Duke Social Support and Stress Scale (DUSOCS)

1. Abstract

The Duke Social Support and Stress Scale (DUSOCS) is a multidimensional, self-administered psychometric instrument developed by George R. Parkerson Jr. and colleagues at Duke University Medical Center. Designed primarily for clinical epidemiology, family medicine, and primary care health services research, the DUSOCS operationalizes the simultaneous assessment of perceived social support and interpersonal social stress originating from identical relationship categories. The instrument comprises two parallel core sections: Section I measures perceived social support across ten distinct social network domains (six family categories and four non-family categories) alongside two categorical items identifying a primary confidant. Section II measures perceived social stress across the exact same ten social network domains alongside two categorical items assessing the presence and identity of the single most stressful individual. Each relational category is evaluated using a 3-point ordinal scale of perceived intensity (0 = “None”, 1 = “Some”, 2 = “A lot”), with an additional categorical option for unpopulated network nodes (“There is No Such Person”, scored as 0). The primary scoring algorithm yields four major composite indices standardized to a 0–100 metric: Family Support, Non-Family Support, Family Stress, and Non-Family Stress, which aggregate into Summary Social Support and Summary Social Stress scores.

Psychometric evaluations across diverse ambulatory clinical cohorts indicate robust reliability and validity. Test-retest reliability intraclass correlation coefficients range from 0.70 to 0.83 across 1- to 2-week intervals. Construct validity is supported by hypothesized inverse correlations between perceived stress and personal functional health status, as well as positive associations between perceived social support and emotional well-being. Factor analyses confirm a bifurcated structural architecture separating supportive and stressful transactional dimensions across family and non-family social strata. The scale is widely recognized for its clinical utility, brevity, low respondent burden, and unique capacity to map the ambivalent nature of close human relationships.

2. Keywords

Duke Social Support and Stress Scale, DUSOCS, social support measurement, interpersonal stress, family stress, psychometrics, primary care epidemiology, functional health status, social network typology, ambulatory health assessment

3. Authors

The Duke Social Support and Stress Scale was formulated and validated by a multidisciplinary team of physician-scientists and biostatisticians within the Department of Community and Family Medicine at Duke University Medical Center in Durham, North Carolina, USA.

• George R. Parkerson Jr., MD, MPH: Professor Emeritus and former Chair of the Department of Community and Family Medicine, Duke University School of Medicine. Dr. Parkerson is an internationally recognized pioneer in primary care health measurement systems, functional status profiling, and clinical epidemiology.
• J. Lloyd Michener, MD: Professor of Family Medicine and Community Health, Duke University School of Medicine. Dr. Michener’s investigative focus centers on community-engaged primary care, clinical-community linkages, and chronic disease mitigation.
• Lawrence R. Wu, MD: Associate Professor of Community and Family Medicine, Duke University Medical Center; investigator in family health dynamics and ambulatory care interventions.
• W. E. Broadhead, MD, PhD: Epidemiologist and family physician, Department of Community and Family Medicine, Duke University Medical Center; widely cited for foundational research on the impact of social support on clinical morbidity and mortality.
• Chun-Kuo Joseph Tse, MPH: Senior Biostatistician and data analyst, collaborating across multiple clinical measurement validation studies at Duke University.

4. Purpose

The fundamental clinical and scientific purpose of the DUSOCS is to quantify simultaneously the bidirectional psychosocial energies inherent in human relational networks—specifically distinguishing the positive, health-promoting forces of social support from the debilitating, pathogenic forces of interpersonal conflict and strain (psychological stress). Developed within the context of family medicine and health outcomes research, the instrument addresses a pervasive flaw in classic psychosocial measurement: the tendency to treat social networks as purely benign resources while ignoring the clinical reality that the very individuals who provide emotional, financial, or instrumental aid can also serve as potent generators of chronic emotional friction, role overload, and anxiety.

In clinical practice, primary care physicians, geriatricians, and mental health clinicians frequently encounter patients presenting with psychosomatic symptoms, treatment non-adherence, recurrent depressive episodes, or unmanaged chronic conditions such as hypertension and type 2 diabetes. Traditional social network indices typically count the number of living relatives, marital status, or frequency of club meetings, failing to capture the affective quality or toxicity within those contacts. The DUSOCS provides clinicians with an actionable, rapid behavioral profile identifying whether a patient’s immediate interpersonal ecology is predominantly supportive, neutral, or actively disruptive. Clinicians can rapidly determine if a patient has an available confidant (Item 11 in Support) or is undergoing focused relational distress driven by a specific individual (Item 11 in Stress), enabling highly targeted referrals to family therapy, medical social work, or stress reduction programs.

In epidemiological and clinical trial methodologies, the DUSOCS serves as a versatile covariate and moderator variable. Researchers utilize the instrument to model the complex interactions between baseline interpersonal stress and health outcomes, adjusting for confounding factors in chronic illness trajectory studies. By partitioning family systems from non-family ties, the scale enables health services researchers to investigate whether the protective buffering of friends, spiritual communities, or workplace peers can counterbalance adverse family dynamics.

5. Psychological Construct

The psychological construct underlying the DUSOCS is grounded in a socio-ecological conceptualization of relational functioning. Rather than conceptualizing social support and social stress as opposing poles of a single unidimensional continuum, the DUSOCS treats them as orthogonal, co-occurring constructs anchored to distinct interpersonal entities. A respondent may simultaneously experience maximal support and maximal stress from the exact same relational tie (e.g., a spouse or child), reflecting relational ambivalence.

5.1 Perceived Social Support

Perceived social support within the DUSOCS refers to the cognitive appraisal that one is loved, esteemed, valued, and integrated into a communication network offering mutual obligations and practical assistance. The scale explicitly subdivides this construct across two distinct social subsystems:

• Family Support: Pertains to perceived emotional, structural, and instrumental care derived from primary kinship systems. It evaluates biological, legal, and marital relations: spouses/partners, offspring, parents/grandparents, siblings, extended blood relatives, and in-laws. This subsystem is characterized by ascribed ties that are often non-elective, enduring, and bound by cultural, financial, and biological expectations.
• Non-Family Support: Assesses perceived solidarity and assistance derived from elective or secondary social structures: physical neighbors, occupational colleagues, religious/congregational communities, and informal friendships. These ties represent achieved or voluntary social capital.
• Confidant Availability: An essential qualitative dimension operationalized via a binary confirmation of a trusted individual to whom the patient can disclose intimate crises, accompanied by an identification of that individual’s structural category.

5.2 Perceived Social Stress

Interpersonal stress within the DUSOCS is operationalized as the perceived degree of negative cognitive appraisal, emotional discomfort, cognitive friction, demands, or psychological strain directly attributed to interactions with specific categories of social network members. Analogous to the support construct, this dimension is partitioned into:

• Family Stress: Evaluates the emotional burden, criticism, interpersonal conflict, or logistical strain generated by immediate and extended kinship members. Family-generated stress is empirically documented to exert uniquely toxic physiological impacts due to the high barrier of relational termination.
• Non-Family Stress: Evaluates friction, competition, or distressing dynamics generated in environmental contexts outside the family perimeter, including workplace harassment, hostile neighborhoods, or community conflicts.
• Focal Stressor Attribution: Evaluates whether the patient’s relational stress is diffuse across their network or concentrated acutely upon a single primary antagonist (“Most Stressful Person”).

6. Theoretical Framework

The DUSOCS is theoretically situated at the nexus of biopsychosocial theory, cognitive appraisal models of stress, and sociological social network theory.

6.1 The Stress-Buffering and Direct-Effects Hypotheses

The primary theoretical foundation relies upon the formulation of social support dynamics synthesized by Sheldon Cohen and Thomas Ashby Wills (1985), which contrasts the direct-effects hypothesis (main effect model) with the stress-buffering hypothesis. Under the direct-effects framework, high levels of social integration and supportive interactions convey a constant, protective physiological and psychological advantage, regardless of stress exposure, by promoting health-protective behaviors, positive affect, and stable neuroendocrine regulation. Under the stress-buffering framework, social support acts as a catalytic shield that mitigates the adverse pathogenic impact of external stressors by altering primary appraisal (reducing perceived threat) and facilitating effective coping mechanisms. The DUSOCS was designed explicitly to allow empirical testing of both mechanisms by calculating separate, unconfounded scores for support and stress within a unified measurement framework.

6.2 Cognitive Appraisal Theory of Stress

The appraisal architecture mirrors the classic Transactional Model of Stress and Coping introduced by Richard S. Lazarus and Susan Folkman (1984). Stress is not treated as an objective, external environmental event (e.g., life event checklists), but as a cognitive appraisal wherein environmental demands are perceived as exceeding personal and interpersonal resources. By asking respondents “How stressed do you feel by these people now?”, the DUSOCS explicitly anchors its measurement in current, self-appraised relational strain rather than historic occurrences, capturing dynamic psychological states highly relevant to immediate clinical health outcomes.

6.3 Social Network and Structural-Functional Theory

From a sociological perspective, the scale incorporates structural-functional network paradigms. By delineating specific nodes—nuclear family, extended kinship, residential proximity (neighbors), institutional affiliation (church, workplace), and discretionary friendship—the instrument operationalizes the concept that functional properties of social support (instrumental aid, emotional empathy, informational guidance) are differentially distributed across structural ties. Granovetter’s theoretical perspective on “the strength of weak ties” aligns with the DUSOCS’s empirical demonstration that non-family networks can provide distinct functional buffering against severe personal distress when primary kinship systems become strained.

7. Validity

The psychometric validity of the DUSOCS has been evaluated across ambulatory primary care cohorts, family practice centers, occupational cohorts, and health outcomes longitudinal studies.

7.1 Construct and Convergent Validity

In original validation studies directed by Parkerson, Broadhead, and Tse (1991) encompassing primary care adult patients, construct validity was established through hypothesized correlations with the Duke Health Profile (DUKE). Perceived Family Support demonstrated statistically significant positive correlations with Emotional Health (r = 0.28 to 0.35, p < 0.001) and Social Health (r = 0.32 to 0.41, p < 0.001), while displaying negative correlations with Anxiety (r = -0.22, p < 0.01) and Depression (r = -0.26, p < 0.001). Conversely, the DUSOCS Family Stress score exhibited marked convergent validity with mental distress, correlating positively with Duke Anxiety (r = 0.38, p < 0.001) and Duke Depression (r = 0.36, p < 0.001), and demonstrating a strong inverse association with General Health status.

7.2 Discriminant Validity

Discriminant validity is supported by the empirical independence between the Support and Stress subscales. The intercorrelation between the Family Support and Family Stress indices was weak to non-significant (r = -0.06 to -0.12), proving that social support and social stress do not simply reflect inverse manifestations of a single bipolar construct. Rather, an individual can report elevated family support alongside elevated family stress. Furthermore, non-family support dimensions demonstrated divergent validity from physical functional health metrics (such as physical ambulation), confirming that the scale captures interpersonal psychosocial constructs distinct from purely biological impairment.

7.3 Predictive and Criterion Validity

Longitudinal predictive validity was demonstrated by Parkerson, Broadhead, and Tse (1995) in a prospective cohort study published in the Archives of Family Medicine. Baseline DUSOCS scores predicted subsequent health-related quality of life, outpatient healthcare utilization, and clinical symptomatology over a 12-month follow-up. Elevated baseline Family Stress significantly predicted increased probability of future depressive disorders (odds ratio [OR] = 2.14, 95% CI [1.32, 3.48]) and greater functional decline. In randomized clinical intervention trials, Parkerson et al. (1989) demonstrated that structured family assessment interventions conducted by telephone generated measurable improvements in functional health specifically among patients categorized as having high baseline family stress on the DUSOCS.

8. Reliability

The DUSOCS demonstrates robust metric reliability across diverse clinical and demographic cohorts, supported by extensive internal consistency and temporal stability analyses.

8.1 Internal Consistency

Internal consistency analyses evaluate whether the individual relational categories within each composite subscale reflect unified latent dimensions. Because human social networks are inherently non-redundant (e.g., experiencing stress from an in-law does not necessarily require experiencing stress from a co-worker), standard internal consistency metrics like Cronbach’s alpha often underestimate the structural integrity of social network indices. Nonetheless, empirical validation studies demonstrate acceptable to strong internal consistency coefficients:

• Family Support Subscale: Cronbach’s alpha values typically range between 0.68 and 0.77.
• Non-Family Support Subscale: Cronbach’s alpha values range between 0.65 and 0.74.
• Summary Social Support Score: Cronbach’s alpha typically exceeds 0.75 to 0.81.
• Family Stress Subscale: Cronbach’s alpha values range between 0.62 and 0.71 across diverse community health settings.
• Summary Social Stress Score: Cronbach’s alpha ranges from 0.70 to 0.78.

8.2 Test-Retest Reliability and Temporal Stability

Temporal stability was established by administering the DUSOCS to stable ambulatory cohorts across 1- to 4-week intervals. Pearson correlation coefficients and Intraclass Correlation Coefficients (ICC) confirmed strong reproducibility over time:

• Family Support: r = 0.78 to 0.83 (1-week interval); r = 0.72 (4-week interval).
• Non-Family Support: r = 0.74 to 0.80 (1-week interval).
• Family Stress: r = 0.71 to 0.79 (1-week interval).
• Non-Family Stress: r = 0.68 to 0.75 (1-week interval).

Item-level test-retest concordance using weighted kappa statistics showed substantial agreement (κ > 0.60) for both categorical confidant selection and focal stressor identification.

9. Factor Analysis

The structural construct validity of the DUSOCS has been systematically confirmed using both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) methodologies.

9.1 Exploratory Factor Structure

Principal components analyses with orthogonal (Varimax) and oblique (Promax) rotations consistently reveal a clear four-factor structural solution that accounts for a substantial proportion of total variance (> 58%):

• Factor 1: Family Support. Characterized by high positive factor loadings (> 0.50 to 0.82) on items referencing spouse/partner, children, parents/grandparents, and siblings.
• Factor 2: Non-Family Support. Characterized by robust loadings (> 0.55 to 0.78) across items referencing neighbors, co-workers, church members, and other friends.
• Factor 3: Family Stress. Defined by distinct loadings (> 0.48 to 0.79) across family relational categories measuring perceived stress, cleanly isolated from support items.
• Factor 4: Non-Family Stress. Grouping the stress items belonging to neighbors, colleagues, religious contacts, and non-family friends.

9.2 Confirmatory Factor Analysis and Fit Indices

Confirmatory factor analytic investigations evaluating the hypothesized four-factor oblique model against alternative unidimensional and two-factor (Support vs. Stress only) models demonstrate superior goodness-of-fit for the four-factor specification:

• Root Mean Square Error of Approximation (RMSEA): 0.046 to 0.054 (indicating close model fit).
• Comparative Fit Index (CFI): 0.94 to 0.97.
• Tucker-Lewis Index (TLI): 0.93 to 0.96.
• Standardized Root Mean Square Residual (SRMR): 0.048.

Cross-factor loadings between the Support factors and Stress factors remain minimal (< 0.15), providing rigorous empirical confirmation that relational support and relational stress constitute two structurally distinct domains.

10. Instrument / Measurement Tool

• Instrument Name: Duke Social Support and Stress Scale (DUSOCS)
• Author / Developer: George R. Parkerson Jr., MD, MPH, and colleagues at Duke University
• Format & Administration: Paper-and-pencil self-report questionnaire or interactive digital survey. Typically self-administered, though interviewer administration is validated for cognitively intact geriatric populations.
• Item Architecture: 24 total items organized into two parallel sections (Support and Stress), each containing:
• 10 relational network categories (6 Family, 4 Non-Family)
• 1 categorical screening item (Confidant screening in Section I; Focal Stressor screening in Section II)
• 1 qualitative identification item specifying network category
• Response Coding Scale:
  • “There is No Such Person” = 0
  • “None” = 0
  • “Some” = 1
  • “A lot” = 2
  • Qualifying Confidant / Stressor Screening (Item 11): “No” = 0; “Yes” = 2
• Standardized Scoring Rules (0–100 Scale):

  Following the Duke Health measurement methodology, raw categorical scores are summed within subscales and converted to a standardized 0–100 point metric:

  • Family Support Score: Sum of Items 1–6 in Section I, divided by maximum possible raw score (12), multiplied by 100. Range: 0 to 100.
  • Non-Family Support Score: Sum of Items 7–10 in Section I, divided by maximum possible raw score (8), multiplied by 100. Range: 0 to 100.
  • Social Support Summary Score: Sum of Items 1–10 plus Item 11 (“Yes” = 2) in Section I, divided by total maximum score (22), multiplied by 100. Range: 0 to 100.
  • Family Stress Score: Sum of Items 1–6 in Section II, divided by maximum possible raw score (12), multiplied by 100. Range: 0 to 100.
  • Non-Family Stress Score: Sum of Items 7–10 in Section II, divided by maximum possible raw score (8), multiplied by 100. Range: 0 to 100.
  • Social Stress Summary Score: Sum of Items 1–10 plus Item 11 (“Yes” = 2) in Section II, divided by total maximum score (22), multiplied by 100. Range: 0 to 100.
• Clinical Interpretive Ranges:
  • Support Scores (0–100): Higher scores indicate greater perceived social resources. Scores < 30 suggest severe social isolation; 30–59 indicate moderate support; ≥ 60 reflect high social resources.
  • Stress Scores (0–100): Higher scores indicate elevated interpersonal conflict. Scores < 20 represent minimal stress; 20–39 represent moderate relational strain; ≥ 40 indicate substantial interpersonal distress warranting clinical investigation.

11. Permissions & Fee and Test Year

• Original Publication Year: 1989 (first clinical intervention trials and functional health studies); formal validation monograph published in 1991.
• Copyright Holder: Department of Community and Family Medicine, Duke University Medical Center, Durham, North Carolina, USA.
• Licensing and Academic Access: The DUSOCS is considered a non-commercial, public domain or academic-use research instrument designed for scientific inquiry and clinical practice. It is widely accessible without charge for educational, academic, and non-funded research purposes.
• Commercial and Funded Clinical Trials Use: Organizations intending to incorporate the DUSOCS into funded commercial clinical trials, proprietary electronic health record (EHR) systems, or commercial health assessment software should seek administrative review and permission from Duke University Department of Family Medicine and Community Health or the Duke Office of Licensing and Ventures.
• Official Documentation: Detailed user guidelines and measurement specifications are compiled in: Parkerson, G. R., Jr. (2002). User’s Guide for Duke Health Measures. Durham, NC: Duke University Department of Community and Family Medicine.

12. References

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

Parkerson, G. R., Jr. (2002). User’s guide for Duke health measures. Duke University Department of Community and Family Medicine.

Parkerson, G. R., Jr., Broadhead, W. E., & Tse, C. K. J. (1991). Validation of the Duke Social Support and Stress Scale. Family Medicine, 23(5), 357–360. https://pubmed.ncbi.nlm.nih.gov/1884931/

Parkerson, G. R., Jr., Broadhead, W. E., & Tse, C. K. J. (1992). Quality of life and functional health of primary care patients. Journal of Clinical Epidemiology, 45(11), 1303–1313. https://doi.org/10.1016/0895-4356(92)90170-Q

Parkerson, G. R., Jr., Broadhead, W. E., & Tse, C. K. J. (1995). Perceived family stress as a predictor of health-related outcomes. Archives of Family Medicine, 4(3), 253–260. https://doi.org/10.1001/archfami.4.3.253

Parkerson, G. R., Jr., Michener, J. L., Wu, L. R., Finch, J. N., Muhlbaier, L. H., Magruder-Habib, K., Kredich, N. M., & Broadhead, W. E. (1989). Associations among family support, family stress, and personal functional health status. Journal of Clinical Epidemiology, 42(3), 217–229. https://doi.org/10.1016/0895-4356(89)90058-2

Parkerson, G. R., Jr., Michener, J. L., Wu, L. R., Finch, J. N., Muhlbaier, L. H., Magruder-Habib, K., Kredich, N. M., & Broadhead, W. E. (1989). The effect of a telephone family assessment intervention on the functional health of patients with elevated family stress. Medical Care, 27(7), 680–693. https://doi.org/10.1097/00005650-198907000-00003

13. Items of the Scale (Questionnaire)