Duke Activity Status Index (DASI)

Clinical background

Functional capacity is a central variable in preoperative risk assessment before non-cardiac surgery. Both the ESC and the ACC/AHA guidelines use a threshold of around 4 METs to separate patients with poor exercise capacity, who may require further cardiac investigation, from those with sufficient reserve [5]. The clinical problem is that a physician's subjective estimate of the patient's functional capacity is unreliable: in the METS study, the anaesthetists' assessment had a sensitivity of only 19.2 per cent for identifying patients who did not achieve 4 METs on formal exercise testing [3]. The Duke Activity Status Index (DASI) fills this gap by offering a structured, self-administered questionnaire that estimates functional capacity from everyday activities and gives an estimate of peak oxygen uptake.

Calculating the Duke Activity Status Index

The DASI consists of 12 yes/no questions about the ability to perform activities of increasing energy demand. Each question is assigned a weighted score based on the metabolic cost of the corresponding activity. The total score is calculated as:

$$\text{DASI} = \sum_{i=1}^{12} w_i \cdot x_i$$

where $w_i$ is the weight for activity $i$ and $x_i$ is 1 if the patient answers "Yes" and 0 if the patient answers "No". The score ranges from 0 to 58.2, with a higher value indicating better functional capacity.

From the DASI score, the estimated peak oxygen uptake and METs can be derived:

$$\text{VO}_{2\text{max}} = 0{.}43 \times \text{DASI} + 9{.}6 \quad (\text{mL/kg/min})$$

$$\text{METs} = \frac{\text{VO}_{2\text{max}}}{3{.}5}$$

The derivation cohort consisted of 50 adults who underwent exercise testing with measurement of peak oxygen uptake at Duke University Medical Center [1]. All participants were interviewed about their ability to perform a range of everyday activities, and a 12-item scale was constructed that correlated with measured peak oxygen uptake at Spearman r = 0.80. In an independent validation cohort of 50 people who completed the questionnaire themselves and underwent exercise testing, the correlation fell to r = 0.58 (p < 0.0001) [1]. The original cohorts were small and consisted of patients referred for exercise testing, not an unselected surgical population.

Interpretation in practice

The DASI score translates into an estimated VO₂max and METs via the formulae above. The most clinically important threshold in perioperative guidelines is 4 METs, corresponding to a VO₂max of 14 mL/kg/min and a DASI score of approximately 10. The following are practical guides to action based on the estimated exercise capacity:

The DASI should be read as a continuous measure rather than a dichotomous screening test. In an international pooled cohort, the predicted risk at a given DASI score varied considerably with age, the Revised Cardiac Risk Index and the natriuretic peptide, which argues for integrating the score with other risk markers rather than using it alone for decision-making [4].

Validation and performance

Since the original study, the DASI has been validated in several independent cohorts. In a Chinese validation of 107 cardiology patients undergoing cardiopulmonary exercise testing (CPET), a Spearman correlation of r = 0.67 (p < 0.001) was measured between the DASI score and measured peak oxygen uptake, with an AUC of 0.788 for identifying patients with a VO₂max above 16 mL/kg/min [2]. Cronbach's alpha was 0.706, indicating acceptable internal consistency.

In the METS study, a prospective international cohort of 1,401 patients aged ≥40 years undergoing elective major non-cardiac surgery, the DASI was the only method of assessing functional capacity that was associated with the primary outcome of death or myocardial infarction within 30 days (adjusted OR 0.96, 95% CI 0.83–0.99, p = 0.03) [3]. The treating anaesthetist's subjective assessment of METs was not associated with the outcome.

A retrospective cohort study from Duke of 4,199 patients with risk factors for coronary artery disease found that the original DASI predicted the composite outcome of death or myocardial injury within 30 days with an AUC of 0.82 (95% CI 0.73–0.91) [6]. A modified four-question version performed equivalently. The DASI also predicted one-year survival (hazard ratio 0.88, p < 0.001) but not severe complications within 30 days.

The largest analysis to date, a pooled cohort of 3,485 patients from the METS and FIT After Surgery studies, showed that the DASI added incremental prognostic information beyond age, the RCRI and the natriuretic peptide (likelihood ratio test p = 0.009) [4]. Discrimination nonetheless remained modest, with a c-index of 0.70–0.71, and the clinical benefit in decision curve analysis was limited. The authors conclude that the DASI works best as a complement to established risk factors, not as a standalone threshold-based tool.

Limitations

The DASI is a self-assessment instrument and does not replace measured exercise capacity. Its correlation with CPET ranges from 0.46 to 0.80 depending on the population, and is generally lower than in the original cohort [1, 2]. Several factors affect its reliability:

Cultural adaptation. The questionnaire includes activities such as bowling, baseball and doubles tennis, which may be unfamiliar to patients outside a North American setting. In the Chinese validation, patients expressed unfamiliarity with several of these activities, which contributed to a weaker correlation in an earlier study (r = 0.467) before cultural adaptation was carried out [2].

Socioeconomic bias. In the Duke cohort, the Area Deprivation Index was inversely related to the DASI score, with a decrease of 0.8 points per 10-point increase in the deprivation index [6]. Patients of lower socioeconomic status therefore reported poorer functional capacity irrespective of their actual cardiac reserve, which may reflect differences in living environment and access to leisure activities rather than physiological capacity.

Limited discrimination in preoperative use. Although the DASI performs better than subjective assessment, its prognostic discrimination is moderate (c-index 0.70–0.71) and its clinical value as a standalone test is limited [4]. A DASI score corresponding to good functional capacity does not exclude significant cardiac risk, particularly in patients with a high RCRI or raised biomarkers.

Does not apply to all populations. The derivation cohort consisted of patients referred for exercise testing at an American university hospital. The instrument has not been validated for patients with neurological or musculoskeletal impairment limiting mobility irrespective of cardiac status, and its interpretation becomes misleading in patients whose activity limitation is not cardiac in origin.

References

1. Hlatky MA et al. A brief self-administered questionnaire to determine functional capacity (the Duke Activity Status Index). Am J Cardiol 1989. PMID: 2782256
2. Liao Y et al. A Comprehensive Assessment of the Chinese Version of the Duke Activity Status Index in Patients with Cardiovascular Diseases. Rev Cardiovasc Med 2024. PMID: 39077360
3. Wijeysundera DN et al. Assessment of functional capacity before major non-cardiac surgery: an international, prospective cohort study. Lancet 2018. PMID: 30070222
4. Wijeysundera DN et al. Prognostic value of the Duke Activity Status Index for preoperative cardiac risk stratification: an international pooled cohort study. EClinicalMedicine 2026. PMID: 42326382
5. Zehner C et al. Preoperative cardiovascular evaluation in patients with cancer. Front Cardiovasc Med 2026. PMID: 42368860
6. Li MH et al. A Retrospective Cohort Study Examining the Validation of the Modified Duke Activity Status Index in the Non-cardiac Surgical Population. J Perianesth Nurs 2025. PMID: 39387780