ob_convertr | R Documentation |

Child thinness, overweight and obesity are defined as the child's body mass
index (BMI) lying beyond a pre-specified reference cutoff. Three references
are compared: IOTF (International Obesity Task Force), WHO (World Health
Organization) and CDC (US Centers for Disease Control and Prevention), each
of which have their own cutoffs. `ob_convertr`

takes age-sex-specific
prevalence rates of thinness, overweight or obesity based on one of the cutoffs,
and converts them to the corresponding rates based on a different cutoff.
`ob_convertr2`

uses paired prevalence rates of overweight and obesity on
one cutoff to estimate those based on another cutoff.

ob_convertr( age, sex, from, to, pfrom = NA, pto = NA, data = parent.frame(), report = c("vector", "wider", "longer"), plot = c("no", "density", "compare") ) ob_convertr2( age, sex, from, to, pfrom = NA, pto = NA, data = parent.frame(), report = c("vector", "wider", "longer"), plot = c("no", "density", "compare") )

`age` |
vector of ages between 2 and 18 years corresponding to prevalence rates |

`sex` |
vector of sexes corresponding to |

`from` |
name(s) of the BMI cutoff(s) on which the prevalence |

`to` |
name(s) of the BMI cutoff(s) on which to base the predicted prevalence (see Details). |

`pfrom` |
vector of age-sex-specific percentage prevalence rates
based on |

`pto` |
vector (needed for |

`data` |
optional data frame containing |

`report` |
character controlling the format of the returned data: 'vector'
for the estimated prevalence rates, 'wider' for the working tibble in wide
format, i.e. the |

`plot` |
character controlling what if anything is plotted: 'no' for no
plot, 'density' to display the BMI density distributions and cutoffs
corresponding to |

The IOTF cutoffs correspond to the value of BMI
(kg/m^{2}) at age 18: IOTF 35
(morbid obesity), IOTF 30 (obesity), IOTF 25 (overweight), IOTF 18.5 (grade 1
thinness), IOTF 17 (grade 2 thinness) and IOTF 16 (grade 3 thinness).

The WHO cutoffs correspond to BMI z_scores. Age 5-19 years, WHO +2 (obesity), WHO +1 (overweight) and WHO -2 (thinness). Age 0-5 years, WHO +3 (obesity), WHO +2 (overweight) and WHO -2 (thinness).

The CDC cutoffs correspond to BMI centiles: CDC 95 (obesity), CDC 85 (overweight) and CDC 5 (thinness).

Note: the overweight category needs to be analysed as overweight prevalence plus obesity prevalence, i.e. the prevalence above the overweight cutoff. To predict overweight prevalence excluding obesity prevalence, first calculate predicted overweight prevalence including obesity then subtract predicted obesity prevalence.

The algorithms for `ob_convertr`

and `ob_convertr2`

are distinguished
by the number of prevalence rates used for the prediction. For `ob_convertr`

(Cole & Lobstein, 2022) just one
rate is used – in this case the algorithm is commutative, meaning that
converting a prevalence rate from cutoff A to cutoff B and then from B to A
returns the original value. `from`

and `to`

are
the names of the cutoffs, and `pfrom`

and optionally `pto`

are vectors
of percentage prevalence rates.

`ob_convertr2`

uses two known prevalence rates (Cole & Lobstein, 2023),
typically overweight and obesity based on one reference, returning the corresponding
rates based on another reference. It is more accurate than `ob_convertr`

though
not exactly commutative. `from`

and `to`

are the names of the cutoffs as length-2
character strings, while `pfrom`

and optionally `pto`

are length-2
character strings giving the names of the corresponding vector prevalence rates.
For convenience the `from`

or `to`

names 'CDC', 'IOTF' or 'WHO'
expand to the corresponding pairs of cutoffs for overweight and obesity,
e.g. 'CDC' expands to c('CDC 85', 'CDC 95').

Alternatively `ob_convertr2`

can be used to interpolate or extrapolate
to one or more specified z-score cutoffs assuming the same reference for all cutoffs.
Here the values of `from`

and `to`

are numerical z-score cutoffs,
with at least two for `from`

. See the final example.

The algorithms require the prevalences of obesity and overweight net of obesity to be non-zero, and if they are zero they are set to missing.

The predicted prevalence rates, optionally with a plot visualizing the
findings, depending on the `report`

and `plot`

settings. Each
predicted rate is given the name of the relevant cutoff followed by " pred".

With `report`

set to "wider" or "longer", extra information
is returned reflecting the internal workings of the algorithms. In particular
`ob_convertr2`

returns `b`

the regression coefficient of z-score
prevalence on z-score cutoff as described in Cole & Lobstein (2023).

If a `plot`

is selected, the underlying data and plot are returned
invisibly with names `data`

and `plot`

.

Tim Cole tim.cole@ucl.ac.uk

Cole TJ, Lobstein T. Exploring an algorithm to harmonize International Obesity Task Force and World Health Organization child overweight and obesity prevalence rates. Pediatr Obes 2022;In press.

Cole TJ, Lobstein T. An improved algorithm to harmonize child overweight and obesity prevalence rates. Pediatr Obes 2023;In press.

The CDC reference for children aged 2-20 years is: Must A, Dallal GE, Dietz WH. Reference data for obesity: 85th and 95th percentiles of body mass index (wt/ht2) and triceps skinfold thickness. American Journal of Clinical Nutrition 1991; 53: 839-46. Available at: https://academic.oup.com/ajcn/article/53/4/839/4715058

The IOTF reference for children aged 2-18 years is: Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320: 1240-5. Available at doi: 10.1136/bmj.320.7244.1240

The WHO reference for children aged 0-5 years is: WHO Child Growth Standards: Length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age: Methods and development. Geneva: World Health Organization, 2006. Available at: https://www.who.int/toolkits/child-growth-standards/standards

The WHO reference for children aged 5-19 years is: de Onis M, Onyango AW, Borghi E, Siyam A, Nishida C, Siekmann J. Development of a WHO growth reference for school-aged children and adolescents. Bulletin of the World Health Organization 2007; 85: 660-7. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2636412/pdf/07-043497.pdf/

## convert 10% IOTF overweight prevalence (cutoff IOTF 25, including obesity) ## in 8-year-old boys to overweight prevalence for cutoff WHO +1 ob_convertr(age = 8, sex = 'boys', from = 'IOTF 25', to = 'WHO +1', pfrom = 10) ## compare the BMI density functions and cutoffs for IOTF and WHO ## in 8-year-old boys ob_convertr2(age = 8, sex = 'boys', from = 'IOTF', to = 'WHO', plot = 'density') ## convert IOTF overweight prevalence to WHO overweight prevalence ## and compare with true value - boys and girls aged 7-17 (22 groups) ## note the need to first add obesity prevalence to overweight prevalence data(deren) deren <- within(deren, { CDC85 = CDC85 + CDC95 IOTF25 = IOTF25 + IOTF30 `WHO+1` = `WHO+1` + `WHO+2`}) ob_convertr(age = Age, sex = Sex, from = 'IOTF 25', to = 'WHO +1', pfrom = IOTF25, pto = `WHO+1`, data = deren, plot = 'compare') ## convert IOTF overweight and obesity prevalence to WHO using ## ob_convertr2 - which is more accurate than ob_convertr ob_convertr2(age = Age, sex = Sex, from = 'IOTF', to = 'WHO', pfrom = c('IOTF25', 'IOTF30'), pto = c('WHO+1', 'WHO+2'), data = deren, plot = 'compare') ## extrapolate WHO overweight and obesity prevalence (cutoffs +1 and +2) ## to severe obesity prevalence based on cutoffs +2.5 or +3 ob_convertr2(Age, Sex, from = 1:2, to = c(2.5, 3), pfrom = c('WHO+1', 'WHO+2'), data = deren, report = 'wider')

Embedding an R snippet on your website

Add the following code to your website.

For more information on customizing the embed code, read Embedding Snippets.