Proposal Writing

Health Research Proposal Writing

Tips for Writing Quantitative Research Proposals

Centre for Advancement of Health

Knowledge Into Action (K2A)

Calgary Health Research

V. Nadine Gall Dr. Peter Faris (Adapted from MJ Verhoef and RJ Hilsden 2001, 2004)

CALGARY HEALTH RESEARCH A partnership of the Calgary Health Region and the Faculty of Medicine

A note about this document: The examples provided throughout this document have been summarized from successful research proposals (references below). The examples are intended to give you a sense of how to write the various sections and suggestions re the format. It is important to note that these examples are merely summaries of the various sections, so may not represent the full detail that review committees will be looking for. References: Verhoef MJ, Boon HS, Mutasingwa DH. The scope of practice of naturopathic

medicine in Canada: An emerging profession. Soc Sci Med 2006; 63; 409-417 Robert M, Ross S, Brasher P, Fischer B, Jacobs P. Surgical Management of Stress

Incontinence in Women: Randomized Trial of Trans-Obturator Tape versus Tension-free Vaginal Tape http://www.obgyn.ucalgary.ca/Documents/TOTv%5B1%5D.TVT.pdf

Acknowledgements: Special thanks to Dr. Sue Ross, Dr. Marja Verhoef, Dr. Magali Robert, Christy Woolcott, Tanis Fenton, Dr. Cathie Scott, and Judy Seidel for their contributions in developing this document. We welcome your feedback about this document. Please let us know how you are using it: Contact:

The Centre for Advancement of Health

602 South Tower, 1403 29th Street NW, Calgary AB

(403) 944-1093

E-Mail: [email protected]

Download Proposal Writing Guide at: https://my.calgaryhealthregion.ca/http://www.calgaryhealthregion.ca/research/publications.htm

Proposal Writing 2006 NG Centre for Advancement of Health

2

TABLE OF CONTENTS Proposal Writing Overview.. 4 Elements of a Research Proposal Title.. 6

Study Problem / Purpose. 6

Relevance / Study Rationale.. 7

Literature Review.. 8

Research Objectives / Research Question.. 9

Study Methods

Study Design . 10

Subjects . 12

Allocation of Study Groups and Intervention .. 14

Variables / Data Collection. 15

Statistical Analysis and Sample Size 18

Ethical Considerations. 22 Limitations 24 Budget .. 25 Plans for Dissemination . 25 Other Possible Sections . 25 General Tips... 25 Pilot Studies .. 26 Supplementary Reading .. 27


3


4

HEALTH RESEARCH PROPOSAL WRITING Research Proposals are intended to demonstrate:

That your project should be done Specifically what you intend to accomplish and how you will do it A workplan that demonstrates you have thought through all of the

elements of your project

Potential Audience: Funders, committees overseeing institutional approval (science, ethics),

thesis committees etc. Keys to Success in Writing a Good Proposal

Quality of the Presentation

Quality of the Proposal

Overall Quality of the Study

Good research question (relevant, justified) Appropriate research design Rigorous, feasible, and ethical methods Qualified research team

Informative title Clear research questions Thorough and relevant background Convincing rationale/significance Clearly defined (and measurable) research objectives Appropriate population and sample Appropriate measurement and intervention methods Adequate sample size Sound analysis plan Ethical issues well addressed Tight budget Realistic timelines

Clear, concise, well-organized Helpful table of contents and subheadings Good schematic diagrams and tables Consistent throughout

(Adapted from Hulley & Cummings)

Observation/Idea

Informal Questions

No Hypotheses(Exploratory Research)

Study Methods

Survey CohortStudy

Case-ControlStudy

Subjects

Data Analysis

Sample Size Calculations

Hypotheses(Explanatory Research)

Precision (Power) Power

Descriptive (Inferential) Inferential (Descriptive)

Ethical Considerations

Justification of research:Why is the research important?Are there gaps in the literature?Would the answers have clinical value?

Survey CohortStudy

Case-ControlStudy

RandomizedControl Trial

Limitations

Budget

Potential biasesFeasibilityCostLevel of Evidence (Causal Inference)Ethics

Study Design

Research Questions/Objectives

VariablesPrimary Outcome(s)Secondary Outcome(s)Covariates

Variables of Interest(Covariates)

Relevance to questions/hypothesesConfoundingJustification in literature/practicePsychometric properties

Who will be studied?How will you select and recruit them?How will you allocate them to groups?

Issues to consider/discuss

Literature Review

Can your study detect meaningful effects?How precise will estimates be?

How will you record, store, reduce data?How will you assess data quality?What statistical analyses will you use?

Be C

onsistent!

Will data be anonymous?How will you ensure confidentiality?Informed consent (consent form)

Personnel costs/consultants.Overhead, supplies, patient care.Justify!!!

Study Problem

Relevance of Project


5

HEALTH RESEARCH PROPOSAL WRITING Title The title should be informative, succinct and interesting. It should include the population of interest and the condition/issue being investigated. Example 1 Title: The Scope of Naturopathic Practice: An Emerging Profession Reference: Verhoef MJ, Boon HS, Mutasingwa DH. The scope of practice of naturopathic medicine in Canada: An emerging profession. Soc Sci Med 2006; 63; 409-417

Example 2 Title: Surgical Management of Stress Incontinence in Women: Randomized Controlled Trial of Trans-Obturator Tape Versus Tension-Free Vaginal Tape Reference: Robert M, Ross S, Brasher P, Fischer B, Jacobs P. Surgical Management of Stress Incontinence in Women: Randomized Trial of Trans-Obturator Tape versus Tension-free Vaginal Tape http://www.obgyn.ucalgary.ca/Documents/TOTv%5B1%5D.TVT.pdf Common Pitfalls

A title that is too brief or too long Use of incorrect terminology Not specifying the population of interest

Study Problem / Purpose A broad statement indicating the goals of the project.

Eg. exploration, description, explanation, prediction (hypothesis testing)

Keys to success:

Relevant Clear Logically argued


6

Example 1: Naturopathic Practice The purpose of the study is to describe naturopathic practitioners perception of their current scope of practice, and of the quality of the training they receive. These perceptions will be compared with current regulations governing naturopathic practice in the four regulated provinces. Example 2: Surgical Management of Stress Incontinence The purpose of this study is to compare outcome up to one year following surgery for stress incontinence using Trans-Obdurator Tape (TOT) versus Tension-free Vaginal Tape (TVT). Relevance / Study Rationale Why should the research be done?

Why is the research important? Has it been done before? Will the study benefit patients, increase knowledge and/or influence

policy? Will research resolve controversies? Would the answers have clinical value?

Keys to success Lead reviewers to answer the above questions. Should be reasonable given the proposed study Literature review will demonstrate that the research is justified Granting agencies may have specific priority areas, be sure to

explain how your study fits into those areas. Example 1: Naturopathic Practice Relevance / Rationale The results of this study will identify the degree of variation in naturopaths practices, as well as the potential degree of overlap with other practitioners scope of practice. This information is important in the debate about the professions identity, its role in the Canadian health care system and the development of new health care professions. The study results will also generate important information with respect to evaluating and further developing national training programs. This study will allow collection of baseline data that can be used to track the development of naturopathic medicine as a profession over the coming years.


7

Example 2: Surgical Management of Stress Incontinence Relevance/ Rationale Urinary incontinence (UI) is a common and distressing problem, affecting approximately 30 to 40% of older women.1-2 Stress urinary incontinence (SUI) is the most common type of UI in women.2 Conservative treatments include pelvic floor muscle training, and urethral or vaginal devices.3 However conservative treatments are often unsuccessful, and surgery becomes a possible solution. .. Following the introduction of TVT, it has gradually become the surgery of choice across much of Canada, including Calgary, and increasing numbers of women have been able to undergo surgery. TOT, as a new procedure, is being enthusiastically promoted by the manufacturers of the surgical devices used to carry out the procedure. However we need proper independent evaluation to determine which surgical approach is best for women. Before introducing TOT into clinical practice in Calgary, all clinicians have agreed not to use TOT outside the context of a randomized trial. If a trial is not undertaken, it is likely that TVT will remain the procedure of choice for stress incontinence, and that women in Calgary may be denied the optimum surgical treatment. ETC.. Literature Review A critical summary of research on a topic of interest, generally prepared to put a research problem in context or to identify gaps and weaknesses in prior studies so as to justify a new investigation. Generally starts off broadly addressing the problem then becoming more narrow and leading to your specific research question(s). This is your chance to build a case for doing your study! Justify the statements that you have made in your Relevance/Rationale

section. Reviewers will be looking to see whether your proposed work is

appropriate for the level of knowledge currently existing in that area. Keys to success:

- Identify gaps in knowledge or controversies - Focused on and leading naturally to your research questions/objectives - Be thorough, relevant, and up-to-date - Use primary sources of original research - Synthesize and be critical - Provide local background


8

Research Objectives / Research Questions Explicitly state what you propose to study. Form the foundation for the rest of the proposal. Will be used to assess the adequacy/appropriateness of the study's

proposed methods. Testable Logically derived from the literature review

Often includes:

Population of interest (person, place, time) Variables (independent and dependent) Relationship between variables being investigated

Keys to success:

Be clear and consistent Generally have only one or two primary research objectives Include the independent and dependent variables, if applicable Your objectives must be measurable. Objectives must be Relevant or novel (as youve established in the

literature review!). If it is a pilot study, state clearly the pilot objectives (e.g., testing the

feasibility of the study procedures, patient adherence to the study protocol, drop-out rates etc.).

Example 1: Naturopathic Practice Research Objectives:

1. To describe naturopathic practitioners perceptions of their scope of practice 2. To determine naturopathic practitioners opinion about their current scope

of practice 3. To examine naturopathic practitioners satisfaction with their training and

how well it prepared them for their current scope of practice. 4. To compare naturopathic medicines scope of practice as identified by

federal and provincial governing authorities with practitioners perceptions. Example 2: Surgical Management of Stress Incontinence Research Questions: Primary research question How effective is TOT compared to TVT in terms of objective cure (determined using a 1-hour pad test) at 12 months postoperatively?


9

Secondary research questions How effective is TOT compared to TVT in terms of: Subjective cure at 12 months postoperatively? Incontinence-specific quality of life at 6 weeks and 12 months postoperatively? Satisfaction with surgery at 12 months postoperatively? Return to usual activities and usual sex life after surgery? The prevalence of voiding dysfunction at 12 months postoperatively? Surgical complications, both short term and long term? Utility and cost? (an economic evaluation) METHODS

Study Design Issues to consider in design

Exploratory versus explanatory research Potential biases that may emerge Feasibility (logistics) Cost Level of evidence, limitations Ethics

Types of Designs: Descriptive (Survey, cross-sectional, correlational)

Observational/analytic

Cohort studies Prospective Retrospective

Quote of the Day: My life is an experiment I never had the chance to properly design (Diana Ballard) Case-control studies

Interventional

Quasi-experimental Randomized controlled trials

Keys to success: Clearly identify the type of design and justify your choice Describe using standard terminology Make sure what you have named your study fits with what will be

done


10

Ensure the proposed design is appropriate to address study objectives

Address any design issues in the literature review and limitations sections

Example 1: Naturopathic Practice Study Design The study design will consist of two components: 1) a survey of licensed Canadian naturopathic practitioners; and 2) an analysis of pertinent national and provincial documents in the public domain. SURVEY: Data will be collected by means of a standardized questionnaire, which will be mailed to all practitioners in the study group. Questionnaires will be numbered to identify those who have responded and, thus, to facilitate follow-up. A reminder will be sent 2 weeks after the initial mail out. Non-responders will be sent a second questionnaire 4 weeks after the initial mailing. Practitioners who have not responded 6 weeks after the initial mail-out will be contacted by phone and asked whether they would be willing to participate in a brief telephone interview to assess selected socio-demographic and practice characteristics, two brief questions assessing opinion about scope of practice, and the reason for not responding. This information will assist in assessing whether the sample of respondents is similar to the target group. The questionnaire will include DOCUMENT COMPARISON: The document analysis will include the collection of documents in the public domain that address issues around the scope of practice of naturopathic medicine. Example 2: Surgical Management of Stress Incontinence Research Design: The proposed study is a randomized controlled trial in which women with stress incontinence will be allocated either to receive either TOT or TVT procedures. The main outcome is effectiveness of the procedure, determined using a 1-hour pad test, and women will be followed up for 1 year after surgery. The study will be carried out according to the ICH Good Clinical Practice Guidelines. All seven surgeons participating in the trial are competent to undertake both the TOT and TVT procedures. Methods (in brief):


11

After patients agree to take part in the study and sign a consent form (should be included as an appendix ), baseline information will be collected and patients will be asked to complete a brief questionnaire (should be included as an appendix ). Women will be randomized to either TOT or TVT procedures (randomization described below). Hospital outcomes, including length of stay (hours for those passed to home care, days for any who require longer hospital stay) and surgical complications will be documented from hospital charts. All women receive a 6-week follow-up as standard of care: a structured data collection form will be used by surgeons to collect information on short-term outcome. At 12 months postoperatively, women will be asked to attend hospital for objective measurement of incontinence, and to complete a questionnaire: by that time, women will have recovered from surgery, and their incontinence (if remaining) will have stabilised. Subjects

Study Design - Common Pitfall The Incorrect Design (or the wrong objective) The primary objective is to determine if coffee drinking

causes pancreatic cancer. A case-control study will be conducted

Better: The primary objective of this study is to determine if there is an association between coffee consumption and pancreatic cancer.

Who will you study? Why? Specify eligible subjects - Target population: clinical & demographic characteristics - Accessible population: temporal & geographic characteristics Keys to success:

Clearly describe the study population Provide appropriate inclusion/exclusion criteria Justify the use of the study population and the sampling method

(consider potential bias, generalizability)


12

Provide all details of the sampling and recruitment methods (including any barriers to recruitment specific to the population and strategies you will use to address these - e.g., cultural barriers)

Do not confuse randomization with random selection Example 1: Naturopathic Practice Study Subjects SURVEY: The study group will consist of all licensed naturopathic practitioners (approximately 400) in Canada. Addresses will be obtained from a database maintained by the Canadian College of Naturopathic Medicine. Letters of support have been obtained from the Canadian Naturopathic Association and the Canadian College of Naturopathic Medicine. (Note: no inclusion/exclusion criteria - as all members of the target population are eligible for inclusion in this case) Example 2: Surgical Management of Stress Incontinence Patient Population: All women who elect for surgical management of their type II stress incontinence are eligible to participate in the trial. Evaluation of stress incontinence prior to surgery will be carried out according to the SOGC guidelines.22 4.1 Inclusion criteria Women may be included if they: - have type II stress incontinence, defined as leaking with increased abdominal pressure.22 - are eligible for both types of surgery. 4.2 Exclusion criteria Women will be excluded if they: - have vaginal prolapse requiring surgical repair - have had previous incontinence surgery - have overactive bladder or incontinence is caused only by bladder overflow - intend to have further children - have Alzheimers or Parkinsons disease, progressive neurological disease such as multiple sclerosis, or are immunocompromised - are unable to understand English - will be unavailable for follow-up.


13

4.3 Identifying and approaching patients Women referred to any of the collaborating clinicians, who fulfill the inclusion and exclusion criteria will be offered the opportunity to join the trial. During the period of recruitment to the study, TOT will be available only as a study intervention. The womans clinician will briefly introduce the study, then the woman will be referred to the study nurse not involved in the womans routine care, who will explain the trial in full and provide detailed information. Women are free to consult with their family before agreeing to join the study. Women who decide to join will sign a consent form. Allocation to Study Groups and Intervention Example 2: Surgical Management of Stress Incontinence Baseline, randomization and interventions 5.1 Baseline data collection After consenting to join the study, data will be extracted from the patients case notes about baseline details, including date of birth, weight, smoking status, and results from their 3 day voiding diary. All women will be asked to complete a questionnaire including incontinence-related quality of life and expectations of surgery. 5.2 Randomization Randomization will take place on the day of surgery. Women will be allocated to one of the two study groups: the TOT or TVT group. Permuted block randomisation with blocks of varying size (2 8) will be employed. Randomization will be stratified by surgeon. Next treatment allocation will not be known by the surgical team or the patient. The randomisation list will be generated by the study statistician (using ralloc procedure in Stata). The list will be kept by the Department of Obstetrics and Gynaecology data manager (or designated delegate not involved in the trial or patient care if she is absent). Women will be randomized to one procedure or the other: participating surgeons are competent to undertake both procedures, and so patients will receive their surgery from their chosen surgeon. That is, patients are allocated to a procedure and not to a surgeon who specialises in only one of the procedures. Neither surgeons nor patients will be blinded to the treatment they receive, however both procedures are minimally invasive, and the incisions required for the surgery are similarly small. 5.3 Intervention Women in both groups will have their surgery performed according to the usual practice of the operating surgeon. Anaesthesia will be either general or local, depending on the clinical state and choice of the patient, and according to the usual clinical practice of the anaesthetist. Where possible, the operations will be planned as day procedures.


14

5.3.1 TOT Group The TOT procedure will be carried out according to the recommendations of the device manufacturer. 5.3.2 TVT Group The TVT procedure will be carried out according to the recommendations of the device manufacture. 5.3.3 Post-operative care for all patients As usual in Calgary for day case patients, participants in the trial will be cared for in their homes. If necessary for clinical reasons, women will remain in hospital. 5.3.4 Further surgery If women require re-operation for incontinence during the 12 months following surgery, the surgery should use the treatment as randomized, unless in the surgeons clinical opinion, another operation is required. Randomization/Random Selection

Common Pitfall We will randomize 50 patients to either treatment or control group. During the 4 weeks of the study, it is estimated that approximately 80 patients will be eligible. Therefore, a random sample of fifty will be chosen. Be clear about the difference between randomization and random

selection. They are not the same thing. If Randomizing patients state how random allocation will be

performed. If random sample - indicate why and how a random sample will

be obtained (note: consecutive sample is not random)

Variables / Data Collection Identify the: Independent variable (exposure, intervention) Dependent variable (outcome) Covariates Where will the information come from (data sources)?

In descriptive and exploratory studies, this distinction is not made.

Exactly how will the information be collected (data collection methods)?


15

Keys to success: - Indicate relevance to research questions/objectives - Describe the variables in detail

Primary, secondary, covariates - Discuss psychometric properties (for measurement instruments, scales) - Provide justification for use (gold standards, other research, clinical practice). - Include letters of agreement if you will get data from a third party (eg. AHW)

Example 1: Naturopathic Practice Data Collection 1) SURVEY: Primary Variables (what will be measured):

Scope of practice (treatment modalities, types of health conditions treated, referral practices) Other Relevant Variables:

Sociodemographic information (age, gender, marital status, membership of professional organization, and license status)

Practice Characteristics (number of years in practice, location of practice, involvement in teaching/research, number of patients seen each week)

Education

Measurement Tool: The questionnaire was developed by the research team and has used valuable information by Boon (5) in qualitative interviews with Canadian Naturopaths. A draft copy of the questionnaire is included for review, along with discussion of the psychometric properties. The questionnaire will be pre-tested using a number of naturopaths, as well as independent assessors representing medicine, research, a wide range of other health professions, and government. 2) DOCUMENT COMPARISON: The document comparison will include legislation from the regulated provinces, the CAN Code of Ethics and CAN Standards of Practice. Variables to be assessed include: Treatment modalities and health conditions seen as part of naturopathic medicines

scope of practice Referral policies Rationale and philosophy underlying scope of practice.


16

Example 2: Surgical Management of Stress Incontinence Data collection methods 6.1 Operative and hospital details Operative details will be collected from the operating room notes. Details will include type of anaesthesia, length of operation, length of time in hospital (hours) and any intra- or post-operative complications. These details are collected rigorously in the case notes. If at the time of surgery compelling clinical circumstances occur that necessitate another procedure being performed than that allocated, the reasons for the change of procedure, and the actual operation performed will be noted. 6.2 Primary outcome: Objective evidence of cure Objective evidence of SUI will be obtained using a standardised pad test undertaken at 12 months following surgery. The pad test will be carried out using a modification 1 of the International Continence Society (ICS) recommendations:22

Description of Pad test: Retrograde filling of bladder with 300ml saline Pre-weighed collecting devide (pad) is put on and the test period begins During a 15 minute period the subject performs the following activities: - walk up and down 1 flight of stairs - standing up from sitting, 10 times - coughing vigorously,10 times - running on the spot for 1 minute At the end of the test the collecting device is removed and weighed

6.3 Secondary outcomes Subjective evidence of cure Subjective symptom assessment will take place at 12 months after surgery. Subjective cure is defined as either no experience of stress incontinence, or if urine loss has been no problem at all: Etc (Data collection methods provided separately for each secondary outcome)

Data Collection - Common Pitfall The Unknown Instrument The primary objective is to determine the degree of satisfaction patients have with outpatient surgery. A questionnaire will be mailed to patients asking about their degree of satisfaction with their hospital stay. Where did the questionnaire come from?


17

A Note about Questionnaires A questionnaire is a data collection instrument What reviewers are looking for:

How was the questionnaire developed (and by whom)? Are you using a Standardized Instrument? Has it been tested for Reliability Validity Has the questionnaire been pre-tested among the patient population

being studied? How will the questionnaire be administered?

Statistical Analyses and Sample Size In many proposals, the sample size section precedes the data analysis section. However, it is advisable to consider the analysis section first, as the sample size calculation should be based on the analysis of the primary research question. Data Analysis Section A detailed plan for handling and analyzing data, including procedures for:

Recording, storing and reducing data Assessing data quality Statistical analyses

Keys to success Indicate how you will test assumptions Provide reasonable detail Make sure you have described the analysis for each stated

objective Tips for the Data Analysis Section Make sure your objectives are clear, simple and testable. Think carefully about your outcome measure.

What constitutes clinical relevance? Is it appropriate, relevant and measurable?

Be concrete and explicit.

Turn your research question into something testable. Take the essential features of your problem and turn it into something

measurable. Think carefully about your outcome measure.


18

What constitutes clinical relevance? Is the outcome measure appropriate? Select an appropriate analysis tool.

Think about how you would present your final results. What convincing evidence would you present? What could you measure and report that would demonstrate

clinical/economic impact? Make table shells (these do not need to be included in the proposal).

How will you present the data from your study? What features of your patients would you need to describe in a manuscript?

How will tests of significance or models be presented? Comparison of means? Comparison of rates or proportions?

Sample Size Estimates: Inference Inference

In research, we usually cant measure everyone. Forced to make inferences regarding true or underlying

characteristics of a population on the basis of data collected from a sample.

The more subjects we use or measure, the more accurate our estimates will be.

If we measure too many, we will waste resources. If we measure too few, we wont be able to detect effects of interest.

Sample Size Calculations Keys to success:

If possible, consult a biostatistician (there are also some online sample size calculators available)

Justify the elements that you put into your calculation. For example:

control group is expected to have this response based on literature (eg. meta-analysis) .wed like to detect this response in the intervention group based on previous literature and clinical significance ..the standard deviation around the change is based on the literature

Base the sample size on the primary objective(s) and on the test of significance that will be used to test your primary study hypothesis.

If you have a fixed sample size, do a power calculation. If you are estimating a prevalence or mean score in a cohort study

or survey, justify the sample size on the widths of confidence


19

intervals (e.g. for prevalence, number of subjects needed to obtain 95% confidence intervals of +/- 10%.).

Account for anticipated dropout rates and or non-response rates in the sample size estimate.

Example 1: Naturopathic Practice Analysis Plan SURVEY: The data analysis will be mainly descriptive. Frequencies and summary measures will be selected according to the level of measurement for each of the key variables of interest (components of scope of practice, satisfaction with scope of practice and training and preparedness for practice). If numbers are sufficient, contingency analyses will be conducted to identify whether socio-demographic, education and practice characteristics are associated with the key variables of interest. Depending on the level of measurement Chi-squared (discrete data, difference in proportions), t-test (continuous data) or one-way ANOVA will be used. Data will be analyzed using SPSS software. DOCUMENT COMPARISON: Treatment modalities and health conditions will be assessed using quantitative content analysis where possible. Quantitative content analysis is a research technique for the objective, systematic and quantitative description of the manifest content of communication (9). The analysis consists primarily of counting occurrences in each identified category (eg. Scope of practice). The remaining variables will be analyzed using qualitative content analysis. This involves analyzing the content of the narrative to determine themes or patterns. A theme is a recurring regularity emerging from an analysis of qualitative data and may be a phrase, sentence or paragraph embodying ideas or making an assertion about a topic. The distinction between quantitative and qualitative content analysis is often more a mater of degree than a clear dichotomy. The quantitative results of the document analysis will be compared with the survey participants responses in terms of descriptive summary measures. Where required (ie. when documents reveal major differences across provinces), the responses of the survey participants will be stratified by province.


20

Example 2: Surgical Management of Stress Incontinence Sample Size and Analysis Plan 7.1 Sample size The collaborating urogynaecologists agree that a 15% difference in cure rate would be required to change clinical practice in Calgary: that is either to introduce TOT into usual practice, or to restrict its introduction. Published cure rates for TVT vary considerably, depending on the definition of cure used, however we believe that using our definition, the cure rate for TVT will be 73% (based on the negative pad test results of Ward et al 12). We wish our study to detect a 15% higher cure rate for TOT. In order to be able to detect a difference of this order, assuming 80% power and a significance level of 0.05, a sample of 100 patients per group would be required (total 200). We believe that we shall be able to recruit this number of patients in one year, based on our previous TOT study, where we recruited 52 patients over the period of 3 months. The importance of follow-up will be stressed to patients so that we can reduce the possibility of drop-outs during the 12 month follow-up period. 7.2 Analysis Analysis will be by intention to treat: that is, women will remain in the group to which they have been allocated by randomization, no matter what surgery they receive. A single analysis is planned: when all women have completed their 12 month follow-up. Descriptive statistics (means, standard deviations, proportions) will be calculated to check for any major dissimilarities in the study groups with regard to baseline information. The primary analysis will be the comparison of the proportion of patients who demonstrate cure on the 1-hour pad test (pad weight gain less than 1g over the test period) at 12 months following surgery. The proportion will be compared between treatment groups using Fishers Exact Test. Point and interval estimates for the differences will be determined. It is expected that almost all patients will be available for follow-up at 12 months. Analyses of the secondary outcomes will be as follows: Subjective urinary incontinence: the proportions of patients reporting no problematic

stress incontinence in the past 7 days, at 12 months following surgery. Incontinence-related quality of life: the mean scores of UDI-6 and IIQ-7 will be

compared between the two groups. Satisfaction with surgical outcome: will be compared at 12 months between

treatment groups for the proportions of patients who (a) would have the same operation again, (b) would recommend the same operation to someone else, and (c) whose outcome met expectation prior to surgery.


21

Return to usual activities after surgery: the mean days to return to usual activities will be compared between the two groups.

Prevalence of voiding dysfunction: the proportions of patients experiencing voiding dysfunction will be compared between groups at 12 months postoperatively.

Surgical complications: the proportions of patients experiencing surgical complications will be compared between groups.

Ect. 7.3 Data management All data will be managed in Access, with analysis carried out using SAS. Data will be double entered, and discrepancies reviewed by an independent third person. Data forms will be checked prior to data entry, and range and logic checks carried out regularly on entered data to ensure the quality of the data. Any results that appear to be out of range or do not comply with the logic checks will be checked with the hospital or clinic case notes and the womans surgeon.

Charlie Brown was addressing his baseball team at the end of the season. He recited numerous dismal statistics such as: Runs scored by us 12, by opponents 125! At the end of the speech he yells out: And what are we going to do about it? .to which the team answers in unison: Get a new Statistician!

Ethical Consideration Keys to success

Describe the recruitment procedures Describe procedures for maintaining subject privacy Follow the guidelines for your organisation (consent form

templates) Describe harms and benefits Describe alternatives to study participation If asking for a waiver of consent, justify explicitly why you cannot

get consent (HIA) If using a placebo group, make sure it fits into the TCPS guidelines


22

Example 1: Naturopathic Practice Ethical Considerations An introductory letter will be included with the questionnaire and will describe the purpose of the study, the research team, and confidentiality of the data (see Appendix X). Follow-up letters are also included in Appendix X. Return of the questionnaire is considered to be consent. Practitioners who do not respond to the questionnaire 6 weeks after two reminder letters will be telephoned and asked for a brief telephone interview. A script outlining how practitioners will be approached is included in Appendix Y. Example 2: Surgical Management of Stress Incontinence Ethical considerations All women considering taking part in the trial will be given a consent form describing the surgical options being evaluated, and given ample time to consider whether they wish to join the trial. TOT will be offered to women in Calgary only as part of the trial: this is a new technology, that has been evaluated in few very limited settings. It would be unethical to open the use of this technology more widely until it has been more thoroughly tested. However both procedures are currently in use in the Canadian health care system, and therefore neither technique can be considered as experimental treatments. An important consideration for the trial is ensuring that women are not coerced to take part: it is for this reason that we have stipulated that a research nurse not involved in the woman's care is assigned the task of describing the study to women and obtaining their consent. Women may take as long as they wish to decide whether to take part in the study, and are free to discuss their decision with friends, family members or their family physician, before deciding. The risks of taking part in the trial are the same as the usual risks of having surgery for stress incontinence. The additional requirements of the trial over and above those generally associated with stress incontinence surgery are an additional visit at 12 months for pad testing and uroflow, and additional questionnaires prior to surgery and later about the outcome of surgery and the associated costs. Women may benefit from the closer follow-up required by the trial. All data for the trial will be collected on data forms designed specifically for the trial. Personal details will be collected to ensure that the research nurse can maintain contact with the women over the full 12 months of follow-up. However these personal details


23

will be stored separately from the study data. Study data will be entered into a password-protected database, that will be accessible only to study staff. Data will be used only for the purpose of the study, and will not be passed to anyone else. Results of the study will be presented with the data from the groups aggregated together: no individual patient will be identified in any way in reports of the study.

http://www.mediclicks.net/Nexium/cartoons.asp

Limitations

- Be upfront about the limitations of your study. - This is one of the most important sections of the proposal (shows

reviewers you have thought through all aspects of your study) - Be clear about why you are not able to overcome these limitations a priori. - How might these impact the findings? - If there are clear issues, better to talk about them than leave them open. - Put a positive spin on any limitations

While this design fails to control for . . . , it allows us to . . .


24

Budget Proposals are usually submitted to secure funds for research Keys to success

Read instructions carefully Itemize costs

Personnel, consultants, equipment, supplies, travel, patient care costs, indirect costs, other expenses.

Do not inflate costs Justify the budget!

Plans for Dissemination Developing a dissemination plan is a key part of the collaborative research process. While publication of the research in a peer reviewed journal is one method of dissemination, it is also important to think of other strategies to get your research findings to those who can use it (i.e., decision-makers, healthcare providers, health organizations etc.). Working through an initial dissemination plan can help you focus the project and identify key audiences. The Canadian Health Services Research Foundation (CHSRF) has put together some tips for developing a dissemination plan (available through their website). Communication Notes: Developing a Dissemination Plan (CHSRF) http://www.chsrf.ca/knowledge_transfer/pdf/dissemination_plan_e.pdf Other Possible Sections Introduction or abstract Timelines Research team (describing the research expertise of the team and the

contributions each member will make to the study) Strengths (can be incorporated into the Limitations section) Discussion of the Broader Research Program

The specific format will depend on the guidelines of the agency/organization to which you are submitting.


25

General Tips Consistency is important - refer back to your objectives as you develop

the proposal. Avoid using too much jargon

http://www.mediclicks.net/Nexium/cartoons.asp If proposing to do a Pilot Study note:

Pilot Studies are:

Preliminary studies to provide guidance in deciding how to conduct a future full-scale study (e.g., testing feasibility, recruitment, adherence to study protocol, etc.).

Pilot studies are NOT:

To test specific hypotheses; Studies that are called pilots to justify using an inadequate

number of subjects to test a hypothesis (to avoid doing a sample size calculation);

Studies that happen to only need a small number of subjects to have the power to address the research question.


26

For Further Reading on Proposal Writing Lock LF, Spirduso WW, Silverman SJ. Proposals that work. Newbury Park, CA: Sage Publications, 1993.

Gitlin LN, Lyons KJ. Successful grant writing. Strategies for health and human service professionals. New York: Springer Publishing Co., 1996. Granstmanship and Review Processes

Barnard J. Keys to writing a competitive grant. Journal of Pediatric Gastroenterology & Nutrition. 35(2):107-10, 2002 Aug. (Available online through UofCs e-journals) Bartz C. It all starts with an idea. Alzheimer Disease and Associated Disorders 1999;13(Suppl 1):S106-10. (Available in print in UofCs Health Sciences Library. A good article to review proposal writing and points to remember. Eg. identifying limitations before your reviewers do.) Calgary Health Region. Writing Successful Grant Applications http://www.calgaryhealthregion.ca/hecomm/comdev/pdf/grantsbooklet_Jan2005.pdf Canadian Institutes of Health Research. Peer Review Process http://www.cihr-irsc.gc.ca/e/4656.html Ferrara JL, Schmaier AH. Hematology grants workshop. I. Writing successful NIH grant applications: a primer. Hematology (Am Soc Hematol Educ Program). 2002;:484-9. (Available online through UofCs e-journals. A nice article describing and justifying the process of NIH grant reviews. It could be applicable to other granting agency reviews.) Grey M. Top 10 tips for successful grantsmanship. Res Nurs Health. 2000 Apr;23(2):91-2. (Available online through UofCs e-journals) Hilsden RJ. A beginners guide to the research proposal. An online guide with tips and common pitfalls. http://www.ucalgary.ca/md/CAH/research/files/hilsden/res_prop.htm Hurley AC, Wells N. The grant application: making yours stand out across the review cycle. Alzheimer Disease and Associated Disorders 1999;13(Suppl1):S120-2. (Available in print at UofCs Health Sciences Library. This article provides a description of what sorts of reviews your application would go through when it reaches the NIH.) Koren G. How to increase your funding chances: 15 pitfalls in medical grant applications. Annals of the Royal College of Physicians and Surgeons of Canada 1997;30(6):335-7. (Available in print at UofCs Health Sciences Library. A list of frequently forgotten tips.) Kundel HL, Walsh C. Preparing a research grant application budget. Investigative Radiology 1993;28(Suppl 2):S13-6. (Available in print at UofCs Health Sciences Library.) Lusk SL. Developing an outstanding grant application. Western Journal of Nursing Research 2004;26(3):367-73. (Available from publisher website easiest to link through PubMed. An article based on developing a successful proposal for nursing research.) Rasey JS. The art of grant writing. Current Biology. 9(11):R387, 1999 Jun 3. (Available online through UofCs e-journals)


27

Overall Quality of the StudyQuality of the ProposalQuality of the Presentation

Proposal Writing

Documents

Transcript of Proposal Writing