Supplementation and Change of Nutritional Habits for the Prevention and Treatment of Iron Deficiency Anaemia in Gaza Children: A Case Study

Iron deficiency anaemia (IDA) is one of the most severe and widespread nutritional disorders in the world. Children and pregnant women in resource-poor areas represent the most vulnerable groups. Iron deficiency impairs the cognitive development of children from infancy through to adolescence. It damages immune mechanisms and is associated with increased morbidity rates. Iron deficiency commonly develops after six months of age if complementary foods do not provide sufficient absorbable iron, even for exclusively breastfed infants (World Health Organization [WHO] et al, 2001). The WHO recommends universal iron supplementation when prevalence of anaemia is more than 40% (WHO, 2004). In the Eastern Mediterranean regions there is an endemic high prevalence of iron-deficiency anaemia (Verster, 1996) due to low total iron intake, low bioavailability (in many diets over 80% of iron is of non-haem origin) and high intake of inhibitors of iron absorption (unleavened bread and tea are severe inhibitors of iron absorption and are consumed in large amounts everywhere). Anaemia and stunting prevalence in Gaza have always been found to be very high in recent years (Abdeen, 2002; Rahim et al 2009).


Introduction
Iron deficiency anaemia (IDA) is one of the most severe and widespread nutritional disorders in the world.Children and pregnant women in resource-poor areas represent the most vulnerable groups.Iron deficiency impairs the cognitive development of children from infancy through to adolescence.It damages immune mechanisms and is associated with increased morbidity rates.Iron deficiency commonly develops after six months of age if complementary foods do not provide sufficient absorbable iron, even for exclusively breastfed infants (World Health Organization [WHO] et al, 2001).The WHO recommends universal iron supplementation when prevalence of anaemia is more than 40% (WHO, 2004).In the Eastern Mediterranean regions there is an endemic high prevalence of iron-deficiency anaemia (Verster, 1996) due to low total iron intake, low bioavailability (in many diets over 80% of iron is of non-haem origin) and high intake of inhibitors of iron absorption (unleavened bread and tea are severe inhibitors of iron absorption and are consumed in large amounts everywhere).Anaemia and stunting prevalence in Gaza have always been found to be very high in recent years (Abdeen, 2002;Rahim et al 2009).

Emergency situation
In the Gaza Strip the basic living conditions of all the inhabitants have deteriorated constantly in recent years, particularly after the "Cast Lead" operation in January 2009: the blockade and the closure of terminals for the movement of goods and people created a very tense situation which severely affected the wellbeing of all the inhabitants: 98% of private businesses closed and the unemployment rate increased to 48.8%, while 80% of the population lives below the poverty line and 79% is aid-dependent.The rate of food insecure households in Gaza has also increased to 75%, up from 56% before the Cast Lead operation.Furthermore, the growing inability of the population to consume iron-rich animal proteins and fresh fruit and vegetables, which contain the vitamins required for iron absorption, is bound to have a critical impact on the already high prevalence of mild and moderate iron deficiency anaemia in the Gaza Strip, habitually already about 20% higher than in the West Bank (WHO, 2009;World Bank, 2009).The National Nutrition Surveillance System Report (PNA, MOH, 2010) has confirmed a worsening level of anaemia and chronic malnutrition in the Gaza Strip, with an overall anaemia prevalence of 76.2% (45.5% in the West Bank) among children 9-12 months old and 58.6% (9.5% in the West Bank) among school children.Stunting prevalence in school children was 7.9% in the Gaza Strip and 4.4% in the West Bank.The national survey does not provide any data for children between the ages of 12 months and 5 years.

Present humanitarian intervention
Terre des Hommes Italy 1 (Tdh-It) and its Palestinian partner Palestinian Medical Relief Society 2 (PMRS) started operating with several humanitarian projects in Gaza in 2009, targeting pre-school children in a holistic way, where the prevention and treatment of anaemia have played a fundamental role.The projects were supported by the Italian Cooperation and other European donors.It is worth pointing out that the Tdh-It and PMRS projects were designed and implemented as (and in the framework of) humanitarian interventions and not study; nevertheless, the projects have also been supported by a strong monitoring and evaluation system that has provided us with a massive and structured quantity of information allowing us to present the projects' impact and data as a case study, although the possibility of bias in the sample selections has to be borne in mind.There is a strong need for a more evidence-based approach in humanitarian medical work and although a substantial body of knowledge has been accumulated regarding the effectiveness of interventions in acute emergencies, especially in refugee settings, the evidence base is much weaker for situations of protracted conflict with longer-term programmes in less controlled settings.(Banatvala and Zwi, 2000;Robertson et al, 2002;Roberts and Hofmann, 2004)   2. HOME VISIT (annex-2): details of the families of anaemic children were collected twice during home visits (coinciding roughly with the start and end of treatment) and concerned mother's knowledge, nutritional habits of all the children, iron treatment compliance and adoption of healthy life styles.In the project implemented in northern Gaza only some families received home visits and there was also a problem with coding of the children, so the link between data collected during screening and the home visit was not available; for this reason we analysed only data from the kindergarten visits carried out in southern Gaza.A total of 1,733 families of anaemic children screened in kindergartens were visited at home.ANNEX 2. Extract from the HOME VIST FILE (nutritional questionnaire and mother's knowledge assessment).

Nutritional health projects
Data were collected for 3,619 children (2,024 anaemics and 1,595 non-anaemic siblings) of these families concerning:  drug adherence, tolerance, storage and administration  mother's knowledge of anaemia and nutrition  nutritional habit (24-hour recall nutritional questionnaire) The second visit took place on average 111 days (SD=46) after the first visit.The home visit data included children on iron preventive treatment, not only anaemics.

Haemoglobin assessment
Blood samples taken at kindergarten were analysed using the Haemocue rapid test.Blood samples in clinics were tested using an aXE-2100D automated haematology analyser (Sysmex).

Definition of anaemia and malnutrition
Anaemia.Children with a haemoglobin level below 11g/dl were considered anaemic.Anaemia was defined as mild for a haemoglobin level of 10-11g/dl, moderate for 7-9.9 gm/dl and severe for less than 7gm/dl.Malnutrition.The software used to calculate Z score was "WHO ANTHRO, Software for Calculating Anthropometry, Version 2.0" and "WHO ANTHROPLUS".Segments of the population below -2 Z score (2SD) were considered as suffering from wasting (acute malnutrition, weight/height), underweight (weight/age) and stunting (chronic malnutrition, height/age).Segments of the population above 2Z of body mass index for age were considered as overweight (WHO, 2009).

Statistical analyses
Statistical analyses were performed using STATA software (Stata Statistical Software release 9.2, 2007;Stata Corporation, College Station, Texas).Uni-and multivariate binary regression and chi square test were used where appropriate.All statistical tests were two sided, and P values of < 0.05 were considered significant.

Main objectives
As already mentioned, the intervention was not conceived or performed as a study, thus the monitoring and evaluation (M&E) system was a tool for correct activity management and for evaluating the impact of the project, but it did allowed us to gather useful information:  for comparing the prevalence of anaemia and malnutrition before and after the project in the pre-school child population;  for identifying risk factors for anaemia;  for assessing compliance and tolerability of treatment and their association with lack of improvement;  for assessing change in the families' knowledge of anaemia and nutrition (mothers);  for assessing nutritional habits and evaluating changes promoted by intervention;  for evaluating anaemia prevalence 1 year after intervention (long-lasting impact);  for monitoring and evaluating anaemic children who did not improve during the first phase of the project, including identification of non-iron deficiency anaemia (e.g.thalassaemia).

Ethical approval
The Helsinki Committee of Palestinian Ministry of Health gave approval for publication of present paper.

RESULTS
3.1 Anaemia prevalence at screening 10,445 children were screened for anaemia between October 2009 and March 2011: 3,941 (37.7%) of them were screened at the PMRS paediatric clinics in northern Gaza, while the other 6,504 (including siblings) were screened at the 22 kindergartens in southern Gaza.51.6% (5,391) of the screened children were male.The mean age of the screened children was 39.7 months (SD=18.0),with no difference between the sexes.5.877 (56.3%) of the tested children were not anaemic and 4,568 (43.7%) were anaemic (HB level <11g/dl); 421 of the anaemics (4.1%) had a haemoglobin level below 9g/dl.The prevalence of anaemia was similar in males (44.0%) and females (43.4%, p=0.5) and strongly and inversely associated with age, as shown in Figure 1 and Table 1: anaemia had a very high prevalence in children below 24 months, peaking at 6-11 months (76.2%) and 12-23 months (72.2%).The percentage was much lower in older children (17.0% for children > 5 years old).
When considering only the under-5 population, the prevalence of anaemia was 49% (4,271/8,709), but it should be noted that children below 12 months of age were underrepresented and our sample did not adequately represent the under-5 population of Gaza.Prevalence of underweight and wasting (acute malnutrition) were low (around 2-3%), similar to the level registered in the normal healthy population according to WHO standards; prevalence of stunting was high (8.7%), and overweight was moderately higher (6.2%) than in the healthy population.

Anaemia improvement after intervention
4,077 of the anaemic children were monitored until a second haemoglobin test was performed, on average after 175 days (SD=43) of treatment.

Adherence to iron supplementation
In the first two phases of the project (South Gaza-1 and North Gaza) we were unable to associate improvement with good adherence to iron supplementation; this was due to the fact that almost all the mothers reported having given the iron as prescribed.A more careful investigation in a subsample of still anaemic children showed that in order to obtain more reliable answers:  the questions on adherence had to be more precise and more specific  the investigator was not to blame the mothers.For this reason a more precise and more sensitive data collection method was introduced in SOUTH GAZA-2; therefore, with regard to drug adherence, we present data limited to this project.Information on drug adherence was collected for all the 2,804 children enrolled, during each distribution at the kindergarten and during the home visit.Specific questions were asked, such as whether the drugs had been taken regularly (>5 days/week), irregularly (4-2 days/week) or not at all (< 1/week) during the previous week.To allow a better comparison we created a comprehensive index of drug adherence by combining all assessments performed, giving a score of 2 when the drugs were taken regularly, 1 when taken irregularly and 0 when not taken or the follow-up was missed.As shown in Table 5, there were 136 children who always took the drugs regularly, (score=10) and 31 who never took the medicine.We can further classify in 3 categories the level of adherence

Drug tolerance, storage and administration
In the previous subsection we presented data on drug adherence recorded for anaemic children; the data here include non-anaemic children undergoing preventive treatment (prophylaxis).There were very few reported complaints related to drug intake: around 2% in children with anaemia who received a higher dosage of iron and less than 1% for children on preventive treatment.
Vomiting and diarrhoea were the most common symptoms reported.Drug storage was adequate in more than 90% of the cases during the first visit, the figure dropping slightly declined at the second visit.A similar pattern was noticed with regard to correct drug administration.

Mother's knowledge of anaemia and nutrition
A total of 1,724 mothers answered a questionnaire on anaemia and nutrition (annex-2) twice during the two home visits.
The questions were open-ended and the social workers did not prompt any answers to them.
Considering the average number of good/correct answers given by mothers, it is clear for each section that there was a significant increase in knowledge (p-value paired t-test was always <0.0001).
The average number of good answers increased by 35%, from 7.7 to 10.4.The number of mothers improving their score was 1,205 (70%), while 247 showed the same and 272 a lower score.When we considered as having a "good basic knowledge" women with a score of > 10 with at least one good answer for every section, we found that the percentage of mothers with a good basic knowledge was only 20.2% at the first home visit, rising to 51.6% at the second home visit (p<0.0001).The level of "good basic knowledge" was strongly related to mother's education at the first home visit, ranging from 10% in women with primary education to 28% for the highly educated (p>0.0001).At the second visit no difference in good basic nutritional knowledge was seen between mothers with different standards of education (apart from the 12 illiterate subjects).This is particularly important since:  less educated mothers displayed a proportionally higher increase in knowledge than highly educated ones;  disparity was reduced at the end of the project;  we proved that the nutritional messages given were well understood even by the less advantaged, who are more in need, Fig. 5. Good-basic knowledge at the first and second home visit by mother's education level.
We found an association between improvement in anaemic status and mother's good basic knowledge at the second home visit: after adjustment for child's age, mothers with a good-basic knowledge were 24% more likely to have a child with improved status than mothers without a good basic knowledge (odds ratio=1.24,p=0.013).Furthermore, in a sub-sample of children with available data, we found a significant association (p=0.016) between improvement of anaemia with mothers' and fathers' participation in awareness sessions at the kindergarten.The percentages of improvement stratified for participation were:  78.3% for the 69 children whose mother and father both participated in the awareness session  69.1% for the 162 children whose mother participated in the awareness session  62.3% for the 533 children whose parents did not participate in the awareness session.

Change in children's nutritional habits
In order to appreciate a possible impact in nutritional habit we compared the results of the 24-hour recall nutritional survey done during the first home visit with those of the same survey repeated 3-4 months later.The overwhelming majority of children were having an average of 3 meals a day before and after the project.When considering the average number of portions consumed the previous day, we noticed a 25% increase in the consumption of fruit and vegetables, +29% for staple food, +19% for animal products and +25% for haem-rich animal food.The increase in fruit and vegetable consumption may not be an effect of the project since we recorded at the same time an increase in staple food that the project did not promote.Junk food consumption declined slightly, with chips and soft drinks up and candies down.Tea consumption decreased by 35%, and tea with meals decreased even more (-45%).At the same time water consumption increased.Similar results were recorded in the non-anaemic population.
As shown in Table 8, the children of parents who participated in awareness sessions seem to have had a better improvement of nutritional habits, particularly in terms of reducing tea consumption during meals: -72% versus -32% when considering cups; -55% versus -10% when considering the percentage of children drinking tea.-55% Table 8.Food consumption during previous 24 hours stratified for participation of parents in awareness sessions.

Children below 24 months of age at screening
The overwhelming majority of children were having an average of 3 meals a day and the percentage was stable.However, the percentage of children receiving fewer than 3 meals a day was reduced from 15.5% to 8.4.% (breast feeding was not considered).
As expected, the consumption of all types of food increased, with the exception of junk food, which remained stable.The consumption of soft drinks and candies, however, increased significantly.
Tea consumption was high, even among small children, half of whom had drunk it the previous day.There was an 18% decrease in quantity, and tea with meals decreased even more (-40%).At the same time water consumption increased.Similar values were found in the non-anaemic population (144 children).

WE COULD NOT FIND ANY ASSOCIATION BETWEEN IMPROVEMENT OF ANAEMIC STATUS AND CHANGE IN NUTRITIONAL HABIT
3.7 Anaemia prevalence after 1 year of intervention: a random sample of Phase 1 KGs children One of the biggest challenges of any medical intervention is to maintain the benefit obtained in the short term also in the long term.This is particularly important for nutritional supplementation intervention, such as treatment for iron deficiency anaemia: it is reasonable to have an improvement after iron supplementation, but what happens then?It is true that even a temporary improvement in anaemia at a crucial age with regard to growth can have long-lasting benefits, but our task was to assess the level of anaemia 1 year after the end of the intervention.

Methods
We randomly selected 178 children who were anaemic when enrolled in October 2009 during the SOUTH GAZA-1 project and had improved by the end of the project (May 2010), and we re-tested them in May 2011, one year after the end of the project.To evaluate improvement we compared anaemia prevalence and haemoglobin level at the three different time points using the pair t-test (comparison of each subject with himself/herself).
 there was still a fraction of children who regressed to anaemia after stopping supplementation, but their Hb levels were much higher than before project implementation.

Follow-up of still anaemic children
Children who were still anaemic after intervention underwent a more thorough clinical investigation, treatment and a longer follow-up.
Only 50 were diagnosed as having thalassaemia: the 0.48% of the 10.445 children screened and the 1.01% of the children found to be anaemic.
Out of the 296 children found to be still anaemic during follow-up screening at the end of the SOUTH GAZA-1 project, 159 (from 140 families) were still attending kindergarten in October 2010 and were enrolled in Phase 2 activities; the others had left, mainly to go to school.One hundred and twenty-nine of them were tested again; the other 33 refused to enter the new program.
After intensive counselling, the majority of children with no improvement in anaemic status were found to:  have had poor drug adherence and/or  have high tea consumption during meals and/or  drink large amounts of tea.Extra counselling was given to the mothers of these children.The results for the 129 children re-tested in October 2010 showed a substantial change from their previous status ( The lower prevalence noticed in our children over 5 years of age could be explained by the fact that there have been several instances in the last 2 years of ironfortified-food distribution in kindergartens, and we found that children receiving fortified food had a significantly lower level of anaemia.However the fortified food distribution, which was not implemented within a public health scheme, does not seem to be able to tackle completely the problem and several kids were found anaemic despite it. In addition to child's age and utilisation of fortified food, we found a significant association of anaemia with mother's poor education (an indication that low-social-status subjects are more vulnerable) and with stunting (not surprising since anaemia and stunting both reflect poor quality nutrition).
It is worth noting that children aged between 12 and 48 months, particularly those below 24 months, have very high prevalence of anaemia but are particularly difficult to reach because they do not attend the health clinic regularly (the vaccination program ends in the first year) or go to kindergarten: specific actions to target them should be implemented because anaemia can have very negative consequences for them (Walter, 2003).
Universal growth monitoring at least once a year for all under-5s, including haemoglobin level testing, could be one of the measures to take in the Palestinian context, where medical facilities and health workers are readily available and could easily provide this service.This would also provide the opportunity to monitor and contrast stunting, an age-old chronic problem, at an early stage, and also obesity, the new rampant one.

Anaemia improvement
Overall performance for anaemia was very good.Nearly 70% of the children treated were cured within 4-6 months.A review of the literature shows that this rate is in line, if not better, with what has been achieved in specific studies elsewhere (Rosadoet al, 2010) or in the region where children received iron daily or weekly (Faqih et al, 2006;Tavil et al, 2003), but these were clinical trials with a small number of participants and effectiveness in the field is always more complicated.One good result seen is an impact of the supportive counselling, including home visits: this is backed up by the fact that mothers who gained a better knowledge had an additional 24% chance of having a child who recovered from anaemia.The effect of well-motivated parents -something rarely studied in trials, where almost all participants are well motivated -is confirmed by the better results achieved when parents participated in awareness sessions.
Of course we have to be careful in considering this difference as a result of the awareness sessions, since it is likely that we had a strong selection bias: parents most interested in "nutritional" topics even before the project probably attended more awareness sessions, and they were also more attentive in monitoring their children's adherence and nutritional habits.
The lower rate of improvement in children below 24 months of age confirms the high vulnerability of this age group.The use of more palatable iron with fewer side effects, such as that used in our projects (Toblli et al, 2007), can explain the relatively good adherence and impact: and the link between adherence and improvement was clearly proven during our interventions.
As found in another study (Zlotkin et al, 2003), further supplementation is not needed to maintain non-anaemic status in most children previously treated for anaemia: almost all the children who recovered from anaemia were not anaemic 1 year later.

Nutritional habits: knowledge and practice
Knowledge of anaemia and nutrition was quite low, particularly for less educated mothers, but health education achieved a substantial increase in the level of knowledge, particularly in the less educated.
The 24-hour nutritional questionnaire was designed as a tool for diagnosis and family counselling, not for gathering information, but it provided some interesting data on food consumption:  high prevalence of junk food in the under-5s, as also noticed for school children (PNA MOH, 2011);  low consumption of fruit and vegetables;  high consumption of tea, half of the children below 24 months of age having drunk some during the previous day.During the second home visit we found a consistent decrease in tea consumption in all age categories, particularly during meals, but we were unable to establish a real reduction in junk food consumption and there was only a minimal increase in the intake of fruit and vegetables.
It is important to point out that food consumption was reported by the mothers, so the reported "good change" should be treated with caution because this could in part be the result of the mother's desire to give the (counselling) interviewer a good impression.Our data confirm that increased knowledge did not immediately result in improved feeding habits, a constraint found in many interventions that try to address chronic nutritional problems such as obesity in children (Branca et al, 2007).Little change in nutritional habits and the weakness of having only two monitoring measures of habits could explain why we were unable to establish a significant association between change in nutritional habit and anaemia improvement.

Acknowledgment
The article preparation was not funded by the donors in any way, and all its contents are published under the responsibility of Terre des Hommes Italia and the Palestinian Medical Relief.Thanks go to the staff of PMRS, Tdh-It and kindergarten who implemented the projects
Figure 3 below shows that:  drugs were taken regularly by 57% of the children after one month, the percentage decreasing constantly to 45.7%;  the percentage of children who took drugs irregularly rose with time from 21.7% to 32.8%;  the percentage of children not taking drugs or not showing for follow-up increased with time.The major reason mentioned for not taking drugs regularly were careless mother and/or child's refusal.

Fig. 4 .
Fig. 4. Mother's knowledge of anaemia and nutrition at the first and second home visit.

Table 1 .
Summary of Tdh-It/ PMRS nutritional projects in Gaza.

Table 2 .
Prevalence of anaemia and malnutrition by age group.Supplementation and Change of Nutritional Habits for the Prevention and Treatment of Iron Deficiency Anaemia in Gaza Children: A Case Study 219As shown in Table3, anaemia was also associated with: www.intechopen.com

Table 3 .
Anaemia prevalence by age group and other variables.

Table 4 .
Anaemia status before and after treatment.Supplementation and Change of Nutritional Habits for the Prevention and Treatment of Iron Deficiency Anaemia in Gaza Children: A Case Study 221

Table 5 .
Adherence scores for anaemic children.

Table 6 .
Drug tolerance, storage and administration among anaemic and non-anaemic children.

1 Children over 24 months of age at screening
Since children below 2 years of age, who are at the weaning stage, can have a substantial change in their nutritional habit independently of the project, we analysed in a stratified way the results for children below 24 months of age (where considerable change is naturally expected) and those for older children (where changes can reflect project impact).

Table 7 .
Food consumption during previous 24 hours as recorded during first and second home visit for children over 24 months of age, anaemic (7a) and not anaemic (7b).

Table 9 .
Food consumption during previous 24 hours as recorded during first and second home visit for anaemic children below 24 months of age.

62 g/dl 10.91 g/dl <0.0001Table 11 .
Haemoglobin level and anaemic status of children not improving in the first phase.