Puberty constitutes a critical period of life when adolescents are particularly prone to development of obesity(Reference Dietz1, Reference Dietz2). Several studies have suggested that up to 80 % of overweight adolescents will become obese adults(Reference Whitaker, Wright and Pepe3, Reference Daniels4). Thus, the incidence of overweight and obesity among adolescents is of public health concern and should be closely monitored.
A study carried out in Poland in 1995 on a population of two million schoolchildren and adolescents aged 6–17 years in almost all regions of Poland revealed gender and regional differences(Reference Oblacinska, Wroclawska and Woynarowska5). The present study population was divided into five age groups: 6–7, 9–10, 14–15 and 16–17 years. There were 439 262 adolescents in the age group of 14–15 years. School nurses measured their height and weight. The national charts of weight for stature for boys and girls were used. Pupils above the 90th percentile were identified as overweight and those above the 97th percentile as obese(Reference Oblacinska, Wroclawska and Woynarowska5).
In the age group of 14–15 years, the prevalence of overweight (including obesity) was 8·9 % for the whole sample, 7·8 % for boys and 9·9 % for girls (P < 0·001). The prevalence of obesity was 3·8 %; 3·5 % for boys and 4·1 % for girls (P < 0·001). The prevalence of overweight and obesity in the whole study population (all age groups) was higher in urban (9·4 %) than in rural areas (7·1 %; P < 0·001). The highest overweight rate was reported in the Podlaskie region and the lowest overweight rate was reported in the Kujawy region (11·0 % and 7·7 %, respectively).
The aim of the present study was 3-fold. First, to assess the prevalence of overweight and obesity among adolescents (13–15 years) in five regions of Poland based on both the International Obesity Task Force (IOTF) cut-offs and national references. Second, to compare these data to the data from the 1995 study (14–15-year-olds). Finally, to assess whether there were differences in overweight and obesity by gender and place of residence in the 2005 study.
Subjects and methods
The study design and participants
A cross-sectional study was carried out in the autumn of 2005. A two-stage sampling method was used. In the first stage, five regions from an overall sixteen regions in Poland were chosen. These regions were selected in a judgement sampling because of the results of the previous study from 1995. One region with the highest overweight rate in 1995 (Podlaskie), one with the lowest rate (Kujawsko-Pomorskie) and three regions with an average rate (Malopolskie, Lubuskie and Pomorskie) were selected.
The theoretical sample size was planned to be 10 000 subjects at the age of 13, 14 and 15 years (the pupils of grades I, II and III of lower secondary schools). The class size was assumed to be about thirty pupils, and the number of classes to be drawn was calculated [10 000: (3 × 30) ≈112] based on this.
In the second stage, a sample of 112 schools, out of all the secondary schools from the selected regions, was drawn. The number of schools drawn from each region was proportional to the total number of pupils in the region. The final sample consisted of 8065 pupils. The sample was smaller than planned because the average number of pupils in the class was lower (about twenty-five pupils) than expected. Out of 8384 pupils attending the selected classes, 319 did not participate due to their absence from school on the day of measurement. The study was approved by the Bioethics Committee of Research Institute of Mother and Child in Warsaw (Opinion no. 24/2004).
The measurement and definition of overweight and obesity
Trained school nurses conducted anthropometric measurements: height and weight of all the participating pupils. Weight was recorded to the nearest 0·1 kg using a mechanical column scale with participants standing without shoes in the minimum clothing possible. Height was measured to the nearest 0·5 cm using non-stretch tape fixed to a flat wall.
BMI was calculated. The definition of overweight and obesity used in the present study was based on the age- and gender-specific cut-off points corresponding to a BMI of 25 and 30 kg/m2 at 18 years of age (the IOTF cut-offs)(Reference Cole, Bellizzi and Flegel6). Adolescents were categorised as normal weight, overweight or obese according to these cut-off values.
To accommodate a direct comparison of the current results with the ones published from the 1995 study, the data from the present study were also analysed using the methodology of the 1995 study(Reference Oblacinska, Wroclawska and Woynarowska5). Pupils’ weight and height from the 2005 study were applied to the weight-for-stature tables used in the 1995 study(Reference Kurniewcz-Witczakowa, Niedzwiecka and Pietrzak7). The 90th and 97th percentiles were cut-off values defining overweight and obesity, respectively. Only the 14- and 15-year-old participants from the 2005 study were included for comparison with the respective population from the 1995 study.
Statistical methods
The Statistical Package for Social Sciences statistical software package version 14·0 (SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. The statistical significance of differences between boys and girls and differences in the prevalence rates of overweight and obesity between urban and rural areas and between regions were examined with the χ 2 test. Percentage agreement was calculated and the McNemar test was used to test agreement between the two methods of defining overweight and obesity.
Logistic regression was used to examine the association of region of residence with overweight and obesity. The regression analyses were conducted for the whole sample, and also stratified by gender. The cut-off point for statistical significance was P < 0·05.
Results
The sample included 8065 lower secondary-school students: 3980 boys and 4085 girls, aged 13–15 years. The mean age was 14·3 (sd 0·9) years. Of the entire sample 61 % (n 4958) of the students lived in rural areas and 39 % (n 3107) lived in urban areas. These pupils lived in five regions of Poland (selected regions and number of participants are shown in Fig. 1).
Fig. 1 Selected regions and number of students examined in each region (POM, Pomorskie; K-P, Kujawsko-Pomorskie; POD, Podlaskie; MA, Malopolskie; LUB, Lubuskie)
Using IOTF cut-offs, the overall prevalence of overweight (including obesity) was 12·5 % and the prevalence of obesity was 1·9 %. No significant age or gender differences were found for overweight or obesity in this sample (Table 1).
Table 1 Prevalence of overweight and obesity among adolescents aged 13–15 years in Poland stratified by gender and age, using IOTF cut-offs
IOTF, International Obesity Task Force.
The results of the analyses for adolescents aged 14–15 years only using the national reference used in the 1995 study are shown in Table 2. The prevalence of overweight (including obesity) and obesity was 11·1 % and 5·5 %, respectively. Significant gender differences were found. The prevalence of overweight and obesity in the analysed age group increased by 2·2 % and 1·7 %, respectively, between 1995 and 2005. The overall percentage agreement between the two methods (IOTF and national reference) was 97 % for overweight (McNemar’s test, P < 0·001), 98 % for girls (P = 0·80), 97 % for boys (P < 0·001) and 96 % for obesity (McNemar’s test, P < 0·001), 95 % for girls (P < 0·001), 96 % for boys (P < 0·001).
Table 2 Comparison of the prevalence of overweight and obesity among adolescents aged 14–15 years in Poland, in 1995 (n 439 262) and 2005 (n 5434), using weight for stature tables
The 90th and 97th percentiles were cut-off values defining overweight and obesity.
Only smaller, non-significant differences in the prevalence of overweight and obesity rates was observed for adolescents 13–15 years residing in urban and rural areas (Table 3).
Table 3 Prevalence of overweight and obesity among adolescents stratified by urban–rural area and regions in adolescents aged 13–15 years in Poland using IOTF cut-offs (n 8065)
IOTF, International Obesity Task Force.
*Urban–rural differences are statistically not significant.
†Regional differences overall; P = 0·003 and 0·012 for overweight and obesity, respectively.
‡Regional differences for overweight in girls (P = 0·002).
§Regional differences for obesity in girls (P = 0·023).
A significant regional difference in the prevalence of overweight was seen among Polish adolescents (P = 0·003), but when analysing separately for gender, the difference was significant only for girls (P = 0·002). The highest percentage of overweight pupils was seen in the sample from the Podlaskie region, and the lowest from Kujawsko-Pomorskie (Table 3).
Similarly, a significant regional difference in the prevalence of obesity was seen (P = 0·012), and as for overweight, the difference was significant for girls only (P=0·023). The highest percentages of obese adolescents were in the Lubuskie and Podlaskie regions, and the lowest in Kujawsko-Pomorskie (Table 3).
The above results were confirmed by logistic regression analyses adjusted for gender and age, revealing that adolescents in the Podlaskie region were more likely to be overweight than adolescents in Kujawsko-Pomorskie (OR = 1·30, 95 % CI 1·03, 1·66). Furthermore, pupils in the Lubuskie and Podlaskie regions were more likely to be obese than their peers in the Kujawsko-Pomorskie region (OR = 2·23, 95 % CI 1·24, 4·01 and OR = 2·11, 95 % CI 1·21, 3·69, respectively).
After stratification for gender, no association between regional residence and incidence of overweight was found for boys (Table 3), but the girls in Pomorskie and Podlaskie had a greater probability of being overweight than girls in Kujawsko-Pomorskie (OR = 1·53, 95 % CI 1·13, 2·06 and OR = 1·42, 95 % CI 1·01, 2·01, respectively). Boys in Podlaskie and girls in Lubuskie had higher odds of being obese than their peers in Kujawsko-Pomorskie (OR = 2·23, 95 % CI 1·04, 4·79 and 2·79, 95 % CI 1·23, 6·30, respectively).
Discussion
The present study provides the baseline to study future trends in overweight and obesity among adolescents using IOTF cut-off values in Poland. The prevalence of overweight and obesity was 12·5 % and 1·9 %, respectively. Comparison of anthropometric criteria for assessing overweight and obesity has shown significant differences between the two methods (IOTF and Polish national references). It seems that the application of IOTF’s cut-off values may result in the underestimation of obesity.
These results are in agreement with those of the WHO Health Behaviour in School-aged Children (HBSC) study, conducted in Poland in early 2006, despite the different methods of data collection (self-reported in the HBSC study and measured in the present study). Based on self-reported data on height and weight, the prevalence of overweight and obesity among 13–15-year-olds was 12 % and 1·7 %, respectively(Reference Mazur, Woynarowska and Kololo8). When comparing overweight and obesity rates across countries, it seems that overweight and obesity rates among adolescents in Poland are relatively low(Reference Lissau, Overpeck and Ruan9–Reference Wang and Beydoun11).
The results of the present study indicate that an urban–rural place of residence was not a risk factor for overweight and obesity in the study sample. The study from 1995 did find a rural–urban difference; overweight and obesity rates were significantly higher in urban areas(Reference Oblacinska, Wroclawska and Woynarowska5). Obesity has been more commonly observed in children and adolescents in urban areas in developed countries before 1980(Reference Dietz and Gortmaker12). However, studies from the USA and developed European countries indicate a reversal of the situation in which overweight and obesity in children and adolescents appear to be worse in rural areas(Reference Levin, Mayer-Davis and Ainsworth13–Reference Ekblom, Oddsson and Ekblom15). There is evidence that rural life presents special cultural and structural challenges in maintaining a healthy weight (higher dietary fat and energy consumption, lower level of physical activity, lack of sport facilities or further distance from them)(Reference Ekblom, Oddsson and Ekblom15, Reference Woynarowska, Jodkowska and Oblacinska16). It is highly probable, therefore, that a similar trend has been occurring currently in Poland. At the beginning of the 21st century, significant differences between urban and rural areas have not been observed, but one may expect a shift towards higher overweight and obesity prevalence in rural areas in the near future. It is necessary to closely monitor this trend.
The study shows regional differences in overweight and obesity among adolescents in Poland. The three independent studies carried out in Poland in the past decade (1995(Reference Oblacinska, Wroclawska and Woynarowska5), 2005 and 2006(Reference Mazur, Woynarowska and Kololo8)) have shown that regional differences in overweight and obesity rates among adolescents aged 13–15 years still persist. No differences were found in food habits (fruit and vegetable consumption, soft drinks and breakfast skipping), physical activity and sedentary behaviours among adolescents in these regions(Reference Mazur, Woynarowska and Kololo8). Thus, further investigations are called for in order to explain these differences.
Researchers from other countries have also reported on regional differences in overweight and obesity among adolescents, and it has been observed both in large(Reference Shields and Tjepkema17) and smaller countries(Reference Schober, Rami and Kirchengast18). Today’s challenge is to find the causes of these regional differences, especially in countries where minority ethnic groups only account for a small proportion of the total population.
In the present study, the observations were limited to only five regions. It is important to monitor the incidence of overweight in all sixteen regions of Poland. We believe that to be more effective, obesity prevention programmes should take into account these regional differences.
The main strength of the present study is that it represents the first Polish report of measured height and weight among adolescent school pupils applying the IOTF cut-off points for overweight and obesity. This allows for direct comparisons with similar measured data for the same age group from other parts of Europe. A limitation of the study, however, is that pupils from only five regions were included, making the comparison to the 1995 study complicated as that study included most regions of Poland. Furthermore, the statistically significant regional differences observed in the present study might be exaggerated, as we did not adjust for the school-level cluster effect (thus reducing the error term). Finally, data on socio-economic status and other background factors that might help explain the observed differences were not available from the present data set.
In conclusion, the observed prevalence of overweight and obesity among 13–15-year-old adolescents in Poland was seen as relatively low compared to that of other European countries. While no significant gender or urban–rural differences were seen, regional differences consistent with those observed a decade earlier were found. A modest increase in the prevalence rates of overweight and obesity (about 2 %) among 14–15-year-olds was seen from 1995 to 2005. It is recommended that overweight and obesity among adolescents in all regions of Poland be continued to be monitored using the IOTF reference criteria, and that further research be conducted in order to investigate the potential causes of the observed regional differences.
Acknowledgements
Source of funding: The present study was supported by the Nutricia Poland Research Foundation Grant no. 1/2005, and has been facilitated by the EU-funded HOPE project: ‘Health promotion through Obesity Prevention across Europe (the Commission of the European Communities, SP5A-CT-2006-044128)’. The study does not necessarily reflect the Commission’s views and in no way anticipates the Commission’s future policy in this area. Conflict of interest declaration: None of the authors has any conflicts of interest. Authorship responsibilities: M.J. was responsible for the study concept, and did the data analyses and manuscript preparation. A.O. was the director of the study and critically reviewed the analyses and the manuscript. I.T. participated in the study, provided expert advice on data interpretation and contributed to the editing of the paper. Acknowledgement: The authors thank Knut-Inge Klepp and Nanna Lien for valuable comments and support in the preparation of the manuscript.
