Standard operating procedures
This leaf answers how to write a standard operating procedure that the people doing the task can follow, how to bring it into use and keep it current, and why a procedure that is followed to the letter can still miss its purpose. The general craft of an instruction is in clear instructions, where procedures sit among a clinical trial's records is in essential records, and the evidence an inspector asks for that a procedure was followed is in the quality and compliance process.
Know what a procedure is and what it is for
The fullest account here is Paula Hayes's chapter on writing and implementing procedures in The Complete Guide to Medical Writing (2007), and most of this leaf follows it. A standard operating procedure describes how to carry out a defined task consistently. A policy establishes the objective and acceptable approach; a procedure translates it into operational steps. Procedures can support training, role clarity, delegation, audit, and investigation of failures, but their existence alone does not establish safe practice. The copy of the book admitted here is incomplete, and two of its missing-page gaps fall inside this chapter, one in its account of a procedure's limitations and sections and one after its paired examples, so a point absent below may be absent from the copy rather than from the chapter.
Hayes writes from British pharmacy practice of that period, and the requirements the chapter names illustrate settings where procedures support governance: the Royal Pharmaceutical Society of Great Britain's requirement, from 1 January 2005, that pharmacies establish and operate written procedures covering the dispensing process, and the procedures that unlicensed aseptic facilities in National Health Service hospitals were required to implement after the Farwell report of 1995. Actual obligations depend on the applicable setting and rules.
In a regulated clinical trial the term is defined and the documents are inventoried. ICH E6(R3) defines standard operating procedures as detailed, documented instructions to achieve uniformity in the performance of a specific activity. It counts them among the records that are not specific to one trial and may be retained outside the trial-specific repositories, which essential records sets out in full. It also expects a trial's monitoring plan to reference the sponsor's applicable policies and procedures, as monitoring and audit describes. The guideline is read here from a supplied Markdown conversion whose fidelity has not been independently reviewed.
Define the work and responsibility
State the purpose, intended standard, and scope, including relevant boundaries. Identify what happens, when and where it happens, who performs it, and the competence or authorization needed. Describe responsibility under normal conditions and how exceptional circumstances are handled. Limit the procedure to a manageable task while preserving its essential dependencies.
The chapter's template includes process steps, responsibility, audit, additional information, author and authorization, version, and review date. These fields connect the instructions to ownership and maintenance.
Write instructions that can be followed
Use direct, precise steps in task order, with necessary conditions visible. A narrative can conceal individual actions; the chapter's paired prescription-intake examples show how separating them helps execution. Explain necessary terms and cross-reference related procedures rather than reproducing competing versions of shared instructions. See clear instructions.
A diagram can clarify a physical action, and a flowchart can expose decisions, alternative paths, and where they rejoin. The chapter's spillage example demonstrates this structure; its chemical concentrations and timings are not a current operational protocol.
Implement and maintain the procedure
Allow time for development and involve the people who perform the task. A competent colleague can draft with appropriate guidance, but the procedure still needs review and practical testing. Walk through a draft with users to find missing actions, ambiguity, and unrealistic assumptions. Explain its purpose as well as its steps.
Keep approved instructions accessible and identifiable. Maintain readable copies and synchronize updates wherever copies are used. Train users and check understanding: a list that staff sign confirms that they have read a procedure, and Hayes adds a quiz on its content to test that they understood it, which the signature does not show.
Distinguish an authorized response to an exceptional situation from simply disregarding the procedure. Define escalation and professional responsibility, and record deviations and their reasons for review. Repeated deviations may reveal problems in the procedure, training, resources, or conduct. Maintain communication between the document's owner and its users.
Align two groups' procedures when the groups merge
A merger or an internal reorganisation leaves two sets of procedures describing one body of work, and a poster abstract by Nicola Bond and JoAnn Tuan of Thousand Oaks, California, printed in AMWA's journal in 2009, describes a method for aligning them. The starting condition is predictable: when groups merge, their processes and the documents describing those processes overlap and are misaligned, and the goal is to align the processes while ensuring the process documents still cover all of the work between them. Coverage is the part that gets lost, because deduplication naturally removes documents and nothing naturally notices the gap that leaves. The abstract is all the record gives of the poster, and it reports its result without data. The issue reached the tree through a layout-aware conversion of its printed pages.
How they organised it is transferable to any documentation merger. A project team was assembled from subject matter experts in both groups, with objectives set to ensure compliance and efficiency, and five areas were put in scope: document management, change control, planning, standards, and content. Each area was given its own team leader, who evaluated the current status and the business needs and produced a plan carrying requirements, resources, approach, deliverables and a timeline. Splitting the work by area rather than by originating group is what stops the exercise becoming a negotiation between the two sets of authors, and the reported result was improved processes and more efficient use of resources with minimal disruption to the business.
Expect a trained user to turn the procedure into a ritual
Training is not the end of the problem, and the failure mode it leaves behind is invisible from inside the document. Hayes's remedies for a procedure that sits unused are the ones above: involve the people who perform the task, keep the document to hand, and test that it was understood. Jack Aslanian, a medical graduate who divided his time between clinical practice and medical editing, wrote in EMWA's journal in 2012 about a failure that training does not prevent. His essay describes watching surgery clinic nurses disinfect a room after a procedure: gloved, they sprayed disinfectant and wiped the surfaces down, performing more or less as they had been trained to, then still gloved gathered the soiled linen and instruments, left the room and pulled the door shut — and later opened the same door, turning the contaminated handle, without gloves. His conclusion is not that the procedure was wrong but that it had become a ritual, and that despite time spent in training the nurses were unaware they had turned a scientifically reasonable prescription into one. His account of how that happens names three routes into ritual, and each is recognisable in a procedure that has stopped working: the absence of a more sustainable alternative, the compulsion to appear militant in the face of peril, and procedural and political correctness. What the piece shows about its form is in essays and creative nonfiction. The issue that carries it was admitted as an unrepaired conversion, so broken ligatures and run-together words may remain in its text.
What that costs a procedure writer is a specific check rather than a general worry. The steps were followed and the objective was not achieved, because the procedure covered the actions inside the room and not the boundary the user crosses on the way out. Aslanian's parallel from airport security is the same shape: take off your shoes, and pick up whatever is on the airport floor; have a bag hand-searched by an inspector who does not change gloves between one bag and the next. Where a procedure's purpose is containment, the transitions — leaving, handing over, moving to the next case — are where it fails, and they are also the steps least likely to be written down because nobody thinks of them as part of the task.
He is explicit that the answer is not to abandon the procedures: without safety policies and procedures we would be worse off. What he objects to is the belief that a written procedure can fully allay the risk, given a human element that professional and personal lives — increasingly complex, overloaded with information, bounded by time limits and expectations of productivity — reliably degrade. The two questions he sets at the head of the essay are worth asking of any procedure being drafted: how much will this solution cost, and how reliable is it? He closes on a third, aimed at the apparatus rather than at the step: is your safety net safe?