Regional differences in the prevalance of overweight, obesity and underweight among polish children and adolescents*

Różnice regionalne w występowaniu nadwagi, otyłości i niedoboru masy ciała wśród polskich dzieci i młodzieży

Aneta Grajda1, Zbigniew Kułaga1, Beata Gurzkowska1, Ewelina Napieralska1, Mieczysław Litwin2
1Public Health Division, The Children’s Memorial Health Institute, Warsaw
Head of Division: Z. Kułaga, MD
2Department of Research, The Children’s Memorial Health Institute, Warsaw
Head of Division: M. Litwin, MD, PhD

* Acknowledgments. The OLAF study was supported by a grant from Iceland, Liechtenstein and Norway through the EEA Financial Mechanism and the Norwegian Financial Mechanism, and the Ministry of Science and Higher Education of Poland; grant number: PL0080.


Aim: The aim of the study was to compare the prevalence of overweight, obesity, and underweight in children and adolescents from different provinces in Poland.
Materials and methods: Data from the recent, large, population-representative sample of school-aged children and adolescents (N=17573) OLAF study: “Elaboration of the reference range of arterial blood pressure for the population of children and adolescents in Poland” – PL0080 OLAF were used in the analysis. The survey was conducted in all provinces of Poland (N=16). Data were analyzed using SAS 9.2, EpiInfo 3.5.1 and LMSgrowth software packages. The frequency of overweight, obesity, and underweight were determined. For overweight and obesity (jointly) and underweight the odds ratio (OR) was calculated for gender and voivodship of residence. The body mass index (BMI) was standardized and expressed as a z-score. The statistical significance of differences between BMI z-scores depending on voivodship of residence was assessed by the t-test.
Results: Significant differences were found in the occurrence of overweight and obesity among the analysed regions, and voivodships with a lower (małopolskie, świętokrzyskie, lubelskie, and podkarpackie) and higher (mazowieckie) risk of overweight and obesity were identified. In case of underweight, łódzkie and podkarpackie (<0.040) provinces were higher risk areas, while mazowieckie (<0.001) had lower risk. The prevalence of overweight (including obesity) was higher compared to the prevalence of underweight in the majority of provinces (15 out of 16).
Conclusions: Analysis of regional differences in the prevalence of obesity, overweight and underweight in children and adolescents may point to the direction in which nationwide and local efforts should be made to reduce the inequalities stemming from nutritional status.

Key words: overweight, obesity, underweight, children, regional differences

Cel: Celem pracy było porównanie częstości występowania nadwagi, otyłości i niedoboru masy ciała wśród dzieci i młodzieży pomiędzy województwami w Polsce.
Materiał i metody: W pracy wykorzystano wyniki aktualnego, dużego badania OLAF zrealizowanego na próbie reprezentatywnej dla populacji dzieci i młodzieży w wieku szkolnym (N=17573) w Polsce pn. „Opracowanie norm ciśnienia tętniczego dla populacji dzieci i młodzieży w Polsce” – PL0080 OLAF. Badanie zostało przeprowadzone we wszystkich województwach w Polsce (N=16). Dane analizowano z użyciem pakietu statystycznego SAS 9.2, EpiInfo 3.5.1 oraz LMSgrowth. Wyznaczono częstość występowania nadwagi i otyłości oraz niedoboru masy ciała. Dla nadwagi i otyłości (łącznie) oraz niedowagi wyliczono iloraz szans (IS) w zależności od płci badanych i województwa. Wskaźnik masy ciała (BMI) dla wieku i płci standaryzowano i wyrażono w postaci z-score. Istotność statystyczną różnic średnich standaryzowanych BMI, w zależności od województwa, badano testem t.
Wyniki: Wykazano istotne różnice w częstości występowania nadwagi i otyłości w analizowanych regionach, wyznaczając województwa o mniejszym (małopolskie, świętokrzyskie, lubelskie i podkarpackie) i większym ryzyku wystąpienia nadwagi i otyłości (mazowieckie). W przypadku niedoboru masy ciała województwa łódzkie i podkarpackie należą do obszarów większego ryzyka (p<0,040), z kolei województwo mazowieckie wykazuje mniejsze ryzyko wystąpienia niedowagi (<0,001). Częstość występowania nadwagi (w tym otyłość) był wyższy w porównaniu do częstości niedoboru masy ciała w większości województw (15 spośród 16).
Wnioski: Analiza regionalnego zróżnicowania występowania nadwagi i otyłości oraz niedoboru masy ciała w populacji wieku rozwojowego może wskazać kierunek dostosowywania krajowych oraz lokalnych wysiłków w celu niwelowania nierówności związanych ze stanem odżywienia.

Słowa kluczowe: nadwaga, otyłość, niedobór masy ciała, dzieci, różnice regionalne

The prevalence of overweight and obesity among children and adolescents is increasing in many countries around the world, including European countries (1). Central and Eastern Europe are characterized by lower levels of overweight and obesity, whereas countries in the southern and western regions of the continent, by higher prevalence levels. In northern European countries, 10- 20% of children are overweight, in southern European countries, 20-40%. The observed trend of prevalence of overweight and obesity depending on the region is related not only to genetic differences among these populations, but also to environmental dissimilarities – differences in socioeconomic conditions, urbanization, culture, and health-promoting behaviours, including dietary habits and level of physical activity (2). Differences in the prevalence of overweight (including obesity) among regions of a country may reflect the diversity of obesity risk factors that occur at the level of the individual, on one hand, and at the level of the region in question, on the other. In Poland, determining the prevalence of excessive body weight is the subject of numerous analyses conducted by regional centres (3, 4, 5); direct comparison of the results of these studies is not possible because of the lack of uniformity of criteria adopted to define overweight and obesity, as well as the differences in the age of the studied subjects. The data collected in the nationwide, representative OLAF study, that was conducted to establish blood pressure reference ranges for this population of children and adolescents in Poland, also makes it possible to determine the prevalence of overweight and obesity in respect to the region and thus to identify areas with higher and lower risks of these conditions. Information about regional differences in the health of the paediatric population may help in recognizing obesity-promoting environments, and, consequently, in undertaking measures related to the allocation of resources to support organised social and political efforts aimed at not only stopping the rise in obesity and overweight, but at reducing their prevalence, which is currently a national health priority (6).

The objective of this study was to compare the prevalence of overweight, obesity, and underweight among children and adolescents from different provinces in Poland.

Data from a representative sample of elementary, middle school and secondary school pupils (N=17 573) obtained from 2007 to 2009 in the OLAF study, “Elaboration of reference blood pressure ranges for Polish children and adolescents” PL0080 OLAF were used for this analysis. The data from 16 pupils were excluded because measurements of either height and/or weight were missing, also excluded were the data from 7 girls who were pregnant. Finally the data from 17550 pupils (8386 boys, 47.8%, and 9164 girls, 52.2%) were used in the first analysis. In the analysis of differences in standardized average BMI values depending on province voivodship of residence, the data from 17 427 pupils aged 7-18 years (8327 boys, 9100 girls), were used, excluding those of 47 participants below the age of 6.5 years and 76 aged over 18.5 years. The survey was conducted throughout all provinces of Poland (N=16). Geographical representativeness of the OLAF study has been presented in another publication (7). Approval was obtained from the Children’s Memorial Health Institute Ethics Committee before the study commenced. Study participants were selected by two-staged random sampling in which the school was the primary sampling unit. Height and weight were measured in accordance with the OLAF study protocol by trained teams of researchers using standardized equipment. Average body heights and weights were calculated using the obtained measurements. The sample selection and measurement methods have been described in detail in a previous report (8). The calendar age of a child was calculated from the difference between the examination and birth dates. The result was expressed as a decimal and the calendar age in the middle of the range was assigned to the child (e.g., children in the age range of ?6.5 and <7.5 years were recorded as 7 years old). The results were grouped according to gender and one-year age categories according to voivodship of residence. The results were analyzed using the SAS 9.2, EpiInfo 3.5.1., LMSgrowth (downloaded from: tjcole/FileSharing1.html) software packages. Overweight and obesity were defined using the BMI (body mass index) cut-offs given in the international definition of obesity for children and adolescents (9), whereas underweight, on the basis of the BMI cut-offs proposed by Cole TJ et al. (10). The frequencies of overweight and obesity and of underweight were calculated. For overweight and obesity (jointly) and underweight the odds ratio (OR) and 95% confidence interval were calculated for children and adolescents aged 6-19 years depending on gender and voivodship of residence. Reference BMI values (developed using the OLAF Study data) for ages 7-18 years were available (11), therefore the BMI for age and gender was standardized for a given age range and expressed as a z-score. The statistical significance of differences in the frequency of overweight and obesity between the voivodship and the rest of the country was assessed using the Chi-square test for the entire studied sample and separately for gender. The statistical significance of differences in standardized average BMI values depending on voivodship of residence was assessed using the t-test. The results were considered significant at p<0.05; values p<0.1 were considered statistical trends.

The sample size, mean and standard deviation BMI of Polish boys and girls aged 6-19 years are shown in table I.
In Poland the prevalence of overweight and obesity in 6-19-year-old children and adolescents was 16.4% (18.7% and 14.3%, boys and girls respectively), and underweight – 12.0% total (10.0% – boys, 13.7% – girls).

Region of residence - voivodship
Analysis of the frequency of overweight and obesity (jointly) according to voivodship showed that pupils in the małopolskie, świętokrzyskie and lubelskie voivodships have a significantly lower risk of overweight and obesity, and statistical trend for lower risk was found in the podkarpackie voivodship. In turn, children and adolescents from the mazowieckie, lubuskie, dolnośląskie, and kujawsko-pomorskie voivodships had the highest frequency of overweight (including obesity) (>17.5%) in comparison with the rest of the country (tab. II). Figure 1 shows the frequency of overweight and obesity depending on the voivodship of residence (fig. 1). In each voivodship (excluding the świętokrzyskie voivodship), the frequency of overweight and obesity was higher among boys than in girls, this difference being significant in the dolnośląskie, lubelskie, małopolskie, mazowieckie, wielkopolskie, and zachodniopomorskie voivodships (p<0.05).
Comparison of the occurrence of overweight and obesity (jointly) and underweight showed that in the majority of voivodships, excessive body weight was more frequent, although there were those in which these frequencies were similar (the podkarpackie, lubelskie, and małopolskie voivodships), and in the świętokrzyskie voivodship, underweight was more frequent. The highest share of underweight was found in the łódzkie, lubuskie and podkarpackie (?14.0%), the lowest, in the mazowieckie and opolskie (<10.0%) voivodships (fig. 2), while in the łódzkie (14.8%; OR=1.30; 95%Cl: 1.09-1.55; p=0.003), podkarpackie (14.0%; OR=1.21; 95%Cl: 1.01-1.45; p=0.040) and mazowieckie (9.3%; OR=0.72; 95%Cl: 0.62-0.84; p<0.001) statistical significance was noted.
Analysis of BMI z-scores for age and gender showed that the voivodship of residence may have a significant influence on the BMI z-score of pupils aged 7-18. Average BMI z-scores were significantly higher than zero in the dolnośląskie and mazowieckie voivodships, whereas in the małopolskie, świętokrzyskie, lubelskie and podkarpackie voivodships, significantly lower than zero. Table III presents the BMI z-scores by gender and voivodship of residence.

The main result of this analysis is the demonstration that there are areas of high (the mazowieckie voivodship) and low (ma.opolskie, świętokrzyskie, lubelskie, and podkarpackie voivodships) risk of overweight and obesity (jointly). The observed differences were particularly significant in boys. Moreover, it was shown that in Poland, similarly as in other countries, the share of children with overweight and obesity is higher that of those who are underweight (12, 13). The results of the analysis relating to regional differences and identification of areas with a higher risk of overweight are in line with the results of the nationwide study of children and adolescents aged 7-17 years conducted in Poland in 1995 (14), in which it was found that the share of overweight (including obesity) was smaller in the eastern voivodships of Poland (including the former łomżyńskie voivodship) in comparison with the rest of the country (<9.5%). A similar study from 2005 conducted in five voivodships (kujawsko-pomorskie, lubuskie, małopolskie, podlaskie, and pomorskie) confirmed the existence of regional differences in the frequency of overweight and obesity and of underweight among adolescents, but that study pointed to different at-risk areas (15, 16). This discrepancy may result from the different age range of the studied population and different overweight and obesity cut-offs: 13-15 years vs 6-19 years; ?85 centile and <5 centile vs International Obesity Task Force cut-offs (for overweight and obesity) and thinness cut-offs proposed by Cole T.J. (9, 10). The differences in nutritional status according to the region of residence, has been in the interest of researchers in many countries. English studies on adolescents, young adults and adults demonstrate a significantly higher share of overweight and obesity in northern England in comparison with the south (17). Hawkins et al. point to eastern England as a region with the lowest risk of overweight and obesity in early childhood, also after correcting for selected risk factors (maternal prepregnancy overweight, maternal smoking during pregnancy, birth weight, breastfeeding duration, introduction of solid foods, daily television viewing), ethnic group, and sociodemographic factors (18). The podkarpackie, świętokrzyskie, lubelskie, małopolskie voivodships, in which the present analysis showed the lowest risk of overweight and obesity, are regions characterized by a higher percentage of rural residents, thus with an urbanization index of under 50% (19). This suggests that the level of urbanization may be of importance in identifying areas at risk of excessive body weight, but the trend between overweight and obesity in relation to urbanization varies in different countries. In Swedish studies, the reported average BMIs, as well as the frequency of overweight and obesity, by region of residence were higher among children from schools in rural areas and small towns than among the population of large cities; the differences in frequencies were 5.0% and 10.7%, for boys and girls, respectively.
The authors assessed the relationship between regional differences in the frequency of overweight with selected socioeconomic factors like membership of children in sports clubs and suggest that regional differences can not be explained by a correlation between the analyzed factors (20). In the mazowieckie voivodship, where a significantly higher risk of overweight and obesity was found, the per capita GDP is the highest in Poland, exceeding the national average by 60.1%; in turn, in three of the four voivodships in which a low risk of overweight and obesity was shown, per capita GDP is below 80% of the national average (21). It can not, therefore, be excluded that the socioeconomic status of a family and of the region of residence in general, may influence the frequency of overweight and obesity in children and adolescents in Poland. German studies have noted a higher frequency of overweight and obesity in children in areas belonging to the former West Germany (Munich) in comparison with children from the eastern part of the country (Dresden) (22), with the tendency and magnitude of increases in overweight and obesity frequencies in both areas becoming similar after the reunification of Germany (23). Causes of differences in nutritional status between regions/provinces can be searched also in the environmental factors, health behaviours of local communities, cultural influences and socioeconomic determinants, but this requires further research.
Knowledge of the geographical distribution of overweight and obesity is the starting point for the next steps in identification of at-risk areas and better understanding of the obesity-generating environment. Identification of the risk factors for overweight and obesity, among others, and analysis of the possible interrelationships are now possible through epidemiological studies currently underway. For example, in British studies in the Leeds metropolis, a relationship was found between a high density of fast-food outlets and BMI values in children aged 3 to 15 years (24). In Canadian (Calgary) studies, in turn, analysis of the relationship between access to parks and green recreation areas and the frequency of overweight and obesity in preschool children found no correlation (25). The incorporation of geographical information systems (26, 27) into the field of health care for the purposes of prevention of health risks, including obesity, may make it possible to (1) undertake effective measures aimed at identifying the risk factors of overweight and obesity, (2) monitor trends, and (3) undertake rational decisions as to the implementation of prophylactic measures.

Analysis of regional differences in the prevalence of obesity, overweight and underweight in children and adolescents may point to the direction in which countrywide and local efforts must be made to reduce the inequalities stemming from nutritional status.

1. Wang Y., Lobstein T.: Worldwide trends in childhood overweight and obesity. Int. J. Pediatr. Obes., 2006, 1, 11- 25.
2. Lobstein T., Frelut M.L.: Prevalence of overweight among children in Europe. Obes. Rev., 2003, 4(4), 195-200.
3. Mazur A., Małecka-Tendera E., Lewin-Kowalik J.: Overweight and obesity in primary school children from the Podkarpatian Region. Pediatr. Pol., 2001, LXXVI, 10, 743-748.
4. Chrzanowska M., Gołab S., Zarów R., Sobiecki J., Matusik S.: Trends in body adiposity and overweight and obesity occurrence in children and youth in Cracow in the last thirty years. Pediatr. Pol., 2002, LXXVII (2), 113-119.
5. Lebiedowicz K., Staskiewicz G., Torres K.: Obesity in 15 year old children in Lublin, Poland. Prevalence compared to other countries and associated factors. Nowiny Lekarskie, 2001, 70 Supl.I, 55-60.
6. Rozporządzenie Ministra Zdrowia z dnia 21 sierpnia 2009 w sprawie priorytetów zdrowotnych. Dz. U. 2009, Nr 137 poz. 1126.
7. Kułaga Z., Rózdzynska A., Palczewska I., Grajda A., Gurzkowska B., Napieralska E., Litwin M.: Percentile charts of height, body mass and body mass index in children and adolescents in Poland – results of the OLAF study. Standardy Medyczne 2010, 7, 690-700.
8. Kułaga Z., Litwin M., Tkaczyk M., Rózdzynska A., Barwicka K., Grajda A., Swiader A., Gurzkowska B., Napieralska E., Pan H.: The height-, weight-, and BMI-for-age of Polish school-aged children and adolescents relative to international and local growth references. BMC Public Health, 2010, 10, 109.
9. Cole T.J., Bellizzi M.C., Flegal K.M., Dietz W.H.: Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000, 320, 1240.
10. Cole T.J., Flegal K.M., Nicholls D., Jackson A.A.: Body mass index cut offs to define thinness in children and adolescents: international survey. BMJ 2007, 335, 194.
11. Kułaga Z., Litwin M., Tkaczyk M., Palczewska I., Zajaczkowska M., Zwolinska D., Krynicki T., Wasilewska A., Moczulska A., Morawiec-Knysak A., Barwicka K., Grajda A., Gurzkowska B., Napieralska E., Pan H.: Polish 2010 growth references for school-aged children and adolescents. Eur J Pediatr. 2010 Oct 23. (Epub ahead of print) DOI: 10.1007/s00431- 010-1329-x.
12. Wang Y., Monteiro C., Popkin B.M.: Trends of obesity and underweight in older children and adolescents in the United States, Brazil, China and Russia. Am. J. Clin. Nutr., 2002, 75(6), 971-977.
13. Lazzeri G., Rossi S., Pammolli A., Pilato V., Pozzi T., Giacchi M.V.: Underweight and overweight among children and adolescents in Tuscany (Italy). Prevalence and short-term trends. J. Prev. Med. Hyg., 2008, 49, 13-21.
14. Oblacinska A., Wrocławska M., Woynarowska B.: Frequency of overweight and obesity in the school-age population in Poland and health care for pupils with these disorders. Pediatr. Pol., 1997, LXXII, 3, 241-245.
15. Oblacinska A., Jodkowska M.: Obesity among Polish adolescents: epidemiology, eating behaviours, dieting, physical activity and well-being among overweight and obese adolescents. Instytut Matki i Dziecka. Warszawa 2007.
16. Oblacinska A., Tabak I., Jodkowska M.: Demographic and regional determinants of underweight in Polish teenagers. Przegl. Epidemiol., 2007, 61, 785-793.
17. Scarborough P., Allender S.: The North–South gap in overweight and obesity in England. Br. J. Nutr., 2008, 100, 677-684.
18. Hawkins S.S., Griths L.J., Cole T.J.: Regional differences in overweight: an effect of people or place? Arch. Dis. Child., 2008, 93, 407-413.
19. Population. Size and structure by territorial division 2008. Central Statistical Office 2009: gus/5840_1863_PLK_HTML.htm (access 2010.09.01).
20. Ekblom Ö.B., Oddsson K., Ekblom B.T.: Prevalence and regional differences in overweight in 2001 and trends in BMI distribution in Swedish children from 1987 to 2001. Scand. J. Public Health 2004, 32(4), 257-263.
21. Gross Domestic Product. Regional Accounts in 2007. Central Statistical Office 2009: rde/xbcr/gus/PUBL_rn_pkb_rachunki_regionalne_w_2007.
pdf (access 2010.07.07).
22. Liese A.D., Hirsch T., Mutius E., Weiland S.K.: Burden of overweight in Germany: prevalence differences between former East and West German children. Eur. J. Public Health 2006, 16(5), 526-531.
23. Apfelbacher C.J., Cairns J., Bruckner T., Möhrenschlager M., Behrendt H., Ring J., Krämer U.: Prevalence of overweight and obesity in East and West German children in the decade after reunification: population-based series of crosssectional studies. J. Epidemiol. Community Health 2008, 62(2), 125-130.
24. Fraser L.K., Edwards K.L.: The association between the geography of fast food outlets and childhood obesity rates in Leeds, UK. Health&Place 2010, science?_ob=ArticleURL&_udi=B6VH5-50KWG3X-1&_ user=1514857&_coverDate=07%2F23%2F2010&_rdoc=1&_ fmt=high&_orig=search&_origin=search&_sort=d&_ docanchor=&view=c&_acct=C000053432&_version=1&_
urlVersion=0&_userid=1514857&md5=c6ef95bfce 79b0022d79e2fe5b1324ba&searchtype=a (access 2010.09.03).
25. Potestio M.L., Patel A.B., Powell C.D., McNeil D.A., Jacobson R.D., McLaren L.: Is there an association between spatial access to parks/green space and childhood overweight/ obesity in Calgary, Canada? Int. J. Behav. Nutr. Phys. Act., 2009, 20, 6, 77.
26. Heywood I., Cornelius S., Carver S.: An introduction to Geographical Information System. Prentice Hall. 2nd ed. United Kingdom, 2002.
27. Rezaeian M., Dunn G., St Leger S., Appleby L.: Geographical epidemiology, spatial analysis and geographical information systems: a multidisciplinary glossary. J. Epidemiol. Community Health 2007, 61, 98-102.

Adres do korespondencji / Address for correspondence:
Aneta Grajda
The Children’s Memorial Health Institute
Public Health Division
Al. Dzieci Polskich 20, 04-730 Warsaw
tel.: (+48 22) 815-19-45
[email protected]

The OLAF Study Group
Medical University of Białystok: Agnieszka Rybi-Szumińska, Michał Szumiński, Katarzyna Taranta-Janusz, Edyta Tenderenda, Anna Wasilewska; Regional Children’s Hospital in Bydgoszcz: Beata Jasińska; Medical University of Gdansk: Piotr Czarniak, Dominik Świętoń, Przemysław Szcześniak; Jagiellonian University Medical College in Kraków: Monika Miklaszewska, Anna Moczulska, Katarzyna Wilkosz, Katarzyna Zachwieja, Iwona Ogarek; Medical University of Lublin: Marek Majewski, Aleksandra Sobieszczańska-Droździel, Izabela Szlązak, Paweł Szlązak, Małgorzata Zajączkowska; Polish Mother’s Memorial Hospital Research Institute in Łódz: Monika Pawlak-Bratkowska, Anna Półtorak-Krawczyk, Marcin Tkaczyk; District Hospital in Kedzierzyn-Kozle: Danuta Gmyrek; Poznan Medical University: Tomasz Krynicki, Jolanta Sołtysiak; Regional Children’s Hospital in Torun: Roman Stankiewicz, Sława Zbucka; The Children’s Memorial Health Institute in Warszawa: Robert Pietruczuk, Agnieszka Różdżyńska, Jan Szpor, Anna Świąder; Wroclaw Medical University: Jacek Kleszczyński, Magdalena Naleśniak, Anna Wawro, Irena Wikiera- Magott, Danuta Zwolińska; Medical University of Silesia: Piotr Adamczyk, Tatiana Augustyn-Iwachów, Beata Banaszak, Omar Bjanid, Katarzyna Broll-Waśka, Aurelia Morawiec-Knysak.