### Transcription

Estimating Powerand Sample Size(How to Help Your Biostatistician!)Amber W. Trickey, PhD, MS, CPHSenior Biostatistician1070 Arastradero #[email protected]

Goal: Effective Statistical Collaboration[Pye, 2016]

TopicsResearchData Questions & Measures Hypothesis TestingStatisticalPower Components AssumptionsStatisticalCollaboration Consultation Process Timelines

Surgery Epidemiologist & Biostatistician14years12years14yearsa whileago EpiPhD, MSStatsPhD, MSHSRPhDBioEngBSSurgical HSR: 7 years

Research Question (PICO)1. Patient population Condition / disease, demographics, setting, time2. Intervention Procedure, policy, process, treatment3. Comparison group Control group (e.g. no treatment, standard of care, non-exposed)4. Outcome of interest Treatment effects, patient-centered outcomes, healthcare utilization

Example Research Question Do hospitals with 200 beds perform better than smaller hospitals? More developed question: specify population & outcome Do large California hospitals 200 beds have lower surgical site infectionrates for adults undergoing inpatient surgical procedures? Population: California adults undergoing inpatient surgical procedureswith general anesthesia in 2017 Intervention (structural characteristic): 200 beds Comparison: smaller hospitals with 200 beds Outcome: surgical site infections within 30 days post-op

Internal & External ValidityExternal validity: generalizability to otherpatients & settings Study design Which patients are included How the intervention is implemented Real-world conditionsInternal validity: finding a true cause-effect relationship Study design analysis Specific information collected (or not) Data collection definitions Data analysis methods

Variable Types1. Exposure (Intervention) Predictor / Primary Independent variable (IV) Occurring first Causal relationship (?)2. Outcome Response / Dependent variable (DV) Occurring after predictorsConfounderCause1 IVExposure3. Confounders Related to both outcome and exposure Must be taken into account for internal validityEffectDVOutcome

Variable Measurement ScalesType ofMeasurementContinuousOrdered DiscreteCharacteristicsExamplesWeight, BMIRanked spectrum;quantifiableintervalsNumber of cigs / dayDescriptiveStatsInformationContentMean (SD) all belowHighestMean (SD) all ategoriesASA Physical StatusClassificationMedianIntermediateCategorical Nominal(Polychotomous)UnorderedCategoriesBlood Type, FacilityCounts,ProportionsLowerTwo categoriesSex (M/F),Obese (Y/N)Counts,ProportionsLowCategoricalBinary (Dichotomous)[Hulley 2007]

Measures of Central Tendency1. Mean average Continuous, normal distribution2. Median middle Continuous, nonparametric distribution3. Mode most common Categorical

Variability Averages are important, but variability iscritical for describing & comparingpopulations. Example measures:o SD “average” deviation from meano Range minimum – maximumo Interquartile range 25th - 75thpercentiles For skewed distributions(e.g. , time), range or IQR are morerepresentative measures of variabilitythan SD.Box Plot Components:range(75th percentile)(25th percentile)

HypothesisTesting

Hypothesis Testing Null Hypothesis (H0)o Default assumption for superiority studies Intervention/treatment has NO effect, i.e. no difference b/t groupso Acts as a “straw man”, assumed to be true so that it can be knocked downas false by a statistical test. Alternative Hypothesis (HA)o Assumption being tested for superiority studies Intervention/treatment has an effect Non-inferiority study hypotheses are reversed:alternative hypothesis no difference (within a specified range)

Error TypesProbability α 0.05Type I Error α: False positive Finding an effect that is not true Due to: Spurious association Solution: Repeat the studyType II Error (β): False negative Do not find an effect when one truly exists Due to: Insufficient power, high variability / measurement error Solution: Increase sample size

Hypothesis TestingOne- vs. Two-tailed TestsHA :M1 M2H0 :M1 M20Test StatisticEvaluate association in one directionM1 M2Two-sided testsalmost alwaysrequired – higherstandard, morecautiousTwo-sidedOne-sidedHA :0Test Statistic

P-value DefinitionThe p-value represents the probabilityof finding the observed,or a more extreme, test statisticif the null hypothesis is true.

P-ValueP-value measures evidence against H0 Smaller the p-value, the larger the evidence against H0 Reject H0 if p-value αPitfalls: The statistical significance of the effect does not explain the sizeof the effect Report descriptive statistics with p-values (N, %, means, SD, etc.) STATISTICAL significance does not equal CLINICAL significance P is not truly yes/no, all or none, but is actually a continuum P is highly dependent on sample size

Which Statistical Test?1.2.Number of IVsIVMeasurementScale3.Independentvs. MatchedGroups4.DVMeasurementScale

Common Regression fficientContinuousLinear RegressionSlope (β): How much the outcome increases forevery 1-unit increase in the predictorOdds Ratio (OR): How much the odds for the outcomeBinary /Logistic Regressionincreases for every 1-unit increase in the predictorCategoricalTime-toEventCox Proportional- Hazard Ratio (HR): How much the rate of the outcomeincreases for every 1-unit increase in the predictorHazards RegressionCountPoisson Regression Incidence Rate Ratio (IRR): How much the rate of theor Negative Binomialoutcome increases for every 1-unit increaseRegressionin the predictor

Hierarchical / Mixed Effects ModelsCorrelated Data Grouping of subjects Repeated measuresover time Multiple related outcomesCan handle Missing data Nonuniform measuresOutcome Variable(s) Categorical Continuous CountsNested DataLevel 3:HospitalsLevel 2:SurgeonsLevel 1:Patients

EstimatingPower

Error TypesType I Error (α): False positive Find an effect when it is truly not there Due to: Spurious association Solution: Repeat the studyProbability β 0.20Type II Error β: False negative Do not find an effect when one truly exists Due to: Insufficient power, high variability / measurement error Solution: Increase sample size

Statistical PowerA study with low power has a high probability of committing type II error. Power 1 – β (typically 1 – 0.2 0.8) Sample size planning aims to select a sufficient number of subjects to keepα and β low without making the study too expensive or difficult.How many subjects do I need to find a statistical & meaningful effect size? Sample size calculation pitfalls: Requires many assumptions Should focus on the minimal clinically important difference (MCID) If power calculation estimated effect size observed effect size,sample may be inadequate or observed effect may not be meaningful.

Statistical Power ToolsThree broad categories1. Hypothesis-based Formally testing a hypothesis to determine astatistically significant effect2. Confidence interval-based Estimating a number (e.g. prevalence) with a desired level of precision3. Rules of thumb Based on simulation studies, we estimate (ballpark) the necessary sample size Interpret carefully & in conjunction with careful sample size calculation usingmethod 1 or 2

Components of Power Calculations Outcome of interest Study design Effect Size Allocation ratio between groups Population variability Alpha (p-value, typically 0.05) Beta (1-power, typically 0.1-0.2) 1- vs. 2-tailed test

Effect Size Cohen’s d: comparison between two means d m1 – m2 / pooled SD Small d 0.2; Medium d 0.5; Large d 0.8 Expected values per group (e.g. complications: 10% open vs.3% laparoscopic) Minimal clinically important difference (e.g. 10% improvement) What is the MCID that would lead a clinician to change his/her practice? Inverse relationship with sample size effect size, sample size effect size, sample size

Confidence Interval-Based Power How precisely can you estimateyour measure of interest? Examples Diagnostic tests: Sensitivity / Specificity Care utilization rates Treatment adherence rates Calculation components NVariabilityα levelExpected outcomes

Rule of Thumb Power Calculations Simulation studies Degrees of freedom (df) estimates df: the number of IV factors that can vary in your regression model Multiple linear regression: 15 observations per df Multiple logistic regression: df # events/15 Cox regression: df # events/15 Best used with other hypothesis-based or confidenceinterval-based methods

Collaboration withBiostatisticians

Biostatistics Collaboration 2001 Survey of BMJ & Annals of Internal Medicine re: statisticaland methodological collaboration Stats/methodological support – how often? Biostatistician 53% Epidemiologist 32% Authorship outcomes given significant contribution Biostatisticians 78% Epidemiologists 96% Publication outcomes Studies w/o methodological assistance more likely to berejected w/o review: 71% vs. 57%, p 0.001[Altman, 2002]

Questions from your Biostatistician What is the research question? What is the study design? What effect do you expect to observe? What other variables may affect your results? How many patients are realistic? Do you have repeated measures per individual/analysis unit? What are your expected consent and follow-up completion rates? Do you have preliminary data? Previous studies / pilot data Published literature

Stages of Power CalculationStudy DesignHypothesisSimilarLiteratureSimulation/Rules of ThumbSample Size[Pye, 2016]Feasible?Important?Other Considerations?

Statistical Power Tips Seek biostatistician feedback early Calculations take time and typically a few iterations Without pilot data, it is helpful to identify previous research withsimilar methods If absolutely no information is available from a reasonable comparison study,you can estimate power from the minimal clinically important difference* Calculate power before the study is implemented Post hoc power calculations are less useful, unless to inform the next study Report estimated power as a range w/ varying assumptions/conditions*[Revicki, 2008]

AuthorshipAuthorshipInternational Committee of Medical Journal Editors (ICMJE) rules:All authors must have 1. Substantial contributions to the conception or design of the work; or theacquisition, analysis, or interpretation of data for the work; AND2. Drafting the work or revising it critically for important intellectual content; AND3. Final approval of the version to be published; AND4. Agreement to be accountable for all aspects of the work in ensuring thatquestions related to the accuracy or integrity of any part of the work areappropriately investigated and resolved. Epidemiologist/Biostatisticians typically qualify for authorship Sometimes an acknowledgement is appropriateMust be discussed

Consultation

S-SPIRE BiostatisticiansAmber Trickey,PhD, MS, CPHQian Ding, MSKelly Blum, MS

http://med.stanford.edu/s-spire/research.html

Timelines for Initial Consultation1. Conference abstract deadlines 4 weeks lead time with data ready for analysis (email 6 weeks out for appt)2. Special issue or meeting paper deadlines 6 weeks lead time with data ready for analysis (email 8 weeks out for appt) Depending on the complexity of the analysis proposed, longer lead timesmay be necessary.3. Grant application deadlines 8-12 weeks lead time (email 10-14 weeks out for appt) Statistical tests are tied to the research questions and design;earlier consultations will better inform grant development

Summary Power calculations are complex,but S-SPIRE statisticians can help Effective statistical collaborationcan be achieved Contact us early power/sample calculations areiterative & take time Gather information prior to consult1. Study design2. Expected effect size3. Feasible sample size4. Similar literature5. Pilot data Come meet us at 1070 Arastradero!

Thank [email protected](650) 725-7003

References1.Farrokhyar F, Reddy D, Poolman RW, Bhandari M. Practical Tips for Surgical Research: Whyperform a priori sample size calculation?. Canadian Journal of Surgery. 2013 Jun;56(3):207.2.Revicki D, Hays RD, Cella D, Sloan J. Recommended methods for determining responsiveness andminimally important differences for patient-reported outcomes. Journal of clinical epidemiology.2008 Feb 1;61(2):102-9.3.Hulley SB, Cummings SR, Browner WS, Grady DG, Newman TB. (2007). Designing ClinicalResearch. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins.4.Gordis, L. (2014). Epidemiology. Philadelphia: Elsevier/Saunders.5.Altman DG, Goodman SN, Schroter S. How statistical expertise is used in medical research. JAMA.2002 Jun 5;287(21):2817-20.6.Pye V, Taylor N, Clay-Williams R, Braithwaite J. When is enough, enough? Understanding andsolving your sample size problems in health services research. BMC research notes. 2016Dec;9(1):90.