Breastfeeding indicatoors for use in population surveys in Australia: have we reached consensus?
Article Type: Report
Subject: Breast feeding (Research)
Health status indicators (Research)
Author: Hector, Debra
Pub Date: 11/01/2011
Publication: Name: Breastfeeding Review Publisher: Australian Breastfeeding Association Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2011 Australian Breastfeeding Association ISSN: 0729-2759
Issue: Date: Nov, 2011 Source Volume: 19 Source Issue: 3
Topic: Event Code: 310 Science & research
Geographic: Geographic Scope: Australia Geographic Code: 8AUST Australia
Accession Number: 279262874
Full Text: Introduction

An indicator is described by the World Health Organization as 'a variable with characteristics of quality, quantity and time used to measure, directly or indirectly, changes in a situation and to appreciate the progress made in addressing it. It also provides a basis for developing adequate plans for improvement' (WHO 2001). Simple, valid indicators are crucial to track progress and guide investment to improve nutrition and health during the first 2 years of life. Indicators that assess breastfeeding are useful to monitor trends, to develop and evaluate [policies and] programs, and for advocacy (Arimond et al 2008). Without an accurate national picture of breastfeeding practices and how these are changing over time, it is difficult to make rational decisions about the need for additional/ more effective programs and interventions to achieve policy goals (Webb et al 2001, p13).

History

Breastfeeding indicators are underpinned by definitions of various breastfeeding practices, relating to the initiation, duration and intensity of breastfeeding. Development of definitions around breastfeeding began in the late 1980s. In 1988, a meeting sponsored by the Interagency Group for Action on Breastfeeding (IGAB), an ad hoc working group of representatives from UNICEF, the United States Agency for International Development (USAID), the Swedish International Development Agency (SIDA) and the World Health Organization (WHO), led to the publication of the first schema to categorise breastfeeding practices (Labbok & Krasovec 1990). The meeting was organised to agree upon a set of definitions that could be used as standardised terminology for the collection and description of cross-sectional information on breastfeeding behaviour. This schema and framework: (1) acknowledged that the term 'breastfeeding' alone is insufficient to describe the numerous types of breastfeeding behaviour, (2) distinguished full from partial breastfeeding, (3) subdivided full breastfeeding into categories of exclusive and almost exclusive breastfeeding, (4) differentiated among levels of partial breastfeeding and (5) recognised that there can be token breastfeeding with little to no nutritional impact. The breastfeeding definitions used in this schema are summarised in Table 1.

Another meeting was held, in Geneva in June 1991, again with representatives from WHO, UNICEF, USAID, as well as the Demographic Health Surveys Program of the Institute for Resource Development/Macro International Inc. The purpose of this meeting included standardisation of breastfeeding/ infant feeding definitions but was particularly aimed at reaching consensus on the definitions of key infant feeding indicators and specific methodologies for their measurement: to have a common set of measures to assess breastfeedingpractices and evaluate the progress of promotional programmes. Indicators should be limited in number, relatively easy to measure and interpret, and operationally useful. The focus of the indicators should be on intra-country comparison, although the degree of comparability between countries is also of interest. This meeting resulted in a set of infant feeding definitions (Table 1) and indicators (WHO 1991), as well as direction for measurement of the indicators. Labbok (2000) has critiqued the WHO definitions and indicators as being designed only to study what enters the infant's mouth (a nutrition-focused approach) and not examining the impact of direct breastfeeding compared to indirect breastmilk-feeding. In Australia, Thorley (2011) has recently highlighted the importance of differentiating between breastfeeding and breastmilk-feeding.

Efforts to reach consensus on breastfeeding indicators for use in Australia were first formally documented in 2001 by members of the National Food and Nutrition Monitoring and Surveillance Project, funded by the Commonwealth Department of Health and Aged Care (Webb et al 2001). Webb et al indicated that surveys prior to 2001 had comprised a lack of consistent definitions, that studies were conducted for different purposes and with different methodologies, with different infant age groups, and different reporting periods; rendering reliability and meaningfulness low. The 2001 Report, Towards a national system for monitoring breastfeeding in Australia: Recommendations for population indicators and next steps, aimed to develop national breastfeeding indicators to provide a common set of measures for use by those that collect data (a) on nationally representative samples, and (b) others who wish to compare their sample data with national data about breastfeeding. However, although the focus was largely national comparisons, one of the key selection criteria was that the indicators be 'consistent with WHO wherever possible to meet international reporting obligations'. Other criteria used in selecting indicators included: relevance to key policy recommendations, consistency with previous indicators/data collected in Australia so that trends may be documented, feasibility/ease of collecting the required information on a nationally representative sample and measurable and valid for detecting the direction and magnitude of changes over time and differences between population subgroups. The set of indicators proposed for use in Australia were derived through extensive stakeholder consultation and expert consensus. These indicators were described in full including description of the rationale or purpose of each indicator, how it should be calculated, measurement issues and data requirements. The breastfeeding definitions underpinning these indicators are summarised in Table 1 and the indicators are listed in Table 2. The indicators related to intensity of breastfeeding are determined using 24-hour recall methods (see below).

The 2001 indicators were not formally tested nor adopted in Australia. In 2008 the Department of Health and Ageing commissioned a review and update of the indicators described by Webb et al. Extensive stakeholder consultations during this 2008 review were conducted mainly by email and, although considerable progress was made, full consensus was not reached at the end of the consultation period. A main recommendation to government in the consultation final report (Hector 2008, unpublished) was that a round-table discussion be held to try and reach consensus on breastfeeding indicators. A workshop was subsequently held, in December 2010, attended by some of those who had participated in the 2008 review; federal, state and territory government employees working in nutrition, epidemiology/health data and maternal health areas; and academics and researchers in child health. A set of proposed core indicators were brought to the workshop by the Australian Institute of Health and Welfare (AIHW). Criteria for indicator selection included supporting the reporting of breastfeeding trends in Australia, especially in terms of evaluating the Australian National Breastfeeding Strategy 2010-2015 ('the Strategy', Australian Health Ministers' Conference 2009). The indicators were discussed for suitability, stability, simplicity and measurability resulting in final agreement to a set of six core indicators. The workshop Report (AIHW 2011) provides some of the history, process, discussion points and decisions regarding these core indicators. Naturally, there was debate at the workshop concerning a number of the indicators but agreement ('no workshop attendee had any major issues with') was reached. The agreed core indicators (AIHW 2011) are summarised in Table 1, compared to the 2001 indicators.

The 1991 WHO Infant and Young Child Feeding Indicators have also been reviewed and updated 'to reflect developments in infant and young child feeding recommendations and scientific knowledge about what constitutes optimal breastfeeding and complementary feeding practices' (WHO 2008), for example to reflect the Global Strategy to exclusively breastfeed for the first 6 months of life (WHO/UNICEF 2003). The WHO (2008) emphasises, as in 1991, that the indicators are mainly designed for use in large-scale surveys or national programs. The indicators for breastfeeding are not vastly different to those in 1991; most of the indicators, except 'children ever breastfed' and 'early initiation of breastfeeding', are based on current status/24-hour recall data (see below); a number of new indicators were incorporated to reflect timely, adequate and appropriate complementary feeding.

Comparison of Australian Breastfeeding Indicators: 2001 and 2011

First glance would suggest that these two sets of indicators are very similar and indeed they are. There is broad agreement as to what breastfeeding/infant feeding practices need to be measured and reported. However, there are a number of obvious and less obvious differences between the sets of indicators. Two main differences between 2011 and 2001 are: (1) Use of preposition: Indicators related to breastfeeding intensity, ie exclusive and predominant/full breastfeeding, involve the preposition 'to' rather than 'at'. Choice of preposition was extensively discussed in the 2008 Review (also in Hector 2011) and during the 2010 workshop (AIHW 2011); where it was decided that concepts that involve not having an attribute are best expressed using the preposition 'to'. For example, 'exclusive breastfeeding' is defined as not having anything other than breastmilk (eg exclusively breastfed to 6 months). Use of 'to' accurately reflects infant feeding recommendations to exclusively breastfeed 'to' 6 months (rather than 'at' 6 months which is not a useful indicator; Hector 2011); plus the 'to' implies from birth to the month of age when exclusive breastfeeding ceases (whether 'first' or 'regular' is used to signify deviation from exclusive breastfeeding; Hector 2011). 'To' could be replaced with 'for', 'until' or 'through'--the latter used in the United States Breastfeeding Report (although the term 'through' is used interchangeably with the preposition 'at'; CDC National Immunization Survey; CDC 2011); and, (2) Non-use of the 24-hour period of recall: It was decided, conversely, that concepts which involve having an attribute are best expressed using 'at' (eg receiving soft/semi-soft/solid food ('solids') at 'x' months or receiving breastmilk at 'x' months). For solid foods this reflects the currently recommended practice (NHMRC 2003; currently under revision) to introduce solids at (around) 6 months. Use of the preposition 'at' in these instances is congruent with expression of the indicators in 2001. However, in 2001 indicators related to breastfeeding intensity were described according to practices during the previous 24-hours, ie using recall survey methods (Table 1). This method of measurement is known to produce misclassification errors for breastfeeding practices and does not measure practices since birth (Binns et al 2009; Hector 2011). The 24-hour recall method was chosen by Webb et al (2001) for greater accuracy. Although an earlier options paper had recommended that rates be measured and reported by 24-hour recall and longer-term recalled practice (the two methods are debated fully by Hector 2011; the systematic overestimation of exclusive breastfeeding using longer-term recall methods (retrospective assessment) has been recently reported from recall as short as 9 months (Agampodi et al 2011). Use of the 24-hour recall method was not preferred by stakeholders in the 2008 review or during the 2010 workshop. Regardless, the indicators were not determined at the workshop in relation to likely survey method or data availability. The AIHW considered that the survey method would not be incorporated into or be 'defined' in association with the indicators. Nevertheless, some indication of the survey methods is given in the 2011 Workshop Report (Table 2, AIHW 2011) and this creates some perplexities.

These and other perplexities, discrepancies and details are discussed below in relation to each of the newly agreed indicators in comparison to 2001 and the revised WHO indicators:

Proportion of children ever breastfed: (a) Reducing the recall period from 4 years to 2 years will increase validity; (b) Also, trends over time are easier to report, without having to do complex analyses (Smith et al 2006); (c) This is an optional WHO Indicator (2008; no. 9), although on a minor point of difference, the WHO uses 0-23.9 months, ie does not include recalled practice of mothers of infants aged 24 (24.0-24.9) months.

Proportion breastfed at each month of age, 0-24 months: (a) Extension of reporting of this indicator from 12 to 24 months reflects the WHO recommendation to breastfeed for up to 2 years, or longer; (b) The indicated denominator in the Workshop Report should include those infants/children aged 'x months or older at the time of the survey, if longer-term recall data are used.

Median duration of breastfeeding: (a) Discarded as not pertinent, nor well-understood and communicated. (b) Remains an optional indicator (no. 13) in the WHO 2008 Indicators (0-35.9 months).

Proportion exclusively breastfed to each month, 0-6 months: (a) Use of the preposition 'to' rather than 'at' is described above. (b) The report indicates 7 data points however 6 data points is correct--'to 1 month', 'to 2 months', 'to 3 months', 'to 4 months', 'to 5 months', 'to 6 months'. If using the proposition 'at' (eg if using 24-hour recall data) then there are still 6 data points: 'at <1 month', 'at 1 month', 'at 2 months', 'at 3 months', 'at 4 months', 'at 5 months'; not at 6 months as it is recommended to exclusively breastfeed to 6 months, not at 6 months, when complementary foods are recommended.

Proportion predominantly (fully) breastfed to each month, 0-6 months: (a) The WHO neither uses nor defines the term 'fully' breastfeeding/breastfed in either of their 'indicators' documents (WHO 1991, 2008). This term was used in the 2001 Report, following Cattaneo et al (2000), to be the sum of exclusive and predominant breastfeeding. However in the Strategy and the 2011 Workshop Report it is indicated that 'predominant' breastfeeding is equivalent to 'fully' breastfeeding. Equating 'predominant' and 'fully' is in accordance with the WHO definition of predominant breastfeeding in that 'predominant' requires that the infant receive breastmilk and allows but doesn't require the infant to receive water and other watery drinks, including juice (Table 2). This means that exclusive breastfeeding is then a 'subset' of predominant breastfeeding and predominant breastfeeding is not a distinct breastfeeding practice as interpreted by Webb et al (2001) and Cattaneo (2000). Possibly the 'confusion' has arisen from the terms used by Labbok & Krasovec (1990) where 'full breastfeeding' was defined as the sum of 'exclusive' and 'almost exclusive', where 'almost exclusive' breastfeeding is distinct from exclusive breastfeeding (Table 2). However, confusion persists. Although the WHO measurement guide (WHO 2010) clearly indicates which questions and therefore infant feeding practices are to be used to determine predominant breastfeeding, ie to include those infants who are receiving breast milk as the sole source of nourishment and those who are also receiving water or watery drinks/juice; some researchers are still interpreting predominant breastfeeding as distinct from exclusive breastfeeding using the WHO methods (eg Hanif 2011); (b) It is worth mentioning here the main objective of the Strategy: 'to increase the percentage of babies who are fully breastfed from birth to 6 months of age, with continued breastfeeding and complementary foods to twelve months and beyond'. It would have been preferable for the Strategy to have 'exclusive' rather than 'full breastfeeding to 6 months' as the core objective/indicator to measure progress as this is the recommended practice for optimal infant health; (c) The specifications for this indicator in the report indicate that it would be calculated using survival analysis; but the impression from the workshop was that all of the indicators would be determined using longer-term recall/retrospective assessment and therefore survival analysis (no specification was given for exclusive breastfeeding; also see next two indicators).

Receiving soft/semi-solid/solid foods (solids) at each month, 0-12 months: (a) 'Soft' is now included in the definition of 'solids' (WHO 2008, AIHW 2011); (b) It is indicated that this (and the next) indicator be determined from 24-hour recall data, but this is likely an inaccuracy in the report. If 'complementary' feeding (ie introduction of solids) gives the deviation from exclusive breastfeeding according to recommendations, ie to exclusively breastfeed from birth to 6 months and then continue breastfeeding while introducing solids, then it makes sense to use the same survey questions and analytical methods to report against both indicators, exclusive breastfeeding and timing of introduction of solids; (c) In the report 'complementary feeding' should be changed to 'require that the infant receive soft, solid or semi-solid food' to be consistent with the Strategy and WHO; (d) Additionally, the Strategy and the Report indicate that 'partial' breastfeeding is equivalent to 'complementary'. However partially breastfed infants are not necessarily receiving solids in addition to breastmilk (more than 'token' but not as much as fully, Labbok & Krasovec 1990; partial breastfeeding is defined within WHO 1991, 2008); they could be receiving non-human milk or formula, or watery liquids, as well as breastmilk. Complementary foods are any food suitable to be fed to infants as a complement to breastmilk or to infant formula when milk alone becomes insufficient to meet the nutritional needs of the infant. Partial breastfeeding has been incorrectly defined as being when the infant receives breastmilk in addition to complementary foods in other studies (eg Mirshahi et al 2008). Often the terms 'complementary' and 'supplementary' are incorrectly used interchangeably; supplementary feeds are classified as fluids other than breastmilk, such as artificial formula or water, which could be given to a baby after a breastfeed or instead of a breastfeed. The WHO (2009) only uses complementary feeding to relate to appropriate feeding practices in breastfed children aged 6-23 months and has replaced the 1991 indicator 'timely complementary feeding rate' with the indicator 'introduction of solid, semi-solid or soft foods', which is a measure of a single feeding practice.

Proportion receiving non-human milk or formula at each month, 0-12 months: (a) As per the previous indicator, this would best be determined using the same methods as for exclusive breastfeeding, ie longer-term recall; and, (b) 'Non-human milk or formula' is now used instead of 'breastmilk substitutes'.

Ongoing developments

The above narrative, including the differences among breastfeeding and infant feeding definitions and indicators shown in Table 1 and Table 2, provides insight into some of the intricacies of measuring and reporting on breastfeeding practices in cross-sectional population level surveys. The variations in indicators and definitions for describing breastfeeding arise and continue despite each group of definitions and indicators being determined with similar aims and objectives and by consensus-forming processes.

The AIHW have decided not to prescribe how the data is to be collected, analysed and interpreted in order to report on the indicators. However, the critical need to work towards harmonisation of methodologies for measuring indicators has been stressed by the WHO (2008) and has resulted in the publication of their detailed measurement guide (WHO 2010). Attention to correct use of preposition ('to' or 'at') within indicator phrasing in accurate reflection of the data collected will increase comparability of indicators between surveys. However, differences in survey questions, survey and analytical methods and interpretation of the data, including the boundary points (Hector 2011), renders comparisons often dubious at best. Indicators such as 'proportion exclusively breastfeeding in the previous 24 hours at 'x' months of age' (eg Webb et al 2001), 'exclusive breastfeeding from birth to '6' months' and/or breastfeeding definitions which reflect the period over which exclusive breastfeeding is measured, eg total breastfeeding = exclusively breastfed in the previous 7 days (but not since birth necessarily) (Breastfeeding Committee for Canada 2006), have the potential to provide more accurate comparisons across surveys; however, such indicators are not as 'simple' as those currently agreed upon.

The upcoming report on the recent National Infant Feeding Survey of over 50,000 Australian infants aged 0-2 years (AIHW, October 2010) will provide insight into the success of the agreed indicators. Comparisons of the survey questions, methods and indicators with those used in ongoing population health surveys in the jurisdictions--particularly in Queensland and New South Wales--may support further indicator refinement. Indeed, the AIHW (2011) stresses that indicator development is a cycle, and that application of the indicators to current and future data collections including the recent Australian National Infant Feeding Survey, may lead to refinement of the indicators and/ or improved data collections as resources permit. The report also indicates that it will be important to standardise the indicator definitions under appropriate governance arrangements; and that the AIHW and the Australian Department of Health and Ageing have initiated a project to formalise the specifications for the indicators. This is occurring as part of a broad government effort to improve the comparability, consistency and relevance of national information on the health and wellbeing of Australians. For example, a set of indicators for children's health--the Children's Headline Indicators--exist in Australia. These are a set of 19 indicators designed to focus policy attention on priorities for children's health, development and wellbeing, one of which is for breastfeeding: 'proportion of infants exclusively breastfed at 4 months of age'.

Indicators must be broadly consistent with standards, recommendations and best practice. The draft report of the updated Infant Feeding Guidelines for Health Workers has been released for public consultation (NHMRC 2011) and included in these draft Guidelines is the main recommended practice: In Australia, it is recommended that as many infants as possible be exclusively breastfed until around six months of age (22-26 weeks) when spoon foods are introduced. This recommendation, if accepted post-consultation, will mean that a single indicator 'percentage of infants exclusively breastfed until/to 6 months of age' will no longer be useful, as solid foods will be introduced once the infant has turned 5 months of age, assuming a proxy of 5 months is used for '22-26 weeks'; technically 22 weeks is equal to 5 months plus 1 week if a month has an average of 4.33 weeks. A single indicator relating to the recommendation about exclusive breastfeeding would then be 'percentage of infants exclusively breastfed until/ to 5 months of age' or if using the preposition 'at' then the useful single indicator would be 'exclusive breastfeeding at 4 months' (as per the current headline indicator). Similarly 'introduction of solid foods before 5 months' would be the corresponding single indicator in order to show adherence to recommended practice.

Stakeholders in the 2008 Review and subsequent workshop, as reflected in the 2011 Report, highlighted the desirability of collecting indicators from maternity services in a nationally standardised manner. Many hospitals are not accredited as 'Baby-Friendly' in Australia, yet may implement several or many of the component Ten Steps. Australia could implement routine data collection of hospital practices which support breastfeeding ie 24-hour rooming-in and early skin-to-skin contact or undermine breastfeeding, ie supplementation in hospital. Such a system could possibly also report on exclusive breastfeeding during the hospital period, although it is important that any such system does not necessitate health professionals making 'judgements' on the category of breastfeeding but that the questions enable later categorisation of breastfeeding practices.

There was agreement at the workshop that further attention needs to be given to devise and implement appropriate data collection methods that better capture information from population sub-groups, such as those residing in remote areas, indigenous Australians, those from culturally diverse groups and those from low socio-economic groups.

Summary points

Breastfeeding indicators for population monitoring have been developed over time, nationally and internationally, using consensus-forming processes. However, there are subtle and palpable differences between the various sets of indicators as well as the underpinning breastfeeding definitions.

Provision of a set of nationally agreed indicators does not mean that the same survey questions and survey, analytical and interpretive methods, are used to report against the indicators across different surveys, hence comparisons (eg across jurisdictions or with international data) should be made cautiously.

A set of indicators has been agreed for use nationally in Australia (AIHW 2011) for which indicator specification is occurring. Indicator definitions will be standardised under appropriate governance arrangements according to a broader government effort to improve the comparability, consistency, and relevance of national information on the health and wellbeing of Australians.

Reporting of particular indicators may need to be revised in accordance with changing recommended practices.

References

Agampodi SB, Fernando S, Dharmaratne SD, Agampodi TC 2011, Duration of exclusive breastfeeding: validity of retrospective assessment at nine months of age. BMC Pediatrics 11: 80.

Arimond M, Daelmans B, Dewey K 2008, Indicators for feeding practices in children. The Lancet 371(9612): 541-2.

Australian Health Ministers' Conference 2009. The Australian National Breastfeeding Strategy 2010-2015. Australian Department of Health and Ageing, Canberra.

AIHW, Australian National Infant Feeding Survey, conducted October 2010. http://www.abs.gov.au/ausstats/ abs@.nsf/Lookup/3414.0main+features1512011%20(Edition%202), paged updated 28 March 2011.

AIHW 2011. National breastfeeding indicators: workshop report. Cat. No. PHE 146. Canberra: AIHW.

Binns CW, Fraser ML, Lee AH, Scott J 2009, Defining exclusive breastfeeding in Australia. J Paediatr Child Health 45: 174-180.

Breastfeeding Committee for Canada 2006. Breastfeeding definitions and data collection periods. The National Authority for the WHO/Baby Friendly Hospital Initiative in Canada.

Cattaneo A, Davanzo R, Ronfani L 2000, Are data on the prevalence and duration of breastfeeding reliable? The case of Italy. Acta Paediatr 89: 88-93.

CDC 2011. Centers for Disease Control and Prevention, Division of Nutrition, Physical Activity and Obesity, National Centre for Chronic Disease Prevention and Health Promotion, www.cdc.gov/breastfeeding/data/NIS_data/index.htm

Hanif HM 2011, Trends in breastfeeding and complementary feeding practices in Pakistan, 1990-2007. Int Breastfeeding J 6:15.

Hector DJ 2011, Complexities and subtleties in the measurement and reporting of breastfeeding practices. Int Breastfeed J 6:5.

Hector DJ (2008 unpublished consultancy report) Review and update of the breastfeeding indicators contained in the report 'Towards a national system for monitoring breastfeeding in Australia: Recommendationsfor population indicators, definitions and next steps'. Submitted to the Australian Department of Health and Ageing.

Labbok MH, Krasovec K 1990, Towards consistency in breastfeeding definitions. StudFam Plan 21(4): 226-30.

Labbok M 2000, What is the definition of breastfeeding? Breastfeeding Abstracts 19(3): 19-21.

Mihrshahi S, Oddy WH, Peat JK, Kabir I 2008, Association between infant feeding patterns and diarrhoeal and respiratory illness: a cohort study in Chittagong, Bangladesh. Int Breastfeed J 3: 28.

NHMRC 2003, Dietary Guidelines for children and adolescents in Australia, incorporating the infant feeding guidelines for health workers. National Health & Medical Research Council, Commonwealth of Australia, pp. 288-291.

NHMRC 2011, Infant feeding guidelines for health workers. Draft for public consultation, October 2011, National Health & Medical Research Council.

Smith PJ, Zhao Z, Wolter KM, Singleton JA, Nuorti JP 2006, Age-period-cohort analyses of public health data collected from independent serial cross-sectional complex probability sample surveys. Seattle, WA: Joint Statistical Meeting; 2006.

Thorley V 2011, The dilemma of breast milk feeding Breastfeed Rev 19(1): 5-7.

Webb K, Marks GC, Lund-Adams M, Abraham B 2001, Towards a national system for monitoring breastfeeding in Australia: recommendations for population indicators, definitions and next steps. Woden ACT: Australian Food and Nutrition Monitoring Unit.

WHO 1991, Indicators for assessing breastfeeding practices: Reprinted report of an informal meeting, June 11-12 1991, Geneva Switzerland. Division of Child Health and Development. World Health Organization, Geneva, WHO/ CDD/SER/91.14.

WHO 2001, Glossary, Health Systems Performance. www.who.int/health-systems-performance/docs/glossary. htm#indicator, accessed 27 October 2011.

WHO, UNICEF 2003, Global strategy for infant and young child feeding. World Health Organization, Geneva.

WHO, UNICEF, USAID, AED, UCDAVIS, IFPRI 2008, Indicators for assessing infant and young child feeding practices. Part 1: Definitions. Conclusions of a consensus meeting held 6-8 November 2007 in Washington DC, USA World Health Organization, Geneva.

WHO 2010, Indicators for assessing infant and young child feeding practices. Part 2: Measurement. World Health Organization, Geneva.

Debra J Hector PhD, MPH Senior Research Fellow, Prevention Research Collaboration, School of Public Health, University of Sydney NSW 2006
Table 1: Comparison of 'consensus' definitions of breastfeeding

                     Labbok &          WHO (1991)
                     Krasovec (1990)

Exclusive            Requires:         Requires:
breastfeeding        breastmilk from   breastmilk
                     the breast        (including milk
                                       expressed or
                     Allows: nothing   from a wet
                     else              nurse)

                                       Allows: drops
                                       or syrups
                                       consisting of
                                       vitamins,
                                       mineral
                                       supplements or
                                       medicines

Almost exclusive     Vitamins,
                     minerals,
                     water, juice,
                     and ritualistic
                     fluids given
                     infrequently in
                     addition to
                     breastfeeds

Predominant                            Breastmilk
breastfeeding                          (including milk
                                       expressed or
                                       from a wet
                                       nurse) as the
                                       predominant
                                       source of
                                       nourishment

                                       Allows: liquids
                                       (water, water-
                                       based drinks,
                                       juice, ORS),
                                       ritual fluids,
                                       drops or syrups
                                       (vitamins,
                                       minerals,
                                       medicines)

Full breastfeeding   Exclusive
                     Breastfeeding
                     plus Almost
                     Exclusive
                     breastfeeding

Partial              Breastfeeding
breastfeeding        at high,
                     medium, or low
                     intensity
                     Distinct from
                      'full
                     breastfeeding'
                     (see above) and
                     from 'token
                     breastfeeding'
                     which relates
                     to minimal,
                     occasional,
                     irregular
                     breastfeeds

Complementary                          Requires that
breastfeeding                          the infant
                                       receives breast
                                       milk and solids
                                       or semi-solid
                                       foods Allows:
                                       any food or
                                       liquid
                                       including non-
                                       human milk

                     Webb et al        WHO (2008)
                     (2001)

Exclusive            Requires:         Requires:
breastfeeding        breastmilk        breastmilk
                     (including        (including milk
                     colostrum, milk   expressed or
                     expressed or      from a wet
                     from a wet        nurse)
                     nurse)

                     Allows: drops,    Allows: ORS,
                     syrups            drops, syrups
                     (vitamins,        (vitamins,
                     minerals,         minerals,
                     medicines)        medicines)

Almost exclusive

Predominant          An infant's       Requires:
breastfeeding        predominant       breastmilk
                     source of         (including
                     nutrition has     expressed
                     been breastmilk   breast milk or
                     but may also      from a wet
                     have received     nurse) as the
                     water and         predominant
                     water-based       source of
                                       nutrition

                     drinks            Allows: certain
                     (sweetened and    liquids (water
                     flavoured         and water-
                     water, teas,      based drinks,
                     infusions,        fruit juice),
                     etc); fruit       ritual fluids
                     juice; ORS;       and ORS, drops
                     drops and         or syrups
                     syrups in         (vitamins,
                     limited           minerals,
                     quantities        medicines)

                     Note--this is
                     the definition
                     in the glossary
                     not as used
                     within the
                     indicator
                     definitions

Full breastfeeding   Exclusive
                     Breastfeeding
                     plus
                     Predominant
                     Breastfeeding
                     (in glossary
                     and indicator
                     definitions)

Partial
breastfeeding

Complementary        Receives:         Requires:
breastfeeding        breastmilk and    breastmilk
                     solid or semi-    (including milk
                     solid food        expressed or
                                       from a wet
                     Allows: 'this     nurse) and
                     may include any   solid or semi-
                     food or liquid    solid foods
                     including non-
                     human milk'       Allows: any
                                       food or liquid
                                       including non-
                                       human milk and
                                       formula

                                       Used to
                                       describe infant
                                       feeding 6-23
                                       months

                     National          Breastfeeding
                     Breastfeeding     Indicators
                     Strategy (2009)   Workshop Report
                                       (2011)

Exclusive            Requires:         Requires:
breastfeeding        breastmilk        breastmilk
                     (including        (including
                     expressed         expressed
                     breastmilk) and   breastmilk)
                     medicines         Allows: ORS,
                     (including ORS,   drops, syrups
                     vitamins,         (vitamins,
                     minerals)         minerals,
                                       medicines)

Almost exclusive

Predominant          Receives:         Receives:
breastfeeding        breastmilk and    breastmilk
                     medicines         (including
                     Allows: water,    expressed
                     water-based       breastmilk) as
                     drinks, tea or    the predominant
                     fruit juice       source of
                     (which is not     nourishment
                     recommended for
                     babies)

                                       Allows: certain
                                       liquids (water,
                                       water-based
                                       drinks, fruit
                                       juice1), ritual
                                       fluids, ORS,
                                       drops or syrups
                                       (vitamins,
                                       minerals,
                                       medicines)

Full breastfeeding   Same as           Same as
                     Predominant       Predominant
                     Breastfeeding     Breastfeeding

Partial              Requires: solid   Receives:
breastfeeding        or semi-solid     breastmilk
                     food plus         (including
                     breastmilk        expressed
                     (including        breastmilk)
                     expressed         Allows: any
                     breastmilk)       food or liquid
                     Allows: any       including non-
                     food or liquid    human milk and
                     including non-    formula [same
                     human milk and    as
                     formula [same     complementary]
                     as
                     complementary]

Complementary        Requires:         Receives:
breastfeeding        breastmilk        breastmilk
                     (including        (including
                     expressed         expressed
                     breastmilk)       breastmilk)
                     plus solid or     Allows: any
                     semi-solid food   food or liquid
                     Allows: any       including non-
                     food or liquid    human milk and
                     including non-    formula [same
                     human milk and    as partial]
                     formula [same
                     as partial]

(1) Of concern to some workshop participants (as reflected in the
review , Hector 2008, unpublished) was the inclusion of fruit juice as
allowable for predominant/full breastfeeding; the debate is partly
about when it is appropriate to introduce juice (AIHW 2011)

Table 2: Comparison of breastfeeding indictors developed for population
monitoring in Australia in 2001 and 2011

Webb et al (2001)      AIHW (2011)            Comments

Per cent ever          Proportion of          Shorter recall
breastfed (1)          children ever          period will provide
                       breastfed (2)          more valid data
                                              Also can identify
                                              trends over time
                                              without having to do
                                              complex analyses
                                              (age-period cohort
                                              analyses to
                                              determine year of
                                              breastfeeding
                                              practice according
                                              to age of child at
                                              time of survey)

Per cent breastfed     Proportion of          The denominator
AT each completed      children breastfed     indicated in the
month of age to 12     AT each month of age   AIHW report is:
months (1)             to 24 months (2)       number of children
                                              aged x months--this
                                              should include the
                                              number of children
                                              aged x months or
                                              older at time of
                                              survef if longer-
                                              term recalled
                                              practices are
                                              included

Median duration of                            Not included in AIHW
breastfeeding among                           indicators (not
 'ever breastfed'                             often used nor well
children (1)                                  understood (2008
                                              review))

Per cent exclusively   Proportion of          Changed from 24-
breastfeeding in the   children exclusively   hour recall to
previous 24 hours      breastfed TOb each     include recalled
among infants AT       month of age, 0-6      practices (where
each completed month   months (2)             recall period is up
of age to 6 months                            to 2 years)
(3)                                           Denominator should
                                              include number of
                                              children aged x
                                              months or older at
                                              the time of the
                                              survey (as per
                                              above (a)) Report
                                              indicates 7 data
                                              points (0-6 months);
                                              but should be 6 (b)

Per cent fully         Proportion of          Issues with
breastfeeding in the   children               differences in
previous 24 hours      predominantly          interpretation of
among infants AT       breastfed TO (b)       WHO definitions of
each completed month   each month of age,     predominant
of age to 6 months     0-6 months (2)         breastfeeding
(3)                                           between the two
                                              reports--in AIHW
                                               'fully ' is
                                              equivalent to
                                              predominant whereas
                                              in Webb et al
                                               'fully ' equals the
                                              sum of predominant
                                              plus exclusive
                                              Changed from 24-
                                              hour recall to
                                              include longer-term
                                              recalled practices
                                              (where recall period
                                              is up to 2 years)

                                              Report indicates 7
                                              data points (0-6
                                              months); but should
                                              be 6 (b)

Per cent receiving     Proportion of          Relates to single
solid foods in the     children receiving     infant feeding
previous 24 hours      soft-semi-solid-       recommendation about
among infants AT       solid food AT (a)      age of introduction
each completed month   each month of age,     of solids Could be
of age to 6 months     0-12 months            expressed according
(3)                                           to those giving-
                                              not-giving breast
                                              milk at the time of
                                              solids introduction
                                              (nearly all data
                                              collections would
                                              allow this sub-
                                              analysis)
                                              Determine using
                                              survival analysis
                                              as per other
                                              indicators; not as
                                              indicated in
                                              the Report which is
                                              on 24-hour recall
                                              basis?

Per cent receiving     Proportion of           'Breast milk
breast milk            children receiving     substitutes' and
substitutes in the     non-human milk or       'non-human milk or
previous 24 hours      formula At (a) each    formula' are
among infants AT       month of age, 0-12     equivalent--non-
each completed month   months                 human milk was
of age to 6 months                            majority decision in
(3)                                           December meeting
                                              Determine using
                                              survival analysis as
                                              per other
                                              indicators; not as
                                              indicated in the
                                              Report which is on
                                              24-hour recall
                                              basis?

(1) Based on maternal recall among children aged 0-47 months

(2) Based on maternal recall among children aged 0-24 months [point
estimates for various indicators includes longer-term recalled
practice as well as current practice in survival analysis (eg the
estimate of the per cent breastfeeding at each month of age includes
both those currently breastfeeding (censored) and those that have
ceased (uncensored)]

(3) Based on 24-hour recall (current practice)

(a) If the Indicator is expressed on an 'AT' basis then the data
points refer to 'at <1 month, at 1 month, at 2 months, at 3
months etc

(b) If the indicator is expressed on a 'TO' basis then the data points
refer to 'to 1 month, to 2 months, to 3 months etc
Gale Copyright: Copyright 2011 Gale, Cengage Learning. All rights reserved.