JD

Single Subject Design, Questionnaires, and Case Reports – DPT 6112 Review

Single Subject Design

  • Definition & Core Concept

    • Research strategy that uses repeated measures on one participant across time to evaluate an intervention.
    • The individual serves as their own control by comparing data from baseline (A) to treatment phase(s) (B, C…).
    • Primary sources referenced:
    • CREd Library – introductory definition.
    • “Single Subject Design for Evidence-Based Practice.”

  • Rationale / When to Use

    • Captures individual differences unobtainable in group means.
    • Reduces inter-subject variability; emphasizes intra-subject variability.
    • Useful when groups are unavailable / sample sizes are tiny.
    • Two broad variants:
    • Interrupted Time-Series (e.g., AB, ABA, ABAB).
    • Intra-subject Replication (e.g., ABC, ABCBC, A-BC).

  • Typical Design Codes

    • A = \text{Baseline Period}
    • B = \text{First Treatment}
    • C = \text{Second Treatment}
    • Common sequences & inherent problems:
    • AB: simple time-series; no return to baseline.
    • ABA / ABAB: allows observation of treatment withdrawal but raises ethical issues about removing effective therapy.
    • ABC / ABCBC / A-BC: multiple treatments; risk of carry-over (B → C) interactions.

  • Methodological Steps

    • Establish reliability & validity of measurement tool(s).
    • Detail the temporal sequence (sampling frequency: hourly, daily, etc.).
    • Obtain a sufficiently long, stable baseline.
    • Introduce treatment while continuing identical measurement schedule.

  • Interpretation (Primarily Visual)

    • Examine raw graphs for:
    • Upward trend.
    • Steady state / plateau.
    • Negative trend.
    • Ideal baseline: stable or trending opposite the expected treatment direction.
    • Treatment deemed effective if:
    • Minimal variance within treatment phases.
    • Distinct change in level and/or slope vs. baseline.
    • Withdrawal (A2) & second treatment (B2) replicate original AB pattern.

  • Validity & Reliability Concerns

    • Inter-rater reliability is critical because small data sets magnify observer error.
    • Internal validity threats:
    • History, uncontrolled co-events, inconsistent procedures, selection bias, multiple treatment interference.
    • External validity: generalization across subjects, settings, behaviors remains challenging—necessitating replication.

  • Statistical Analysis

    • Use is controversial; visual inspection often preferred.
    • The greater the data overlap between A & B, the stronger the argument for formal statistics.
    • Tiny mathematical differences may achieve significance yet be clinically trivial.
    • Several quantitative methods exist (slope analysis, percentage of non-overlapping data, two-standard-deviation-band method) → collectively termed “trend statistics.”

  • Generalizability Questions

    • Will findings:
    • Apply to other patients?
    • Translate when delivered by other clinicians?
    • Hold in different environments?

Questionnaires

  • Key Difficulties

    • Designing an instrument that is simultaneously reliable & valid.
    • Motivating respondents to reply willingly, promptly, honestly.

  • Questionnaire vs. Interview

    • Questionnaire:
    • Focused, limited depth → efficient scope definition.
    • Minimal interviewer bias; cheap; anonymity; large samples quickly.
    • Limitations: low return rates, unverifiable accuracy, weak follow-up.
    • Interview:
    • Rich, qualitative depth; allows probing & reading non-verbals; can salvage low response rates.
    • Limitations: requires interviewer training, costly, time-intensive, sometimes hard to quantify.

  • Questionnaire Design Principles

    • Constant revision; brevity matters (longer forms ↓ response probability).
    • Each question must be relevant, standalone, unambiguous.
    • Content development strategies: literature review, brainstorming, sub-samples, pilot testing.
    • Hypothesis clarity: Define both conceptual & operational variables.
    • Single-idea questions; words with one meaning; concise; grammatically correct.
    • Avoid value-laden wording, absolutes (always, never), & double negatives.
    • Decide open- vs. closed-ended based on analytic goals.

  • Readability – FOG Index

    • Formula:
      \text{Fog Index} = 0.4\left[\left(\frac{\text{total words}}{\text{total sentences}}\right) + 100\left(\frac{\text{complex words}}{\text{total words}}\right)\right]
    • Transforms to an approximate school-grade reading level (e.g., score 19+ ≈ doctoral; 12 ≈ high-school senior; 6 ≈ seventh grade).
    • Example provided: \text{Fog}=9 → ninth-grade reading difficulty.

  • Closed-Ended Items

    • Pros: simple statistics; forced choice boosts reliability; limits “fudging.”
    • Cons: threatens validity; sacrifices data richness.

  • Open-Ended Items

    • Pros: ↓ investigator bias; allows nuanced, continuous data; easy to pose.
    • Cons: complex coding/analysis; off-topic answers; writing skill variance; longer respondent time.

  • Ordering & Presentation

    • Sequence cues memory: interesting → chronological → broad → specific; cluster by theme.
    • Avoid splitting thematic blocks across pages; exclude mere “fun to know” curiosities.
    • Layout: clear headings/instructions, minimal page count, light paper, graphics/cartoons optional.
    • Distribution: weigh costs (printing, postage); plan two-week reminders or phone follow-ups; time of year matters.

Case Reports

  • Purpose & Role

    • Mechanism for clinicians to explore, document, and analyze individual patient scenarios.
    • Bridge between theory, practice, and research (see modified Figure 12$-$1 cycle).
    • Provide teaching material, inspire hypothesis generation, inform practice guidelines, and motivate peers.

  • Case Report vs. Case Study

    • Case Report: detailed description of clinical care in one patient.
    • Case Study: broader qualitative approach; may include multiple patients sharing a condition.

  • Cautions / Limitations

    • Haynes (1990): heavy reliance risks self-deception & patient harm; most reports do not translate directly into management rules.
    • Better suited for hypothesis generation than definitive evidence.

  • Common Foci

    • Unique or common clinical problems; measurement conundrums; differential diagnoses; decision-making processes; administrative or community programs; ethical dilemmas.

  • Standard Sections (Outline)

    1. Abstract
    2. Introduction – Why this case matters.
    3. Framework / Literature Context – Mechanistic rationale.
    4. Patient Description & History – exhaustive demographics & background.
    5. Examination / Evaluation – objective findings.
    6. Diagnosis / Prognosis – PT Guide alignment.
    7. Intervention – what, how often, duration.
    8. Re-examination / Outcomes – post-data & comparisons.
    9. Discussion / Conclusions – interpretation, implications, research needs.

  • Data Types to Include

    • Narrative notes, standardized outcome measures (ROM, MMT, posture, weight, etc.), lab/imaging, photographs, videos, audio, patient-reported perceptions.

  • Example Highlight (Poster Summary)

    • 26-yr-old male, chronic stroke + HIV + neurosyphilis.
    • Intervention: 10 sessions (5-week), speed-dependent locomotor training + ancillary tasks.
    • Outcomes:
    • Postural Assessment for Stroke improved 13 \rightarrow 25 (\Delta=12 > MDC).
    • Berg Balance Score 8 \rightarrow 19 (\Delta=11 > MDC & MCID).
    • Became ambulatory >300\,\text{ft} with rolling walker & minimal assist; gained independence in dressing & transfers.
    • Maintained 80\text{–}85\% max HR on treadmill; used 30\% body-weight support throughout.
    • Conclusion: Speed-dependent locomotor training may enhance mobility in similar comorbid populations.

Practical / Course Logistics

  • Upcoming Course Tasks

    • Assignment 1 due 6/7 at 11:59 pm (submit to Moodle).
    • Quiz 1 scheduled 6/14.
    • Future activity: Students will craft their own case report from clinical experiences (seek interesting diagnoses or novel interventions; ex: Sunken Flap Syndrome case shared by instructor).

  • Ethical / Professional Implications

    • Balance need for withdrawal phases in single-subject designs against ethical duty not to withhold beneficial treatments.
    • Case reports’ anecdotal nature demands critical appraisal before altering practice.
    • Questionnaire construction must respect participant burden, comprehension, and confidentiality.

Quick Reference Equations & Numbers

  • Fog Index Formula: \text{Fog}=0.4\left[\dfrac{\text{words}}{\text{sentences}} + 100\left(\dfrac{\text{complex words}}{\text{words}}\right)\right]
  • Design Codes: A,B,C map to baseline and successive treatments.
  • Example Outcome Changes: \text{PASS}:13\to25,\; \text{BBS}:8\to19.
  • Body-weight support during treadmill: 30\%; training intensity 80\text{–}85\% HRmax.

Study Tips & Connections

  • Link single-subject design principles to clinical goal writing: each treatment block parallels personalized goal phases.
  • Use readability metrics (FOG) not only for questionnaires but also patient education materials.
  • When writing your case report, embed AB/ABC logic to illustrate pre/post outcome trajectories.

Questions to Self-Quiz

  1. In an ABAB design, what ethical issues arise when you withdraw a helpful intervention?
  2. How does visual inspection determine a “change in level” vs. a “change in trend”?
  3. List three tactics to improve questionnaire return rates without interviewer bias.
  4. Why can’t case reports alone dictate clinical practice changes?
  5. Apply the Fog formula to a 250-word paragraph containing 15 sentences & 20 complex words—what grade level results?