A study to determine a practical method for weight estimation of deceased persons

Background

Body weight is an important parameter for estimating the postmortem interval (PMI) at a crime scene. However, a challenge arises at crime scenes when a weight scale for measuring the total body weight is unavailable. Anthropometry-based models to estimate body weight have been developed in previous studies. This study aims to determine the accuracy of body weight estimations by practitioners, test the anthropometry-based models for applicability to deceased individuals, and examine a potential new method based on applying heel weight.

Methods

A prospective study was conducted at the Isala Hospital in the Netherlands. During the study period, deceased people that were admitted to the hospital mortuary were included consecutively. The body weight of deceased persons estimated by practitioners was compared to the actual body weight. Anthropometric measurements were taken and used to perform eight sex dependant anthropometry-based models, with accuracy for the actual body weight calculated using RMSE values. A Pearson's correlation test was used to determine the correlation between heel weight and total body weight.

Results

During the study period, a total of 100 cases, 56 males and 44 females, were included. Overall, only 33.3 % of practitioners' estimations were within 5 % of the actual measured weight. The model based on abdominal and thigh circumference performed best for weight estimation in males and the models based on mid-arm circumference, abdominal circumference, calf circumference and, in one model, subscapular skinfold performed best in females. A Pearson's correlation test revealed a weak positive correlation between weight of the heel and total body weight (Pearson's correlation coefficient: 0.214).

Discussion

Estimations of underweight or obese patients posed a challenge for weight estimation. Especially in these cases, study results showed that anthropometry-based models have potential for daily practice. However, additional research is required to assess the reliability of the best performing models before implementation in forensic casework. The correlation between the weight of the heel and body weight was low, therefore implementation of the current method is not recommended, and further research is required.