Protocol precision: why "close enough" does not exist in clinical research

In clinical practice, a physician exercises judgment. A blood pressure of 142/88 measured after two minutes of rest is close enough to a blood pressure measured after five minutes of rest — the clinical decision will be the same. The physician adjusts. The physician contextualizes. The physician treats the patient, not the number.

In clinical research, the protocol replaces judgment with specification. And this is not a bureaucratic inconvenience — it is the foundation of data integrity across a multi-site trial. Consider a Phase III cardiovascular trial running at 120 sites across 15 countries. If Site 42 in Munich measures blood pressure after two minutes of rest while Site 97 in Philadelphia measures after five minutes of rest, and Site 11 in Tokyo measures with the participant standing, the blood pressure data across those sites are not comparable. The statistical analyses that will determine whether this drug works — analyses that will inform whether the FDA approves it, whether patients receive it — depend on every site measuring blood pressure under the same conditions.

This is what ICH E6(R3) Annex 1, Section 2.5.2 means when it states that "the investigator should comply with the protocol, GCP and applicable regulatory requirements." Compliance is not approximate. It is specific to each protocol-defined procedure, each specified condition, each required time point.

I want to be direct about a tension I have observed throughout my career. Coordinators who come from strong clinical backgrounds — nurses, medical assistants, respiratory therapists — sometimes struggle with this transition more than those who come from non-clinical backgrounds. The reason is counterintuitive. Clinical experience gives you excellent procedural skills but also gives you the habit of exercising clinical judgment about how to perform procedures. In clinical practice, that judgment is valued. In clinical research, that judgment — when it overrides protocol specifications — creates deviations.

The discipline is this: read the protocol's procedure specification as written. Execute it as written. Document it as performed. If the specification seems unnecessary, impractical, or clinically questionable — raise it through the proper channels (query to the sponsor, protocol amendment request). But do not quietly substitute your clinical judgment for the protocol's specification. That is where deviations are born.

Vital signs and physical measurements: every specification matters

Let me walk through the major screening procedures and highlight where protocol specifications typically diverge from routine clinical practice. This is not an exhaustive catalog of every possible protocol requirement — your protocol is the authoritative source for your study. But these are the patterns I see most frequently, and the deviations they produce when coordinators operate on clinical autopilot rather than protocol precision.

Blood pressure

Blood pressure measurement is, in my experience, the single most common source of procedural deviations at screening visits. Not because coordinators cannot measure blood pressure — they can — but because protocols specify conditions that clinical practice does not require.

Position and rest period. Most protocols specify seated blood pressure after a defined rest period — typically three to five minutes. Some protocols additionally require supine or standing measurements, often in a specified sequence (seated, then supine, then standing). The rest period is not optional. A blood pressure measured immediately after the participant walks from the waiting room is a post-exertion measurement, not a resting measurement. If the protocol specifies five minutes of seated rest, set a timer. Document the start of the rest period. Measure at five minutes — not at three, not at "about five."

Replicate measurements. Many protocols require duplicate or triplicate blood pressure readings at defined intervals — often one to two minutes apart. All readings must be recorded. Do not discard the first reading because it seems high. Do not average the three readings and record only the average (unless the protocol specifically instructs averaging). Record each measurement individually with its timestamp.

Device specification. If the protocol specifies an automatic oscillometric device, do not use a manual sphygmomanometer — even if you believe your auscultatory technique is superior. If the protocol specifies a particular manufacturer or model, use that device. Check calibration status before every screening visit.

Arm specification. Many protocols designate the arm (left or right) or specify "same arm for all measurements." Document which arm was used. If the protocol says left arm and the participant has a peripherally inserted central catheter in the left arm, do not simply switch to the right arm — contact the sponsor's medical monitor to determine the appropriate approach, and document the query and response.

Heart rate, respiratory rate, and temperature

These measurements appear deceptively simple, but protocols often specify measurement methods that differ from routine clinical shorthand. Heart rate may be required from a full 60-second radial pulse count rather than a 15-second count multiplied by four. Respiratory rate may need to be measured over 60 seconds without the participant's awareness (to prevent voluntary rate changes). Temperature may need to be measured orally, tympanically, or temporally — with the protocol specifying the method, and consistency across visits required.

Height, weight, and derived measurements

Height is typically measured once at screening and not repeated. Weight is measured at screening and at subsequent visits. The protocol may specify whether shoes are on or off, whether the participant should be in a hospital gown or street clothes, and whether the same scale must be used for all measurements. Body Mass Index (BMI) calculations, if protocol-required, use the protocol-specified formula. Waist circumference, if required, follows the anatomical landmarks specified in the protocol — the World Health Organization (WHO) midpoint between the lower rib margin and the iliac crest is not the same as the National Institutes of Health (NIH) measurement at the superior iliac crest.

ECG procedures: timing, positioning, and quality

Electrocardiogram acquisition during screening visits follows a pattern similar to vital signs — the procedure itself is routine, but the protocol's specifications transform it into a precisely controlled measurement.

Timing requirements. Many protocols specify ECG timing relative to other events: pre-dose (at screening, this is the baseline), at specific time windows post-dose in later visits, or at specific clock times. At the screening visit, the ECG is typically a resting baseline, but the protocol may specify that it be obtained after a rest period, before physical exertion (including the physical examination), or at a specific point in the visit sequence. Follow the sequence specified in the schedule of assessments.

Replicate ECGs. Some protocols — particularly in cardiology and oncology where QTc prolongation is a concern — require triplicate 12-lead ECGs at defined intervals, often one minute apart. Each tracing is a separate acquisition. Do not obtain one ECG and print three copies. Each must be a distinct recording, obtained at the specified interval, with the timestamp documented.

Participant positioning. The standard is supine for 12-lead ECGs, but the protocol may specify a minimum rest period in the supine position before acquisition — typically five to ten minutes. The participant should not be speaking during acquisition, as this introduces artifact. Hands should be at the sides or resting on the abdomen, not gripping the edges of the examination table.

Lead placement. Electrode placement must follow the standard 12-lead configuration. Lead reversal — placing electrodes in the wrong position — is a common error that produces recognizable but misleading tracings. If the ECG is being transmitted to a central reading laboratory, the central reader will identify lead reversal and request a repeat, which may fall outside the protocol's timing window. Verify placement before acquisition.

Quality assessment. Before the participant leaves the ECG suite, review the tracing for quality. Is the baseline stable? Is there excessive artifact from muscle tremor? Are all 12 leads recording? Is the paper speed correct (typically 25 mm/sec)? A tracing that is uninterpretable — due to artifact, baseline wander, or lead malfunction — must be repeated immediately. Discovering quality issues after the participant has left creates a logistical problem that may not be resolvable within the screening window.

Laboratory specimen collection: precision from tourniquet to transport

Laboratory specimen collection at the screening visit is often the most logistically complex procedure — and the one with the most opportunities for deviation. The complexity arises from the number of specifications that must be satisfied simultaneously: fasting status, tube types, order of draw, processing requirements, storage conditions, and shipping logistics.

Fasting verification

If the protocol requires fasting specimens, verify fasting status before the draw — and document the verification. This means asking the participant directly: "Can you confirm that you have had nothing to eat or drink except water since [time]?" Record the participant's response and the duration of the fast. As you learned in the previous lesson, "fasting" means different things to different participants. Some believe coffee without sugar qualifies. Others assume that medication taken with juice is acceptable. Be specific in your verification question, and document the participant's answer verbatim if there is any ambiguity.

Tube types and order of draw

Every protocol has a laboratory manual that specifies the exact specimen tubes required for each test. This is not interchangeable. A serum separator tube (SST, gold or red-and-grey top) is not the same as a sodium citrate tube (light blue top). An EDTA tube (lavender top) is not the same as a sodium heparin tube (green top). Using the wrong tube type invalidates the specimen — the laboratory cannot run the assay, the result is missing, and the specimen must be recollected (which may require another visit and another fast).

The order of draw — the sequence in which different tube types are filled — follows a standard phlebotomy sequence designed to prevent cross-contamination between tube additives. Most protocols follow the Clinical and Laboratory Standards Institute (CLSI) recommended order, but some protocols specify a modified order. Follow the protocol's laboratory manual.

Processing, storage, and shipment

The moment blood leaves the participant's vein, a clock starts. Many specimens require processing within a defined window — centrifugation within 30 minutes, aliquoting within 60 minutes, freezing within two hours. These windows are not suggestions. They are specifications validated during assay development, and specimens processed outside these windows may produce unreliable results.

Know your protocol's processing requirements before the draw — not after. If the protocol requires centrifugation at a specific speed and duration, confirm that the centrifuge is calibrated to the correct settings. If the protocol requires freezing at -70 degrees Celsius, confirm that the ultra-low freezer has available space and is functioning within the temperature range before you collect the specimen. Discovering that the freezer is full or malfunctioning after you have drawn the blood creates an unrecoverable situation.

For studies using a central laboratory, shipping logistics require advance planning. Know the shipping schedule — most central laboratories have designated pickup days and times. Know the packaging requirements: ambient, refrigerated (wet ice or cold packs), or frozen (dry ice). Know who is responsible for scheduling the courier. A perfectly collected, perfectly processed specimen that sits on a loading dock for 48 hours because the courier was not scheduled is a wasted specimen and a potential protocol deviation.

Questionnaires and patient-reported outcomes: standardization is everything

Questionnaires and patient-reported outcome (PRO) instruments at the screening visit serve two purposes: they generate baseline data that will be compared to post-treatment measurements, and they may contain scores that function as eligibility criteria (a depression scale score above a threshold, a pain score within a specified range, a cognitive assessment above a minimum).

The cardinal rule is this: administer the instrument exactly as the protocol and the instrument's administration manual specify. This sounds obvious, but the deviations I have seen in this area are remarkably creative.

Self-administered versus interviewer-administered. The protocol specifies which mode of administration is required. A self-administered questionnaire is completed by the participant independently — the coordinator provides the instrument, explains any general instructions, and confirms completion, but does not read the questions aloud or interpret them. An interviewer-administered instrument is read to the participant by a trained administrator, exactly as written, without paraphrasing or elaboration. Mixing these modes — reading a self-administered questionnaire aloud because the participant asks for help, or handing an interviewer-administered instrument to the participant to save time — changes the measurement properties of the instrument and introduces bias.

Completion verification. Before the participant leaves, review every questionnaire for completeness. Check for skipped items, double-marked responses, and illegible handwriting. A missing response on a validated instrument may invalidate the total score. It is far easier to ask the participant to complete a missed item while they are still in the clinic than to discover the gap during data entry and attempt a retrospective correction.

Scoring. If the protocol requires the site to score the instrument (not all do — some are scored centrally), perform the scoring calculation per the instrument's manual. Scoring errors are a form of data error that can affect eligibility determinations. Double-check arithmetic. If the instrument uses reverse-scored items, confirm you have applied the reverse scoring correctly.

Version control. Validated instruments have specific versions. If the protocol specifies the PHQ-9 (Patient Health Questionnaire-9), do not use the PHQ-2. If the protocol specifies version 3.0 of a quality-of-life instrument, do not use version 2.0. Using the wrong version produces data that are not comparable to other sites and may not be scorable using the protocol's specified scoring algorithm.

Contemporaneous documentation: ALCOA in practice

Every procedure you perform during the screening visit must be documented — and documented as it occurs, not from memory at the end of the visit, not the next morning, not reconstructed from partial notes scribbled on a glove or a paper towel.

This is the principle of contemporaneous documentation, and it is one of the five pillars of the ALCOA framework that governs data integrity in clinical research. ICH E6(R3) Annex 1, Section 2.12.1 requires that investigators "ensure the integrity of data under their responsibility, irrespective of the media used." Section 2.12.5 further specifies that the investigator should "ensure the accuracy, completeness, legibility and timeliness of the data reported." The ALCOA acronym — often extended to ALCOA-C or ALCOA-CCEA — captures these requirements in a memorable framework.

Putting ALCOA into the screening visit

In practical terms, contemporaneous ALCOA-compliant documentation during a screening visit means carrying your source document through the visit and recording each data point as it is generated. For a typical screening visit, this looks like the following.

Before you begin the blood pressure measurement, document the start of the rest period with the time. When the first reading appears on the monitor, record it immediately — the value, the time, the arm, the position. When the second reading appears two minutes later, record it immediately. When the third reading appears, record it. Initial the entry. Move to the next procedure.

When you draw laboratory specimens, document the time of the first tube drawn, the time of the last tube drawn, the tubes collected, and the fasting duration confirmed by the participant. When you process the specimens, document the centrifuge start time, the centrifuge settings, and the time aliquots were frozen. Each of these entries is made at the time the event occurs — not reconstructed later.

This discipline feels burdensome at first. But it becomes automatic with practice, and it produces source records that withstand monitoring visits, audits, and regulatory inspections without requiring explanations, corrections, or justifications. In the words of one experienced monitor I have worked with for years: "The best source documents are the ones I never have to ask about."

Common procedural deviations: recognition and prevention

Understanding where deviations happen most frequently during screening visits allows you to build prevention into your workflow rather than discovering errors after the fact. The table below captures the deviations I see most often — not theoretical risks, but patterns that have appeared repeatedly across sites, therapeutic areas, and sponsor organizations.

The culture of deviation prevention

I want to note something that extends beyond checklists and timers. The sites that consistently produce clean screening visit data — the sites where monitors find few deviations and auditors find fewer — share a cultural characteristic. They treat protocol fidelity as a professional standard, not an administrative burden. The coordinator at these sites does not view the five-minute rest period before blood pressure as a waste of time. They view it as a specification that exists for a scientific reason, one that they are professionally obligated to execute.

This mindset is trainable. It begins with understanding why each specification exists — that triplicate blood pressures reduce measurement variability, that fasting specimens eliminate dietary confounders, that standardized questionnaire administration ensures measurement equivalence across sites. When you understand the reason behind the requirement, compliance becomes purposeful rather than mechanical. And purposeful compliance is sustainable in a way that mechanical compliance is not.

ICH E6(R3) Section 3.10 and Principle 7 frame this as proportionate quality management. Not every procedure carries equal risk to data integrity. But screening procedures — the procedures that determine whether a participant enters the trial, that establish the baseline against which all treatment effects are measured — carry disproportionate weight. A deviation at screening can propagate through the entire dataset. An inaccurate baseline blood pressure distorts the change-from-baseline analysis at every subsequent visit. An improperly scored screening questionnaire may enroll a participant who does not meet the study population criteria, diluting the treatment effect signal.

Prevention is not perfectionism. It is proportionate attention to the procedures that matter most.

Protocol precision: why "close enough" does not exist in clinical research

In clinical practice, a physician exercises judgment. A blood pressure of 142/88 measured after two minutes of rest is close enough to a blood pressure measured after five minutes of rest — the clinical decision will be the same. The physician adjusts. The physician contextualizes. The physician treats the patient, not the number.

In clinical research, the protocol replaces judgment with specification. And this is not a bureaucratic inconvenience — it is the foundation of data integrity across a multi-site trial. Consider a Phase III cardiovascular trial running at 120 sites across 15 countries. If Site 42 in Munich measures blood pressure after two minutes of rest while Site 97 in Philadelphia measures after five minutes of rest, and Site 11 in Tokyo measures with the participant standing, the blood pressure data across those sites are not comparable. The statistical analyses that will determine whether this drug works — analyses that will inform whether the FDA approves it, whether patients receive it — depend on every site measuring blood pressure under the same conditions.

This is what ICH E6(R3) Annex 1, Section 2.5.2 means when it states that "the investigator should comply with the protocol, GCP and applicable regulatory requirements." Compliance is not approximate. It is specific to each protocol-defined procedure, each specified condition, each required time point.

I want to be direct about a tension I have observed throughout my career. Coordinators who come from strong clinical backgrounds — nurses, medical assistants, respiratory therapists — sometimes struggle with this transition more than those who come from non-clinical backgrounds. The reason is counterintuitive. Clinical experience gives you excellent procedural skills but also gives you the habit of exercising clinical judgment about how to perform procedures. In clinical practice, that judgment is valued. In clinical research, that judgment — when it overrides protocol specifications — creates deviations.

The discipline is this: read the protocol's procedure specification as written. Execute it as written. Document it as performed. If the specification seems unnecessary, impractical, or clinically questionable — raise it through the proper channels (query to the sponsor, protocol amendment request). But do not quietly substitute your clinical judgment for the protocol's specification. That is where deviations are born.

Vital signs and physical measurements: every specification matters

Let me walk through the major screening procedures and highlight where protocol specifications typically diverge from routine clinical practice. This is not an exhaustive catalog of every possible protocol requirement — your protocol is the authoritative source for your study. But these are the patterns I see most frequently, and the deviations they produce when coordinators operate on clinical autopilot rather than protocol precision.

Blood pressure

Blood pressure measurement is, in my experience, the single most common source of procedural deviations at screening visits. Not because coordinators cannot measure blood pressure — they can — but because protocols specify conditions that clinical practice does not require.

Position and rest period. Most protocols specify seated blood pressure after a defined rest period — typically three to five minutes. Some protocols additionally require supine or standing measurements, often in a specified sequence (seated, then supine, then standing). The rest period is not optional. A blood pressure measured immediately after the participant walks from the waiting room is a post-exertion measurement, not a resting measurement. If the protocol specifies five minutes of seated rest, set a timer. Document the start of the rest period. Measure at five minutes — not at three, not at "about five."

Replicate measurements. Many protocols require duplicate or triplicate blood pressure readings at defined intervals — often one to two minutes apart. All readings must be recorded. Do not discard the first reading because it seems high. Do not average the three readings and record only the average (unless the protocol specifically instructs averaging). Record each measurement individually with its timestamp.

Device specification. If the protocol specifies an automatic oscillometric device, do not use a manual sphygmomanometer — even if you believe your auscultatory technique is superior. If the protocol specifies a particular manufacturer or model, use that device. Check calibration status before every screening visit.

Arm specification. Many protocols designate the arm (left or right) or specify "same arm for all measurements." Document which arm was used. If the protocol says left arm and the participant has a peripherally inserted central catheter in the left arm, do not simply switch to the right arm — contact the sponsor's medical monitor to determine the appropriate approach, and document the query and response.

Heart rate, respiratory rate, and temperature

These measurements appear deceptively simple, but protocols often specify measurement methods that differ from routine clinical shorthand. Heart rate may be required from a full 60-second radial pulse count rather than a 15-second count multiplied by four. Respiratory rate may need to be measured over 60 seconds without the participant's awareness (to prevent voluntary rate changes). Temperature may need to be measured orally, tympanically, or temporally — with the protocol specifying the method, and consistency across visits required.

Height, weight, and derived measurements

Height is typically measured once at screening and not repeated. Weight is measured at screening and at subsequent visits. The protocol may specify whether shoes are on or off, whether the participant should be in a hospital gown or street clothes, and whether the same scale must be used for all measurements. Body Mass Index (BMI) calculations, if protocol-required, use the protocol-specified formula. Waist circumference, if required, follows the anatomical landmarks specified in the protocol — the World Health Organization (WHO) midpoint between the lower rib margin and the iliac crest is not the same as the National Institutes of Health (NIH) measurement at the superior iliac crest.

ECG procedures: timing, positioning, and quality

Electrocardiogram acquisition during screening visits follows a pattern similar to vital signs — the procedure itself is routine, but the protocol's specifications transform it into a precisely controlled measurement.

Timing requirements. Many protocols specify ECG timing relative to other events: pre-dose (at screening, this is the baseline), at specific time windows post-dose in later visits, or at specific clock times. At the screening visit, the ECG is typically a resting baseline, but the protocol may specify that it be obtained after a rest period, before physical exertion (including the physical examination), or at a specific point in the visit sequence. Follow the sequence specified in the schedule of assessments.

Replicate ECGs. Some protocols — particularly in cardiology and oncology where QTc prolongation is a concern — require triplicate 12-lead ECGs at defined intervals, often one minute apart. Each tracing is a separate acquisition. Do not obtain one ECG and print three copies. Each must be a distinct recording, obtained at the specified interval, with the timestamp documented.

Participant positioning. The standard is supine for 12-lead ECGs, but the protocol may specify a minimum rest period in the supine position before acquisition — typically five to ten minutes. The participant should not be speaking during acquisition, as this introduces artifact. Hands should be at the sides or resting on the abdomen, not gripping the edges of the examination table.

Lead placement. Electrode placement must follow the standard 12-lead configuration. Lead reversal — placing electrodes in the wrong position — is a common error that produces recognizable but misleading tracings. If the ECG is being transmitted to a central reading laboratory, the central reader will identify lead reversal and request a repeat, which may fall outside the protocol's timing window. Verify placement before acquisition.

Quality assessment. Before the participant leaves the ECG suite, review the tracing for quality. Is the baseline stable? Is there excessive artifact from muscle tremor? Are all 12 leads recording? Is the paper speed correct (typically 25 mm/sec)? A tracing that is uninterpretable — due to artifact, baseline wander, or lead malfunction — must be repeated immediately. Discovering quality issues after the participant has left creates a logistical problem that may not be resolvable within the screening window.

Laboratory specimen collection: precision from tourniquet to transport

Laboratory specimen collection at the screening visit is often the most logistically complex procedure — and the one with the most opportunities for deviation. The complexity arises from the number of specifications that must be satisfied simultaneously: fasting status, tube types, order of draw, processing requirements, storage conditions, and shipping logistics.

Fasting verification

If the protocol requires fasting specimens, verify fasting status before the draw — and document the verification. This means asking the participant directly: "Can you confirm that you have had nothing to eat or drink except water since [time]?" Record the participant's response and the duration of the fast. As you learned in the previous lesson, "fasting" means different things to different participants. Some believe coffee without sugar qualifies. Others assume that medication taken with juice is acceptable. Be specific in your verification question, and document the participant's answer verbatim if there is any ambiguity.

Tube types and order of draw

Every protocol has a laboratory manual that specifies the exact specimen tubes required for each test. This is not interchangeable. A serum separator tube (SST, gold or red-and-grey top) is not the same as a sodium citrate tube (light blue top). An EDTA tube (lavender top) is not the same as a sodium heparin tube (green top). Using the wrong tube type invalidates the specimen — the laboratory cannot run the assay, the result is missing, and the specimen must be recollected (which may require another visit and another fast).

The order of draw — the sequence in which different tube types are filled — follows a standard phlebotomy sequence designed to prevent cross-contamination between tube additives. Most protocols follow the Clinical and Laboratory Standards Institute (CLSI) recommended order, but some protocols specify a modified order. Follow the protocol's laboratory manual.

Processing, storage, and shipment

The moment blood leaves the participant's vein, a clock starts. Many specimens require processing within a defined window — centrifugation within 30 minutes, aliquoting within 60 minutes, freezing within two hours. These windows are not suggestions. They are specifications validated during assay development, and specimens processed outside these windows may produce unreliable results.

Know your protocol's processing requirements before the draw — not after. If the protocol requires centrifugation at a specific speed and duration, confirm that the centrifuge is calibrated to the correct settings. If the protocol requires freezing at -70 degrees Celsius, confirm that the ultra-low freezer has available space and is functioning within the temperature range before you collect the specimen. Discovering that the freezer is full or malfunctioning after you have drawn the blood creates an unrecoverable situation.

For studies using a central laboratory, shipping logistics require advance planning. Know the shipping schedule — most central laboratories have designated pickup days and times. Know the packaging requirements: ambient, refrigerated (wet ice or cold packs), or frozen (dry ice). Know who is responsible for scheduling the courier. A perfectly collected, perfectly processed specimen that sits on a loading dock for 48 hours because the courier was not scheduled is a wasted specimen and a potential protocol deviation.

Questionnaires and patient-reported outcomes: standardization is everything

Questionnaires and patient-reported outcome (PRO) instruments at the screening visit serve two purposes: they generate baseline data that will be compared to post-treatment measurements, and they may contain scores that function as eligibility criteria (a depression scale score above a threshold, a pain score within a specified range, a cognitive assessment above a minimum).

The cardinal rule is this: administer the instrument exactly as the protocol and the instrument's administration manual specify. This sounds obvious, but the deviations I have seen in this area are remarkably creative.

Self-administered versus interviewer-administered. The protocol specifies which mode of administration is required. A self-administered questionnaire is completed by the participant independently — the coordinator provides the instrument, explains any general instructions, and confirms completion, but does not read the questions aloud or interpret them. An interviewer-administered instrument is read to the participant by a trained administrator, exactly as written, without paraphrasing or elaboration. Mixing these modes — reading a self-administered questionnaire aloud because the participant asks for help, or handing an interviewer-administered instrument to the participant to save time — changes the measurement properties of the instrument and introduces bias.

Completion verification. Before the participant leaves, review every questionnaire for completeness. Check for skipped items, double-marked responses, and illegible handwriting. A missing response on a validated instrument may invalidate the total score. It is far easier to ask the participant to complete a missed item while they are still in the clinic than to discover the gap during data entry and attempt a retrospective correction.

Scoring. If the protocol requires the site to score the instrument (not all do — some are scored centrally), perform the scoring calculation per the instrument's manual. Scoring errors are a form of data error that can affect eligibility determinations. Double-check arithmetic. If the instrument uses reverse-scored items, confirm you have applied the reverse scoring correctly.

Version control. Validated instruments have specific versions. If the protocol specifies the PHQ-9 (Patient Health Questionnaire-9), do not use the PHQ-2. If the protocol specifies version 3.0 of a quality-of-life instrument, do not use version 2.0. Using the wrong version produces data that are not comparable to other sites and may not be scorable using the protocol's specified scoring algorithm.

Contemporaneous documentation: ALCOA in practice

Every procedure you perform during the screening visit must be documented — and documented as it occurs, not from memory at the end of the visit, not the next morning, not reconstructed from partial notes scribbled on a glove or a paper towel.

This is the principle of contemporaneous documentation, and it is one of the five pillars of the ALCOA framework that governs data integrity in clinical research. ICH E6(R3) Annex 1, Section 2.12.1 requires that investigators "ensure the integrity of data under their responsibility, irrespective of the media used." Section 2.12.5 further specifies that the investigator should "ensure the accuracy, completeness, legibility and timeliness of the data reported." The ALCOA acronym — often extended to ALCOA-C or ALCOA-CCEA — captures these requirements in a memorable framework.

Putting ALCOA into the screening visit

In practical terms, contemporaneous ALCOA-compliant documentation during a screening visit means carrying your source document through the visit and recording each data point as it is generated. For a typical screening visit, this looks like the following.

Before you begin the blood pressure measurement, document the start of the rest period with the time. When the first reading appears on the monitor, record it immediately — the value, the time, the arm, the position. When the second reading appears two minutes later, record it immediately. When the third reading appears, record it. Initial the entry. Move to the next procedure.

When you draw laboratory specimens, document the time of the first tube drawn, the time of the last tube drawn, the tubes collected, and the fasting duration confirmed by the participant. When you process the specimens, document the centrifuge start time, the centrifuge settings, and the time aliquots were frozen. Each of these entries is made at the time the event occurs — not reconstructed later.

This discipline feels burdensome at first. But it becomes automatic with practice, and it produces source records that withstand monitoring visits, audits, and regulatory inspections without requiring explanations, corrections, or justifications. In the words of one experienced monitor I have worked with for years: "The best source documents are the ones I never have to ask about."

Common procedural deviations: recognition and prevention

Understanding where deviations happen most frequently during screening visits allows you to build prevention into your workflow rather than discovering errors after the fact. The table below captures the deviations I see most often — not theoretical risks, but patterns that have appeared repeatedly across sites, therapeutic areas, and sponsor organizations.

The culture of deviation prevention

I want to note something that extends beyond checklists and timers. The sites that consistently produce clean screening visit data — the sites where monitors find few deviations and auditors find fewer — share a cultural characteristic. They treat protocol fidelity as a professional standard, not an administrative burden. The coordinator at these sites does not view the five-minute rest period before blood pressure as a waste of time. They view it as a specification that exists for a scientific reason, one that they are professionally obligated to execute.

This mindset is trainable. It begins with understanding why each specification exists — that triplicate blood pressures reduce measurement variability, that fasting specimens eliminate dietary confounders, that standardized questionnaire administration ensures measurement equivalence across sites. When you understand the reason behind the requirement, compliance becomes purposeful rather than mechanical. And purposeful compliance is sustainable in a way that mechanical compliance is not.

ICH E6(R3) Section 3.10 and Principle 7 frame this as proportionate quality management. Not every procedure carries equal risk to data integrity. But screening procedures — the procedures that determine whether a participant enters the trial, that establish the baseline against which all treatment effects are measured — carry disproportionate weight. A deviation at screening can propagate through the entire dataset. An inaccurate baseline blood pressure distorts the change-from-baseline analysis at every subsequent visit. An improperly scored screening questionnaire may enroll a participant who does not meet the study population criteria, diluting the treatment effect signal.

Prevention is not perfectionism. It is proportionate attention to the procedures that matter most.