Putting IFE guidance into practice: operational challenges in Myanmar
Victoria Sibson has been the emergency nutrition adviser for Save the Children UK since April 2007, with a focus on treatment of acute malnutrition and infant and young child feeding in emergencies. Victoria led Save the Children's nutrition response to cyclone Nargis in June and July 2008.
The tireless efforts and innovation of the programme manager, Khin Sett Linn, and her team were essential in realising the response reviewed in this article. The work of Bienfait M'mbakwa, SCUK emergency response nutritionist, in evaluating the response and the insight of infant feeding consultant, Nina Berry, is gratefully acknowledged.
This article outlines the challenges in implementing a response on infant and young child feeding at scale in Myanmar in 2008. It draws upon a review by Save the Children UK on the Save the Children Alliance 'IFE response' to cyclone Nargis.
Cyclone Nargis was the worst natural disaster in the history of Myanmar. It struck on 2nd and 3rd May 2008, causing devastation across the vast and remote Ayeyarwady delta and in the city of Yangon. The scale of human loss and suffering was enormous. Nearly 140,000 people were killed or remain missing and 2.4 million people were severely affected. Save the Children had been working in Myanmar for 13 years, in a large-scale multi-sectoral programme. Immediately after the cyclone, Save the Children started relief operations, redeploying about half the 500 national staff in efforts to deliver an emergency response. A seven-sector cyclone response was implemented, involving shelter, food, livelihoods support, nutrition, health, child protection and education, focused on 12 of the 15 worst-affected townships.
The emergency nutrition response focused on treatment of acute malnutrition as part of a primary health care response in two townships, and protection and support of safe and appropriate infant and young child feeding in the emergency (IFE) in 10 townships. The IFE response was guided by the provisions of Operational Guidance on Infant and Young Child Feeding in Emergencies (Ops Guidance on IFE)1 that addresses policy, coordination, assessment and monitoring, basic interventions and technical interventions. This article focuses on the IFE elements of the Save the Children response and the challenges of putting guidance into practice in this context.
Victoria Sibson/SCUK, Mayanmar, 2008
Needs assessment An IFE response was prioritised for three main reasons:
- Secondary data suggested that pre-cyclone infant and young child feeding practices were poor nationally, with low rates of exclusive breastfeeding (0-3 months only 16%, MICS 2000), early or late introduction of complementary foods (40-50% before 4 months, 15% at 9 months, MICS 2000) and poor dietary diversity and feeding frequency for the complementary feeding age child.
- Since the cyclone, there were cases of young children and infants less than 6 months of age separated from their mothers and being brought to Save the Children child protection staff who did not know how best to manage them.
- Save the Children's rapid assessments confirmed inappropriate, and sometimes dangerous, feeding practices for young children, e.g. commonly infants less than 6 months were being given water, milk made from powder or condensed milk, often from bottles, and mashed rice ('kazi'), at as young as 1 month of age.
- After the cyclone, there was distribution of infant formula by some humanitarian agencies and commercial companies. A Nutrition Cluster was activated and led by UNICEF, but awareness of optimal infant and young child feeding and best practice in emergency situations seemed low amongst humanitarian actors and was not being adequately addressed by any one agency.
Nature of the response
The IFE response comprised four main components:
- Advocacy at Cluster level for an appropriate response to IFE as per the Ops Guidance on IFE and in accordance with the International Code of Marketing of Breast Milk Substitutes and subsequent relevant World Health Assembly Resolutions (the Code).
- Sensitisation of Save the Children staff on safe and appropriate IFE.
- Inclusion of IFE in assessments.
- Integration of IFE into child protection programming in 10 townships where the child protection response was in place. This included a breastfeeding support programme and a programme to minimise the risks of artificial feeding.
Save the Children led a working group on IFE and focused its efforts on working through a list of activities drawn directly from the recommendations of the Ops Guidance on IFE. Advocacy was focused on the clusters, principally the Nutrition Cluster members, but also across sectors and with the Ministry of Health (MoH). It also involved direct lobbying of three infant formula distributing companies undertaking inappropriate distribution, resulting in cessation. Another key activity was promotion of key messages, such as the value of wet-nursing orphaned babies, through communication with the media. Tailored materials were developed for advocacy across health, nutrition and water, sanitation and hygiene programmes (WASH). A 'Joint Statement on IFE', based on the model statement produced at the regional IFE workshop in Bali2 and on the provisions of the Ops Guidance on IFE, were translated and disseminated. This helped to limit inappropriate distribution of breast milk substitutes (BMS).
Since child protection and logistics staff were at the frontline seeing 'problem' cases or inappropriate distribution of BMS, awareness of IFE issues was strengthened amongst this group through targeted internal sensitisation activities. In addition, Save the Children health and nutrition staff were targeted for orientation on IFE. Sensitisation of the Logistics Cluster resulted in a clear statement to logistics cluster members about breast milk substitute control.
There were no inter-sectoral rapid assessments from which to draw data but early qualitative assessments provided sufficient information to broadly determine need and the type of response required. The assessments identified the need for urgent support for non-breastfed infants and for displaced breastfeeding mothers. However, they could not provide adequate information on the scale of the problem or numbers of children in need. A 'consolidated assessment' was undertaken in early July 2008, summarising available assessment data and findings on the pre-cyclone and post-cyclone situation and the response so far. The assessment identified gaps in the wider response, documented lessons learned for future responses and provided recommendations for ongoing Save the Children programming.
Integrated programming on IFE
The direct programme of support on infant and young child feeding capitalised on a large-scale child protection response, and was the first of its kind to be implemented by Save the Children. The child protection team's initiation of 'child friendly spaces' across multiple townships suggested that infant and young child feeding support could be a natural complement to their child protection activities. Programmes of support (see Box 1) and follow-up for both breastfed and non-breastfed infants were established.
Follow up was managed through home visits. These were generally undertaken weekly for infants being artificially fed, and less frequently for breastfed infants. Mothers were paired with a local volunteer who was trained 'on the job' by the counsellor as a lay advisor, to provide daily support to the breastfeeding or artificially feeding mother or carer. By the end of August 2008, 3 months after the cyclone stuck, 110 volunteers were identified, to be paired with mothers receiving skilled support.
Some referrals of malnourished infants were made to MoH hospitals but this was limited by the fact that the referral facilities did not have the expertise to care for these infants.
It was intended to link lactating mothers with relevant food distributions and screen and refer malnourished women for supplementary feeding, but due to limited implementation of targeted feeding programmes this did not occur.
The aim of this programmatic response was to cover the catchment area of the child friendly spaces in 10 townships. The result was estimated to be higher, with township coverage at an average of 18% of all villages by August 2008 (n=447 out of 2456 villages), and 40% of all villages with a Save the Children presence in at least one other sector (n=447 out of 1095 villages). In the assessed villages, 1.3% of infants under 6 months were separated from their mothers or orphaned (n=27).
A total of 67% (n=8,055 out of 12,065) of mothers with children under 2 years of age had attended one education session by the end of A total of 67% (n=8,055 out of 12,065) of mothers with children under 2 years of age had attended one education session by the end of August 2008 (a total of 365 education sessions held). Counsellors undertook 568 Simple Rapid Assessments and 378 Full Assessments5. Of these, 53 caregiver/infant pairs were receiving skilled support to artificially feed or to relactate. A total of 192 mothers were recorded as receiving 'basic aid' to breastfeeding. An unknown number received alternative skilled assistance from breastfeeding counsellors. Overall, 2.1% of mothers of children under 2 in all assessed villages were registered for 'continuing supportive care'.
Kit for minimising the risk of artificial feeding
Operationalisation of these interventions has highlighted important challenges to successful implementation of the Ops Guidance on IFE related to the context (vast geographic area and widely dispersed population), the breadth, scale and focuses of the wider emergency response, 'internal' agency issues and the lack of programme models and tools for mounting an IFE response at scale.
A key gap was the lack of a designated agency with clear responsibility on IFE from the outset of the emergency. Advocacy work progressed with Save the Children's involvement but was very slow, due to the informal nature of Save the Children's IFE leadership role (no terms of reference), lack of time, lack of clarity about how best to engage the MoH, and lack of IFE emergency preparedness (both within Save the Children Alliance and externally).
Specific weaknesses related to the lack of designated coordinating agency included that the military were not sensitised and distributed infant formula; there was no mechanism to control BMS donations that did arrive; and there was poor initial sensitisation of the other Clusters on IFE (logistics, food aid, health, child protection). Also, within Save the Children there was a different management grouping for nutrition (with health) and food security and livelihoods/food aid (with education and protection), which hindered creation of programming links.
Poor agency awareness
Sensitisation of Save the Children staff proved a key challenge for nutrition staff. Mixed awareness and understanding of IFE among all other Save the Children staff was complicated by no prior sensitization on the issues, vast numbers of mobile staff and rapidly expanding staffing structures as well as high staff turnover. Lack of 'internal' awareness also meant that integrated programming 'on paper' was not so easily reflected operationally - in this instance, the IFE programme overlapped with the operational areas of the child protection teams but was not so simple to integrate 'on the ground'. On the positive side, whilst a multi-sectoral response requires negotiation between sectors within an agency (each sector effectively 'competing' for their share of the proposed funding), the coordinators of the Save the Children response did facilitate inclusion of IFE into proposals at the request of the technical adviser.
Donor driven programming
Assessing and defining need was also a key challenge. A limited timeframe to submit proposals meant there was great pressure to estimate target populations and numbers to 'treat', with no time for detailed assessment to provide this information. The multi-sectoral initial rapid assessment (IRA) tool developed by the Nutrition Cluster, that includes infant and young child feeding indicators, was not used in Myanmar. A sector specific rapid assessment was not possible given limited available time and staff and assessment sensitivities peculiar to Myanmar, where the Government does not generally approve or facilitate assessments. An externally led wide assessment (the 'Village Tract Assessment') included questions on infant and young child feeding but insufficient data on the demography of the included households and other contextual information severely limited the utility of the findings (e.g. it was found that 2% of households used infant formula prior to the cyclone, and 1% after, but additional information to understand this finding was lacking). Overall, the data available were not sufficient to determine scale of need or areas for geographic focus.
Possible low coverage of those most at risk
Although programme coverage, when estimated 3 months post cyclone, was higher than aimed for, it was slow to achieve due to the ongoing process of community assessments leading to individual assessments and provision of care rather than rapid initial assessment covering the whole target area. Community assessments ceased in September 2008 for practical reasons and recognising that the period of highest risk for cyclone affected infants had passed. In addition, at 18% of total township populations by the end August, coverage was still low. There is a high probability that the response missed high-risk infants who may have died in the early post cyclone weeks/months, due to slow roll out over the large affected area and poor quantification and targeting of the most vulnerable.
Box 1: Programmes of support for breastfed and non-breastfed infants
Support to breastfeeding women
A 'trainer of trainers (ToT)' was hired in the first weeks to provide a rapid 1 day training for 22 nurses and midwives to become breastfeeding counsellors. This was followed up after a couple of months with an additional training, focusing on counselling skills when it became clear that this would be beneficial.
Participants undertook all aspects of basic aid and skilled support as outlined in the IFE Modules 1 and 23 and the WHO 40 hour breastfeeding counselling training course4. Tailored materials were developed by the ToT to support this abbreviated training. Posters with six key infant and young child feeding messages were developed within 2 weeks and flip charts for counselling developed within 1 month (both were adequate for first phase IEC (Information, Education and Communication)).
The breastfeeding counsellors were deployed to all ten child protection townships. Their main tasks were to:
- Undertake community assessments in the catchment areas of 'child friendly spaces'.
- Identify carers/infants who required additional assessment and potential support.
- Provide basic aid and if necessary, further skilled support to these carers and infants.
The counsellors assessed communities spreading outwards from the new offices (counting children <2 years, infants <1 year, numbers of infants artificially fed or requiring assistance, numbers of pregnant women and also identifying wet nurses). On the same day as a community assessment, counsellors called mothers with children under 2 years of age for an education session and asked who in the group was having breastfeeding problems. Those women highlighting problems and wanting help met with the counsellor individually for a brief chat about the details of the problem, modelled on a 'simple assessment'.
Where indicated, a more extensive 'one to one' session was implemented with scheduled, time to observe, talk and discuss the child's feeding and associated problems, modelled on a 'full assessment'. A plan was made for follow up as appropriate.
Simple Rapid Assessments and Full Assessments of infant feeding with individual carers and their children were based on the guidance and tools provided in IFE Modules 1 and 2. Mothers and all those attending education sessions were given High Energy Biscuits (and shop bought biscuits in their absence). Babies were weighed - it was generally reported that there was not a problem of weight loss (suggesting support choices were good and feeding adequate).
Support to artificially fed infants
A need for small-scale support to non-breastfed infants had been identified through the cases presenting to the child protection team and the rapid assessments. An infant formula distribution system and support mechanism was set up to meet the needs of infants who were not breastfed based on defined criteria:
- Mother is unavoidably absent or dead, or mother is critically ill and incapable of breastfeeding (temporary use may be all that is needed). Mothers who are unwell should be encouraged to breastfeed frequently.
- Mother rejects infant (temporary use may be all that is needed).
- Baby artificially fed before the emergency (temporary use may be all that is needed).
- Mother is relactating, so until breastfeeding is reestablished.
- Mother tests HIV positive, artificial feeding is Acceptable, Feasible, Affordable, Sustainable and Safe, mother chooses to feed her baby infant formula, and neither a (full time) wet nurse nor breast milk donor can be found.
The team developed a spreadsheet for calculating infant formula needs. Locally purchased infant formula, meeting Codex Alimentarius standards, was relabelled in Burmese (meeting the Code requirements). The distribution system was informal (small quantities of relabelled formula given to each carer by the dedicated counsellor following assessment) but seemed appropriate to need and was manageable at this scale.
Because feeding bottles are very dangerous in unsanitary conditions, a 'bottle amnesty' was offered. Beneficiary mothers/carers of the IFE programme were asked to surrender their bottles in return for cups, and supported to increase their breastmilk production so that BMS (and feeding bottles) were no longer necessary for the child. The bottle amnesty was not explicitly planned in the response but was a necessary component.
Technical programming challenges
In devising programmes to meet the needs of breastfed and non-breastfed infants, there were many practical challenges, including:
- The programme was understaffed for the vast geographic area, the difficulties in staff movements and the intensity of their required daily activities. Outreach workers and peer educators weren't hired as planned (due to prioritisation by programme staff of community and individual assessments and provision of care), adequate staff supervision was a challenge and community mobilisation was not sufficiently prioritised in the programme design. Also, the training of the midwives and nurses in breastfeeding counselling did not adequately cover counselling skills at first.
- IEC materials were not used across sectors (health, nutrition, child protection and education) as envisaged, due to limited crosssectoral briefing and training on IFE.
- The group process of identifying those requiring simple assessments may have missed out infants requiring support. Also the completion of full assessments on the same day as the community assessment, education session and simple assessments may have been over ambitious and so risked compromising their quality.
- Wet nursing was promoted and a wet nurse was sought as a first option for all identified infants separated from their mothers. However, full time wet nursing was not practical in any single case encountered, due to social and economic/time constraints.
- The criteria for artificial feeding support were not always adhered to, due largely to inability to provide daily support to establish breast feeding, in the absence of recruitment of volunteer peer educators. This has resulted in more babies being artificially fed than strictly indicated (53 instead of 43).
- Due to limited staffing and a large operational area, education sessions were not often repeated. Instead, efforts focused on sensitisation and identification of infants in need and their follow up. In effect, community messaging and sensitisation was compromised by the need for limited staff to provide skilled life saving support to those few requiring it.
- There were no inpatient facilities for appropriate care for malnourished infants <6 months.
Failure to meet basic food needs
Linking women to the general food distribution (GFD) proved impossible - criteria were not flexible and the GFD had low coverage, was irregular and of poor quality, thereby also not meeting the needs of the complementary fed child. A concept note was submitted to UNICEF for the provision of vouchers for complementary foods but was not well received, as it was considered inappropriate for the initial phase response. Save the Children's supplementary feeding programme (SFP) did not admit malnourished pregnant or lactating women in the early months and SFPs were not present in other operational areas. In devising the IFE response, it was not judged necessary to give food to mothers with adequate nutritional status. However, in practice, this fell short of carers expectations and, coupled with lack of both supplementary feeding for malnourished mothers and general ration targeting for mothers with young children, there was a shortfall in food availability.
Speed of response
Some of the response was fast once technical support was available in country, e.g. initiation of advocacy, translation of a Joint Statement on the response to infant and young child feeding, agreement on key advice and messages and development of IEC material. However, because there was no emergency preparedness for IFE and no pre-designed programming models or tools available, implementation of some activities such as staff sensitisation, procurement and targeted BMS distribution were relatively slow or were not effectively carried out. Whilst initiation of the direct programme was as fast as it could be in the programming context, the roll out was slow relative to the acute needs requiring response. Again, this was due to the large areas and dispersed population being targeted, the relatively few staff hired and the lack of programming tools.
Strategic challenges to IFE response
Convincing decision makers of need for direct support
In developing countries in n on-emergency situations, a non-breastfed infant under 6 months of age is 14 times more likely to die than an exclusively breastfed child. Save the Children's experiences in the Myanmar context reinforces the need for programmes of direct active support for breastfeeding and to minimise the risks of artificial feeding in this context.
Determining the programmatic model for provision of direct infant and young child feeding support: inpatient vs. community level care
Typically, mothers with breastfeeding problems or malnourished infants under six months are referred to inpatient facilities. However, these are not sufficiently accessible to the wider populations (only those close to the hospital choose to attend). Furthermore, inpatient facilities may not have the capacity or expertise to deal with the cases that do present.
Although necessitated by the identification of infants with acute needs by the child protection programme staff, the provision of community level support in Myanmar highlights major challenges. In this instance, cost was a real challenge given the staff to beneficiary ratio needed to deliver a skilled and intensive programme at such large scale amongst a widely dispersed population.
Balance of activities between advocacy, 'basic' interventions and 'technical' interventions providing direct support
Broadly speaking, there are three categories of IFE activity reflected in the Operational Guidance on IFE: advocacy/orientation, basic interventions (e.g. newborn registration, shelter, security, general food ration access, complementary food provision) and technical interventions (e.g. skilled support to breastfed and non-breastfed infants and children). In Myanmar, advocacy and a technical intervention were prioritised by Save the Children, with less attention paid to ensuring effective basic interventions. Whilst it is likely that the restrictive environment for imports was a major advantage in preventing any massive influx of BMS in to Myanmar, there are many countries in the region where this would not be the case. With hindsight, greater attention to basic interventions that potentially impact a greater number of infants and their carers - including plans for handling donations of BMS and to address shortfalls in food aid to pregnant and lactating women and complementary feeding - should feature earlier and higher on the agenda.
Fundamental to meeting the provisions of the Ops Guidance on IFE is a coordinated effort that was lacking in this emergency response. As an agency concerned with IFE, Save the Children have identified the following areas of action needed:
- Early identification of a lead agency for IFE, ideally UNICEF, as stipulated in the Ops Guidance on IFE and for whom protection, promotion and support of optimal infant and young child feeding is a core commitment in emergencies (Core Commitments for Children in Emergencies, UNICEF 2005).
- Practical guidance on when and how to prioritise different types of IFE activity, particularly how to manage an appropriate balance between basic and technical interventions and advocacy/sensitisation.
- Emergency preparedness work involving country level assessments of infant feeding risks and the modelling of appropriate responses given different emergency scenarios.
- Development of emergency plans and proce dures to support feeding of separated/ orphaned infants.
- Development of tools to help project caseloads for direct programmes of support for breast feeding and artificial feeding.
- Urgent development of programming models for supporting breastfeeding and artificial feeding in generic emergency health, nutrition and child protection programmes. Along with the development of associated tools (e.g. for M&E) and ready to use programme specific raining materials, programmatic recommendations would aid the implementation of the guidance provided in the Ops Guidance on IFE and IFE Modules 1 and 2.
- Work on defining most appropriate indicators and output/impact targets for IFE responses.
To address these areas of development and to aid IFE responses in future emergencies, Save the Children is keen to collaborate with the IFE Core Group6, interested operational agencies and notably UNICEF.
Save the Children's IFE response in Myanmar contributed to protecting, promoting and supporting breastfeeding and minimising the risks of artificial feeding among vulnerable cyclone affected babies and young children. However, Save the Children's efforts on IFE to take on external advocacy, internal sensitisation, consider basic interventions, and set up a technical response highlighted several important lessons. Most importantly, there were key gaps in the wider cyclone Nargis response in respecting key provisions of, and co-ordinating efforts to implement, current operational guidance. Secondly, the response also highlights the importance of emergency preparedness for an IFE response and lastly, the urgent need for the development of programming models and tools if actors are to effectively fulfil the provisions of the Ops Guidance on IFE.
For further details, contact: Vicky Sibson, email: V.Sibson@savethechildren.org.uk
1Operational Guidance on IFE for emergency relief staff and programme managers. Available at www.ennonline.net in 11 languages.
2Link to Joint Statement
3IFE Module 1 is being updated and a new orientation package is due out in August 2009. The 2001 version, along with IFE Module 2 in English and French is available at: http://www.ennonline.net/resources/
5Individual level assessments included in IFE Module 2.
6Following on from the Myanmar experiences, Save the Children UK and Save the Children US have both become full members of the IFE Core Group.
More like this
By UNICEF Myanmar As reflected in the field article by Save the Children, there were many elements to coordinating a response on infant and young child feeding (IFE) during...
By Ali Maclaine and Mary Corbett Ali Maclaine has a MSc in Human Nutrition from the London School of Hygiene and Tropical Medicine. She has been involved in infant feeding and...
Maaike's observations concur with many of our experiences around infant feeding in emergencies (IFE) in previous humanitarian interventions. Failure to include breastfeeding...
FEX: Save the Children’s IYCF programme and linkages to Protection, Food Security and Livelihoods in Haiti
A breastfeeding mother who received support in the baby tent By Lucia Pantella Lucia holds a M.Sc. in Humanitarian Programme Management at the Liverpool School of Tropical...
Summary of meeting1 One hundred and twelve participants from 16 countries and special territories, together with regional and international representatives of United Nations...
By Suzanne Mboya Suzanne Mboya is a consultant nutritionist. In 2014 she completed a sixth month mission supporting the Syrian crisis IYCF-E response through a partnership...
A mother and her baby in a temporary shelter following Cyclone Nargis Summary of evaluation report1 An evaluation was undertaken by the Emergency Nutrition Network (ENN) to...
By Susan Thurstans and Vicky Sibson Susan Thurstans has been part of the emergency response team for nutrition with Save the Children UK since January 2009 and previously...
By Isabelle Modigell, Christine Fernandes and Megan Gayford Isabelle Modigell is the IYCF-E Adviser for the Technical Rapid Response Team (Tech RRT) at Save the Children. She...
This mother had exclusively breastfed all her children, including this baby whom she breastfed throughout the war in Lebanon. The aim of the International Code of the...
FEX: The Haiti Earthquake - Country and Global level Cluster Coordination Experiences and Lessons Learnt
By Carmel Dolan and Mija Ververs Carmel Dolan was Global Nutrition Cluster Coordinator-Consultant at the time of the earthquake and located from the UK to UNICEF Headquarters...
Summary of international meeting Media/communication Working Group in the foreground and Implementation/Capacity building Working Group in the background On 1st and 2nd...
Resource: WHO/UNICEF joint statement on appropriate IYCF in the current emergency, and caution about unnecessary use of milk products
Joint statement by WHO / UNICEF for Myanmar, Cyclone Nargis, May 2008. Myanmar Tropical Cyclone WHO and UNICEF joint statement on appropriate infant and young child feeding...
Published Report By Marie McGrath The importance of infant feeding in emergencies has been highlighted during recent emergencies in countries such as Iraq and Bosnia, where...
en-net: Breastfeeding Tent Picture
I am currently working on a project to compile an advice leaflet for local charities with the aim of giving them the knowledge and understanding to better support...
en-net: Writing a university text book on NIE in Arab and Middle East Region, any suggestions, ideas or information? Thank you
Hi, Quite a lot of work was done in the Lebanon crisis in 2006 by Save the Children on infant and young child feeding in emergencies. This was reported in a field exchange...
By Sura Alsamman Sura Alsamman is nutrition supervisor at Save the Children Jordan, responsible for the overall all coordination of the IYCF technical functions and activities...
This tool is designed to assist countries in: summarizing current data with regard to infant and young child feeding practices, assessing the strengths and weaknesses of their...
Summary of published experience1 Location: Haiti What we know: Sub-optimal breastfeeding practices increase risk of infant morbidity, malnutrition and death. Emergencies...
By Bushra Rashid, Nicki Connell and Caroline Wilkinson Bushra Rashid is working as a Nutrition Advisor in the Humanitarian Surge Team of Save the Children UK. She has been...
Reference this page
Victoria Sibson (2009). Putting IFE guidance into practice: operational challenges in Myanmar. Field Exchange 36, July 2009. p30. www.ennonline.net/fex/36/putting