Reference Data for new Phenotypes
Help Index > Reference Data for new Phenotypes
The results are shown in the “Percentiles” and “Distribution” tab. The former displays a plot comprising the percentile curves alongside the entered personal data. The percentile curves are broken down into more detail at the end of the tab in the form of a table below the plot itself. There is also an evaluation table right below the plot regarding the entered personal data e.g. showing the percentile that the individual is on at each age.
Distribution of the reference data can also be displayed by using the “Distribution” tab. The first plot covered by this tab displays how the individuals of the reference data distribute over certain age categories allowing for a better classification of the personal data. The second plot shows the hand grip strength values of all individuals within the reference data taken into account by the options set earlier.
1. Background
Reference intervals and normative data are important in medicine sciences and clinical care to characterize measured data, e.g., to find outliers.
A typical way to represent such reference intervals in medical sciences is to use percentile curves specifically for each gender and separated by age bands
based on a large population. In this way, a new measured value can be compared to most frequent values of this population.
2. Problem
Reference intervals and normative data are traditionally available for anthropometry and within laboratories in order to decide whether measured values
are out of range and to trigger actions if necessary. However, such data are not available for new phenotypes as they have been measured in large
epidemiological studies, such as LIFE Adult and LIFE Heart . That makes it difficult to characterize such data on individual level. For example, given
a female person of 42 years for who a hand grip power of 32 kg is measured. The question whether 32 kg is low, high or normal in contrast to other
people in Mid Europe can be only answered by taking reference intervals or normative data into account.
3. Solution
We used data of 10,000 adults out of the epidemiological study LIFE Adult to generate reference intervals / normative data for different new
phenotypes including hand grip power, anthropometry, blood pressure, different measurements of eyes and in phonatry (voices). We used the widely
accepted GAMLSS approach to analyze raw data and to generate percentile curves according to gender and age bands. The percentile curves are
used to visualize data, i. e., individual data of a person which are then related to the LIFE Adult population data and percentile curves.
4. Guided Tour
- Head over to ‘Models’ using the menu bar
- Find the ‘Reference Data For Hand Grip Strength’ tool using the search, or the link below
(https://www.health-atlas.de/en/lha/7rvk32xtur-4) - on the right side click the link ‘Hand Grip Strength Percentiles Demonstrator’ in the tools section, or use the link below
(https://apps.health-atlas.de/hand-grip-strength/)
The results are shown in the “Percentiles” and “Distribution” tab. The former displays a plot comprising the percentile curves alongside the entered personal data. The percentile curves are broken down into more detail at the end of the tab in the form of a table below the plot itself. There is also an evaluation table right below the plot regarding the entered personal data e.g. showing the percentile that the individual is on at each age.
Distribution of the reference data can also be displayed by using the “Distribution” tab. The first plot covered by this tab displays how the individuals of the reference data distribute over certain age categories allowing for a better classification of the personal data. The second plot shows the hand grip strength values of all individuals within the reference data taken into account by the options set earlier.
5. Results
Based on conceptual work and study data of LIFE Adult we designed demonstrators allowing to characterize and to compare individual data
with percentile curves as representations of reference intervals and normative data.
6. References
- Löffler M, Engel C, Ahnert P et al.: The LIFE-Adult-Study: objectives and design of a population-based cohort study with 10,000 deeply phenotyped adults in Germany. BMC Public Health, 15, 2015
- Beutner F, Teupser D, Gielen S, Holdt LM, Scholz M, Boudriot E, Schuler G, Thiery J: Rationale and Design of the Leipzig (LIFE) Heart Study: Phenotyping and Cardiovascular Patients with Coronary Artery Disease. PloS One 6(12), 2012
- Stasinopoulos M D, Rigby R A, Heller G A, Voudouris V, De Bastiani F: Flexible Regression and Smoothing: Using GAMLSS in R. Chapman and Hall/CRC, 2017
