START NOW

Please start below.

If you feel unready to get started then just put “TBD” for “to be determined” for sections currently unknown, and prefacing sections that are still under development with “(under development)”.  Then it can be shared which will foster collaboration as early in the process as possible.

After clicking “submit”, invite others to collaborate, and work toward research-design-excellence, rather than mere research-design-completion.  Inviting others in on the design phase, before it’s fully “ready”, is a great way to get key players on board, improve buy-in, and improving study design before it’s too late to make design changes.

  • RECOMMENDATION:

    Enter only the items with a red star for now and press submit. Then after you receive an email to validate your account you can edit the other sections, then invite your collaborators to assist in the design of the study.

  • We recommend you include a unique identifier so you can easily find it with a search
  • Please provide any details necessary to do the research: Protocols, dosage, frequency, special instructions
  • Provide the justification for the range of values being tested. In general the values should range from the lowest safe limit to the highest safe limit with minimal granularity in the different experimental split groupings to identify any sweet spots. Include the source for these upper and lower limits. If exploratory in nature these limits may safely test ranges previously untested.
  • Enter which grants, if any, that you will be pursuing, as well as other funding sources
  • List all diseases, illnesses, health conditions, symptoms for which this study is intended to impact.
  • Persons Affected

    The total number of persons affected with a condition that may benefit if the hypothesis is true.

  • in the US
  • in the world
  • Recommendation: When determining what variables to measure, unless you are using statistically significant sample sizes, please use "continuous variables" (example: a real number where it is unlikely than any patient will have the same exact value) instead of using "categorical variables" (ie. where the values are likely to be repeated among the patients, such as mortality which only has 2 possible values: survived vs died). Even if sample sizes are expected to be large enough for "categorical variables" to reach statistical significance, if there is even a chance that the sample size might be too small (for example: patient enrollment decreases over time) then use continuous variables instead.

    Make at least one of these based on directly measurable human-function / human-wellness indicators (example: normal-cell survival, QLQ-c30 score, tissue / organ functional rating like SOFA score, etc). Avoid using biomarkers that are pathology dependent wherein the presumed pathology may be incorrect, but rather use such markers as secondary outcome measurements.

    If a pathology-dependent variable is used for the Primary Outcome Measure (eg. IL-6) provide a source or rational argument that removes all doubt that it will correlate perfectly with the desired outcome. If a perfect level of confidence can't be reached, move the variable to a Secondary Outcome Measure.

    Cautions: Avoid using primary measurements that may indicate negative response in spite of providing an overall benefit to the patient. This can happen when an incorrect pathology is presumed, or when the metric is too specific.

    Provide instructions when and how measurements are taken, as well as any other instructions that can insure "all other things being equal". Select a metric that is easily, consistently, and reliably taken, with as few uncontrolled factors as possible.

    Note that if a primary metric fails the whole study is generally reported as failing regardless whether the intervention was beneficial or not.
  • These metrics can often be more instructive and beneficial to understanding the pathology of a disease or the patient's response mechanism, but may also have a more tenuous relationship with the outcome than might be presumed. If (and this is extremely important) there are measurements that may not correlate with the desired outcome then include them here rather than as Primary Outcome Measures.

    Also, as was the case with Primary Outcome Measures, if the sample population is too low for a specific measurement to be statistically significant, avoid even stating them as a secondary outcome measurement. In practice this means avoiding "categorical variables" unless the sample size is large enough to demonstrate a statistical significance in most cases with those variables.
  • Please identify any variables (other medications taken, patient age, lifestyle factors, co-morbid factors, other co-factors, etc) that might influence efficacy (+ or -). This can be use for statistical matching to reduce model dependence, bias, group imbalances, and improve statistical significance (so long as there isn't an endogeneity or simultaneity problem). This is useful for all studies, but especially observational studies making their significance approach that of an RCT. After identifying these variables make them required information for each patient. See this video for more info on best use of statistical matching (note: PSM should be avoided for endogeneity or simultaneity cases).
  • Please include studies related to this study that demonstrate presumed efficacy of treatment. Provide citations to peer reviewed journals (AMA format is preferred), and PMID with links whenever possible. Separate each with a blank line.
  • Include the treatments that provide the best outcomes, as well as those which are most cost effective. This includes conventional treatments as well as other treatments current under investigation.
  • Include all potential life-threatening contraindicating conditions, as well as patient criteria and ideal and acceptable limits for participating in the treatment. This criteria must be explicitly stated, not subject to interpretation.
  • Estimated cost per patient participating in the study. Include all clinical costs, setup costs, training costs, equipment costs, pharmaceutical costs.
  • Cost per patient - in practice

    Total estimated cost for this treatment, per patient, after the treatment is given FDA approval, presuming that happens. This should include clinical costs as well as pharmaceutical costs.

  • Minimum
  • Typical
  • Maximum
  • Treatment Savings Per Patient, if any

    The expected savings per patient for this treatment compared to standard care. This should include clinical costs as well as drug and medical technology costs.

  • Minimum
  • Typical
  • Maximum
  • Registration, media creation, engagement / recruiting / networking / outreach, regulatory compliance, legal, documentation, publishing, etc.
  • If this experiment is designed to validate the claims of a specific protocol that has previously been tested, please indicate the person(s) who designed and/or administered the protocol, who might be consulted to verify that the exact protocol is followed in this validating study. If no such person(s) exist, enter your own name in this section.
  • * The "CF Study-Validation Verification" form ensures that everything, from timing, to dose, to patient criteria, to things like hospital food, contraindications, and a range of doses, etc, are all reasonable and are within the parameters most likely to demonstrate efficacy.
  •  
    Save Draft