Writing a research protocol, writing a grant application, obtaining funding

1
Writing a research protocol, writing a grant
application, obtaining funding

• Tony Kendrick
• Professor of Primary Medical Care
• Member, NIHR Research Training Fellowships Panel

2
Learning outcomes

• By the end of the session you will be expected to
  be able to
• Develop a research question
• Recall commonly used research designs
• Match the question to a research design
• Structure a research proposal
• Identify the elements of a successful application
  for a doctoral fellowship

3
Qualities of a good question

• Important
• - seriousness of topic
• - frequency with which it happens
• Of general interest and applicability
• Answerable
• Useful in terms of changing practice and/or
  improving care

4
Poor research questions

• Uninformed
• Undoable
• Unfocused
• Unclear
• Unsupported

5
Exercise

• Good or bad questions?
• Why are they good?
• Why are they bad?
• Could the bad questions be turned into good ones?

6
Good or bad research questions?

• Why do some people with asthma still smoke?
• Why do people want to see the doctor when they
  have no obvious physical illness?
• Do older people prefer older doctors?
• Is asthma more common in only children?
• Are home visits worthwhile?
• Does an injection of vitamin D given in the
  autumn reduce the incidence of hip fracture in
  the elderly?

7
Framing a research question
Broad question
Dissect the question
What is the key question?
8
Refining the question

• Discuss with colleagues, collaborators,
  (potential) supervisors
• Review the criteria of importance
• Review the literature
• what we know/dont know?
• Be critical and realistic
• So what?
• Is it possible?

9
Choosing a study design

• Have you any experience of study designs?
• What designs do you know about?
• How familiar are you with these designs?

10
Quantitative studies

• Test a research hypothesis
• For example
• Experiments
• Trials
• Surveys using questionnaires
• Structured interview studies

11
Qualitative studies

• Are used to explain and give an understanding of
  a persons beliefs and experiences
• To understand human behaviour from the subjects
  own frame of reference
• To explore in-depth beliefs, attitudes and
  motivations of participants

12
Systematic review

• An overview of primary studies that used explicit
  and reproducible methods
• Only high quality trials are included
• Limits bias
• Reduces chance effects
• Provides more reliable results upon which to draw
  conclusions and make decisions
• Can include a meta-analysis
• a mathematical synthesis of the results of two or
  more primary studies that addressed the same
  hypotheses in the same way.

13
Randomised controlled trial

• A study that randomly assigns patients to a
  treatment (intervention) or control group and
  follows the groups up for the outcomes of
  interest
• Used to determine the efficacy or effectiveness
  of
• Different treatments or approaches to disease
  management
• Service developments including education of
  health professionals in disease recognition or
  management

14
Cohort study

• A study that follows up people with and without
  exposure and sees if they develop the outcome of
  interest
• Ideal for finding out what happens over time
  (prospectively or retrospectively)

15
Case-control study

• A study that identifies patients who have the
  outcome of interest (cases) and control patients
  without the same outcome, and examines if they
  have been exposed to suspected causative factors
• Ideal for finding out why patients are different
  from controls

16
Survey

• A study that observes a defined population at a
  single point in time or time interval
• Ideal for describing what is going on now

17
Case report

• A study of one case describing the presenting
  signs and symptoms of a disease, its progress, or
  its response to treatment
• Ideal to monitor (new) clinical interventions

18
Exercise

• Match the design to the question

19
Writing research proposals and protocols

• Whats the difference between a proposal and a
  protocol?
• Why do we have to write
• a proposal?
• a protocol?

20
Differences between a proposal and a protocol

• Initial proposal to secure
• Funding
• Ethics
• Research Governance approval
• Later, more detailed protocol
• So all the researchers know what to do
• Researchers must always stick to the protocol!

21
Research ethics and governance

• Must have ethical approval if it involves NHS
  patients, staff, or premises/equipment
• Must comply with the DH Research Governance
  Framework
• Needs approval from the Trust RD department
• Register (if applicable) with the National
  Research Register, the Clinical Trials Register,
  Cochrane Collaboration etc.

22
Proposal structure

• Background
• Aims/objectives/hypothesis
• Methodology and Method
• Research governance and ethics
• Milestones/timetable
• Strengths of the team
• Dissemination
• Costing of the research
• Must be understandable to a range of
  professionals and lay people

23
Background

• The rationale for the study
• Importance - so what, why bother?
• How the research idea was formed
• Current knowledge (critical appraisal of
  literature)
• Gaps in knowledge that require further research
• Aims of proposal

24
Why is it important?

• Provides or improves the knowledge base
• Important to patient care or satisfaction
• Important to the NHS as an organisation or to
  local models of care delivery
• Important to development of health
  professionals/the workforce
• Value in economic terms

25
Aims and objectives

• Logically following on from the background
• The research question or aim
• Then the objectives (primary and secondary
  outcomes)
• Concise but understandable
• Use bullet points (quantitative)
• Describe (qualitative)

26
Sample and recruitment

• Sampling frame who and why
• Inclusion/exclusion criteria
• Sample size/power calculation
• Selection
• A consecutive sample, a random sample etc
• Qualitative
• Purposive, theoretical, snowball, opportunistic
  etc
• Recruitment
• How will people be recruited
• Gatekeepers
• Data protection issues and confidentiality
• Consent

27
Randomisation and blindness

• Method used to implement the random allocation
  sequence?
• e.g. numbered containers, central telephone
  service
• Ideally the sequence will be concealed until the
  interventions have been assigned
• who will generate the allocation sequence?
• who will enrol the participants?
• who assigns the participants to the group?
• Are the participants blind to the group
  allocation?
• Are those administering the interventions blind
  to the group allocation?
• Are those assessing the outcomes blind to the
  group allocation?

28
Data collection

• Location of data collection
• Clinic
• Patients home
• What data will be collected
• Observations (standardised)
• Questionnaires (standardised)
• Interviews type, how long, where
• Baseline and outcome measures

29
Data management and analysis

• How are the data to be managed?
• Input into analysis computer package such as SPSS
  or Stata
• Qualitative data transcription of the
  interviews
• Transcription rules
• Use of computer software NuDist, NVivo, Atlas
  etc
• The analysis
• Statistical which tests
• Qualitative - how
• Who will perform each task?
• Confidentiality
• Code numbers
• Identifying data locked away

30
Dissemination of findings

• To the funders (reports)
• To local health social care services,
  professionals and managers (e.g. seminars)
• At national/international conferences
• Submissions to peer reviewed journals
• To consumers (e.g. lay summaries, presentations)

31
The team

• Who will be involved
• Steering group
• Outline why the team and institution are suited
  to carry out the research
• Multi-disciplinary
• Research and clinical expertise
• Track record of research in the field,
  supervision and dissemination
• External support the team can draw upon

32
Costing the project

• Consider
• The amount of data to be collected
• The rate at which you will be able to recruit
  (and get data)
• The number of researchers needed to collect the
  data (bear in mind blinding)
• Secretarial support
• Salary on-costs and overheads
• Consumables stationery, telephone, postage,
  copying, printing, (incl. inter-library loans)
• Equipment needed, travel expenses
• Costs of dissemination and meetings

33
Milestones/timetable

• Anticipated length of the study
• What will be done when, e.g.
• Research governance requirements and application
  to ethics
• Pilot study
• Recruitment
• Data collection
• Data analysis (interim analysis if applicable)
• Preparation of reports and publications
• Initial dissemination

34
References

• Crombie IK, Davies HTO. Research in health care.
  Design, conduct and interpretation of health
  services research. John Wiley Sons. 1996
• McPherson K, Lord S. Clinicians guide to
  research. Part 2 Matching the method to the
  question. NZ Journal of Physiotherapy. 200028
  (2)20-28.
• McPherson K, Lord S. Clinicians guide to
  research. Part 3 Undertaking research. NZ
  Journal of Physiotherapy. 200028 (3)29-34.
• Moher D, Schulz KF, Altman DG (for the CONSORT
  group). The CONSORT statement revised
  recommendations for improving the quality of
  reports of parallel-group randomised trials.
  Annals of Internal Medicine. 2001134
  (8)657-662.

35
National Institute for Health Research Research
Training Fellowships Scheme

• Open to any individuals working in any scientific
  discipline within an institution in England
  (devolved countries separate)
• Who can demonstrate a contribution to improving
  health or health services
• Clinically active researchers
• Non-clinical researchers in health

36
National Institute for Health Research Research
Training Fellowships Scheme

• Website www.nccrcd.nhs.uk/nihrfellow/
• Read the guidance notes!
• Must not be registered already for a PhD
• 3 years full time or 4 years at 75 fte
• November application for January closing date,
  shortlisting in February/March, interviews in
  June, start October

37
What the award includes

• Salary (100)
• Tuition fees up to Research Council maximum
  (3,300 for 2008/9)
• Up to 10 000 research expenses
• Indirect costs (consumables/Uni support) (80)
• Estate costs (80)
• Up to 1000 towards specialist support (Stats
  etc)
• Up to 1,000 for research facilities (IT, labs)

38
The person

• The application form
• Previous research experience (UG/PG)
• Research training (Masters level)
• Conference presentations
• Publications
• Grants
• At interview
• Knowledge of the topic area and proposed
  methodology
• Ability to communicate research ideas

39
The project

• Topic
• Important area in health terms
• Common and costly problems
• NICE Guidance, National Service Frameworks
• Clear gap in literature
• Previous systematic review?
• Methodology
• Appropriate design
• Theoretical framework
• Evidence of pilot/preparatory work
• Consumer involvement
• Need for further funding
• Contingency plans in place for likely problems

40
The environment

• Track record in relevant research RAE rating
• Quality of supervision
• Trained supervisors with regular timetabled
  contact
• Own project rather than supervisors
• Training plan
• Some relevant to project (timed to fit with
  schedule)
• Some to broaden you as a researcher generally
• Highest quality courses you can identify, e.g.
• LSHTM for Epidemiology/Statistics
• ESRC Research Methods Centre in Southampton
• University of Surrey for qualitative research
• Placements in centres of excellence (in UK or
  elsewhere) if justified

41
Criteria for judgement (NIHR-RTF scheme)

• Candidates potential to make a long-term
  contribution to (NHS) RD in their field
• Appropriateness of host institution
• Quality of training and development plan
• Quality of proposed project
• Relevance to patient care (within 5 years)
• Likelihood of getting external funding if needed
• Quality and appropriateness of proposed mentoring
  and institutional support

42
Chances of success

• Around 100 applications per year
• Around 1 in 3 short listed
• Around 16 awards
• HOWEVER medics have better than average success
  rates