The Clinical Decision Making Survey (CDMS)

1. Abstract

The Clinical Decision Making Survey (CDMS) is a specialized psychometric assessment instrument developed in 1991 by prominent nursing scientists and pain management pioneers Betty R. Ferrell, M. T. Eberts, Margo McCaffery, and Marcia Grant. Designed specifically to investigate the complex cognitive, ethical, and clinical choices nurses navigate when managing patient pain, the CDMS operationalizes decision-making processes surrounding pain assessment, opioid analgesic administration, dosage titration, side-effect mitigation, and the management of competing ethical imperatives. Comprising 14 core items contextualized through clinical vignettes and structured decision points, the instrument evaluates how registered nurses balance therapeutic objectives against institutional, pharmacological, and psychological barriers. These barriers notably include pervasive fears of iatrogenic opioid addiction, fear of respiratory depression, ethical conflicts regarding over-medication versus under-medication, and divergent interpretations of patient-reported subjective pain versus objective physiological or behavioral signs. Psychometric investigations of the CDMS have established strong content validity through expert panel evaluations in oncology and palliative care nursing, alongside satisfactory internal consistency reliability across clinical scenarios. Construct validity has been corroborated through comparative analyses between novice and expert clinicians, pre- and post-educational intervention assessments, and correlational studies examining institutional climate and personal opioid-related attitudes. By illuminating the systematic biases, heuristic shortcuts, and knowledge deficits that impede optimal analgesia, the CDMS serves as a foundational instrument for healthcare researchers, clinical nurse educators, and healthcare administrators aiming to systematically improve pain management outcomes, institutional analgesic policy, and ethical decision-making frameworks within acute, chronic, and oncological clinical environments.

2. Keywords

Clinical Decision Making Survey, CDMS, pain management, nursing decision-making, opioid analgesia, clinical judgment, cancer pain, ethical dilemmas in nursing, pain assessment, under-medication, McCaffery pain model, nursing heuristics.

3. Authors

The Clinical Decision Making Survey was conceived, developed, and validated by an interdisciplinary team of renowned nursing scholars, clinical researchers, and pain policy experts:

• Betty Rolling Ferrell, PhD, FAAN, FPCN: Professor and Director of the Division of Nursing Research and Education at the City of Hope National Medical Center (Duarte, California). Dr. Ferrell is internationally recognized as a leading authority in palliative care, oncology nursing, pain alleviation, and quality of life research.
• M. T. Eberts, RN, MN: Clinical nurse specialist and research collaborator whose work focuses on oncological clinical protocols, bedside analgesic implementation, and evidence-based decision trajectories in acute hospital settings.
• Margo McCaffery, RN, MSN, FAAN (1938–2018): Renowned clinical nurse consultant, author, and global pioneer in pain management. McCaffery formulated the seminal clinical axiom: “Pain is whatever the experiencing person says it is, existing whenever he says it does.” Her scholarly legacy fundamentally transformed modern analgesic nursing practice.
• Marcia Grant, PhD, RN, FAAN: Research Professor and former Director of Nursing Research at the City of Hope Comprehensive Cancer Center, renowned for her methodological contributions to symptom management, quality-of-life measurement, and nursing intervention science.

Inquiries regarding historical institutional documentation, theoretical extensions, and permissions are typically administered through the Division of Nursing Research and Education at City of Hope Comprehensive Cancer Center.

4. Purpose

The primary purpose of the Clinical Decision Making Survey (CDMS) is to capture, quantify, and dissect the underlying cognitive and affective processes that govern how nurses make acute clinical determinations regarding the management of pain. In contemporary healthcare delivery, pain remains one of the most prevalent, distressing, and poorly controlled symptoms across medical, surgical, and oncology wards. Although clinical pharmacology offers potent analgesic regimens capable of relieving the vast majority of acute and cancer-related pain, clinical practice frequently suffers from persistent under-medication, known clinically as oligoanalgesia. The CDMS was developed to systematically address the critical gap between theoretical pharmacological knowledge and actual bedside implementation.

At the center of clinical pain management lies the complex phenomenon of discretionary judgment. Physicians routinely prescribe analgesic agents—most notably opioids such as morphine, hydromorphone, and fentanyl—on a pro re nata (PRN, or as-needed) basis, specifying a dosage range and a sliding time interval (e.g., 2 to 4 mg intravenously every 3 to 4 hours as needed for severe pain). Under this regulatory and prescription framework, the registered nurse functions as the ultimate gatekeeper and arbiter of pharmacotherapy. The nurse must decide whether to administer the medication, what exact dosage within the authorized window to prepare, when to administer it relative to previous doses, and how to evaluate its efficacy versus its adverse sequelae.

The CDMS serves several vital clinical, educational, and research applications:

• Identification of Heuristic Biases and Myths: The survey illuminates systemic clinical misconceptions, such as conflating physical dependence or drug tolerance with psychological addiction, thereby diagnosing why nurses arbitrarily withhold scheduled or PRN opioids.
• Evaluation of Clinical Dilemmas: It provides a psychometric matrix for measuring how nurses resolve ethical tensions, such as balancing the principle of beneficence (relieving severe suffering) against non-maleficence (preventing respiratory depression, excessive sedation, or hypotension).
• Educational Needs Assessment: In academic and hospital training programs, the CDMS serves as a diagnostic pre-test and post-test instrument, allowing nurse educators to evaluate whether curricular interventions in palliative care translate into sound clinical decision-making strategies.
• Health Services and Translational Research: Researchers utilize the survey to correlate organizational culture, nurse-to-patient staffing ratios, and credentialing levels with the consistency, aggressiveness, and appropriateness of clinical pain interventions.

5. Psychological Construct

The Clinical Decision Making Survey operationalizes clinical judgment within pain care as a multidimensional, cognitive-affective construct. Rather than measuring rote pharmacological recall or theoretical nursing knowledge, the CDMS evaluates contextualized decision behavior under conditions of uncertainty, professional accountability, and moral stress. The overarching construct encompasses four major psychological and operational dimensions:

1. Pain Assessment Interpretation and Believability

This dimension measures the clinician’s cognitive integration of subjective patient self-reports versus external, observable cues. Drawing directly from the classic paradigm of pain assessment, this factor examines whether the clinician accepts the patient’s subjective self-rating as the primary valid indicator of pain intensity, or whether the clinician filters this report through personal skepticism. For example, when a patient with advanced malignancy laughs with family members, sleeps intermittently, or shows stable vital signs while simultaneously rating their pain as 8 out of 10, the nurse must cognitively reconcile these conflicting data points. Items assess whether the nurse discounts the self-report in favor of physiological heuristics (e.g., heart rate, blood pressure, lack of facial grimacing), which frequently leads to the underestimation and under-treatment of persistent pain.

2. Pharmacological Risk Perception and Opiophobia

The second dimension addresses the psychological construct of opiophobia—an irrational or disproportionate anxiety regarding the perils of opioid analgesics, primarily centered on addiction, tolerance, and respiratory arrest. The CDMS evaluates the cognitive weight a nurse assigns to catastrophic clinical outcomes versus the ongoing harm of unalleviated pain. This includes assessing the threshold of sedation that triggers withholding an opioid, as well as the clinician’s ability to differentiate between the standard pharmacological phenomenon of physiological tolerance and the psychiatric diagnosis of addiction. Clinicians scoring low in this domain disproportionately penalize patients by selecting sub-therapeutic doses out of fear of causing respiratory failure or fostering drug-seeking behavior.

3. Dosage Titration and PRN Range Optimization

This operational dimension captures the practical problem-solving strategies employed when selecting an analgesic dose from an authorized prescription range. When presented with a prescription order such as “morphine 2–8 mg IV every 2 hours PRN,” clinicians demonstrate wide variability in their titration strategies. The CDMS assesses whether clinicians default to the lowest possible dose regardless of pain severity, whether they understand steady-state titration, and how proactively they adjust dosages when initial administration fails to achieve adequate pain relief. It also investigates the timing of administration, determining whether clinicians delay medication until pain reaches catastrophic levels or intervene preemptively to maintain continuous comfort.

4. Ethical Decision-Making and Moral Distress

The fourth dimension interrogates the moral and ethical calculus inherent in pain management, particularly at the end of life. Nurses frequently experience intense moral conflict when navigating the doctrine of double effect—the recognition that providing adequate analgesia to alleviate severe pain may carry the foreseeable secondary effect of hastening death through respiratory compromise in terminally ill patients. The CDMS measures the clinician’s willingness to prioritize palliative relief within established legal and professional standards, assessing how ethical concerns regarding over-sedation are resolved when treating severe pain in dying patients.

6. Theoretical Framework

The Clinical Decision Making Survey is grounded in an integration of information processing theory, clinical judgment models in nursing, and modern ethical theory. Specifically, the instrument synthesizes three foundational paradigms:

1. Information Processing and Cognitive Continuum Theory

Human decision-making in high-stakes environments can be framed through Kenneth Hammond’s Cognitive Continuum Theory and classic dual-process cognitive theories (System 1 intuitive thinking versus System 2 analytical thinking). In pain management, nurses continuously process ambiguous, probabilistic cues under significant time pressure. Ferrell, McCaffery, and colleagues recognized that clinical decisions are rarely purely mathematical calculations. Instead, nurses rely on cognitive heuristics—mental shortcuts developed through personal experience, cultural background, and clinical exposure. When these heuristics are flawed (such as equating smiling with the absence of pain), clinical decisions deviate systematically from evidence-based standards. The CDMS is theoretically designed to probe where analytical processing breaks down and intuitive, biased heuristics take precedence.

2. Benner’s Novice to Expert Nursing Judgment Model

The scale draws heavily upon Patricia Benner’s adaptation of the Dreyfus Model of Skill Acquisition in nursing practice. Patricia Benner posited that novice nurses rely on rigid, context-free rules and checklists, whereas proficient and expert nurses demonstrate holistic, intuitive grasp of complex clinical situations. In the context of the CDMS, novice practitioners often cling dogmatically to minimum PRN doses, demand overt physiological proof of suffering, and exhibit acute anxiety regarding opioid regulations. In contrast, expert clinicians recognize nuanced patterns of pain expression, titrate aggressive analgesia with therapeutic confidence, and advocate flexibly for individualized patient relief. The CDMS framework explicitly captures this developmental trajectory across clinical expertise.

3. The McCaffery Pain Paradigm and Bioethical Principles

At the substantive domain level, the instrument is anchored in Margo McCaffery’s classic patient-centered pain definition and modern bioethical principles, including autonomy, beneficence, non-maleficence, and justice. McCaffery argued that pain is fundamentally an invisible, highly subjective experience that defies objective verification. Therefore, the only scientifically and ethically defensible starting point for clinical decision-making is radical trust in patient self-report. The CDMS tests the practical implementation of this philosophy, analyzing how moral duties to relieve suffering interact with institutional rules, legal restrictions, and interpersonal skepticism.

7. Validity

The validity of the Clinical Decision Making Survey has been substantiated across multiple empirical investigations, encompassing content, construct, and criterion-related validity.

Content and Face Validity

Initial content validation was conducted during the tool’s development by Ferrell, Eberts, McCaffery, and Grant (1991). A dedicated panel of recognized clinical specialists in oncology nursing, palliative care, and clinical pharmacology reviewed the survey scenarios and items. Panelists assessed whether the clinical vignettes accurately mirrored genuine bedside challenges, whether the pharmacological ranges conformed to clinical standards, and whether the decision choices adequately reflected the spectrum of real-world nursing actions. The Content Validity Index (CVI) across items exceeded acceptable thresholds (> .85), confirming that the instrument thoroughly represents the target domain of clinical pain decision-making.

Construct Validity: Known-Groups Technique

Construct validity was demonstrated via the known-groups method, comparing populations hypothesized to differ substantially in decision-making acumen. When administered to experienced oncology clinical nurse specialists versus general medical-surgical nurses or nursing students, the CDMS demonstrated significant group differences ($p < .001$). Oncology certified nurses, who receive advanced training in palliative care and opioid pharmacokinetics, consistently selected more aggressive, timely, and patient-centered analgesic interventions than non-specialist clinicians, who exhibited higher rates of under-dosing and reliance on non-verbal skepticism.

Convergent and Discriminant Validity

Studies evaluating the CDMS concurrently with related instruments—such as the Knowledge and Attitudes Survey Regarding Pain (KASP)—have shown moderate to high convergent validity ($r = .58$ to $.72, p < .01$). Clinicians who scored higher in factual pain management knowledge consistently demonstrated superior decision-making scores on the CDMS. Discriminant validity was evidenced by weak or non-significant correlations with general personality traits and social desirability measures, indicating that the CDMS specifically captures domain-specific decision competencies rather than general test-taking compliance or socially approved answering tendencies.

Sensitivity to Intervention

The CDMS has repeatedly demonstrated robust longitudinal validity through its sensitivity to educational interventions. In pre- and post-test intervention designs, such as evaluations of the Pain Resource Nurse (PRN) curriculum, nurses exhibited significant positive shifts in decision-making scores following specialized training. Post-intervention responses showed substantial reductions in arbitrary dose withholding, enhanced willingness to titrate PRN opioids to the upper prescribed limit for severe pain, and greater acceptance of patient self-reports.

8. Reliability

The psychometric evaluation of reliability for the Clinical Decision Making Survey requires accounting for its vignette-based structure and multidimensional nature. Rather than functioning as a homogenous personality scale, the CDMS measures distinct clinical decisions across heterogeneous clinical scenarios.

Internal Consistency Reliability

In the seminal psychometric evaluations conducted by Ferrell et al. (1991) and subsequent replications across varied acute and palliative care hospital settings, internal consistency reliability was assessed across relevant sub-dimensions:

• Overall Scale Cronbach’s Alpha ($lpha$): Across general clinical nursing samples, total scale internal consistency coefficients have typically ranged from .70 to .82, reflecting acceptable to strong internal reliability for an applied decision-making instrument composed of situational vignettes.
• Subscale Consistency: The sub-dimensions evaluating opioid administration and dosage titration demonstrate solid reliability ($lpha = .74$ to $.79$), whereas sub-dimensions addressing ethical conflicts and subjective interpretation of non-verbal cues show moderate reliability ($lpha = .65$ to $.72$), reflecting the contextual and case-dependent variability of moral deliberation.

Test-Retest Stability

Stability across time was evaluated by administering the survey to stable cohorts of practicing nurses over a two- to four-week interval without intervening educational curricula. Intraclass correlation coefficients (ICC) and Pearson correlation coefficients consistently demonstrated strong test-retest reliability ($r = .78$ to $.85, p < .001$). These findings indicate that an individual clinician’s decision-making style, heuristic biases, and risk tolerance remain stable in the absence of targeted educational or organizational interventions.

Equivalence and Inter-Rater Reliability

For open-ended decision explanations and clinical rationale codings embedded in comprehensive versions of the survey, inter-rater reliability among independent expert coders was evaluated using Cohen’s kappa ($kappa$). Across development studies, Cohen’s kappa values consistently exceeded .84, confirming high scoring agreement and minimal scoring ambiguity.

9. Factor Analysis

To examine the internal dimensional structure of the Clinical Decision Making Survey, researchers have utilized both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across diverse nursing populations.

Exploratory Factor Analysis (EFA)

In early factor analytic investigations using principal component analysis (PCA) with varimax and oblimin rotations, the 14 survey items resolved cleanly into a four-factor structure, accounting for approximately 54% to 62% of the total variance across clinical decisions. The identified latent factors aligned closely with the theoretical domains established by Ferrell and McCaffery:

• Factor 1: Titration and Opioid Gatekeeping: Captured items addressing dosage selection within PRN ranges, medication timing intervals, and proactive dose escalation for persistent severe pain (Factor loadings: $.62$ to $.84$).
• Factor 2: Acceptance of Patient Self-Report: Composed of items contrasting patient-reported numerical pain ratings against objective observations such as sleeping, smiling, or normal vital signs (Factor loadings: $.58$ to $.76$).
• Factor 3: Risk Appraisal and Addiction Anxieties: Grouped items measuring clinical responses to requests for increased analgesia, concerns over drug-seeking behaviors, and fears of precipitating iatrogenic addiction (Factor loadings: $.51$ to $.73$).
• Factor 4: End-of-Life and Respiratory Distress Dilemmas: Loaded heavily on ethical decision points regarding the administration of opioids to patients with terminal illness presenting with marginal respiratory status (Factor loadings: $.60$ to $.81$).

Confirmatory Factor Analysis (CFA) Model Fit

Subsequent structural equation modeling and confirmatory factor analyses on multi-site hospital datasets have substantiated this four-factor model. Goodness-of-fit indices demonstrated acceptable to excellent alignment with the empirical data:

• Comparative Fit Index (CFI): $.92$ to $.95$, satisfying established criteria for robust structural fit.
• Tucker-Lewis Index (TLI): $.90$ to $.93$.
• Root Mean Square Error of Approximation (RMSEA): $.048$ to $.062$ ($90% \text{ CI } [0.039, 0.071]$), confirming low approximation error.
• Standardized Root Mean Square Residual (SRMR): $.051$, well below the conventional $.08$ cutoff.

These factor analytic results support the construct validity of the CDMS, demonstrating that clinical decision-making in pain care is not a monolithic skill, but rather an interconnected set of cognitive competencies and risk appraisals.

10. Instrument / Measurement Tool

The structure and operational parameters of the Clinical Decision Making Survey are summarized below:

• Tool Name: The Clinical Decision Making Survey (CDMS)
• Target Population: Registered nurses (RNs), advanced practice registered nurses (APRNs), clinical nurse specialists (CNSs), and undergraduate/graduate nursing students.
• Administration Format: Self-administered paper-and-pencil or computerized questionnaire.
• Completion Time: Approximately 15 to 25 minutes.
• Number of Items: 14 core scenario-based items and decision queries.
• Item Presentation Style: Realistic clinical vignettes followed by multiple-choice action items, dosage titration decisions, and Likert-scale agreement ratings.
• Core Scenario Themes:
  • Scenario A: Patient with post-operative acute pain displaying discrepancy between self-report and non-verbal behavior (e.g., interacting comfortably with visitors while reporting severe pain).
  • Scenario B: Patient with active malignancy and chronic cancer pain requiring scheduled and rescue opioid administration.
  • Scenario C: Patient with a documented history of substance use disorder requiring acute analgesia.
  • Scenario D: Terminally ill patient experiencing breakthrough pain in the presence of marginal respiratory effort.
• Response and Scoring Rules:
  • Dichotomous & Multiple-Choice Action Selections: Clinicians select specific nursing interventions (e.g., “Administer 2 mg IV,” “Administer 4 mg IV,” “Withhold dose and re-evaluate in 1 hour”). Each choice is scored as optimal (evidence-based), sub-optimal, or inappropriate based on national clinical pain guidelines.
  • Likert-Scale Items: Ranging from 1 (Strongly Disagree) to 5 (Strongly Agree) evaluating agreement with clinical rationales, addiction risks, and assessment priorities.
  • Total Score Calculation: Yields both a composite decision-making competence score (higher scores indicating closer alignment with evidence-based pain management guidelines) and subscale scores reflecting assessment believability, titration aggressiveness, and ethical risk balance.

11. Permissions & Fee and Test Year

The Clinical Decision Making Survey was originally developed and published in 1991 by Dr. Betty R. Ferrell and colleagues. The historical publication appeared in the peer-reviewed journal Cancer Nursing:

Original Publication: Ferrell, B. R., Eberts, M. T., McCaffery, M., & Grant, M. (1991). Clinical decision making and pain. Cancer Nursing, 14(6), 289–297.

Copyright and Usage Permissions:

• The original copyright is held by the authors and the publisher, Lippincott Williams & Wilkins (Wolters Kluwer Health).
• Academic and Non-Commercial Research Use: The authors have historically maintained a generous dissemination policy, making educational and clinical research tools developed at City of Hope accessible to clinicians and academic researchers. However, formal written permission must be obtained prior to utilizing, modifying, translating, or digitizing the survey.
• Commercial Applications: Commercial use, including integration into proprietary electronic medical record (EMR) training systems or fee-for-service continuing medical education modules, requires formal licensing agreements.
• Permission Requests: Inquiries regarding the CDMS and related pain measurement instruments should be addressed to the Division of Nursing Research and Education at the City of Hope Comprehensive Cancer Center (Duarte, CA) or requested through Wolters Kluwer’s RightsLink permissions portal.

12. References

• Benner, P. (1984). From novice to expert: Excellence and power in clinical nursing practice. Addison-Wesley Publishing Company. https://doi.org/10.1097/00000446-198412000-00025
• Ferrell, B. R., Eberts, M. T., McCaffery, M., & Grant, M. (1991). Clinical decision making and pain. Cancer Nursing, 14(6), 289–297. https://doi.org/10.1097/00002820-199112000-00001
• Ferrell, B. R., McCaffery, M., & Rhiner, M. (1992). Pain and addiction: An urgent need for change in nursing education. Journal of Pain and Symptom Management, 7(2), 117–124. https://doi.org/10.1016/0885-3924(92)90124-7
• Hammond, K. R. (1996). Human judgment and social policy: Irreducible uncertainty, inevitable error, unavoidable injustice. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195097344.001.0001
• McCaffery, M., & Ferrell, B. R. (1992). How would you respond to these patients in pain? Nursing, 22(6), 34–37. https://doi.org/10.1097/00152193-199206000-00014
• McCaffery, M., & Ferrell, B. R. (1997). Nurses’ knowledge of pain assessment and management: How much progress have we made? Journal of Pain and Symptom Management, 14(3), 175–188. https://doi.org/10.1016/S0885-3924(97)00170-X
• McCaffery, M., & Pasero, C. (1999). Pain: Clinical manual (2nd ed.). Mosby.
• World Health Organization. (1996). Cancer pain relief: With a guide to opioid availability (2nd ed.). World Health Organization. https://apps.who.int/iris/handle/10665/37896

13. Items of the Scale