Determinants of reliability of self-reported height and weight and their impact on medication dosing: a cross-sectional study

Objective Patient-reported anthropometric measures, such as height and weight, are frequently used in clinical practice but are susceptible to reporting biases. This study aims to investigate the determinants of reliability of patient-reported anthropometric measures in patients in cardiology and general practice and their impact on potential medication dosing.Design Cross-sectional study.Setting and methods 730 patients were recruited at the Clinic of Cardiology, Angiology and Intensive Care Medicine of Saarland University Hospital and a general medicine practice from November 2015 to December 2018. We assessed self-reported height and weight and compared them to calibrated measures immediately afterwards. Weight and height (optional with medical history) were self-reported via questionnaire. Interviews were conducted by female or male nursing staff or physicians.Outcome measures The main outcomes were the deviation between patients' self-reported height and weight from objective calibrated measures, as well as the amount of misdosing of exemplary drugs based on this deviation.Results The mean height (SD) of the participants (36% were patients) was 170.92 (9.34) cm. Patients significantly overestimated their height by 1.82 cm (range: -8.00 to 11.00 cm). Misreporting was best predicted by age, with older patients providing more height overestimations. The mean weight was 84.25 (17.41) kg and was significantly underestimated by 1.49 kg (range: -36.00 to 26.00 kg). Misreporting was best predicted by higher body mass index, cognitive impairment and a longer duration since the last weighing, and self-reporting by questionnaires was associated with a higher under-reporting of weight. Unlike females, male patients exhibited a more pronounced tendency to under-report their weight when responding to questionnaires compared with face-to-face interviews. Comparison of doses for low-molecular-weight heparin according to self-reported versus calibrated weight revealed potential underdosing and overdosing in 17% and 77% of all patients, respectively. For the cytostatic agent doxorubicin, for instance, underdosing and overdosing would have been applied in 40% and 43% of all patients, respectively.Conclusions and relevance Self-reported height and weight are often invalid, especially in patients who are older and overweight. Misreporting can lead to inappropriate drug dosing. Calibrated measurement of height and weight remains part of good clinical practice, and if self-reporting is unavoidable, personal interviews should be preferred over questionnaires.Trial registration number https://clinicaltrials.gov/study/NCT04321057