China
Africa
Explore chapters and articles related to this topic
Critical appraisal of cross-sectional surveys
Published in O. Ajetunmobi, Making Sense of Critical Appraisal, 2021
It is most important that researchers should consult with a statistician regarding the sample size calculation of a survey. Inadequate sample size leads to the collection of unrepresentative information called ‘sampling error’.
HPB Surgery
Published in Tjun Tang, Elizabeth O'Riordan, Stewart Walsh, Cracking the Intercollegiate General Surgery FRCS Viva, 2020
London Lucien Ooi Peng Jin, Teo Jin Yao
What is the benefit of EUS-FNA and what are the limitations?An EUS allows for a more direct assessment of the tumour in the head of the pancreas, and may show the tumour in relation to adjacent critical structures like the duodenum and portal vein to assess for invasion and also the potential for achieving an R0 resection. It is also useful in preoperative assessment of a possible need for concomitant portal vein resection. In addition, the EUS may show up possible lymphadenopathy although this should not affect operability.An additional benefit of the EUS is to allow FNA for cytology or core biopsy of the tumour, if needed. The transduodenal puncture reduces the risk of possible tumour seeding that exists with a percutaneous approach. However, the main issues of sampling error and false negatives need to be considered. There is also a small risk of bleeding and pancreatitis.
Experimental Design, Evaluation Methods, Data Analysis, Publication, and Research Ethics
Published in Yuehuei H. An, Richard J. Friedman, Animal Models in Orthopaedic Research, 2020
The basic objective of sampling is that a sample should be chosen to represent its population. An estimate of a population parameter that is determined from a random sample will generally differ from the true value to some extent. This difference is referred to as the sampling error. Sampling error reflects the inherent uncertainty of conclusions about a population based solely on information gained from sample data (a subset of the population). The magnitude of sampling error is a function of sample size. This is due to the large amount of inherent variation in estimates based on small samples as compared to the smaller inherent variation seen with large sample sizes. The amount of statistical uncertainty associated with a particular study can be expressed in the form of confidence intervals. These intervals are largely determined by the sample size. Intervals based on small samples are relatively wide, reflecting a relatively large sampling error. Intervals based on adequate sample sizes are more narrow, reflecting a smaller sampling error.
Reproductive outcomes of infertile couples undergoing assisted reproductive technology who are carriers of chromosomal abnormalities: a retrospective cohort study
Published in Annals of Medicine, 2022
Ling Cui, Fang Wang, Yonghong Lin, Min Li
This was a retrospective cohort study. The exposed group comprised infertile couples with chromosomal abnormalities. The control group comprised infertile couples without chromosomal abnormalities. Bias was due mostly to sampling error. To minimize the sampling error, we matched 1:2 data by female age, type of infertility (primary, secondary), and type of ART, namely, intrauterine insemination (IUI), IVF, or intra-cytoplasmic sperm injection (ICSI), was conducted. Including criteria was infertile patients who carriers of chromosomal abnormalities refuse PGT after genetic counselling and request random selection of embryos when they seek ART. A total of 4656 infertile couples came to our centre (Department of Reproduction and Infertility, Chengdu Women’s and Children’s Central Hospital) for ART in the past 3 years (1 January 2017 to 31 December 2020) and were followed-up with phone calls.
The association between self-Esteem, stigma, and mental health among South African youth living with HIV: the need for integrated HIV care services
Published in AIDS Care, 2022
Latoya A. Small, Alexis K. Huynh, Tyrone M. Parchment
Certain limitations influence the interpretation of this study's findings. The small sample size and recruitment method may not reflect the larger population of pYLHIV in Durban, S.A., or Sub-Saharan Africa. However, it provides insight into mental health symptoms existing within this subgroup. The use of purposive sampling discounts random selection. This increases the possibility of sampling bias and sampling error. These youth and their caregivers were recruited from HIV clinics where they were already involved in medical treatment and may not reflect the needs of those not involved in care. Among participants, social desirability bias may also have occurred as participants were interviewed during data collection. The use of baseline data also prevents the inference of causal relationships between variables.
Five-year review of intraoperative pathology consultation in a single institution
Published in Baylor University Medical Center Proceedings, 2021
Liping Wang, Bing Leng, Debby Rampisela
For frozen section discordances with major clinical impact, the most commonly involved organ system was skin (37.8%, skin margin), followed by lymph node (24.4%, lymph node metastasis) and head and neck (13.5%). The most common cause of error was histologic sampling error (51.2%), followed by misinterpretation (35.1%) and gross sampling error (8.1%). Most frozen section errors were false-negative errors (91.9%); only two cases were false-positive errors (5.4%), and one case was a misclassification error (2.7%). An example of a false-positive case was a lymph node from a patient with invasive ductal carcinoma of breast, which was initially diagnosed as positive for metastatic carcinoma in the frozen section, while the final diagnosis was benign lymph node with subcapsular nevus (Figure 4). An example of a false-negative error was a distal ureteral margin from a patient with bladder urothelial carcinoma, which was misinterpreted as benign in the frozen section, while the final margin was positive for urothelial carcinoma, plasmacytoid variant (Figure 5).