Health Opinion Survey (HOS)

1. Abstract

The Health Opinion Survey (HOS) is a seminal 20-item screening questionnaire developed by Alexander M. Macmillan in 1957 as part of the groundbreaking Stirling County Study of psychiatric epidemiology in Nova Scotia, Canada. Designed to identify individuals suffering from psychoneurotic disorders and generalized psychological distress within community populations, the HOS represents one of the earliest standardized instruments in psychiatric epidemiology. The inventory assesses manifestations of emotional disturbance, primarily operationalized through psychophysiological, vegetative, and psychosomatic symptomatology, alongside subjective experiences of nervous breakdown and functional impairment. Each item is rated on a three-point frequency continuum: 1 (hardly ever), 2 (sometimes), and 3 (often), yielding total composite scores ranging from 20 to 60. Extensive psychometric evaluations conducted over several decades demonstrate that the HOS possesses acceptable internal consistency (Cronbach’s alpha typically ranging between .75 and .86) and moderate to high test-retest reliability (.70 to .87). Although initially conceived as a unidimensional indicator of neurotic distress or general psychopathology, subsequent exploratory and confirmatory factor analyses have identified multidimensional substructures, commonly distinguishing somatic complaints, physiological reactivity, sleep disruption, and cognitive-affective anxiety. While the instrument exhibits high sensitivity in detecting individuals with diagnosable neuroses and stress reactions in public health surveys, psychometric research has highlighted that its heavy reliance on physical and autonomic complaints can conflate somatic illness with psychiatric disturbance, particularly among elderly populations and individuals with chronic physical health conditions. Nevertheless, the HOS remains a historically vital and widely cited instrument that paved the way for modern mental health screening inventories such as the General Health Questionnaire (GHQ) and the Symptom Checklist-90 (SCL-90).

2. Keywords

Health Opinion Survey, HOS, psychiatric epidemiology, psychoneurotic disorder, psychosomatic symptoms, psychophysiological distress, Stirling County Study, psychological screening, neurosis, mental health assessment, somatic complaints.

3. Authors

The Health Opinion Survey was conceived and formulated by Alexander M. Macmillan, Ph.D., who served as a senior research associate and psychometrician with the Cornell Program in Social Psychiatry. Working in close collaboration with Alexander H. Leighton, the principal investigator of the landmark Stirling County Study, Macmillan designed the instrument to bridge clinical psychiatric assessment and large-scale sociological survey methodology.

Macmillan’s academic appointments included affiliations with the Department of Sociology and Anthropology at Cornell University and the Department of Psychiatry at the Cornell University Medical College in New York. Later psychometric investigations, refinements, and normative validations were conducted by independent researchers, including Arthur M. Semmence in general practice settings in the United Kingdom, Michel Tousignant, Guy Denis, and Réjean Lachapelle in Canadian francophone populations, and Mark C. Butler and Allan P. Jones within occupational and military organizational contexts.

4. Purpose

The primary purpose of the Health Opinion Survey is to provide a standardized, brief, and cost-effective screening mechanism capable of estimating the prevalence of psychoneurotic disorders, emotional distress, and functional psychological impairment in non-institutionalized community populations. During the mid-twentieth century, psychiatric epidemiology faced severe logistical and methodological bottlenecks: direct psychiatric diagnostic interviews conducted by licensed clinicians were prohibitively expensive, time-consuming, and subject to marked inter-rater variance when applied to thousands of community residents.

To overcome these obstacles, Macmillan and the Stirling County research team sought to construct an objective instrument that could be administered by trained non-medical interviewers or completed as a self-administered questionnaire. The specific objectives and applications of the HOS encompass:

• Epidemiological Case Identification: Functioning as a first-stage triage screening mechanism to identify respondents who demonstrate a high probability of meeting clinical criteria for neurosis, anxiety reactions, or psychoneurotic personality disturbances according to nosological frameworks of the era.
• Community Health Surveillance: Enabling public health authorities and social epidemiologists to assess the mental health status of diverse geographical, socio-economic, and cultural cohorts, thereby delineating demographic correlations between social stress, urbanization, cultural disintegration, and psychological morbidity.
• Primary Care Triage: Assisting general medical practitioners in identifying hidden psychiatric distress, somatization, and functional overlay among ambulatory medical patients presenting with ambiguous physical complaints (Semmence, 1969).
• Occupational and Stress Research: Evaluating the psychological toll of environmental stressors, hazardous occupations, military assignments, and organizational changes (Butler & Jones, 1979).

The theoretical rationale underlying the HOS posits that psychiatric disturbance, specifically psychoneurosis, manifests predominantly in individuals’ self-perceptions of altered physiological function, autonomic nervous system lability, and subjective fatigue. By measuring the frequency of these physical and subjective signals, the HOS captures the behavioral and physiological manifestations of chronic emotional strain without relying on overt psychiatric terminology that might elicit defensive denial or social desirability bias.

5. Psychological Construct

The core psychological construct operationalized by the Health Opinion Survey is psychoneurotic distress, expressed primarily through the nexus of somatic symptoms, autonomic hyperarousal, and perceived vulnerability to nervous collapse. At the time of its development, psychiatric theory conceptualized psychoneurosis as an enduring state of emotional maladjustment characterized by anxious apprehension, internal conflict, and vegetative nervous system dysfunction, operating without gross distortions of external reality (in contrast to psychosis).

Although initially evaluated as a global index of psychiatric impairment, empirical investigations have revealed that the HOS measures several interconnected sub-constructs:

Autonomic and Physiological Reactivity

This dimension encompasses hyperactivation of the sympathetic nervous system in response to chronic environmental or psychological stressors. The construct is manifested in physical signs of autonomic arousal, including peripheral vasomotor changes, cardiac palpitations, and hyperhidrosis. Specific questionnaire items tapping this domain include:

• Item 5: “Are you ever troubled by your hands sweating so that they feel damp and clammy?”
• Item 8: “Do your hands ever tremble enough to bother you?”
• Item 14: “Have you ever been bothered by your heart beating hard?”
• Item 17: “Have you ever been bothered by shortness of breath when you were not exercising or working hard?”
• Item 20: “Have you ever been troubled by ‘cold sweats’?”

Gastrointestinal and Vegetative Disturbance

Chronic anxiety and neurotic conflict frequently disrupt basic biological rhythms and vegetative bodily systems. Gastrointestinal distress has long been recognized as a primary somatic correlate of affective dysregulation. The HOS captures this domain through:

• Item 1: “Do you have loss of appetite?”
• Item 2: “How often are you bothered by having an upset stomach?”
• Item 15: “Do you tend to lose weight when you have important things bothering you?”

Sleep Disturbance and Chronic Asthenia

Impairment of restorative biological functions constitutes another fundamental facet of the neurotic construct. Persistent tension impairs sleep architecture and leads to chronic morning fatigue and subjective lethargy, operationalized by:

• Item 7: “Do you ever have any trouble in getting to sleep and staying asleep?”
• Item 11: “Are you ever bothered by having nightmares?”
• Item 18: “Do you tend to feel tired in the mornings?”

Subjective Nervousness and Cognitive Dread

Beyond visceral symptoms, the HOS evaluates conscious awareness of emotional disequilibrium, subjective tension, and fear of catastrophic psychological decompensation:

• Item 4: “Have you ever felt that you were going to have a nervous breakdown?”
• Item 10: “Do you ever take weak turns?”
• Item 13: “Have you ever had spells of dizziness?”
• Item 16: “Are you ever bothered by nervousness?”

Functional Impairment and General Health Perception

The final facet addresses the behavioral consequence of distress—specifically, the degree to which health issues constrain everyday vocational functioning and subjective vigor:

• Item 3: “Has any ill health affected the amount of work you do?”
• Item 6: “Do you feel that you are bothered by all sorts of ailments in different parts of your body?”
• Item 9: “Do you have any particular physical or health trouble?”
• Item 19: “For the most part, do you feel healthy enough to carry out the things that you would like to do?”

6. Theoretical Framework

The conceptual foundation of the Health Opinion Survey rests upon mid-twentieth-century psychosomatic medicine, stress theory, and the social psychiatric paradigms championed by Alexander Leighton and Adolf Meyer. The survey was constructed at the intersection of three major theoretical orientations:

Adolf Meyer’s Psychobiology

Adolf Meyer, the architect of American psychobiology, argued that mental disorders are not discrete biological disease entities localized within specific brain organs, nor are they merely intrapsychic fantasies. Instead, Meyer posited that psychiatric disorders represent holistic “reaction types” of the total integrated organism (the person) adapting to life situations and social environments. Macmillan and Leighton operationalized this framework by conceptualizing psychiatric morbidity as an integrated psychobiological continuum wherein social stress induces observable autonomic, somatic, and behavioral reactions.

Walter Cannon and Hans Selye’s Stress Paradigms

The physiologic items of the HOS directly reflect the classic physiological formulations of Walter Cannon‘s “fight-or-flight” emergency response and Hans Selye‘s General Adaptation Syndrome (GAS). Under prolonged social and psychological stress, continuous activation of the sympathetic-adrenal-medullary (SAM) axis and hypothalamic-pituitary-adrenal (HPA) axis generates persistent physiological dysregulation. Symptoms such as tremors, cold sweats, palpitations, tachypnea, and gastrointestinal hypermotility are interpreted as direct markers of organismic strain in the resistance or exhaustion phases of chronic stress.

Sociocultural Disintegration Theory

At the macro-sociological level, the HOS was designed to test Leighton’s overarching thesis in the Stirling County Study: that sociocultural disintegration (characterized by poverty, broken homes, cultural confusion, weak leadership, and secularization) impairs individual psychological functioning. Leighton posited that human beings possess fundamental psychological needs for physical security, affection, membership in a definite human group, and a sense of belonging. When societal fragmentation prevents the satisfaction of these essential striving sentiments, the individual experiences heightened anxiety and psychobiological disruption, which registers as elevated scores on the HOS.

7. Validity

Over the decades following its publication, the Health Opinion Survey was subjected to rigorous empirical examinations of construct, criterion, convergent, and discriminant validity across diverse clinical and community cohorts.

Criterion and Predictive Validity

In Macmillan’s original 1957 validation study, the instrument was administered to known clinical psychiatric patients and a matched non-patient community sample. Using a cut-off score of 30 or higher, the HOS demonstrated remarkable sensitivity, correctly identifying over 85% of diagnosed psychoneurotic individuals. In a subsequent clinical trial conducted in British general practice, Semmence (1969) administered the HOS to 411 patients. Patients classified by general practitioners as exhibiting neurotic or psychiatric symptoms scored significantly higher (mean = 35.8, SD = 6.4) than healthy controls (mean = 27.2, SD = 4.8), yielding strong criterion discrimination (p < .001).

Convergent and Discriminant Validity

Convergent validity has been established through moderate-to-strong correlations with other validated psychometric inventories measuring anxiety, depression, and neuroticism. In comparative investigations, HOS scores correlate substantially with the Taylor Manifest Anxiety Scale (TMAS, r = .68 to .74), the Eysenck Personality Inventory Neuroticism (N) scale (r = .62 to .71), and the Langner 22-Item Index (r = .78 to .84). Butler and Jones (1979) observed significant correlations between HOS scores and self-reported job dissatisfaction, emotional exhaustion, and somatic complaints among military personnel.

Critiques and Discriminant Limitations

Despite its sensitivity, the discriminant validity of the HOS has encountered substantial scholarly critique. In a landmark paper, Tousignant, Denis, and Lachapelle (1974) re-examined the scale and demonstrated that the HOS suffers from a high rate of “false positives” among individuals with genuine physical illnesses. Because 14 of the 20 items focus on physical and autonomic complaints (such as stomach upset, shortness of breath, trembling, and heart pounding), respondents suffering from organic medical conditions—such as cardiovascular disorders, asthma, endocrine dysfunctions, or chronic gastrointestinal ailments—obtain highly elevated HOS scores in the absence of psychiatric neurosis. Furthermore, elderly populations systematically score higher on the HOS due to age-related physical infirmity rather than elevated psychopathology, indicating that the instrument measures a combined construct of physical and emotional ill-health rather than pure psychological distress.

8. Reliability

The reliability of the Health Opinion Survey has been demonstrated across multiple community surveys, clinical cohorts, and occupational investigations.

Internal Consistency

Macmillan’s (1957) initial investigations yielded split-half reliability coefficients ranging from .82 to .88, corrected by the Spearman-Brown formula. Subsequent studies utilizing modern psychometric indices have consistently corroborated the internal consistency of the 20-item scale:

• Butler and Jones (1979) reported a Cronbach’s alpha of .83 in a sample of 2,122 US Navy personnel.
• Tousignant et al. (1974) identified an internal consistency coefficient of .81 in a representative Canadian community sample.
• McDowell (2006) summarized multiple cross-sectional population studies, showing alpha coefficients consistently clustering between .75 and .86.

Item-total correlations for the majority of items range between .32 and .61. Items such as Item 4 (nervous breakdown), Item 14 (heart pounding), Item 16 (nervousness), and Item 18 (morning fatigue) demonstrate the highest corrected item-total correlations, whereas Item 12 (excessive smoking) and Item 15 (weight loss when bothered) display lower correlations (.18 to .29), reflecting their behavioral rather than direct affective-physiological character.

Test-Retest Stability

Temporal stability assessments have shown that the HOS demonstrates moderate-to-high stability across short- and medium-term intervals. Macmillan reported a 6-week test-retest reliability coefficient of r = .87 among non-patient volunteers. Butler and Jones evaluated test-retest reliability over a 6-month deployment interval, observing a stability coefficient of r = .70, indicating that while the HOS captures relatively enduring neurotic dispositions, scores fluctuate predictably in response to substantial shifts in environmental stress levels.

9. Factor Analysis

While Macmillan initially treated the HOS as an essentially unidimensional screening metric for psychoneurotic status, subsequent exploratory factor analyses (EFA) and confirmatory factor analyses (CFA) have conclusively demonstrated its multidimensional nature.

Multidimensional Structural Findings

Butler and Jones (1979) conducted a principal components factor analysis with varimax rotation on HOS responses from 2,122 subjects. Their analysis refuted the strict unidimensional assumption, identifying four distinct, interpretable factors accounting for 48.6% of the common variance:

Tousignant et al. (1974) similarly identified a primary somatic-physical health factor distinct from a psychological distress factor, underscoring that the scale conflates organic physical complaints with purely functional psychological reactions.

10. Instrument / Measurement Tool

• Instrument Name: Health Opinion Survey (HOS)
• Alternative Names: Macmillan Health Opinion Survey, Stirling County Health Opinion Questionnaire
• Author: Alexander M. Macmillan, Ph.D.
• Original Publication Year: 1957
• Assessment Type: Self-administered questionnaire or structured interviewer-administered screening schedule
• Construct Measured: Psychoneurotic disorder, psychosomatic symptoms, autonomic hyperreactivity, and general emotional distress
• Target Population: Adults (aged 18 and older) in general community surveys, primary healthcare, or epidemiological research
• Item Count: 20 items
• Response Continuum: 3-point frequency response scale:
  • 1 = Hardly ever
  • 2 = Sometimes
  • 3 = Often
• Scoring Protocol:
  • Each item is scored from 1 to 3 according to symptom frequency.
  • Item 16: Marked in historical scoring documentation as reverse-coded or specifically evaluated for directionality depending on administration convention (note: in standard epidemiological scoring, all items are aligned such that higher numerical values indicate greater pathology; positively worded health items such as Item 19 are routinely reversed in standard implementations: 1 = Often, 2 = Sometimes, 3 = Hardly ever).
  • Total Score Range: 20 to 60 points.
  • Clinical Cut-Off Thresholds:
    • Scores < 30: Within normal community limits; low probability of psychiatric caseness.
    • Scores 30 to 34: Borderline / intermediate distress; elevated risk of psychoneurotic disturbance requiring further diagnostic exploration.
    • Scores ≥ 35: High probability of clinically significant psychoneurotic disorder or severe psychosomatic distress.
• Administration Time: Approximately 5 to 10 minutes.

11. Permissions & Fee and Test Year

The Health Opinion Survey was first published by Alexander M. Macmillan in 1957 in the journal Psychological Reports. Developed within the context of university-sponsored, publicly funded epidemiological research (the Stirling County Study under Cornell University), the HOS resides in the public domain for academic, scientific, and non-commercial clinical research purposes.

No user fees, royalties, or formal commercial licensing agreements are required to utilize the instrument in academic or healthcare settings. Researchers and clinicians citing the tool should reference Macmillan’s foundational 1957 article and subsequent validation monographs (such as McDowell, 2006). For modified versions or inclusion within proprietary commercial digital platforms, consultation of institutional copyright guidelines is recommended.

12. References

Butler, M. C., & Jones, A. P. (1979). The Health Opinion Survey reconsidered: Dimensionality, reliability, and validity. Journal of Clinical Psychology, 35(3), 554–559. https://doi.org/10.2466/pr0.1957.3.3.325

McDowell, I. (2006). Measuring Health: A Guide to Rating Scales and Questionnaires (3rd ed.). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195165678.001.0001

Semmence, A. M. (1969). The Health Opinion Survey: A psychiatric screening instrument. Journal of the Royal College of General Practitioners, 18(89), 344–348. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2236965/

Tousignant, M., Denis, G., & Lachapelle, R. (1974). Some considerations concerning the validity and use of the Health Opinion Survey. Journal of Health and Social Behavior, 15(3), 241–252. https://doi.org/10.2307/2137021

13. Items of the Scale