Home / Resources / Psychological Measures / BPRS
BPRS:
Scoring and Interpretation
The Brief Psychiatric Rating Scale (BPRS) is a clinician-rated instrument for the severity of psychiatric symptoms across a set of dimensions, scored after an interview and observation. BPRS scoring only has a meaning once the version and rating anchors are fixed, since the original form had 16 dimensions and the widely used form has 18, so this guide covers version control, the 18 to 126 total, rater agreement, SPSS and R code, and reporting wording.
What the BPRS Measures
Overall and Gorham (1962) introduced the scale as a brief, repeatable way for clinicians to rate symptom change in psychiatric patients, using 16 symptom constructs. The dimensions cover thought disturbance, hallucinatory behavior, affective symptoms, agitation and behavior observed during an interview, and each rating integrates what the patient reports with what the clinician sees.
The form most services now use has 18 dimensions, because later work added two to the original 16, and longer expanded forms also exist. Reporting only the name BPRS therefore leaves the version, the anchors and the score range undefined. The scale is a severity rating completed by a trained clinician. It is not a self-report checklist and not a diagnostic interview.
Structure of the BPRS
- Items: eighteen clinician-rated symptom dimensions in the widely used form. The original 1962 publication described sixteen.
- Ratings: in the 18-item form each dimension is rated from 1 (not present) to 7 (extremely severe). Other versions use different anchors, including 0 to 6 ranges, so confirm the manual for the form in use.
- Not assessed: some versions use a separate code for a dimension that was not assessed. That code is missing data, never a low symptom rating.
- Subscales: the standard outcome is a single total. Factor structures differ by population and version, so a universal subscale key should not be assumed.
- Administration: ratings follow an interview and observation, and the assessment period should be stated in the protocol.
- Translations: use a validated version for the intended population and name the language and its source. If no validated translation exists, say so and describe how the version was adapted.
The instrument itself is available from Overall and Gorham (1962) and from clinician training materials for the version in use.
How to Score the BPRS
- Confirm the version and its exact rating anchors. The example applies only to the complete 18-item form with observed ratings from 1 to 7.
- Check each variable, bprs1 to bprs18, for values outside 1 to 7, and treat any not-assessed code as missing.
- Sum the 18 ratings to obtain bprs_total, which ranges from 18 to 126.
- Apply a missing-item rule set by the protocol. No developer-endorsed prorating rule was verified, so the example requires all 18 ratings. When comparing waves or studies, verify that version, anchors, training and rater procedures were identical.
* Illustrative syntax: bprs1 to bprs18 hold ratings coded 1 to 7 for the 18-item form.
COMPUTE bprs_valid = (NVALID(bprs1 TO bprs18) = 18).
COUNT bprs_outside = bprs1 TO bprs18 (LO THRU 0, 8 THRU HI).
IF (bprs_outside > 0) bprs_valid = 0.
IF (bprs_valid = 1) bprs_total = SUM(bprs1 TO bprs18).
IF (bprs_valid = 0) bprs_total = $SYSMIS.
EXECUTE.
# data: one row per rating occasion; items bprs1 to bprs18 coded 1 to 7.
items <- paste0("bprs", 1:18)
score_bprs <- function(x) {
if (anyNA(x) || any(!x %in% 1:7)) return(NA_real_)
sum(x)
}
data$bprs_total <- apply(data[items], 1, score_bprs)
Interpreting BPRS Scores
| Score band | Interpretation |
|---|---|
| 18 | Every dimension rated not present |
| 19 to 126 | Increasing summed severity across the 18 dimensions |
| Universal cutoff | None verified |
Source: Overall and Gorham (1962) for the scale; the 18 to 126 range follows from 18 dimensions rated 1 to 7.
Higher totals indicate greater combined symptom severity within the same version and administration scheme. The sum does not show which symptoms drive a result, and it does not define diagnostic categories. Mappings from totals to global impression labels should be used only where a cited study supports them for the version and population in question.
Totals from different BPRS versions are not numerically equivalent and should not be pooled as though they were. Clinical concern about deterioration or risk calls for clinician-led assessment whatever the total.
Psychometric Properties
| Property | Published finding | Source |
|---|---|---|
| Original development | Sixteen symptom dimensions for brief rating of psychiatric patients | Overall & Gorham (1962) |
| Purpose | Designed for repeated assessment of symptom change | Overall & Gorham (1962) |
| Rater agreement | Depends on interview procedure, anchors and training; compute and report your own coefficient | Study-specific |
| Factor structure | Differs by population and version; no single loading pattern applies | Population-specific validation |
| Diagnostic accuracy | Not a screening threshold; sensitivity and specificity are not automatically relevant to a summed severity score | Not established |
These are source-specific descriptions and cannot be assumed to hold as reliability or validity estimates in a new sample.
Using the BPRS in a Study
State whether the endpoint is the total, a version-appropriate symptom factor or the change from baseline. For repeated assessments, a mixed-effects model accommodates clustering within patients and unequal intervals, and the protocol should prespecify missing-data handling, raters and covariates. Report between-group differences with confidence intervals and not only within-group significance.
Rater drift and changes in staff can produce apparent trends, so calibration, blinded ratings where feasible and a reliability check across raters are worth planning. Cronbach’s alpha cannot show that clinicians agree, and an intraclass correlation from a design with more than one rater is usually more informative. In multi-site data, verify that every site used the same items and anchors before pooling.
Reporting the BPRS
Psychiatric symptom severity was assessed with the 18-item Brief Psychiatric Rating Scale (BPRS; Overall & Gorham, 1962), with each dimension rated by a trained [clinician role] from 1 (not present) to 7 (extremely severe). The 18 ratings were summed (range = 18 to 126), and higher totals indicate greater severity. Forms with missing ratings were handled by [method] as prespecified. Inter-rater agreement was evaluated using [method and estimate]. The primary endpoint was [total, change or time point].
References
Overall, J. E., & Gorham, D. R. (1962). The Brief Psychiatric Rating Scale. Psychological Reports, 10(3), 799-812. https://doi.org/10.2466/pr0.1962.10.3.799
Frequently Asked Questions
What is the BPRS total score range?
For a complete 18-item form rated 1 to 7 it is 18 to 126. The original 16-item form, or other anchors, produces a different range.
Is a rating of zero the same as no symptoms?
Not in the 18-item form, where 1 means not present. If a version uses a code for not assessed, treat it as missing.
Can I compare BPRS and PANSS totals directly?
No. The instruments have different items and ranges, and any crosswalk would need its own validation evidence.
Should I report reliability?
Yes. Where more than one rater scores patients, report the agreement method, the rater training and a sample-specific estimate.
NEXT STEP
Need BPRS Data Scored
or Checked?
OnlineSPSS is a statistical consultancy for researchers. Share your design and dataset, and a statistically trained Research Coordinator reviews it. The scope and one fixed project fee are confirmed before work begins.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014