No abstract

No abstract

No abstract

Introduction The trend of prescribing VD preparations for nonspecific body aches and self-medication has increased significantly. The importance of vitamin D toxicity (VDT) has been underestimated and under recognized. This study was done to determine the frequency toxicity (>150 ng/ml) in subjects for 25-hydroxyvitamin D (25OHD) and evaluate the vitamin D (VD) supplements used by these subjects. Methodology This descriptive cross-sectional study was conducted at the Section of Chemical Pathology, Aga Khan University Hospital Karachi from April 2020 to March 2021. Subjects with 25OHD toxicity were contacted and information related to history of calcium and VD supplementation were collected. The statistical analysis was performed using the Microsoft Excel 2016. Results Over a year period 105398 subjects were tested for serum 25OHD, of which 0.34% (n = 364) subjects had 25OHD level of >150 ng/ml. After satisfying exclusion criteria 186 subjects (78 were <18 years of age and 108 were adults) were included in final analysis. All of these were using VD supplements and the main indications were delayed growth/short height (43.7%, n = 34) and aches or pains in (54.6%, n = 59) in pediatric and adult subjects respectively. Most of the subjects were taking supplements orally (74.1%, n = 138). Commonly prescribed preparation in adults and pediatric was 200,000 IU (70.4%, n = 76) and 400 IU (35.9%, n = 28) respectively. Most subjects took supplements for 1–3 months (68.3%, n = 127). Stated total supplementation ranged from 20,000 IU to 3600,000 IU in pediatric subjects and 200,000 IU to 96,00,000 IU in adults. Conclusions Supplementation is a leading cause of potential toxic levels of 25OHD. The condition can be prevented by careful use of VD supplements and consistent monitoring.

In this paper, we propose an exponential cum ratio-product type class of estimators for population median under simple random sampling scheme using the supplementary variable. Expressions for bias and mean square error (MS E) are obtained up to first order of approximation. The proposed class of estimators is more efficient as compared to all considered estimators under certain conditions. Four real data sets and simulation studies are carried out to observe the performances of the estimators. Both numerical and simulation studies show that the proposed class of estimators performs better as compared to all considered estimators.