ARTICLE IN PRESS

Midwifery (2010) 26, e6 e13

www.elsevier.com/midw

Incorrect advice: the most signicant negative determinant on breast feeding in Malta
Simon Attard Montalto, MBChB, MD (Lpool), FRCP, FRCPCH, DCH (Professor)a,, Helen Borg, RM, BSc (Infant Feeding Specialist Midwife)b, Mary Buttigieg-Said, SRN, RM, Dappsocst (Practice Development Midwife)b, Edward J Clemmer, BA, MS (R), PhD (Medical Statistician)c
a

Department of Paediatrics, St. Lukes & Mater Dei Hospitals, Malta Breastfeeding Counselling Unit, Department of Midwifery, St. Lukes & Mater Dei Hospitals, Malta c The Medical School, University of Malta, Msida, Malta
b

Corresponding author.

E-mail address: simon.attard-montalto@gov.mt (S. Attard Montalto).

Received 25 October 2007; received in revised form 7 June 2008; accepted 15 June 2008

Abstract Objective: this study reviewed breast-feeding rates in Malta for the rst six months of life and identied reasons why mothers discontinue breast feeding in this small island state. Design and participants: a random sample of 405 new mothers who chose to breast feed in the only major state hospital were contacted by phone one week postnatally and again each month up to six months, and presented with a questionnaire relating to their feeding experience. Findings: breast-feeding attrition rates were high with just 152 (38%) of 403 analysable babies still breast feeding at six months. The reasons for stopping breast feeding were categorised by: maternal choice; medical reasons; lack of information; social reasons, incorrect advice and no reason provided. 200 (50%) of the total cohort stopped breast feeding following incorrect advice from health professionals. Just 14 (3.5%) and 17 (4.2%) mothers stopped as a result of their own choice or a medical problem, respectively. Of the total of 403, 77 (19%) mothers introduced supplementary bottle feeds in hospital; of these, 70 (91%) stopped breast feeding altogether soon afterwards. This compared with just 180 (55%) of 326 women who did not introduce bottle feeds in hospital yet subsequently discontinued breast feeding (po0.001). Key conclusions: many health professionals in Malta are not sufciently committed to supporting breast-feeding mothers, and articial feeds are widely recommended without any scientic-based rationale. Incorrect advice on breast feeding is often given early prior to discharge from hospital. As a result, many Maltese mothers introduce supplementary articial milk feeds in hospital, and this is signicantly associated with subsequent cessation of breast feeding within six months of discharge. Implications for practice: the introduction of a clear hospital breast-feeding policy and appropriate education for all health professionals involved in maternity care is strongly recommended. & 2008 Published by Elsevier Ltd. Keywords Incorrect advice; Breast feeding

0266-6138/$ - see front matter & 2008 Published by Elsevier Ltd. doi:10.1016/j.midw.2008.06.002

ARTICLE IN PRESS
Incorrect advice: the most signicant negative determinant on breast feeding in Malta e7

Introduction
Many countries including Malta are not yet compliant with the WHO/UNICEF declaration and global strategy in support of breast feeding (1989, 1990, 2003). Malta is a small island state in Europe with a population of 402,000, and had one major national hospital (St. Lukes with 890 beds in 20045), a small general hospital on the sister island of Gozo and three small private hospitals. It has yet to achieve the Ten Steps toward the Baby Friendly Hospital Initiative (BFHI), (Naylor, 2001). Despite unequivocal benets supporting breast feeding over articial milk feeds (Ball and Bennett, 2001; Heinig, 2001; Labbok, 2001; Picciano, 2001; Reynolds, 2001; Leon-Cava et al., 2002), Malta retains one of the lowest rates for breast-feeding initiation in Europe (Fig. 1). There is a steady decrease in breast feeding even in countries with initial high rates of breast feeding (Cattaneo et al., 2005; Taveras et al, 2005), and this situation is no different in Malta. Although recent reports have suggested a slow but steady improvement in breast-feeding initiation locally (Attard Montalto, 2002), this upward trend has not continued over the past three years (Table 1). Whilst in 1995, 45% of Maltese mothers were breast feeding (exclusively or mixed feeding) at the time

of discharge from St. Lukes Hospital, this gure rose to 56% in 2000 and to 61% in 2006. Furthermore, in 2001, only 18% of Maltese mothers were still exclusively or partially breast feeding nine months after birth, and this subgroup was just 17% in 2006 (National Obstetric Information and Statistics, 19952007). Most mothers who stop breast feeding do so within the rst few weeks after birth (Colin and Scott, 2002), and the instruction, encouragement and attitude of staff toward mothers whilst still in the hospital, in Malta as elsewhere (Meek, 2001; Cattaneo et al, 2005), is likely to have a signicant impact on future attrition rates in breast feeding. This study was designed to review the reasons for cessation of breast feeding in Malta within the rst six months after birth, in order to highlight the more important and potentially remediable problems.

Methods Participants
At the time of the study, St. Lukes Hospital was the only large state hospital in Malta and catered for approximately 8090% of all deliveries, totalling 35004000 per annum. A sample of every fourth new mother who chose to breast feed was selected sequentially after they gave birth in St. Lukes Hospital during the period June 2004 to March 2005. The choice of feeding was left entirely up to each individual mother, and only those who indicated that they had chosen to breast feed were invited to enter the study prospectively. Care was taken to ensure that any mother whose baby may have died/was stillborn was not enrolled into the study. Data were collected on maternal age, nationality, social class, mode of birth, parity and number of babies. Verbal consent was obtained prior to enrolment, whilst complete anonymity was

100 90 80 70 60 50 40 30 20 10 0
U K ly n ce ta al M d w ay an ni ai Ita an ua Po l Sp N or th Ire Fr la nd a

percentage of live births

Fig. 1 Breast-feeding rates at outset in selected European countries.

Table 1 Institution

Li

Percentage of mothers breast/mixed feeding in Malta and Gozo. Total births (%) 1995 3 days 1 mo 20 15 45 22 1998 3 days 46 31 64 47 1 mo 26 19 51 28 2000 3 days 56 35 75 56 1 mo 31 22 55 33 2006 3 days 61 37 77 60 1 mo 33 25 57 35

St. Lukes Hospital (state) Gozo hospital (state) Private hospitals National rate

81 9 10 100

45 29 55 45

Total live births: 1995 5234; 1998 4488; 2000 4205; 2006 3857; mo month.

ARTICLE IN PRESS
e8 respected and all subjects were free to exit the study without justication at any stage. S. Attard Montalto et al. applied to compare the curves, using SPSS 15.0 for Windows. The comparative likelihood for discontinuing breast feeding between those who gave supplementary feeds and those who did not was analysed using w2, taking pp0.05 to represent signicance in all analyses.

Data collection
All those entered into the study were contacted by phone at one postnatally and then again every month up to six months, or earlier if breast feeding was discontinued (i.e. at time 0, 1, 4, 8, 14, 18, 22 and 26 weeks). They were asked to provide details relating to the birth and their feeding practice, namely: mode of birth and birth experience; mode of feeding (breast, articial milk, mixed); time interval to rst breast feed; advice received relating to breast feeding; events impacting on feeding practice, and, where appropriate, reasons for discontinuing breast feeding. The latter were analysed and could be grouped into the following arbitrary but tightly dened categories: maternal choice; medical reasons; lack of information; social reasons; incorrect advice; and no reason. An explanatory denition for each category is presented in Table 2. Furthermore, the study compared all mothers who had introduced any supplementary bottle feeds with those who breast fed exclusively prior to discharge, and, for these two groups, analysed the likelihood of discontinuing breast feeding altogether within the six month study period.

Findings
A total of 3124 mothers with an age range of 1446 years (mean 28, median 29), of whom 94% were Maltese nationals, 52% primigravidae and 98.6% had singleton pregnancies, delivered during the 10month study period. Of these, 1650 (53%) chose to breast feed, and every fourth mother up to a total of 405 (aged 1942, mean and median 29 years; 99% Maltese, 51% primigravidae and 100% singleton pregnancies) were invited to enter the study. All 405 mothers had given birth to live, surviving babies and, of those, just two declined to participate after entry, with 403 successfully completing the six month study. The study group was representative of the total 1650 who chose to breast feed as well as the 3124 mothers who delivered during the recruitment period, as there was no signicant difference in maternal age, mode of birth, nationality, social class, parity and number of babies between the groups.

Statistical analysis
Attrition rates for breast feeding were analysed using non-parametric univariate KaplanMeier survival curves where the number of subjects who persisted with breast feeding (the survivors) was compared at the designated study time intervals between the study groups. The log-rank test was

Reasons for discontinuing breast feeding

The reasons for stopping breast feeding over the six month study were analysed and divided into the six groups: maternal choice, medical reasons, lack of information, social reasons, incorrect advice and no reason. Details of these replies are listed in Table 3. There was no difference in mean age,

Table 2 Category

Detailed explanation of study categories. Explanatory denition Breast feeding stopped due to mothers own personal choice, independent of any external inuence Breast feeding stopped following advice based on the medical condition of the mother or infant Breast feeding stopped as a result of mothers concerns or misconceptions without consultation for professional advice Breast feeding stopped after social pressure/inuence from relatives and/or society, against mothers personal choice Breast feeding stopped after advice from health professionals with no scientic basis, no clinical evidence-based reasoning, clear misconceptions and incorrect presumptions Breast feeding stopped without any explanation given

Maternal choice Medical reasons Lack of information Social reasons Incorrect advice No reason

ARTICLE IN PRESS
Incorrect advice: the most signicant negative determinant on breast feeding in Malta
Table 3 Reasons for stopping breast feeding. Number (total 403) 14 17 Breakdown of replies 8 work commitments; 2 pressure of other children; 4 decided ow inadequate Infant: 2 required urgent surgery overseas Mother: 11 breast complications (e.g. cracked nipples, difculties latching on); 1 unrelated medical disease; 3 depression Opted to discontinue (but did not seek advice): concerns/ misconceptions included: 1 previous breast feeding failure; 1 on special incompatible diet; 3 engorgement only relieved by stopping; 3 more calories in formula milk 6 advised to stop by relatives/family; 2 altered social circumstances 15 told breasts too small; 139 insufcient amount of milk; 9 milk of poor quality (no evidence); 9 poor weight gain (not validated); 5 maternal antibiotics (not contraindicated); 21 maternal illness (20 mild mastitis, 1 unconrmed anaemia); 2 infant colic 4 no reason provided

e9

Replies by category Maternal choice Medical reasons

Lack of information

Social reasons Incorrect advice

8 200

No reason

One hundred and fty-two women continued to breast feed until the study end point at six months.

500 400 number 300 200 100 0 0 1 4 8 14 18 study time point (weeks) 22 26

(4.2%) for a valid medical reason, eight (2%) due to lack of Information, eight (2%) for social reasons and four (1%) for no reason. The attrition in breast feeding for each individual group is shown on the KaplanMeier curve in Fig. 3. The difference in attrition rates between the groups and those who continued breast feeding was highly signicant (w2 301, d.f. 1, po0.001).

still breastfeeding stopped for other reasons

stopped due to incorrect advice

Differences in attrition rates by cause at different time points

As shown in Figs. 2 and 3, although few mothers stopped breast feeding due to medical complications, they did so very early after the birth; 15 of 17 women in this category had stopped breast feeding by the rst week. Similarly, those who did not seek advice or support (lack of information) also discontinued breast feeding relatively early, and all bar one had stopped by eight weeks. In contrast, those mothers who stopped through their own choice tended to do so toward the end of the study period, with four stopping at the halfway mark and 10 stopping by six months. Half of the entire study cohort, however, stopped after receiving incorrect advice, and the attrition rate of this group was distributed across the study period with a plateau toward the later phase.

Fig. 2 Breast-feeding attrition due to incorrect advice vs other reasons.

mode of birth, nationality, social class, parity and number of babies between the six groups. Breast-feeding attrition rates were high; from the initial 403 subjects, just 152 (37.7%) were still breast feeding at six months. On analysing the attrition rates by underlying cause, the most common reason for introducing articial feeding in 200 babies (49.6%) was on the incorrect advice of health professionals, usually without any medical indication. As shown in Fig. 2, incorrect advice was the most signicant factor that correlated with the overall decrease in breast feeding over the study period (po0.001). Just 14 (3.5%) mothers discontinued breast feeding due to maternal choice, 17

ARTICLE IN PRESS
e10 S. Attard Montalto et al.

Fig. 3 Survival functions.

Table 4

Breast-feeding attrition related to exclusive or supplemented breastfeeding in hospital. Exclusive breast feeding 326 145 Supplemented breast feeding 77 7 w2 32, po0.001 Total 403 152

Practice prior to discharge Breast feeding at discharge Breast feeding at six months

Relationship of discontinuing breast feeding with supplementary feeds in hospital

Finally, the study questionnaire conrmed that, from the total of 403 breast-feeding mothers, 77 (19.1%) had introduced supplementary bottle feeds prior to discharge from hospital; of those, 70 admitted to lacking in condence and discontinued breast feeding within the six month study period, regardless of the reason given for stopping breast feeding (Table 4). In contrast, 326 (80.9%) mothers breast fed exclusively whilst in hospital and 181 of those subsequently stopped breast feeding. The likelihood of subsequently stopping breast feeding was signicantly greater in those who introduced supplementary feeds whilst still in hospital (w2 32, po0. 001, Table 4).

Discussion
Breast-feeding rates in Malta have been relatively low since records have been collected over the past 15 years. Although more recent national statistics

would suggest an increasing trend in the rst month after birth (National Obsteric Information and Statistics, 19952007; Attard Montalto, 2002), this improvement is not sustained. Hence, although the national current rate of breast feeding at hospital discharge is around 60%, this study has reported that just 38% of babies were still breast feeding at six months; a gure well below the Healthy People 2010 goal of 50% at six months. The reasons for this stubbornly low rate is probably multifactorial and this study was designed to explore these reasons in the rst six months after birth, as reported by the mothers themselves. The reasons given for discontinuation of breast feeding were categorised into six groups that highlighted some interesting trends. Hence, although few mothers stopped breast feeding due to medical complications, these did so relatively early after the birth. This was not surprising as most of the medical complications were related to problems with the breasts and signicant systemic medical and/or surgical disorders in the baby or mother. By their very nature, these generally presented soon after birth or may have had a signicant impact on maternal health, either way

ARTICLE IN PRESS
Incorrect advice: the most signicant negative determinant on breast feeding in Malta increasing the chances of early weaning off the breast. Similarly, mothers who did not seek advice or support also discontinued breast feeding by 12 weeks, presumably as they represented a relatively unmotivated subgroup with an increased likelihood to give up. In contrast, those mothers who stopped by choice tended to do so toward the end of the six month study period. As many in this group stopped breast feeding due to work commitments, this suggests a general lack of support for breast feeding in the workplace, a critical determinant for successful breast feeding after 3 months or longer (Meek, 2001). Those who stopped after receiving incorrect advice from health professionals amounted to half of the entire group, but the attrition rate of this group was evenly distributed over the study period. This suggests that poor advice is being given consistently at all time intervals after birth, and no single unit (e.g. ward staff, primary doctor, community services, etc.) is solely responsible. In the incorrect advice group, 148 of 200 women reported that they were instructed to stop breast feeding due to insufcient or poor quality milk. Occasionally, this was justied by inadequate weight gain although, in all cases, these instructions could not be validated on assessing case notes. Indeed, although problems attributed to the mothers milk are very commonly offered to justify discontinuing breast feeding (Colin and Scott, 2002), genuine problem cases are extremely uncommon (Amir and Cwikel, 2005; McCann et al., 2007). Hence, addressing and eliminating this single misconception alone would signicantly increase breast-feeding rates in this group fourfold. In addition, this study, like Humenick et al. (1998), has conrmed the importance of establishing breast feeding in hospital as those mothers who introduced supplementary bottle feeds prior to discharge (regardless of the frequency) were far more likely to discontinue breast feeding altogether before the six month study endpoint. Others have shown that successful and sustained breast feeding is dependent on education, employment, marital status, attendance at antenatal classes and, importantly, access to appropriate advice postnatally (Deshpande and Gazmararian, 2000). The latter may be provided during routine community-based postnatal preventive visits run by a doctor or trained counsellor (Graffy et al., 2004; Laberere et al., 2005), and possibly augmented by a peer-support group (Ingram et al., 2005). At all times, sustained communication, encouragement and support for the mother would appear to be critical for successful breast feeding (Humenick e11

et al., 1998; Taveras et al., 2005). In Malta, this postpartum advisory role is presently shared amongst postnatal obstetric staff, midwives, primary doctors, in- and out-of-hospital paediatricians and community midwives/nurses. However, this study has clearly shown that, locally, the majority of these health professionals are not providing appropriate postpartum advice, resulting in the cessation of breast feeding in 50% of the study cohort. Furthermore, analyses of the reasons that staff provided to justify discontinuation of breast feeding (Table 2) would suggest a lack of a unied educational programme, inaccuracies and inconsistent advice on their behalf. The current negative attitude to breast feeding by those expected to be its strong proponents is of concern. Historically, breast feeding has been supported in the Maltese islands since prehistory with evidence of cults dedicated to suckling and use of wet nurses dating to several centuries BC (Savona-Ventura, 2004). This practice continued to be strongly encouraged throughout the 18th, 19th and 20th Centuries where much published data exist, and include information on the benets and possible contraindications of breast milk, and perils of articial feeding. Indeed, exclusive breast feeding was encouraged for at least nine months (before weaning) and, if possible, for the rst year (Borg, 1884, Manche, 1911), whilst wet nursing and midwifery were promoted in preference to articial feeds if direct feeding at the breast was not possible. Importation of condensed milk to Malta commenced in 1905 and escalated by the 1950s. This, together with signicant social upheaval, changing attitudes toward the womans role in society and women entering the workforce, resulted in a signicant shift toward more convenient articial milks for baby feeding and an initial breast-feeding rate of just 19.3% (Savona-Ventura and Grech, 1990). From the 1950s until the 1980s, articial feeding predominated and was associated with lower social class, multiparity, smoking and incorrect advice including insufcient milk (Savona-Ventura and Grech, 1990). Unfortunately, the converts to articial feeding from the 19501980 era are now grandmothers, and retain their antipathy for breast feeding. Further discouragement for new mothers comes from a still-prevailing cultural distaste for breast feeding in public, and sustained pressure from milk companies whose inuence on the public and health professionals remains strong. It is not surprising, therefore, that the trend toward articial feeds has prevails despite advice to the contrary and those in favour of breast feeding being eminent medical doctors, health authorities and the Catholic Church, the

ARTICLE IN PRESS
e12 predominant religion in the country (SavonaVentura, 2004). It has only started to reverse following a national multi-stakeholder educational campaign since the early 1980s to promote breast feeding. Notwithstanding, this study would suggest that the campaign still has some way to go before Malta attains breast-feeding rates comparable to other European countries. This study has clearly shown that incorrect advice from health professionals, often based on a lack of education, is of critical importance as a determinant of breast-feeding rates. Nevertheless, this problem could be eliminated with relative ease through an educational training programme for professional staff (Ekstro m et al., 2005; Ingram, 2006), which would not be particularly difcult to set up in Malta, given the small size of the country (122 square miles) with easy access to the great majority of the population. As highlighted by this study, this programme would need to be applied to all health professionals and at all stages in breast-feeding management. Although improved knowledge is not necessarily associated with breast-feeding promotion (Cantrill et al., 2003), this exercise could potentially improve the low national rates of breast feeding by an additional 50%. Moreover, this study has shown that if this measure was combined with a serious exercise in improving support for breast-feeding in the workplace (admittedly, that is more difcult to implement), this could translate into breast-feeding rates from the current 38% to 90% at six months. S. Attard Montalto et al.
Ball, T.M., Bennett, D.M., 2001. The economic impact of breastfeeding. Pediatric Clinics of North America 48, 253262. Borg, F., 1884. Kelmtein fuq is-sahha tal ulied. C Busuttil, Malta. Cantrill, R.M., et al., 2003. An Australian study of midwives breast-feeding knowledge. Midwifery 19, 310317. Cattaneo, A., et al., 2005. Protection, promotion of breastfeeding in Europe: current situation. Public Health Nutrition 8, 3946. Colin, W.B., Scott, J.A., 2002. Breastfeeding: reasons for starting, reasons for stopping and problems along the way. Breastfeed Review 10, 1319. Deshpande, A.D., Gazmararian, J.A., 2000. Breast-feeding education and support: association with the decision to breast-feed. Effects of Clinical Practice 3, 141143. Ekstro m, A., et al., 2005. Breastfeeding attitudes among counselling health professionals. Scandinavian Journal of Public Health 33, 353359. Graffy, J., et al., 2004. Randomised controlled trial of support from volunteer counsellors for mothers considering breast feeding. BMJ 328, 26. Heinig, M.J., 2001. Host defence benets of breastfeeding for the infant: effect of breastfeeding duration and exclusivity. Pediatric Clinics of North America 48, 105124. Humenick, S.S., et al., 1998. Breastfeeding and health professional encouragement. Journal of Human Lactation 14, 305310. Ingram, J., 2006. Multi-professional training for breastfeeding management in primary care in the UK. International Breastfeeding Journal 28, 9. Ingram, J., et al., 2005. Breastfeeding peer supporters and a community support group: evaluating their effectiveness. Maternal Child Nutrition 1, 111118. Labbok, M.H., 2001. Effects of breastfeeding on the mother. Pediatric Clinics of North America 48, 143158. Laberere, J., et al., 2005. Efcacy of breastfeeding support provided by trained clinicians during an early, routine, preventive visit: a prospective, randomised, open trial of 226 mother-infant pairs. Pediatrics 115, 139146. Leon-Cava, N., et al., 2002. Quantifying the Benets of Breastfeeding: A Summary of the Evidence. Pan American Health Organisation, Washington, DC. /http://www.patho. org/English/HPP/HPN/Benets_of_BF.htmS. Manche, L., 1911. Mard t-trabi. Moghdija taz-zmien 110, 3740. McCann, M.F., Baydar, N., Williams, R.L., 2007. Breastfeeding attitudes and reported problems in a national sample of WIC participants. Journal of Human Lactation 23, 314324. Meek, J.Y., 2001. Breastfeeding in the workplace. Pediatric Clinics of North America 48, 461474. National Obstetric Information and Statistics: Quarterly and Annual Reports, 19952007, Division of Health, Malta. Naylor, A.J., 2001. Baby Friendly Hospital Initiative: protecting, promoting and supporting breastfeeding in the twenty-rst century. Pediatric Clinics of North America 48, 475484. Picciano, M.F., 2001. Nutrient composition of human milk. Pediatric Clinics of North America 48, 5367. Reynolds, A., 2001. Breastfeeding and brain development. Pediatric Clinics of North America 48, 159172. Savona-Ventura, C., 2004. Breast versus bottle. A History of Infant Feeding in Malta. Association for the Study of Maltese Medical History, Proprint Ltd., Malta. Savona-Ventura, C., Grech, E.S., 1990. Infant feeding in Malta. Journal of Psychosomatic Obstetrics and Gynaecology 11, 107117.

Conclusion
This study highlights that, even in a small country with easy access to the population such as Malta, breast-feeding rates are low if the majority of health professionals are not sufciently committed to supporting breast-feeding mothers. The introduction of a clear hospital policy and compulsive education for all health professionals involved in maternity care within the framework of the BFHI is urgently required. Finally, this needs to be combined with a serious, community-based initiative to ensure a breast-feeding friendly workplace.

References
Amir, L.H., Cwikel, J., 2005. Why do women stop breastfeeding? A closer look at not enough milk among Israeli women in the Negev region. Breastfeed Review 13, 713. Attard Montalto, S.P., 2002. Breastfeeding in Malta: a review. Malta Medical Journal XIII, 4548.

ARTICLE IN PRESS
Incorrect advice: the most signicant negative determinant on breast feeding in Malta
Taveras, E.M., et al., 2005. Mothers and clinicians perspectives on breastfeeding counselling during routine preventive visits. Pediatrics 113, 405411. WHO, 1989. Protecting, promoting and supporting breastfeeding: the special role of maternity services: a joint WHO/ UNICEF statement. World Health Organization, Geneva /http://www.unicef.org/newsline/tenstps.htmS.

e13

WHO/UNICEF, 1990. The Innocenti Declaration on the Protection, Promotion and Support of Breastfeeding. Florence, Italy, 1 August /http://www.waba.org.br/inno.htmS WHO/UNICEF, 2003. Global Strategy for Infant and Young Child Feeding. World Health Organization, Geneva.