CSI stands for the Central Sensitization Inventory, a 25-item patient questionnaire that screens for symptoms linked to central sensitization. Each item is scored 0 (never) to 4 (always), so Part A totals 0 to 100. A score of 40 or higher was the original cutoff for flagging central sensitivity syndromes (81% sensitivity, 75% specificity), and five severity bands run from subclinical (0 to 29) to extreme (60 to 100). It screens; it does not diagnose.
This guide gives clinicians the full picture in one place: how to score the CSI, which cutoff to use and why they differ, what the new 7-item CSI-7 adds, how to document it, and what billing codes do and do not apply. Use the calculator below to score a patient. Every figure is tied to its source in the table of references at the end.
Central Sensitization Inventory at a glance
| Item | Detail |
|---|---|
| Full name | Central Sensitization Inventory (CSI) |
| Developed by | Mayer, Neblett, Cohen and colleagues, Pain Practice, 2012 |
| Format | Self-report. Part A has 25 scored items; Part B lists 10 diagnoses and is not scored |
| Scoring | Each Part A item 0 (never), 1 (rarely), 2 (sometimes), 3 (often), 4 (always). Total 0 to 100 |
| Original cutoff | 40 or higher. Sensitivity 81%, specificity 75%, AUC 0.86 (Neblett et al., 2013; 121 patients and 129 non-patients) |
| Severity bands | Subclinical 0 to 29, mild 30 to 39, moderate 40 to 49, severe 50 to 59, extreme 60 to 100 (Neblett et al., 2017) |
| Newer cutoffs | 30 overall, 33 for women, 25 for men (Schuttert et al., 2023) |
| Reliability (English) | Cronbach's alpha 0.88; test-retest r 0.82. Translations: alpha 0.88 to 0.91, test-retest 0.85 to 0.97 |
| Short form | CSI-7, 7 items, developed with 7,862 participants from 23 countries (Neblett et al., 2026) |
| Access | The CSI user's manual lists many translations and no licensing restrictions; forms are available from the Pride Dallas questionnaire page |
CSI score calculator
CSI score calculator (Part A)
Enter the patient's 25 Part A answers (0 = never, 1 = rarely, 2 = sometimes, 3 = often, 4 = always) from the completed CSI form, or type a total you already scored. The tool adds the score and shows the severity band and the published cutoffs. It is a screening aid, not a diagnosis.
Severity bands: Neblett et al., Pain Practice 2017. Cutoff of 40: Neblett et al., Journal of Pain 2013. Overall, female and male cutoffs: Schuttert et al., Journal of Clinical Medicine 2023. Use the official CSI form for the questions; item numbers here follow that form. This tool does not store any data.
How to administer and score the CSI
- Hand over both parts. The patient completes Part A (25 symptoms) and Part B (diagnosis history) on their own, on paper or in a digital form.
- Score Part A only. Add the 25 answers. Each is 0 to 4, so the total is 0 to 100. Part B is information only.
- Check for blanks. The total needs every item answered. Ask the patient to complete any they skipped.
- Place the total in a severity band and compare it with the cutoff you have chosen (see the next sections).
- Read Part B. Prior diagnoses such as fibromyalgia, irritable bowel syndrome, temporomandibular disorder, migraine, depression or anxiety add context to the score.
- Document and repeat. Record the total, date and band, then repeat at set points such as discharge to see change.
Part A covers symptom areas such as widespread pain, sleep, concentration, digestive and urinary symptoms, sensitivity to light, sound and smells, skin sensitivity, mood, anxiety, headaches, jaw symptoms and fatigue. The form is the patient's own report of how often each occurs. To send the form and track scores across visits, see how PROMs in SPRY automate collection, and browse our full PROMs and functional outcome library.
CSI severity levels and what they suggest
The five severity bands were derived by Neblett and colleagues and confirmed in a separate chronic pain sample. Higher bands were linked to more central sensitivity syndrome diagnoses, anxiety, depression, sleep disturbance, disability and pain intensity. The bands describe symptom burden. The "suggested next steps" column is clinical reasoning, not a validated treatment rule.
| Score | Band | Suggested next steps (clinical reasoning aids) |
|---|---|---|
| 0 to 29 | Subclinical | Symptom burden is low. Continue the usual examination and look for peripheral contributors. |
| 30 to 39 | Mild | Below the original cutoff but near the newer cutoffs. Screen sleep, stress and activity patterns and monitor. |
| 40 to 49 | Moderate | At or above the original cutoff. Consider pain neuroscience education, sleep and stress strategies and graded activity alongside tissue-based care. |
| 50 to 59 | Severe | Central sensitization may be a major driver of symptoms. Consider collaboration with behavioral health or pain psychology and review how much weight passive treatment should carry. |
| 60 to 100 | Extreme | Consider a multidisciplinary pain approach and medical review of the whole presentation. Document the reasoning. |
Pair the CSI with measures of the factors it correlates with. Our guides to the Pain Catastrophizing Scale, Fear Avoidance Beliefs Questionnaire, PHQ-9, GAD-7, FACIT-F and fibromyalgia criteria cover the usual companions.
Which CSI cutoff should you use?
Three published sets of numbers are often mixed together. They answer different questions, so name the one you used in the note.
| Source | Sample and comparison | Numbers | Use it for |
|---|---|---|---|
| Neblett et al., 2013, Journal of Pain | 121 patients (89 with at least one central sensitivity syndrome) versus 129 non-patients | Cutoff 40. Sensitivity 81%, specificity 75%, AUC 0.86 | The original screening cutoff |
| Neblett et al., 2017, Pain Practice | Derived levels, confirmed in a separate chronic pain sample (58% with a central sensitivity syndrome diagnosis) | Subclinical 0 to 29; mild 30 to 39; moderate 40 to 49; severe 50 to 59; extreme 60 to 100 | Describing severity and tracking symptom burden |
| Schuttert et al., 2023, Journal of Clinical Medicine | 1,730 patients with chronic pain and a central sensitivity syndrome versus 250 healthy, pain-free controls; Dutch-speaking | Overall 30 (85% sensitivity, 93% specificity); women 33 (83%, 97%); men 25 (89%, 91%) | A lower, sex-specific screening threshold. The authors note Dutch-only participants and low negative predictive values |
In practice, a score of 40 or higher is the long-standing flag. The 2023 study suggests lower thresholds, and different ones for men and women, but it used healthy controls and one language group, so treat those numbers as extra context rather than a replacement. Whatever you choose, record the number, the source and the date.
CSI-7: the new short form
In 2026, Neblett and colleagues in The Journal of Pain published the CSI-7, a 7-item version built with data from 7,862 participants in 23 countries using Rasch analysis. The authors report internal consistency of 0.85 (alpha and omega) and, for separating fibromyalgia from healthy controls, an area under the curve of 0.98 with 92% sensitivity and 93% specificity. They describe it as an efficient alternative to the 25-item CSI for large-scale use. The abstract does not give severity cutoffs, so keep scoring the full 25-item CSI with the bands above until the CSI-7 scoring guidance is available to you.
How well does the CSI work? Evidence and limits
| Property | What is known | Practical meaning |
|---|---|---|
| Reliability | English version: alpha 0.88 and test-retest r 0.82 (translations 0.88 to 0.91 and 0.85 to 0.97) | Consistent enough for screening and repeat measurement |
| Convergent validity | The user's manual reports strong links with pain intensity, depression, anxiety, sleep disturbance, catastrophizing and disability | A high score often travels with these factors, so measure them too |
| Discrimination | Scores separate fibromyalgia, regional chronic pain and healthy controls | Useful for identifying symptom-heavy presentations |
| Responsiveness | In 763 patients with chronic spinal pain, CSI scores and other psychosocial measures fell by discharge from a functional restoration program (Neblett et al., 2017) | Can be used to track change, but no minimal clinically important difference has been verified |
| Limits | It measures reported symptoms, not nerve function. It does not diagnose. Negative predictive values were low in the 2023 study | Do not rule central sensitization in or out on the CSI alone |
CSI versus quantitative sensory testing
Quantitative sensory testing measures how a patient responds to calibrated stimuli and needs equipment and training. The CSI asks about symptoms and takes no equipment, so it fits outpatient and telehealth intake. They measure different things and are not interchangeable: a patient can have a high CSI and normal sensory testing, or the reverse. Use the CSI to screen and track symptom burden, and add sensory or movement testing when the answer changes the plan.
Documenting the CSI in your notes
Put the facts in the objective section, the reasoning in the assessment, and the decision in the plan. A worked example with made-up numbers:
| SOAP section | Example |
|---|---|
| Objective | CSI Part A completed on 10/04/2026: total 52 of 100, severe band (Neblett 2017); above the cutoff of 40. |
| Assessment | High symptom burden across sleep, fatigue and sensitivity items suggests central sensitization may be contributing to persistent pain. Findings on examination are less severe than the reported pain. |
| Plan | Add pain neuroscience education and graded activity; address sleep and stress; discuss collaboration with the referring provider; repeat the CSI at discharge. |
Our SOAP note and documentation guide and physical therapy EMR show how to build score fields into the chart, and reporting and analytics shows how to review outcomes across patients. Book a demo to see CSI forms sent and scored automatically.
Can you bill for the CSI? What each code does and does not cover
Billing rules depend on the payer, the state and the clinician's license, so check each payer's policy first. The table covers the codes people ask about.
| Code | What it is | Does it fit the CSI? |
|---|---|---|
| CPT 96127 | Brief emotional or behavioral assessment with scoring and documentation, per standardized instrument | Some practices report it for scored screeners. Whether a payer accepts it for a physical therapist, or for the CSI, varies. Confirm in writing before billing. |
| CPT 97129 and 97130 | Therapeutic interventions that focus on cognitive function and compensatory strategies, first 15 minutes and each added 15 minutes | No. These are for cognitive function interventions, not pain neuroscience education. |
| Pain neuroscience education | No separate code | Bill the code that matches the service delivered and documented, according to payer policy. |
| HCPCS G2211 | CMS add-on to office and outpatient or home E/M visits | No. It attaches to E/M visits, which physical therapists do not bill. |
For a broader view of therapy coding, see our physical therapy CPT codes reference sheet, the CPT code library and our advice on reducing claim denials.
Using the CSI to guide care
- Use it at intake and discharge. The 2017 spinal pain study measured CSI at admission and discharge, which is a practical pattern to follow.
- Do not let one number set the plan. A high score tells you to look harder at sleep, stress, beliefs and pacing. It does not mean peripheral treatment will fail.
- Share the result. Explaining that the nervous system can amplify signals helps many patients understand why pain and examination findings may not match.
- Escalate when needed. Severe and extreme scores are reasons to discuss behavioral health or multidisciplinary support with the referring provider.
Frequently asked questions
What does CSI stand for in medical terms?
In pain and rehabilitation, CSI stands for the Central Sensitization Inventory, a 25-item self-report questionnaire that screens for symptoms related to central sensitization.
What CSI score indicates central sensitization?
The original cutoff is 40 or higher (81% sensitivity, 75% specificity). A 2023 study proposed lower cutoffs of 30 overall, 33 for women and 25 for men. These are screening thresholds, not a diagnosis.
What is a normal CSI score?
There is no single normal score. Scores of 0 to 29 fall in the subclinical band. In the 2023 cutoff study, healthy pain-free adults were the comparison group.
What are the CSI severity levels?
Subclinical 0 to 29, mild 30 to 39, moderate 40 to 49, severe 50 to 59 and extreme 60 to 100, from Neblett and colleagues in Pain Practice, 2017.
Is Part B of the CSI scored?
No. Part B lists previous diagnoses of seven central sensitivity syndromes and three related conditions (depression, anxiety or panic attacks, and neck injury). It adds context and is not part of the 0 to 100 score.
Is the CSI diagnostic?
No. It screens for symptoms associated with central sensitization. Diagnosis needs a full history and examination, and the CSI should be read with other findings.
How often should I repeat the CSI?
No verified schedule exists. Studies have measured it at admission and discharge, so a baseline and a discharge score is a reasonable minimum, with extra points if the plan changes.
Is there a minimal clinically important difference for the CSI?
None has been verified in the sources reviewed for this page. Interpret change alongside function, pain and other outcome measures.
Is the CSI free to use?
The CSI user's manual lists translations and states no licensing restrictions. Forms are on the Pride Dallas questionnaire page. Check the terms on the version you use.
What is the CSI-7?
A 7-item short form published in The Journal of Pain in 2026 and developed with 7,862 participants from 23 countries. It had an AUC of 0.98 for separating fibromyalgia from healthy controls.
Can physical therapists bill for the CSI?
It depends on the payer. CPT 96127 is sometimes used for scored screeners, but acceptance varies. CPT 97129 and 97130 are for cognitive function interventions, and G2211 is an E/M add-on, so neither fits.
Sources and further reading
- Neblett R. The Central Sensitization Inventory (CSI): A User's Manual. Journal of Applied Behavioral Research, 2018
- Neblett R, et al. The Central Sensitization Inventory (CSI): establishing clinically significant values. Journal of Pain, 2013
- Neblett R, et al. Establishing clinically relevant severity levels for the CSI. Pain Practice, 2017
- Schuttert I, et al. Validity of the CSI: newly proposed clinically relevant values and associations. Journal of Clinical Medicine, 2023
- Neblett R, et al. Development of a CSI short form (CSI-7) using data from 23 countries. Journal of Pain, 2026
- Neblett R, et al. Use of the CSI as a treatment outcome measure in a functional restoration program. Spine Journal, 2017
- CMS: HCPCS code G2211 frequently asked questions
- Pride Dallas: CSI forms and translations
This page is for clinicians and is informational. The CSI screens for central sensitization symptoms and does not replace clinical judgment. Billing rules vary by payer, so verify them before submitting claims.
Reduce costs and improve your reimbursement rate with a modern, all-in-one clinic management software.
Get a DemoWebinar
From Claims Delays to Clean Approvals: How AI Helps Clinics Win
