BIOSTATISTICS FOR MEDICAL STUDENTS : TYPES OF DATA

Visual Analogue Scale & Statistical Concerns

🔸A frequently used tool in anaesthesia research is the 100 mm visual analogue scale (VAS). 

🔸This is most commonly used to measure postoperative pain, but can also be used to measure a diverse range of (mostly) subjective experiences such as preoperative anxiety, postoperative nausea, and patient satisfaction after ICU discharge. 

🔸Because there are infinite possible values that can occur throughout the range 0-100 mm, describing a continuum of pain intensity, most researchers treat the resulting data as continuous. 

🔸If there is some doubt about the sample distribution, then the data should be considered ordinal.

🔸When small numbers of observations are being analysed (say, less than 30 observations), it is preferable to consider VAS data as ordinal. 

🔸For a number of practical reasons, a VAS is sometimes converted to a 'verbal rating scale', whereby the subject is asked to rate an endpoint on a scale of 0-10 (or 0-5), most commonly recorded as whole numbers. In this situation it is preferable to treat the observations as ordinal data.

🔸There has been some controversy in the literature regarding which statistical tests should be used when analysing VAS data. 

🔸Mantha et al surveyed the anaesthetic literature and found that approximately 50% used parametric tests. 

🔸Dexter and Chestnuts used a multiple resampling (of VAS data) method to demonstrate that parametric tests had the greater power to detect differences among groups. 

🔸Myles et al have recently shown that the VAS has properties consistent with a linear scale, and thus VAS scores can be treated as ratio data. This supports the notion that a change in the VAS score represents a relative change in the magnitude of pain sensation. This enhances its clinical application.

Reference: Statistical Methods for Anaesthesia and Intensive Care, Paul S Myles and Tony Gin

Don't run away; it's simple : PARALLEL Vs CROSS OVER DESIGNS IN RANDOMIZED CONTROLLED TRIALS

 CONTROLLED TRIALS 

⏸Parallel groups 

➖➖➖➖➖➖➖

(👳🏻🔷vs 👳▪️ ) Simple, head-to-head comparison of two or more treatments 

1️⃣Subjects are allocated at random to a single treatment or a single treatment programme for the duration of the trial 

🆓The groups are independent  of each other

 🔀Crossover trials 

➖➖➖➖➖➖➖

(💂🔷  ➡ ️vs  💂▪️)This involves a single group study, where each patient receives two more treatments in turn; i.e. Each patient acts as their own control and comparisons of treatments are made within patients 

2️⃣Two or more treatments are given to each patient in random order 

✅useful for chronic conditions  such as pain relief in long-term illness or the control of high blood pressure where the outcome can be assessed relatively quickly

✴️It may not be feasible for treatments for short-term illnesses that once treated are cured, for example antibiotics for infections

😃Advantages of parallel group designs

➖➖➖➖➖➖➖➖➖➖➖➖➖➖

✔️The comparison of the treatments takes place concurrently 

✔️Can be used for any condition, especially an acute condition which is cured or self-limiting such as an infection 

✔️No problem of carry-over effects 

😬Disadvantages of parallel group designs 

➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖

✖️The comparison is between patients and so usually needs a bigger sample size than the equivalent cross-over trial 

😃Advantages of crossover designs 

➖➖➖➖➖➖➖➖➖➖➖➖➖

✔️Treatments are compared within patients and so differences between patients are accounted for explicitly 

✔️Usually need fewer subjects than the equivalent parallel group trials 

✔️Can be used to test treatments for chronic conditions 

😬Disadvantages of crossover designs 

➖➖➖➖➖➖➖➖➖➖➖➖➖➖

✖️Cannot be used for many acute illnesses 

✖️Carry-over effects need to be controlled 

✖️Likely to take longer than the equivalent parallel designs 

✖️Statistical analysis is more complicated if subjects do not complete all periods

ѦԀԀIṬIȎṄѦʟ IṄҒȎ➕ 

〰〰〰〰〰〰〰

🏃🏾In cross over trials, it is important to avoid the 'carry-over effect' of one treatment into the period in which the next treatment is allocated. 

↔️This is usually achieved by having a gap or 'washout period'  between treatments

#

S͙U͙P͙E͙R͙I͙O͙R͙I͙T͙Y͙ A͙N͙D͙ E͙Q͙U͙I͙V͙A͙L͙E͙N͙C͙E͙ T͙R͙I͙A͙L͙S͙ ⬆️↔️

⬆️S͙U͙P͙E͙R͙I͙O͙R͙I͙T͙Y͙ T͙R͙I͙A͙L͙S͙

✔️Seek to establish that one treatment is better than another 

✔️The sample size is set so that there is high statistical power to detect a clinically meaningful difference between the two treatments 

↔️E͙Q͙U͙I͙V͙A͙L͙E͙N͙C͙E͙ T͙R͙I͙A͙L͙S͙

✔️Seek to test if a new treatment is similar  in effectiveness to an existing one

✔️Appropriate if the new treatment has certain benefits such as fewer side effects, being easier to use, or being cheaper 

✔️Designed to be able to demonstrate that, within given acceptable limits, the two treatments are equally effective 

✔️Equivalence is a pre-set maximum difference between treatments such that, if the observed difference is less than this, the two treatments are regarded as equivalent . The tighter the limits of equivalence are set, the larger the sample size that will be required 

✔️A serious condition requires tighter limits for equivalence than a less serious condition. 

✔️The calculated sample size tends to be bigger for equivalence trials than superiority trials 

🔴T͙H͙I͙N͙G͙S͙ T͙O͙ R͙E͙M͙E͙M͙B͙E͙R͙

👉🏿In general the design and implementation of equivalence trials is less straight forward than superiority trials 

👉🏿If patients are lost to follow-up or fail to comply with the trial protocol, then any differences between the treatments is likely to be reduced and so equivalence may be incorrectly inferred.

👉🏿So equivalence trials need very strict management and good patient follow-up to minimize these problems 

👉🏿It is often helpful to include a secondary analysis where subjects are analysed according to the treatment they actually received, ‘per protocol’ analysis 

#MedicalResearch ,#ClinicalResearch , #MedicalStatistics , #BioStatistics , #AnaesthesiaResearch , #Statistics ,#research

Reference: Oxford Handbook of Medical Statistics, Janet L. Peacock , Philip J. Peacock

EXAMPLES OF RESEARCH QUESTIONS🔢

(From : Oxford Handbook of Medical Statistics, Janet L. Peacock, Philip J. Peacock, P:5)

❓What is the prevalence of diabetes mellitus in the population? 

🔴This is a simple descriptive study 

❓How effective is influenza vaccination in the community-based elderly? 

🔴This is a comparative study, comparing individuals who had vaccines with those who did not 

❓Does lowering blood pressure reduce the risk of coronary heart disease? 

🔴This is an evaluative study, investigating the effi cacy of lowering blood pressure 

❓Is prognosis following stroke dependent on age at the time of the event? 

🔴This is an observational study 

❓Why does smoking increase the risk of heart disease? 

🔴This is an explanatory study investigating the mechanism behind an observed relationship 

❓What evidence is there for the effectiveness of antidepressants in treating depression? 

🔴This study is a meta-analysis of existing interventional studies