Use of a child health surveillance instrument focusing on growth. A cross-sectional study

ABSTRACT BACKGROUND: Proper use of a child health handbook is an important indicator of the quality of care provided to children at healthcare services. This study aimed to evaluate the use of child health surveillance tool (by health professionals?), especially focusing on growth. DESIGN AND SETTING: Cross-sectional study carried out in the context of the Family Health Strategy in two municipalities in Paraíba, Brazil. METHODS: Three hundred and twenty-one children under five years of age from areas covered by health workers were included in the study. Mothers answered a questionnaire asking for information on sociodemographic characteristics. Growth charts, records of iron and vitamin A supplementation and notes on immunization schedules registered in the instrument were analyzed. In the case of children for whom the third version of the child health handbook was used, the association between completion of this handbook and sociodemographic characteristics was analyzed. RESULTS: All the parameters studied showed high frequencies of inadequate data entry, ranging from 41.1% for the weight-versus- age chart to 95.3% for the body mass index-versus-age chart. Higher frequency of inadequate data entry was found among children aged 25 months and over and among those living in areas of these municipalities with minimal numbers of professionals in the healthcare teams. CONCLUSIONS: The use of a child health handbook to monitor children’s growth in the municipalities studied appeared to be faulty. Data entry to this instrument was better at locations with larger healthcare teams.


INTRODUCTION
Healthcare for children comprises development of health promotion and preventive care actions through highly qualified assistance, in order to contribute to adequate child growth and development, and to maintenance of health and quality of life. 1 Primary healthcare services are the preferred gateway to the system, and links from this level to higher levels of care need to be established so as to offer comprehensive healthcare. 2,3[7] Use of a specific instrument for monitoring children's health is not exclusive to Brazil.
Other countries such as the United Kingdom, Sweden, Greece, Portugal, France, Canada, Indonesia, Japan, Australia, New Zealand and the United States (some states) also use similar technologies, thus showing the relevance of such tools. 8In Indonesia, for example, this instrument has been the principal means of keeping health records since 2004 and its use has been correlated with better child immunization coverage. 9oper use of a child health handbook, with complete and correct recording of information, not only is essential to the dialogue between healthcare professionals and users regarding its content but also can result in adherence and appreciation of the instrument by families, as well as shared responsibility for actions. 10,11With handbook data at hand, both parents and healthcare professionals have the opportunity to monitor the entire process of children's growth and development, thus enabling early identification of health problems such as malnutrition and delayed growth, so that these do not become irreversible. 4,7Thus, use of a handbook is an important indicator of the quality of care provided to children at healthcare services. 5,11However, studies on this topic have noted that a variety of circumstantial difficulties may exist and have emphasized the need for further research. 7,11,12e objective of the present study was to evaluate the use of child health surveillance, especially focusing on growth.There were more healthcare professionals (general physicians and pediatricians, nurses and nutritionists) for every 1,000 inhabitants, within the Brazilian National Health System (Sistema Único de Saúde, SUS), in municipality 1 than in municipality 2 at the time of this study, and the family health strategy teams are larger in municipality 1, with inclusion of nutritionists.

This
Children under five years of age were the study population.
For the present study, a convenience sample was used, obtained from the sample of a larger study of which this formed part, which had the aim of evaluating healthcare service users' perceptions about the quality of children's health care.Parameters obtained from another study evaluating the quality of healthcare offered to children within the family health strategy, according to patients' point of view, in the city of Montes Claros (Minas Gerais), 13 were used to estimate the sample size in the major study.
In each municipality, nine family health strategy teams were randomly selected to represent at least one third of the total number of teams.Each team contributed 18 children on average.For the total sample, two criteria were followed: 1.An intentional sample of all nursing services that were part of routine visits on the day of data collection, on a typical working day; and 2. Selection of 18 out of all the children referred to healthcare services by community health workers, according to their work routine, to be included in the study based on the number of children intentionally sampled.
At the end of the data collection period, the sample consisted of 321 children under five years old, of whom 153 were from municipality 1 and 168 from municipality 2. Of this total, five children (one from municipality 1 and four from municipality 2) attended the healthcare unit without taking their handbook or card.Therefore, there were 316 evaluations for this study.
Data collection was carried out in the healthcare units between July and December 2014.The field team was composed of healthcare professionals and students who had had previous experience of fieldwork, and the team was supervised by a trained professional.
The quality control for the study included: training and standardizing of interviewers, building an instruction manual and carrying out a pilot study in the city of Campina Grande, Paraíba.In the case of children who had the proper instrument at hand, i.e. the third version of the child health handbook (n = 259), the completion of the instrument was classified as adequate or inadequate.To this end, a scoring system that allowed comparisons between different locations and over time was used. 17In order to calculate the score, a value of one was attributed for items correctly entered and zero for items incorrectly entered, and the total score was expressed as the sum of the values for these items.Thus, variation from 0 to

RESULTS
Among the 321 children studied, the majority (75.1%) were under 25 months of age.A high frequency of households with four or more people (67.0%) was observed.The vulnerability of the population was also evident in the proportion of the families enrolled in the family grant program (59.8%) and the proportion with family income below two minimum monthly wages (27.8%) (Table 1).
Among the children who had the instrument at hand (316), 18% had inadequate versions (i.e.out-of-date versions).Regarding completion, all the parameters studied showed high frequencies of inadequate data entry, ranging from 41.1% for completion of the weight-versus-age chart to 95.3% for completion of the BMIversus-age chart (Table 2).Table 3 presents the analyses on the associations between sociodemographic characteristics and completion of the child health handbook.The results show that there were statistically significant differences regarding the variables of child's age and municipality of residence.
Through fitting the Poisson regression model, as shown in Table 4, it was observed that being a child under 25 months of age was a protective factor against inadequate data entry in the handbook (PR = 0.65; 95% CI: 0.53-0.80).Furthermore, children living in municipality 2, in relation to those living in municipality 1, had a higher frequency of inadequate data entry (PR = 1.55; 95% CI: 1.33-1.78).

DISCUSSION
It was observed that, even though nine years have elapsed since implementation of use of the child health handbook, 5 15.2% of the children in this study were still using the child card that preceded this.This result raises questions about the replacement of the card and the distribution of the handbook, as reported in a previous study. 189][20] These findings suggest that there is non-compliance with the minimum number of consultations recommended by the Ministry of Health for the first five years of life, 14 as also suspected by other researchers, 21 and/or suggest that monitoring of growth has not yet received proper attention among healthcare professionals. 5e abovementioned situation, especially the disuse of height and body mass index charts, is a cause for concern because of the nutritional status of Brazilian children, which is characterized by coexistence of significant prevalence of growth retardation and overweight. 22ese disorders, when untreated during childhood, can have irreversible consequences. 23This is why monitoring of growth is so essential. 7e higher percentage of completion of the weight-versusage chart, compared with the other growth charts in this study, has been reported in a similar manner by other researchers, in Brazil 18,24 and in other countries. 8This finding may be related to the fact that weight-versus-age curves have been used for a long time, i.e. since the implementation of the child card in 1984.It may also be associated with inclusion of new curves with concepts that are unfamiliar to professionals, such as representation as z scores. 6th regard to information on head circumference, the results from the municipalities studied here showed lower frequencies of correct data entry than was reported in other cities. 4,18 These low percentages need to be analyzed with caution, because deviations from the recommended measurements may be associated with neurological diseases, such as microcephaly and hydrocephaly, which need better evaluation and referral. 15ke in previous studies, 20 a low percentage of completion of the height-versus-age chart was seen in the present study.Nonavailability of equipment and lack of professional training for making height measurements are possible factors relating to this problem. 25stematic evaluation of the height-for-age index deserves greater attention at healthcare facilities because linear growth is an important expression of health and of the coverage of public policies. 26 this study, almost all BMI-versus-age charts were inadequately filled out.There is a lack of analyses in the literature in this regard. 23 evaluation of the medical records of children under two years of age in Cuiabá (Mato Grosso) showed that only 22.7% of the documents had registered BMI. 27Use of BMI is recommended from the time of birth onwards as a control measurement regarding overweight and adiposity, and as a predictor of obesity in adulthood. 23e greater adequacy of completion of data entry in the child health handbook among younger children that was observed in the present study has similarly been reported in previous investigations. 24This can possibly be explained by the higher frequency of scheduling of consultations during children's first months of life, as a reflection of the greater biological risk that is typical of this extremely young age. 23,24Moreover, during the first year of life, children receive most of their vaccines, thus implying that there is a need to return to the healthcare service more often, which gives rise to opportunities for use of this instrument. 20It should be noted that the Poisson regression with robust variance that was used in the present study to obtain this result has been suggested as an appropriate method for cross-sectional designs with binary outcomes, which should be based on calculation of prevalence ratios. 28nsidering the particular features of the municipalities analyzed, it is possible that the significant difference in the rate of inadequate completion of data entry in the child health handbook that was observed between these two municipalities was related to differences between them regarding the structure of larger teams and the training of professionals to work within primary healthcare and develop food and nutrition actions.Both of these municipalities had adequate availability of materials, inputs and equipment, including anthropometric instruments and handbooks.These conditions may favor interaction and minimize the influence of factors that hinder completion of the handbook, at least as far as food and nutrition issues are concerned.Experiences from other countries have shown that interactions among primary healthcare professionals promote improvement of growth monitoring and, in turn, the nutritional  status of children. 29However, further research in the Brazilian context should be conducted to address the possible matters of controversy.
The need for caution in interpreting the findings from the present study has to be emphasized, given that the responsibility for using the child health surveillance instrument is not restricted to healthcare professionals but should be shared with families. 23,30wareness needs to be raised among all parties involved regarding the importance of making proper use of the handbook and of effectively achieving its purpose. 5 interpreting the results from this study, some limitations should be considered.Firstly, it should be noted that data entry in the child health handbook was restricted only to some of the parameters included in the instrument.Furthermore, some factors relating to completion of the data entry in the handbook were not considered here, such as birth weight and other conditions inherent to child health, maternal care during the prenatal and postpartum period, the area of residence (urban or rural) and the parents' marital status.Furthermore, because of the attempt to preserve the work routine through use of a convenience sample, the possibility of bias in selecting the sample of children may be presumed.

CONCLUSIONS
Although the child health handbook is one of the most prominent aspects of public health policies in Brazil, the use of this instrument to monitor children's growth within the family health strategy in the municipalities studied is flawed.The precariousness of the records suggests that there is a need to raise awareness about the importance of the handbook and of larger healthcare teams.
was a cross-sectional study carried out in two municipalities in the state of Paraíba.These municipalities were chosen based on their similarities regarding factors such as geographical location (metropolitan region of the state capital, with access to the service network available in the city), degree of urbanization (almost 100%), sociodemographic indicators, economic resources and tradition regarding organization of primary healthcare services (the "family health strategy" covers nearly 100% of the population).Municipality 1 has a population of 57,944 inhabitants, of whom 4,596 are children under five years of age.This municipality has a healthcare system composed of 19 family health strategy teams.Municipality 2 has a population of 99,716 inhabitants, of whom 7,862 are children under five years of age.Its healthcare system is composed of 28 family health strategy teams.

Figure 1 .
Figure 1.Adequacy parameters considered in determining the adequacy of records in growth charts and in charts on preventive iron and vitamin A supplementation, according to the child's age.
included in multivariate analysis, in which P values according to the chi-square test were considered.In all statistical analyses, the significance level accepted was 5%.Analyses were performed using the Statistical Package for the Social Sciences (SPSS) software, version 13.0.This study was approved by the Ethics Committee of the State University of Paraíba, under protocol number 19689613.3.0000.5187.
Data were organized in spreadsheets and double-entered.The validate application of the Epi Info software, version 3.3.2,wasused to analyze data consistency, thus generating the final database used in the statistical analysis.Estimates of prevalence ratios (PR) and their respective 95% confidence intervals (CI) were used to analyze the association between completing the third version of the child health handbook and sociodemographic characteristics.For adjustment of confounding factors, we used Poisson regression with robust variance adjustment.Variables that were associated with the level of 25% (P < 0.25) in bivariate analysis were

Table 1 .
Sociodemographic characteristics of children under five years of age assisted by healthcare teams within the family health strategy in two municipalities in the state of Paraíba, 2014

Table 2 .
Child health surveillance instrument: version at hand and level of completion among children under five years of age who were treated by healthcare teams within the family health strategy in two municipalities in the state of Paraíba, 2014 CHH = child health handbook; *values that differ from the total were due to the version of the instrument used and/or the child's age; † completion level among children who had the third version of the child health handbook.

Table 3 .
Association between level of completion of data entry in the child health handbook and sociodemographic characteristics of children under five years of age who were treated by healthcare teams within the family health strategy in two municipalities in the state of Paraíba, 2014 P = P-values for chi-square test; PR (95% CI) = prevalence ratio (with 95% confidence interval).

Table 4 .
Adjusted association (Poisson regression) between inadequate completion of data entry in the child health handbook and sociodemographic characteristics of children under five years of age who were treated by healthcare teams within the family health strategy in two municipalities in the state of Paraíba, 2014