Health Perceptions Questionnaire (HPQ)

1. Abstract

The Health Perceptions Questionnaire (HPQ) is a landmark psychometric instrument developed by Dr. John E. Ware, Jr. and colleagues at the RAND Corporation as part of the groundbreaking National Health Insurance Study (HIS). Constructed to measure subjective health status comprehensively across general non-institutionalized populations, the HPQ operationalizes personal health assessments beyond the mere presence or absence of clinically diagnosed pathology. The instrument evaluates six primary dimensions: Current Health, Prior Health, Health Outlook, Resistance to Illness, Health Worry/Concern, and Sickness Orientation (alongside behavioral attitudes addressing rejection of the sick role and reluctance toward medical consultation). Comprising 32 core self-report items rated on a 5-point Likert-type agreement scale ranging from Definitely True (1) to Definitely False (5), the questionnaire also yields a widely used composite metric known as the General Health Rating Index (GHRI).

Extensive psychometric investigations across large-scale, demographically diverse community cohorts (with empirical sample sizes exceeding $N = 1,790$, $N = 4,700$, and $N = 1,200$) demonstrate solid measurement properties. Internal consistency estimates (Cronbach’s alpha) range from adequate to exceptional across subscales, typically achieving $\alpha = .88$ to $.91$ for Current Health and $\alpha = .89$ for the aggregate GHRI, though narrower behavioral scales such as Sickness Orientation yield more modest coefficients ($\alpha \approx .53 – .59$). Construct, convergent, and discriminant validities are established against comprehensive physical functioning indices, physician assessments, mental health inventories, and longitudinal healthcare utilization records. As a seminal foundation for modern patient-reported outcome measures (PROMs)—including the Medical Outcomes Study (SF-36)—the HPQ remains an enduring paradigm for evaluating self-perceived wellness, health-related quality of life (HRQoL), and subjective disease vulnerability.

2. Keywords

Health Perceptions Questionnaire, HPQ, General Health Rating Index, John E. Ware Jr., RAND Health Insurance Experiment, Self-Rated Health, Health-Related Quality of Life, Resistance to Illness, Health Outlook, Psychometrics, Patient-Reported Outcome Measures, Sickness Orientation

3. Authors

The Health Perceptions Questionnaire was developed primarily by John E. Ware, Jr., Ph.D., in collaboration with Allyson Ross Davies (formerly Davies-Avery), Ph.D., Cathy A. Donald, M.A., and Robert H. Brook, M.D., Sc.D., under the auspices of the RAND Corporation in Santa Monica, California, with funding from the U.S. Department of Health, Education, and Welfare (now the Department of Health and Human Services).

• John E. Ware, Jr., Ph.D.: Senior Research Psychologist at the RAND Corporation during instrument inception; later Professor and Chief of the Outcomes Management System at New England Medical Center and Harvard School of Public Health; founder of QualityMetric Inc.
• Allyson Ross Davies, Ph.D.: Health Services Researcher at the RAND Corporation; specialized in patient satisfaction, health status assessment, and clinical trial outcomes.
• Cathy A. Donald, M.A.: Psychometrician and Policy Analyst at the RAND Corporation; co-author of the RAND Health Insurance Study monograph series on adult health measurement.
• Robert H. Brook, M.D., Sc.D.: Corporate Senior Chair in Health Care Services at the RAND Corporation and Professor of Medicine and Health Services at the University of California, Los Angeles (UCLA); Principal Investigator of the clinical components of the RAND Health Insurance Experiment.

Institutional Address: RAND Corporation, 1776 Main Street, Santa Monica, CA 90401-3208, USA.

4. Purpose

The primary purpose of the Health Perceptions Questionnaire is to provide an empirically standardized, psychometrically rigorous, and comprehensive assessment of subjective health status for use in general population surveys, clinical trials, and health services research. Prior to the design of the HPQ in the mid-1970s, population health evaluations relied almost exclusively on objective physiological endpoints, clinical diagnoses, mortality rates, or single-item subjective global ratings (e.g., “In general, would you say your health is excellent, good, fair, or poor?”). While single-item metrics offered quick administration, they suffered from unknown measurement error, restricted reliability, ceiling effects, and an inability to disentangle distinct temporal and cognitive facets of personal health appraisal.

The HPQ was conceived to operationalize the World Health Organization definition of health—a state of complete physical, mental, and social well-being, rather than merely the absence of disease or infirmity. Ware and colleagues recognized that individuals construct personal evaluations of their health through dynamic cognitive appraisals combining biological sensations, knowledge of medical diagnoses, personal health history, anticipated biological vulnerability, and subjective distress regarding illness. Consequently, the instrument was engineered to achieve several key objectives:

• Population Health Monitoring: To quantify health disparities across socioeconomic, age, and geographical strata within large-scale epidemiologic and health policy studies, such as the RAND Health Insurance Experiment.
• Policy Evaluation and Health Economics: To evaluate how alternative health insurance financing mechanisms (e.g., cost-sharing, fee-for-service, health maintenance organizations) influence longitudinal perceptions of personal well-being and health security over several years.
• Clinical Research and Trials: To serve as a multidimensional patient-reported outcome measure capable of detecting subtle shifts in health trajectories among individuals suffering from chronic medical conditions (e.g., cardiovascular disease, diabetes, arthritis) versus healthy controls.
• Behavioral Prediction: To forecast healthcare utilization patterns, including physician ambulatory visits, adherence to medical regimens, preventive screening uptake, and the adoption or rejection of the formal “sick role.”

By disaggregating general health into temporal vectors (past, present, and future) and behavioral-affective constructs (resistance, worry, and orientation to illness), the HPQ allows researchers to understand not only how healthy an individual feels right now, but how confident they are in their biological resilience and how heavily health anxieties weigh upon their daily functioning.

5. Psychological Construct

The psychological construct underlying the HPQ is Self-Rated Health (SRH), conceptualized as an integrative cognitive-perceptual process. Rather than acting as a simple proxy for medical diagnosis, self-rated health represents an individual’s internal synthesis of somatic sensations, physiological markers, psychological affect, functional capacity, and sociocultural expectations. The HPQ operationalizes this macro-construct through distinct, interrelated sub-dimensions:

Current Health

This core dimension evaluates the respondent’s contemporaneous evaluation of their present biological and physical status. Items in this subscale measure the subjective sense of vigor, wellness, and absence of current disease manifestations (e.g., “My health is excellent,” “I am somewhat ill”). It serves as the primary direct barometer of everyday vitality and somatic equilibrium, capturing both positive health states and debilitating physical impairments.

Prior Health

Prior Health captures the retrospective baseline against which present health is cognitively evaluated. Individuals evaluate their present physical state relative to their personal biological trajectory (e.g., “I have never been seriously ill,” “I was so sick once I thought I might die”). Longitudinal health psychology reveals that a history of traumatic medical events or chronic childhood vulnerability fundamentally recalibrates a patient’s internal standard of reference, altering their tolerance for symptoms and their interpretive framework for current somatic signals.

Health Outlook

Health Outlook measures the prospective, cognitive expectations regarding one’s future physical condition and longevity (e.g., “I expect to have a very healthy life,” “I think my health will be worse in the future than it is now”). This dimension taps into subjective biological optimism versus pessimism. An adverse health outlook frequently functions as an early psychological indicator of subclinical decline, demoralization, or perceived vulnerability to terminal illness, and correlates strongly with health-related self-efficacy.

Resistance to Illness

Resistance to Illness assesses an individual’s subjective perception of their innate immune competence and biological hardiness (e.g., “My body seems to resist illness very well,” “I seem to get sick a little easier than other people”). Unlike objective immunological assays, this construct measures psychological perceived vulnerability to contagion and disease susceptibility, influencing protective health behaviors, social withdrawal during epidemics, and somatosensory amplification.

Health Worry/Concern

This affective construct measures the cognitive burden, anxiety, and distress elicited by real or imagined physical vulnerabilities (e.g., “I worry about my health more than other people do,” “I never worry about my health”). Distinct from generalized anxiety, health worry reflects specific hypochondriacal rumination, disease anxiety, and the psychological salience of somatic vulnerability, which can independently impair quality of life regardless of objective diagnostic severity.

Sickness Orientation and Sick-Role Behaviors

Sickness orientation evaluates an individual’s cognitive and behavioral posture toward being ill, specifically examining whether sickness is accepted as an inevitable biological occurrence or actively resisted (e.g., “I accept the fact that I get sick”). Supplementary items in the HPQ pool probe Rejection of the Sick Role (the tendency to maintain normal daily roles despite functional impairment: “I try to avoid letting illness interfere with my life”) and Attitude Toward Going to the Doctor (reluctance or willingness to seek clinical intervention: “I don’t like to go to the doctor”).

6. Theoretical Framework

The structural design of the HPQ draws upon several foundational models within health psychology, medical sociology, and cognitive psychometrics:

The Biopsychosocial Model

Formulated by George L. Engel in the late 1970s, the biopsychosocial model posits that health and illness cannot be understood solely in terms of molecular, microbiological, or anatomical deviations. Rather, psychological states (affect, health beliefs, cognitive expectations) and social contexts (interpersonal roles, cultural norms of health) interact continuously with biological substrates. The HPQ serves as an empirical instrument of this paradigm by validating that a patient’s personal evaluation of health contains reliable variance independent of physical pathology, yet possessing profound prognostic significance.

Parsons’ Sick Role Theory

Sociologist Talcott Parsons introduced the conceptual framework of the “sick role,” describing illness not merely as a biological condition but as a institutionalized social role characterized by distinct rights (exemption from normal social obligations) and duties (the obligation to desire recovery and seek competent medical assistance). The HPQ operationalizes Parsonian sociology through its explicit measurement of Sickness Orientation and Rejection of the Sick Role. By measuring whether an individual fights disease manifestations, conceals symptoms, or readily relinquishes occupational tasks, the HPQ bridges subjective cognitive appraisal with social behavior.

The Common-Sense Model of Self-Regulation of Health and Illness

Pioneered by Howard Leventhal, the Common-Sense Model (CSM) asserts that individuals construct parallel cognitive and emotional representations of health threats. These representations comprise identity (symptoms and labels), cause, timeline (acute, chronic, cyclical), consequences, and controllability. The multi-temporal architecture of the HPQ directly mirrors Leventhal’s framework: Current Health maps onto symptom identity and immediate status; Prior Health and Health Outlook capture the timeline and chronic disease expectations; Resistance to Illness reflects perceived immunological control; and Health Worry represents the parallel emotional processing of health threats.

Cognitive Appraisal and Psychometric Scale Theory

Ware incorporated psychometric tenets from Thurstone scaling, Likert summated ratings, and multi-trait scaling analysis. He addressed common response biases—notably acquiescence response set (the tendency to agree with statements regardless of content)—by deliberately constructing balanced subscales featuring equal proportions of favorably and unfavorably worded items. This balance ensures that latent traits represent genuine perceptual dimensions rather than artifactual response tendencies.

7. Validity

The validity of the Health Perceptions Questionnaire has been substantiated across decades of empirical testing in population-based, clinical, and health services research cohorts.

Construct and Convergent Validity

Extensive multitrait-multimethod correlation matrices conducted during the RAND Health Insurance Study established strong convergent validity. Current Health and the composite General Health Rating Index (GHRI) correlate substantially with objective physical functioning scales (such as limitations in vigorous activities, mobility, and self-care; correlations typically ranging from $r = .50$ to $.70$). Furthermore, HPQ scores correlate moderately to strongly with clinician-derived physical health assessments ($r = .40$ to $.60$) and physiological bio-markers, including systolic and diastolic blood pressure, functional lung capacity ($FEV_1$), and serum cholesterol levels.

The convergent validity of the Health Worry/Concern subscale is evidenced by its robust correlations ($r = .45$ to $.62$) with standardized measures of emotional well-being and psychological distress, such as the Mental Health Inventory (MHI) Anxiety and Depression subscales. Conversely, Resistance to Illness demonstrates specific negative associations with acute illness symptom checklists ($r = -.35$ to $-.48$).

Discriminant Validity

Multi-trait scaling criteria demonstrated that individual survey items load consistently higher on their hypothesized construct than on competing dimensions. For example, items measuring Resistance to Illness (such as “My body seems to resist illness very well”) correlate significantly higher with the resistance dimension than with general psychological distress or prior medical history, confirming discriminant validity. Additionally, while the HPQ correlates with mental health scales, the magnitude of association between Current Health and physical impairment ($r \approx .65$) is demonstrably higher than its association with neuroticism or depression ($r \approx .35$), confirming that respondents clearly distinguish physical vitality from emotional dysphoria.

Predictive and Criterion Validity

The predictive power of the HPQ is among its most celebrated attributes in public health. Longitudinal investigations spanning three- to five-year follow-up intervals demonstrated that baseline HPQ scores independently predicted:

• Healthcare Utilization: Low scores on Current Health and high scores on Health Worry forecasted elevated rates of ambulatory physician visits, emergency department visits, and inpatient hospital admissions, even after statistically controlling for baseline chronic diagnostic status and insurance coverage.
• Longitudinal Mortality: In landmark epidemiologic studies, baseline self-rated health on instruments derived from the HPQ significantly predicted all-cause mortality over 5- to 12-year follow-up periods. Individuals reporting poor current health and negative health outlooks exhibited relative risks of mortality 2 to 3 times higher than those with favorable perceptions, independent of smoking status, age, sex, and clinical disease markers.
• Disability and Occupational Absence: Scores on Rejection of the Sick Role and Sickness Orientation demonstrated criterion validity by predicting days lost from work and bed-disability days during acute respiratory and musculoskeletal episodes.

8. Reliability

The reliability of the Health Perceptions Questionnaire has been extensively documented across distinct demographic groups, socioeconomic strata, and clinical populations in the RAND Health Insurance Experiment.

Internal Consistency Reliability

Internal consistency was calculated using Cronbach’s coefficient alpha ($\alpha$) across multiple large-scale sampling waves. The table below outlines the empirical alpha coefficients observed in three primary validation cohorts reported by Ware and colleagues ($N = 1,790$, $N = 4,700$, and $N = 1,200$):

The 22-item composite General Health Rating Index exhibits an alpha of $.89$ in broad community cohorts, confirming that aggregate subjective health can be measured with high precision suitable for group-level policy comparisons. The lower reliability observed for short subscales (e.g., Sickness Orientation, $\alpha = .53 – .59$) is a direct statistical consequence of their brief length (2 to 4 items) and the complex, heterogeneous nature of behavioral illness attitudes. In disadvantaged or lower-literacy sub-samples (such as Cohort 3), alphas decreased moderately, emphasizing the necessity of standardized administration and clear respondent instructions.

Test-Retest Stability

Test-retest reliability assessments conducted over intervals ranging from one year to several years during the Health Insurance Experiment revealed remarkable stability. For Current Health and the GHRI, annual stability coefficients approached $r = .65$ to $.75$. Over a three-year period, stability coefficients remained above $r = .55$, confirming that while health perceptions fluctuate in response to acute biological shocks or onset of chronic conditions, they represent enduring psychological traits rather than fleeting mood states.

9. Factor Analysis

The structural dimensionality of the HPQ was thoroughly evaluated through both exploratory factor analysis (EFA) and confirmatory multi-trait scaling techniques during its original development and subsequent re-analyses.

Exploratory Factor Analyses

Early psychometric evaluations conducted by Ware (1976) utilized principal components analysis followed by orthogonal (Varimax) and oblique (Promax) rotations on correlation matrices derived from large community survey administrations. Initial unrotated factor extraction consistently revealed a dominant general factor accounting for approximately 35% to 45% of the total common variance, on which items assessing current health status, energy, and overall wellness loaded heavily. This strong first principal component provided empirical justification for constructing the unweighted aggregate General Health Rating Index (GHRI).

Subsequent rotations of components with eigenvalues exceeding 1.0 ($> 1.0$) isolated distinct, replicable factor groupings that aligned with hypothesized theoretical dimensions:

• Factor 1: Current Somatic Health: High positive loadings ($> .60$) for items expressing current wellness (“My health is excellent,” “I feel about as good now as I ever have”) and negative loadings for items indicating illness (“I am somewhat ill,” “I have been feeling bad lately”).
• Factor 2: Future Health Trajectory: Prominent loadings ($> .55$) from prospective items (“I will probably be sick a lot in the future,” “I expect to have a very healthy life”).
• Factor 3: Biological Resistance / Vulnerability: Marked by items addressing immune resilience (“My body seems to resist illness very well” loading positively, “When there is something going around, I usually catch it” loading negatively).
• Factor 4: Health Anxiety / Ruminative Concern: Marked by affective items assessing preoccupation with physical integrity (“I worry about my health more than other people worry about their health”).
• Factor 5: Retrospective Health History: Composed of items referencing childhood illness or historical severity (“I have never been seriously ill,” “I was so sick once I thought I might die”).

Multitrait Scaling Analysis

To verify that items formed valid summated rating scales without arbitrary weighting, Ware applied multi-trait scaling criteria: (1) Item-convergent validity was satisfied if an item correlated substantially with its hypothesized scale ($r ge .30$, corrected for item-scale overlap); (2) Item-discriminant validity was confirmed if an item correlated significantly higher with its hypothesized scale than with other subscales (known as scaling “successes”). In the primary adult cohorts, scaling successes exceeded 90% across the core constructs, demonstrating that the constructs are empirically distinguishable despite being moderately intercorrelated.

Confirmatory Factor Analysis (CFA) Fit

Modern structural equation modeling and CFA evaluations of the HPQ and its derivative scales (such as the Medical Outcomes Study General Health subscale) support a hierarchical model comprising six correlated first-order factors under a secondary overarching construct of General Health Perceptions. Standard goodness-of-fit parameters across modern community datasets typically yield acceptable fit:

• Comparative Fit Index (CFI): $\approx .92 – .95$
• Tucker-Lewis Index (TLI): $\approx .91 – .94$
• Root Mean Square Error of Approximation (RMSEA): $le .055$ ($90%\text{ CI: } .048 – .062$)
• Standardized Root Mean Square Residual (SRMR): $le .050$

10. Instrument / Measurement Tool

• Instrument Name: Health Perceptions Questionnaire (HPQ)
• Alternate / Derivative Names: General Health Rating Index (GHRI); Form B of the Health Insurance Study Health Status Survey
• Target Population: Non-institutionalized adults (ages 14 and older); parent-proxy versions exist for pediatric populations
• Administration Format: Self-administered paper-and-pencil questionnaire, interviewer-administered survey, or digital electronic assessment
• Total Item Count: 32 core standardized items (plus optional supplementary items evaluating health worry frequency over the past 3 months)
• Response Scale: 5-point Likert-type agreement scale:
  • 1 = Definitely True
  • 2 = Mostly True
  • 3 = Don’t Know
  • 4 = Mostly False
  • 5 = Definitely False

  (Alternative coding frames utilize 1 = Strongly Agree to 5 = Strongly Disagree)

• Subscale Composition:
  • Current Health: 9 items (Items 1, 4, 9, 12, 14, 18, 20, 24, 27)
  • Prior Health: 3 items (Items 11, 23, 31)
  • Health Outlook: 4 items (Items 5, 10, 16, 28)
  • Resistance to Illness: 4 items (Items 3, 7, 21, 30)
  • Health Worry/Concern: 4 items (Items 6, 17, 22, 29)
  • Sickness Orientation: 4 items (Items 2, 8, 13, 26)
  • Behavioral / Doctor Attitude Supplements: 4 items (Items 15, 19, 25, 32 assessing sick-role rejection and medical consultation attitudes)
  • General Health Rating Index (GHRI Composite): 22 selected items encompassing Current Health, Prior Health, Outlook, and Resistance
• Scoring and Transformation Rules:
  • Directionality: Items are scored so that higher numerical values reflect more positive health perceptions or higher levels of the named construct.
  • Reverse Scoring: Items expressing poor health, elevated disease vulnerability, or high worry must be reverse-coded prior to subscale summation. When using the original 1 to 5 scale where 1 represents “Definitely True,” recoding transforms raw responses such that favorable statements receive a value of 5 and unfavorable statements receive a value of 1.
  • Linear Standardization: Raw subscale scores can be converted to a 0–100 scale using the standard transformation formula:
    
    $$\text{Transformed Score} = \left( \frac{\text{Actual Raw Score} – \text{Lowest Possible Raw Score}}{\text{Highest Possible Raw Score} – \text{Lowest Possible Raw Score}} \right) \times 100$$

11. Permissions & Fee and Test Year

• Publication Year: The instrument was formally introduced and published by John E. Ware, Jr. in 1976 in Health Services Research, with exhaustive technical monographs published by the RAND Corporation between 1978 and 1981.
• Copyright & Intellectual Property: The HPQ was developed under contract with the United States Department of Health, Education, and Welfare (Grant No. 016B-7901) by the RAND Corporation. Under federal funding terms, research instruments produced directly within the RAND Health Insurance Experiment reside in the public domain.
• Permissions & Fees: The original 32-item Health Perceptions Questionnaire is available for academic, clinical, and non-commercial research use free of charge without royalty or licensing fees. Researchers do not require written permission from the RAND Corporation for standard non-commercial academic research administrations. Commercial vendors integrating the instrument into proprietary commercial software suites should consult the RAND Corporation and provide appropriate bibliographic attribution.
• Historical Lineage: Concepts and item formulations from the HPQ were subsequently refined by Ware and colleagues into the Medical Outcomes Study (MOS) health measures, culminating in the copyrighted SF-36® and SF-12® instruments managed by QualityMetric Inc. and Optum. The original HPQ remains an open, non-proprietary instrument.

12. References

• Brook, R. H., Ware, J. E., Jr., Davies-Avery, A., Stewart, A. L., Donald, C. A., Rogers, W. H., Williams, K. N., & Johnston, S. A. (1979). Overview of adult health status measures fielded in RAND’s Health Insurance Study. Medical Care, 17(7 Suppl), 1–131. https://www.jstor.org/stable/3763784
• Davies, A. R., & Ware, J. E., Jr. (1981). Measuring health perceptions in the Health Insurance Experiment (Publication No. R-2711-HHS). Santa Monica, CA: RAND Corporation. https://www.rand.org/pubs/reports/R2711.html
• Eisen, M., Ware, J. E., Jr., Donald, C. A., & Brook, R. H. (1979). Measuring components of children’s health status. Medical Care, 17(9), 902–921. https://doi.org/10.1097/00005650-197909000-00003
• 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
• Stewart, A. L., Hays, R. D., & Ware, J. E., Jr. (1992). Health perceptions, energy/fatigue, and health distress measures. In A. L. Stewart & J. E. Ware, Jr. (Eds.), Measuring functioning and well-being: The Medical Outcomes Study approach (pp. 143–172). Duke University Press.
• Ware, J. E., Jr. (1976). Scales for measuring general health perceptions. Health Services Research, 11(4), 396–415. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1071941/
• Ware, J. E., Jr., Davies-Avery, A., & Donald, C. A. (1978). Conceptualization and measurement of health for adults in the Health Insurance Study: Vol. V, General health perceptions (Publication No. R-1987/5-HEW). Santa Monica, CA: RAND Corporation. https://www.rand.org/pubs/reports/R1987.5.html
• Ware, J. E., Jr., Manning, W. G., Jr., Duan, N., Wells, K. B., & Newhouse, J. P. (1984). Health status and the use of ambulatory mental health services. American Psychologist, 39(10), 1090–1100. https://doi.org/10.1037/0003-066X.39.10.1090

13. Items of the Scale (Questionnaire)