Статистическая викторина Британского медицинского журнала

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Sampling distributions


Researchers obtained birth weights for 4300 men born between 1973 and 1975. The mean birth weight was 3471 grams and the standard deviation was 540 grams. The 95% confidence interval for the mean was 3455 grams to 3487 grams.

Which of the following statements best describes the information given by the 95% confidence interval?


a) 95% of men in the sample had a birth weight between 3455 and 3487 grams

b) There was a probability of 0.95 that the sample mean birth weight would fall between 3455 to 3487 grams

c) There is a probability of 0.95 that the 95% confidence interval will contain the population mean birth weight for men born between 1973 and 1975

d) 95% of men in the population born between 1973 and 1975 had a birth weight between 3455 and 3487 grams

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Study design


A clinical trial investigated the effectiveness of topical versus oral ibuprofen for chronic knee pain in people over 50. In designing a trial to decide the best treatment approach, which answer best describes how confounding could be minimised?


a) Double blind trial

b) Intention to treat analyses

c) Matching by age and sex

d) Randomisation of research participants

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Number needed to treat


Two weeks after entry to a trial of antibiotics to treat symptoms of chest infection, 25% of people receiving placebo and 15% of people receiving antibiotics still had symptoms. What is the number needed to treat in this trial for antibiotics to cure symptoms at two weeks?


a) 10

b) 1.18

c) 10%

d) 4

e) 6.7

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Correlation


Which of the following, if any, does not measure correlation?


a) Kendall’s

b) Spearman’s

c) Cronbach’s

d) Pearson’s r

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Choosing a statistic


Which statistical test is best for each outcome measure?


Outcome measure

a) Comparing improvement in pain scores over 4 weeks using a visual analogue scale in a trial of 100 people with back pain.

b) Comparing the effect of two drugs on time to reduce fever below 38ºC in two groups of 100 febrile infants.

c) Comparing for 100 countries the number of doctors trained per million population with the infant mortality rate of that country.

d) Comparing death rates at 30 days after discharge from hospital of 100 patients having hip surgery and 100 patients having eye surgery.


Statistic

e) Chi square test (2)

f) Fisher’s test of exact probability

g) Cox regression

h) Correlation coefficient (r)

i) Student’s t test

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Study design


A randomised double blind trial compared the effects and tolerability of omeprazole and cisapride for the management of heartburn. Which of the following statements best describe the rationale for conducting this trial using the double blind technique?


a) Elimination of allocation bias

b) Minimisation of confounding between treatment groups

c) Minimisation of response bias

d) Minimisation of assessment bias

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Allocation bias


Which, if any, of the following methods is unlikely to introduce bias in the allocation of people to treatments in a trial?


a) People are matched according to age and sex, and for each person allocated to one treatment, a matching person is allocated to the other

b) The clinician treating the person tosses a coin to decide which treatment to give

c) People whose initial letter of their last name is in the first half of the alphabet receive one treatment; those in the second half of the alphabet receive the other

d) People with an even date of birth receive one treatment, those with an odd date of birth receive the other

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Standardised mortality ratios


The standardised mortality ratio for a certain town, using England and Wales as the standard, is reported as 70. Which of the following statements, if any, are true?


a) The low standardised mortality ratio could be because the average age of the town’s population is low

b) The death rate in each age band is 70% of that seen in England and Wales

c) The number of deaths in the town is 70% of what would be expected if the age specific death rates for England and Wales are applied to the town population

d) The standardised mortality ratio for the town is 30% lower than the average for England and Wales

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Likelihood ratios


In a study of general practice records in the UK, the symptom of dysphagia was found to have a positive likelihood ratio of 300 for oesophageal cancer, which itself had an incidence of 1 case in every 10 000 persons per year. Which of the following statements, if any, are true?


a) Among people with dysphagia, the ratio of those with cancer to those without is 300 to 1

b) Cancer is 300 times more likely in patients with dysphagia than in the general population

c) More than 95% of people with dysphagia seeing their general practitioner will have cancer

d) More than 95% of people with dysphagia seeing their general practitioner will not have cancer

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Type I and type II errors



Which of the options below best describes a type II error?


a) The probability that the null hypothesis is true

b) The probability that the null hypothesis is false

c) It is made if the null hypothesis is accepted when it is false

d) It is made if the null hypothesis is rejected when it is true

e) None of the above

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Correlation


A study finds a correlation between two variables (r=0.06), but this outcome is not statistically significant. Which, if any, of the following could be correctly used to describe the data?


a) High values of one variable are associated with low values of the other

b) For every unit change in one variable the other variable increases by 0.07

c) There is a strong linear association but the sample size is too small to demonstrate it

d) The significance should be quoted as 0.06 rather than "not significant"

e) There does not appear to be a linear association between the variables

[Наталья П.]

[Ссылки доступны только зарегистрированным пользователям ]

Risk


In a prospective cohort study, researchers reported that the risk of fatal coronary heart disease was increased for women with diagnosed diabetes compared with women without (relative risk adjusted for age 3.50, 95% confidence interval 2.70 to 4.53). Which of the following statements accurately describes the reported risk of fatal coronary heart disease?


a) For women with diagnosed diabetes, the risk of having fatal coronary heart disease is 3.5 times that of women without diagnosed diabetes

b) The reported risk of fatal coronary heart disease is statistically significant at the 5% level

c) In women, diabetes causes fatal coronary heart disease

d) 95% of women with diabetes have an increased risk of fatal coronary heart disease between 2.70 and 4.53 times that of women without diabetes

[Наталья П.]

Number needed to treat

In a trial comparing a drug with a placebo for the prevention of myocardial infarction, the incidence of myocardial infarction was 5% in the group receiving drug A (treatment group) and 10% in the group receiving placebo (control group); P=0.01.

How many patients need to be treated with drug A to prevent one additional case of myocardial infarction?


a) 5

b) 10

c) 20

d) 30


[Ссылки доступны только зарегистрированным пользователям ]