HoNOS: Scoring, Subscales and Psychometrics

The Health of the Nation Outcome Scales (HoNOS) is a 12-item clinician-rated instrument developed by the Research Unit of the Royal College of Psychiatrists (RCPsych; Wing et al., 1998) to assess health and social functioning in working-age adults (aged 18 to 65) with mental illness. Mandated across healthcare systems in the United Kingdom, Australia, and New Zealand, HoNOS provides a standardized metric for tracking recovery, evaluating service outcomes, and assessing clinical complexity across behavioral, impairment, symptom, and social domains.

Definition in One Paragraph

The Health of the Nation Outcome Scales (HoNOS) is a standardized clinician-rated outcome measure that quantifies mental health and social problem severity in working-age adults. Administered by trained mental health professionals following routine assessment, it evaluates 12 clinical indicators on a 5-point severity continuum from 0 (no problem) to 4 (severe to very severe problem), yielding four domain subscales and an aggregate total score from 0 to 48. Rather than diagnosing specific psychiatric disorders, HoNOS measures biopsychosocial difficulties across a specified rating window—typically the preceding two weeks in ambulatory care or the preceding three days upon inpatient discharge—enabling clinical teams and researchers to track therapeutic change over time.

Formula and Scoring Structure with Every Symbol Explained

HoNOS contains 12 items ($X_1$ through $X_{12}$) scored on an ordinal integer scale:

0: No problem during the rating window.
1: Minor problem requiring no formal clinical action.
2: Mild problem, but definitely present.
3: Moderately severe problem.
4: Severe to very severe problem.
9: Not known or missing (must be resolved before summing).

Items map to four subscales formalized by Te Pou and AMHOCN:

Behavioural Problems ($S_{text{behaviour}}$): Items 1 to 3 (overactive/aggressive behaviour, self-injury, substance misuse).

$$S_{text{behaviour}} = X_1 + X_2 + X_3 quad (0 le S_{text{behaviour}} le 12)$$
Impairment Problems ($S_{text{impairment}}$): Items 4 and 5 (cognitive deficits, physical illness/disability).

$$S_{text{impairment}} = X_4 + X_5 quad (0 le S_{text{impairment}} le 8)$$
Symptom Problems ($S_{text{symptom}}$): Items 6 to 8 (hallucinations/delusions, depressive symptoms, other emotional symptoms).

$$S_{text{symptom}} = X_6 + X_7 + X_8 quad (0 le S_{text{symptom}} le 12)$$
Social Problems ($S_{text{social}}$): Items 9 to 12 (relationships, daily living activities, living conditions, occupation/activities).

$$S_{text{social}} = X_9 + X_{10} + X_{11} + X_{12} quad (0 le S_{text{social}} le 16)$$

The total score ($T_{text{HoNOS}}$) is the sum across all 12 items:

$$T_{text{HoNOS}} = sum_{i=1}^{12} X_i = S_{text{behaviour}} + S_{text{impairment}} + S_{text{symptom}} + S_{text{social}} quad (0 le T_{text{HoNOS}} le 48)$$

Two essential scoring rules govern item coding under RCPsych protocols:

Severity Hierarchy: Score the peak severity observed in the rating window, rather than an average over time.
Mutual Exclusivity: Do not score the same problem on multiple items. Prior items take precedence (e.g., aggression secondary to delusions is rated under Item 1, while hallucinations/delusions are rated under Item 6). Item 10 (activities of daily living) serves as the sole overall functioning exception.

Worked Example with Step-by-Step Arithmetic

An adult outpatient is evaluated at intake and following a 12-week intervention. Clinicians recorded the following ratings across all 12 items:

Scale Item	Clinical Focus	Baseline ($X_i$)	Follow-up ($X_i’$)
Item 1	Overactive, aggressive, or agitated behaviour	1	0
Item 2	Non-accidental self-injury	0	0
Item 3	Problem drinking or drug-taking	2	1
Item 4	Cognitive problems	0	0
Item 5	Physical illness or disability problems	1	1
Item 6	Problems with hallucinations and delusions	0	0
Item 7	Problems with depressive symptoms	3	1
Item 8	Other mental and behavioural problems (anxiety)	2	1
Item 9	Problems with relationships	2	1
Item 10	Problems with activities of daily living (ADL)	1	0
Item 11	Problems with living conditions	0	0
Item 12	Problems with occupation and activities	2	1

Baseline Arithmetic:

$S_{text{behaviour}} = 1 + 0 + 2 = 3$
$S_{text{impairment}} = 0 + 1 = 1$
$S_{text{symptom}} = 0 + 3 + 2 = 5$
$S_{text{social}} = 2 + 1 + 0 + 2 = 5$
Total Baseline: $T_{text{baseline}} = 3 + 1 + 5 + 5 = 14$

Follow-Up Arithmetic (12 Weeks):

$S_{text{behaviour}}’ = 0 + 0 + 1 = 1$
$S_{text{impairment}}’ = 0 + 1 = 1$
$S_{text{symptom}}’ = 0 + 1 + 1 = 2$
$S_{text{social}}’ = 1 + 0 + 0 + 1 = 2$
Total Follow-Up: $T_{text{follow-up}} = 1 + 1 + 2 + 2 = 6$

Longitudinal Change:

$$Delta T = T_{text{baseline}} – T_{text{follow-up}} = 14 – 6 = +8$$

The patient exhibited an 8-point absolute reduction in problem severity (a 57.1% relative reduction, $[8 / 14] times 100$), reflecting marked recovery in depressive symptoms and daily functioning.

Interpretation and Score Reading Guidance

HoNOS scores offer standardized benchmarks across clinical settings:

Score Band	Severity Level	Typical Setting / Presentation	Clinical Reading
0–4	Minimal	Remission; primary care discharge	Minimal functional disruption; monitor maintenance
5–9	Mild	Primary care mental health, mild outpatient care	Low-intensity psychological interventions
10–14	Moderate	Community Mental Health Team intake	Structured multidisciplinary care management
15–19	Marked	Crisis intervention teams, day hospitals	Intensive community support; risk of breakdown
20–48	Severe	Acute psychiatric inpatient admission	Pervasive functional impairment; acute crisis

Evaluating subscale profiles alongside total scores avoids obscuring divergent domain changes. Researchers should also account for possible floor and ceiling effects in mild ambulatory cohorts, where low baseline scores on psychotic scales may restrict measurable improvement. For comprehensive assessment, HoNOS is frequently paired with patient-reported instruments like PROMIS measures.

Assumptions and Common Mistakes

Standardized HoNOS administration requires avoiding key methodological errors:

Diagnostic Conflation: HoNOS measures problem severity and functioning, not psychiatric diagnoses. Identical scores can reflect entirely different underlying conditions.
Double-Counting Across Scales: Rating a single symptom on multiple items (e.g., scoring alcohol complications under Item 3 and Item 5) violates mutual exclusivity and inflates scores.
Averaging Fluctuating Symptoms: Raters must document the peak severity observed in the evaluation window, not an average.
Misinterpreting Code 9: Rating 9 denotes “not known.” Converting 9 to zero or adding it into sums distorts scores; missing items require resolution or formal imputation.
Rater Drift: Clinician scoring requires periodic calibration. Evaluating inter-rater reliability via intraclass correlation coefficients ensures measurement consistency across sites.

How to Report It (APA Example)

In APA 7th style, report sample size, means, standard deviations, test statistics, degrees of freedom, p-values, and effect sizes for total and subscale scores:

“To evaluate changes in functioning following a 12-week intervention, clinicians administered the Health of the Nation Outcome Scales (HoNOS; Wing et al., 1998) at admission and discharge. Total scores decreased significantly from admission (M = 14.28, SD = 4.12) to discharge (M = 6.45, SD = 3.20), t(48) = 11.34, p < .001, Cohen’s d = 2.12. Subscale improvements occurred in symptoms (admission M = 5.24 vs. discharge M = 1.88; p < .001) and social problems (admission M = 4.88 vs. discharge M = 2.22; p < .001). Inter-rater agreement yielded an intraclass correlation coefficient (ICC [2,1]) of .81.”

Studies should also evaluate convergent and discriminant validity and report test-retest reliability in stable samples.

How to Calculate It in Excel, R, or Python

Automated calculations must validate inputs and prevent missing values from distorting aggregate scores.

Excel Formula

Assuming Items 1 through 12 are recorded in cells B2:M2:

Behavioural Subscale: =SUM(B2:D2)
Impairment Subscale: =SUM(E2:F2)
Symptom Subscale: =SUM(G2:I2)
Social Subscale: =SUM(J2:M2)
Total Score: =IF(COUNTIF(B2:M2, 9)>0, "Code 9 Present", IF(OR(MIN(B2:M2)4), "Invalid Score", SUM(B2:M2)))

R Function

calculate_honos <- function(items) {
  if (length(items) != 12) stop("HoNOS requires exactly 12 items.")
  if (any(items == 9, na.rm = TRUE)) stop("Contains code 9 (not known).")
  if (any(items < 0 | items > 4, na.rm = TRUE)) stop("Scores must be 0 to 4.")

  list(
    behavioural = sum(items[1:3]),
    impairment  = sum(items[4:5]),
    symptom     = sum(items[6:8]),
    social      = sum(items[9:12]),
    total_score = sum(items[1:12])
  )
}

# Example usage
scores <- calculate_honos(c(1, 0, 2, 0, 1, 0, 3, 2, 2, 1, 0, 2))
print(scores)

Python Implementation

def calculate_honos(items: list[int]) -> dict[str, int]:
    if len(items) != 12:
        raise ValueError("HoNOS requires exactly 12 items.")
    if 9 in items:
        raise ValueError("Item list contains code 9 (not known).")
    if any(x  4 for x in items):
        raise ValueError("Item ratings must be between 0 and 4.")

    return {
        "behavioural": sum(items[0:3]),   # Items 1-3 (0-12)
        "impairment":  sum(items[3:5]),   # Items 4-5 (0-8)
        "symptom":     sum(items[5:8]),   # Items 6-8 (0-12)
        "social":      sum(items[8:12]),  # Items 9-12 (0-16)
        "total_score": sum(items),        # Items 1-12 (0-48)
    }

# Example calculation
scores = calculate_honos([1, 0, 2, 0, 1, 0, 3, 2, 2, 1, 0, 2])
print(scores)

Frequently Asked Questions

Who owns the copyright and licensing for HoNOS?

The Royal College of Psychiatrists (RCPsych) holds the copyright. HoNOS is free for routine clinical practice within publicly funded health systems and academic research. Integration into electronic records for care delivery requires no licensing fees. Commercial resale or for-profit software distribution requires written RCPsych authorization.

What time window should clinicians evaluate when scoring HoNOS?

The standard assessment period covers the preceding two weeks. For acute inpatient discharge reviews, the window is shortened to the preceding three days to evaluate pre-discharge clinical stability.

Can service users complete HoNOS as a self-report measure?

No. HoNOS is strictly a clinician-rated scale completed by trained healthcare professionals using multidisciplinary observations and clinical records. Service-user perspectives are assessed using parallel patient-reported outcome measures.

Why is Cronbach’s alpha often moderate for the total HoNOS score?

Studies (e.g., Pirkis et al., 2005) report Cronbach’s alpha values between 0.59 and 0.76. This reflects the scale’s multidimensional architecture across clinical, cognitive, physical, and social domains rather than poor reliability. Subscale profiles should be interpreted alongside total scores.

What adaptations exist within the HoNOS family?

The RCPsych family of measures includes HoNOS 65+ (older adults), HoNOSCA (children and adolescents), HoNOS-Secure (forensic units), and HoNOS-LD / HoNOS-ID (learning and intellectual disabilities).

Sources

Australian Mental Health Outcomes and Classification Network (AMHOCN). (2020). Health of the Nation Outcome Scales (HoNOS) Training Manual and Clinical Guide. Australian Government Department of Health. https://www.amhocn.org/training-resources/honos-working-age-adults (Accessed 2026-11-09)
Pirkis, J. E., Burgess, P. M., Kirk, P. K., Dodson, S., Coombs, T. J., & Williamson, M. K. (2005). Review of the psychometric properties of the Health of the Nation Outcome Scales (HoNOS) family of measures. Australian & New Zealand Journal of Psychiatry, 39(3), 127–139. https://doi.org/10.1080/j.1440-1614.2005.01538.x
Royal College of Psychiatrists (RCPsych). (2018). Health of the Nation Outcome Scales (HoNOS): 2018 Glossary and Guidance. https://www.rcpsych.ac.uk/improving-care/ccqi/outcome-measures/honos (Accessed 2026-11-09)
Te Pou o te Whakaaro Nui. (2021). HoNOS Reference Guide and Scoring Rules. New Zealand Ministry of Health. https://www.tepou.co.nz/resources/honos (Accessed 2026-11-09)
Wing, J. K., Beevor, A. S., Curtis, R. H., Park, S. B. G., Hadden, S., & Burns, A. (1998). Health of the Nation Outcome Scales (HoNOS): Research and development. British Journal of Psychiatry, 172(1), 11–18. https://doi.org/10.1192/bjp.172.1.11