Form library, copying and eCRF from Excel

The YouEDC standard forms (CDASH codes, Russian and English), a copy of a form or of a whole study, forms and visits from an Excel workbook.

Registries and observational studies look alike: demographics, diagnosis, therapy, adverse events, outcomes. The form library saves you from typing the same again: take a standard form of YouEDC or one of its fields, copy your own or another study's, take the whole eCRF of another study or upload a specification from Excel. Open the eCRF builder and click Form library.

YouEDC standard forms

One library for the whole portal: 165 forms in 18 groups. The core forms by CDASH (visit, consent, eligibility, demographics, medical history, family history, surgery, allergies, vaccination history, medications, procedures, vital signs, orthostatic test, height and weight, anthropometry, physical, eye and neurological exam, ECG, Holter, imaging, adverse events, end of treatment and of participation, survival status, smoking, alcohol, other substances, social characteristics); study conduct and GCP (protocol deviations, re-consent, inclusion and exclusion log, unblinding, the 24-hour SAE report, death, contacts with the patient, adjudication); laboratory (blood counts, chemistry, liver, inflammation, iron, cardiac markers, vitamins, blood group, insulin, the glucose tolerance test, coagulation, urine, lipids, HbA1c, kidneys, thyroid, serology and others; every lab form has the sample time and fasting); study drug (randomisation, exposure with route and batch, dosing log, compliance, infusion, overdose, accountability, PK and biospecimens); infections; vaccines; oncology (diagnosis, RECIST 1.1 target, non-target and new lesions, radiotherapy, surgery, ECOG and Karnofsky, biomarkers, lines of therapy); cardiometabolic (the diabetes, hypertension, heart failure, kidney and atrial fibrillation forms, CHA2DS2-VASc, HAS-BLED, MACE); neurology and psychiatry (Glasgow coma scale, modified Rankin, Hoehn and Yahr, stroke, seizures, headache diary, CGI); respiratory (spirometry, mMRC, exacerbations, peak flow diary); gastroenterology and hepatology (Child-Pugh, MELD, symptoms, endoscopy); rheumatology (DAS28); dermatology and cosmetics (PASI, phototype, instrumental measurements, tolerance, patch test); pregnancy and children; medical devices; health economics; registries (Charlson comorbidity index, lines of therapy, discharge); patient-reported (own questionnaire, symptoms, diary, satisfaction, adherence, the patient's global assessment and two validated questionnaires, the PHQ-9 and the GAD-7).

• Variable codes by CDASH: AETERM, VSDAT, SYSBP. Answer codes are short (M, MILD, Y) and every list is matched to CDISC Controlled Terminology, so the data later goes into SDTM without manual work.
• Two languages: every text of the forms exists in Russian and English. A form is added in the language of the interface; the preview switches between RU and EN.
• Search and filters: the groups are on the left with the number of forms in each; the list filters itself while you type a name, a question or a code (“haemoglobin”, AETERM, “ECG”), with no reload. Ticks narrow it to forms the patient fills in, logs, or forms the study does not have yet. The study type keeps only the fields it needs: for a registry a form comes without the fields only an interventional trial needs, and the other way round.
• Preview beside the list: a click on a form opens it on the right, the list stays where it was. Questions, codes, types, units, ranges, answer codes of scored questions, importance (core, recommended, optional), where the field goes in SDTM and the built-in edit checks. Every check is verified on test cases. On a phone the preview opens under the list.
• One field: the Field button in the preview adds one question to your form together with its answer list, for example HbA1c into a visit form. A calculated field is not taken this way: it refers to other fields of its form. Nor is a question of a validated questionnaire: only the whole form.
• Several forms at once: tick the forms in the list and click Add the chosen forms at the bottom: they go into the eCRF draft in one step, with one record in the audit trail.

Fields marked “not added with the form” are taken one by one. These are race and ethnicity: in Russia they are a special category of personal data (152-FZ, art. 10) and are collected only when the protocol needs them. The library holds no text of licensed scales (EQ-5D, SF-36 and others). WHO-5, AUDIT and EPDS are templates with the right number of questions and answers and the total score, but without the questions' text: paste the official text from the owner's permission or licence before the release. For the other scales the “Questionnaire result” form keeps only the date and the total score, and the questionnaire itself is used under your organisation's licence. When the form code is taken, the new form gets a code from its name.

Validated questionnaires: PHQ-9 and GAD-7

The group “Patient-reported” has two free validated questionnaires: the PHQ-9 (depression) and the GAD-7 (anxiety). Their owner, Pfizer Inc., lets anyone reproduce, translate, display and distribute them with no permission. The library marks them validated and English only; the preview shows the owner, the terms, the reference and the credit line “Developed by Drs. Robert L. Spitzer, Janet B.W. Williams, Kurt Kroenke and colleagues…”, which the form carries under the questions.

• Word for word. The instructions, the questions and the answers are the official English text; the item codes are those of CDISC (PHQ0101–PHQ0109, the difficulty question PHQ0110, GAD0101–GAD0107). The form comes into the study in scale mode, with its passport filled in and its texts locked, so nobody rewords it by accident.
• English only. The library has no official Russian version of them yet. In a study in Russian, or one that offers a Russian translation, the builder marks the form “English only” in the list of forms and explains it on the form page; the questions and answers stay in English, word for word, also for a patient who reads Russian. Do not translate them yourself: a validated questionnaire is translated only by its owner’s official translation. The name of the form, the score and the checks are ours and are in both languages.
• Score and band. The total (PHQ0111, 0–27; GAD0109, 0–21) is counted on save when every item is answered, and so is the severity band: PHQ-9 1 minimal (0–4), 2 mild (5–9), 3 moderate (10–14), 4 moderately severe (15–19), 5 severe (20–27); GAD-7 1 minimal (0–4), 2 mild (5–9), 3 moderate (10–14), 4 severe (15–21). The patient sees neither.
• Item 9. An answer above “Not at all” to item 9 of the PHQ-9 (thoughts of being better off dead or of self-harm) opens a query to the site: “PHQ-9 item 9 positive: the investigator assesses suicide risk”. The patient does not see it.

Licensed instruments are not in the library: SF-36 and SF-36v2, EQ-5D, HADS and the like. Nor, for now, is RAND-36: RAND asks commercial users to write to it first.

Disease packs

A pack is the forms of one disease with a schedule of visits to start from. There are thirteen now: two basic sets to start a study and eleven by disease:

• Type 2 diabetes: diabetes history and the classes of glucose-lowering drugs, complications, HbA1c, fasting glucose, lipids, chemistry, urine albumin and creatinine, a log of hypoglycaemia with levels 1–3 and the summary of continuous glucose monitoring (time in range, below and above it, variability).
• Arterial hypertension: hypertension history with the grade and drug classes, office blood pressure of three readings with the mean of the second and third, 24-hour monitoring, a home blood pressure diary the patient fills in through a link, organ damage, ECG and kidneys.
• Chronic heart failure: history and the phenotype by ejection fraction, NYHA class, signs of congestion, echocardiography, NT-proBNP and BNP, potassium and creatinine, the six-minute walk test and a log of heart failure events (admissions, urgent visits with intravenous diuretics).
• Chronic kidney disease: history with the GFR and albuminuria categories, creatinine, eGFR as reported by the laboratory, cystatin C and electrolytes, albuminuria, blood count, lipids and a log of kidney replacement therapy.
• Interventional trial: the basics and Registry or observational: the basics: the forms any study of that type needs (consent, eligibility, demographics, history, medications, adverse events, protocol deviations, end of participation; for a registry also the source of data, comorbidity and outcomes). Start with one of them, then add a disease pack.
• Atrial fibrillation: history of AF, CHA2DS2-VASc, HAS-BLED, ECG and Holter, coagulation and kidneys, a log of strokes and of major cardiovascular events.
• Asthma and COPD: spirometry, the mMRC breathlessness scale, exacerbations and a peak flow diary.
• Solid tumour, RECIST 1.1: diagnosis, target, non-target and new lesions, ECOG, prior therapy, surgery, radiotherapy, biomarkers and survival.
• Psoriasis and Rheumatoid arthritis: PASI or DAS28, inflammation markers, the patient's global assessment, infections and screening for them, lines of therapy.
• Cosmetic product: efficacy and tolerance: phototype, the dermatologist's assessment, instrumental measurements, tolerance, patch test, product use and the participant's own view.
• Vaccine: vaccination, reactogenicity diary, immunogenicity samples, vaccination history, pregnancy and its outcome.

Every pack has an enrolment visit and follow-up visits with windows (the four cardiometabolic packs at 3, 6 and 12 months, the hypertension and heart failure packs also at one month), logs of therapy and of the disease's events, an adverse event and the end of participation. It is a reasonable start for a registry or a one-year study, not a protocol: everything changes in the builder.

A pack is taken in two ways:

• When the study is created: the “Template” step of the wizard has a group “By disease, from the form library”. The study gets all forms and visits of the pack at once.
• In a study that already exists: the form library of the builder has a section Disease packs with the button Add the pack. Only the forms the study has not taken from the library yet are added: demographics or adverse events do not appear twice. With With its visits ticked, the visits of the pack come after yours, with your forms of the same kind already in them. Without it only the forms come: put them into visits in the matrix.

The forms of the packs are in the library in their groups (the lab tests in “Laboratory”) and can be taken one by one too. Classifications with well-known thresholds (the levels of hypoglycaemia, the grades of hypertension, NYHA, the heart failure phenotypes by ejection fraction, the KDIGO categories) are described in our own words. Licensed questionnaires (KCCQ, MLHFQ and the like) and the Borg scale are not in the packs: their total score goes into the “Questionnaire result” form under the owner's licence. Echocardiography measurements and the glucose monitoring summary have short YouEDC codes: the sponsor's plan says how they go to SDTM.

The link with the library

A form from the library remembers where it came from. Its page in the builder shows “From the YouEDC library: DM 1.0.0” and next to it either “as in the library” or how many fields the team has changed since. When the library has a newer version of the form, that shows too.

The Compare with the library link shows your form next to the library field by field: what matches, what was changed (for example the question: “Systolic blood pressure” in the library, “Sitting SBP” in yours), which field of the library your form lacks and which fields are your own. Nothing changes by itself: a field of the library is taken with the Take from the library button, in the language the form was taken in, and goes into the eCRF draft like any change.

The library of the organisation

An organisation can have a form library of its own: the forms it adopted as the standard for its studies. It opens from Form library in the menu on the left and is seen only by the people of the organisation; another organisation does not see it.

How a form gets into the library. A person of the organisation finishes a form in the builder of one of its studies (or takes a YouEDC standard form and adapts it), presses Propose to the library on the form’s page, gives the form’s code in the library and writes what changed and why. A new code adds a new form; the code of a form already there proposes its next version. Nothing changes in the study.

Approval. A steward of the library sees the fields of the version, its checks and what changed against the last approved version, and approves it with an electronic signature: the password (and the e-mail for the first signature of a session). The signature keeps the name, the time and the SHA-256 of exactly this version. The steward can also reject the version with a reason, which the author reads. The author can withdraw a proposal while it waits. The stewards are the admins of the organisation and the people an admin names on the library’s page. With two stewards or more, nobody approves their own version.

In the studies. The approved forms show in the Library of the organisation section of the Form library page in the builder of every study of the organisation; only the latest approved version is offered. A form is taken with Add and stays linked to the library (lib4:AE@2): its page shows “From the library of the organisation”, how many fields the team changed and whether a newer version is out, and the field-by-field comparison works as for the YouEDC standard forms.

Retiring. A steward can retire a form with a reason: new studies are no longer offered it, and the studies that took it keep it. Every action in the library goes to the audit trail.

A demo study has the library of an invented demo sponsor: forms approved before the study and while it ran, and one form waiting for approval. You build the demo’s eCRF, so you are its steward and can approve it with your password.

A copy of a form of this study

The Copy button next to a form copies it with all fields, show-if conditions, checks and answer lists. For example, a “Month 6” and a “Month 12” form that share most questions: copy and adjust.

From another study

Choose a study where you are the project administrator and copy one form or the whole eCRF: visits with windows, repeats and conditions, forms, fields, checks, answer lists and the visit matrix. A visit counted from a date in the data (“30 days after surgery”) or with an “expected only if” condition points to the field of the new study after the copy. Patients and data never move.

While the study has no patients, the eCRF of another study can go in place of the current one: the template's visits and forms are removed. Once there are patients, an eCRF can only be added.

A “From a dictionary” field keeps its dictionary only if you may use it yourself; otherwise it becomes short text and you choose the dictionary again.

eCRF from Excel

eCRF specifications often come in Excel. The Forms sheet has one row per field.

• FORM, VARIABLE, LABEL are needed; FORM_NAME is enough in the first row of a form.
• TYPE: a builder type (number, date, choice…) or a word: “number”, “integer”, “date”, “text”, “drop-down list”… An empty type with answer codes means single choice.
• CODES: 1=Yes; 0=No. REQUIRED: Y or “yes”. MIN, MAX: numbers. SHOW_IF: SMOKER=1, several joined by AND or OR, a field of another form as DM.SEX=F (conditions). Also UNIT, SECTION, HELP.
• The Visits sheet (optional): VISIT, KIND (planned, recurring, event, log), DAY, BEFORE, AFTER, FORMS separated by commas.

A row that cannot be read as it is gets read in the nearest sensible way, with a remark: “row 6: unknown type “magic”, read as short text”. Calculated and dictionary fields are set up in the builder.

The This study’s eCRF in this layout button downloads your eCRF in the same layout: a template, and a way to move an eCRF to another YouEDC installation.

What next

Everything added goes into the eCRF draft and into the audit trail as one record naming the source. A form or field from the library remembers where it came from (youedc:DM@1.0.0), so it can always be compared with the library (see “The link with the library”). Put the new forms into visits in the visit × form matrix. If the eCRF is already released, the sites get the changes in the next version.