Digital PharmD

Informatics in Pharmacy

biostatistics

What Statistical Test To Use

By

I memorized this table and wrote it out on my noteboard during the test. It makes it easy to pick out which test without much mental effort

Type of Data2 Independent SamplesRelated or Paired Samples3 or More Independent Samples3 or More Related SamplesMeasures of Correlation
Nominal
Example: male/female, yes/no, dead/alive		Chi-Square Fisher’s Exact (for small samples)McNemar testChi-Square for K-independent t-samplesCochran Qcontingency coefficient
Ordinal
Example: Class I, II, II Heart failure		Mann Whitney U Willcoxen Rank SumSign test Willcoxen Signed RankKruskal Wallis one way ANOVAFriedman 2 way ANOVASpearman Kendall rank Kendell COE
Continuous (aka “parametric”)
Example: temperature, heart rate, blood pressure		Student T-test Mann Whitney U (only if parametric with outliers)Paired t-test1-way ANOVA2-way ANOVAPearson correlation

Study Type Strength or Weakness:

They may ask what kind of study would you use for X problem.

*Top is the highest quality, the bottom is the least

  • Cohort and cross-sectional studies are usually prospective, case-control and case-report are usually retrospective.
  • Cohort studies are best suited determining the association between exposures/factors and diseases/conditions (prospective. You follow a cohort and see if they get the disease).
  • Case-control studies are best to determine the association between a rare event and a potential cause (retrospective.  You see the rare event and work backward because the event is too rare to find in enough numbers in a random cohort).
  • Cross-sectional studies are data collected from a population, or subset, at one specific point in time.
  • Survival analysis – time to an event, not just death or MI, etc. Can be anything.  Kaplan Meir curves estimate the percentage of people saved per time

Is it Intention to Treat or Per Protocol?

  • Intention to treat: Analyze data from all randomized patients, regardless of the completion of the trial.  This increases external validity (is this applicable to the real world)
  • Per-protocol: This increase internal validity (is this a well-done study for whatever the hypothesis is). Only count those who followed the trial to completion.  This reflects the actual treatment, not real-world results.

Pharmacoeconomic Studies

By

The easy thing is to memorize this chart and pick out the keywords from the questions. These are easy to get correct.

Method:

Cost-minimizationDrugs are equalOnly cost is considered. For example, choosing which ACEI goes on the formulary 
Cost-effectivenessDrugs are not equal, $$$Looking at outcomes (decrease in BP, decrease in glucose). For example, deciding between classes like ARBs and ACEIs. 
Cost-benefitUsed to evaluate programs, not drugsOutcome expressed as benefit: cost 
Cost-utilityLooking at life yearsUsually expressed in QALYs (quality-adjusted life years) 

Basic Biostatistics

By

The BCPS is heavily weighted on Biostats, study design, and regulatory issues.  I recommend you buy the little biostatistics book ACCP offers (it’s cheap and it has some good practice problems in it).   It was worth more than any other book I purchased for the BCPS.  This is also a really good study guide and here’s a really, really simple sheet. This study guide has more topics in biostatistics (including the “which statistical test to pick” questions).

Biostats Definitions:

Here are some basic things in case you have forgotten them:

  • Nominal date – data with no order (yes/no, male/female)
  • Ordinal data – data with order, but no consistent difference in magnitude change (classes of heart failure, pain scales)
  • Interval data – continuous data with consistent interval difference (temperature)
  • Ratio data – continuous data with consistent interval difference, but zero is the starting point (HR, BP)
  • Mean– “Average.” Only with continuous data (parametric, normally distributed)
  • Median – 50th percentile. The data point exactly in the middle of the data points.  Usually only used with ordinal data or continuous data that is not normally distributed
  • Mode – the most frequently occurring value
  • Range – how far apart the data points are
  • Interquartile range – related to the median.  Most data is in the 25-75 percentile.
  • Standard deviation (SD) – only applicable to parametric data.  Measures how data points scatter around the mean.  Not available for nominal data or ordinal data.  99% of the data should be found in +/-3 SDs, 95% of the data in +/- 2 SDs, 68% in +/- 1 SD.
  • Standard error the mean (SEM) – smaller than the standard deviation. Average variability of data.
  • Parametric data – continuous data that is normally distributed (like a parabola)
  • Nonparametric data – continuous data that does not follow a normal distribution (skewed)
    • If mean > median, skewed to the right
  • Correlation coefficient (R) – how variables relate.  The closer the number is to 1, the stronger the relationship.
  • R2 – How much of the relationship is due to Y (ie: 70% of weight gain is due to calories, 30% is unknown).
  • Narange Scale – the probability that an adverse effect is related to drug toxicity.
  • Lot Proportional Hazard – survival data

Therapeutic Index

  • Therapeutic Index = Median Toxic Dose / Medicontinuousive Dose (also Lethal dose 50/effective dose 50)
  • The higher the TI, the safer the drug
  • IF LD50>ED50 TI is large so the drug is safe

Specificity, Sensitivity, Predictive Values and Accuracy

  • Sensitivity = True Positives / (True Positives + False Negatives) Sensitivity measures true positives.  If a highly sensitive test is negative, you can be sure they don’t have the disease (SNOUT Rules Things Out)
  • Specificity = True Negatives / (True Negatives + False Positives) Specificity measures true negative.  If a highly specific test is positive, you can be sure they have the disease (SPin Rules Things In)
  • Bigger Numbers are more significant.
  • Positive Predictive Value = True Positives / (True Positives + False Positives)  This is the percentage of people who test positive who actually have the disease. *Bigger numbers are more significant*
  • Negative Predictive Value = True Negatives / (True Negatives  + False Negatives )  This is the percentage of people who test negative who don’t have the disease. *Bigger numbers are more significant*
  • Accuracy = (True Positives + True Negatives) / Total
  • When you increase sensitivity, you decrease specificity.  You get more diagnosis but more false positives.
  • Prevalence = the number of cases / total at risk; incidence = new cases/total population at risk

Some people like to set up a table (this is called a “confusion matrix” in the statistics world):

 +– 
+TPFPTP+FP
–FNTNFN+TN
 TP+FNTN+FPTP+FP+FN+TN
  • Sensitivity= TP/(TP+FN) (column 1)
  • Specificity = TN/ (TN+FP) (column 2)
  • PPV = TP/ (TP+FP) (row 1)
  • NPV = TN/(TN+FN) (row 2)
  • Accuracy = diagonal down chart : TP+TN/ (TP+FP+FN+TN)

Hypothesis Testing:

The null hypothesis is that there is no difference between groups in a study. In order to find significance, you need to REJECT the null. This can be a little confusing.

 Null is TrueNull is False
Accept Nullcorrect decision Type II Error (β)
Reject NullType I error (alpha)correct decision
  • Type 1 error: Reject the null hypothesis when it is true.  A difference is found where none exists. The maximum acceptable alpha error is usually 0.05 (think of alpha as the p-value you are designing the study to obtain).
  • Type 2 Errors: Accept the null hypothesis when it is not true.  No difference found when one exists.  The maximum acceptable probability of a Type II error should be 20% (β = 0.2).
    • Beta errors are usually due to sample size or a poorly powered study. The easiest way to decrease Beta is to increase the sample size. Alpha and sample size have the greatest impact on study power.
  • You’ll always have the risk of making either a Type 1 or Type 2 error, but never have the risk of making both. If the p-value is significant, you have the risk of making a Type 1 error. If it is not, you have the risk of making a Type 2. For example, a p-value of 0.01 would mean there is a chance of committing a Type I error (i.e.: you found the p was significant, rejected the null and stated there was a different between the groups. In real life, there was no difference between the two groups).
  • “If the p-value is low, the null must go.” If the p-value is less than alpha, the null is rejected.
  • P VALUES DO NOT SUGGEST CLINICAL SIGNIFICANCE, just statistical significance. Clinical significance can only be assessed by reading the study and finding the methods, inclusion criteria, etc.
  • Confidence intervals:
    • The closer a data point lies to the 95% confidence interval, the more likely it represents the population.
    • For a ratio confidence interval, if it includes 1, it’s not significant (think 1/1 = 1, no difference)
    • For a continuous confidence interval, if it includes 0, it’s not significant (think 1-1 = 0, no difference)

Relative Risk

  • Relative Risk = incidence in exposed patients / incidence in non-exposed patients
    • >1 incidence in the exposed group is higher
    • <1 incidence in the exposed group is lower
  • Absolute risk reduction = Risk Reduction in the control group – Risk Reduction in the treatment group
  • For this one, use a table for sure (these will likely be on the test):
TreatmentDiseaseNo Disease
ExposedAB
UnexposedCD
  • Relative Risk = A/(A+B)
  • Absolute Risk Reduction (ARR) = C/(C+D)
  • Odd Ratio = AD/CB
    • Only used in retrospective studies
    • If used in a prospective study, odds ratio overestimates the risks. They may try to trip you up on this.
  • Number needed to treat or Number Needed to Harm: 1/ARR (it’s a decimal, not a percentage)
    • Include duration of study = must treat 10 patients for 5 years
  1. AR (absolute risk) = the number of events (good or bad) in treated or control groups, divided by the number of people in that group.
  2. ARC = the AR of events in the control group.
  3. ART = the AR of events in the treatment group.
  4. ARR (absolute risk reduction) = ARC – ART.
  5. RR (relative risk) = ART / ARC.

Unsafe Abbreviations

By

Read through the ISMP’s Unsafe Abbreviation Guide.  It’s more comprehensive than the Joint Commission’s and they overlap.  Here are a few things you should know.

  • You may use range orders according to your hospital protocol.
  • PRN range orders must include a symptom. Pain PRN meds must include some type of scale.  You shouldn’t have multiple meds for the same scale or indication without a reason (ie: one is IV and one is PO)
  • If the patient can take PO and there are two meds for the same indication, PO is preferred.
  • Don’t use:
    • U – write unit
    • IU – write international unit
    • QD, QOD – write daily, every other day, etc.
    • 2.0 – just write 2
    • .3 – write 0.3
    • MS, MS04, MgSO4 – write morphine, magnesium
    • ug – write microgram
    • HS – write bedtime or half-strength
    • S.C or SQ – write sub-q or subcutaneously
    • D/C – write discharge
    • cc – write ml
    • AS, AD, AU, OS, OD, OU – write right ear, left eye, etc
    • > and < – write out less than or greater than
    • / – write per

References to Know

By

There may be questions on the test about “where would you go to find [info]” or “Which online database would this study be found in.”  The are easy to guess if you know at least what most databases contain and how often they are updated.

  • Medline: Most of us know how Medline works, but:
    • Know how to vaguely use MeSH terms.  MeSH (medical subject headings) is the National Library of Medicine’s controlled vocabulary thesaurus. Each bibliographic reference is associated with a set of MeSH terms assigned to describe the content of an article. There are more than 19,000 main headings in MeSH, as well as thousands of cross-references that assist in finding the most appropriate MeSH headings. MeSH terms are arranged in a hierarchy, or “tree structure,” that permits searching at various levels of detail, from the most general to more narrow levels, to find the most precise terms. Subject specialists at the National Library of Medicine update MeSH annually.
    • Pubmed is an interface used to search Medline, as well as additional biomedical content.  Ovid is an interface for searching only Medline content. Pubmed is more user-friendly and allows you to search through more content than Ovid. However,  Ovid Medline allows you to perform a more focused search. You will get slightly different results by searching in each database.
    • High-profile journals like JAMA or the New England Journal of Medicine are indexed within days, but other journals take weeks to months.
  • International Pharmaceutical Abstracts: pharmaceutical abstracts from 750 journals including foreign and state pharmacy journals, in addition to key US medical and pharmacy journals.  Many are not included in Medline, but subject descriptions are inconsistent.
  • Iowa Drug Information Service Database: Full-text article from 1966 to present in about 200 medical and pharmacy journals (mostly US).  Updated monthly.
  • Clin-Alert: more than 100 medical and pharmacy journals. Mosty focused on adverse events, drug interactions and medical/legal issues.  Good for recent reports of adverse events.
  • Excerpta Medical: more than 7000 journals from 74 countries from 1974 to the present.  Articles appear within 10 days of publication, often before Medline.
  • Lact-Med is a free online resource for lactation guidelines.
  • Drugs in Pregnancy and Lactation is a subscription-based service.
  • Orange Book – therapeutic equivalence
  • Purple Book – biosimilars
  • Red Book – a collection of package inserts & pricing info (historical and current).
  • US Pharmacopeia – Drug info and storage info
  • Facts and Comparisons – Drug info grouped by therapeutic category
  • Trissel’s: IV Compatability, storage and handling