HCL-32: Scoring, Cutoffs, and Research Guide

Q: What is the HCL-32 used for?

The HCL-32 is used in clinical research and psychiatric outpatients to screen for lifetime hypomanic symptoms, distinguishing bipolar spectrum disorders (particularly Bipolar II) from unipolar depression.

Q: What is the cutoff score on the HCL-32?

The standard cutoff is ≥ 14 out of 32 points. Scores of 14 or higher indicate a positive screen, yielding approximately 80%–82% sensitivity and 57%–67% specificity in psychiatric services.

Q: What is the difference between the HCL-32 and the MDQ?

The HCL-32 contains 32 items and provides higher sensitivity for Bipolar II disorder (81% vs. 66%). The Mood Disorder Questionnaire (MDQ) contains 13 items plus functional impairment criteria, offering higher specificity (79% vs. 67%) but missing subtle hypomanic presentations.

Q: Does the HCL-32 distinguish between Bipolar I and Bipolar II?

No. Research by Angst et al. (2005) established that total scores reflect general bipolar spectrum liability and cannot distinguish Bipolar I from Bipolar II. Structured clinical interview is required to assess mania severity and hospitalization history.

Q: Is the HCL-32 a diagnostic instrument?

No. The HCL-32 is strictly a screening checklist. A positive score flags individuals who require structured clinical evaluation (e.g., SCID-5 or MINI) but does not constitute a DSM-5-TR diagnosis on its own.

The Hypomania Checklist-32 (HCL-32) is a 32-item self-report questionnaire developed by Jules Angst and colleagues (2005) to screen for lifetime hypomanic symptoms in adult psychiatric outpatients. It addresses the diagnostic misclassification of bipolar spectrum disorders (particularly Bipolar II) as unipolar depression. Patients complete 32 dichotomous (“Yes” or “No”) items assessing cognitive, affective, and behavioral shifts during “high” states. Total scores range from 0 to 32, with an operational cutoff of ≥ 14 indicating a positive screen for potential bipolarity warranting clinical interview confirmation.

Definition

The Hypomania Checklist-32 (HCL-32) is an internationally validated psychometric screening instrument designed to detect lifetime hypomanic episodes. Developed across psychiatric outpatients in Italy and Sweden, the instrument captures both energized, expansive states (“active/elated” hypomania) and impulsive, irritable behaviors (“risk-taking/irritable” hypomania). The tool functions as a screening instrument rather than a definitive diagnostic test, prioritizing sensitivity (80%–82%) to identify patients warranting comprehensive diagnostic evaluation under DSM-5-TR criteria.

Formula and Scoring Architecture

The total HCL-32 score is an unweighted sum of endorsed hypomanic symptoms:

$$S = sum_{i=1}^{32} x_i$$

Where:

S: Total HCL-32 score, ranging from 0 to 32 ($0 le S le 32$).
x_i: Binary indicator for item $i$, where $x_i = 1$ if endorsed (“Yes”) during a high period, and $x_i = 0$ if absent (“No”).
i: Item index running from 1 to 32.

Psychometric evaluations and confirmatory factor analysis demonstrate that the 32 items reflect two correlated latent dimensions:

Active/Elated Hypomania (Factor 1, 20 items): Measures increased energy, mood elevation, sociability, physical activity, and optimism. Computed as $S_{text{active}} = sum_{j in F_1} x_j$ where $F_1$ comprises 20 active/elated items (score range: 0–20).
Risk-Taking/Irritable Hypomania (Factor 2, 12 items): Captures impulsivity, reckless spending, fast driving, substance use, impatience, and irritability. Computed as $S_{text{risk}} = sum_{k in F_2} x_k$ where $F_2$ comprises 12 risk-taking items (score range: 0–12).

Evaluating both subscales alongside the total score strengthens the construct validity of mood disorder phenotyping in quantitative studies.

Worked Example

Consider an adult participant (Participant RP-409) completing the HCL-32 during research intake for an affective disorders study.

The participant reports the following binary responses across the 32 items:

Factor 1 (Active/Elated): Endorses 11 items ($x_i = 1$) out of 20 (including reduced sleep need, heightened energy, increased sociability, and creative ideas). Denies 9 items ($x_i = 0$).
Factor 2 (Risk-Taking/Irritable): Endorses 6 items ($x_i = 1$) out of 12 (including faster driving, impulsive spending, impatience, and interpersonal friction). Denies 6 items ($x_i = 0$).

The scoring arithmetic is calculated directly:

$$S_{text{active}} = 11$$

$$S_{text{risk}} = 6$$

$$S = S_{text{active}} + S_{text{risk}} = 11 + 6 = 17$$

Comparing the total score against the operational screening threshold ($S ge 14$):

$$17 ge 14 implies text{Positive Screening Result}$$

Participant RP-409 meets the criteria for a positive screen. Per research protocol, the participant is referred for a semi-structured diagnostic interview (e.g., SCID-5) to verify episode duration (≥ 4 consecutive days) and confirm clinical diagnosis.

Interpretation

HCL-32 total scores are evaluated against validated diagnostic cutoffs established in clinical trials:

Total Score Range	Classification	Clinical Interpretation	Recommended Action
0–13	Screening Negative	Low likelihood of lifetime hypomania. Endorsement pattern consistent with unipolar depression or healthy controls.	Retain in unipolar cohort; no immediate structured hypomania interview indicated.
14–20	Screening Positive (Moderate)	Elevated probability of bipolar spectrum condition (especially Bipolar II). Meets standard cutoff (Angst et al., 2005).	Administer structured clinical diagnostic interview (SCID-5, MINI) to assess episode duration and functional criteria.
21–32	Screening Positive (High)	Substantial hypomanic symptom burden across both active/elated and risk-taking/irritable dimensions.	Prioritize comprehensive psychiatric diagnostic evaluation; examine medication history and behavioral risks.

Large-scale meta-analyses compare the accuracy of the HCL-32 against the Mood Disorder Questionnaire (MDQ) and Bipolar Spectrum Diagnostic Scale (BSDS):

Instrument	Recommended Cutoff	Summary Sensitivity	Summary Specificity	Comparative Evidence
HCL-32	≥ 14 / 32	81%–82%	57%–67%	Significantly higher sensitivity for Bipolar II disorder compared to MDQ in psychiatric services (Carvalho et al., 2015; Wang et al., 2019).
MDQ	≥ 7 / 13 + criteria	66%–80%	70%–79%	Higher specificity than HCL-32, reducing false positives, but higher false-negative rate for Bipolar II (Carvalho et al., 2015).
BSDS	≥ 13 / 19	69%	86%	High specificity in outpatient cohorts, but lower sensitivity relative to HCL-32 and MDQ (Carvalho et al., 2015).

When applying the instrument across heterogeneous cohorts, researchers must account for floor and ceiling effects. In community cohorts, scores frequently cluster at floor levels (0–5), whereas inpatient cohorts may produce false positives driven by general acute distress.

Assumptions and Common Mistakes

Appropriate application of the HCL-32 requires understanding its psychometric assumptions and common interpretation errors.

Key Methodological Assumptions

Retrospective Recall Accuracy: Assumes participants accurately remember and characterize past “high” states. Concurrent severe depression can introduce recall bias.
Unit Item Weighting: Assumes all 32 items contribute equally to latent hypomanic liability ($x_i = 1$).
Dichotomous Response Sufficiency: Assumes binary (“Yes”/”No”) responses adequately capture hypomanic symptom expression.

Common Mistakes

Treating Screening as Diagnosis: A score ≥ 14 is not a diagnosis. DSM-5-TR requires symptom persistence for at least 4 consecutive days and observable functional change without psychotic features.
Subtype Classification (Bipolar I vs. Bipolar II): Angst et al. (2005) demonstrated that the HCL-32 cannot reliably distinguish Bipolar I from Bipolar II disorders; it measures general bipolar spectrum diathesis.
Ignoring Factor 2 Scores: Focusing only on total scores overlooks risk-taking and irritability symptoms, which are critical predictors of clinical severity.
Unadjusted Community Application: Applying the clinical cutoff of 14 to healthy student or community samples produces elevated false-positive rates due to non-pathological extraversion.

How to Report It

When documenting HCL-32 data in manuscripts following APA 7th edition guidelines, report instrument details, scoring cutoff, sample reliability, and statistical comparisons.

Methods Section Example

“Lifetime hypomanic symptoms were assessed using the Hypomania Checklist-32 (HCL-32; Angst et al., 2005), a 32-item self-report questionnaire scored dichotomously (0 = No, 1 = Yes; score range: 0–32). Scores ≥ 14 indicate a positive screen for bipolar spectrum illness. The scale encompasses two dimensions: Active/Elated Hypomania (20 items) and Risk-Taking/Irritable Hypomania (12 items). In this study, the full scale showed good internal consistency (Cronbach’s alpha α = .85), with subscale reliabilities of α = .82 for Active/Elated and α = .78 for Risk-Taking/Irritable.”

Results Section Example

“Participants with diagnosed bipolar spectrum disorder scored significantly higher on the HCL-32 (M = 18.64, SD = 4.82) than participants with unipolar major depression (M = 11.23, SD = 5.15), t(184) = 10.08, p < .001, Cohen’s d = 1.48. At the cutoff of 14, sensitivity was 82.1% and specificity was 68.3%. Correlation with Young Mania Rating Scale scores supported convergent and discriminant validity.”

How to Calculate It in Excel, R or Python

Scoring workflows can be implemented across Excel, R, and Python environments.

Excel Implementation

With binary item responses in cells B2 through AG2:

# Total Score (Cell AH2) and Screening Result (Cell AI2):
=SUM(B2:AG2)
=IF(AH2>=14, "Positive", "Negative")

# Subscale Scores (Factor 1 in AJ2, Factor 2 in AK2):
=SUM(B2:G2, K2:U2, W2, Y2, AC2)
=SUM(H2:J2, V2, X2, Z2:AB2, AD2:AG2)

R Implementation

Scoring a data frame of binary items and calculating reliability in R:

item_cols <- paste0("item_", 1:32)
df$hcl32_total  <- rowSums(df[, item_cols], na.rm = TRUE)
df$hcl32_screen <- ifelse(df$hcl32_total >= 14, "Positive", "Negative")

library(psych)
psych::alpha(df[, item_cols])$total$raw_alpha

Python Implementation

Scoring a dataset with Python and pandas:

import numpy as np

def score_hcl32(df):
    item_cols = [f"item_{i}" for i in range(1, 33)]
    df["hcl32_total"] = df[item_cols].sum(axis=1)
    df["hcl32_screen"] = np.where(df["hcl32_total"] >= 14, "Positive", "Negative")
    return df

Frequently Asked Questions

What is the HCL-32 used for?