China
Africa
Explore chapters and articles related to this topic
External validity and public health
Published in Sridhar Venkatapuram, Alex Broadbent, The Routledge Handbook of Philosophy of Public Health, 2023
Even though this categorization framework is an improvement on approaches that simply list all known threats to external validity, it is still based on a problematic underlying assumption. This assumption is that the best way of thinking about the problem is to consider, when making an EV inference, whether the target population is sufficiently similar to the studied population. This emphasis on the idea of similarity of environment is questioned by Nancy Cartwright and Jeremy Hardie (2012) in their criticism of established methods for dealing with external validity: In the orthodoxy, a study has external validity when the “same treatment” has the “same result” in a specific target as it did in the study. The orthodox advice is that external validity can be expected if the target population is “sufficiently similar” to the study population. For us, the key question is how good a job this advice does in getting you from “it worked there” to “it will work here.” The answer: you are lucky if it gets you anywhere.(Cartwright and Hardie 2012: 46)
Athlete Monitoring
Published in Michael H. Stone, Timothy J. Suchomel, W. Guy Hornsby, John P. Wagle, Aaron J. Cunanan, Strength and Conditioning in Sports, 2023
Michael H. Stone, Timothy J. Suchomel, W. Guy Hornsby, John P. Wagle, Aaron J. Cunanan
Regardless of whether a project is research, service, or a combination of the two, the tools used to measure the variables of interest must be valid and reliable. Validity refers to whether or not the instrument is actually measuring what it is supposed to be measuring. There are basically four different types of measurement validity: Internal validity refers to how well the tool measures the variable in question (e.g., strength, power, speed, endurance).External validity concerns the ability of the tool to predict changes in a population other than the one being studied (e.g., when investigators measure strength in one group and then generalize to another group).Prediction validity refers to the ability to predict one variable from another (e.g., when investigators measure strength to predict the vertical jump).Ecological validity deals with how the parameters of the intervention/test approximate real-world setting. This is particularly important in sport science application and research.
Measurement Models for Patient-Reported Outcomes and Other Health-Related Outcomes
Published in Douglas D. Gunzler, Adam T. Perzynski, Adam C. Carle, Structural Equation Modeling for Health and Medicine, 2021
Douglas D. Gunzler, Adam T. Perzynski, Adam C. Carle
In psychometrics, internal validity exists when responses to a set of questions measure a construct (or constructs) as expected. This differs from the more general concept of internal validity in research methods. In research methods, a study is internally valid when the research design rules out alternative explanations for a cause and effect relationship. Relatedly, in research methods, external validity relates to how generalizable the study findings are to a population beyond the study’s sample.
Are antimuscarinic effects common in hydroxyzine overdose? A cohort analysis of antimuscarinic effects in hydroxyzine and diphenhydramine-poisoned patients
Published in Clinical Toxicology, 2023
Furthermore, threats to external validity are those that affect whether a study is generalizable to daily practice [23]. The NPDS was chosen to help with generalizability as it captures all calls made to United States poison centers. However, some exposures are not reported to poison centers. Again, there is no reason to believe that there are categorical differences in hydroxyzine and diphenhydramine reporting. Furthermore, pediatric patients were excluded from the NPDS exposures in an attempt to minimize the effect of minor exploratory ingestions on the data. Because of this, we chose to include all ages from the ToxIC Core dataset to capture pediatric cases that developed findings. The limited pediatric sample size means that these results do not allow us to draw conclusions on pediatric patients, and further research is required in the pediatric population. Finally, the data from ToxIC does not represent all exposures like the NPDS. The ToxIC registry only captures patients who had a medical toxicology consultation. In most cases, this implies that patients were more complicated or ill, for which a primary team sought a formal expert consultation. Therefore, it may be expected that this dataset captures a sicker subset of exposures. Finally, some patients could have been present in both datasets. However, the data from these datasets were not pooled, and the calculations were done separately.
Prescription opioid misuse among university students: A systematic review
Published in Journal of American College Health, 2022
Lisa L. Weyandt, Bergljot Gyda Gudmundsdottir, Emily Z. Holding, Marisa E. Marraccini, Megan Keith, Shannon E. May, Emily Shepard, Alyssa Francis, Elizabeth D. Wilson, Isabella Channell, Caroline Sweeney
Two independent reviewers assessed the quality of the included studies using a validated 27-item checklist that has high internal consistency, inter-rater reliability, and test-retest reliability.31 The checklist includes items addressing quality of reporting, external validity, internal validity, selection bias, randomization, and statistical power, with items scored as either yes (1), no (0), or unable to determine (0). Because the designs of studies included in this review varied considerably, including observational and intervention studies, a code of “not applicable” was used for items that did not relate to the study design and was not weighted in the final score (for example, if questions asked about treatment or intervention explicitly this item was coded as not applicable for observational studies). Accordingly, we calculated quality scores by averaging the individual binary scores to generate a scale from 0-1, with 1 indicating the highest quality and 0 the lowest. We report on the following domains for all studies: reporting (e.g., are the characteristics of the participants clearly described?) external validity, (e.g., were the participants representative of the entire population?), and statistical power. For intervention studies, we also calculated internal validity bias (e.g., was an attempt made to blind participants to the intervention they received?).
Vasoactive pharmacologic therapy in cardiogenic shock: a critical review
Published in Journal of Drug Assessment, 2021
Rasha Kaddoura, Amr Elmoheen, Ehab Badawy, Mahmoud F. Eltawagny, Mohamed A. Seif, Khalid Bashir, Amar M. Salam
The assessment of methodological quality using a reliable and valid instrument (e.g. Jadad scale) is essential to capture the variations in the quality of the studies which may affect the overall conclusion of the results30. Quality assessment investigates the validity constructs of a study which includes the internal validity (i.e. study’s methods), external validity (i.e. study’s results), and statistical analysis97. The elements that have shown to change the treatment effects are lack of randomization98, inadequate allocation concealment99,100, absence of blinding100, and inadequate sample size101,102. The quality across the studies in this review was variable. Randomization and blinding are challenging in the context of CS. Thus, such obstacles would probably be reflected in the slow enrollment that causes study termination or withdrawal. Basic methodological standards support the consideration of other elements that may empirically influence the quality of the study. Elements including appropriate patient disposition description, inclusion, and exclusion criteria definitions, statistical analysis, outcomes objectivity, ITT, and baseline comparability can affect the quality of a study as well. However, this was not supported by respective studies30.