-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.Rmd
86 lines (59 loc) · 7.79 KB
/
index.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
---
title: "Child Development with the D-score: Turning Milestones into Measurement"
author: "Authors: Stef van Buuren & Iris Eekhout"
description: "The D-score is a one-number summary that quantifies generic development of children, closely modelled after body measurements like height and weight. This booklet is an accessible introduction to the D-score. The text illustrates the main concepts with real data collected on the Dutch Development Instrument for children aged 0-2 years."
site: bookdown::bookdown_site
url: https://d-score.org/dbook1
github-repo: "D-score/dbook1"
cover-image: "fig/shirleypd-1.png"
apple-touch-icon: "fig/D.png"
apple-touch-icon-size: 72
favicon: "favicon.ico"
documentclass: book
bibliography: [book.bib]
biblio-style: apalike
link-citations: yes
always_allow_html: true
---
```{r initbook, include = FALSE}
opts_chunk$set(comment = NA, background = "#FBFBFB",
fig.align = 'center', cache = TRUE,
echo = FALSE, out.width = "100%")
# silence dplyr grouping warning
# https://stackoverflow.com/questions/62140483/how-to-interpret-dplyr-message-summarise-regrouping-output-by-x-override
options(dplyr.summarise.inform = FALSE)
```
# Turning milestones into measurement {-}
Children learn to walk, speak, and think at an astonishing pace. The D-score captures this process as a one-number summary. The D-score book explains why we need the D-score, how we construct it, and how we calculate it. Application of the D-score enables comparisons in child development across populations, groups and individuals.
We are preparing four D-score book chapters under the following titles:
I. [Turning milestones into measurement](https://d-score.org/dbook1)
II. [Tuning instruments to unity](https://d-score.org/dbook2)
III. [Tailoring tests to fit the occasion](https://stefvanbuuren.name/dbook3)
IV. [Taking off the hood](https://stefvanbuuren.name/dbook4)
This chapter is written by [Stef van Buuren](https://stefvanbuuren.name) and [Iris Eekhout](https://www.iriseekhout.com), with contributions from Marrianne de Wolff to Section 1, Manon Grevinga to Section 8, Paula van Dommelen to Section 9, and Maria C. Olthof to Section 9.
[Chapter I](https://d-score.org/dbook1) and [Chapter II](https://d-score.org/dbook2) are currently available as a complete draft. We still need to write most of Chapter III and IV. The series addresses conceptual aspects of the D-score, discusses practical issues, and introduces a dedicated set of `R` packages.
The *Health Birth Growth and Development knowledge integration (ki)* program of the *Bill & Melinda Gates Foundation* kindly supports the work.
If you have any suggestions or comments, please let us know.
## About our work {-}
The first 1000 days of human life cover the period between conception and the second birthday. Proper development during this period contributes to future health, happiness, and productivity, so it is essential to track the child's progress during infancy and early childhood.
But did you know that more than 150 instruments exist that quantify child development? And are you aware that many of these tools produce not just one, but many scores? Such an overwhelming choice may seem a luxury until you realise that we cannot compare their ratings. Of course, we could settle on just one instrument ...., but that's never going to happen.
Our work on the D-score explores an alternative strategy—modern data science methods aid in connecting instruments through shared milestones. We present a unified framework that places children and milestones from different tools onto the same scale. As a result, we can measure child development by just one number, the D-score. Separating the scale from the instrument is a revolutionary concept. Application of the D-score enables comparisons in child development across populations, groups and individuals, even when we measure by different tools.
The new "unit for child development" has exciting implications. We may:
- Track child development over time, as in growth charts;
- Construct age-related references for healthy development;
- Adjust the D-score for age;
- Select an instrument that is precise enough for the setting at hand;
- Compare developmental trajectories between children;
- Compare child development between countries;
- Derive concise tools by picking only well-targeted milestones;
- Study the impact of interventions on child development;
- Predict future health from the current D-score.
Our ongoing work addresses conceptual aspects of the D-score, discusses practical issues, and introduces a dedicated set of R packages.
We aim for three audiences:
1. Professionals in child development who wish to familiarise themselves with a new approach to measure child development in early childhood. Separating the tools from the scales allows the professional to select the means most suited for a particular setting. These chapters give professionals the conceptual background of the D-score.
2. Policymakers in international settings who need to weigh the effect of interventions on child development. The existence of different instruments severely hampers their ability to obtain insight into the results of these interventions. The ability to place measurements onto the same scale allows for a more accurate understanding of policy effects, thus supporting the setting of priority levels.
3. Data scientists who can transform a vector of milestone data into a one-number summary with an unambiguous unit. The techniques have a solid psychometric backing, and also apply to other types of problems. These chapters explain this conversion process in detail, thereby opening up the way for the application of precise analytic techniques in many different settings.
Additionally, parents are always eager to follow every step of their child. While we do not target this work to parents, our methodology may spark the interest of authors, app writers, and instrument creators that do address the interests and needs of parents. Hence, the publication of these chapters may have additional societal impact.
## About the Authors {-}
Stef van Buuren is a professor of Statistical Analysis of Incomplete Data at the University of Utrecht and statistician at the Netherlands Organisation for Applied Scientific Research TNO in Leiden. His interests include the analysis of incomplete data and child growth and development (h-index 61). Van Buuren is the inventor of the MICE algorithm for multiple imputation of missing data (>85.000 downloads per month) and has written the accessible monograph Flexible Imputation of Missing Data. Second Edition, CRC/Chapman & Hall. He designed the growth charts for the Dutch child health care system and invented the D-score, a new method for expressing child development on a quantitative scale. He consults for the World Health Organization and the Bill & Melinda Gates Foundation. More background at [https://stefvanbuuren.name](https://stefvanbuuren.name), and software at [https://github.com/stefvanbuuren](https://github.com/stefvanbuuren).
Iris Eekhout holds a double masters in clinical psychology and methodology and statistics of psychology (Leiden University). She obtained her PhD at the Department of Epidemiology and Biostatistics of the VU University medical centre in Amsterdam. Her dissertation work resulted in novel ways of dealing with missing data in questionnaire items and total scores. Currently, Iris teaches a course on missing data analysis in the epidemiology master’s program at VU University medical centre. At TNO, Iris works on a variety of projects as a methodologist and statistical analyst related to child health, e.g., measuring child development (D-score) and adaptive screenings for psycho-social problems (psycat). More background at [https://www.iriseekhout.com](https://www.iriseekhout.com), and software at [https://github.com/iriseekhout](https://github.com/iriseekhout).