Aimone Ashley M, Bassani Diego G, Qamar Huma, Dasiewicz Alison, Perumal Nandita, Namaste Sorrel M L, Shah Devanshi, Roth Daniel E
Centre for Global Child Health, The Hospital for Sick Children, Toronto, Ontario, Canada.
Department of Paediatrics, The Hospital for Sick Children and University of Toronto, Toronto, Ontario, Canada.
PLOS Glob Public Health. 2023 Apr 17;3(4):e0001766. doi: 10.1371/journal.pgph.0001766. eCollection 2023.
Stunting prevalence is commonly used to track population-level child nutritional status. However, other metrics derived from anthropometric datasets may be used as alternatives to stunting or provide complementary perspectives on the status of linear growth faltering in low- and middle-income countries (LMICs). Data from 156 Demographic and Health Surveys in 63 LMICs (years 2000 to 2020) were used to generate 2 types of linear growth metrics: (i) measures of location of height distributions (including stunting) for under-5 years (<5y) and 2 to 5 years (2-5y); (ii) model-derived metrics including predicted mean height-for-age z-score (HAZ) at 0, 2, and 5 years; interval slopes of HAZ, height-for-age difference (HAD), and growth delay (GD) from 1 month to 2 years (1mo-2y) and 2-5y; and the SITAR intensity parameter (SITAR-IP) for <5y. Using Spearman's rank correlation coefficient (r), metrics were considered alternatives to stunting if very strongly correlated with stunting (|r|≥0.95) and at least as strongly correlated as stunting with selected population indicators (under 5y mortality, gross domestic product, maternal education). Metrics were considered complementary if less strongly correlated with stunting (|r|<0.95) yet correlated with population indicators. We identified 6 of 15 candidate metrics (stunting 2-5y, mean HAZ <5y and 2-5y, p25 HAZ <5y and 2-5y, predicted HAZ at 2y) as potential alternatives to stunting and 6 as complementary metrics (SITAR-IP, predicted HAZ at 5y, HAZ slope 1m-2y, HAD slope 1m-2y, GD slopes 1m-2y and 2-5y). Three metrics (HAZ slope 2-5y, HAD slope 2-5y years and predicted HAZ at birth) had weak correlations with population indicators (|r| ≤ 0.43). In conclusion, several linear growth metrics could serve as alternatives to stunting prevalence and others may be complementary to stunting in tracking global progress in child health and nutrition. Further research is needed to explore the real-world utility of these alternative and complementary metrics.
发育迟缓患病率通常用于追踪人群层面的儿童营养状况。然而，从人体测量数据集得出的其他指标可作为发育迟缓的替代指标，或为低收入和中等收入国家（LMICs）线性生长发育迟缓状况提供补充视角。利用63个低收入和中等收入国家（2000年至2020年）156份人口与健康调查的数据，生成了两种线性生长指标：（i）5岁以下（<5岁）和2至5岁（2 - 5岁）儿童身高分布位置的测量指标（包括发育迟缓）；（ii）模型衍生指标，包括0岁、2岁和5岁时预测的年龄别身高z评分（HAZ）；1个月至2岁（1个月 - 2岁）和2至5岁期间HAZ、年龄别身高差值（HAD）和生长延迟（GD）的区间斜率；以及5岁以下儿童的SITAR强度参数（SITAR - IP）。使用斯皮尔曼等级相关系数（r），如果指标与发育迟缓高度相关（|r|≥0.95）且与选定的人群指标（5岁以下儿童死亡率、国内生产总值、母亲教育程度）的相关性至少与发育迟缓一样强，则该指标被视为发育迟缓的替代指标。如果指标与发育迟缓的相关性较弱（|r|<0.95）但与人群指标相关，则该指标被视为补充指标。我们确定了15个候选指标中的6个（2 - 5岁发育迟缓、<5岁和2 - 5岁的平均HAZ、<5岁和2 - 5岁的第25百分位HAZ、2岁时预测的HAZ）为发育迟缓的潜在替代指标，6个为补充指标（SITAR - IP、5岁时预测的HAZ、1个月 - 2岁时的HAZ斜率、1个月 - 2岁时的HAD斜率、1个月 - 2岁和2 - 5岁时的GD斜率）。三个指标（2 - 5岁时的HAZ斜率、2 - 5岁时的HAD斜率以及出生时预测的HAZ）与人群指标的相关性较弱（|r|≤0.43）。总之，在追踪儿童健康和营养方面的全球进展时，几种线性生长指标可作为发育迟缓患病率的替代指标，其他指标可能是发育迟缓的补充指标。需要进一步研究以探索这些替代指标和补充指标在现实世界中的实用性。
