Completeness and accuracy of anthropometric measurements in electronic medical records for children attending primary care

Abstract

Background Electronic medical records (EMRs) from primary care may be a feasible source of height and weight data. However, the use of EMRs in research has been impeded by lack of standardisation of EMRs systems, data access and concerns about the quality of the data. Objectives The study objectives were to determine the data completeness and accuracy of child heights and weights collected in primary care EMRs, and to identify factors associated with these data quality attributes. Methods A cross-sectional study examining height and weight data for children <19 years from EMRs through the Electronic Medical Record Administrative data Linked Database (EMRALD), a network of family practices across the province of Ontario. Body mass index z-scores were calculated using the World Health Organization Growth Standards and Reference. Results A total of 54,964 children were identified from EMRALD. Overall, 93% had at least one complete set of growth measurements to calculate a body mass index (BMI) z-score. 66.2% of all primary care visits had complete BMI z-score data. After stratifying by visit type 89.9% of well-child visits and 33.9% of sick visits had complete BMI z-score data; incomplete BMI z-score was mainly due to missing height measurements. Only 2.7% of BMI z-score data were excluded due to implausible values.

Is it time to stop sweeping data cleaning under the carpet? A novel algorithm for outlier management in growth data

“…The NLME-A method detected duplications, decimal point and unit errors in all datasets but transpose and addition errors seemed to be unique to Dogslife and CLOSER data with simulated errors. The prevalence of errors in our datasets ranged from 0.25% to 3.31%, which is within the same range as previous studies that have identified implausible values in growth data [9][10][11][12][13]. The presence of duplications and errors in medical records emphasises the importance of cleaning datasets even if they have been recorded by professionals.…”

“…It is difficult to distinguish errors from genuine anomalies in certain types of data, such as height and weight records, because biological data is heterogenous and may contain unusual but plausible values. These datasets are variable in terms of how accurate they are, with authors estimating error rates to be anything from 0.03% to 4.5% [9][10][11][12][13]. Since the first computational cleaning method for longitudinal growth [14] there have been enormous technological advancements, yet there remains no standardised data cleaning method.…”

Temporal trends in severe obesity prevalence in children and youth from primary care electronic medical records in Ontario: a repeated cross-sectional study

“…38 The completeness and accuracy of child height and weight measurements in EMRALD were examined previously and shown to be of high quality. 39 Another analysis of weight status that used EMRALD data highlighted the strengths and limitations of this data source: 40 a strength is the adequate sample size to investigate severe obesity prevalence, and a limitation is the lack of generalizability to all children in the Ontario and Canadian populations. Our EMRALD study population differed from the overall Ontario population on factors related to obesity 41 and to routine use of primary care such as age, immigration status and neighbourhood income.…”

In this issue – patient risks are ignored at organisation level

“…sets out to report completeness of data about obesity in childhood records. They found that over 90% of records had valid information about their Canadian network 8. My sense is that the completeness of data in computerised medical records is improving over time, and with obesity, such an international problem having data about complete cohorts is really valuable.…”