Written by: Nayaonika Vasishtha
Edited by: Isabel Angres
Edited by: Isabel Angres
Selection bias refers to the distortion introduced in research studies when the process of participant selection systematically deviates from a truly representative sample. It occurs when the individuals chosen for a study do not accurately represent the target population, leading to biased and potentially misleading conclusions. Selection bias can arise from various factors, including flawed sampling methods, participant non-response, or exclusion criteria that disproportionately affects certain groups. In research, it is crucial to strive for representative samples that adequately reflect the population of interest to ensure the validity and generalizability of findings. Failure to address selection bias can compromise the reliability of research outcomes and hinder progress in understanding and addressing complex issues.
The type of selection bias this article focuses on is sampling bias. Sampling bias arises when the method of selecting participants leads to a non-representative sample. For instance, convenience sampling and purposive sampling can introduce biases by favoring easily accessible individuals or specific characteristics. Sampling bias compromises the accuracy and external validity of study findings. Since they are not representative of the entire population, results inferred through such research may not be true for everyone and can have negative effects on certain individuals’ lives (Shringarpure and Xing 2014). Non-random sampling, also known as non-probability sampling, is a method of selecting participants for a study that does not involve random selection. Unlike random sampling, which provides each member of the target population with an equal chance of being included, non-random sampling relies on subjective criteria and deliberate selection methods.
Such biases have ethical implications in the medical field that continue to affect lives for generations. An important example to discuss here is the research behind the development of the pulse oximeter that is used to measure blood oxygen levels. Oxygen saturation levels tell us the amount of hemoglobin that is bound to oxygen compared to how much is unbound. Along with temperature, heart rate, breathing rate, and blood pressure, blood oxygen is known as the fifth vital sign. Pulse oximeters were indispensable during the COVID-19 pandemic to monitor conditions and prioritize patients. Its alternative, arterial blood gas, is an invasive procedure that measures oxygen levels in blood taken from arteries. This procedure is limited to emergency situations as pulse oximeters are used more frequently to measure blood oxygen levels.
The sensors in pulse oximeters use a technology called pulse oximetry, which measures the amount of oxygen carried by red blood cells by passing light through the skin. Darker skin tones have higher levels of melanin, which absorbs and scatters light differently than lighter skin tones. Consequently, the pulse oximeter may struggle to accurately detect and measure the light absorbed and reflected by darker skin, resulting in inaccurate readings of oxygen saturation levels (Valbuena et al. 2020). Studies suggest that pulse oximeters underestimate hypoxia in patients with darker skin tone by overestimating their oxygen saturation (Brownscombe et al. 2022). Such errors in measurement are detrimental to the health of patients as they can lead to misdiagnosis and improper treatment.
While non-random sampling may be more practical and cost-effective in certain research contexts, it introduces the risk of selection bias and limits the generalizability of findings to the larger population. Researchers using non-random sampling must exercise caution and acknowledge the limitations of their results, as the sample may not fully represent the diversity and characteristics of the target population. Recognizing and mitigating selection bias is essential to uphold the integrity of scientific inquiry. It is imperative to promote evidence-based decision-making that upholds justice for underrepresented minorities in medicine in various fields, including healthcare, social sciences, and public policy.
References
Brownscombe JJ, Loane H, Honan B. 2022 May 6. COVID ‐19 highlights the need for action on pulse oximeter accuracy in people with dark skin. Medical Journal of Australia. doi:https://doi.org/10.5694/mja2.51522.
Shringarpure S, Xing EP. 2014. Effects of Sample Selection Bias on the Accuracy of Population Structure and Ancestry Inference. G3: Genes|Genomes|Genetics. 4(5):901–911. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4025489/#:~:text=It%20affects%20the%20internal%20validity,not%20generalize%20to%20the%20population.
Valbuena VSM, Barbaro RP, Claar D, Valley TS, Dickson RP, Gay SE, Sjoding MW, Iwashyna TJ. 2021 Sep. Racial Bias in Pulse Oximetry Measurement Among Patients About to Undergo Extracorporeal Membrane Oxygenation in 2019-2020. Chest. doi:https://doi.org/10.1016/j.chest.2021.09.025.
The type of selection bias this article focuses on is sampling bias. Sampling bias arises when the method of selecting participants leads to a non-representative sample. For instance, convenience sampling and purposive sampling can introduce biases by favoring easily accessible individuals or specific characteristics. Sampling bias compromises the accuracy and external validity of study findings. Since they are not representative of the entire population, results inferred through such research may not be true for everyone and can have negative effects on certain individuals’ lives (Shringarpure and Xing 2014). Non-random sampling, also known as non-probability sampling, is a method of selecting participants for a study that does not involve random selection. Unlike random sampling, which provides each member of the target population with an equal chance of being included, non-random sampling relies on subjective criteria and deliberate selection methods.
Such biases have ethical implications in the medical field that continue to affect lives for generations. An important example to discuss here is the research behind the development of the pulse oximeter that is used to measure blood oxygen levels. Oxygen saturation levels tell us the amount of hemoglobin that is bound to oxygen compared to how much is unbound. Along with temperature, heart rate, breathing rate, and blood pressure, blood oxygen is known as the fifth vital sign. Pulse oximeters were indispensable during the COVID-19 pandemic to monitor conditions and prioritize patients. Its alternative, arterial blood gas, is an invasive procedure that measures oxygen levels in blood taken from arteries. This procedure is limited to emergency situations as pulse oximeters are used more frequently to measure blood oxygen levels.
The sensors in pulse oximeters use a technology called pulse oximetry, which measures the amount of oxygen carried by red blood cells by passing light through the skin. Darker skin tones have higher levels of melanin, which absorbs and scatters light differently than lighter skin tones. Consequently, the pulse oximeter may struggle to accurately detect and measure the light absorbed and reflected by darker skin, resulting in inaccurate readings of oxygen saturation levels (Valbuena et al. 2020). Studies suggest that pulse oximeters underestimate hypoxia in patients with darker skin tone by overestimating their oxygen saturation (Brownscombe et al. 2022). Such errors in measurement are detrimental to the health of patients as they can lead to misdiagnosis and improper treatment.
While non-random sampling may be more practical and cost-effective in certain research contexts, it introduces the risk of selection bias and limits the generalizability of findings to the larger population. Researchers using non-random sampling must exercise caution and acknowledge the limitations of their results, as the sample may not fully represent the diversity and characteristics of the target population. Recognizing and mitigating selection bias is essential to uphold the integrity of scientific inquiry. It is imperative to promote evidence-based decision-making that upholds justice for underrepresented minorities in medicine in various fields, including healthcare, social sciences, and public policy.
References
Brownscombe JJ, Loane H, Honan B. 2022 May 6. COVID ‐19 highlights the need for action on pulse oximeter accuracy in people with dark skin. Medical Journal of Australia. doi:https://doi.org/10.5694/mja2.51522.
Shringarpure S, Xing EP. 2014. Effects of Sample Selection Bias on the Accuracy of Population Structure and Ancestry Inference. G3: Genes|Genomes|Genetics. 4(5):901–911. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4025489/#:~:text=It%20affects%20the%20internal%20validity,not%20generalize%20to%20the%20population.
Valbuena VSM, Barbaro RP, Claar D, Valley TS, Dickson RP, Gay SE, Sjoding MW, Iwashyna TJ. 2021 Sep. Racial Bias in Pulse Oximetry Measurement Among Patients About to Undergo Extracorporeal Membrane Oxygenation in 2019-2020. Chest. doi:https://doi.org/10.1016/j.chest.2021.09.025.
Proudly powered by Weebly