Applied Clinical Research

What kind of study would look at how many people in a population, how many male/female etc?

A descriptive study

What kind of study would look at associations, such as associating hair length with gender?

An analytical study

What kind of study tries to measure exposure or outcome on a collective level? Such as cases of asthma and measuring pollution in a given area

An ecological analytic study

What kind of studies measure all of their variables at the same time instead of measuring outcome after exposure?

Cross-sectional analytical studies

What studies select people based on the outcome of interest and ask them to look back to see if they experienced a particular exposure

Case-control studies

What studies recruit people without a condition of interest and study their habits over a period of time until some of them develop the condition?

Cohort studies

Do RCTs, case-control studies, cohort studies and time series analyses come under qualitative or quantitative methods of evaluating intervention methods?

Quantitative

What is a limitation of using a randomised control trial to help a certain group of people?

These trials often have very strict inclusion criteria, meaning that the people who the trial would benefit may be very different to those involved in the trial

What is the main benefit of randomisation?

Avoiding selection bias, whether it be conscious or unconscious

What is the ICH GCP?

A set of internationally recognised ethical and scientific quality requirements for designing, conducting, recording and reporting studies that involve participation of humans

What is the Hawthorne effect?

The idea that we change people by studying them

Give some limitations of trials

Efficacy is measured rather than effectiveness

Often too short or too small

Ignores patient preferences

Too much irrelevant data is often collected

What is the belief in a single identifiable reality? ie There is a single truth that can be measured and studied- the purpose of research is to predict and control nature

Positivism

What is the belief that multiple realities exist? ie Realities that are dependent on interactions between the individual and the social world

Contructionism

Give some examples of social science methodologies

Questionnaires, interviews and ethnography

What are some methodological limitations of questionnaires?

Social desirability bias

Some things are difficult to measure on a numerical scale

People’s feelings can change throughout the day

People who feel strongly positively or strongly negatively are more likely to fill out questionnaires

Different groups may have different interpretations of what is ‘good’ and ‘bad’ etc

What are the strengths of using qualitative data?

Understanding the perspective of those in a situation

Accessing information not revealed by quant variables

Explaining relationships between variables

Empowering marginalised groups

Considering researcher’s relationship to the data

What are some weaknesses of using qualitative research?

Finding consistent relationships between variables

Generalisability/representativeness

What are the advantages of interviews?

Depth- obtaining detailed data about individual experiences

Naturalistic- reflects everyday conversations

Dynamism- researchers can probe or change direction

Access- interviewers can access things that may be difficult to observe but a person can talk about

Equity- easier to access marginalised groups

Flexible- scheduled into people’s lives

Ethical- clear terms around what counts as data

What are the disadvantages of interviews?

Breadth- samples are rarely representative and often small

Artificial- not a transparent window, people may describe things differently to how they normally would

Replicability- difficult to replicate because results are dependent on contingent interactions

Anonymity- can be difficult to maintain, especially in small communities

Time consuming- for researchers and participants

Why would the elderly likely benefit more from interviews than surveys?

They may struggle with the technology often required for surveys

What is ethnography?

The recording and analysis of a culture or society, usually based on participant-observation and resulting in a written account of a people, place or institution

What kind of study is inductive and open-ended, rather than testing a number of hypotheses formulation in advance of the fieldwork?

An ethnographic analysis

What are the advantages of ethnography?

Depth- can obtain detailed data about individual experiences/perspectives

Naturalistic- very familiar social situation that reflects everyday conversations

Access to behaviour- individuals in interviews may provide biased accounts (even if they are unaware) but observing them directly limits this

Understanding of culture- can offer insights into symbols, rituals, rules etc

Exploring ambiguity- help to define a problem when it is not clear or is very complex

Long-term- able to capture processes including changes in outcomes and institutions over a longer period

Thick context- situates findings in deep historical and cultural context

What are the disadvantages of ethnography?

Breadth- samples are rarely representative and often small

Hard to plan for and get funded- difficult to anticipate findings

Replicability- difficult to replicate because the results are dependent on contingent interactions and interviewer subjectivity

Anonymity/ethics- can be difficult to maintain, especially in small communities. Difficult to set boundaries

Time-consuming- for researchers (transcribing) and participants

Why does ethnography make replicability even harder than interviews do?

The findings highly depend on the relationship the researcher is able to build with their participants over the long period of time

How does ethnography put both the researcher and participants at potential risk?

The researcher can end up harmed depending on what situation they find themselves in (eg if they wanted to study life as a gang member)

Potential emotional risk to participants- vulnerable groups may form an attachment to the researcher and feel abandoned when the study is over

Give some disadvantages of qualitative research

May miss wider trends that would only be observable in larger quantitative datasets

Small samples may limit generalisability (and therefore power of the study)

Participants may lack insight into causes so interviews might be limited

Observing patient interactions may change change the nature of the interaction

Social desirability bias

What is social desirability bias?

Participants may tell the interviewer what they think the interviewer wants to hear rather than being honest

Give four methods for ensuring qualitative research is rigorous

Triangulation

Member checking

Multi-coding/Inter-rater reliability

Reflexivity

What is triangulation?

Using more than one method of data collection to answer a research question (eg- satisfaction levels collected via questionnaire vary between areas- conducting interviews may give light to why this is)

What is member checking?

Researcher presents data transcripts or data interpretations to all or some participants for comment

PPI is an example- it is insider interpretation from the group that your participants fall into

In what circumstance would conducting member checking not be beneficial?

In a small community- chance of breaking anonymity

What is multi-coding/inter-rater reliability?

The use of more than one analyst- can improve the consistency or reliability of analyses

What is reflexivity?

A process that enables researchers to consider their position and influence during a study

Helps to communicate to the reader where blind spots or biases may lie

What way of collecting information can take many forms- qualitative or quantitative; postal, face-to-face, hand-out, telephone or online; at a single or multiple points in time?

Surveys

When is the best time to conduct a survey?

When your research questions are best answered by the people of interest themselves

In terms of surveys, what is a construct?

The idea you are trying to measure

In terms of surveys, what is an instrument?

The item(s) you use to measure the construct

In terms of surveys, what is a questionnaire?

A collection of instruments

Would depression and gender be examples of a construct, an instrument or a questionnaire?

A construct

Would PHQ-9 and ‘what is your gender?’ be examples of a construct, an instrument or a questionnaire?

An instrument

What is the difference between a questionnaire and a survey?

The questionnaire is the set of questions created for the survey, the survey is the whole process.

What are the six stages of conducting a survey?

1. Planning and survey design

2. Data collection

3. Data preparation and management

4. Data analysis

5. Reporting

6. Dissemination

Why do we need to be specific about how we define a population when conducting a survey?

We will generalise the findings of the survey within this population

What are the advantages of conducting online surveys?

Cheapest option

Fastest option

Social desirability bias is low

No chance of interviewer bias

What are the disadvantages of conducting an online survey?

Low to moderate response rate

Moderate respondent burden

High survey complexity

Requires moderate understanding of technology

What are the advantages of conducting a postal survey?

Cheap option

Low social desirability bias

No chance of interviewer bias

What are the disadvantages of conducting a postal survey?

Low response rate

High respondent burden

Poor survey complexity

What are the advantages of conducting a phone survey?

Fast option

Good survey complexity

Moderate response rate

What are the disadvantages of conducting a phone survey?

Quite expensive

Moderate respondent burden

Social desirability bias is moderate

Moderate chance of interviewer bias

What are the advantages of conducting a face-to-face survey>

High response rate

Low respondent burden

High survey complexity

What are the disadvantages of conducting a face-to-face survey?

Expensive

Slow option

Social desirability bias is high

High chance of interviewer bias

What should you include in an introduction to the respondent of your survey?

Why the survey is being conducted and how the respondent’s information will be used

What is the advantage of using existing questions from UK data service variable and question bank in your survey?

Existing questions are more reliable and also reduce the time spent coming up with appropriate new questions

What is the disadvantage of using existing questions from UK data service variable and question bank in your survey?

Existing questions are often sets of questions, so using them may increase the length of the questionnaire

What three things will piloting a survey help to identify?

Questions receiving poor response

Confusion over branching questions

Failure to follow instructions

What is the difference between reliability and validity?

Validity is how close to the correct answer you are, reliability is how often you come up with the same answer

What is testing concurrent validity?

Testing against an existing instrument- if the results are very similar, we can say our instrument has high concurrent validity

What is testing predictive validity?

Testing against a potential future measure

What kind of validity is implied when there is a lack of correlation with theoretically unrelated constructs?

Discriminant validity

What kind of validity is implied when there is correlation with similar constructs?

Convergent validity

What is the aim of using explanatory factor analysis (EFA)? How is this done?

Reducing the complication in the instrument(s). This is done by reducing the overall number of observed variables into latent factors based on commonalities within the data

What is the aim of using confirmatory factor analysis (CFA)? How is this done?

Being able to confirm one group to be asked about interview. This is done by using fit indices to investigate a previously known construct of latent factors which are generated by observed variables

What is content validity? Who do you need to consult when assessing content validity?

A formal examination of whether the content relates to the construct of interest and declines all necessary facets. You generally need to consult experts who will provide feedback on whether you have truly captured everything in your areas of interest

What is face validity?

The degree to which a test appears to be valid at ‘face video’

What is temporal validity?

Assessing whether a study designed in the past can still be used to predict behaviour today, which would mean it’s applicable across timing

What is evidence synthesis?

The process of combining data from multiple sources to provide an overall summary of existing knowledge

What other names is evidence synthesis also called by?

Literature review or narrative synthesis

Why do we need evidence synthesis in health and social care?

They provide the best evidence on the effectiveness and cost-effectiveness of treatments, tests and other interventions to support decision-making across health, public health and social care

What are the three main types of literature used in evidence synthesis?

Theoretical (describes expected and unexpected relationships about the way things happen)

Research (observation and measurement of the world)

Policy (tells practitioners how to act in a certain set of circumstances)

What are the limitations of having a narrative review?

Not comprehensive- doesn’t cover all of the literature

No information on how the authors selected the studies

There can be biases present in the selection and framing of the literature

No systematic quality check

Can’t quantitatively synthesise the data

What is the main cause of most of the limitations of conducting a narrative review?

The author selecting the papers to review themselves

How is a systematic review different from a literature/narrative review?

It involves systematically searching all of the literature available for a specific aim or question and includes a methodology section explaining how the author carried out the review. Literature reviews look at a much broader question than systematic reviews

What are the strengths of systematic reviews?

Transparent and reproducible

Involves strict and explicit eligibility criteria for inclusion and exclusion

Often includes systematically assessing the quality of studies

Highlights differences in findings across papers

Identifies gaps in the literature

What is a meta-analysis?

A statistical combination of results from two or more separate studies

What are the advantages of using a meta-analysis?

Improvement of precision in results

The ability to answer questions not proposed by individual studies

Has the opportunity to settle controversies arising from conflicting claims

What type of systematic review would answer these questions:

‘What is the prevalence of asthma internationally?’

‘Which intervention is most effective at reducing PTSD symptoms in young people?’

Meta-analysis

What is a qualitative evidence synthesis?

Combinations of multiple qualitative research studies

What is the purpose of qualitative evidence synthesis?

To examine patterns across studies

What is a scoping review?

Looking at the body of literature as a whole

What makes a scoping review different from other methods of evidence synthesis?

A scoping review doesn’t have a specific research question in mind

When is it useful to conduct a scoping review?

When examining evidence when it is still unclear what other, more specific questions can be addressed by a systematic review

When looking into the factors that cause a specific issue and there isn’t enough literature to conduct a systematic review for one specific factor

What is an umbrella review?

A review of current reviews

What condition is required to successfully conduct an umbrella review?

We know there is a lot of research (and meta-analyses) about our topic of interest

Give the methodological approach of systematic reviews (important to memorise)

1. Selecting a research question

2. Searching the literature

3. Screening for inclusion

4. Data extraction

5. Synthesis

6. Quality assessment

What is the FINER criteria for choosing a research question?

Feasible

Interesting

Novel

Ethical

Relevant

What is the purpose of pre-registering your protocol before conducting your review?

To prevent unethical practice, such as changing your mind to exclude specific data

What is involved in pre-registering a protocol?

Publishing the protocol online. This involves outlining the research question, methodology and analysis plans in advance.

What is the PICO framework for searching literature?

Population

Intervention

Comparator

Outcome

What is inclusion criteria?

Every included article must satisfy this criteria

What is exclusion criteria?

Any article containing this criteria must not be included

What is synthesis in terms of conducting a systematic review?

The process of analysing the findings from all the included studies

What are the three types of synthesis used when conducting systematic reviews?

Narrative synthesis

Quantitative synthesis/Meta-analysis

Qualitative synthesis

What is mentioned during a quality assessment?

Quality of reporting

Quality of the study design

Risk of bias

What type of study is the Cochrane Risk of Bias tool used to assess the quality of?

Randomised controlled trials

What type of studies (qualitative or quantitative) are quality assessed using the Modified Newcastle-Ottawa Quality Assessment Scales?

Quantitative

What type of studies (qualitative or quantitative) are quality assessed using the Critical Appraisal Skills Programme Qualitative Checklist?

Qualitative

What two pieces of information can be found using a meta-analysis?

1. The statistical combination of results from two or more studies, used to calculate the overall effect- what size is the effect?

2. Whether the variation across studies is small enough to be considered random noise, or large enough to suggest the studies are finding genuinely different effects- how confident are we in the effects?

Give some examples of the different types of data we ca meta-analyse

Proportion/prevalence estimates

Differences in groups with continuous data

Differences in groups with binary outcomes (event data)

Continuous association studies (correlations)