- 67
Papousek  Mvon Hofacker  N Persistent crying in early infancy: a non-trivial condition of risk for the developing mother-infant relationship. Child Care Health Dev. 1998;24395- 424Article
Long  TJohnson  M Living and coping with excessive infantile crying. J Adv Nurs. 2001;34155- 162Article
Clifford  TJCampbell  MKSpeechley  KNGorodzinsky  F Infant colic: empirical evidence of the absence of an association with source of early infant nutrition. Arch Pediatr Adolesc Med. 2002;1561123- 1128Article
Spielberger  CDGorsuch  RLLuschene  R The State-Trait Anxiety Inventory.  Palo Alto, Calif Consulting Psychologists Press1970;
Cox  JLHolden  JMSagovsky  R Detection of postnatal depression: development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry. 1987;150782- 786Article
Brown  MA Social support during pregnancy: a unidimensional or multidimensional construct. Nurs Res. 1986;354- 9Article
Barr  RGKramer  MSBoisjoly  CMcVey-White  LPless  IB Parental diary of infant cry and fuss behaviour. Arch Dis Child. 1988;63380- 387Article
Ames  EWBradley  C Infant and parent characteristics related to parents' reports of colic in one-month-old and three-month-old infants.  Paper presented at Annual Meeting of the Canadian Psychological Association June 10, 1983 Winnipeg, Manitoba
Dillman  D Mail and Telephone Surveys: The Total Design Method.  New York, NY John Wiley & Sons Inc1978;
Brazleton  TB Crying in infancy. Pediatrics. 1962;29579- 588
Paradise  JL Maternal and other factors in the etiology of colic. JAMA. 1966;197123- 131Article
Stahlberg  M Infantile colic: occurrence and risk factors. Eur J Pediatr. 1984;143108- 111Article
Michelsson  KPaajanen  SRinne  ATervo  HKunnamo  IKiminkinen  T Mothers' perceptions of and feelings towards their babies' crying. Early Child Dev Care. 1990;65109- 116Article
Rebelsky  FBlack  R Crying in infancy. J Genet Psychol. 1972;12149- 57Article
Bernal  JF Night waking in infants during the first 14 months. Dev Med Child Neurol. 1973;15760- 769Article
Hunziker  UABarr  RG Increased carrying reduces infant crying: a randomized controlled trial. Pediatrics. 1986;77641- 648
Barr  RG The normal crying curve: what do we really know? Dev Med Child Neurol. 1990;32356- 362Article
Barr  RGRotman  AYaremko  JLeduc  DFrancoeur  TE The crying of infants with colic: a controlled empirical description. Pediatrics. 1992;9014- 21
St James-Roberts  IHalil  T Infant crying patterns in the first year: normal community and clinical findings. J Child Psychol Psychiatry. 1991;32951- 968Article
St James-Roberts  IPlewis  I Individual differences, daily fluctuations and developmental changes in amounts of infant waking, fussing, crying, feeding and sleeping. Child Dev. 1996;672527- 2540Article
Alvarez  MSt James-Roberts  I Infant fussing and crying patterns in the first year in an urban community in Denmark. Acta Paediatr. 1996;85463- 466Article
Barr  RGElias  MF Nursing interval and maternal responsivity: effect on early infant crying. Pediatrics. 1988;81529- 536
St James-Roberts  IConroy  SWilsher  K Links between maternal care and persistent infant crying in the early months. Child Care Health Dev. 1998;24353- 376Article
Lester  BMBoykydis  CJZGarcia-Coll  CTHole  WT Colic for developmentalists. Infant Ment Health J. 1990;11321- 331Article
Fisichelli  VFisichelli  RKarditz  SCooper  J The course of individual crying activity in the first year of life. Pediatr Res. 1974;8921- 928Article
Hide  DWGuyer  BM Prevalence of infant colic. Arch Dis Child. 1982;57559- 560Article
Rubin  SPPrendergast  M Infantile colic: incidence and treatment in a Norfolk community. Child Care Health Dev. 1984;10219- 226Article
Wurmser  HLaubereau  BHermann  MPapousek  Mvon Kries  R Excessive infant crying: often not confined to the first 3 months of age. Early Hum Dev. 2001;641- 6Article
Gormally  SBarr  RG Of clinical pies and clinical clues: proposal for a clinical approach to complaints of early crying and colic. Ambul Child Health. 1997;3137- 153
Shepherd  CKPower  KGCarter  H Characteristics of responders and non-responders in an infant feeding study. J Public Health Med. 1998;20275- 280Article
Larroque  BKaminski  MBouvier-Colle  MHHollebecque  V Participation in a mail survey: role of repeated mailings and characteristics of nonrespondents among recent mothers. Paediatr Perinat Epidemiol. 1999;13218- 233Article