Obesity status transitions across the elementary years: use of Markov chain modelling

Objective: The objective of this study was to assess overweight and obesity status transition probabilities using first-order Markov transition models applied to elementary school children.

Method: Complete longitudinal data across 11 assessments were available from 1494 elementary school children (from 7599 students in 41 out of 45 schools in a Southeast Texas school district) from kindergarten to the beginning of the fifth grade. Heights and weights were measured by trained school nurses using standard procedures at the beginning and end of each school year for the 11 consecutive assessments. To estimate the transition probabilities, first-order three-state (healthy weight, overweight and obese) Markov transition models were fit to the longitudinal weight status data of all assessment periods.

Results: While there was a gradual shift to more children in the overweight or obese category over 5 years, children were most likely to stay in the same weight category as the previous assessment. A consistent seasonal difference in the probability of changing weight status category was seen, with a greater probability of becoming overweight and obese during the summer compared with the school year. The transition probabilities to obesity were higher among boys, Hispanic and non-Hispanic Black, and lower socioeconomic status children.

Conclusions: This study provides the first application of a Markov transition model to child weight status data. The transitions into the overweight and obese categories were small, but persistent, with smaller percentages transitioning out of overweight or obese. Early monitoring and summer intervention strategies are needed to prevent the slow, but relentless, transition into the overweight and obese categories.