Scheduling the clinical study report
A clinical study report is written against a final date that the sponsor fixes before drafting begins. The schedule is planned from the report's complexity and from what drives each task. The analysis and reporting tasks take very different times across the industry, a date that will not move is met by persuading colleagues and managing clients' expectations, and the writing itself can be shortened by taking waste out of it.
The final date
The sponsor predetermines the date of the final report, and that date should take into account the requirement to disclose summary results publicly within one year of the end of the trial. The report's authors have to meet it by scheduling the preparation of the integrated report's components effectively.
The reporting scenario has to be considered early, because reporting comes after clinical, regulatory and statistical tasks that may be delayed during the project. It seems reasonable to expect a delay upstream to delay the final report date as well, but this cannot be assumed, and it rarely happens.
Planning around complexity and task drivers
Planning has to take account of the study's complexity and the resulting complexity of the report. It also has to take account of the precursors and drivers of each medical-writing task, such as the database lock and the availability of the final tables, figures and listings. Report complexity is estimated from study and reporting variables: the number of subjects, the study phase, the indication, the number of secondary efficacy variables, the number of unique tables, figures and listings, and the complexity of the analysis. A high-complexity report takes longer to author than a medium or simple one.
Breaking the integrated report into smaller deliverables, each prepared on a timescale that fits the final date, is recommended. The front end, including the unpopulated in-text summary results tables, and the appendices should be completed before the final tables, figures and listings, which drive draft reporting, arrive. Regulatory writers call this practice frontloading. Because timelines are often tight, it is common practice, done to alleviate stress later on. In a clinical study report, the patients and methods section, or front end, is largely independent of the actual results and so can be written before the final statistical outputs are available. Too much frontloading, though, can waste time if the work has to be thrown away, and can be dangerous if many fiddly adjustments are needed later, with the risk that they are overlooked.
Reporting from draft tables, figures and listings is a common strategy, but it is not efficient and is not recommended, because changes between the draft and final data will mandate changes in the report.
Typical durations of the analysis and reporting tasks
There are no industry standard durations for the analysis and reporting tasks. Between 2008 and 2013, industry professionals, predominantly regulatory medical writers working for sponsoring pharmaceutical companies, contract research organisations or independently, gave the typical average duration of each task for a moderate-complexity phase III study in 200 to 400 subjects. Each gave it in working days rather than as a range, for their own organisation or, for a freelancer, from experience of working with a range of organisations.
| Task | Respondents | Mean (SD) | Range |
|---|---|---|---|
| Last subject data in-house to database lock | 77 | 6.6 (5.6) days | 1–30 days |
| Database lock to draft tables, figures and listings | 77 | 13.9 (9.6) days | 1–56 days |
| Draft to final tables, figures and listings | 75 | 10.6 (6.1) days | 1–30 days |
| Final tables, figures and listings to first draft report | 78 | 16.9 (8.2) days | 5–45 days |
| First draft report to final report | 78 | 25.7 (21.1) days | 3–120 days |
| Client review cycles | 77 | 2.6 (0.8) cycles | 1–5 cycles |
| Quality assurance on the final integrated report | 77 | 6.6 (5.6) days | 1–30 days |
Preparing a first draft from the final outputs took a mean of just over three working weeks, and its range of 5 to 45 working days shows how greatly the task varies across the industry. The time from first draft to final report was also wide ranging, possibly because the number of client review cycles between report versions varies.
Refining the timeline and negotiating to meet the date
Algorithm-generated reporting timelines offer a useful guide at the outset of a project, and project-specific refinements afterwards add value. When the final date cannot move, the author has to bring wider skills to bear: persuading colleagues in related but independent functions to keep to timelines, and managing client expectations, are integral to producing the final report on time. By understanding the regulatory requirements and using project intelligence, a writer can schedule and author a high-quality final report on time for a preregistration drug trial in the EU.
Lean methods in the writing process
Developing a clinical study report is time-consuming, and key results and messages may be missed because of extensive documentation and overwriting. Lean principles such as Six Sigma, from the manufacturing industries, are designed to track and reduce sources of error and waste. In medical writing, these occur in rewriting, redoing, redundancy, verbosity and review cycles, for example.
In one application to writing the report, lean methods were applied in these areas:
- To standardise, the report template carried embedded instructions, and plain language was recommended and used.
- To streamline, the team used collaborative review tools, held data review meetings before drafting the results, and highlighted changes after each review.
- To simplify, numbers already summarised in tables and graphs were not repeated in the text.
- To leverage technology and tools, the team developed macros for project management, for creating mapping tables, for finding abbreviations and for extracting comments, and acquired tools for splitting PDFs and comparing documents to eliminate tedious tasks and compare draft versions. Time-zone differences among global teams were also used to optimise turnaround, through efficient hand-offs, around-the-clock work and clear communication.
These process improvements and lean thinking saved 10% to 15% of the time from drafting to approval and minimised variation across clinical teams for one type of report, and quality-control checks indicated a reduction in error rates. The most important time savings came from macros for tedious tasks: for 50 phase 1 reports, 200 hours were cut from building mapping tables, 37.5 hours from abbreviation checks and 37.5 hours from collating review comments. Systematically eliminating waste brought a focus on what is important in writing the report while ensuring consistent quality.