EDAC cheat sheet

Evidence-based design (EBD) is the process of basing decisions about the built environment on credible research to achieve the best possible outcomes.

The Fable Hospital business case (2004) modeled a 300-bed hospital in which about $12 million in design innovations were projected to return roughly $41 million per year, driven mainly by fewer falls, fewer infections, and lower nurse turnover.

ROI = (benefits minus costs) / costs x 100%. Payback period = up-front investment / annual savings.

A credible business case compares first cost against lifecycle operating cost: staffing and turnover, healthcare-associated infections, falls, length of stay, energy, and maintenance.

Triple Aim (IHI) = improve population health, improve the patient experience of care, reduce per-capita cost. Quadruple Aim adds clinician and staff well-being.

The Pebble Project (Center for Health Design, launched 2000) has partner healthcare organizations measure and publish EBD outcomes so results ripple through the industry.

Ulrich's 1984 study found that patients recovering from cholecystectomy with a nature view had shorter stays, fewer analgesic doses, and better mood than patients facing a brick wall.

Decision rule: a design investment is strongest when it targets a high-cost, high-frequency problem (falls, infections, turnover, transfers) and is supported by consistent evidence and measurable baseline data.

The EBD process opens with defining evidence-based goals and objectives that connect organizational strategy to measurable built-environment outcomes.

SMART objectives are Specific, Measurable, Achievable, Relevant, and Time-bound; non-SMART goals cannot be evaluated in a post-occupancy study.

Goal = broad desired end state; objective = specific, measurable, time-bound step that operationalizes the goal.

Design hypothesis format: 'If [design intervention], then [expected outcome], as measured by [metric].'

Standard outcome domains for goal setting: patient safety, clinical quality, patient experience, staff satisfaction and efficiency, financial performance, environmental impact.

Goals should be tied to metrics the organization already tracks (fall rates, HCAHPS scores, CLABSI/CAUTI rates, turnover) so comparison data exist.

Baseline data must be collected before construction or occupancy; without a pre-design measure, post-occupancy comparison is not possible.

Stakeholder input at the goal-setting stage aligns project values with the evidence and increases the odds the objectives are used later.

Prioritize objectives by expected impact and feasibility so the project team knows which outcomes drive design trade-offs.

Evidence hierarchy (high to low): systematic review or meta-analysis, randomized controlled trial, cohort, case-control, cross-sectional, case report, expert opinion.

PICO frames a searchable question: Population, Intervention, Comparison, Outcome.

Search technique: combine keywords with Boolean operators (AND narrows, OR broadens), add MeSH or controlled vocabulary terms, and document inclusion/exclusion criteria.

Core sources: PubMed/MEDLINE, CINAHL, Cochrane Library, InformeDesign and health design knowledge repositories, and the journal HERD.

Critical appraisal covers design appropriateness, sample size and power, randomization and blinding, bias, confounders, effect size, confidence intervals, and relevance to the project.

Internal validity = confidence that the intervention caused the observed effect; external validity = generalizability to other populations, settings, and times.

Reliability = consistency or repeatability of a measure; validity = the measure captures what it claims to capture.

Decision rule: no single study justifies a design decision; weigh consistency across studies, size of effect, and similarity of setting and population to the project.

Publication bias favors positive findings, and industry funding can skew conclusions; check for gray literature, conflicts of interest, and contradictory studies.

Translation runs evidence to design concept to tested detail; mock-ups and simulations are the standard test step before construction documents.

Evidence is context dependent: match study population, care model, acuity, culture, and building type before applying a finding.

Features with substantial evidence: single-patient rooms (fewer healthcare-associated infections, better privacy and communication), noise reduction (sleep), daylight and nature access (stress, pain, length of stay), visible hand-hygiene stations.

Acuity-adaptable rooms let patients remain in one room from intensive through med-surg care, reducing transfers, handoffs, and transfer-related errors.

Decentralized nursing shortens walking distance but can reduce peer support and sight lines; the trade-off must be weighed against unit size and staffing model.

A design decision matrix scores options against weighted criteria: strength of evidence, capital cost, operational impact, flexibility, and stakeholder priorities.

Document design rationale, linking each major decision to the evidence cited and the goal it serves, so the assumption can be tested later.

Where evidence is weak, absent, or conflicting, pair innovation with measurement: state the design hypothesis and evaluate it in post-occupancy.

Full-scale mock-ups reveal usability, safety, and workflow problems before construction, reducing late change orders and rework.

Post-occupancy evaluation (POE) is a systematic assessment of building performance after occupancy against explicit criteria such as outcomes, satisfaction, workflow, and operating cost.

POE levels (Preiser): indicative (rapid walkthrough and interviews), investigative (more data collection and analysis), diagnostic (thorough, causal explanation).

Common measures: HCAHPS, fall rate per 1,000 patient days, CLABSI/CAUTI rates, length of stay, staff turnover, noise levels in dB, walking distance, medication errors.

Quasi-experimental pre/post designs dominate EBD research because randomization is rarely feasible; secular trends and case-mix shifts are the main threats.

Triangulation across surveys, observation, archival records, and environmental sensors increases confidence when findings converge.

Use validated, standardized instruments where they exist; report reliability (e.g., inter-rater agreement) for observational measures.

Human-subjects data require IRB approval and HIPAA compliance; patient and staff data must be de-identified before analysis.

Statistical significance (p < 0.05) is not practical significance; report effect sizes and confidence intervals alongside p values.

Behavioral mapping, time-and-motion studies, and space-utilization counts quantify how settings are actually used rather than how they were intended to be used.

A charrette is an intensive, time-boxed multidisciplinary workshop used to generate, critique, and converge on design solutions with stakeholders present.

Core stakeholder groups: patients and families, clinicians, facilities and engineering, IT, administration and finance, and community representatives.

Recurring user group meetings and mock-up reviews give operational feedback that single briefing sessions do not capture.

Implementation requires a communication plan, staff training, phased occupancy, and change management, or the intended outcomes will not be realized in operation.

Dissemination channels: peer-reviewed journals (HERD), conferences (Healthcare Design, IHI), Pebble Project case studies, and organizational knowledge repositories.

Report negative and null findings as well as positive ones, and describe project context so others can judge transferability.

Close the loop: post-occupancy findings feed back into organizational design standards and the shared EBD knowledge base.

Record decisions, rationale, and metrics during the project so the evaluation can be replicated and defended later.