In depth
Rheumatoid arthritis and blood tests
About a third of people with rheumatoid arthritis test negative for the antibody. Normal blood markers at diagnosis are common as well, and both of those facts get missed more often than they should. So this page goes through what each test can settle and what it can't settle.
What the research found.
One review pooled 37 studies and 14,949 patients, and anti-CCP caught 67 percent of the cases. It cleared 95 percent of the people who didn't have the disease. Rheumatoid factor caught 69 percent of the cases and cleared only 85 percent, which is what puts anti-CCP ahead of it.
A larger review pooled 138 studies, and anti-CCP caught 67 percent of the cases there too. It cleared 96 percent of the people who didn't have the disease. Early disease is harder than that, and there the cohort studies put the catch rate at just 54 percent of the cases.
Two cohorts were followed in usual care, and CRP was under 10 mg/L at the first measurement in 44 percent of one. It was under 10 in 58 percent of the other. ESR and CRP were both normal in 33 and 42 percent, and all three of ESR, CRP, and rheumatoid factor were normal in about 15 percent.
Some trials treated to a set target instead of by impression, and the target was a score built partly from a blood marker. Those trials raised remission by a great deal. TICORA reported 65 percent against 16 percent on usual care, and FIN-RACo reported 37 percent against 18 percent.
Hock and colleagues, SN Comprehensive Clinical Medicine, 2021
People with rheumatoid arthritis had lower vitamin D than healthy controls, at a pooled difference of 16.52 nmol/L. The level barely tracked how active somebody's disease was. The correlation came out at negative 0.13, which is close enough to zero to mean no useful link between the two.
What the number means.
| Test | What it measures | What a positive or raised result means | What a normal result rules out |
|---|---|---|---|
| Anti-CCP | Antibodies to citrullinated proteins | Strong evidence for rheumatoid arthritis. It clears 95 to 96 percent of people who don't have it | Little. About a third with the disease test negative, and over 40 percent in early cohort studies |
| Rheumatoid factor | An antibody aimed at other antibodies | Supports the diagnosis. It clears 85 percent of people who don't have it | Little, and this is the weaker of the two tests |
| CRP | A protein the liver makes when there's inflammation | Inflammation somewhere, and not always in the joints | Nothing. It was under 10 mg/L in 44 to 58 percent at the first measurement |
| ESR | How fast red cells settle, which measures inflammation indirectly | Inflammation. It also rises with age, low blood counts, and pregnancy | Nothing on its own, and the age-adjusted limit differs from the printed one |
| Vitamin D | Stored vitamin D, as 25-hydroxyvitamin D | A deficiency worth correcting for bone health | Nothing about disease activity. The correlation with DAS28, the score built from a 28-joint count, a blood marker, and your own health rating, is about negative 0.13 |
| DAS28 | A composite score, and not a blood test | Higher disease activity, which drives treatment choices | It falls with treatment, and it's what treat-to-target steers by |
A reference range is not a target. Where a range on this page differs from the one your laboratory prints, the difference is explained above and sourced below. Where no trial has tested a target, this page says so rather than offering one.
What each test is for
The blood tests in rheumatoid arthritis do two different jobs, and keeping them apart is the first useful thing to know. Anti-CCP and rheumatoid factor are diagnostic tests, used to work out whether you have the disease. CRP and ESR are activity measures, used to work out how much inflammation is present right now rather than whether the diagnosis is right.
Confusing those two jobs produces most of the trouble people have with their results. An antibody test doesn't tell you whether your disease is active today, and it doesn't need repeating once it's known. An inflammatory marker doesn't tell you whether you have rheumatoid arthritis, and a normal one at diagnosis is common enough to be unremarkable.
The third thing on a rheumatology form isn't a blood test at all. DAS28 is a composite score that combines a joint count, a global assessment, and a blood marker into one number. That number is what modern treatment is steered by, which is why it gets calculated at every visit.
Anti-CCP, and why a negative result settles so little
Anti-CCP is the most useful single blood test for rheumatoid arthritis, and its value appears in the positive result. One review pooled 37 studies and 14,949 patients, and anti-CCP cleared 95 percent of the people who didn't have the disease. That's the clear rate, and it's what makes a positive result mean something. Rheumatoid factor cleared only 85 percent, and a larger review of 138 studies put the anti-CCP clear rate at 96 percent. So a positive anti-CCP in somebody with swollen joints is strong evidence.
The other half of the picture is the catch rate, which is the share of people with the disease the test finds. It's the weak side here, and weaker than the headline suggests. The pooled figure is 67 percent, so about a third of people with the disease test negative. Early disease is when the test is wanted most, and there the cohort studies gave a catch rate of 54 percent, which means nearly half of early cases test negative.
Study design accounts for a good deal of the spread in those numbers. Anti-CCP caught 68 percent of the cases in case-control studies, 69 percent in studies that looked at people at one moment, and 60 percent in cohort studies. Cohort studies most resemble how the test gets used in practice, which is why that 60 percent is the figure to hold. Across all 138 studies the catch rate ranged from 12 to 93 percent and the clear rate from 63 to 100 percent, so any single pooled number is smoothing over a lot.
Inflammatory markers, and how often they are normal
One fact about CRP and ESR counts for more than any other here, and it's how often they come back normal. Two cohorts were followed in usual care, and in Finland CRP was under 10 mg/L at the first measurement in 44 percent of patients. In the United States it was 58 percent. Both ESR and CRP were normal in 33 and 42 percent, and all three of ESR, CRP, and rheumatoid factor were normal in about 15 percent.
Only 28 percent and 23 percent of those two cohorts had all three tests abnormal. That's the opposite of the impression most people are left with, which is that inflammatory arthritis always appears in the blood. Often it doesn't, so a swollen joint on examination outranks a normal number every time.
ESR needs one more piece of context, because it rises with age on its own and apart from any disease. So there's a usual age rule, which for men is to take your age and halve it, and for women to add ten first and then halve it. Applying that rule changed the picture in one study a great deal. Unadjusted, 85 percent of late-onset patients were above the limit, and so were 71 percent of younger-onset patients, while after adjustment the two figures were 65 and 68 percent, which removed the apparent age difference entirely.
Why the score counts more than any single result
Treating to a target beats treating to clinical impression, and that's one of the better-established ideas in rheumatology. The trials behind it used composite scores, and each score contained a blood marker of inflammation. TICORA reported remission in 65 percent on intensive protocol treatment against 16 percent on usual care, CAMERA reported 50 against 37, a DAS28-driven strategy reported 38 against 21, and FIN-RACo reported 37 against 18.
So your appointment involves counting joints and calculating a number rather than asking how you feel and adjusting by judgment. That number exists for one reason and one reason only. Strategies built around a score produced better remission rates than strategies built around impression, which makes it a tool for a decision rather than a grade.
That also explains something about the single marker, which gets more attention than its accuracy alone would justify. Inside a composite score, a marker that's often normal still adds useful information, because the joint count and the global assessment are counted alongside it. On its own the marker is a good deal weaker than it looks in that context.
Vitamin D, which is worth correcting and won't move your disease
People with rheumatoid arthritis do have lower vitamin D than healthy controls, and the pooled difference was 16.52 nmol/L. That's a real difference, and what doesn't follow is the conclusion usually drawn from it. The level barely tracks how active the disease is, at a correlation of negative 0.13 against DAS28 and negative 0.12 against CRP. One pooled odds ratio for the relationship came out at 1.06, with the true value ranging from 0.8 to 1.4, and that range crosses one.
The supplement trials line up with those figures. A meta-analysis found supplements raised vitamin D in the blood and improved pain scores and CRP significantly. DAS28 was statistically unchanged at a standardized mean difference of negative 0.09, and ESR didn't reach significance either, so the thing that moved was the level plus two softer measures.
Correcting a deficiency is worth doing for bone health, and that goes double for anyone who has taken corticosteroids. Expecting it to change your disease activity goes beyond what these studies show. Knowing that in advance saves a good deal of disappointment at the next blood test.
Common misconceptions.
Myth. A negative anti-CCP means I don't have rheumatoid arthritis.
Reality. Around a third of people with the disease test negative, and early disease is worse than that. There the cohort studies say the test caught 54 percent of the cases, so nearly half of early cases test negative. Seronegative rheumatoid arthritis is a real diagnosis, made on the joint exam and the history. A negative antibody test lowers the odds, and it doesn't close the question.
Myth. Normal CRP and ESR mean my arthritis isn't active.
Reality. CRP was under 10 mg/L at the first measurement in 44 percent of one cohort and 58 percent of another. All three tests were normal in about 15 percent of people who had the disease, and those three are ESR, CRP, and rheumatoid factor. The markers describe the moment blood was taken, so swollen joints on exam outrank a normal number.
Myth. The published accuracy figures tell me how the test performs on me.
Reality. They're better than clinic performance, and the reason is how the studies were built. Anti-CCP caught 68 percent of the cases in case-control studies and 60 percent in cohort studies, and cohort studies are the ones that resemble real practice. Across 138 studies the spread was wide, because the test caught between 12 and 93 percent of the cases and cleared between 63 and 100 percent of the people who didn't have it. So a single pooled figure hides a great deal.
Myth. My vitamin D level explains my disease activity.
Reality. People with rheumatoid arthritis do have less vitamin D than controls, at a pooled difference of 16.52 nmol/L. The level barely tracks how active the disease is, scoring negative 0.13 against the DAS28, the score built from a 28-joint count, a blood marker, and a health rating, and negative 0.12 against CRP, where zero would mean no link at all. Supplements raised the level and improved pain scores and CRP, and DAS28 barely moved.
Questions patients ask.
What does a positive anti-CCP mean?
It's the strongest single blood result pointing at rheumatoid arthritis, and the reason is how well it clears people who don't have the disease. Anti-CCP cleared 95 to 96 percent of them, and rheumatoid factor cleared 85 percent. That difference is why anti-CCP counts for more. A positive result in somebody with swollen small joints is meaningful, and a positive result in somebody with no joint symptoms means a good deal less.
What does a negative anti-CCP mean?
Much less than people assume, because anti-CCP caught 67 percent of the cases when the studies were pooled. So about a third of people with the disease test negative, and in early disease the cohort studies put it at 54 percent. Seronegative rheumatoid arthritis is diagnosed on exam and history. It isn't a milder or lesser form of the disease, and waiting for a positive antibody before treating costs joints.
Why are the accuracy numbers different in different places?
Because study design changes them a great deal. In the largest review, anti-CCP caught 68 percent of the cases in case-control studies, 69 percent in studies that looked at people at one moment, and 60 percent in cohort studies. Cohort studies most resemble how a test is used in practice. Across all 138 studies the spread was wider still, at 12 to 93 percent of the cases, so any single quoted figure is a summary of something very varied.
What's the difference between CRP and ESR?
CRP measures a protein the liver makes when there's inflammation, and it moves within a day or two. ESR measures how fast red cells settle, which is affected by inflammation and also by age, low blood counts, and pregnancy, and it moves slowly. So the two disagree with each other often. The usual age rule for ESR has two forms, which are to take your age and halve it for men, and to add ten first and then halve it for women.
My inflammatory markers are normal. Should I be reassured?
Reassured about this moment, and not about the diagnosis. Two cohorts were followed in usual care, and CRP was under 10 mg/L at the first measurement in 44 and 58 percent of patients. Both ESR and CRP were normal in 33 and 42 percent, and all three of ESR, CRP, and rheumatoid factor were normal in about 15 percent. So exam findings count for more than a normal number.
Why does my rheumatologist keep calculating a score?
Because treating to a target beats treating to impression, and the evidence here is unusually clear. Some trials followed a set protocol and aimed at a score built partly from a blood marker, and those trials beat usual care on remission every time. TICORA reported 65 percent against 16 percent, CAMERA reported 50 against 37, and FIN-RACo reported 37 against 18. So the score isn't paperwork, and it's what those trials steered by.
Should I supplement vitamin D for my rheumatoid arthritis?
For bone health, yes, and don't expect it to move the disease itself. Supplements raised vitamin D in the blood and improved pain scores and CRP too, while the DAS28, the score built from a 28-joint count, a blood marker, and a health rating, barely moved at all. That change came out at negative 0.09, too small to be sure of. The level barely tracks how active the disease is, at a link of around negative 0.13 against DAS28.
References.
- Nishimura K; Sugiyama D; Kogata Y et al. Meta-analysis: diagnostic accuracy of anti-cyclic citrullinated peptide antibody and rheumatoid factor for rheumatoid arthritis. Annals of internal medicine. 2007;146:797-808. 10.7326/0003-4819-146-11-200706050-00008Diagnostic accuracy meta-analysis
- Whiting PF; Smidt N; Sterne JA et al. Systematic review: accuracy of anti-citrullinated Peptide antibodies for diagnosing rheumatoid arthritis. Annals of internal medicine. 2010;152:456-64; W155-66. 10.7326/0003-4819-152-7-201004060-00010Diagnostic accuracy systematic review and meta-analysis of 151 studies
- Sokka T; Pincus T. Erythrocyte sedimentation rate, C-reactive protein, or rheumatoid factor are normal at presentation in 35%-45% of patients with rheumatoid arthritis seen between 1980 and 2004: analyses from Finland and the United States. The Journal of rheumatology. 2009;36:1387-90. 10.3899/jrheum.080770Retrospective analysis of two consecutive usual-care RA databases
- Ranganath VK; Elashoff DA; Khanna D et al. Age adjustment corrects for apparent differences in erythrocyte sedimentation rate and C-reactive protein values at the onset of seropositive rheumatoid arthritis in younger and older patients. The Journal of rheumatology. 2005;32:1040-2. PMID 15940764Observational cohort analysis of 263 patients with early seropositive RA enrolled within 14 months of symptom onset
- Hock E; Martyn-St James M; Wailoo A et al. Treat-to-Target Strategies in Rheumatoid Arthritis: a Systematic Review and Cost-Effectiveness Analysis. SN Comprehensive Clinical Medicine. 2021;3:838-854. 10.1007/s42399-021-00727-4Systematic review of randomised controlled trials
- Lin J; Liu J; Davies M et al. Serum Vitamin D Level and Rheumatoid Arthritis Disease Activity: Review and Meta-Analysis. PLOS ONE. 2016;11:e0146351. 10.1371/journal.pone.0146351Meta-analysis of observational studies
- Khatirnamani Z; Hajian-Tilaki K; Heidari B et al. The Effect of Vitamin D Supplementation on Clinical and Laboratory Biomarkers in Patients with Rheumatoid Arthritis: A Systematic Review and Meta-Analysis of Randomised Controlled Trials. Mediterranean Journal of Rheumatology. 2025;36:428. 10.31138/mjr.090725.larSystematic review and meta-analysis of 11 randomised controlled trials
This page gathers the published research on this subject into one place. The studies behind it were published between 1989 and 2026, and every figure links to the paper it came from. Those studies were peer reviewed. This summary of them was not. Dr. Sarah Luebker is reviewing these pages one at a time and has not reached this one yet, so it carries no medical review date and nothing here is her opinion or her advice to you. Each page gets updated as she reaches it. It is here in the meantime because the science is worth having in one organized place that is easy to find and easy to read. Talk to your own clinician before acting on any of it.
Across the site