Cohen-Hoberman Inventory of Physical Symptoms (CHIPS)

Abstract

The Cohen-Hoberman Inventory of Physical Symptoms (CHIPS) is a standardized, self-report psychometric instrument developed in 1983 by social psychologist Sheldon Cohen and clinical psychologist Harry M. Hoberman. Designed to quantify perceived physical symptom burden without contaminating physical health indices with explicit indicators of psychological distress, the instrument comprises 33 common somatic complaints rated on a 5-point Likert scale (ranging from 0 = “not at all” to 4 = “an extreme bother”) with respect to distress experienced over the preceding two-week period. Unlike general somatization indices derived from psychiatric diagnostic batteries—which frequently conflate affective states with physiological symptoms—CHIPS was explicitly engineered to exclude items with overt psychological content (such as “feeling nervous” or “feeling depressed”), while retaining somatic manifestations frequently associated with stress, autonomic arousal, and organic illness (e.g., headache, gastrointestinal disturbances, fatigue, musculoskeletal aches). Psychometric evaluations across diverse collegiate and community populations establish high internal consistency (Cronbach’s α = .88 in original validation cohorts; .86 to .92 across subsequent empirical studies) and robust predictive validity, demonstrated by statistically significant prospective correlations with objective health service utilization, prospective healthcare facility visits, and immune functional markers. Factor-analytic investigations have revealed both an overarching unidimensional somatic burden factor and multidimensional sub-factors reflecting gastrointestinal distress, musculoskeletal and pain symptoms, autonomic/cardiovascular reactivity, and generalized fatigue/malaise. The CHIPS remains an indispensable tool across health psychology, psychosomatic medicine, behavioral medicine, and psychoneuroimmunology, particularly within research paradigms investigating the stress-buffering role of social support and cognitive coping mechanisms.

Keywords

Cohen-Hoberman Inventory of Physical Symptoms, CHIPS, somatic symptom burden, health psychology, stress-buffering hypothesis, psychosomatic assessment, somatic distress, psychoneuroimmunology, behavioral medicine, physical symptom inventory.

Authors

The Cohen-Hoberman Inventory of Physical Symptoms was conceptualized and validated by:

• Sheldon Cohen, Ph.D.: Robert E. Doherty University Professor of Psychology Emeritus at Carnegie Mellon University (Pittsburgh, Pennsylvania, USA). Renowned for pioneering contributions to health psychology and psychoneuroimmunology, Dr. Cohen has extensively investigated the biological and behavioral pathways linking psychological stress, social network diversity, social support, and susceptibility to infectious disease and somatic pathology.
• Harry M. Hoberman, Ph.D.: Clinical psychologist, researcher, and adjunct faculty member affiliated with the University of Minnesota Medical School (Minneapolis, Minnesota, USA). Dr. Hoberman’s clinical and research trajectory has encompassed adolescent depression, suicide prevention, psychosomatic symptomatology, and the therapeutic modulation of stress through social and cognitive resources.

Institutional origins: Department of Psychology, Carnegie Mellon University, and Department of Psychology, University of Oregon. Inquiries regarding scale distribution and experimental implementation are commonly directed through academic repositories associated with Carnegie Mellon University’s Laboratory for the Study of Stress, Immunity, and Disease.

Purpose

The primary purpose of the Cohen-Hoberman Inventory of Physical Symptoms (CHIPS) is to assess the subjective distress and functional burden associated with common physical complaints over a circumscribed, two-week recall window. In the decades preceding the instrument’s publication in 1983, behavioral science and psychosomatic literature frequently grappled with methodological circularity when attempting to demonstrate associations between psychological stressors (e.g., negative life events) and physical health outcomes. Many existing somatic symptom checklists—such as components of the Hopkins Symptom Checklist (HSCL) or the Minnesota Multiphasic Personality Inventory (MMPI) Somatization subscales—included items that inherently reflected psychological states, affective disturbances, or psychiatric demoralization (e.g., crying easily, feeling anxious, experiencing existential dread).

Consequently, observed statistical correlations between life stress and physical illness were frequently confounded by shared method variance and overlapping conceptual definitions. Cohen and Hoberman developed CHIPS precisely to dismantle this measurement confound. The instrument provides a pure physical symptom checklist that measures genuine physical bother while strictly excising transparent emotional, cognitive, and psychiatric states. This separation enables researchers to test whether psychological stress or psychosocial buffers (such as perceived social support or positive life events) prospectively predict physical symptomatology without artificially inflating effect sizes through affective item overlap.

In clinical and experimental contexts, CHIPS fulfills several operational roles:

• Testing the Stress-Buffering Hypothesis: Serving as an uncontaminated physical outcome criterion in models evaluating whether social support, dispositional optimism, or emotional regulation attenuates the pathogenic influence of life changes.
• Behavioral Medicine and Primary Care Screening: Measuring baseline physical complaints and tracking symptom reduction or exacerbation across longitudinal lifestyle, pharmacotherapy, or cognitive-behavioral interventions.
• Objective Healthcare Utilization Prediction: Demonstrating how subjective physical symptom distress translates into health-seeking behaviors, such as physician consultations, infirmary visits, and prescription medication adherence.
• Psychoneuroimmunological Trials: Assessing self-reported physical symptoms alongside physiological biomarkers, including inflammatory cytokines (e.g., IL-6, TNF-alpha), cortisol rhythms, and viral challenge susceptibility.

Psychological Construct

The psychological construct assessed by the CHIPS is perceived somatic symptom burden (or subjective physical symptom bother). Within psychometrics and behavioral medicine, somatic complaints are recognized not merely as direct physiological reflections of tissue pathology, but as complex biobehavioral phenomena shaped by peripheral sensory input, central sensory processing, neuroendocrine activation, and cognitive appraisal. CHIPS captures the severity of psychological distress and behavioral hindrance elicited by somatic perturbations.

Deconstruction of Somatic Domains

Although CHIPS was initially deployed as a unidimensional composite index reflecting overall physical symptom burden, its 33 items broadly sample multiple distinct physiological systems, capturing a comprehensive spectrum of bodily complaints:

• Autonomic and Cardiovascular Arousal: Items such as “Heart pounding or racing” (Item 16), “Shortness of breath when not exercising or working hard” (Item 18), “Hands trembling” (Item 15), “Faintness” (Item 7), and “Hot or cold spells” (Item 14). These items tap into sympathetic nervous system hyperactivity, vasomotor instability, and acute hyperarousal.
• Gastrointestinal Distress: Items including “Constipation” (Item 4), “Diarrhea” (Item 6), “Nausea and/or vomiting” (Item 11), “Acid stomach or indigestion” (Item 12), and “Stomach pain (e.g., cramps)” (Item 13). The enteric nervous system is acutely responsive to psychological stress via the brain-gut axis, making functional GI disturbances prime markers of stress-induced physical strain.
• Musculoskeletal Discomfort and Pain: Items such as “Back pain” (Item 3), “Muscle tension or soreness” (Item 25), “Muscle cramps” (Item 26), “Severe aches and pains” (Item 27), and “Pulled (strained) muscles” (Item 31). Chronic muscle hypertonicity is a well-established physiological consequence of sustained hypothalamic-pituitary-adrenal (HPA) axis activation and autonomic outflow.
• Cephalic and Neurological Complaints: Items encompassing “Dizziness” (Item 5), “Headache” (Item 9), “Migraine headache” (Item 10), “Blurred vision” (Item 24), and “Numbness or tingling in parts of your body” (Item 19). These reflect central nervous system sensitivity, vascular changes in cranial blood vessels, and peripheral paresthesias.
• Constitutional Fatigue and Energy Depletion: Items such as “Constant fatigue” (Item 8), “Felt weak all over” (Item 20), “Feeling low in energy” (Item 22), “Sleep problems (can’t fall asleep, wake up in middle of night or early in morning)” (Item 1), and “Weight change (gain or loss of 5 lbs. or more)” (Item 2). These somatic markers represent systemic neuroendocrine exhaustion, vegetative disruptions, and circadian disturbances.
• Infectious, Dermatological, and Mucosal Symptoms: Items including “Stuffy head or nose” (Item 23), “Cold or cough” (Item 33), “Acne” (Item 28), “Bruises” (Item 29), and “Nosebleed” (Item 30). These complaints monitor immune-related susceptibility, upper respiratory infections, microvascular fragility, and cutaneous reactivity.

By assessing both classic “psychosomatic” symptoms (e.g., tension headaches, indigestion) and more objective physical ailments (e.g., nosebleeds, cough, muscle strains), CHIPS ensures that the construct captures both functional bodily complaints and actual physical vulnerability.

Theoretical Framework

The Cohen-Hoberman Inventory of Physical Symptoms is firmly rooted in the Stress-Buffering Model (Cohen & Wills, 1985) and the broader Transactional Model of Stress and Coping articulated by Richard Lazarus and Susan Folkman (1984). According to this theoretical perspective, environmental demands (stressors) produce adverse physiological and psychological consequences only when appraised as taxing or exceeding an individual’s coping resources.

The Stress-Buffering Hypothesis

Cohen and Hoberman (1983) originally introduced CHIPS to investigate how psychosocial assets—specifically interpersonal social support and positive life events—protect individuals exposed to severe life change stress. The buffering model posits that psychosocial resources intervene at two critical junctures:

1. Primary/Secondary Appraisal: Social support and internal resilience alter the perception of stressors, attenuating the initial neuroendocrine cascade (sympathetic-adrenal-medullary [SAM] axis and hypothalamic-pituitary-adrenal [HPA] axis activation).
2. Physiological Mitigation: When stress is appraised as threatening, coping resources facilitate behavioral adaptation (e.g., sleep hygiene, nutritional maintenance, medical compliance) and dampening of prolonged autonomic arousal, thereby preventing the transition from acute physiological reactivity to sustained somatic symptoms.

To rigorously test whether a moderator (such as social support) interacts with stress to predict physical pathology, the outcome measure must independently reflect somatic illness rather than being an artifactual proxy for depressive or anxious demoralization. CHIPS provided the empirical anchor for this methodology. If social support buffered against the manifestation of CHIPS scores in individuals undergoing major negative life events, researchers could conclusively claim a somatic protective effect rather than merely a reduction in affective complainers.

The Allostatic Load Formulation

In modern psychoneuroendocrinology, CHIPS corresponds closely to the theoretical concept of allostatic load proposed by Bruce McEwen. Allostatic load refers to the cumulative wear and tear on organ systems resulting from repeated or sustained physiological adaptation to chronic environmental challenges. High scores on CHIPS represent the symptomatic manifestation of allostatic state alterations: dysregulated circadian cortisol rhythms, prolonged pro-inflammatory cytokine activity (interleukin-1, interleukin-6), hyperdynamic beta-adrenergic cardiovascular states, and musculoskeletal tension. CHIPS serves as the subjective phenomenological endpoint of this allostatic dysregulation.

Validity

The psychometric validity of the CHIPS has been established through empirical research spanning construct, predictive, convergent, and discriminant validation frameworks.

Predictive and Criterion Validity

The foundational validation studies conducted by Cohen and Hoberman (1983) demonstrated strong prospective criterion validity. In two independent samples of university students, baseline CHIPS scores prospectively predicted objective, verified visits to student health centers over the subsequent five-week period, yielding statistically significant Pearson correlation coefficients of r = .22 (p < .05) and r = .29 (p < .01). These findings confirmed that subjective ratings of symptom distress on CHIPS correspond to actual health-seeking behaviors and verifiable medical consultations, rather than operating merely as transient verbal complaining.

Convergent Validity

Convergent validity has been repeatedly demonstrated through significant, theory-consistent associations with established indices of negative life stress, psychological strain, and somatic symptom scales:

• Correlations with Negative Life Events: In the initial validation cohorts, CHIPS correlated positively and significantly with negative life change scores on the Life Experiences Survey (LES; r values ranging from .30 to .42, p < .001).
• Correlations with Perceived Stress: Strong convergence has been established with the Perceived Stress Scale (PSS), with coefficients typically spanning r = .45 to .58 across diverse cohorts.
• Correlations with Depressive Symptomatology: CHIPS correlates moderately with the Beck Depression Inventory (BDI; typically r = .38 to .49) and the Center for Epidemiologic Studies Depression Scale (CES-D). Because physiological disruptions (sleep disturbance, fatigue, appetite alterations) are shared features of somatic illness and depression, moderate convergence is expected.

Discriminant Validity

Crucially, CHIPS demonstrates discriminant validity from pure affective measures. In multiple hierarchical regression models examining the stress-buffering effect, the interaction between negative life events and social support predicted CHIPS scores even after statistically partialling out baseline depressive symptoms, trait neuroticism, or general negative affectivity. Furthermore, when compared against psychiatric distress inventories (such as the SCL-90-R anxiety and depression subscales), CHIPS consistently emerges as a distinct empirical factor, confirming that it measures bodily symptom burden rather than nonspecific psychological distress.

Reliability

The reliability of the CHIPS has been thoroughly examined across diverse international cohorts, clinical populations, and normative academic samples.

Internal Consistency

In the original validation studies by Cohen and Hoberman (1983), the scale exhibited high internal consistency, yielding a Cronbach’s alpha (α) of .88 across two collegiate samples. Subsequent independent investigations have affirmed the high homogeneity of the 33-item inventory:

• In occupational stress studies evaluating corporate and healthcare personnel, Cronbach’s alpha coefficients for the total CHIPS score consistently range between .87 and .92.
• In cohorts experiencing chronic environmental stressors, internal consistency has been reported at .89 to .91, indicating that the items reliably co-vary to reflect an overarching construct of somatic distress.
• Split-half reliability coefficients, calculated via the Spearman-Brown prophecy formula, have routinely exceeded .85.

Test-Retest Reliability and Stability

Because CHIPS assesses symptoms experienced during a defined past two weeks window, it functions as a state-sensitive measure of somatic strain rather than a fixed personality trait. Consequently, test-retest reliability varies predictably as a function of the retest interval:

• Across brief intervals (e.g., 2 to 3 weeks), test-retest reliability is moderately high, with correlation coefficients ranging from r = .68 to .76, reflecting moderate stability in physical symptom burden under unchanging environmental conditions.
• Across extended periods (e.g., 8 to 12 weeks), test-retest correlations attenuate to between r = .45 and .55, consistent with theoretical expectations that physical symptoms fluctuate in response to dynamic life stressors, acute illnesses, and lifestyle shifts.

The standard error of measurement (SEM) remains low relative to the scale’s total variance, providing researchers and clinicians with sufficient precision to detect meaningful intra-individual symptom changes over time.

Factor Analysis

Although the CHIPS was initially proposed as a single global composite score summing distress across all 33 symptoms, extensive exploratory factor analyses (EFA) and confirmatory factor analyses (CFA) have elucidated its underlying latent architecture.

Exploratory Factor Analyses (EFA)

Principal Axis Factoring and Principal Component Analyses with oblique (Promax/Oblimin) rotations have typically extracted between four and six correlated first-order factors, which account for approximately 45% to 58% of the total variance:

• Factor 1: Gastrointestinal Symptoms (GI): High factor loadings (> .50) from items such as Acid stomach or indigestion, Stomach pain, Nausea and/or vomiting, Constipation, and Diarrhea.
• Factor 2: Musculoskeletal / Pain Complaints: Substantial loadings from Back pain, Muscle tension or soreness, Muscle cramps, Severe aches and pains, and Pulled muscles.
• Factor 3: Autonomic / Cardiovascular Reactivity: Characterized by high loadings from Heart pounding or racing, Shortness of breath, Faintness, Dizziness, Hands trembling, and Hot or cold spells.
• Factor 4: Fatigue / Neuromuscular Malaise: Defined by strong loadings on Constant fatigue, Feeling low in energy, Felt weak all over, and Sleep problems.
• Factor 5: Cephalic / Sensory Disturbances: Capturing Headache, Migraine headache, Blurred vision, and Stuffy head or nose.

Confirmatory Factor Analysis (CFA) and Bifactor Modeling

Confirmatory factor analytic investigations have compared single-factor, correlated multi-factor, and hierarchical/bifactor structures:

CFA results routinely demonstrate that while correlated multi-factor models yield acceptable fit, a bifactor model provides the superior statistical representation of the data. In the bifactor formulation, each item loads simultaneously on a broad, general “Physical Symptom Burden” factor and on its specific physiological sub-domain. Because the general factor accounts for the preponderance of common variance (explained common variance [ECV] typically > .65), psychometricians endorse the standard practice of calculating and interpreting the total CHIPS composite score for overarching stress-health research, while allowing researchers focused on specific body systems to isolate distinct sub-domain scores.

Instrument / Measurement Tool

• Instrument Name: Cohen-Hoberman Inventory of Physical Symptoms (CHIPS)
• Alternative Names: CHIPS Inventory; Cohen-Hoberman Somatic Symptom Scale
• Target Construct: Perceived physical symptom burden and bodily distress (excluding explicit psychological symptoms)
• Target Population: Adults and adolescents aged 16 years and older; validated in undergraduate students, working adults, community samples, and clinical health cohorts
• Administration Format: Self-administered paper-and-pencil or secure digital survey
• Completion Time: Approximately 5 to 8 minutes
• Total Number of Items: 33 items
• Time Recall Window: The past two weeks, including today
• Response Format: 5-point Likert-type scale scored from 0 to 4:
  • 0 = Have not been bothered by the problem / Not at all
  • 1 = Bothered very little
  • 2 = Bothered moderately
  • 3 = Bothered quite a bit
  • 4 = The problem has been an extreme bother / Extremely
• Scoring Rules:
  • Total Score Calculation: Calculate the sum across all 33 items. Total scores range from 0 to 132.
  • Directionality: Higher scores denote greater perceived physical symptom burden, somatic distress, and functional bother.
  • Symptom Count Scoring (Alternative): Some researchers compute a dichotomous symptom count by summing items rated ≥ 1 (any presence) or ≥ 2 (moderate bother), yielding a range from 0 to 33. However, continuous summation (0–132) is standard as it retains nuanced distress variance.
  • Missing Data Imputation: If fewer than 10% of items (≤ 3 items) are missing, person-mean substitution across remaining completed items is standard; if > 3 items are omitted, the protocol is considered invalid.

Permissions & Fee and Test Year

The Cohen-Hoberman Inventory of Physical Symptoms was developed in 1983 by Sheldon Cohen and Harry M. Hoberman and published in the Journal of Applied Social Psychology. The instrument is considered an open-access, non-commercial psychological assessment available for academic, scientific, and clinical research purposes.

Dr. Sheldon Cohen and Carnegie Mellon University maintain an open psychometrics resource page providing the CHIPS scale freely to the scientific community. No licensing fee or royalty payment is required for non-profit academic research. Researchers utilizing the instrument are expected to provide full academic attribution by citing the foundational 1983 validation article. Commercial deployment, inclusion within commercial diagnostic software platforms, or third-party fee-based distribution requires explicit formal permission from the scale authors.

References

• Cohen, S., & Hoberman, H. M. (1983). Positive events and social supports as buffers of life change stress. Journal of Applied Social Psychology, 13(2), 99–125. https://doi.org/10.1111/j.1559-1816.1983.tb02325.x
• Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98(2), 310–357. https://doi.org/10.1037/0033-2909.98.2.310
• Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24(4), 385–396. https://doi.org/10.2307/2136404
• Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. Springer Publishing Company.
• McEwen, B. S. (1998). Protective and damaging effects of stress mediators. New England Journal of Medicine, 338(3), 171–179. https://doi.org/10.1056/NEJM199801153380307
• Sarason, I. G., Johnson, J. H., & Siegel, J. M. (1978). Assessing the impact of life changes: Development of the Life Experiences Survey. Journal of Consulting and Clinical Psychology, 46(5), 932–946. https://doi.org/10.1037/0022-006X.46.5.932

Items of the Scale