Background

Examine the evolution of stunting in <5 years of age in Peru and its relationship with birth interval (BI), and associated factors from 1996 to 2014.
Methods

Data were extracted from the National Demographic and Health Survey of Peru (DHS) for the years 1996, 2000, 2005, 2010 and 2014 on children <5 years of age for stunting (height / age ≤ 2DS). The main independent variable of interest was BI dichotomized by interval between date of last and penultimate birth (≤24 months vs. >24 months) and others sociodemographic variables. Crude and adjusted odds ratios with 95% confidence intervals were estimated in logistic regression for the effect of BI and other variables on stunting by each survey year.
Results

Between 1996 and 2014, stunting declined in both BI groups: ≤24 months: 35.8% to 21.8%; >24 months: 29.5% to 14.3%. BI was associated with stunting after adjustment for other variables in each survey except 2005. Of the other factors, sex, birth order (fourth child or more), maternal education and poverty were independently associated with stunting in all survey years. Residence in rural areas and in Amazon forest and highland regions was associated with stunting 1996 and 2000. Maternal age was not independently associated with stunting.
Conclusions

While stunting showed a decreasing trend from 1996 to 2014, birth interval exceeding 24 months exerted a protective effect on stunting across the years surveyed. Poverty, low maternal education and high birth order were associated with stunting in all survey years.