Subjects and methods

We conducted a survey during the routine sociomedical examinations of infants aged 2–3 months in six out of 21 child health clinics belonging to the Municipal Health Service of Amsterdam. The combined population of these six child health clinics is representative of all infants in Amsterdam with respect to sex, ethnic origin, and maternal educational level.

DATA COLLECTION

The study was conducted between January 1995 and December 1996. During this period, 2180 infants aged 2–3 months and their mothers were invited to visit the child health clinics at the addresses registered in the municipal population register. All residents of the city of Amsterdam are registered in the municipal population register. The nurses and physicians from the collaborating child health clinics filled out a questionnaire on sociodemographic characteristics and the child’s crying history. Interpreters were used if the mother did not speak Dutch: for non-Dutch speaking Turkish and Moroccan mothers a female interpreter was present during the sociomedical examination; for other ethnic groups the health centres called in the help of an interpreter by telephone. All Surinamese residents speak Dutch. A second questionnaire on soothing techniques was given to the mother to be filled out at home and returned to the Municipal Health Service by prepaid mail. The questionnaire was available in Dutch, Dutch–Turkish, Dutch–Moroccan or Dutch–English, as appropriate. Although neither questionnaire included the name of the subject, they were linked using a unique serial number.

QUESTIONNAIRE

Data on infant crying were collected by means of a retrospective questionnaire. This method is fast, convenient, and suitable for large scale use in measuring mother’ perceptions of and responses to infant crying.4 Other methods often used to obtain information on the amount of infant crying are audiotape recordings and prospective diaries. Audiotape recordings are too expensive for use in large samples. Prospective diaries are unlikely to be completed in low socioeconomic groups and ethnic minorities.21Furthermore, the amount of crying reported by Western mothers in diaries and questionnaires is comparable.4 22

Mothers were asked to report the total amount of crying in hours/24 hour day, on average, in the previous week. Excessive crying was defined as crying for three or more hours/24 hour day. Mothers were also asked whether their baby cries a lot, and whether their baby is easy or difficult to soothe. The interpretation of infant crying was left to the mothers themselves. Sociodemographic data included age and sex of the child, birth order, ethnic origin, educational level of the mother, and type of infant feeding. Ethnic origin was defined according to the mother’s country of origin: the Netherlands, Surinam (former Dutch colony in South America), Morocco, Turkey, and other countries. The mother’s educational level was defined as poor (primary school or lower secondary school) or high (higher secondary school or post-secondary education), based on the highest educational level achieved.

The questionnaire on soothing techniques listed various responses to crying. The sensitive questions were hidden within a set of other, more neutral questions. Mothers were asked whether they had ever used each of the soothing techniques: for example, “Have you ever slapped your baby to make it stop crying?”.

STATISTICAL ANALYSES

We used the statistical package SPSS/PC to analyse the data.23 Chi-squared tests were used to examine univariate associations between crying behaviours and various sociodemographic variables. Next, logistic regression was used to assess the relative importance of each of the sociodemographic variables on the prevalence of problematic infant crying. Logistic regression was also used to determine the relative effect of sociodemographic variables and crying behaviours on the prevalence of various soothing techniques. A regression was carried out for each individual soothing technique. The results of these analyses are expressed as adjusted odds ratios. The level of significance used was p < 0.05.

Results

RESPONSE RATES

A questionnaire on sociodemographic characteristics and crying behaviours was filled out for 1826 infants (84%). No information is available for the remaining infants. For 1142 (63%) of the 1826 infants, a questionnaire about soothing techniques was returned as well. The mothers who did not return the questionnaire were more frequently of non-Dutch origin (p < 0.001) and more had a lower level of education (p < 0.001) and a first born child (p < 0.001). No differences in response rate were found for crying behaviour and other sociodemographic characteristics.

CRYING BEHAVIOUR

Table 1 presents the study sample according to sociodemographic characteristics and reported crying behaviour. Overall prevalences of “crying for three or more hours/24 hour day”, “crying a lot”, and “difficult to comfort” were 7.6%, 14.0%, and 10.3%, respectively (table 1). Problematic infant crying, according to at least one criterion, was reported by 20.3% of the mothers. Of these crying infants, only 14% met all three inclusion criteria. Sex, parity, ethnic origin, and type of feeding were all related to crying behaviour (table 2). Perceived crying for three or more hours a day occurred less frequently in girls compared with boys, Surinamese infants compared with Dutch infants, and breast fed infants compared with formula or mixed fed infants. The same held true for infants whose mothers stated that their infant cries a lot. Furthermore, primiparous mothers stated more frequently that their infant cries a lot compared with multiparous mothers. Surinamese mothers reported less frequently that their infant was difficult to comfort compared with Dutch mothers.

SOOTHING TECHNIQUES

Mothers used various soothing techniques to calm their babies. In addition to harmless methods, mothers also used techniques that could affect the infant’s health negatively (table 3). Most soothing methods were related to sociodemographic variables (table 4). Reported shaking, slapping, and putting the baby to sleep in a prone position were more common among non-Dutch mothers than among Dutch mothers. This held true especially for Turkish mothers (OR = 6.36 for shaking; OR = 6.37 for slapping; and OR = 3.93 for prone sleeping position). Poorly educated mothers also reported slapping their baby more often than highly educated mothers (OR = 4.08). Moroccan mothers rocked (0.56), held, and carried their baby less frequently than Dutch mothers in response to infant crying (OR = 0.24).

Discussion

This study shows that the overall prevalence of reported problematic crying in babies aged 2–3 months differs according to the definition used: 7.6% for crying for three or more hours/24 hour day, 14.0% for crying a lot, and 10.3% for inconsolable crying. Problematic infant crying according to one or more criteria was reported by 20.3% of the mothers. Many mothers used soothing techniques that could affect the infant’s health negatively. Mothers’ reports of infant crying and soothing varied sociodemographically. Excessive crying was less common among girls, breast fed infants, and Surinamese infants. Soothing methods that can be injurious to infants’ health, such as slapping, shaking, and prone sleeping position, were more prevalent among non-Dutch mothers (especially Turkish) than among Dutch mothers.

Theoretically, a selective non-response among non-Dutch families might bias overall prevalences of crying babies. A selective high non-response among non-Dutch families is a common problem in surveys carried out in the multicultural cities of the Netherlands. However, in this study the non-response rate for the verbal questionnaire was very low, probably as a result of the reputation of the child health clinics, as well as the experience of the nurses and physicians at these clinics in contacting and interviewing respondents. Apart from this, a previous survey in these six child health centres showed that attendance rates vary little according to ethnic group.24

A selective non-response was found for the mail questionnaire on soothing techniques. As a result, overall prevalences might be biased. Furthermore, the results might be biased by differences in socially desirable responses. For instance, it is possible that Dutch mothers slap and shake their children more often than they reported. It is also possible that “slapping” and “shaking” might have different meanings in different ethnic groups.

Of all problematic crying infants, only 14% met all three criteria. We find it remarkable that only 39% of the infants who cry a lot are also reported to cry for three or more hours/24 hour day. Conversely, of the infants who cry for three or more hours/24 hour day, 78% were also reported to cry a lot. Apparently, some parents cannot tolerate the least amount of crying, while others do not experience long crying periods as troublesome. Therefore, intervention efforts should be guided by parents’ perceptions of infant crying, irrespective of the actual amount of crying.

EXCESSIVE CRYING BY SOCIODEMOGRAPHIC CHARACTERISTICS

Different definitions of a crying infant lead to different conclusions about the sociodemographic factors associated with reported crying. For example, there was no association between parity and crying for three or more hours/day. This is in agreement with other studies.2 4 25-27 However, crying a lot was reported to be more common among infants of primiparous mothers. Perhaps primiparous mothers experience the same amount of crying as more troublesome. Furthermore, primiparous mothers were more likely to seek referral for infant crying than mothers of later born infants.2 4

Reported excessive crying appeared to be more common in boys than in girls. Other studies reported no sex differences in the prevalence of excessive crying1 28 29 or mean amount of crying.2 4 30 31 However, most of these studies were based on small selective groups and data were sampled for purposes other than establishing sex differences in the amount of infant crying. One previous study found modestly increased crying in boys.32

Reported excessive crying was more frequent among Dutch infants than among Surinamese infants. We found only two other studies in Western multicultural societies dealing with ethnic differences in excessive infant crying. In an Australian study, excessive crying was reported more commonly in infants with parents born in Lebanon, Asia, and South and Central America.33 A study in the USA among low birthweight infants showed that excessive crying was much less frequent among black infants than among white infants.27 This might reflect ethnic differences in genetic predisposition, care giving practice, or soothing techniques. Given that these studies were based on maternal reports, ethnic differences in maternal perceptions of crying might also be an explanation.

Poorly educated mothers did not differ from highly educated mothers in their reports of excessive crying. This is in accordance with other studies.25 28 30 In some studies, however, more infant crying was reported in higher socioeconomic groups10 29or lower socioeconomic groups.26 These inconsistent findings can be explained partly by different indicators of socioeconomic status in different studies. Socioeconomic health differences in children vary strongly depending on the indicator of socioeconomic status.34

Breast fed infants cried less than formula or mixed fed babies. This may not be related to the content of feeding. For instance, excessive infant crying is a common reason for mothers to switch from breastfeeding to bottle feeding. Furthermore, bottle fed infants might cry excessively because the child health clinic prescribes an insufficient amount of formula feeding,35 whereas breast fed babies can control the amount they feed. A Dutch study has shown that breast fed infants feed more frequently and/or were more frequently fed on demand than those who were bottle fed.36 These two factors are associated with a decrease in infant crying.16 37

SOOTHING TECHNIQUES

There is evidence that picking up and carrying,31 38 39 rocking,40 auditory stimula-tion,41 42 pacifiers,43 44and herbal tea45 are effective soothers. Whether these techniques reduce crying in the long run is not known. In addition, some soothers that have been shown to be highly effective in experimental research are not perceived as such by mothers.46

Some soothing methods must be strongly discouraged, because of their possible injurious effect on the child’s health. The most dangerous techniques a parent can use are to smother, slap, or shake the baby to make it quiet. Nonetheless, these techniques were reported by 2%, 3%, and 5% of mothers, respectively. Among mothers of infants with problematic crying, these prevalences were two to three times higher. Excessive crying has been identified as a major cause of child abuse,47 which could be fatal in infants, because of their inability to escape harm and their physical vulnerability to serious trauma.48

Some studies suggest that ignoring a crying baby is effective in reducing the amount of crying.49 50 Ignoring the infant’s crying or leaving the infant alone when crying is a common response in Western societies.31 51 In our study, 45% of mothers let the baby cry on at least one occasion. It is unclear how this behaviour affects the parent–child relationship. In a study dealing with the circumstances of 222 sudden unexpected deaths, five infants died in the night when the parents had let it continue to cry for the first time.52

Crying in breast fed infants can be a symptom of allergy to cow’s milk products consumed by the mother.53 It is better to eliminate these products from the mother’s diet than to stop breastfeeding,54 which protects children from various diseases, especially infectious ones.55 Moreover, infants who had been switched to formula feeding were no more quiet than infants who were exclusively breast fed.36

Swaddling (wrapping the baby in a blanket) appears to be soothing,56 but some forms of wrapping increase the risk of a congenital dislocation of the hip.57

Laying infants prone to sleep increases the risk of cot death,58 so it is not acceptable to place the baby in this position to stop it crying. In a series of 108 sudden unexpected deaths, six infants died during their sleep when they were put to sleep in a prone position for the first time.59 In a series of 222 other sudden unexpected deaths this was the case with eight infants.52

Some mothers assume that infant crying is caused by constipation, but the lack of evidence means laxatives should be discouraged. Hypnotics and sedatives should not be used because of a possible relation to brain damage and cot death.60

SOOTHING TECHNIQUES BY SOCIODEMOGRAPHIC CHARACTERISTICS

Primiparous mothers were shown to admonish their baby more often in response to crying than multiparous mothers. Similarly, Boukydis and Burgess61 noted that first time parents respond to crying with greater arousal and more irritation than multiparous parents.

It is notable that Turkish and Moroccan mothers relatively frequently admonished, slapped, and shook their infant in response to crying. Perhaps punishment is a common child rearing practice with young children in those cultural groups. In a study of bed wetting in Amsterdam, relatively more Dutch parents praised their toddlers for a dry night, while relatively more immigrant parents punished their toddlers for a wet night.62

Many Turkish mothers placed their baby to sleep in the prone position in response to crying. This is probably one reason why, in the Netherlands, Turkish infants are almost twice as likely to suffer cot death than Dutch infants.63 It is possible that the more frequent use of slapping and shaking among Turkish mothers also contributes to this higher risk.64

IMPLICATIONS

Much harm may be prevented by counselling parents (especially immigrants) on how and how not to respond to infant crying. It is important to identify inappropriate parental responses and to suggest alternatives. Furthermore, it is practical to ask parents whether they detect differences in the effectiveness of their soothing methods, because effective soothing strategies differ for each child.65 Health education should start before the child’s birth, because some soothing techniques can be fatal, even when practised for the first time, such as the prone sleeping position, allowing the baby to continue to cry, smothering, and shaking.