View Article PDF

Dear SirThe recent editorial by Kelly and Flint1 is a classic that should be included in the next edition of the book ‘How to Lie with Charts' by Gerald Everett Jones.2 Their graph is an excellent teaching example of turning good data into misleading information. A first-year student would quickly identify the glaring errors.A close look at the x-axis of their graph shows inconsistent year intervals of eight years (1989–1997), five years (1997–2002, 2002–2007, 2007–2012), one year (2012–2013), three years (2013–2016), and four years (2016–2020). The so-called projections of obesity rates compare two different models: a linear model for New Zealand and a log model of the USA using the different time intervals. It is also difficult to decipher the authors' statement ‘that the rate of obesity in New Zealand rises by 1% each year' from the data presented in their paper.Graphing the data correctly (Figure 1) still suggests an increase in the obesity rate, but not the same dramatic increase suggested by Kelly and Flint. The correct graph provides less support for the authors' proposition that in 5 years the New Zealand rate will exceed the rate in the USA. Figure 1. National adult obesity prevalence trends, New Zealand and USA Notes: 95% confidence intervals included on data points. New Zealand data is from comparable and consistent surveys.Sources: 1997 New Zealand National Nutrition Survey, New Zealand Health Survey (2002/03, 2006/07, 2011/12, 2012/13, 2013/14)3,4; NHANES data5; OECD Health Statistics 2014.6 While the authors have nobly promoted the need for increasing bariatric operations, their message is severely compromised by the inclusion of misleading and inappropriate data. Undoubtedly, obesity is, and will continue to be, a major public health issue in New Zealand.Kelly and Flint have done a major disservice to those involved in developing prevention strategies by the lack of rigor in the presentation of data. It is a prime example of producing unreliable facts from reliable data.Barry BormanAssociate ProfessorKylie MasonPrincipal AnalystCentre for Public Health Research Massey University-Wellington Wellington, New ZealandReferences1. Kelly S, Flint R. New Zealand is far behind Australia in offering weight-loss surgery [editorial]. N Z Med J 2015;128(1408):10–2. http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2015/vol-128-no-1408/64082. Jones GE. How To Lie With Charts. LaPuerta; 2007.3. Ministry of Health. The Health of New Zealand Adults 2011/12: Key findings of the New Zealand Health Survey. Wellington: Ministry of Health; 2012.4. Ministry of Health. Annual Update of Key Results 2013/14: New Zealand Health Survey. Wellington: Ministry of Health; 2014.5. Fryar CD, Carroll MD, Ogden CL. Prevalence of Overweight, Obesity, and Extreme Obesity Among Adults: United States, 1960–1962 Through 2011–12. 2014. http://www.cdc.gov/nchs/data/hestat/obesity_adult_11_12/obesity_adult_11_12.pdf6. OECD. OECD Health Statistics 2014. http://www.oecd.org/health/healthdataResponse from Kelly and FlintThis is our reply to the questions about the diagram.The x-axis data values are not linear as the data points are projected directly on the graph. However this does not affect the statistical modelling, it only reflects the graphical projection. Your graph projects both New Zealand and USA obesity rates versus time as being linear. This is true for New Zealand, as it has linearly increasing obesity rates of 1% every year since 1989. However it is a fact that the USA has had a stable obesity rate for the last 10 years as published by Cynthia Ogden from the CDC1 and also published by the OECD.2 Therefore USA has a logarithmic trend line as depicted on the editorial paper.From a simplistic standpoint if New Zealand is currently 30% obese and increases by 1% per year, and USA is currently 35% obese and stable, then it will take 5 years until NZ will then become most obese.No-one knows if or when New Zealand's obesity rates will start to plateau particularly with a government that is doing little to address the problem. If the rate however does start to plateau we will not know this for another 5 years.More importantly the point behind the editorial was not to debate details of future obesity predictions. It was to raise awareness of the single biggest health threat to New Zealand which is currently not being addressed.Steven Kelly & Richard FlintBariatric Surgeons Department of Surgery, Christchurch Public Hospital Christchurch, New ZealandReferences1. Ogden CL, Carroll MD, Kit BK, Flegal KL. Prevalence of Childhood and Adult Obesity in the United States, 2011-2012. JAMA. 2014;311(8):806-814. doi:10.1001/jama.2014.732. http://jama.jamanetwork.com/article.aspx?articleid=18325422. OECD. Obesity Update. June 2014. http://www.oecd.org/els/health-systems/Obesity-Update-2014.pdf

Summary

Abstract

Aim

Method

Results

Conclusion

Author Information

Acknowledgements

Correspondence

Correspondence Email

Competing Interests

For the PDF of this article,
contact nzmj@nzma.org.nz

View Article PDF

Dear SirThe recent editorial by Kelly and Flint1 is a classic that should be included in the next edition of the book ‘How to Lie with Charts' by Gerald Everett Jones.2 Their graph is an excellent teaching example of turning good data into misleading information. A first-year student would quickly identify the glaring errors.A close look at the x-axis of their graph shows inconsistent year intervals of eight years (1989–1997), five years (1997–2002, 2002–2007, 2007–2012), one year (2012–2013), three years (2013–2016), and four years (2016–2020). The so-called projections of obesity rates compare two different models: a linear model for New Zealand and a log model of the USA using the different time intervals. It is also difficult to decipher the authors' statement ‘that the rate of obesity in New Zealand rises by 1% each year' from the data presented in their paper.Graphing the data correctly (Figure 1) still suggests an increase in the obesity rate, but not the same dramatic increase suggested by Kelly and Flint. The correct graph provides less support for the authors' proposition that in 5 years the New Zealand rate will exceed the rate in the USA. Figure 1. National adult obesity prevalence trends, New Zealand and USA Notes: 95% confidence intervals included on data points. New Zealand data is from comparable and consistent surveys.Sources: 1997 New Zealand National Nutrition Survey, New Zealand Health Survey (2002/03, 2006/07, 2011/12, 2012/13, 2013/14)3,4; NHANES data5; OECD Health Statistics 2014.6 While the authors have nobly promoted the need for increasing bariatric operations, their message is severely compromised by the inclusion of misleading and inappropriate data. Undoubtedly, obesity is, and will continue to be, a major public health issue in New Zealand.Kelly and Flint have done a major disservice to those involved in developing prevention strategies by the lack of rigor in the presentation of data. It is a prime example of producing unreliable facts from reliable data.Barry BormanAssociate ProfessorKylie MasonPrincipal AnalystCentre for Public Health Research Massey University-Wellington Wellington, New ZealandReferences1. Kelly S, Flint R. New Zealand is far behind Australia in offering weight-loss surgery [editorial]. N Z Med J 2015;128(1408):10–2. http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2015/vol-128-no-1408/64082. Jones GE. How To Lie With Charts. LaPuerta; 2007.3. Ministry of Health. The Health of New Zealand Adults 2011/12: Key findings of the New Zealand Health Survey. Wellington: Ministry of Health; 2012.4. Ministry of Health. Annual Update of Key Results 2013/14: New Zealand Health Survey. Wellington: Ministry of Health; 2014.5. Fryar CD, Carroll MD, Ogden CL. Prevalence of Overweight, Obesity, and Extreme Obesity Among Adults: United States, 1960–1962 Through 2011–12. 2014. http://www.cdc.gov/nchs/data/hestat/obesity_adult_11_12/obesity_adult_11_12.pdf6. OECD. OECD Health Statistics 2014. http://www.oecd.org/health/healthdataResponse from Kelly and FlintThis is our reply to the questions about the diagram.The x-axis data values are not linear as the data points are projected directly on the graph. However this does not affect the statistical modelling, it only reflects the graphical projection. Your graph projects both New Zealand and USA obesity rates versus time as being linear. This is true for New Zealand, as it has linearly increasing obesity rates of 1% every year since 1989. However it is a fact that the USA has had a stable obesity rate for the last 10 years as published by Cynthia Ogden from the CDC1 and also published by the OECD.2 Therefore USA has a logarithmic trend line as depicted on the editorial paper.From a simplistic standpoint if New Zealand is currently 30% obese and increases by 1% per year, and USA is currently 35% obese and stable, then it will take 5 years until NZ will then become most obese.No-one knows if or when New Zealand's obesity rates will start to plateau particularly with a government that is doing little to address the problem. If the rate however does start to plateau we will not know this for another 5 years.More importantly the point behind the editorial was not to debate details of future obesity predictions. It was to raise awareness of the single biggest health threat to New Zealand which is currently not being addressed.Steven Kelly & Richard FlintBariatric Surgeons Department of Surgery, Christchurch Public Hospital Christchurch, New ZealandReferences1. Ogden CL, Carroll MD, Kit BK, Flegal KL. Prevalence of Childhood and Adult Obesity in the United States, 2011-2012. JAMA. 2014;311(8):806-814. doi:10.1001/jama.2014.732. http://jama.jamanetwork.com/article.aspx?articleid=18325422. OECD. Obesity Update. June 2014. http://www.oecd.org/els/health-systems/Obesity-Update-2014.pdf

Summary

Abstract

Aim

Method

Results

Conclusion

Author Information

Acknowledgements

Correspondence

Correspondence Email

Competing Interests

For the PDF of this article,
contact nzmj@nzma.org.nz

View Article PDF

Dear SirThe recent editorial by Kelly and Flint1 is a classic that should be included in the next edition of the book ‘How to Lie with Charts' by Gerald Everett Jones.2 Their graph is an excellent teaching example of turning good data into misleading information. A first-year student would quickly identify the glaring errors.A close look at the x-axis of their graph shows inconsistent year intervals of eight years (1989–1997), five years (1997–2002, 2002–2007, 2007–2012), one year (2012–2013), three years (2013–2016), and four years (2016–2020). The so-called projections of obesity rates compare two different models: a linear model for New Zealand and a log model of the USA using the different time intervals. It is also difficult to decipher the authors' statement ‘that the rate of obesity in New Zealand rises by 1% each year' from the data presented in their paper.Graphing the data correctly (Figure 1) still suggests an increase in the obesity rate, but not the same dramatic increase suggested by Kelly and Flint. The correct graph provides less support for the authors' proposition that in 5 years the New Zealand rate will exceed the rate in the USA. Figure 1. National adult obesity prevalence trends, New Zealand and USA Notes: 95% confidence intervals included on data points. New Zealand data is from comparable and consistent surveys.Sources: 1997 New Zealand National Nutrition Survey, New Zealand Health Survey (2002/03, 2006/07, 2011/12, 2012/13, 2013/14)3,4; NHANES data5; OECD Health Statistics 2014.6 While the authors have nobly promoted the need for increasing bariatric operations, their message is severely compromised by the inclusion of misleading and inappropriate data. Undoubtedly, obesity is, and will continue to be, a major public health issue in New Zealand.Kelly and Flint have done a major disservice to those involved in developing prevention strategies by the lack of rigor in the presentation of data. It is a prime example of producing unreliable facts from reliable data.Barry BormanAssociate ProfessorKylie MasonPrincipal AnalystCentre for Public Health Research Massey University-Wellington Wellington, New ZealandReferences1. Kelly S, Flint R. New Zealand is far behind Australia in offering weight-loss surgery [editorial]. N Z Med J 2015;128(1408):10–2. http://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2015/vol-128-no-1408/64082. Jones GE. How To Lie With Charts. LaPuerta; 2007.3. Ministry of Health. The Health of New Zealand Adults 2011/12: Key findings of the New Zealand Health Survey. Wellington: Ministry of Health; 2012.4. Ministry of Health. Annual Update of Key Results 2013/14: New Zealand Health Survey. Wellington: Ministry of Health; 2014.5. Fryar CD, Carroll MD, Ogden CL. Prevalence of Overweight, Obesity, and Extreme Obesity Among Adults: United States, 1960–1962 Through 2011–12. 2014. http://www.cdc.gov/nchs/data/hestat/obesity_adult_11_12/obesity_adult_11_12.pdf6. OECD. OECD Health Statistics 2014. http://www.oecd.org/health/healthdataResponse from Kelly and FlintThis is our reply to the questions about the diagram.The x-axis data values are not linear as the data points are projected directly on the graph. However this does not affect the statistical modelling, it only reflects the graphical projection. Your graph projects both New Zealand and USA obesity rates versus time as being linear. This is true for New Zealand, as it has linearly increasing obesity rates of 1% every year since 1989. However it is a fact that the USA has had a stable obesity rate for the last 10 years as published by Cynthia Ogden from the CDC1 and also published by the OECD.2 Therefore USA has a logarithmic trend line as depicted on the editorial paper.From a simplistic standpoint if New Zealand is currently 30% obese and increases by 1% per year, and USA is currently 35% obese and stable, then it will take 5 years until NZ will then become most obese.No-one knows if or when New Zealand's obesity rates will start to plateau particularly with a government that is doing little to address the problem. If the rate however does start to plateau we will not know this for another 5 years.More importantly the point behind the editorial was not to debate details of future obesity predictions. It was to raise awareness of the single biggest health threat to New Zealand which is currently not being addressed.Steven Kelly & Richard FlintBariatric Surgeons Department of Surgery, Christchurch Public Hospital Christchurch, New ZealandReferences1. Ogden CL, Carroll MD, Kit BK, Flegal KL. Prevalence of Childhood and Adult Obesity in the United States, 2011-2012. JAMA. 2014;311(8):806-814. doi:10.1001/jama.2014.732. http://jama.jamanetwork.com/article.aspx?articleid=18325422. OECD. Obesity Update. June 2014. http://www.oecd.org/els/health-systems/Obesity-Update-2014.pdf

Summary

Abstract

Aim

Method

Results

Conclusion

Author Information

Acknowledgements

Correspondence

Correspondence Email

Competing Interests

Contact diana@nzma.org.nz
for the PDF of this article

Subscriber Content

The full contents of this pages only available to subscribers.
Login, subscribe or email nzmj@nzma.org.nz to purchase this article.

LOGINSUBSCRIBE