Adding Questionnaires

This guide explains how to create Patient Reported Outcome Measures (PROMs) questionnaires in SATHI. Creating questionnaires involves a structured four-step process that must be followed in order.

Accessing the Questionnaire Creation Guide

  1. Navigate to the main menu

  2. Click on “Questionnaires” or “PROM Management”

  3. Select “Questionnaire Creation Guide” to access the comprehensive creation workflow

Understanding the Creation Process

Creating a questionnaire requires completing four key steps in the correct order:

Step 1: Create Construct Scale (Foundation)

Define the latent trait you want to measure. This is the theoretical foundation of your questionnaire.

Step 2: Create Response Scales (Optional, if needed)

Set up Likert or Range scales if your questions will use these response types.

Step 3: Create Items (Questions) (Building blocks)

Create individual questions that will measure your construct.

Step 4: Add to Questionnaire (Final assembly)

Combine your items into a complete questionnaire and arrange them in order.

Important

These steps must be completed in order. You cannot create items without first creating the construct scale, and you cannot create a questionnaire without first creating items.

Understanding Latent Traits

Before creating questionnaires, it’s important to understand what you’re measuring.

What are Latent Traits?

Latent traits are characteristics that cannot be measured directly and require patient-reported outcome measures to quantify them. These are subjective in nature, and PROMs help obtain a quantitative value that can be measured and compared.

Key Characteristics:

  • Cannot be measured with physical instruments

  • Subjective in nature

  • Require multiple questions to capture fully

  • Need patient self-reporting

Examples of Latent Traits:

  • Pain: Intensity, quality, and impact of pain experiences

  • Quality of Life: Overall well-being and life satisfaction

  • Depression: Mood, emotional state, and psychological well-being

  • Anxiety: Worry, fear, and stress levels

  • Physical Function: Ability to perform daily activities

  • Fatigue: Energy levels and tiredness

Comparison:

  • Objective Measurement: Blood pressure can be measured using a sphygmomanometer (direct, quantitative)

  • Latent Trait Measurement: Pain is measured using the Visual Analog Scale (indirect, patient-reported)

Step 1: Creating Construct Scales

A construct scale is designed to measure a specific latent trait using a set of questions (items).

What is a Construct Scale?

A construct scale defines:

  • What latent trait is being measured

  • How to calculate a score from individual item responses

  • Clinical thresholds and reference values

  • Scoring direction (higher/lower is better)

Creating a New Construct Scale

  1. Click “Create Construct Scale” from the creation guide

  2. Fill in the required information:

Basic Information:

  • Name: The name of the construct (e.g., “Physical Function”, “Pain Intensity”)

  • Instrument Name: The PROM instrument this belongs to (e.g., “EORTC QLQ-C30”)

  • Instrument Version: Version number of the instrument

  • Scale Equation: Mathematical formula to calculate the score from items (e.g., (Q1 + Q2 + Q3) / 3)

  • Minimum Items: Minimum number of questions that must be answered to calculate a valid score

Clinical Parameters:

  • Score Direction: Whether higher or lower scores indicate better outcomes

    • Higher is Better (e.g., quality of life)

    • Lower is Better (e.g., symptom severity)

    • Middle is Better (e.g., some psychological measures)

  • Threshold Score: The clinical significance threshold (scores beyond this indicate clinical concern)

  • Minimum Clinical Important Difference (MCID): The smallest change in score that is clinically meaningful

  • Normative Score Mean: Average score in a healthy population

  • Normative Score Standard Deviation: Variation in the healthy population

Examples:

  • Physical Function (from EORTC QLQ-C30)

  • Pain Intensity (from Brief Pain Inventory)

  • Depression Severity (from PHQ-9)

  1. Click “Save” to create the construct scale

Step 2: Creating Response Scales

Response scales define how patients will answer questions. There are two main types:

Likert Scales

Likert scales are the most common response type, consisting of ordered response options with numerical values and text descriptions.

When to Use:

  • Structured agreement scales (e.g., Strongly Disagree to Strongly Agree)

  • Frequency scales (e.g., Never, Rarely, Sometimes, Often, Always)

  • Severity scales (e.g., None, Mild, Moderate, Severe)

Creating a Likert Scale:

  1. Click “Create Likert Scale”

  2. Enter a Scale Name (descriptive label for identification)

  3. Add response options:

    • Option Order: Display order (1, 2, 3, etc.)

    • Option Value: Numerical value for scoring

    • Option Text: Text description shown to patients

    • Emoji (optional): Visual indicator (use cautiously as meaning may vary)

    • Option Media (optional): Image, audio, or video to explain the option

  4. Add multiple options as needed

  5. To delete an option, check the “Delete” checkbox and save

Common Examples:

  • Severity: None (0), Mild (1), Moderate (2), Severe (3)

  • Frequency: Never (0), Rarely (1), Sometimes (2), Often (3), Always (4)

  • Agreement: Strongly Disagree (1) to Strongly Agree (5)

Note

Likert scales can be reused across multiple items and questionnaires if they share the same response structure.

Range Scales

Range scales present a numeric ruler, typically used for visual analog scales like pain ratings.

When to Use:

  • Pain scales (0-10)

  • Quality of life ratings (0-100)

  • Any continuous numeric measurement

Creating a Range Scale:

  1. Click “Create Range Scale”

  2. Enter the scale details:

    • Scale Name: Descriptive label

    • Minimum Value: Starting number (e.g., 0)

    • Maximum Value: Ending number (e.g., 10)

    • Increment: Step size between numbers (e.g., 1)

    • Minimum Value Text: Label for minimum (e.g., “No Pain”)

    • Maximum Value Text: Label for maximum (e.g., “Worst Pain”)

Common Examples:

  • Pain Scale: 0 (No Pain) to 10 (Worst Pain), increment 1

  • Quality of Life: 0 (Worst) to 100 (Best), increment 10

Step 3: Creating Items (Questions)

Items are the individual questions that patients will answer. Each item belongs to a construct scale.

Creating a New Item

  1. Click “Create New Item (Question)”

  2. Fill in the item details:

Basic Information:

  • Construct Scale: Select which construct this item measures

  • Abbreviated Item ID: Short code for data export (e.g., “PF1”, “Q1”) - use official abbreviations from scoring manuals when available

  • Item Name: The complete question text

    Important

    Include any question root text. For example, if the questionnaire has “In the past 7 days:” as a common prefix, include it in each question for accuracy.

  • Item Media (optional): Upload image, audio, or video to help explain the question or assist patients with limited literacy

Response Type:

Choose how patients will answer this question:

  • Likert: Select from a pre-defined Likert scale (requires Step 2)

  • Range: Use a numeric slider (requires Step 2)

  • Text: Free-form text entry

  • Numeric: Numeric value input

  • Media: Audio or video recording by the patient

Clinical Scoring Parameters (optional but recommended):

  • Score Direction: Higher/Lower/Middle is better

  • Threshold Score: Clinical significance threshold for this specific item

  • Minimum Clinical Important Difference: Meaningful change for this item

  • Normative Score Mean: Expected value in healthy population

  • Normative Score Standard Deviation: Variation in healthy population

  1. Click “Save” to create the item

Tip

Multi-Language Support: After creating an item, you can add translations in different languages. The system maintains scoring consistency across all language versions.

Managing Items

  • View All Items: Click “Manage Items” to see all created questions

  • Edit Items: Click on any item to modify it

  • Reuse Items: Items can be used in multiple questionnaires

  • Question Bank: Build a library of validated questions for reuse

Step 4: Creating the Questionnaire

A questionnaire (also called an instrument) is a collection of items presented to patients in a specific order.

Creating a New Questionnaire

  1. Click “Create Questionnaire”

  2. Enter questionnaire details:

Basic Settings:

  • Name: The questionnaire name (e.g., “EORTC QLQ-C30”, “Brief Pain Inventory”)

  • Description: Explain what the questionnaire measures, how long it takes, and why it’s being used

    Note

    This description is visible to patients, so use clear, easy-to-understand language.

  • Questionnaire Answer Interval: Minimum time between submissions (prevents patients from answering too frequently)

  • Questionnaire Order: Display order number (controls the sequence patients see questionnaires)

  • Questionnaire Redirect: Optionally select another questionnaire to automatically redirect to after completion

Adding Items to the Questionnaire

  1. Select Items: Check the box next to each item you want to include

    • Selected items appear in the side panel on the right

    • Use the construct scale filter to narrow down items

    • Use the search box to find specific items by text

  2. Arrange Order: Drag and drop items in the side panel to change their order

    Important

    Match the official order from the validated PROM instrument when possible.

  3. Remove Items: Click the cross (X) icon to deselect an item

  4. Click “Save” to create the questionnaire

Managing Questionnaires

  • View All Questionnaires: Click “Manage Questionnaires”

  • Edit Questionnaires: Modify items, order, or settings

  • Assign to Patients: After creation, questionnaires can be assigned to specific patients

Advanced Features

Multi-Language Translations

The system supports comprehensive multi-language functionality:

Translatable Elements:

  • Questionnaire titles and descriptions

  • Item text and media

  • Likert scale response option text and media

  • Range scale minimum and maximum value text

How to Add Translations:

  1. Navigate to “Manage Translations” from the main menu

  2. Select the language you want to translate to

  3. Enter translations for each element

  4. The system maintains scoring consistency across all languages

Note

Available languages are configured by the system administrator.

Equation Editor

Advanced scoring capabilities using mathematical equations to calculate construct scores.

Supported Operations:

  • Basic operators: +, -, *, /, ^

  • Functions: sqrt(), abs()

  • Item references: Q1, Q2, Q3, etc.

  • Parentheses for order of operations

Example Equations:

  • Simple sum: Q1 + Q2 + Q3

  • Average score: (Q1 + Q2 + Q3) / 3

  • Transformed score: 100 - ((Q1 + Q2) * 25 / 2)

  • EORTC scoring: (1 - (Q1 + Q2 + Q3) / 12) * 100

The equation editor includes real-time validation to ensure your formula is correct.

Conditional Logic

After creating a questionnaire, you can define rules to control question display based on patient responses.

Use Cases:

  • Show follow-up questions only if a specific answer is given

  • Skip irrelevant questions based on previous responses

  • Create branching logic for complex questionnaires

How to Set Up:

  1. Create the questionnaire first

  2. Navigate to the questionnaire’s rule settings

  3. Define conditions (e.g., “If Q1 = ‘Yes’, show Q2”)

  4. Test the logic before assigning to patients

Composite Construct Scores

Composite scores combine multiple individual construct scores into a single higher-level measurement.

What is a Composite Construct Score?

A composite construct score is calculated from one or more individual construct scores, providing a broader view of patient functioning.

Example:

The FACT TOI (Trial Outcome Index) score is calculated as: FACT-G (General) + FACT-B (Breast Cancer) + BCS (Breast Cancer Subscale)

Creating a Composite Score

  1. Click “Create Composite Construct Score”

  2. Select the Scoring Type:

    • Sum: Adds all selected construct scores together

    • Average: Calculates the mean of all scores

    • Median: Finds the middle value

    • Mode: Identifies the most frequent score

    • Minimum: Takes the lowest value

    • Maximum: Takes the highest value

  3. Select which Construct Scales to include in the calculation

  4. Click “Save”

Common Use Cases:

  • Creating overall quality of life scores from multiple domains

  • Combining physical and emotional wellbeing measures

  • Generating summary scores for clinical trial endpoints

  • Creating simplified reporting metrics from complex questionnaires

Warning

Only create composite scores when there is theoretical and statistical justification. Refer to the original questionnaire validation papers for guidance.

Best Practices

When Creating Construct Scales:

  • Use validated constructs from published PROM instruments

  • Include all available clinical parameters (thresholds, MCID, normative scores)

  • Verify the scoring equation matches the official scoring manual

  • Document the source of your construct definition

When Creating Items:

  • Use exact wording from validated questionnaires

  • Include question roots and timeframes in the item text

  • Use official abbreviated IDs from scoring manuals

  • Add media to assist patients with limited literacy

  • Test questions with a small group before full deployment

When Creating Questionnaires:

  • Match the official item order from validated instruments

  • Write clear, patient-friendly descriptions

  • Set appropriate answer intervals to prevent survey fatigue

  • Use questionnaire redirects to create smooth workflows

  • Test the complete questionnaire before assigning to patients

For Multi-Language Support:

  • Use professionally validated translations when available

  • Have translations reviewed by native speakers

  • Test questionnaires in each language

  • Ensure cultural appropriateness of all content

Troubleshooting Common Issues

Cannot create items without construct scale

You must create the construct scale first (Step 1) before creating items (Step 3).

Likert/Range scale not appearing in item creation

Create the response scale first (Step 2) before creating items that use it.

Equation validation errors

  • Check that all item references (Q1, Q2, etc.) are correct

  • Ensure parentheses are balanced

  • Verify mathematical operators are valid

  • Test with sample values

Items not appearing when creating questionnaire

  • Verify items were saved successfully

  • Check that items belong to the expected construct

  • Use the search and filter functions to locate items

Translations not displaying

  • Ensure translations are saved for the active language

  • Check that the language is enabled in system settings

  • Verify all required fields are translated

Next Steps

  • Review Reviewing Results to understand how scores are displayed

  • Review Getting Started for general navigation tips

  • Explore the questionnaire guidance page within the system for detailed examples

Additional Resources

Within the System:

  • Manage Construct Scales: View and edit all construct scales

  • Manage Likert Scales: View and edit all Likert response scales

  • Manage Range Scales: View and edit all range response scales

  • Manage Items: View and edit all questions

  • Manage Questionnaires: View and edit all questionnaires

  • Manage Translations: Add and edit multi-language translations

  • Composite Scoring: Create and manage composite construct scores

External Resources:

  • Refer to original PROM validation papers for construct definitions

  • Consult official scoring manuals for equation verification

  • Review published translations for language accuracy