Correlation Between Flat Feet and Obesity in Children


This research analyses nine studies to investigate the effect of obesity on the foot structure of children. The findings of nine studies show a significant contribution from obesity on the foot structure. However, the studies may also have limitations the reader must consider before drawing meaningful conclusions. This research uses the PICO method to discuss the findings of the nine studies. The research focuses on the question: does obesity affect the foot structure of children?


There is evidence of a wide variation in the sample sizes considered ranging from 26 in Dowling et al. (2001), to 2083 in Chang et al. (2010). Riddiford-Harland et al. (2000), use a sample size of 124 while Villarroya et al. use 58 in (2008a), and 245 in (2008b). Chen et al. (2009) use the second-largest sample size of 1024, whereas Dowling et al. (2005) and Evans (2011) use 45 and 140 children respectively. Morrison et al. (2007) use a sample size of 200 children. The correlation between obesity prevalence to various anthropometric foot measurements determined the sample size. However, in Chen et al. (2009; 2010) obesity prevalence changes with age difference complicated the studies hence a large sample size facilitated stratification of sample into various ages to allow independent study on a specific age.

There was a significant variation in the age and gender distribution of subjects used in the experiments. In Riddiford-Harland et al. (2000), out of 124 children, 62 children formed the experimental group, whereas 62 were the control group. Villarroya et al. (2008a) use 33 male and 25 female children, representing 57% and 43% of the sample respectively. Villarroya et al. (2008b) used 130 male and 115 female children, representing about 57% and 43% of the population respectively. Chen et al. (2009) had a gender distribution of 53% (male) versus 47% (female). This gender distribution between studies by these two groups was a deliberate measure to allow easy comparison of the studies. Dowling et al. (2005) used 33.3% (male) versus 66.7% (female), whereas Evans (2011) used a 48.5% (males) versus 51.5% (females) representation. The gender distribution in these two studies depended on the prevalence of obesity in each gender.

Morrison et al. (2007) had four categories in their population in which females/males representation was 57.3%/42.7% in normal weight category, 56.7%/43.3% in the overweight category, 51%/49% in the obese category, and 50%/50% in the severely obese category. The rest of the studies did not specify the gender distribution of the population under study.

The population in the studies consisted of children between 7 and 16.5 years. The mean ages were 8.5 in Ruddiford-Harland et al. (2000), 8.1 in Dowling et al. (2001), 10.5 in Morrison et al. (2007), 12.75 in Villaroya et al. (2008a), 13.22 in Villaroya et al. (2008b), 9 in Chen et al. (2009), 9.5 in Chang et al. (2010), 8.5 in Evans (2011), and 9.2 in Dowling et al. (2005). The standard deviation in the ages of the participants in all nine studies was 5.52 years. It reflects that some studies used a distributed age aspect whereas others used a very narrow age difference.


The studies’ common goal was to establish the prevalence of obesity depending on the foot structure of children between 7 and 16.5 years. However, the researchers conducted each of the studies under different environs. This analysis notes factors varying between the studies for their significance in the outcome of the research. These factors include the duration of each study and the time of conduction, the authors of each study, and examiners.

Duration of research

Most studies do not specify the duration of the research. These studies failed to incorporate other factors that might contribute to the occurrence of flatfeet over a period, such as family history, weak arch, injury, and nervous system or muscle disease.

Riddiford-Harland et al. (2000), collected data once in 2000 and they have not collected the latest statistics. The study fails to establish the actual cause regardless of research showing a higher frequency of flatfeet in obese prepubescent children.

Dowling et al. (2001), also failed to establish the cause of the lowering of the MLA for the same reason. Failure by other studies to discuss study duration has a major impact on the outcome of the studies because research duration may define the completeness and viability of the study. Studies conducted within a short period fail to address the vital issue because of time constraints. However, adequate time of study would facilitate in-depth research to get a concrete conclusion on the causes and prevalence of obesity among children.

Authors of the studies

There is a variation in researchers in each of the nine studies. There were four authors in Morrison et al. (2007), whereas Riddiford-Harland et al. (2000), Dowling et al. (2001), Chen et al. (2009), and Dowling et al. (2005) studies had three researchers each. Five authors contributed in Villarroya et al. (2008a) and (2008b), and Chang et al. (2010) individual study, whereas Evans (2011) had a single author. The outcome of studies whose authority is under a single author may be more prone to the bias of the author because of the possibility to manipulate the result without consulting anybody else. In multiple authors’ studies, thorough consultations, and teamwork would ensure a conclusive study free from bias. However, the honesty and integrity among authors could give space for research ethics to eliminate such mentality on biasing research results.

Examiners in each study

None of the studies gave information about the people who conducted the examinations in each study. This information is important and the researchers should not have omitted it because the studies may suffer bias depending on the number of examiners and their specialization areas. People with prior knowledge of obesity were better at offering vital information for this study. However, examination of the affected persons could have an adverse effect because of trauma. As a result, the information given by these affected could ruin the results.


Apart from the external factors, there are also internal factors influencing the outcome of the studies. These factors include variations within the control and the experimental groups, testing methods, and the geographical location of the study.

Control and Controlled groups

Five of the nine studies mention the use of a control group in the experiment. However, only three of these define the size of the control group. Riddiford-Harland et al. (2000) used a control group of 62 non-obese children. Dowling et al. (2001) used 13 out of 26 children as the control group, whereas Dowling et al. (2005) used 45 obese children. Villarroya et al. (2008b) used two control groups, whereas Villarroya et al. (2008a) used a control group sample size presented in percentages. Other studies mentioned the presence of both obese and non-obese children. In addition, the treatment given to the control groups in different geographical locations was not specific. However, none of the studies made these considerations in their outcomes.

Testing methods

Although the studies made the same investigations, they used different methods to measure, test, and analyze the results. Most of the studies used a pedograph to measure the Footprint Angle (FA) and Body Metabolic Index to measure foot anthropometric measurements. An exception to this is Chen et al. (2009) which uses a digital tape measure and a 3D measuring probe to measure foot dimensions and Evans (2011) who uses Foot Posture Index (FPI-6) to measure foot posture. The different anthropometric measures used in this study complicated comparison with other studies.

The geographical location of the studies

Finally, the researched country may also influence how the control group is treated. Out of the nine studies, two were conducted in Taiwan Chang et al. (2010) and Chen et al. (2009). Other studies were in different countries in which the researcher gathered further information on this issue. It is feasible to conduct the studies over a wide geographical location and compare the outcomes to determine if there are any other factors based on the geographical location, apart from obesity, which affect the foot structure. Other causes of flatfeet emanate from another body malfunctioning that might be influenced by geographical conditions. For instance, family history as well as arthritis, and diabetic prevalence can dictate the occurrence of flatfeet during childhood. Therefore, children born in areas with a high case of these diseases might develop symptoms of flatfeet if the right measures fail to adhere in time.


The nine studies acknowledged the possibility of obesity affecting the foot structure in children. There was a close linkage between obesity and change in foot structure as revealed by some studies, although body weight cannot entirely be the major cause for flatfeet.

Riddiford-Harland et al. (2000), shows a significant difference between the Footprint Angle (FA) between the experimental group (obese children) and the control group (non-obese children); with the former having, a lower FA and a higher CSI score. Dowling, et al. (2001) and Villarroya et al. (2008a;2008b) used similar measurement tools in studying obese children.

Chen et al. (2009) suggested a higher frequency of flatfeet in obese children (56%) against overweight children (31%) and the normally weighing children constituting 27%. They further suggested that outcome is more likely to occur in boys than in girls and may vary with age, weight, and race (Chen et al. 2009).

Research by Dowling et al. (2005) also indicates a higher plantar contact area in obese children compared to non-obese children (Arch Index values of 0.23±0.05 against 0.17±0.08). This translates to more flatfeet occurrence in obese children. Chang et al. (2010), Evans (2011), and Morrison et al. (2007) also indicate a higher frequency of flatfeet among obese children compared to non-obese ones.

Other considerations

Studies this review sought to find the relationship between obesity and the foot structure. However, most of the studies have not considered other factors influencing feet structure to compare them with the effect of obesity. For example, most of the researchers conducted their studies in a single place. This limited the outcomes to only the areas of study without considering differences in geographical conditions and prevalence rate.

Generalization of the results from reviewed studies in this study may fail validity requirements based on the above limitation. Some of the outcomes also fail to cite enough evidence to conclude. For example, the conclusion by Morrison et al. (2007) that foot length and forefoot width increase with weight is just a postulation. Readers cannot use this and other similar studies to confirm that obesity is purely influential in the determination of the foot structure. Studies quoted in this literature review exhibited no concurrence of the age at which MLA is wholly developed.

Recommendation for Further Research

This review alone cannot be sufficient to make a meaningful conclusion about the causes of flat feet because of the uncertainties identified in the outcomes. The research recommends further in-depth research on all the possible contributors to flatfeet apart from obesity. This will be beneficial in devising ways of reducing flatfeet among children who might be a victim of the disease because of obesity. In addition, preventive measures can protect newly born children from developing flatfeet where possible.


Basing observations on the nine studies, this review concludes obesity is certainly a possible contributor to the foot structure in young children. In efforts to reduce flatfeet, the review postulates that controlling obesity may be a good option. However, readers must keenly consider the limitations of the research such as the tools and pieces of equipment used and their human safety when used in this population Villarroya et al. (2008a), took feet radiographic images that may expose children to X-Ray radiation. Radiations may have adverse effects over a long period to the children’s failure to take proper measures.

Demographical location is another aspect one must take into consideration. There is no universal definition of flat feet and thus different countries and different ideas may give different outcomes. There is no evidence that the same researchers performed the actual studies and one cannot draw meaningful conclusions that the process of the studies was accurate and can reflect the results presented.

Reference list

Chang, J, Wang, S, Shen, H, Hong, Y & Lin, L 2010, ‘Prevalence of flexible flatfoot in Taiwanese school-aged children in relation to obesity, gender, and age’, European Journal of Pediatrics, pp. 447-452.

Chen, J, Chung, M & Wang, M 2009, ‘Flatfoot Prevalence and Foot Dimensions of 5-13 Year old Children in Taiwan’, Foot and Ankle International , pp. 326-332.

Dowling, AM, Steele, JR & Baur, LA 2001, ‘Does obesity influence foot structure and plantar pressure pattern in prepubescent children?’, International Journal of Obesity, vol. 25, pp. 845-852.

Dowling, A, Steele, J & Baur, L 2005, ‘How does obesity and gender affect foot shape and structure in children?’, ISB XXth Congress- ASB 29th Annual Meeting, pp. 845-852.

Evans, AM 2011, ‘The Paediatric flat foot and general Anthropometry in 140 Australian children aged 7-10 years’, Journal of Foot and Ankle Research, pp.140.

Morrison, C, Durward, R, Watt, G & Donaldosn, M 2007, ‘Anthropometric Foot Structure of Peripubescent Children with Excessive versus Normal Body Mass’, Journal of American Pediatric Medical Association: United Kingdom, pp. 366-370.

Riddiford-Harland, D, Steele, J & Storlien, L 2000, ‘Does obesity influence foot structure in prepubescent children?’, International Journey Obesity, Australia, pp.120-124.

Villarroya, MA, Esquivel, JM, Concepción, T, Buenaféé, A & Moreno, L 2008a, ‘Foot structure in overweight and obese children’, International Journal of Pediatric Obesity, pp. 39-45.

Villarroya, MA, Esquivel, JM, Tomáás, C, Buenaféé, A & Moreno, L 2008b, ‘Assessment of the Medical Longitudinal Arch in Children and Adolescents with Obesity: Footprints and Radiographic Study’, European Journal of Pediatrics, pp. 32-48.