Research and analysis

Breastfeeding at 6 to 8 weeks: a comparison of methods

Published 7 September 2023

Applies to England

Executive summary

This report compares the outputs and methodologies of the 2 current national sources of data for prevalence of breastfeeding at 6 to 8 weeks to determine if the NHS England maternity statistics publication (produced from record level data) is sufficiently established to supersede the Office for Health Improvement and Disparities (OHID) interim reporting system for children’s public health 0 to 5 years (produced from aggregate data).

The analysis presented in this report is based on a comparison of data from NHS England and OHID covering the year April 2020 to March 2021.

Breastfeeding status is recorded by health visitors when they carry out an infant’s 6 to 8 weeks health visitor review. Breastfeeding status is defined in this report using 4 categories:

• full: infants who were breastfed, received breast milk from a bottle or both of these
• partial: infants who were fed both breast and formula or liquid milk
• any: infants where there was full or partial breastfeeding
• none: infants who received no breast milk

Record level data for breastfeeding status is recorded by health visiting service providers and submitted each month to NHS England’s Community Services Data Set (CSDS).

Health visiting service providers also submit aggregate data to local authority commissioners as a key performance indicator (KPI) for contract monitoring. This aggregate data is forwarded to OHID as part of the interim reporting system for children’s public health 0 to 5 years.

Providers of maternity services submit data about mothers and newborn infants to NHS England’s Maternity Services Data Set (MSDS).

NHS England links CSDS with MSDS to allow the inclusion of disparities data for the mother (deprivation, ethnicity and age). The data from the linked CSDS and MSDS asset is published in NHS Maternity Statistics, England - 2020 to 2021. For the 146,910 infant records where a breastfeeding status was known, the prevalence of breastfeeding at 6 to 8 weeks was reported as: 36.5% full, 17.7% partial, 54.2% any and 45.8% none. All proportions are of those infants with known breastfeeding status.

From OHID’s interim reporting system for children’s public health 0 to 5 years, in 2020 to 2021, any breastfeeding at 6 to 8 weeks was 47.6% as a proportion of infants due a 6 to 8 weeks health visitor review. These figures were recalculated based on the total number of infants with known breastfeeding status so that they could be compared with NHS England’s figures. This gave any breastfeeding at 6 to 8 weeks 54.6% which is similar to NHS England’s figure of 54.2%.

The analysis of these data sources resulted in the following main findings:

• breastfeeding is associated with deprivation: rates increase as deprivation decreases
• mothers with white ethnicity were less likely to breastfeed than those mothers from other ethnic groups
• younger mothers were less likely to breastfeed than older mothers

This comparison also offers findings which could inform the improvement of data collection and reporting:

• the NHS England methodology is stronger for reporting on disparities than the OHID interim methodology, as it is constructed from record level data
• low counts for breastfeeding status within CSDS are currently not sufficient to provide the level of coverage required for production of official statistics at local level
• breastfeeding rates were similar when comparing CSDS figures to adjusted OHID figures. This suggests the potential for a linked CSDS and MSDS asset to become the source of data for breastfeeding in the future, given improved data submissions to CSDS

Introduction

Purpose

This report compares the outputs and methodologies of the 2 current national sources of data for breastfeeding at 6 to 8 weeks for the time period 2020 to 2021 in order to determine if the NHS England maternity statistics publication (produced from record level data) is sufficiently established to supersede the OHID interim reporting for children’s public health 0 to 5 years (produced from aggregate data).

The report will be of interest to public health commissioners and health intelligence analysts in local authorities, providers of health visiting services and those working to produce data at a national and local level in NHS England and OHID.

Data collection and reporting in England

The prevalence of breastfeeding at 6 to 8 weeks is reported at national and local level in the Public Health Outcomes Framework. These statistics are derived from OHID’s interim reporting system for children’s public health 0 to 5 years. This voluntary collection of aggregate data from local authorities matches the KPIs recommended for use in commissioning contracts with health visiting service providers. Stakeholder feedback suggests that interim reporting is highly valued, leading to most local authorities submitting data. Data on breastfeeding at 6 to 8 weeks is published by OHID on GOV.UK and on the Fingertips public health data tool. It is available for England with a deprivation breakdown, and for local authorities.

Health visiting service providers also submit record level data to NHS England’s Community Services Data Set (CSDS). NHS England publishes annual maternity statistics which include breastfeeding prevalence at 6 to 8 weeks. These use CSDS as the source of breastfeeding status with a record level link to the Maternity Services Data Set (MSDS). This creates a single data asset which allows reporting on maternal population subgroups by deprivation, ethnicity and age.

Analysis

This report presents findings from the 2 current national sources, at England level, of data for breastfeeding at 6 to 8 weeks:

• NHS Maternity Statistics, England - 2020 to 2021 (NHS England)
• Breastfeeding at 6 to 8 weeks after birth: annual data 2020 to 2021 (based on OHID’s interim reporting system)

It makes comparisons where appropriate and explores potential explanations for similarities and differences between data from OHID and NHS England.

The report considers what data is available to look at disparities in outcomes for different population groups, including deprivation, ethnicity and age of mother.

On 1 February 2023, NHS Digital became part of NHS England. The paper refers throughout to ‘NHS England’ as the organisation which is now responsible for CSDS and MSDS, even where analysis is discussed which was published by NHS Digital before its merger. NHS Digital is acknowledged only where formal references are given.

NHS England maternity publication 2021 to 2022

Since the start of this analysis, NHS England has published the latest edition of annual maternity statistics for 2021 to 2022. In comparison to the findings in 2020 to 2021, there is no substantial change in the trends for breastfeeding or disparities. One difference is a decrease in missing breastfeeding status, accounting for 74.6% of all records in 2020 to 2021 to 68.1% in 2021 to 2022.

Overview of the data collection process

Breastfeeding data collection

Data about breastfeeding status is recorded by health visiting service providers and submitted each month to NHS England’s CSDS using the Strategic Data Collection Service (SDCS). This gives data at record level for each individual and means that patterns for different population groups can be investigated. For example, analysis is possible which looks at the relative rates of breastfeeding for infants with different characteristics such as ethnicity or deprivation. It is also possible to look at factors to do with the mother such as her age, ethnicity or deprivation and how those characteristics might interact, such as patterns for older mothers with different levels of deprivation.

Service providers also submit aggregate data to local authority commissioners as a KPI for contract monitoring. This aggregate data is forwarded to OHID via the Local Government Association as part of OHID’s interim reporting system for children’s public health 0 to 5 years. The aggregate data collates individual records and reports this as a simple count for each local authority. The data does not include any information about deprivation, ethnicity or age.

Breastfeeding status is recorded by health visitors when they carry out an infant’s 6 to 8 weeks universal health visitor review. Breastfeeding status is defined in this report using 4 categories:

• full: infants who were breastfed, received breast milk from a bottle or both of these
• partial: infants who were fed both breast and formula or liquid milk
• any: infants where there was full or partial breastfeeding
• none: infants who received no breast milk

Maternity data collection

Providers of maternity services also submit maternity data to NHS England’s MSDS through the SDCS. This data set contains details of the pregnant woman throughout pregnancy from antenatal care, delivered through outpatient services and in the community, to delivery, and details of the child at and immediately after birth. Demographic details include age, ethnicity and deprivation of the mother, recorded at the midwifery booking appointment, which is recommended to take place within the first 10 weeks of pregnancy.

Linking MSDS and CSDS data

MSDS and CSDS also contain identifiers for both mother and infant which provide a mechanism to link records: mother to mother, infant to infant and mother to infant. NHS England links breastfeeding records from the infant in CSDS to the matching records for mother and infant in MSDS, using matching pseudonymised identifiers. These linked records were used to produce the NHS Maternity Statistics, England - 2020 to 2021 report and form the breastfeeding cohort.

The breastfeeding cohort includes infants if they were 63 days old between April 2020 and March 2021 inclusive. The ages of infants were derived from either the date of delivery in MSDS or date of birth in CSDS. Data reported to either MSDS or CSDS after March 2021 was not included in the cohort. Data reported to MSDS back to January 2020 was used to identify both eligible infants and for matching back from CSDS. Data reported to CSDS before April 2020 was used to identify infants who would be aged 63 days old between April 2020 and March 2021. Infants with no CSDS activity between April 2020 and March 2021 who have CSDS activity before April 2020 were included in the cohort if they would be 63 days old between April 2020 and March 2021 and were due their 6 to 8 week review. There were 577,460 infant records in the population cohort, sourced from MSDS and CSDS. These break down as follows:

• there were 466,415^{[footnote 1]} (81%) infants’ records which were found in MSDS. Where data was also found in CSDS, the records were linked between the 2 data sets. Unpublished analysis from NHS England indicates that approximately 40,000 infants were not found in CSDS and so could not be linked from MSDS
• there were 111,0551 (19%) infants who only appeared in CSDS

Analysis of disparity breakdowns include maternal age, ethnicity and deprivation (based on the mother’s residential postcode). Not all deliveries were recorded in MSDS. For those 111,055 infants recorded in CSDS but whose delivery was not recorded in MSDS, disparities data for mothers was unavailable. A breakdown of the breastfeeding status for these infants is presented in Appendix 1, table A.

Some of the difference in cohort size between MSDS and CSDS (MSDS has a cohort which is closer to the actual population of infants indicated by the Office for National Statistics (ONS)) could be explained by differences in incentives for maternity service providers and health visiting service providers to submit data. The Maternity Incentive Scheme is an incentive programme to encourage maternity providers to submit good quality data to MSDS. While NHS information standards require community service providers to submit health visitor and breastfeeding data to CSDS, there are no incentives similar to the Maternity Incentive Scheme or penalties associated with not submitting data.

Summary and comparisons of breastfeeding prevalence

Breastfeeding prevalence data from NHS England

NHS England provides numbers for breastfeeding with disparities breakdowns. For the purposes of this report, infants with full, partial or where the breastfeeding status was recorded as ‘none’, have been aggregated into a metric ‘infants whose breastfeeding status is known’, to allow for proportions to be calculated and comparisons with OHID figures made.

In 2020 to 2021 the NHS England data showed 79,615 infants (54.2%) where there was any breastfeeding (full and partial) as a proportion of those where breastfeeding status was known. There were 53,590 (36.5%) infants where there was full breastfeeding, 26,025 (17.7%) where there was partial breastfeeding, and 67,295 (45.8%) where there was no breastfeeding (table 1).

Table 1: breastfeeding at 6 to 8 weeks - numbers and percentages, from MSDS and CSDS linked

Breastfeeding status	Numbers	Percentage of all records	Percentage of records where breastfeeding status is known
Full	53,590	9.3	36.5
Partial	26,025	4.5	17.7
Any	79,615	13.8	54.2
None	67,295	11.7	45.8
Missing value or value outside reporting parameters	430,550	74.6	Not applicable
Total	577,460	Not applicable	Not applicable

Deprivation

Based on the index of multiple deprivation (IMD), the linked asset data showed:

• the 10% most deprived areas in England had the lowest rates of full (23.2%) and any (38.3%) breastfeeding, and the highest rates where there was none (61.7%) (figure 1)
• the 10% least deprived areas had the highest rates of full (50.9%) and any (71.2%) breastfeeding, as well as the lowest rates where there was none (28.8%)

Further breastfeeding statistics by deprivation can be found in Appendix 1, table B.

Figure 1: breastfeeding at 6 to 8 weeks in 2020 to 2021, by deprivation decile (IMD-15) of mother from CSDS and MSDS linked

IMD-15 decile	Any	None
1 - most deprived	38.26%	61.74%
2	46.19%	53.81%
3	51.92%	48.08%
4	53.89%	46.11%
5	57.00%	43.00%
6	59.95%	40.05%
7	60.84%	39.16%
8	61.31%	38.69%
9	64.14%	35.86%
10 - least deprived	71.19%	28.81%

Note: rates use total where IMD and breastfeeding status is known as a denominator (117,540).

In looking at deprivation, NHS England uses the 2015 index of multiple deprivation (IMD-15) rather than the more up-to-date IMD-19 (2019) ranking. As NHS England has person level data, the method assigns a deprivation decile based on lower-layer super output area (LSOA) of residence to each mother. This gives a good assessment of breastfeeding rates by deprivation decile as it is based on an individual’s data.

Ethnicity

The linked data asset showed:

• mothers from a white ethnic group had the lowest rate of any breastfeeding (49.8%) and the highest rate where there was none (50.2%) (figure 2)
• mothers from a black or black British ethnic group had the highest rate of any breastfeeding (85.3%) and the lowest rate where there was none (14.7%) (figure 2)
• mothers from an Asian or Asian British (69.0%) ethnic group, mixed ethnic group (66.4%) and those from all other ethnic groups (combined 75.8%) had high rates of any breastfeeding and lower rates of none (figure 2)

Further breastfeeding statistics based on ethnicity of mother can be found in Appendix 1, table C.

Figure 2: breastfeeding at 6 to 8 weeks in 2020 to 2021 by ethnic group of mother from CSDS and MSDS linked

Ethnicity	Any	None
Asian or Asian British	69.0%	31.0%
Black or black British	85.3%	14.7%
Mixed	66.4%	33.6%
White	49.8%	50.2%
Any other ethnic group	75.8%	24.2%

Note: rates use total where ethnicity and breastfeeding status is known as a denominator (107,895 infants).

Age of mother

The linked data asset showed:

• as mothers’ age increases, the rate of any breastfeeding increases. Rates of any breastfeeding were lowest among young mothers under 20 (21.2%), increasing with age and reaching their highest levels among mothers aged 45 and over (67.6%) (figure 3)
• rates for infants who were not breastfed were highest in younger mothers under 20 (78.8%). As mothers’ age increased, the number of infants who were not breastfed decreased, reaching its lowest level with mothers aged 45 and over (32.4%) (figure 3)

Further breastfeeding statistics on age of mother can be found in Appendix 1.

Figure 3: breastfeeding at 6 to 8 weeks in 2020 to 2021 by age of mother from CSDS and MSDS linked

Age	Any	None
Under 20	21.15%	78.85%
20 to 24	34.22%	65.78%
25 to 29	48.97%	51.03%
30 to 34	61.88%	38.12%
35 to 39	67.11%	32.89%
40 to 44	65.96%	34.04%
45 or over	67.65%	32.35%

Note: rates use total where age and breastfeeding status is known as a denominator (117,540 infants).

Breastfeeding prevalence data from OHID’s interim reporting system for children’s public health 0 to 5 years

In 2020 to 2021, OHID’s interim reporting system for children’s public health 0 to 5 years found that there was any breastfeeding for 47.6% of infants as a proportion of infants due a 6 to 8 weeks health visitor review. Further values for England and regions can be found in table 3.

OHID’s interim reporting system uses a 3-stage process of data validation to ensure that the data which is published is of a suitable quality. Local authorities which are unable to meet all 3 stages of validation have their value (percentage) supressed. Their value is not included in the national or regional calculations and is not published as an official statistic at local authority level (see validation rules in Appendix 2).

OHID publishes breakdowns by deprivation for the local authority of residence. This is a relatively crude analysis of deprivation and less sensitive than building from record level data at LSOA of residence. Data for 2020 to 2021 indicates lower rates of any breastfeeding at 6 to 8 weeks in more deprived areas. Higher rates of any breastfeeding were found in less deprived local authorities. Further information on data published by OHID about deprivation can be found in Fingertips and in Appendix 2 to this report.

Comparisons of statistics between OHID and NHS England

Data published by OHID includes infants with unknown breastfeeding status in the denominator when calculating the proportion of infants where there was any breastfeeding. This is different from NHS England’s methodology which only uses records for infants with known breastfeeding status in the denominator. For the purposes of comparison, methodologies have been standardised. OHID figures have been recalculated so breastfeeding status is reflected as a proportion of infants whose breastfeeding status is known. This allows for like-for-like comparison between data from the 2 sources (table 2).

The comparison displayed in table 2 shows that in 2020 to 2021:

• based on OHID data published as official statistics for 2020 to 2021, 262,777 (47.6%) infants had any breastfeeding at 6 to 8 weeks
• analysis by NHS England, based on CSDS, reported that there was any breastfeeding for 79,615 (54.2%) infants. This is similar to the OHID value above
• when figures were recalculated using OHID’s data but with the total number of infants where breastfeeding status is known as a denominator, a new prevalence figure of 54.6% was generated for infants receiving any breastmilk. This is similar to the NHS England value (54.2% for any breastfeeding at 6 to 8 weeks)

Table 2: breastfeeding at 6 to 8 weeks - comparison of breastfeeding rates from OHID and NHS England

Breastfeeding status	OHID (published method)	OHID (standardised method for comparison purposes)	NHS England (maternity statistics method)	Percentage point difference between OHID (standardised method) and NHS England method
Full	176,093 (31.9%)	176,093 (36.6%)	53,590 (36.5%)	0.1
Partial	86,683 (15.7%)	86,683 (18.0%)	26,025 (17.7%)	-0.3
Any	262,777 (47.6%)	262,777 (54.6%)	79,615 (54.2%)	0.4
None	218,223 (39.5%)	218,223 (45.4%)	67,295 (45.8%)	0.4
Total	480,999	480,999	146,910	Not applicable

Note: the OHID published method uses ‘infants due a 6 to 8 weeks health visitor review’ as a denominator. The OHID standardised method uses the total where breastfeeding status is known as a denominator.

Breastfeeding rates which have been standardised between publications are broadly similar at England level, with a difference of 0.4 percentage points or less for all status of breastfeeding.

Comparisons at regional level are not published in OHID’s official statistics for breastfeeding at 6 to 8 weeks, as data for most regions does not pass stage 3 of the validation rules (see validation rules in Appendix 2). Stage 3 of the validation rules is not relevant for the standardised method so it is possible to reanalyse data at regional level using this methodology.

In addition, NHS England’s maternity statistics publication uses NHS England regions as the mid-level grouping. This combines data from the North East with Yorkshire and the Humber, and the East Midlands with the West Midlands. For the purposes of comparison, OHID’s data has been combined into the NHS England regional geographies and the standardised method has been applied.

There is significant variation at regional level between NHS England’s figures and OHID’s figures (table 3).

Table 3: regional and national figures: comparison of any breastfeeding prevalence at 6 to 8 weeks

NHS England region	OHID (published method)	OHID (standardised method for comparison purposes)	NHS England (maternity statistics method)	Percentage point difference between OHID (standardised method) and NHS England method
North East and Yorkshire and the Humber	Data failed to pass validation	41.4%	44.7%	-3.3
North West	Data failed to pass validation	44.3%	42.0%	1.3
Midlands	Data failed to pass validation	51.3%	48.1%	3.2
East of England	Data failed to pass validation	55.4%	69.4%	-14.0
London	Data failed to pass validation	79.3%	74.0%	5.3
South East	Data failed to pass validation	59.0%	60.0%	-1.0
South West	Data failed to pass validation	58.2%	58.0%	0.2
England	47.6%	54.6%	54.2%	0.4

Notes:

• the OHID published method uses ‘infants due a 6 to 8 weeks health visitor review’ as a denominator. The OHID standardised method uses the total number of records where breastfeeding status is known as a denominator
• information about data validation rules is in Appendix 2

The OHID interim reporting system for children’s public health 0 to 5 years does not collect data about ethnicity or mother’s age. However, it can provide data by deprivation. As OHID only has data aggregated at local authority level, it assigns deprivation to each local authority and not to records of individual mothers.

Figures 4 and 5 show the proportion of infants where there was any breastfeeding at 6 to 8 weeks by deprivation decile of local authority, for NHS England and OHID respectively. OHID’s report showed that the proportion of infants where there was any breastfeeding varied with deprivation, with any breastfeeding increasing as deprivation decreased. NHS England data also showed this trend but with a more consistent increase in rates of any breastfeeding as deprivation decreased.

Figure 4: NHS England breastfeeding at 6 to 8 weeks in 2020 to 2021 by deprivation decile (IMD-15)

IMD-15 decile	CSDS
1 - most deprived	38.3%
2	46.2%
3	51.9%
4	53.9%
5	57.0%
6	59.9%
7	60.8%
8	61.3%
9	64.1%
10 - least deprived	71.2%

Note: deprivation decile is based on LSOA deprivation scores derived from mother’s postcode.

Figure 5: OHID breastfeeding at 6 to 8 weeks in 2020 to 2021 by deprivation decile (IMD-19) of mother

IMD-19 decile	OHID
1 - most deprived	42.1%
2	38.3%
3	47.8%
4	50.4%
5	41.2%
6	50.4%
7	47.6%
8	47.0%
9	53.7%
10 - least deprived	56.5%

Findings

The description and analysis presented in this report offers findings, both about levels and patterns of breastfeeding but also where further work is required to achieve long-term, comprehensive data for this topic.

Breastfeeding in England

Comparing the data on breastfeeding from NHS England and OHID confirms some important findings about the general patterns of breastfeeding in the population. These include:

• breastfeeding rates were associated with deprivation: rates increase as deprivation decreases
• mothers from a white ethnic group were less likely to breastfeed than those mothers from other ethnic groups
• younger mothers were less likely to breastfeed than older mothers

Data quality and consistency

A comparison of the data on breastfeeding from these 2 sources also offers findings which could inform the improvement of data collection and reporting. These include:

• the NHS England approach has a stronger methodology for reporting on disparities than the OHID interim methodology, as it is constructed from record level data. Assigning deprivation at LSOA level also allows for a more detailed and robust analysis
• coverage of breastfeeding status from CSDS is low, with 74.6% of records missing a breastfeeding status. This suggests that providers are not submitting breastfeeding status to CSDS consistently. Low counts for breastfeeding within CSDS are currently not sufficient to provide the level of coverage required for publication as official statistics at local level. However, at a national level, births data categorised by maternal age appears to be broadly comparable to births data published by ONS for 2020, and comparisons and conclusions about breastfeeding prevalence at a national level are appropriate

Conclusions

NHS England’s maternity statistics are not yet sufficiently established to supersede the OHID interim reporting methodology for production of official statistics on breastfeeding at 6 to 8 weeks, especially at local authority level. However, the NHS England methodology has some advantages for exploring inequalities by population sub-groups. Any future decision to switch data source would be dependent upon improvements to the coverage and data quality of CSDS and MSDS. These data sets need to be as complete as possible in order to become the source of data for breastfeeding official statistics at national, regional and local level.

NHS England and OHID provide 2 different reporting pathways for breastfeeding at 6 to 8 weeks, each with their own strengths and limitations (table 4). The analysis shows that national breastfeeding prevalence was similar using both methodologies when OHID figures were standardised to align with the NHS England methodology. The NHS England methodology is stronger for reporting on disparities than the OHID interim methodology, as it is constructed from record level data. Using LSOA deprivation scores based on IMD-19 would give a more nuanced and accurate measure of rates by deprivation deciles.

OHID is currently undertaking a detailed reconciliation exercise at local authority level between data from the OHID interim reporting system for children’s public health 0 to 5 years and other statistics produced from CSDS by NHS England. This work is intended to inform the development of practical solutions to improving data quality in the CSDS so that it can replace the OHID interim reporting system for children’s public health 0 to 5 years as the source of official statistics for breastfeeding at 6 to 8 weeks and for other indicators.

Value and limitations of NHS England’s methodology and OHID’s interim methodology for reporting prevalence of breastfeeding at 6 to 8 weeks

NHS England (linked MSDS and CSDS)

Value:

• coverage of MSDS is high (86.2% of deliveries in Hospital Episode Statistics (HES) in 2020 to 2021)
• supports segmentation of statistics by maternal age, ethnicity and deprivation
• national breastfeeding prevalence is similar to that calculated by standardising OHID methodology to align with the NHS England methodology
• can form the basis of breastfeeding prevalence at national level
• deprivation data is derived from record level LSOA which is more accurate than the OHID methodology
• automated data validation at record level
• provider incentive to submit to MSDS
• aggregated from record level data

Limitations:

• 74.6% of infants in CSDS have breastfeeding status missing or unknown
• cannot yet support production of statistics at local authority level due to data coverage issues
• deprivation disparities used by NHS England are the 2015 IMD rankings, not the latest available from 2019

OHID (interim reporting system)

Value:

• supports the current production of official statistics on breastfeeding at 6 to 8 weeks for the Public Health Outcomes Framework
• high participation rate from local authorities submitting aggregate data on a voluntary basis
• can be used to produce statistically significant prevalence at national, regional and local authority level
• feedback demonstrates high value to local authorities

Limitations:

• relies on a voluntary submission of aggregate data to OHID
• can only support segmentation by deprivation at local authority level
• cannot support segmentation by maternal age, ethnicity or deprivation derived from record level LSOA
• high level validation only
• validation and data quality controls can suppress regional and local data
• process was established as a temporary solution while CSDS coverage and data quality was improving

Appendix 1: supporting and background information about NHS England’s breastfeeding reporting

For the purposes of its reporting, NHS England offers guidance on when and what data should be recorded and submitted to national systems about breastfeeding.

All mothers and children are eligible for 5 universal health visitor reviews:

• antenatal
• new birth visit
• 6 to 8 weeks review
• one year review
• 2 to 2 and a half year review

Breastfeeding status is obtained at the earliest opportunity between 42 and 63 days, during the mandated 6 to 8 weeks health visitor review.

At these reviews, health visitors and their teams encourage breastfeeding and record the breastfeeding status of the infant as either:

• full: infants who are breastfed, received breast milk from a bottle or both of these. NHS England refers to this as ‘exclusively breastfed’. In OHID’s reporting, it is referred to as ‘totally breastfed’
• partial: infants who are fed both breast and formula or liquid milk
• any: for the purposes of standardisation across publications, this refers to infants where there was full or partial breastfeeding. In OHID publications, this is referred to as ‘totally or partially breastfed’
• none: infants who receive no breast milk. NHS England refers to this as ‘no breast milk feeding’ and OHID refers to this as ‘not at all breastfed’
• missing value or value outside reporting parameters: this publication also includes counts for missing data. Missing data includes missing breastfeeding status, value falling outside of reporting parameters (such as breastfeeding status recorded at 12 weeks) or where there is no record in MSDS but there is a record in CSDS^{[footnote 2]}

Breastfeeding metadata

In the information that supports the analysis NHS England has published in this area, the technical process used for identifying relevant records is described as follows:

• identifying the cohort of infants (maternity services field MSD401, community services field CYP001) uses maternity services MSD401 records submitted between January 2020 and March 2021 with date of birth between 29 January 2020 and 27 January 2021, or community services CYP001 records submitted up to March 2021 with a date of birth between 29 January 2020 and 27 January 2021, such that the infant would be 9 weeks (63 days) old between 1 April 2020 and 31 March 2021

• if MSD401 or CYP001 records exist for the infant where stillbirth or termination is recorded, or there are data quality issues with the recording of the date of birth or date of death, or a date of death that has been recorded less than 63 days after the date of birth, the infant record is not included in the 2020 to 2021 analysis

• identifying breast feeding status (CYP201, CYP202, CYP610):
  Each of the infant records (MSD401, CYP001) are linked to community activity recorded in CSDS for that Person_ID_Infant / Person_ID. Breastfeeding status is taken from the earliest contact date (CYP201) between 42 and 63 days after the date of birth (MSD401, CYP001) for which information about whether the infant is breastfed or not is recorded in CYP202 or CYP610^{[footnote 2]}

Table A: breastfeeding at 6 to 8 weeks: numbers and percentages by breastfeeding status, where there is an infant record in CSDS only

Breastfeeding status	Number	Percentage of all infants in CSDS only (%)
Full	10,425	9.4
Partial	4,770	4.3
None	14,175	12.8
Missing value or value outside reporting parameters	81,735	73.6

Table B: breastfeeding at 6 to 8 weeks - numbers and percentages of any breastfeeding by deprivation decile, where breastfeeding status is known

IMD-15 decile	Full	Partial	Any	None
1 - most deprived	3,970 (23.2%)	2,585 (15.1%)	6,555 (38.3%)	10,580 (61.7%)
2	3,635 (28.4%)	2,275 (17.8%)	5,910 (46.2%)	6,885 (53.8%)
3	4,325 (33.3%)	2,425 (18.7%)	6,750 (51.9%)	6,250 (48.1%)
4	4,620 (35.9%)	2,315 (18.0%)	6,935 (53.9%)	5,935 (46.1%)
5	4,520 (38.5%)	2,175 (18.5%)	6,695 (57.0%)	5,050 (43.0%)
6	4,525 (41.5%)	2,015 (18.5%)	6,540 (59.9%)	4,370 (40.1%)
7	4,355 (42.2%)	1,930 (18.7%)	6,285 (60.8%)	4,045 (39.2%)
8	4,370 (42.6%)	1,920 (18.7%)	6,290 (61.3%)	3,970 (38.7%)
9	4,265 (45.4%)	1,755 (18.7%)	6,020 (64.1%)	3,365 (35.9%)
10 - least deprived	4,500 (50.9%)	1,790 (20.3%)	6,290 (71.2%)	2,545 (28.8%)

Note: percentages use total number of infants for the deprivation decile of mother’s residence as a denominator.

Table C: breastfeeding at 6 to 8 weeks - numbers and percentages of any breastfeeding by ethnic group of mother, where breastfeeding status is known

Ethnicity	Full	Partial	Any	None
Asian or Asian British	3,505 (33.7%)	3,670 (35.3%)	7,175 (69.0%)	3,225 (31.0%)
Black or black British	1,695 (47.9%)	1,325 (37.4%)	3,020 (85.3%)	520 (14.7%)
Mixed	860 (43.1%)	465 (23.3%)	1,325 (66.4%)	670 (33.6%)
White	31,270 (35.6%)	12,470 (14.2%)	43,740 (49.8%)	44,020 (50.2%)
Any other ethnic group	1,940 (46.2%)	1,245 (29.6%)	3,185 (75.8%)	1,015 (24.2%)

Note: percentages use total where ethnicity and breastfeeding status is known as a denominator.

Table D: breastfeeding at 6 to 8 weeks - numbers and percentages of any breastfeeding by age of mother, where breastfeeding status is known

Age	Full	Partial	Any	None
Under 20	380 (12.9%)	245 (8.3%)	625 (21.2%)	2,330 (78.8%)
20 to 24	3,450 (22.5%)	1,805 (11.8%)	5,255 (34.2%)	10,100 (65.8%)
25 to 29	10,770 (33.0%)	5,230 (16.0%)	16,000 (49.0%)	16,675 (51.0%)
30 to 34	16,945 (42.6%)	7,660 (19.3%)	24,605 (61.9%)	15,155 (38.1%)
35 to 39	9,685 (44.5%)	4,905 (22.6%)	14,590 (67.1%)	7,150 (32.9%)
40 to 44	1,825 (38.7%)	1,285 (27.3%)	3,110 (66.0%)	1,605 (34.0%)
45 and over	110 (32.4%)	120 (35.3%)	230 (67.6%)	110 (32.4%)

Note: percentages use total where age and breastfeeding status is known as a denominator.

NHS England metadata for sourcing IMD

DeprivationDecileAtBooking

Indicates the 2015 IMD Decile of the mother’s residence at booking

“Breakdown:

CASE WHEN Rank_IMD_Decile_2015 IS NOT NULL THEN Rank_IMD_Decile_2015

WHEN Rank_IMD_Decile_2015 IS NULL AND LSOAMother2011 IS NOT NULL THEN ‘Resident

Elsewhere in UK, Channel Islands or Isle of Man’

WHEN Rank_IMD_Decile_2015 IS NULL AND Upper(PostcodeDistrictMother) = ‘ZZ99’ THEN ‘Pseudo postcode recorded (includes no fixed abode or resident overseas)’

ELSE ‘Missing Value / Value outside reporting parameters’ END

COUNT(DISTINCT Person_ID_Mother, UniqPregID)”

Appendix 2: supporting and background information about OHID’s breastfeeding reporting

Table showing breastfeeding at 6 to 8 weeks - numbers and percentages of any breastfeeding by deprivation decile, from OHID data

IMD 2019	Number	Percentage (%)
1 - most deprived	26,493	42.1
2	14,954	38.3
3	22,468	47.8
4	24,572	50.4
5	22,203	41.2
6	23,650	50.4
7	35,964	47.6
8	29,502	47.0
9	38,009	53.7
10 - least deprived	24,962	56.5
England	262,777	47.6

Note: OHID percentages use ‘infants due a 6 to 8 weeks review’ as a denominator.

Validation rules

Stage 1 : number of infants totally or partially breastfed at 6 to 8 weeks less than or equal to number of infants due a 6 to 8 weeks review. Both data items a valid integer.

Stage 2: number of infants due a 6 to 8 weeks review within 20% of the resident population of infants aged 0 for each local authority, based on 2020 mid-year population estimate. Annual estimate is divided by 4 to provide a quarterly estimate.

Stage 3: percentage of infants whose breastfeeding status was known greater than or equal to the data quality standard of 95% (85% for England). This means the sum of those infants recorded as totally breastfed, partiality breastfed, or not at all breastfed is between 95% and 100% of the total number of infants due a 6 to 8 weeks review.

Appendix 3: differences between reporting pathways

Healthcare providers submit data directly to MSDS, CSDS and HES (NHS England). Although all maternity providers submit data to MSDS, not all deliveries are included in the records. In data submitted for 2020 to 2021, 86.2% of the deliveries recorded in HES were also included in MSDS. This means that, while MSDS has coverage of all providers, it currently does not cover all births that occurred in England. In addition, community service providers are required monthly to submit data describing referrals and service contacts for infants to CSDS.

Community providers also submit health visiting and breastfeeding data to their local authority commissioners, who, in turn, forward this data, in aggregate format, on a voluntary basis to OHID’s interim reporting system for children’s public health 0 to 5 years (via the Local Government Association) for publication.

Statistics published by OHID for 2020 to 2021 show there were 552,003 infants due a 6 to 8 weeks review, of which 480,999 had a known breastfeeding status (87.1%).

Data published in the NHS England maternity statistics publication has poorer coverage with 146,910 infants whose breastfeeding status was known (25.4%) and with 430,550 infants whose breastfeeding status was unknown (74.6%) in 2020 to 2021.

Data flow for community services data on breastfeeding status and maternity data

The data collected by front line services is collated, analysed and published by national organisations for planning and research purposes.

Sources of data

Midwives and health visitors provide care to pregnant women and their babies, and collect data as clinical records of care:

• midwives collect record level data on pregnant women and babies which is submitted to NHS England’s MSDS
• health visitors collect record level data on the care of babies, infants and children which is submitted to NHS England’s CSDS

NHS England publication

NHS England links MSDS and CSDS data. They then use breastfeeding status from CSDS and the mothers’ characteristics from MSDS to produce breastfeeding prevalence at 6 to 8 weeks, broken down by mothers’ characteristics.

NHS England publishes these metrics in the annual NHS maternity statistics publication.

OHID publication

Separately, providers of health visiting services send aggregate data on breastfeeding to their local authority commissioners.

Local authorities submit the aggregate breastfeeding data to OHID.

OHID publishes official statistics on breastfeeding at 6 to 8 weeks, based on the aggregate data from local authorities.

1. These figures are subtotals based on rounded figures so they do not add up to the total number of infants in the cohort. ↩

2. NHS Digital (2021). NHS Maternity Statistics, England - 2020 to 2021 - see MSDS metadata in the ‘Resources’ section (accessed 3 February 2023). ↩ ↩²