Printer Friendly

Weaning: what influences the timing?

Introduction and background

Appropriate nutrition in the early years is essential for optimal growth and development. Infants are dependent initially on milk, with breastmilk the best choice. (1,2) Weaning is a gradual process through which milk is replaced with solid food. (3) The aim is to introduce food that ensures adequate nutrition for growth and development, and that is safe. The risks of weaning too early (putting stress on an immature gastro-intestinal tract, kidney and immune systems) need to be balanced with the risks of weaning too late (undernutrition and feeding problems). (5,6) Although World Health Organization guidelines, adopted by the UK and Scottish governments, suggest delaying the introduction of solid food until the infant reaches six months of age, there is evidence that compliance is currently limited. (4,7)

In general, it is mothers who decide how, when and what to feed their infants. (8) This choice is dependent on a complex interaction of social and psychological factors. (9) However, demographic characteristics influence all aspects of infant feeding significantly. (7) Mothers who are younger, lone parents, with lower educational attainment or lower family income have been found consistently more likely to wean early. (3,6,10,11) Living in socially deprived circumstances increases the tendency to wean early. (1,2)

Knowledge and understanding of healthy eating recommendations influences an individual mother's feeding decision. (1) Providing nutritional information in a variety of settings has been a longstanding role for health visitors. (9) However, the recent policy shift requiring health visitors to concentrate on vulnerable families has meant that the onus is often on parents to access appropriate support and advice. (12) Informal networks of friends and family are considered to be important sources. (13) Unfortunately, reliance on such advice may perpetuate weaning traditions that are now considered to be inappropriate. (1)

Despite recommendations to commence weaning at six months, earlier introduction of solids is widespread. (6) While there is limited evidence that weaning between the age of four and six months in developed countries is harmful, (14) introducing solids before the age of four months is recognised universally as inappropriate, (15) and linked with increased risk of childhood obesity. (16) With mothers living in socially deprived circumstances particularly likely to wean early, health inequalities linked to obesity are likely to be perpetuated. Understanding the factors that contribute to the decision-making process to start weaning is essential if effective interventions are to be designed. (17)

Study aim

This study investigated parental decisions to introduce solid food to their infants. It aimed to explore the effect of demographic variables and influence of external sources of information on the timing of weaning.

It was conducted as part of an MSc in public health, undertaken while employed as a health visitor in North Ayrshire.


This locally-based study used a cross-sectional survey design. This is able to capture standardised information about a particular phenomenon, its frequency and natural variation in a defined population at a given time. (18)

Sample and recruitment

Families with infants born from 1 March to 30 June 2006 and registered with one of eight GP practices were invited to take part. Each practice was located within one community health partnership (CHP) in North Ayrshire, where 24% of the population live in the 15% most deprived areas of Scotland. (19) With funding devolved from NHS boards, CHPs are responsible for a range of community health services.

At the time of the survey (January to March 2007), each infant was aged between seven and 10 months. It was expected that weaning would have commenced by seven months, and a maximum of 10 months was chosen to minimise difficulties in recall.

Participating practices were asked to identify families with infants born within the four-month period (n=203). A pack containing an invitation letter, information sheet, questionnaire and stamped addressed envelope was sent by the practice to each of these families. They were contacted on two occasions, three weeks apart, with a copy of the questionnaire.

Data collection

Data was collected by a self-administered postal questionnaire. An extensive literature search failed to identify a suitable validated survey. Findings from a review of literature and the author's experiential knowledge were therefore used to design a 22-item questionnaire. Previous studies had highlighted demographic characteristics that were likely to influence feeding practices. (6,10,11) Data were thus collected about the main carer's age, parity, income, marital status, educational level and the infant's gender. Respondents were asked what age they had given their baby any food apart from milk. Closed questions were also asked about what prompted the parent to introduce solids and what sources of information about feeding had been accessed. For some questions, respondents were able to give more than one answer, and there was an 'other (please state)' option to accommodate unanticipated responses.

Time constraints meant that it was not possible to pilot the questionnaire, so a draft was distributed to six health visiting colleagues to assess content validity. Comments were invited about the appropriateness, ease of understanding and layout of both questions and answer options, and it was altered in response to these. The questionnaire was considered easy to read when checked by the readability statistical tool in Microsoft Word 2002. Different coloured paper was used for the two distributions of the questionnaire, so responses to each could be distinguished.

Data analysis

Returned questionnaires were numbered on receipt and coded manually according to a pre-designed coding frame. These codes were entered into the Statistical Package for Social Sciences (SPSS) for Windows (version 14.0). Infants' dates of birth and respondents' demographic details were compared to enable the identification of duplicate returns. Not all returned questionnaires were complete, so some data were missing. For most, less than 4% of variables were omitted. However, for the question about the age of the infant when weaning advice was received from a healthcare professional, only 85 (75%) questionnaires were complete.

Initially, to gain an overall picture of the ages at which parents were introducing solids to their infants and what had influenced that decision, descriptive statistics were performed. These were examined by means of frequency tables and measures of central tendency and dispersion. Based on previous research, it was anticipated that the distribution of the 'age of weaning' variable were likely to be non-normal, so non-parametric techniques of data analysis were chosen and performed. (21) The age of weaning was then converted into a categorical variable to allow its relationship with other variables to be examined. In this study, infants who were introduced to solids before the age of four months were considered to have been weaned early. Adopting this as the primary outcome measure took likely significant discrepancies between feeding guidelines and actual practice into consideration. (15) Data were categorised into three groups--before four months, between four and six months, and at six months. Pearson's chi-square test of independence was performed to determine whether there was any evidence of an association between the age of weaning and demographic variables. The reason for weaning and the influence of sources of information was explored in the same way. The relationship between the age of weaning and the timing of any professional advice was examined by the Wilcoxon signed-rank test. Associations were considered to be significant at a level of p [less than or equal to] 0.05. (21)


Ethical considerations

Agreement to take part was presumed by the return of the completed questionnaire. Family health visitors attached to participating practices were invited to withdraw families who they considered inappropriate to approach, but none did. The personal details of recipients were unknown to the researcher, so it was possible to assure participants of confidentiality and anonymity. The study was approved by the local research ethics committee and NHS board research and development department.


In all, 56% (114) of questionnaires were returned. Apart from one, all respondents were mothers of the infant. The carers were aged between 17 and 41 years, and 15% (n=17) were lone parents. More than half (54%, n=61) were first-time parents. Measured by the Scottish Index of Multiple Deprivation, 46% (n=51) lived in one of the 15% most deprived areas in Scotland. Also, 43% (n=45) of respondents were considered to be on a low income. Annual family income was estimated from reported receipt of tax credits. At the time of the study, parents with an annual income less than 15,000 [pounds sterling] were eligible for working tax credit or Healthy Start vouchers. Those with an income over 60,000 [pounds sterling] were unable to claim tax credits.

When solids were introduced

All respondents reported having introduced solids to their infant by the time of the survey. Although timing varied from four to 26 weeks (median=16, interquartile range=14 to 21), there was a marked increase in frequency of weaning at three and four months (see Figure 1).

By four months, more than half (57%, n=63) had been introduced to solids. Of these, 40% (n=25) had commenced weaning by 13 weeks. All had been weaned by six months. Early weaning was found to be significantly associated with the carer's annual family income, whether she was a lone parent or had never breastfed her baby (see Table 1). Although respondents who reported lower qualifications or were younger tended to have weaned their baby earlier, this was not statistically significant. First-time parenthood and the baby's gender also did not significantly affect the timing of weaning.

Why infants were weaned

The most common reported reasons for weaning infants were internal influences, including the impression that the infant was ready (91%, n=103) and that it would help the infant sleep better (40%, n=45). The most common external reported influences were friends and family (34.5%, n=39) and health visitors (34.5%, n=39).

The influence of having received or sought a form of formal weaning advice was explored. The majority of respondents (89%, n=101) reported having been recipients of verbal advice from a health visitor at either home or clinic, or being given written information. Even though no significant relationship was found between the receipt of advice and timing of weaning, when the advice was received was significantly associated with when infants were weaned (p=0.005).

Of respondents who had received weaning advice when their infant was aged between three and four months, more than half (n=13) had introduced solids at that time (see Figure 2).



There has been increased recent interest in the contribution of early nutrition to the optimal growth and development of infants and young children, though the emphasis has been on breastfeeding initiation and duration. Given that inappropriate feeding practices in infancy may have important effects on long-term health, weaning is beginning to be recognised as an important dietary event. (15) However, data about the age of weaning are not collected routinely. This study's findings support other research that has found early weaning to be widespread. The pattern found in this survey is broadly in line with the Scottish data from the seventh Infant feeding survey. (4) Overall, while the reported age of weaning has become progressively higher over the last decade, high proportions of parents continue to wean earlier than recommended. However, a higher proportion of parents in this study had weaned their infant by the age of three months (see Table 2). This may reflect the relatively high levels of social deprivation in the locality.

Early weaning was found to be socially patterned. Consistent with previous research, younger mothers, lone parents, those with lower educational attainment, or lower family income were found to be more likely to wean early. (6,7) The finding that early weaning is more prevalent in families living in deprived circumstances supports the focus of policy on these families.

That the predominant reported influence on mothers' decisions to wean was the perception of their child's readiness is complex to interpret. Perceptions of readiness may be a function of the infant reaching a certain age, weight or size, in conjunction with parental opinion that their baby was hungry and no longer satisfied with milk. (6,22,23) It may reflect feelings that settling the baby is synonymous with being a good parent. (22) The change in feeding practices at weaning may also be seen as moving onto the next developmental stage, and as such, symbolic of an infant's general progress.

Perceptions of readiness were influenced by external sources. Carers listened to advice and information from family and friends as well as healthcare professionals. (13) The prevalence of early weaning may be influenced by a belief that the guidelines are wrong when considered in conjunction with previous generations being weaned earlier without any apparent health consequences. (22) In addition, manufacturers of commercial baby food label their products as being suitable from four months of age.

As healthcare professionals with considerable contact with families with infants, health visitors have the potential to be key educators about infant nutrition. (13,23) It is widely assumed that anticipatory advice given early and consistently has the potential to influence parental feeding behaviours. (5) Conversely, this study suggests that early advice might have the unintended effect of encouraging early weaning.

Although local policy promoted delaying weaning until six months, the advice offered by individual health visitors might have differed from this. The questionnaire also failed to distinguish between advice delivered as part of the health promotion programme and that which was sought by the parent. While mothers might have interpreted receiving weaning advice from the health visitor as an indication that it was time to start weaning, it is also feasible that mothers sought health visitor advice once they were considering to begin weaning.

Parents of infants born after 1 October 2006 are offered a health promotion programme that includes discussion of weaning when the infant reaches three months of age. (12) Further research into the effect of the timing of health visitor advice would be valuable to clarify the nature of its relationship with the timing of weaning.

This research was a small locally-based study, and generalising the findings is problematic. The sample was not selected at random--details of all pre-school children are kept on the local child health surveillance database, but access to this was denied due to data protection protocol. Time limitations meant that recruitment was restricted to parents with infants born over a four-month period, which may have introduced a seasonal bias. Also, while the response rate was reasonable, respondents may have differed from non-respondents. Compared to mothers who registered a birth in 2006 in North Ayrshire, those in the survey sample were older, less likely to be single or were educated to a higher level. (24) Taking the social patterning of early weaning into account, the proportion of infants being weaned early may have been greater than was found.

As a survey, the reliability of this study depended on the accuracy of respondents' answers. (18) Data were collected retrospectively, so relied on respondents' accurate recall, though the relatively short time between weaning and the survey was intended to minimise this potential bias. Participants' awareness of the researcher's profession as a health visitor may have compromised accuracy. Wright et al (6) found that participants who weaned earlier than recommendations were aware they were doing so-respondents may have answered in line with guidelines rather than reality.


Further research is needed to confirm the findings of this study, particularly the effects of the timing of advice. In addition, the reasons why mothers wean early in different social groups needs further exploration. The social patterning of early weaning suggests that health visitors may need to provide additional support to those families most likely to wean early. The possibility that giving weaning advice may encourage the introduction of solids needs to be taken into account when health visitors are deciding when to offer this advice. There may be risks in offering information at an early stage.

Key points

* Weaning is an important dietary event and often occurs earlier than recommended

* The age of weaning is socially patterned, with those who are living in socially deprived circumstances most likely to wean early

* This study found that the timing of weaning advice may influence the timing of weaning, though further research is needed


The author would like to thank everyone who made this study possible--in particular, the respondents who took time to answer the questionnaire, GP practices, line managers and colleagues at North Ayrshire CHP, and MSc supervisor Marion Welsh.

The author also thanks Glasgow Caledonian University PhD supervisors Flora Cornish and Susan Kerr, who commented on early drafts.


(1) Higginson C. Nutrition in the under-fives. Edinburgh: Health Education Board for Scotland, 2001.

(2) World Health Organization. Global strategy for infant and young child feeding. Geneva: World Health Organization, 2003.

(3) Davies DB, O'Hare B. Weaning: a worry as old as time. Current Paediatrics, 2004; 14(2): 83-96.

(4) Bolling K, Grant C, Hamlyn B, Thornton A. Infant feeding survey 2005. London: Information Centre, 2007.

(5) Hart CN, Drotar D. Maternal knowledge of nutrition, problem-solving abilities and the introduction of complementary food into infants' diets. Early Child Development and Care, 2006; 176(7): 693-705.

(6) Wright CM, Parkinson KN, Drewett RF. Why are babies weaned early? Data from a prospective population based cohort study. Archives of Disease in Childhood, 2004; 89(9): 813-6.

(7) Anderson S, Bradshaw P, Cunningham-Burley S, Hayes F, Jamieson L, MacGregor A, Marryat L, Wasoff F. Growing up in Scotland. Edinburgh: Scottish Executive, 2007.

(8) Scottish Executive. Infant feeding strategy for Scotland: a consultation paper. Edinburgh: Scottish Executive; 2006.

(9) Murphy E. Expertise and forms of knowledge in the government of families. Sociological Review, 2003; 52(4): 434-62.

(10) Hamlyn B, Brokker S, Oleinikova K, Wands S. Infant feeding 2000. London: Stationery Office, 2002.

(11) Dubois L, Girard M. Social inequalities in infant feeding during the first year of life: the longitudinal study of child development in Quebec (LSCDQ 1998-2002). Public Health Nutrition, 2003; 6(8): 773-83.

(12) Scottish Executive. Health for all children 4: guidance for implementation in Scotland. Edinburgh: Scottish Executive, 2005.

(13) Gildea A, Slaon S, Stewart M. Sources of feeding advice in the first year of life: who do parents value? Community Practitioner, 2009; 82(3): 27-31.

(14) MacDonald A. Is breast best? Is early solid feeding harmful? Journal of the Royal Society for the Promotion of Health, 2003; 123(3): 169-74.

(15) Lanigan JA, Bishop JA, Kimber AC, Morgan J. Systematic review concerning the age of introduction of complementary foods to the healthy full-term infant. European Journal of Clinical Nutrition, 2001; 55(5): 309-20.

(16) Sloan S, Gildea A, Stewart M, Sneddon H, Iwaniec D. Early weaning is related to weight and rate of weight gain in infancy. Child: care, health and development, 2007; 34(1): 59-64.

(17) Taylor L, Gallagher M, McCullough FSW. The role of parental influence and additional factors in the determination of food choices for pre-school children. International Journal of Consumer Studies, 2004; 28(4): 337-46.

(18) Alridge A, Levine K. Surveying the social world. Buckingham: Open University, 2001.

(19) Scottish Government. Scottish Index of Multiple Deprivation 2009: general report. Edinburgh: Scottish Government, 2009.

(20) Edwards P, Roberts I, Clarke M, DiGuiseppi C, Pratap S, Wentz R, Kwan I, Cooper R. Methods to increase response rates to postal questionnaires. Cochrane Database of Systematic Reviews, 2003; (4): MR000008.

(21) McNeill P, Chapman S. Research methods (third edition). Oxford: Routledge, 2005.

(22) Anderson AS, Guthrie C, Alder EM, Forsyth S, Howie PW. Rattling the plate: reason and rationales for early weaning. Health Education Research, 2001; 16(4): 471-9.

(23) McDougall P. Weaning: parents' perceptions and practices. Community Practitioner, 2003; 76(1): 25-8.

(24) General Register Office for Scotland. Vital events reference tables 2006.Available at: vital-events/vital-events/reference-tables-2006/section-3-births.html (accessed 12 November 2009).

Jane M White MSc, BSc, BN, RN, HV

PhD student, Glasgow Caledonian University
Table 1. Associations between variables and timing of weaning

 Respondents in each
 weaning timing
Demographic variable
 Before four months

Annual income up to 15000 [pounds sterling] 54% (n=27)
 15000 [pounds sterling] to 38% (n=19)
 60000 [pounds sterling] over 8% (n=4)
 60000 [pounds sterling]

Marital status married or living as married 76% (n=43)
 single 24% (n=13)

Ever breastfeed yes 48% (n=27)
 no 52% (n=29)

Qualification none 9% (n=5)
(or equivalent) standard grade 33% (n=18)
 higher grade 11% (n=6)
 HNC,HND 33% (n=18)
 degree or higher 14% (n=8)

Age less than 25 years 34% (n=19)
 25 to 29 years 21% (n=12)
 30 to 34 years 25% (n=14)
 35 years or more 20% (n=11)

Parity first baby 51% (n=28)
 experienced parent 49% (n=27)

Infant's gender boy 57% (n=32)
 girl 43% (n=24)

Demographic Respondents in each weaning
variable timing category

 Four to six months At six months

Annual income 34% (n=15) 29% (n=2)
 34% (n=14) 57% (n=4)
 32% (n=14) 14% (n=1)

Marital status 93% (n=42) 100% (n=7)
 7% (n=3) 0% (n=0)

Ever breastfeed 70% (n=33) 86% (n=6)
 30% (n=14) 14% (n=1)

Qualification 11% (n=5) 0% (n=0)
(or equivalent) 15% (n=7) 14% (n=1)
 15% (n=7) 14% (n=1)
 22% (n=10) 29% (n=2)
 37% (n=17) 43% (n=3)

Age 17% (n=7) 0% (n=0)
 22% (n=10) 57% (n=12)
 29% (n=13) 14% (n=1)
 33% (n=15) 29% (n=2)

Parity 55% (n=26) 57% (n=4)
 45% (n=21) 43% (n=3)

Infant's gender 51% (n=23) 29% (n=2)
 49% (n=22) 71% (n=5)

Demographic Associations

Annual income Parents on low incomes significantly
 more likely to wean early (p=0.033)

Marital status Lone parents significantly more
 likely to wean early (p=0.031)

Ever breastfeed Never breastfed: significantly more
 likely to wean early (p=0.027)

Qualification Parents with lower educational
(or equivalent) attainment more likely to wean
 early, but not significant (p=0.180)

Age Younger parents more likely to wean
 early but not significant (p=0.086)

Parity Not significant (p=0.885)

Infant's gender Not significant (p=0.346)

Table 2. Timing of weaning: comparison

Timing of weaning Whole of scotland In this study
 19905 (10) 2000 (10) 2005 (4) 2006

By six weeks 8% 4% 1% 1%
By eight weeks 22% 7% 2% 2%
By three months 64% 28% 13% 23%
By four months 91% 83% 60% 57%
By six months n/a 99% 98% 100%
COPYRIGHT 2009 Ten Alps Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:PROFESSIONAL
Author:White, Jane M.
Publication:Community Practitioner
Article Type:Report
Geographic Code:4EUUK
Date:Dec 1, 2009
Previous Article:Breastfeeding peer support: are there additional benefits?
Next Article:Even low lead levels can affect a child's behaviour.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters