Dietary Intake, Gender and Activity Factors Influenced on BMI - Essay Example

Nutrition report Introduction There are many studies that have investigated the distinction in body composition related to gender. In most of thesestudies, women have been found to possess more body fat and percentage of fats than men. Body Mass Index (BMI) is the most common indicator or percentage body fat (Jackson et al., 2003). The differences of fat distribution between men and women is scientifically unknown , however, the variations in hormones, enzyme concentrations and hormone receptors is believed to play a causative role. Although the etiology of too much body weight is scientifically not not well known, dietary intake and energy metabolism are believed to play some role in weight progression and development. Nevertheless, overweight is most importantly attributable to excess intake of energy, which lead to not only a positive energy balance, but also an accumulation of body fats (Moore, 2000). Also, sedentary lifestyle is the other factors that can contribute to a gain in weight. BMI is a reliable and easily obtainable indicator of relative body size. At most times, BMI is directly associated with LDL and total cholesterol plasma concentrations. However, an inverse relationship has been reported between BMI and HDL-cholesterol (McNamara et al. 1992). On the other hand, the effect of gender on the association between blood lipid constraint and BMI has not been evidently recognized because most of the studies that have been conducted in this area are hardly consistent. Participating in physical activities is commonly used as a valuable way of preventing a number of health risks that is especially caused by heavy weight across all genders (Eaton and Eaton, 2003). There are a number of reports that have indicated that youth and children spend most of their leisure time in sedentary engagements such as playing video games of watching television (Moore , 2000). Mounting evidence reveal that sedentary behaviors, which are characterized by lack of physical activities, are attributed to increased risk of physical problems Sedentary behaviors have been proved to, be associated with physical activities, eating habits, and obesity when correlation designs are used (Gortmaker et al., 1996). Although these designs are valuable in determining associations between variables, experimental designs that entails the manipulation of sedentary behaviors are important in determining the causal impact of sedentary habits on energy consumption (Robinson, 1999). Epstein et al. (2005) used an experimental design within subjects that decreased and increased sedentary tendencies in young people, with the aim of determining the impact of sedentary tendencies on energy expenditure and energy consumption. In their findings, they established that there was an increasing sedentary behavior, which was related an increase in energy intake and a decrease in energy expenditure. In addition, they found that increase in sedentary behaviors was more in girls than boys, while overweight youths exhibited more decline in physical activities when sedentary behaviours were more, and vice versa (Epstein et al., 2005). Weight management is mainly managed through diet and physical activities. Weight loss can be achieved by following a recommended dietary strategy, which conventionally has been focused on reducing fat intake and decreasing portion sizes (Gortmaker et al., 1996). As an alternative method of weight management, a number of renowned health-related organizations have come up with a new strategy, which involves intake of foods with a low energy concentration. The implication of this method is that foods with low energy density in a standard amount of food provides relatively little energy (3). Ledikwe et al. (2006) carried out a study to establish the relationship between weight status and energy density, which aimed at determining how key indicators of energy density such as vegetable, fruits, and fat intakes were associated with obesity. In this case, energy density is the amount of energy that a given weight of food contains. Less energy per gram is contained in foods with a low energy density. This study will involve a variety of parametric and non parametric tests, which will be aimed at deciding the outcome of three hypotheses including: (1) Females have a greater percentage body fat compared with males; (2) energy and fat intake is strongly related to body fat; and (3) Sedentary subjects are more likely to be overweight compared with active subjects. METHODS This study will make use of a cross-sectional study design, whereby different groups of people with different variables of interest, but share some characteristics such as educational background or socioeconomic status will be sampled out to participate in the study. Therefore, the people who were selected to participate in this study differed Gender, sedentary lifestyles, and Body Mass Index among other characteristics. By doing this, the researcher intends to attribute differences in the dependent variables to the chosen independent variables. In this design, the researcher observes and records the information that is available from the population without manipulating the variables. The information that will be collected will be tested for statistical relevance – this will be used as a preliminary data to sustain further experimentation and research. The dependent variables include Body Mass Index (BMI) and percentage of body fat. On the other hand, the independent variables will include dietary energy intake, time spent watching television and gender. The researcher will use the simple random sampling method to select the participants that meet all the inclusion criteria. This method of sampling is preferred because of its simplicity. A sample of 53 participants is selected, and assumed to be a good representative of the whole population because the research ensured an unbiased selection. The materials that were used for anthropometric measurements include electronic weighing scales, speedometer, tape measure and bioelectrical impedance machine. Data collection was done by use of a General Practice Physical Activity Questionnaire (GPPAQ) and 24-hours dietary recall. GPPAQ is a tool used used by physicians to take a patient’s level of physical activity. This tool was adopted by London School Of Hygiene and Tropical Medicine in 2002. To ensure ethics, the Body Composition Practical Research Project was approved by the University Ethics Committee. Procedures To get permission to conduct the research, the researcher administered the consent form to the university authority and at the same time followed all the procedures that were necessary before starting such an activity in the institution. When the consent was granted, the researcher gathered all the necessary materials including questionnaires, stop watches and so on. The researcher invited the participants, one after the other, to fill the questionnaire while at the same time the observable variables were recorded accordingly. In this regards, all the details such as those from the stopwatches and work list as well as display monitors were described. The moment the researcher completed collecting the data from the survey, he started by capturing it in a Microsoft Excel spreadsheet. This data would then be analyzed using the IBM SPSS 11.0 package. The data would then be described using descriptive statistics as well as frequency tables. According to Hussey and Hussey (1997), descriptive statistics are ideal for reviewing and displaying of quantitative data, hence producing relationships and patterns that can be explained, something that is not possible with raw data. In order to check data errors and consistency, the researcher will produce frequency tables for the questions posed in the questionnaire. In addition, the researcher would work out the means, sample sizes, standard deviations and any other statistical measures that will help derive more meaning from the questions. DATA ANALYSIS As explained above, the data that was collected was captured in an excel worksheet for further analysis using SPSS among other methods. The statistical analysis that will be undertaken includes regression analysis, ANOVA, and t-tests analysis among other tests, with the aim of finding out whether the stated hypothesis can be supported or rather rejected. Most importantly, the statistical software will be used to conduct complex computations that can help establish whether particular models are statistically significant, or whether the model is fit in predicting the dependent variable. The following chapter will describe, in details, the results of all the tests that were undertaken. RESULTS The first hypothesis that was tested went as follows: Females have a greater percentage body fat compared with males. In this case, the Independent variable was represented by the Gender while the Dependent variable was represented by the percentage of Body fat. The following sections show the results of the regression analysis. Table 1: coefficients of gender and percentage of fats Model Unstandardized Coefficients Standardized Coefficients Sig. Coefficient Std. Error Beta 1 (Constant) 26.731 1.460 .000 Male -12.423 2.086 -.648 .000 Table 1 above shows the coefficients of gender and percentage of body fat. The male variable is represented by a dummy variable (1) while the female gender is represented by a dummy variable (0), therefore, the female gender is not featured in the table. From the coefficients column, the male variables accounts for -12.423, which indicates that for every additional male, the percentage of body fat is reduced by 12.423. The p-value for this coefficient is less than 0.05, hence the relationship is significant. In other words, this shows that there is significant evidence that Females have a greater percentage body more fat than males. Table 2: measure of goodness of fit Model R Square Adjusted R Square Std. Error of the Estimate Sig. F Change 1 .420 .408 7.4463 .000 In table 2 above, Adjusted R squred is 0.408, which means that the independent variables are not very good in predicting the dpendent vriable as a very high percentage (59.2%)er, of the output is attributable to errors. However, this model is siginificant since p-value is less than 0.01. T-test An independent t-test was also used to test the hypothesis that Females have a greater percentage body fat compared with males. The spss output for the independent t-test is as shown in the following tables. Table 3: Group statistics Male N Mean Std. Deviation Body fat % 0 26 26.731 9.2760 1 25 14.308 4.8554 The Male and Female variables are represented by numbers 1 and 0 respectively. The number of males that took part was 25 while the number of females was 26 as shown by the column labeled (N). The mean for females is 26.7 while the mean for males is 14.9. The standard deviation for females is 9.27 while that of males is 4.8. This means that the percentage of fat for the females is almost twice that of the males. Table 4: The independent sample test box Levenes Test for Equality of Variances t-test for Equality of Means F Sig. t Sig. (2-tailed) Body fat % Equal variances assumed 12.357 .001 5.956 .000 The Levene’s Test for Equality of Variances determines if males and females have about the same or different amounts of variability between percentage of body fat. In this case, the sig. value is less than 0.05, which means that the variability between males and females is not the same. Scientifically, this means that the variability of males and females with respect to body fat %  is significantly different. The Sig (2 – Tailed) value shows whether the means of males and females are statistically different. In this case, the Sig (2 – Tailed) value is less than 0.00, which means that there is a statistically significant difference between body fat % of males and females. This also leads to our conclusion that the differences between the means are potentially due to the body fat % manipulation and not due to chance. Because the mean of females is greater than that of the males, we can conclude that Females have a greater percentage body fat compared with males. The second hypothesis was as follows: Dietary energy intake is strongly related to body mass index (BMI). The Independent variable was Dietary energy intake while the Dependent variable was BMI. To establish if there is any significant statistical relationship between BMI and Dietary energy intake, we start by conducting a regression analysis with BMI as the dependent variable and Dietary energy intake as the independent variable. The following tables show the outputs of the regression analysis: Figure 1: scatter plot The scatter plot in figure 1 above shows that there is no evidence of statistical significance between dietary energy intake and BMI because there is no any trend between the two variables. The procedure below will be performed to further confirm absence of relationship between the two variables. Table 5: coefficients Model Unstandardized Coefficients t Sigh. Coefficients 1 (Constant) 23.769 18.931 .000 Dietary energy intake .000 -.473 .638 The coefficient of Dietary energy intake is less than 0.001, which means that change in BMI when dietary energy intake is increased by one unit is insignificant. Again, P-value of this coefficient is greater than 0.05, hence it is not statistically significant. Table 6: model summary Model R R Square Adjusted R Square Change Statistics Sig. F Change 1 .066 .004 -.015 .638 The model summaty above indicates the Adjusted R Square as -0.015. this value is negative indicating that the chosen model is extrenmely poor. It is also very far from one meaning that the prediction is mainly attributable to chance rather than the independent variable. The Sig, F Change is greater than 0.05 indicating that the model is not statistically siginificant. As such, we can conclude that enegy and fat intake is not strongly related to body fat. The thisrd hypothesis aimed at establishing whtheether Sedentary subjects are more likely to be overweight compared with active subjects. In this test, the Independent variable is Time spent watching TV while the Dependent variable is Body fat. The following table shows the results of the regression analysis. Table 7: Coefficients Model Unstandardized Coefficients t Sig. B Std. Error 1 (Constant) 16.120 3.811 4.230 .000 Time spent watching TV 3.014 2.165 1.392 .170 The coefficient for time spent watching television is 3.014 implying that for every hour spent watching television, the percentage of body fat is increased by 3 units. However, this prediction is not significant because p-value is greater than 0.05. Table 8 below choose that this model is not very good in predicting the dependent variable because Adjuted R squired is 0.018, which is very small percentage and which indicates that the biggest percentage of the observed results is attributable to chance and not the independent variable. Table 8: R squired Model R R Square Adjusted R Square Std. Error of the Estimate R Square Change 1 .195a .038 .018 9.597 .038 The implication of these results is that there is no enough evidence to claim that Sedentary subjects are more likely to be overweight compared with active subjects. DISCUSSION AND CONCLUSION We have found enough evidence to prove that Females have a greater percentage body fat compared with males. This is the case because the p-value was less than 0.05 while the average of body fat for females was almost twice that of the males. Nevertheless, the goodness of fit test revealed that this model is relatively poor in making such prediction. The second model that aimed at establishing whether Dietary energy intake is strongly related to body mass index (BMI) appeared to be a poor model and statistically insignificant. Finally, the third hypothesis that aimed at establishing that Sedentary subjects are more likely to be overweight compared with active subjects was scientifically insignificant and hence the relationship is outraged. The outcome of this study is consistent with a multiple of studies that have established that women have been found to possess more body fat and percentage of fats than men. This observation could be attributed to the variations in hormones, enzyme concentrations and hormone receptors between the two sexes, although this claim is not strongly supported by science. Although Epstein et al. (2005) established a strong relationship between sedentary behavior and body weight, this study has failed to significant evidence to support this observation, but essentially this could have been resulted from interference from other uncontrolled variables in this study. Likewise, the findings of this study in regards to the relationship between energy intake and weight level is in contrast with the findings of Ledikwe et al. (2006) on a similar study, which again can be attributed to bias introduced by variables that have not been put into account. Nevertheless, the findings of this study have added some weight to the existing knowledge on the fact that females are subjected to more body weight that their male counterparts, especially by attempting to link this observation to lifestyles such as the sedentary life whereby females spend more time seated than males who perhaps exercise their bodies more. This study has several limitations, which may affect the generalisability of the results. First, investigating whether the independent variables can predict the dependent variables is essentially complex because there are multiple factors that play a part in a variation of the dependent variables, and which may not have been controlled for. Furthermore, the sample size of 53 is very low and may fail to be a good representative of the whole population. These and other difficult situations may limit our understanding of the results, but they are basically weighed down and controlled by the research design. The other limitation which has a significant implication of this study is the existence of confounding variables, which cause effects on the dependent variables that the researcher had not accounted for - this has caused unexpected results in this study. This finding of this study can be used by physicians to encourage their patients to participate in physical activities in order to prevent health risks especially those caused by heavy weight across all genders. These findings also be used by researchers who would wish to conduct further study in a similar area. References Eaton, S.B. and Eaton, S.B. , 2003. An evolutionary perspective on human physical activity: implications for health. Comp Biochem Physiol A Mol Integr Physiol, 136, pp. 153–159. Einstein et al., 2005. Influence of changes in sedentary behavior on energy and macronutrient intake in youth. The American journal of clinical nutrition, 81 (2), pp. 361 – 366. Gortmaker, S.L., Must, A., Sobol. A.M, Peterson, K., Colditz, G.A. and Dietz, W.H., 1996. Television watching as a cause of increasing obesity among children in the United States, 1986–1990. Arch Pediatr Adolesc Med, 150, pp. 356-62. Hussey, J. and Hussey, R., 1997. Business research: a practical guide for undergraduate and postgraduate studies. London: Macmillan Business Ledikwe et al., 2006. Dietary energy density is associated with energy intake and weight status in US adults. The American journal of clinical nutrition, 83 (6), pp. 1362 – 1368. Moore, M.S., 2000. Interactions between physical activity and diet in the regulation of body weight. Proc Nutr Soc 59, pp.193–198. Robinson, T.N., 1999. Reducing children’s television viewing to prevent obesity: a randomized controlled trial. JAMA, 282, pp. 1561-7. Appendices 1. APPENDIX 1: BODY FAT IS DETERMINED BY GENDER (FEMALE HAVE BIGGER BODY FAT THAN MALE) TO INCLUDE REFERENCES. Regression Variables Entered/Removeda Model Variables Entered Variables Removed Method 1 Maleb . Enter a. Dependent Variable: Body fat % b. All requested variables entered. Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate Change Statistics R Square Change F Change df1 df2 Sig. F Change 1 .648a .420 .408 7.4463 .420 35.473 1 49 .000 a. Predictors: (Constant), Male ANOVAa Model Sum of Squares df Mean Square F Sig. 1 Regression 1966.890 1 1966.890 35.473 .000b Residual 2716.914 49 55.447 Total 4683.804 50 a. Dependent Variable: Body fat % b. Predictors: (Constant), Male Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 26.731 1.460 18.305 .000 Male -12.423 2.086 -.648 -5.956 .000 a. Dependent Variable: Body fat % FREQUENCIES VARIABLES=Male Female Bodyfat /STATISTICS=STDDEV MEAN MEDIAN MODE SUM SKEWNESS SESKEW /PIECHART FREQ /ORDER=ANALYSIS. Statistics Male Female Body fat % N Valid 53 53 51 Missing 11 11 13 Mean .47 .53 20.641 Median .00 1.00 18.000 Mode 0 1 16.0 Std. Deviation .504 .504 9.6786 Skewness .117 -.117 .717 Std. Error of Skewness .327 .327 .333 Sum 25 28 1052.7 Male Frequency Percent Valid Percent Cumulative Percent Valid 0 28 43.8 52.8 52.8 1 25 39.1 47.2 100.0 Total 53 82.8 100.0 Missing System 11 17.2 Total 64 100.0 Female Frequency Percent Valid Percent Cumulative Percent Valid 0 25 39.1 47.2 47.2 1 28 43.8 52.8 100.0 Total 53 82.8 100.0 Missing System 11 17.2 Total 64 100.0 requencies T-Test Notes Output Created 31-JAN-2013 09:10:51 Comments Input Active Dataset DataSet0 Filter Weight Split File N of Rows in Working Data File 64 Missing Value Handling Definition of Missing User defined missing values are treated as missing. Cases Used Statistics for each analysis are based on the cases with no missing or out-of-range data for any variable in the analysis. Syntax T-TEST GROUPS=Male(0 1) /MISSING=ANALYSIS /VARIABLES=Bodyfat /CRITERIA=CI(.95). Resources Processor Time 00:00:00.00 Elapsed Time 00:00:00.01 [DataSet0] Group Statistics Male N Mean Std. Deviation Std. Error Mean Body fat % 0 26 26.731 9.2760 1.8192 1 25 14.308 4.8554 .9711 Independent Samples Test Levenes Test for Equality of Variances t-test for Equality of Means F Sig. t df Sig. (2-tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference Lower Upper Body fat % Equal variances assumed 12.357 .001 5.956 49 .000 12.4228 2.0858 8.2312 16.6143 Equal variances not assumed 6.024 38.058 .000 12.4228 2.0621 8.2484 16.5971 1. Appenix 2: DIETARY ENGERGY INTAKE IS STRONGLY RELATED TO BODY MASS INDEX (LINK TO BODY FAT: FROM THE STUDENT’S DIET. [DataSet1] Statistics BMI Dietary energy intake N Valid 53 53 Missing 12 12 Mean 23.23 1924.74 Median 23.00 1744.00 Mode 24 192a Std. Deviation 3.952 940.386 Sum 1231 102011 a. Multiple modes exist. The smallest value is shown T-Test [DataSet1] One-Sample Statistics N Mean Std. Deviation Std. Error Mean BMI 53 23.23 3.952 .543 One-Sample Test Test Value = 0 t df Sig. (2-tailed) Mean Difference 95% Confidence Interval of the Difference Lower Upper BMI 42.799 52 .000 23.234 22.14 24.32 REGRESSION /MISSING LISTWISE /STATISTICS COEFF OUTS R ANOVA CHANGE /CRITERIA=PIN(.05) POUT(.10) /NOORIGIN /DEPENDENT BMI /METHOD=ENTER activity /SCATTERPLOT=(BMI ,*ZPRED) /RESIDUALS HISTOGRAM(ZRESID) NORMPROB(ZRESID). Variables Entered/Removeda Model Variables Entered Variables Removed Method 1 Physical activityb . Enter a. Dependent Variable: BMI b. All requested variables entered. Model Summaryb Model R R Square Adjusted R Square Std. Error of the Estimate Change Statistics R Square Change F Change df1 df2 Sig. F Change 1 .040a .002 -.018 3.987 .002 .081 1 51 .777 a. Predictors: (Constant), Physical activity b. Dependent Variable: BMI ANOVAa Model Sum of Squares df Mean Square F Sig. 1 Regression 1.291 1 1.291 .081 .777b Residual 810.888 51 15.900 Total 812.179 52 a. Dependent Variable: BMI b. Predictors: (Constant), Physical activity Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 23.256 .553 42.026 .000 Physical activity .000 .001 -.040 -.285 .777 a. Dependent Variable: BMI Residuals Statisticsa Minimum Maximum Mean Std. Deviation N Predicted Value 22.11 23.26 23.23 .158 53 Residual -6.055 15.044 .000 3.949 53 Std. Predicted Value -7.143 .141 .000 1.000 53 Std. Residual -1.519 3.773 .000 .990 53 a. Dependent Variable: BMI [DataSet1] Variables Entered/Removeda Model Variables Entered Variables Removed Method 1 Time spent watching TVb . Enter a. Dependent Variable: % of body fat b. All requested variables entered. Model Summaryb Model R R Square Adjusted R Square Std. Error of the Estimate Change Statistics R Square Change F Change df1 df2 Sig. F Change 1 .195a .038 .018 9.597 .038 1.938 1 49 .170 a. Predictors: (Constant), Time spent watching TV b. Dependent Variable: % of body fat ANOVAa Model Sum of Squares df Mean Square F Sig. 1 Regression 178.486 1 178.486 1.938 .170b Residual 4513.081 49 92.104 Total 4691.567 50 a. Dependent Variable: % of body fat b. Predictors: (Constant), Time spent watching TV Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 16.120 3.811 4.230 .000 Time spent watching TV 3.014 2.165 .195 1.392 .170 a. Dependent Variable: % of body fat Residuals Statisticsa Minimum Maximum Mean Std. Deviation N Predicted Value 19.13 25.16 21.08 1.889 51 Residual -13.948 23.352 .000 9.501 51 Std. Predicted Value -1.032 2.158 .000 1.000 51 Std. Residual -1.453 2.433 .000 .990 51 a. Dependent Variable: % of body fat Charts * Chart Builder. GGRAPH /GRAPHDATASET NAME="graphdataset" VARIABLES=TV Fat MISSING=LISTWISE REPORTMISSING=NO /GRAPHSPEC SOURCE=INLINE. BEGIN GPL SOURCE: s=userSource(id("graphdataset")) DATA: TV=col(source(s), name("TV"), unit.category()) DATA: Fat=col(source(s), name("Fat"), unit.category()) GUIDE: axis(dim(1), label("Time spent watching TV")) GUIDE: axis(dim(2), label("% of body fat")) ELEMENT: point(position(TV*Fat)) END GPL. GGraph Notes Output Created 31-JAN-2013 16:59:15 Comments Input Active Dataset DataSet1 Filter Weight Split File N of Rows in Working Data File 64 Syntax GGRAPH /GRAPHDATASET NAME="graphdataset" VARIABLES=TV Fat MISSING=LISTWISE REPORTMISSING=NO /GRAPHSPEC SOURCE=INLINE. BEGIN GPL SOURCE: s=userSource(id("graphdataset")) DATA: TV=col(source(s), name("TV"), unit.category()) DATA: Fat=col(source(s), name("Fat"), unit.category()) GUIDE: axis(dim(1), label("Time spent watching TV")) GUIDE: axis(dim(2), label("% of body fat")) ELEMENT: point(position(TV*Fat)) END GPL. Resources Processor Time 00:00:00.27 Elapsed Time 00:00:00.21 [DataSet1] * Chart Builder. GGRAPH /GRAPHDATASET NAME="graphdataset" VARIABLES=TV Fat MISSING=LISTWISE REPORTMISSING=NO /GRAPHSPEC SOURCE=INLINE. BEGIN GPL SOURCE: s=userSource(id("graphdataset")) DATA: TV=col(source(s), name("TV"), unit.category()) DATA: Fat=col(source(s), name("Fat"), unit.category()) GUIDE: axis(dim(1), label("Time spent watching TV")) GUIDE: axis(dim(2), label("% of body fat")) ELEMENT: point(position(TV*Fat)) END GPL. [DataSet1] Read More