Collecting Data about people and Comparing the Health of Groups - Ovarian Cancer

Question 1

Why compare the health of groups?

• Research questions such as:
  
  • Is this disease increasing in …?
  • Does it occur with … frequency in my local community?
  • Is incidence associated with some suspected … …?
  • Has the outcome changed since … measures were instituted?
• Differences between groups at a … in time / Differences between groups … time

Answer 1

• Research questions such as:
  
  • Is this disease increasing in prevalence?
  • Does it occur with undue frequency in my local community?
  • Is incidence associated with some suspected risk factor?
  • Has the outcome changed since control measures were instituted?
• Differences between groups at a point in time / Differences between groups over time

Question 2

How do we compare the health of groups?

• Cross-sectional study -> one group surveyed to test associations between … and …/s
• Ecological study -> community/population observed to test associations between … and …/s
  
  • Both studies focus on simultaneous observation of … and …, but difference is unit of observation
  • Cross-sectional study – focus on … level data
  • Ecological study – focus on …-level data

Answer 2

• Cross-sectional study à one group surveyed to test associations between exposures and outcome/s
• Ecological study à community/population observed to test associations between exposures and outcome/s
  
  • Both studies focus on simultaneous observation of exposure and outcome, but difference is unit of observation
  • Cross-sectional study – focus on individual level data
  • Ecological study – focus on population-level data

Question 3

How do we compare the health of groups? (2)

• Cohort study -> …-… participants followed up to see if they develop a … (condition/outcome) of interest
  
  • Usually with groups who differ at outset on some …/s of interest
• Case–control study -> groups who differ at outset on … (condition/outcome) …
  
  • Look back at …/s of interest
• Randomised controlled trial (RCT) -> groups who are randomly allocated to receive …/s versus …/s
  
  • Test safety and efficacy/effectiveness of interventions

Answer 3

• Cohort study -> disease-free participants followed up to see if they develop a disease (condition/outcome) of interest
  
  • Usually with groups who differ at outset on some exposure/s of interest
• Case–control study -> groups who differ at outset on disease (condition/outcome) status
  
  • Look back at exposure/s of interest
• Randomised controlled trial (RCT) -> groups who are randomly allocated to receive intervention/s versus comparator/s
  
  • Test safety and efficacy/effectiveness of interventions

Question 4

Case-control studies

• Case–control study -> groups who … at … on disease (condition/outcome) …
  
  • Two groups of participants are selected – one … (cases) and one … (controls)
  • Controls selected to be as … as possible to the cases (e.g. age, gender, occupation, stage of illness, etc.)
    
    • Variables not of interest are matched (i.e. potential …) at selection
    • Are Exposures of interest measured or matched at selection?
• Always … -> Past exposure/s in both groups E.g. interview/survey, historical records

Answer 4

• Case–control study -> groups who differ at outset on disease (condition/outcome) status
  
  • Two groups of participants are selected – one with condition (cases) and one without (controls)
  • Controls selected to be as similar as possible to the cases (e.g. age, gender, occupation, stage of illness, etc.)
    
    • Variables not of interest are matched (i.e. potential confounders) at selection
    • Exposures of interest are not measured or matched at selection
• Always retrospective -> Past exposure/s in both groups E.g. interview/survey, historical records

Question 5

What are case-control studies?

Answer 5

• groups who differ at outset on disease (condition/outcome) status
  
  • Look back at exposure/s of interest

Question 6

What type of study?

Answer 6

Case Control Studies

Question 7

Case-control studies

• We cannot calculate … using case-control data
  
  • Because … = probability of … the outcome of interest
• In case-control studies:
  
  • we have … the outcome
  • i.e. we have selected participants into the case group or control group
  • we have decided the … of the groups
• Therefore, we cannot calculate … …
• Instead we calculate the odds of cases and controls in terms of their … …
  
  • We calculate an … ratio (OR) which is very similar to the … … (RR)

Answer 7

• We cannot calculate risk using case-control data
  
  • Because risk = probability of developing the outcome of interest
• In case-control studies:
  
  • we have determined the outcome
  • i.e. we have selected participants into the case group or control group
  • we have decided the size of the groups
• Therefore, we cannot calculate relative risk
• Instead we calculate the odds of cases and controls in terms of their past exposures
  
  • We calculate an odds ratio (OR) which is very similar to the relative risk (RR)

Question 8

Case-control studies - Example

• Case group - developed ovarian cancer
• Control group - … to case group - no ovarian cancer
• This example study shows that for African-American women, development of ovarian cancer was associated with:
  
  • … odds of 1 year prior to diagnosis, having a BMI of 25 or over
  • … odds of having completed post-high school education

Answer 8

• Case group - developed ovarian cancer
• Control group - similar to case group - no ovarian cancer
• This example study shows that for African-American women, development of ovarian cancer was associated with:
  
  • Greater odds of 1 year prior to diagnosis, having a BMI of 25 or over
  • Reduced odds of having completed post-high school education

Question 9

Strengths of the Case Control Studies

• Can offer some evidence of … – … relationship i.e. association between … and …
• Can identify multiple exposures (both … and … associations)
• Good when disease/outcome is …
• Minimises selection and information …
• R.. - cheaper and typically shorter in duration

Answer 9

• Can offer some evidence of cause – effect relationship i.e. association between exposure and outcome
• Can identify multiple exposures (both positive and negative associations)
• Good when disease/outcome is rare
• Minimises selection and information bias
• Retrospective - cheaper and typically shorter in duration

Question 10

Weaknesses of the Case Control Studies

• Cannot calculate … or …
• Less suitable for … exposures
• Can be hard to ensure exposure occurred … onset
• Retrospective data availability and quality may be …
• Suitable … group may be difficult to find
• Vulnerable to …

Answer 10

• Cannot calculate prevalence or incidence
• Less suitable for rare exposures
• Can be hard to ensure exposure occurred before onset
• Retrospective data availability and quality may be poor
• Suitable control group may be difficult to find
• Vulnerable to confounding

Question 11

The Randomised Controlled Trial

• a study in which participants are allocated randomly between an … (e.g. treatment) and a … … (e.g. no treatment or standard treatment)

Answer 11

• a study in which participants are allocated randomly between an intervention (e.g. treatment) and a control group (e.g. no treatment or standard treatment)

Question 12

Why are RCTs conducted?

• Safety:
  
  • ​Ascertain the safe … of a new drug.
  • Demonstrate safety and t… of a new …
  • Monitor … events profile of a new drug (against an existing drug or placebo)
• Efficacy/Effectiveness​
  
  • Demonstrate efficacy of new drug – does it …?
  • Show that treatment T is … or … to treatment X
  • Demonstrate effectiveness, and …-effectiveness, of A vs. B

Answer 12

• Safety:
  
  • ​Ascertain the safe dose of a new drug.
  • Demonstrate safety and tolerability of a new compound
  • Monitor adverse events profile of a new drug (against an existing drug or placebo)
• Efficacy/Effectiveness​
  
  • Demonstrate efficacy of new drug – does it work?
  • Show that treatment T is superior or equivalent to treatment X
  • Demonstrate effectiveness, and cost-effectiveness, of A vs. B

Question 13

RCTs as an experiment

• RCTs are also a special type of experiment in which randomisation is used
• Randomisation means that potential … variables should be … distributed between groups
• This creates two situations which are identical but:
  
  • One situation in which the supposed cause (intervention of interest) is …
  • One situation in which the supposed cause is …
• RCTs can reduce … and allow identification of exposures which are … related to disease of interest
  
  • i.e. identification of interventions which cause reduction in disease likelihood or severity

Answer 13

• RCTs are also a special type of experiment in which randomisation is used
• Randomisation means that potential confounding variables should be equally distributed between groups
• This creates two situations which are identical but:
  
  • One situation in which the supposed cause (intervention of interest) is present
  • One situation in which the supposed cause is absent
• RCTs can reduce confounding and allow identification of exposures which are causally related to disease of interest
  
  • i.e. identification of interventions which cause reduction in disease likelihood or severity

Question 14

Strengths of the RCT

• Establish the s… and e…/ e… of new interventions
• Minimise selection and information …
• Best single-study evidence for … association between exposure (intervention) and outcome

Answer 14

• Establish the safety and efficacy/ effectiveness of new interventions
• Minimise selection and information bias
• Best single-study evidence for causal association between exposure (intervention) and outcome

Question 15

Weaknesses of the RCT

• …-consuming, difficult and e…
• Not immune to …
• Issues with participant …-…
• Can lack g…

Answer 15

• Time-consuming, difficult and expensive
• Not immune to bias
• Issues with participant drop=out
• Can lack generalisability

Question 16

What data can we collect?

• Data properties to consider:
• Categorical (D…)
  
  • B.. variables- e.g. Are you a parent: yes/no; diagnosis of endometriosis (yes/no)
  • … categories but no order (unordered categorical) – e.g. Sexual orientation; Smoking (never, former, current);
  • Several ordered categories (ordinal) – e.g. Socio-economic status; categorised age; responses on Likert scale
• Continuous variables (S…)
  
  • E.g. age; no. of symptoms; score on a questionnaire such as quality of life or satisfaction

Answer 16

• Data properties to consider:
• Categorical (discrete)
  
  • Binary variables- e.g. Are you a parent: yes/no; diagnosis of endometriosis (yes/no)
  • Several categories but no order (unordered categorical) – e.g. Sexual orientation; Smoking (never, former, current);
  • Several ordered categories (ordinal) – e.g. Socio-economic status; categorised age; responses on Likert scale
• Continuous variables (scale)
  
  • E.g. age; no. of symptoms; score on a questionnaire such as quality of life or satisfaction

Question 17

Using data to compare health outcomes of groups

• Does one group report higher scores (on a … scale) than another?
  
  • E.g. is condition or intervention X associated with reduced symptoms compared to condition/intervention Y?
  • …-test or ANOVA
• Does one group have a higher proportion of an outcome than another (c…)?
  
  • E.g. is condition/intervention X associated with increased frequency of diagnosis P?
  • … … test

Answer 17

• Does one group report higher scores (on a continuous scale) than another?
  
  • E.g. is condition or intervention X associated with reduced symptoms compared to condition/intervention Y?
  • T-test or ANOVA
• Does one group have a higher proportion of an outcome than another (categories)?
  
  • E.g. is condition/intervention X associated with increased frequency of diagnosis P?
  • Chi squared test

Question 18

Error and Power

• Protection against Type I error = threshold for determining when effects are …
  
  • Significance level for p value is typically accepted at …% (or …) i.e. …% chance of type I error
• Protection against Type II error = power of the study to … when … effects are present
  
  • Statistical power is typically accepted at … – …% (or ..-..) i.e. …% or …% chance of type II error
  • The larger the sample, the … the statistical power

Answer 18

• Protection against Type I error = threshold for determining when effects are significant
  
  • Significance level for p value is typically accepted at 5% (or 0.05) i.e. 5% chance of type I error
• Protection against Type II error = power of the study to detect when significant effects are present
  
  • Statistical power is typically accepted at 80 – 90% (or 0.8 – 0.9) i.e. 20% or 10% chance of type II error
  • The larger the sample, the larger the statistical power

Question 19

Two aspects of assessing certainty of estimates - P values and Confidence Intervals

• P values = p…
  
  • When you compare groups using a statistical test (t test, χ2), the result has a p value.
  • The p value is the … that the difference observed (or one more extreme) could have occurred by … if the groups compared were really …
    
    • E.g. P 0.05 = 1/20, P 0.01 = 1/100
• P value = type … … protection
• Larger sample = … p value

Answer 19

• P values = Probability
  
  • When you compare groups using a statistical test (t test, χ2), the result has a p value.
  • The p value is the probability that the difference observed (or one more extreme) could have occurred by chance if the groups compared were really alike.
    
    • E.g. P 0.05 = 1/20, P 0.01 = 1/100
• P value = type I error protection
• Larger sample = smaller p value

Question 20

Two aspects of assessing certainty of estimates - P values and Confidence Intervals

• Confidence Intervals = P…
  
  • The confidence interval describes the range of values with a given … (e.g. 95%) that the true value of a variable is … within that …

Answer 20

• Confidence Intervals = Precision
• The confidence interval describes the range of values with a given probability (e.g. 95%) that the true value of a variable is contained within that range.

Question 21

T-tests – what’s being tested?

• The (…) hypothesis: There is no difference in quality of life between women who have epithelial ovarian cancer (EOC) versus an ovarian germ cell tumour (OGCT)
  
  • The mean quality of life in the EOC group is the same as the mean quality of life in the OGCT group
• The (…) hypothesis: There is a difference in mean quality of life between the groups (greater in EOC group)
• The between-group difference needs to be … if there is more within-group variability
  
  • e.g. if women vary greatly in quality of life within each group, we would need to see a bigger between-group difference to be sure it is a between-group difference
• Test output: …-statistic and accompanying …-value

Answer 21

• The (null) hypothesis: There is no difference in quality of life between women who have epithelial ovarian cancer (EOC) versus an ovarian germ cell tumour (OGCT)
  
  • The mean quality of life in the EOC group is the same as the mean quality of life in the OGCT group

The (alternative) hypothesis: There is a difference in mean quality of life between the groups (greater in EOC group)

• The between-group difference needs to be greater if there is more within-group variability
  
  • e.g. if women vary greatly in quality of life within each group, we would need to see a bigger between-group difference to be sure it is a between-group difference
• Test output: t-statistic and accompanying p-value

Question 22

T-test – Hypothesis: Quality of life is higher for women with epithelial ovarian cancer (EOC) versus an ovarian germ cell tumour (OGCT)

• Conclusion?
• (look at P value)

Answer 22

• Conclusion: There is not enough evidence to suggest quality of life is higher for women with epithelial ovarian cancer (EOC) versus an ovarian germ cell tumour (OGCT) (P=0.10) - can’t reject null / accept alternative

Question 23

Chi-square tests – what’s being tested?

• The (null) hypothesis: There is … … between endometriosis and ovarian cancer.
  
  • The proportion of women with endometriosis who have ovarian cancer is the … as the proportion of women who do not have endometriosis and have ovarian cancer.
• The (alternative) hypothesis: There is … … between endometriosis and ovarian cancer.
  
  • The proportion of women with endometriosis who have ovarian cancer is … … … as the proportion of women who do not have endometriosis and have ovarian cancer.
• Test output: …-… (χ2) statistic and accompanying ….-value

Answer 23

• The (null) hypothesis: There is no association between endometriosis and ovarian cancer.
  
  • The proportion of women with endometriosis who have ovarian cancer is the same as the proportion of women who do not have endometriosis and have ovarian cancer.
• The (alternative) hypothesis: There is an association between endometriosis and ovarian cancer.
  
  • The proportion of women with endometriosis who have ovarian cancer is not the same as the proportion of women who do not have endometriosis and have ovarian cancer.
• Test output: chi-squared (χ2) statistic and accompanying p-value

Question 24

Chi-Square- two groups with categorical outcome - Hypothesis - Endometriosis increases likelihood of having Ovarian cancer

• P value indicates what?
• Reject or accept null hypothesis? (There is no association between endometriosis and ovarian cancer.)

Answer 24

• Strong evidence to reject the null hypothesis of no association between endometriosis and ovarian cancer
  
  • P value is well below the threshold at which is appears that the data we have has not come about by chance

Question 25

P-Value and Statistical Significance

• A p-value less than … (typically ≤ …) is statistically significant.
• It indicates strong evidence against the … hypothesis, as there is less than a …% probability the null is correct (and the results are …)
• Therefore, we reject the … hypothesis, and accept the … hypothesis

Answer 25

• A p-value less than 0.05 (typically ≤ 0.05) is statistically significant.
• It indicates strong evidence against the null hypothesis, as there is less than a 5% probability the null is correct (and the results are random).
• Therefore, we reject the null hypothesis, and accept the alternative hypothesis

Question 26

Confidence intervals

• Confidence intervals (CIs) = range of values that should contain the … population parameter
• Use current sample to estimate shape of actual distribution of variable in population – and imagine how much observed sample values would vary if kept running the study
• Create an … and … bound a certain distance above and below the mean, e.g. 95% - these upper and lower bounds converted back to real scores
• 95% CIs would contain the true population value …% of the time and fail to contain the true value …% percent of the time
• CIs become … as the sample size increases – the larger the sample, the more likely it is that scores will cluster narrowly around the true population mean

Answer 26

• Confidence intervals (CIs) = range of values that should contain the true population parameter
• Use current sample to estimate shape of actual distribution of variable in population – and imagine how much observed sample values would vary if kept running the study
• Create an upper and lower bound a certain distance above and below the mean, e.g. 95% - these upper and lower bounds converted back to real scores
• 95% CIs would contain the true population value 95% of the time and fail to contain the true value 5% percent of the time
• CIs become narrower as the sample size increases – the larger the sample, the more likely it is that scores will cluster narrowly around the true population mean

Question 27

Confidence intervals example

• E.g. Odds of exposures for African-American women with ovarian cancer
  
  • OR < 1 = … association
  • OR > 1 … factor
  • OR = 1, … association
• Precision of example estimates fairly good, i.e. the 95% CIs are not very wide
• But, for most exposures, do contain 1 (i.e. true population could be OR =1, … association)
• So can’t be very sure or confident about making clinical recommendations/ predictions in light of these data

Answer 27

• E.g. Odds of exposures for African-American women with ovarian cancer
  
  • OR < 1 = protective association
  • OR > 1 risk factor
  • OR = 1, no association
• Precision of example estimates fairly good, i.e. the 95% CIs are not very wide
• But, for most exposures, do contain 1 (i.e. true population could be OR =1, no association)
• So can’t be very sure or confident about making clinical recommendations/ predictions in light of these data

Question 28

Summary - Collecting Data about People and Comparing Health of Groups

• To measure existing health needs of populations at one time = … and over time = …
• To compare the health needs/outcomes of different groups = measures of risk, differences in frequency and/or means
• Comparing groups can allow us to identify casual risk factors (…) relevant to the disease or condition of interest
  
  • … studies (e.g. case control) – describe population, identify potential causal exposure factors
  • … studies (e.g. RCT) – test safety and efficacy/effectiveness (and best evidence of cause-effect)
• Selecting appropriate statistical test = consider nature of data
  
  • Continuous data e.g. quality of life = …-test or ANOVA
  • Categorical data e.g. disease presence or absence = …-… test
• In collecting data, we need to consider certainty
  
  • How to design a study that is adequately powered
  • Provide estimates of probability (… value) and estimates of precision (… intervals)

Answer 28

• To measure existing health needs of populations at one time = prevalence and over time = incidence
• To compare the health needs/outcomes of different groups = measures of risk, differences in frequency and/or means
• Comparing groups can allow us to identify casual risk factors (exposures) relevant to the disease or condition of interest
  
  • Observational studies (e.g. case control) – describe population, identify potential causal exposure factors
  • Experimental studies (e.g. RCT) – test safety and efficacy/effectiveness (and best evidence of cause-effect)
• Selecting appropriate statistical test = consider nature of data
  
  • Continuous data e.g. quality of life = t-test or ANOVA
  • Categorical data e.g. disease presence or absence = chi-squared test
• In collecting data, we need to consider certainty
  
  • How to design a study that is adequately powered
  • Provide estimates of probability (p value) and estimates of precision (confidence intervals)