Statistical Ethics in Research

#Dissertation Season is Here—But Where are the Research Ethics? It’s that time of the year when my inbox (and probably yours too) is flooded with Google Forms for research participation. While I love seeing students and researchers actively collecting data, there's something crucial missing in many of these forms: #ResearchEthics. Before hitting ‘Send,’ make sure your Google Form includes: ✔ Informed Consent: For minors (below 18 years): Parental #consent + Child’s #assent (both are required). For adults (18+): A clear consent form detailing participation terms. ✔ Full Researcher Identification: Mention your name, affiliation, #designation, and supervisor’s details (if applicable). Provide a contact email for participant queries. ✔ #Debriefing Statement: Participants should know: Why are they filling this form? What will happen to their data? If deception is used, provide an explanation after participation. ✔ #Confidentiality & Data Protection: Anonymity vs. Confidentiality: Clearly state whether responses are anonymous or confidential. Ensure data storage is secure and inform participants about how long data will be retained. ✔ Right to #Withdraw: Participants should not feel obligated to complete the survey. Inform them that they can exit at any time without penalties. ✔ Fair #Compensation (if applicable): If incentives are offered, clarify eligibility, amount, and disbursement method. ✔ Avoid Leading or Loaded Questions: Ensure that your #survey design doesn’t influence responses or force participants into biased answers. Ethical research isn’t just about ticking boxes—it’s about integrity, respect, and responsibility towards your participants. Let’s make sure our research meets the standards it truly deserves.

The American Statistical Association's Health Policy Statistics Section is raising serious concerns about a proposed NIH/CMS research project on Autism Spectrum Disorder (ASD) (https://lnkd.in/dYfzFr_F). Specifically, the outlined research program by NIH/CMS has serious methodological flaws for prospective research, including: *The sample is unrepresentative: The study only examines Medicaid-eligible families, excluding families with private insurance *The data source is inappropriate: It uses Medicare/Medicaid data designed for aging studies, not autism research affecting primarily children *The data will have gaps: Medicaid's "revolving door" creates data gaps as families change insurance status. *Aspects of the study violate basic ethical guidelines: If the study includes individuals who are younger than 18 years, this would violate the rights of minors with ASD who cannot provide informed consent. As a result, the study cannot determine autism trends across all populations, assess intervention effectiveness fairly, measure real economic burden on families, and investigate actual causes of ASD. That is, the study’s design does not support the study’s stated goals. But there are good alternatives. Social Security disability services data are a strong data source, as are national/state educational databases. Studies can be designed that include both Medicaid AND privately insured populations and do not require a registry. We ask Congress and the Administration to fix these flaws. Our children with autism deserve rigorous, ethical research that can effectively inform policy and improve care—not flawed studies that risk wasting resources and perpetuating inequities. Contact your representatives today. Science integrity matters. Our families matter. #AutismResearch #Congress #NIH #HealthPolicy #DataScience #Statistics #PublicHealth #AutismAdvocacy

Please do NOT start research on human subjects unless you have taken into account the ethics part. I beg you, please! 😂 I've encountered multiple cases of my mentees who started a project without the necessary approvals, and when it came to journal publication, they were stuck! Let's see what we need to get started 👇 1. Informed consent Ensures participants fully understand the research, its potential risks and benefits, and their right to withdraw without consequence (you must include this in your submission!) 2. Privacy and confidentiality Safeguarding participant data, including anonymization, encryption, and secure storage (you'll have to describe this in your method section.) 3. Vulnerable populations If research involves children, the elderly, prisoners, or those with cognitive impairments, additional measures protect their rights and well-being. 4. Benefit-risk assessment Potential benefits or risks to participants considering not only physical harm but also psychological and social impacts. 5. Data integrity and transparency Accurate data collection, analysis, and reporting. 6. Researcher bias and conflicts of interest Addressing personal biases and financial conflicts and transparent disclosure and mitigation strategies. 7. Cultural sensitivity Respecting diverse cultural values and beliefs AND, here comes the tough one 👇 8. Institutional review board (IRB) approval An approval letter generated by an IRB is compulsory for every single submission that involves research on human subjects. ___________________ 🔔 This is Dr. Samira Hosseini. Scholars who took my training published +2,000 articles in top-tier journals. Join my inner circle not to miss even one single bit of learning: https://lnkd.in/eVNSihCM

Excited to share insights from a paper by Virginia Acha et al. on the principles for good practice in the conduct of non-interventional studies (#NIS). This comprehensive reflection paper emphasizes the importance of robust research practices in generating real-world evidence (#RWE) to inform healthcare decisions. 🔍 Key Highlights: Definition and Scope: Non-interventional studies (NIS) involve participants receiving routine clinical care without specific treatment assignments per protocol, crucial for assessing treatment effects in real-world settings. Research Design: The choice of research design should align with the research question's nature and intent, emphasizing clear communication and transparency. Study Protocol: The study protocol, a cornerstone of the research process, must be well-defined before commencement, detailing population selection criteria, key variables, and data sources. Data Quality: High-quality, fit-for-purpose data access is essential, requiring validation and reliability to address research questions effectively. Analytical Methods: Appropriate and verifiable analytical methods are crucial, with a growing trend towards advanced analytics, including AI/ML methods. Bias Reduction: Transparent and reproducible scientific study planning is key to minimizing bias, including confounding factors. Transparency: Ensuring transparency in study conduct and reporting supports the credibility of NIS, with registration and clear reporting enhancing confidence in the research process. Privacy and Ethics: Adherence to privacy requirements and ethics review is vital in NIS to protect individual rights and ensure regulatory compliance.