Normal Curves: Sexy Science, Serious Statistics

Does the temperature of your coffee six months ago really predict whether you feel gassy today? This week we dissect a new nutrition survey study on hot and cold beverage habits that claims to connect drink temperature with gut symptoms, anxiety, and more—despite relying on year-old memories and a blizzard of statistical tests. It’s the perfect case study for our Holiday Survival Guide Part 2, where we teach you how to talk with Uncle Joe at the dinner table about one of the most common—and most fraught—study designs in science: cross-sectional surveys. We walk through our easy checklist for making sense of results, show how recall bias and measurement error can skew the story, and reacquaint you with nonmonogamous Multiple-Testing Dude, who’s been very busy in this dataset. A friendly, practical guide to spotting when researchers are just torturing the data until it confesses.Statistical topicsConfoundingCross-sectional studiesFalse positivesMeasurement errorMultiple testingPICOT / PIVOT frameworkRecall biasResearch hypothesesSample size and powerSignal vs. noiseSMART frameworkStatistical significanceSubgroup analysesSurvey designTransparency and trustworthinessMethodological morals“When your measurement starts with ‘think back to last winter’ you might as well use a random number generator.”“If the effect is only significant in certain subgroups in certain seasons for certain outcomes, it might just be a bad case of gas.”ReferencesWu T, Doyle C, Ito J, et al. Cold Exposures in Relation to Dysmenorrhea among Asian and White Women. Int J Environ Res Public Health. 2023;21(1):56. Published 2023 Dec 30. doi:10.3390/ijerph21010056Wu T, Ramesh N, Doyle C, Hsu FC. Cold and hot consumption and health outcomes among US Asian and White populations. Br J Nutr. Published online September 18, 2025. doi:10.1017/S000711452510514XKristin and Regina’s online courses: Demystifying Data: A Modern Approach to Statistical Understanding Clinical Trials: Design, Strategy, and Analysis Medical Statistics Certificate Program Writing in the Sciences Epidemiology and Clinical Research Graduate Certificate Program Programs that we teach in:Epidemiology and Clinical Research Graduate Certificate Program Find us on:Kristin - LinkedIn & Twitter/XRegina - LinkedIn & ReginaNuzzo.com(00:00) - Intro (04:36) - Did they have real research hypotheses? (10:29) - Observational or randomized experiment? (20:09) - PICOT and PIVOT (26:20) - Memory problems (32:03) - Five outcomes and measurement problems therein (36:56) - SMART (41:50) - Multiple Testing Dude is having a great time (52:36) - How big is the effect? (59:06) - Wrap-up and Irish Coffee rating scale

What do chickenpox and shingles have to do with your brain? This week, we dig into two 2025 headline-grabbing studies that link the shingles shot to lower dementia rates. We start in Wales, where a birthday cutoff turned into the perfect natural experiment, and end in the U.S. with a multi-million-person megastudy. Featuring bias-variance Goldilockses, Fozzy-the-Bear regression discontinuities, a Barbie-versus-Oppenheimer showdown for propensity scores – and the hottest rebrand of inverse-probability weighting you’ll ever hear.Statistical topicsAbsolute vs. relative riskBias–variance tradeoffCausal inferenceCensoringConfoundingFuzzy regression discontinuity designHealthy-user biasInverse probability of treatment weighting (IPTW)Longitudinal studyNatural experimentNegative controlsOptimal bandwidthPropensity scoresSelection biasSubgroup analysisTriangular kernel weightsMethodological morals“Propensity scores are the lipstick you put on observational pigs.”“Natural experiments are a hot flirtation date with causality.”ReferencesEyting M, Xie M, Michalik F, Heß S, Chung S, Geldsetzer P. A natural experiment on the effect of herpes zoster vaccination on dementia. Nature. 2025 May;641(8062):438-446. doi: 10.1038/s41586-025-08800-x. Epub 2025 Apr 2. PMID: 40175543; PMCID: PMC12058522.Polisky V, Littmann M, Triastcyn A, et al. Varicella-zoster virus reactivation and the risk of dementia. Nat Med. Published online October 6, 2025. doi:10.1038/s41591-025-03972-5Sainani KL. Propensity scores: uses and limitations. PM&R 2012; 4:693-97.Detailed Show Notes PageKristin and Regina’s online courses: Demystifying Data: A Modern Approach to Statistical Understanding Clinical Trials: Design, Strategy, and Analysis Medical Statistics Certificate Program Writing in the Sciences Epidemiology and Clinical Research Graduate Certificate Program Programs that we teach in:Epidemiology and Clinical Research Graduate Certificate Program Find us on:Kristin - LinkedIn & Twitter/XRegina - LinkedIn & ReginaNuzzo.com(00:00) - Intro and first gratuitous mention of sex (03:56) - What are shingles, chickenpox, and the vaccines against them? (12:30) - Fun facts about the varicella zoster and herpes viruses (18:00) - A natural experiment in Wales (21:54) - What is the Goldilocks optimal bandwidth? (26:17) - Fuzzy regression discontinuity design demystified (32:43) - Shingles vaccine vs dementia showdown (34:13) - Absolute risk reduction paradox (37:44) - Effects for men and women differ (41:07) - A giant longitudinal study (47:51) - Propensity scores demystified via Barbie and Oppenheimer (53:55) - Using propensity scores to make matches (58:08) - Inverse probability of treatment weighting demystified via more Barbenheimer (01:02:27) - Attempts to rename IPTW for TikTok (01:05:59) - Longitudinal study results (01:10:00) - Smooch ratings and methodological morals: pigs and hot dates

Ultramarathoners push their bodies to the limit, but can a giant pre-race dose of vitamin D really keep their bones from breaking down? In this episode, we dig into a trial that tested this claim – and found a statistical endurance event of its own: six highly interchangeable papers sliced from one small study. Expect missing runners, recycled figures, and a peer-review that reads like stand-up comedy, plus a quick lesson in using degrees of freedom as your statistical breadcrumbs.Statistical topicsData cleaning and validationDegrees of freedomExploratory vs confirmatory analysisFalse positives and Type I errorIntention-to-treat principleMultiple testingOpen data and transparencyP-hackingSalami slicingParametric vs non-parametric testsPeer review qualityRandomized controlled trialsResearch reproducibilityStatistical sleuthingMethodological morals“Degrees of freedom are the breadcrumbs in statistical sleuthing. They reveal the sample size even when the authors do not.”“Publishing the same study again and again with only the outcomes swapped is Mad Libs Science, better known as salami slicing.”ReferencesBoswell, Rachel. Pre-race vitamin D could do wonders for ultrarunners’ bone health, according to science. Runner’s World. September 25, 2025. Mieszkowski J, Stankiewicz B, Kochanowicz A, et al. Ultra-Marathon-Induced Increase in Serum Levels of Vitamin D Metabolites: A Double-Blind Randomized Controlled Trial. Nutrients. 2020;12(12):3629. Published 2020 Nov 25. doi:10.3390/nu12123629Mieszkowski J, Borkowska A, Stankiewicz B, et al. Single High-Dose Vitamin D Supplementation as an Approach for Reducing Ultramarathon-Induced Inflammation: A Double-Blind Randomized Controlled Trial. Nutrients. 2021;13(4):1280. Published 2021 Apr 13. doi:10.3390/nu13041280Mieszkowski J, Brzezińska P, Stankiewicz B, et al. Direct Effects of Vitamin D Supplementation on Ultramarathon-Induced Changes in Kynurenine Metabolism. Nutrients. 2022;14(21):4485. Published 2022 Oct 25. doi:10.3390/nu14214485Mieszkowski J, Brzezińska P, Stankiewicz B, et al. Vitamin D Supplementation Influences Ultramarathon-Induced Changes in Serum Amino Acid Levels, Tryptophan/Branched-Chain Amino Acid Ratio, and Arginine/Asymmetric Dimethylarginine Ratio. Nutrients. 2023;15(16):3536. Published 2023 Aug 11. doi:10.3390/nu15163536Stankiewicz B, Mieszkowski J, Kochanowicz A, et al. Effect of Single High-Dose Vitamin D3 Supplementation on Post-Ultra Mountain Running Heart Damage and Iron Metabolism Changes: A Double-Blind Randomized Controlled Trial. Nutrients. 2024;16(15):2479. Published 2024 Jul 31. doi:10.3390/nu16152479Stankiewicz B, Kochanowicz A, et al. Single high-dose vitamin D supplementation impacts ultramarathon-induced changes in serum levels of bone turnover markers: a double-blind randomized controlled trial. J Int Soc Sports Nutr. 2025 Dec;22(1):2561661. doi: 10.1080/15502783.2025.2561661.Kristin and Regina’s online courses: Demystifying Data: A Modern Approach to Statistical Understanding Clinical Trials: Design, Strategy, and Analysis Medical Statistics Certificate Program Writing in the Sciences Epidemiology and Clinical Research Graduate Certificate Program Programs that we teach in:Epidemiology and Clinical Research Graduate Certificate Program Find us on:Kristin - LinkedIn & Twitter/XRegina - LinkedIn & ReginaNuzzo.com 00:00 Intro & claim of the episode 00:44 Runner’s World headline: Vitamin D for ultramarathoners 02:03 Kristin’s connection to running and vitamin D skepticism 03:32 Ultramarathon world—Regina’s stories and Death Valley race 06:29 What ultramarathons do to your bones 08:02 Boy story: four stress fractures in one race 10:00 Study design—40 male runners in Poland 11:33 Missing flow diagram and violated intention-to-treat 13:02 The intervention: 150,000 IU megadose 15:09 Blinding details and missing randomization info 17:13 Measuring bone biomarkers—no primary outcome specified 19:12 The wrong clinicaltrials.gov registration 20:35 Discovery of six papers from one dataset (salami slicing) 23:02 Why salami slicing misleads readers 25:42 Inconsistent reporting across papers 29:11 Changing inclusion criteria and sloppy methods 31:06 Typos, Polish notes, and misnumbered references 32:39 Peer review comedy gold—“Please define vitamin D” 36:06 Reviewer laziness and p-hacking admission 39:13 Results: implausible bone growth mid-race 41:16 Degrees of freedom sleuthing reveals hidden sample sizes 47:07 Open data? Kristin emails the authors 48:42 Lessons from Kristin’s own ultramarathon dataset 51:22 Fishing expeditions and misuse of parametric tests 53:07 Strength of evidence: one smooch each 54:44 Methodologic morals—Mad Libs Science & degrees of freedom breadcrumbs 56:12 Anyone can spot red flags—trust your eyes 57:34 Outro: skip the vitamin D shot before your next run

Exercise has long been hailed as cancer-fighting magic, but is there hard evidence behind the hype? In this episode, we tackle the CHALLENGE trial, a large phase III study of colon cancer patients that tested whether prescribed exercise could improve cancer-free survival. We translate clinical jargon into plain English, show why ratio statistics make splashy headlines while absolute differences tell the real story, and take a detour into why statisticians think survival analysis is downright sexy. And we even bring in a classic reality show to make sense of the numbers.Statistical topicsData and Safety Monitoring Board (DSMB)Hazard ratiosIntention-to-treat analysisInterim analysesKaplan-Meier curvesPhase III trialsRandomized clinical trialRates and rate ratiosRelative vs absolute differencesStratified randomization with minimizationSurvival analysisTime-to-event variablesMethodological morals“Ratio statistics sell headlines. Absolute differences sell truth.”“Survival analysis is this sexy stats tool that makes every moment and every Cox count.”ReferencesCourneya KS, Vardy JL, O'Callaghan CJ, et al. Structured Exercise after Adjuvant Chemotherapy for Colon Cancer. NEJM. 2025;393:13-25. Rabin RC. Are Marathons and Extreme Running Linked to Colon Cancer? The New York Times. Aug 19, 2025.Sainani KL. Introduction to survival analysis. PM&R. 2016; 8:580-85.Sainani KL. Making sense of intention-to-treat. PM&R. 2010;2:209-13.ThanksThanks to Caitlin Goodrich for the episode topic tip!Kristin and Regina’s online courses: Demystifying Data: A Modern Approach to Statistical Understanding Clinical Trials: Design, Strategy, and Analysis Medical Statistics Certificate Program Writing in the Sciences Epidemiology and Clinical Research Graduate Certificate Program Programs that we teach in:Epidemiology and Clinical Research Graduate Certificate Program Find us on:Kristin - LinkedIn & Twitter/XRegina - LinkedIn & ReginaNuzzo.com(00:00) - Intro (05:42) - Two different types of cancer studies (08:12) - Why might exercise affect cancer? (10:05) - Phase III trials are different (12:40) - Who was in the CHALLENGE trial? (13:31) - Stratified randomization with minimization (15:05) - The exercise prescription (19:20) - What did the CHALLENGE trial measure? (20:07) - Disease-free survival (22:02) - Data and Safety Monitoring Board – what do they do? (24:38) - Participants and adherence to exercise (26:57) - Intention-to-treat analysis (30:01) - Survival analysis overview (31:54) - Kaplan-Meier curves (34:30) - Reality-show analogy (36:57) - Ratio statistics are confusing (39:33) - Hazard ratios (47:06) - Wrap-up, rating, and methodological morals

ChatGPT is melting our brainpower, killing creativity, and making us soulless — or so the headlines imply. We dig into the study behind the claims, starting with quirky bar charts and mysterious sample sizes, then winding through hairball-like brain diagrams and tens of thousands of statistical tests. Our statistical sleuthing leaves us with questions, not just about the results, but about whether this was science’s version of a first date that looked better on paper.Statistical topicsANOVABar graphsData visualization False Discovery Rate correctionMultiple testingPreprintsStatistical SleuthingMethodological morals"Treat your preprints like your blind dates. Show up showered and with teeth brushed.""Always check your N. Then check it again.""Never make a bar graph that just shows p-values. Ever."Link to paperKristin and Regina’s online courses: Demystifying Data: A Modern Approach to Statistical Understanding Clinical Trials: Design, Strategy, and Analysis Medical Statistics Certificate Program Writing in the Sciences Epidemiology and Clinical Research Graduate Certificate ProgramPrograms that we teach in:Epidemiology and Clinical Research Graduate Certificate Program Find us on:Kristin - LinkedIn & Twitter/XRegina - LinkedIn & ReginaNuzzo.com(00:00) - Intro (03:46) - Media coverage of the study (08:35) - The experiment (12:09) - Sample size issues (13:11) - Bar chart sleuthing (19:15) - Blind date analogy (23:54) - Interview results (30:04) - Simple text analysis results (34:04) - Natural language processing results (41:00) - N-gram and ontology analysis results (45:55) - Teacher evaluation results (52:30) - Neuroimaging analysis (01:00:32) - Multiple testing and connectivity issues (01:06:10) - Brain adaptation results (01:09:47) - Wrap-up, rating, and methodological morals

Loyal, funny, hot — you’ve probably got a wish list for your dream partner. But does checking all your boxes actually lead to happily ever after? In this episode, we dive into a massive global study that put the “ideal partner” hypothesis to the test. Do people really know what they want, and does getting it actually make them happier? We explore surprising statistical insights from over 10,000 romantics in 43 countries, from mean-centering and interaction effects to the good-catch confounder. Along the way, we dig into dessert metaphors, partner boat-count regression models, and the one trait that people say doesn’t matter — but secretly makes them happiest.Statistical topicsRegressionRandom Slopes and Intercepts (Random Effects) in RegressionStandardized Beta Coefficients in RegressionInteraction Effects in RegressionMean CenteringExploratory AnalysesMethodological morals“Good science bares it all.”“When the world isn't one size fits all, don't fit just one line; use random slopes and intercepts.”ReferencesEastwick PW, Sparks J, Finkel EJ, Meza EM, Adamkovič M, Adu P, Ai T, Akintola AA, Al-Shawaf L, Apriliawati D, Arriaga P, Aubert-Teillaud B, Baník G, Barzykowski K, Batres C, Baucom KJ, Beaulieu EZ, Behnke M, Butcher N, Charles DY, Chen JM, Cheon JE, Chittham P, Chwiłkowska P, Cong CW, Copping LT, Corral-Frias NS, Ćubela Adorić V, Dizon M, Du H, Ehinmowo MI, Escribano DA, Espinosa NM, Expósito F, Feldman G, Freitag R, Frias Armenta M, Gallyamova A, Gillath O, Gjoneska B, Gkinopoulos T, Grafe F, Grigoryev D, Groyecka-Bernard A, Gunaydin G, Ilustrisimo R, Impett E, Kačmár P, Kim YH, Kocur M, Kowal M, Krishna M, Labor PD, Lu JG, Lucas MY, Małecki WP, Malinakova K, Meißner S, Meier Z, Misiak M, Muise A, Novak L, O J, Özdoğru AA, Park HG, Paruzel M, Pavlović Z, Püski M, Ribeiro G, Roberts SC, Röer JP, Ropovik I, Ross RM, Sakman E, Salvador CE, Selcuk E, Skakoon-Sparling S, Sorokowska A, Sorokowski P, Spasovski O, Stanton SCE, Stewart SLK, Swami V, Szaszi B, Takashima K, Tavel P, Tejada J, Tu E, Tuominen J, Vaidis D, Vally Z, Vaughn LA, Villanueva-Moya L, Wisnuwardhani D, Yamada Y, Yonemitsu F, Žídková R, Živná K, Coles NA. A worldwide test of the predictive validity of ideal partner preference matching. J Pers Soc Psychol. 2025 Jan;128(1):123-146. doi: 10.1037/pspp0000524Love Factually Podcast: https://www.lovefactuallypod.com/Kristin and Regina’s online courses: Demystifying Data: A Modern Approach to Statistical Understanding Clinical Trials: Design, Strategy, and Analysis Medical Statistics Certificate Program Writing in the Sciences Epidemiology and Clinical Research Graduate Certificate Program Programs that we teach in:Epidemiology and Clinical Research Graduate Certificate Program Find us on:Kristin - LinkedIn & Twitter/XRegina - LinkedIn & ReginaNuzzo.com(00:00) - (00:00) - Intro (04:57) - Actual dating profile wishlists vs study wishlists (09:12) - Juicy paper details (18:31) - What the study actually asked – wishlist, partner resume, relationship satisfaction (25:07) - Linear regression illustrated through number of boats your partner has (31:34) - Standardized regression coefficients illustrated through spouse height concordance (35:49) - Good catch confounder: We all just want the same high-quality ice cream / mate (40:43) - Does your personalized wishlist matter? Results (42:58) - Wishlist regression interaction effects: like chocolate and peanut butter (46:48) - Partner traits result in happiness bonus points (50:48) - What do we say we want – and what really makes us happy? Surprise (55:07) - Gender stereotypes and whether they held up (57:48) - Random effects models and boats again (01:00:27) - Other cool things they did (01:01:38) - One-minute paper summary (01:03:20) - Wrap-up, rate the claim, methodological morals

Could a preteen vaccine wipe out a global cancer? In this episode, we examine the bold claim that cervical cancer could be eradicated in much of the world by the end of the century—thanks to the highly effective HPV vaccine. We unpack statistical modeling, microsimulations, and how Markov chains make good date-night conversation. We also explore why vaccine uptake has been uneven, how a splash of vinegar is helping screen for cancer in low-resource countries, and why HPV isn’t just a women’s issue—it now causes more cancer in men than in women. Plus: dangerously tight corsets, allegedly breast-squeezing nuns, and the Cosmo quote we wish we’d written ourselves.Statistical topics:Cancer surveillanceMarkov modelsMicrosimulation modelsSensitivity analysesPassive surveillanceBackground ratesCase reports and case seriesMethodologic morals:“When reality is too complex to test, let microsimulations do the rest.”“Case reports are medicine's equivalent to see something, say something. They call for hard data, not hysteria.”Citations:No cervical cancer cases detected in vaccinated women following HPV immunisation. University of Strathclyde, January 22, 2024.Palmer TJ, Kavanagh K, Cuschieri K, et al. Invasive cervical cancer incidence following bivalent human papillomavirus vaccination: a population-based observational study of age at immunization, dose, and deprivation. J Natl Cancer Inst. 2024;116:857-65.Rigoni-Stern. Statistical facts about cancers on which Doctor Rigoni-Stern based his contribution to the Surgeons' Subgroup of the IV Congress of the Italian Scientists on 23 September 1842. (translation). Stat Med. 1987;6:881-4.Gordan JA, Lenkei SC. Cleanliness, Continence, Constancy, and Cervical Carcinoma. Can Med Assoc J. 1964;90:1132.zur Hausen H. Condylomata acuminata and human genital cancer. Cancer Res. 1976;36:794.Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol. 1999;189:12-9.Chesson HW, Dunne EF, Hariri S, et al. The estimated lifetime probability of acquiring human papillomavirus in the United States. Sex Transm Dis. 2014;41:660-4.Sullivan, Morgan. Let’s Have a Little Chat About the HPV Vaccine. Cosmopolitan. March 19, 2025.Burger EA, Kim JJ, Sy S, et al. Age of Acquiring Causal Human Papillomavirus (HPV) Infections: Leveraging Simulation Models to Explore the Natural History of HPV-induced Cervical Cancer. Clin Infect Dis. 2017;65:893-99.Canfell K. Towards the global elimination of cervical cancer. Papillomavirus Res. 2019;8:100170.World Health Organization. Global strategy to accelerate the elimination of cervical cancer as a public health problem. November 17, 2020.Hall MT, Simms KT, Lew JB, et al. The projected timeframe until cervical cancer elimination in Australia: a modelling study. Lancet Public Health. 2019;4:e19-e27.Burger EA, Smith MA, Killen J, et al. Projected time to elimination of cervical cancer in the USA: a comparative modelling study. Lancet Public Health. 2020 Apr;5(4):e213-e222.Brisson M, Kim JJ, Canfell K, et al. Impact of HPV vaccination and cervical screening on cervical cancer elimination: a comparative modelling analysis in 78 low-income and lower-middle-income countries. Lancet. 2020;395:575-90.Escabí-Wojna E, Alvelo-Fernández PM, Suárez E, et al. Sex differences in parental reasons for lack of intent to initiate HPV vaccination among adolescents ages 13-17 years: National Immunization Survey - Teen 2019-2021. Vaccine. 2025;44:126584. (see supplement) Szilagyi PG, Albertin CS, Gurfinkel D, et al. Prevalence and characteristics of HPV vaccine hesitancy among parents of adolescents across the US. Vaccine. 2020;38:6027-6037.LaPook, Jonathan. Is the HPV Vaccine Safe? CBS Evening News. August 18, 2009.Slade BA, Leidel L, Vellozzi C, et al. Postlicensure safety surveillance for quadrivalent human papillomavirus recombinant vaccine. JAMA. 2009;302:750-7.Kharabsheh S, Al-Otoum H, Clements J, et al. Mass psychogenic illness following tetanus-diphtheria toxoid vaccination in Jordan. Bull World Health Organ. 2001;79:764-70.Jones TF, Craig AS, Hoy D, et al. Mass psychogenic illness attributed to toxic exposure at a high school. N Engl J Med. 2000;342:96-100.Buttery JP, Madin S, Crawford NW, et al. Mass psychogenic response to human papillomavirus vaccination. Med J Aust. 2008;189:261-2.Clements CJ. Gardasil and mass psychogenic illness. Aust N Z J Public Health. 2007;31:387.Simas C, Munoz N, Arregoces L, et al. HPV vaccine confidence and cases of mass psychogenic illness following immunization in Carmen de Bolivar, Colombia. Hum Vaccin Immunother. 2019;15:163-66.Larson HJ. Japan's HPV vaccine crisis: act now to avert cervical cancer cases and deaths. Lancet Public Health. 2020;5:e184-e185.Brinth LS, Pors K, Theibel AC, Mehlsen J. Orthostatic intolerance and postural tachycardia syndrome as suspected adverse effects of vaccination against human papilloma virus. Vaccine. 2015;33:2602-5.Large well-done studies following up on case reports and passive surveillance:Phillips A, Hickie M, Totterdell J, Brotherton J, Dey A, Hill R, Snelling T, Macartney K. Adverse events following HPV vaccination: 11 years of surveillance in Australia. Vaccine. 2020;38:6038-46.Arnheim-Dahlström L, Pasternak B, Svanström H, et al.