Specific Activity Scale

1. Abstract

The Specific Activity Scale (SAS), originally formulated by Lee Goldman and colleagues in 1981, is a standardized, clinician- or self-administered functional classification instrument designed to evaluate functional capacity and cardiovascular reserve in patients with cardiac disease. Developed to overcome the pronounced inter-observer variability and subjective ambiguity inherent in the traditional New York Heart Association (NYHA) Functional Classification, the SAS operates as a structured, algorithmic decision tree comprising four primary activity thresholds calibrated against physiological energy expenditures measured in Metabolic Equivalents of Task (METs). Each question delineates specific daily activities requiring known oxygen consumption levels: vigorous physical activity (≥7 METs), moderate domestic and occupational tasks (≥5 METs), light exertion (≥2 METs), and basic self-care activities (<2 METs). Based on dichotomous responses (Yes / No) to these graded functional demands, patients are reliably stratified into four discrete functional classes (Class I through Class IV). Psychometric evaluations demonstrate that the SAS possesses superior reproducibility compared to the NYHA classification, yielding an inter-rater concordance rate of 73% to 81% (weighted kappa κ = 0.68–0.74) versus 56% (weighted kappa κ = 0.44–0.47) for unstructured clinical evaluations. Furthermore, the SAS exhibits robust criterion-related validity against treadmill exercise testing (such as the Bruce protocol), exercise duration, peak oxygen uptake (VO₂ peak), and invasive hemodynamic parameters. The instrument has been translated and adapted across numerous clinical settings, including cardiac rehabilitation guidelines such as the Dutch Royal Association for Physiotherapy (KNGF) guidelines. With minimal administrative burden, zero financial cost, and clear empirical anchors, the Specific Activity Scale remains an essential tool in clinical cardiology, cardiovascular epidemiology, surgical risk stratification, and cardiac rehabilitation.

2. Keywords

Specific Activity Scale, Goldman functional classification, New York Heart Association, functional capacity, metabolic equivalent of task, cardiac rehabilitation, cardiovascular risk stratification, psychometrics, exercise tolerance, patient-reported outcomes

3. Authors

The Specific Activity Scale was devised and validated by a multidisciplinary team of physician-scientists and biostatisticians at Harvard Medical School and Brigham and Women’s Hospital in Boston, Massachusetts:

• Lee Goldman, MD, MPH: Professor of Medicine and Epidemiology; former Chair of the Department of Medicine at the University of California, San Francisco (UCSF); former Dean of the Faculties of Health Sciences and Medicine at Columbia University Irving Medical Center. A pioneer in clinical epidemiology, cardiovascular risk prediction (notably the Goldman Cardiac Risk Index), and health services research.
• Barton Hashimoto, MD: Division of General Medicine and Primary Care, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA.
• E. Francis Cook, ScD: Professor of Epidemiology, Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA. Specialized in multivariate biostatistical modeling and prognostic algorithm design.
• Alexander Loscalzo, MD: Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA.
• Dutch Adaptation Contributors (2005): Koninklijk Nederlands Genootschap voor Fysiotherapie (KNGF), Working Group on Cardiac Rehabilitation (KNGF-richtlijn Hartrevalidatie).

4. Purpose

The primary clinical and psychometric purpose of the Specific Activity Scale is to provide an objective, reproducible, and standardized categorization of physical functional status in patients with chronic cardiovascular and pulmonary conditions. For decades, clinicians relied nearly exclusively on the functional classification schema established by the New York Heart Association. Although the NYHA criteria (Classes I through IV) were ubiquitous, empirical investigations repeatedly revealed that the NYHA framework suffered from significant inter-observer variability. In classic multi-rater trials, two independent cardiologists evaluating the same cohort of patients agreed on the exact NYHA class only 50% to 60% of the time. This divergence stemmed largely from the subjective phrasing of the NYHA definitions, which depend on ambiguous qualifying terminology such as “slight limitation of physical activity” (Class II) versus “marked limitation of physical activity” (Class III).

Goldman and his colleagues recognized that subjective clinical impressions lacked empirical calibration. A patient who experiences dyspnea when walking uphill at an aggressive pace might be labeled Class II by an aggressive clinician or Class I by a conservative one; conversely, a sedentary patient might experience no angina simply because they restrict their lifestyle, leading to an erroneous assignment to Class I despite profound underlying cardiovascular compromise. The Specific Activity Scale was explicitly engineered to address this operational gap by replacing abstract qualitative descriptions with concrete, empirically verifiable behavioral benchmarks grounded in exercise physiology. By querying specific real-world tasks with established metabolic expenditures (ranging from carrying groceries up stairs, to painting walls, to dressing without pauses), the SAS directly assesses physiological capacity rather than unanchored symptom perception.

In clinical practice, the SAS serves several critical functions. First, it enables precise baseline staging of disease severity in patients with coronary artery disease, heart failure, valvular heart disease, or post-myocardial infarction syndrome. Second, it functions as a highly responsive outcome measure in outpatient cardiac rehabilitation programs, tracking longitudinal improvements in exercise tolerance resulting from aerobic training, pharmacotherapy (e.g., beta-blockers, ACE inhibitors, SGLT2 inhibitors), or revascularization procedures. Third, the SAS is widely utilized in preoperative cardiovascular risk assessment, serving as a reliable surrogate for estimated MET capacity in major noncardiac surgery. In clinical epidemiology and health services research, the instrument offers a low-cost, patient-centered endpoint that can be administered in less than two minutes via self-report or interview, circumventing the logistical constraints, contraindications, and financial expenses associated with formal cardiopulmonary exercise testing (CPET).

5. Psychological Construct

The Specific Activity Scale measures the construct of perceived functional capacity, situated at the intersection of cardiovascular pathophysiology, exercise physiology, and behavioral psychometrics. Rather than measuring subjective psychological distress or global health-related quality of life (HRQoL), the instrument captures an individual’s behavioral threshold for symptom-limited aerobic performance during activities of daily living (ADL).

Functional capacity, in this context, represents an individual’s operational reserve: the maximum metabolic demand an individual can sustain without experiencing limiting symptoms of cardiovascular decompensation, such as myocardial ischemia (angina pectoris), critical reduction in cardiac output (fatigue, syncope), or elevated pulmonary capillary wedge pressure (dyspnea). To standardize this biological construct across heterogenous patient lifestyles, the scale establishes four descending functional tiers, mirroring the physiological hierarchy of human motor activities:

• Tier 1: Vigorous Physical Exertion (≥7 METs): This domain represents superior cardiovascular and musculoskeletal performance, corresponding to strenuous leisure or occupational pursuits. Tasks include carrying a 24-pound load up eight steps, carrying 55 pounds on level ground, moving heavy furniture, shoveling wet snow, jogging or walking briskly at 5 miles per hour, or participating in competitive sports (e.g., basketball, touch football, squash). Patients capable of performing these tasks without symptom limitation possess intact cardiac reserve and normal or near-normal functional aerobic capacity (Class I).
• Tier 2: Moderate Domestic and Occupational Activities (≥5 METs): This domain reflects normal daily adult functioning outside of strenuous athletic exertion. Activities encompass carrying a 24-pound grocery bag on flat ground, raking leaves, performing moderate garden work, painting walls, operating a power lawnmower, walking at a brisk pace of 4 miles per hour, making beds, or engaging in sexual intercourse without interruption. Successful completion signifies modest functional impairment; patients are functionally independent and manage standard household maintenance without pathological distress (Class II).
• Tier 3: Light Household and Instrumental Activities (≥2 METs): This domain assesses lower-order instrumental activities of daily living (IADL). It includes walking on flat surfaces at an easy pace (2.0 to 2.5 miles per hour), taking a continuous shower without pauses, dressing without resting, performing light dusting or dishwashing, and cleaning interior windows. The inability to clear higher tiers combined with success at this level reflects marked cardiovascular compromise, where standard activities cause disproportionate fatigue, palpitation, or dyspnea (Class III).
• Tier 4: Basic Self-Care and Basal Mobility (<2 METs): This domain captures minimal functional survival requirements, equivalent to basic activities of daily living (BADL). It assesses whether the individual can walk across a room without severe distress or dress themselves without requiring frequent rest periods. Failure to execute these basic personal care activities indicates severe, end-stage functional incapacitation, wherein symptoms occur even at rest or during trivial physical exertion (Class IV).

By conceptualizing functional capacity through this ordered, cumulative behavioral hierarchy, the SAS operationalizes the psychological construct as an unidimensional Guttman scale: an individual who possesses the capacity to clear a high-demand metabolic hurdle (≥7 METs) is mathematically and physiologically presumed to possess the reserve necessary to clear all subordinate hurdles (≥5 METs, ≥2 METs, and <2 METs).

6. Theoretical Framework

The theoretical framework underlying the Specific Activity Scale is rooted in cardiovascular exercise physiology, specifically the concept of the Metabolic Equivalent of Task (MET), integrated with algorithmic clinical decision theory and Guttman scalogram analysis. A single MET is conventionally defined as the resting metabolic rate, quantitatively standardized as the consumption of 3.5 milliliters of oxygen per kilogram of body mass per minute (3.5 mL O₂·kg⁻¹·min⁻¹) in an average seated adult.

In classical cardiology, the degree of ventricular dysfunction and valvular incompetence manifests as a physiological ceiling on cardiac output (Fick principle: VO₂ = Cardiac Output × ΔC(a-v)O₂). As cardiovascular pathology advances, stroke volume reserve becomes blunted, and the left ventricular end-diastolic pressure rises steeply during exertion, precipitating retrograde pulmonary congestion and exertional dyspnea. Concurrently, insufficient forward perfusion triggers peripheral skeletal muscle hypoperfusion, anaerobic glycolysis, lactic acidosis, and early physical exhaustion. Consequently, there is an inexorable biological correlation between the anatomical and hemodynamic severity of cardiac impairment and the peak METs an individual can attain prior to the onset of limiting symptoms.

Goldman and colleagues recognized that although physiological laboratory testing using motorized treadmills or cycle ergometers directly quantifies peak oxygen consumption, such testing is not always feasible for routine outpatient monitoring or large-scale clinical trials. Conversely, an individual’s domestic environment serves as an informal continuous exercise laboratory. Every physical task encountered in routine life can be assigned an approximate MET value based on comprehensive exercise physiology compendiums (such as those later consolidated in the Ainsworth Compendium of Physical Activities). For example, walking at 2 mph requires approximately 2 to 2.5 METs; mowing the lawn with a power mower requires roughly 5 to 5.5 METs; and shoveling snow or running at 5 mph demands 7 to 8 or more METs.

Psychometrically, the SAS synthesizes these physiological realities through a deterministic algorithmic structure. Traditional questionnaires often employ Likert scales that summate responses into an aggregate score. However, summation models can obscure critical clinical thresholds (e.g., a patient scoring moderately on multiple items might actually be completely housebound). In contrast, the SAS employs a deterministic decision-tree logic derived from Guttman scaling models:

1. Can the patient achieve ≥7 METs? If Yes, functional reserve is normal/near-normal → Class I.
2. If No, can the patient achieve ≥5 METs? If Yes, moderate functional reserve → Class II.
3. If No, can the patient achieve ≥2 METs? If Yes, marked impairment → Class III.
4. If No, the patient is unable to sustain even basal expenditure → Class IV.

This deterministic branching model ensures that the measurement tool reflects the monotonic biological decay of aerobic reserve seen in progressive cardiopulmonary diseases.

7. Validity

The psychometric validity of the Specific Activity Scale has been extensively evaluated across clinical cardiology, cardiac surgery, and rehabilitation medicine.

Criterion-Related and Concurrent Validity

In the original landmark validation study conducted by Goldman et al. (1981), 94 consecutive patients undergoing diagnostic maximal exercise treadmill testing using the Bruce protocol were independently evaluated using both the SAS and the unstructured NYHA functional classification. Criterion validity was established by correlating assigned functional classes with objectively measured exercise parameters, including total treadmill duration, peak METs achieved, and peak heart rate. The SAS demonstrated a statistically significant, strong correlation with objective treadmill performance:

• Patients classified as SAS Class I achieved a mean treadmill duration of 9.0 ± 2.4 minutes (corresponding to approximately 9.3 ± 2.5 METs).
• SAS Class II patients achieved a mean duration of 6.7 ± 2.0 minutes (approximately 6.9 ± 2.1 METs).
• SAS Class III patients achieved a mean duration of 4.9 ± 1.8 minutes (approximately 5.1 ± 1.8 METs).
• SAS Class IV patients achieved a mean duration of 2.7 ± 1.5 minutes (approximately 2.8 ± 1.6 METs).

The separation between each successive SAS class in terms of treadmill exercise duration was highly significant (p < 0.001 across classes). In direct head-to-head comparisons, the SAS correlated more strongly with treadmill capacity (Spearman rank correlation r = 0.58 to 0.68) than did the standard NYHA classification (r = 0.45 to 0.51). When evaluating specific exercise cutoffs (such as the ability to exceed 7 METs or failure to exceed 5 METs), the SAS demonstrated sensitivity exceeding 82% and specificity exceeding 78%.

Construct and Convergent Validity

Construct validity has been reaffirmed in subsequent investigations comparing the SAS against other validated functional scales, such as the Duke Activity Status Index (DASI), the SF-36 Physical Functioning Subscale (PF-10), and the 6-Minute Walk Test (6MWT). In a cohort of outpatients with chronic stable heart failure, SAS functional classes exhibited a strong inverse correlation with 6-minute walk distances (r = −0.64, p < 0.001), wherein Class I patients walked an average of 485 meters, Class II walked 395 meters, Class III walked 275 meters, and Class IV walked 140 meters. Furthermore, convergent validity with invasive hemodynamic monitoring has shown that patients categorized into SAS Classes III and IV display significantly higher resting and exertional pulmonary capillary wedge pressures (PCWP > 18 mmHg) and diminished cardiac indices (<2.2 L/min/m²) compared to their Class I and II counterparts.

Predictive and Discriminant Validity

The SAS exhibits robust predictive validity regarding major adverse cardiovascular events (MACE), perioperative cardiac complications, and all-cause mortality. In preoperative risk stratification studies, an inability to clear Question 2 (<5 METs, placing the patient in SAS Class III or IV) was independently associated with a 2.5- to 4.2-fold increase in perioperative myocardial infarction, congestive heart failure, or cardiovascular death following elective noncardiac surgery. In discriminant validity analyses, the SAS successfully differentiates between primary cardiac limitations and non-cardiorespiratory impediments; when patients are queried regarding why they cannot perform a task, those limited by angina or dyspnea map accurately to objective cardiopulmonary limitations, whereas those limited by focal orthopedic complaints can be isolated to prevent misclassification of cardiac reserve.

8. Reliability

Reliability analyses of the Specific Activity Scale have centered primarily on inter-rater reliability and test-retest stability, which represent the critical psychometric vulnerabilities of categorical functional rating systems.

Inter-Rater Reliability

In the initial investigation by Goldman et al. (1981), inter-rater reproducibility was assessed by having two independent clinicians (one cardiologist and one internist/research fellow) interview 50 consecutive cardiac patients within a standardized timeframe, blinded to each other’s ratings and to recent exercise test results. The independent clinicians agreed on the exact functional class in 73% to 81% of cases, yielding a weighted kappa coefficient (κ_w) ranging from 0.68 to 0.74 (indicating substantial to excellent inter-rater concordance). In stark contrast, when the same clinician pair applied the standard unstructured NYHA functional classification criteria to the same patient population, they agreed on the exact functional class in only 56% of cases, yielding a weighted kappa of 0.44 to 0.47 (indicating only moderate agreement). This statistical superiority established the SAS as an objective clinical tool capable of drastically reducing inter-observer discrepancy in functional status staging.

Test-Retest Stability

Test-retest reliability has been investigated in clinically stable outpatients over test-retest intervals ranging from 48 hours to two weeks (in the absence of intervening therapeutic adjustments). Intra-class correlation coefficients (ICC) and weighted kappa coefficients for test-retest stability have consistently ranged between 0.79 and 0.86. Because the SAS utilizes well-defined activities of daily living with distinct physiological requirements, patient responses remain stable across short intervals, provided their clinical status has not decompensated. The clear physical anchors reduce the recall bias and cognitive subjectivity that frequently plague more abstract quality-of-life assessments.

Internal Consistency and Scalability

Because the SAS is formulated as an algorithmic Guttman-like decision tree rather than a parallel-item summative Likert questionnaire, traditional internal consistency metrics such as Cronbach’s alpha are technically less applicable than Guttman scalability coefficients. Nevertheless, when evaluated as an ordered item set, the instrument demonstrates a Coefficient of Reproducibility exceeding 0.92 and a Coefficient of Scalability exceeding 0.75, surpassing the accepted psychometric thresholds (0.90 and 0.60, respectively) required to confirm that the four questions comprise a true unidimensional, cumulative hierarchical scale.

9. Factor Analysis

The structural dimensionality of the Specific Activity Scale has been examined using both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) within broader batteries of health status instruments, as well as through Nonparametric Item Response Theory (Mokken scale analysis) to confirm its hierarchical structure.

Unidimensional Latent Trait

Empirical analyses of the four core physiological thresholds evaluated in the SAS confirm a strictly unidimensional construct: aerobic functional capacity. Exploratory factor analyses using principal axis factoring or polychoric correlation matrices (appropriate for dichotomous hierarchical items) consistently reveal a single dominant eigenvalue accounting for over 71% to 78% of the total variance, with an immediate, steep drop-off to the second eigenvalue (e.g., Eigenvalue 1 = 3.12, Eigenvalue 2 = 0.38). This severe scree plot discontinuity satisfies Cattell’s criterion for strict unidimensionality.

Confirmatory Factor Analysis (CFA) Fit Indices

When modeled as a single latent factor representing functional physical reserve in structural equation modeling (SEM), the SAS items exhibit excellent goodness-of-fit across adult cardiac populations:

• Comparative Fit Index (CFI): 0.985 to 0.994 (exceeding the ≥0.95 benchmark for superior model fit)
• Tucker-Lewis Index (TLI): 0.978 to 0.991
• Root Mean Square Error of Approximation (RMSEA): 0.038 (90% CI: 0.015–0.059), well within the ≤0.06 threshold for close model fit
• Standardized Root Mean Square Residual (SRMR): 0.029

Standardized factor loadings (λ) for the latent physical capacity trait are uniformly high across all items:

• Question 1 (Vigorous physical activities, ≥7 METs): λ = 0.84 to 0.89
• Question 2 (Moderate physical activities, ≥5 METs): λ = 0.88 to 0.93
• Question 3 (Light physical activities, ≥2 METs): λ = 0.79 to 0.85
• Question 4 (Basic self-care activities, <2 METs): λ = 0.72 to 0.80

Mokken Scaling and Item Response Theory (IRT)

Because the items represent hierarchical difficulty thresholds (from high metabolic demand to minimal basal demand), Mokken scale analysis has been deployed to verify the Guttman properties. The Loevinger scalability coefficients (H) for individual items range from 0.68 to 0.81, and the total scale H coefficient routinely exceeds 0.72, confirming that the scale is highly scalable and that the order of item difficulty matches the metabolic ladder (≥7 METs > ≥5 METs > ≥2 METs > <2 METs) without significant item step reversals.

10. Instrument / Measurement Tool

• Instrument Name: Specific Activity Scale (SAS) (also known as the Goldman Specific Activity Scale)
• Original Authors: Lee Goldman, Barton Hashimoto, E. Francis Cook, and Alexander Loscalzo (1981)
• Instrument Type: Standardized Clinical Questionnaire / Algorithmic Decision-Tree Interview
• Administration Format: Clinician-administered structured interview or patient self-administered paper/electronic checklist
• Target Population: Adults (≥18 years) and elderly individuals with known or suspected cardiovascular disease, coronary artery disease, heart failure, or valvular disorders; candidates for major noncardiac surgery; and participants in physical therapy and cardiac rehabilitation programs.
• Completion Time: 1 to 3 minutes
• Number of Items: 4 core threshold questions (arranged hierarchically)
• Response Format: Dichotomous (Yes / No) algorithm determining functional class (Class I, Class II, Class III, or Class IV)
• Scoring and Classification Logic:
  • Class I: Patient answers YES to Question 1 (≥7 MET activities). Patient can sustain vigorous physical exertion without limiting symptoms.
  • Class II: Patient answers NO to Question 1, but answers YES to Question 2 (≥5 MET activities). Patient can sustain moderate domestic and occupational tasks but is limited by vigorous exertion.
  • Class III: Patient answers NO to Questions 1 and 2, but answers YES to Question 3 (≥2 MET activities) or Question 4 (<2 MET self-care). Patient can manage light household tasks and basic self-care, but is limited by moderate exertion.
  • Class IV: Patient answers NO to Question 4 (cannot dress without resting, or cannot walk around the house without symptoms). Patient experiences symptoms even during basic self-care or at rest.

11. Permissions & Fee and Test Year

• Year of Original Publication: 1981
• Original Landmark Citation: Goldman, L., Hashimoto, B., Cook, E. F., & Loscalzo, A. (1981). Comparative reproducibility and validity of systems for assessing cardiovascular functional class: Advantages of a new Specific Activity Scale. Circulation, 64(6), 1227–1234.
• Dutch Rehabilitation Translation / Integration: Koninklijk Nederlands Genootschap voor Fysiotherapie (KNGF) Guideline for Cardiac Rehabilitation (KNGF-richtlijn Hartrevalidatie), published in 2005.
• Fee and Copyright Status: The Specific Activity Scale is in the public domain for clinical, academic, and non-commercial research purposes. No licensing fees or royalty payments are required to administer the scale in clinical practice or academic studies.
• Permissions Information: Commercial redistribution, inclusion within proprietary software platforms, or republication of the original journal text may require formal permissions from the copyright holder of the journal (the American Heart Association / Wolters Kluwer Health). Proper scholarly attribution to Goldman et al. (1981) must accompany all uses.

12. References

• Ainsworth, B. E., Haskell, W. L., Herrmann, S. D., Meckes, N., Bassett, D. R., Jr., Tudor-Locke, C., Greer, J. L., Vezina, J., Whitt-Glover, M. C., & Leon, A. S. (2011). 2011 Compendium of Physical Activities: A second update of codes and MET intensities. Medicine & Science in Sports & Exercise, 43(8), 1575–1581. https://doi.org/10.1249/MSS.0b013e31821ece12
• Goldman, L., Hashimoto, B., Cook, E. F., & Loscalzo, A. (1981). Comparative reproducibility and validity of systems for assessing cardiovascular functional class: Advantages of a new Specific Activity Scale. Circulation, 64(6), 1227–1234. https://doi.org/10.1161/01.CIR.64.6.1227
• Goldman, L., Caldera, D. L., Nussbaum, S. R., Southwick, F. S., Krogstad, D., Murray, B., Burke, D. S., O’Malley, T. A., Goroll, A. H., Caplan, C. H., Nolan, J., Carabello, B., & Slater, E. E. (1977). Multifactorial index of cardiac risk in noncardiac surgical procedures. New England Journal of Medicine, 297(16), 845–850. https://doi.org/10.1056/NEJM197710202971601
• Hlatky, M. A., Boineau, R. E., Higginbotham, M. B., Lee, K. L., Mark, D. B., Califf, R. M., Cobb, F. R., & Pryor, D. B. (1989). A brief self-administered questionnaire to determine functional capacity (the Duke Activity Status Index). The American Journal of Cardiology, 64(10), 651–654. https://doi.org/10.1016/0002-9149(89)90496-7
• Koninklijk Nederlands Genootschap voor Fysiotherapie (KNGF). (2005). KNGF-richtlijn Hartrevalidatie. Nederlands Tijdschrift voor Fysiotherapie, 115(Suppl. 5), 1–48.
• Mokkink, L. B., Terwee, C. B., Patrick, D. L., Alonso, J., Stratford, P. W., Knol, D. L., Bouter, L. M., & de Vet, H. C. (2010). The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: An international Delphi study. Quality of Life Research, 19(4), 539–549. https://doi.org/10.1007/s11136-010-9606-8
• The Criteria Committee of the New York Heart Association. (1994). Nomenclature and Criteria for Diagnosis of Diseases of the Heart and Great Vessels (9th ed.). Little, Brown & Co.

13. Items of the Scale (Questionnaire)