4.3 Literature Searches, Academic Databases & Search Strategies

Literature Searches, Academic Databases & Search Strategies

In The Center for Health Design's 8-step Evidence-Based Design process, Step 2 is "Find sources for relevant evidence." Once an interdisciplinary project team defines its strategic vision, project goals, and performance targets in Step 1, the team must systematically search the global body of knowledge to discover what empirical research exists to guide their design decisions.

Executing a rigorous literature search in healthcare architecture presents a unique challenge: the literature is radically interdisciplinary. Relevant studies do not reside in a single architectural library. Instead, crucial findings are dispersed across biomedical science, nursing administration, environmental psychology, human factors engineering, infection epidemiology, and acoustics. To find credible evidence, EBD practitioners must employ structured search frameworks and advanced search mechanics.

---

The Literature Review in EDAC Study Materials

EDAC Study Guide 2 describes the literature review as a review of existing evidence and relevant theoretical thinking on a research topic. After a research question is identified, three steps make up a literature review:

1. Locate information
2. Critically evaluate and interpret the evidence
3. Record and synthesize the information

A well-performed review helps the team identify gaps in current knowledge, determine what relevant research has already been performed, and inform the basis for new research.

Types of Sources

To judge relevance to the research plan, note each source's date, location, scope of work, source, and internet address.

Evidence Sources Named in CHD's Current Exam Content Outline

The outline asks candidates to identify sources and collect and analyze existing evidence related to research questions. It lists a broad set of sources—not only journals:

Benchmarking: Internal and External

The outline repeatedly mentions benchmarking (external and internal):

• Internal benchmarking compares performance within the organization—across units, campuses, or its own past results.
• External benchmarking compares performance with peer organizations or national databases (for example, nurse-sensitive quality databases or publicly reported CMS measures).

Benchmarks help set realistic targets and show whether a problem is local or widespread, but they are not proof that a design caused a difference.

Judging Relevance

The outline lists factors for determining the relevance of evidence to the project: the source, author qualifications and/or experience, appropriateness of the research methodology, replication, and composition of the sample. A well-designed study in a very different population or care setting may be less relevant than a moderately designed study that closely matches the project.

---

Formulating the Search Query: Adapting PICO for Design

In Evidence-Based Medicine (EBM), researchers formulate clinical search queries using the well-established PICO framework:

• P: Patient / Population
• I: Intervention
• C: Comparison / Control
• O: Clinical Outcome

Because design interventions change physical spaces rather than administering drugs, EBD teams often adapt PICO so the elements describe settings and design features. This guide uses the label SECO as a memory aid for that adaptation; it is not an official CHD term:

   CLINICAL PICO                             EBD ADAPTED SECO
┌──────────────────────┐                  ┌──────────────────────┐
│ Patient / Population │  ──────────────> │ Setting & Population │ [S]
├──────────────────────┤                  ├──────────────────────┤
│     Intervention     │  ──────────────> │Environmental Interv. │ [E]
├──────────────────────┤                  ├──────────────────────┤
│ Comparison / Control │  ──────────────> │Comparison / Baseline │ [C]
├──────────────────────┤                  ├──────────────────────┤
│   Clinical Outcome   │  ──────────────> │  Measurable Outcome  │ [O]
└──────────────────────┘                  └──────────────────────┘

The Adapted Components

1. Setting & Population (S): The specific healthcare environment, care acuity level, and human occupants under study.
   • Examples: Adult intensive care unit (ICU), freestanding pediatric emergency department, memory-care assisted living facility, neonatal intensive care unit (NICU), post-anesthesia care unit (PACU); registered night nurses, bariatric surgical patients, cognitively impaired elderly.
2. Environmental Intervention (E): The physical, architectural, spatial, technological, or sensory design element being tested.
   • Examples: High-performance sound-absorbing ceiling tiles, circadian-supportive dynamic LED lighting, decentralized nurse charting stations, same-handed private patient rooms, ceiling-mounted patient lifts, dedicated view of nature.
3. Comparison / Baseline (C): The physical baseline, alternative layout, or control environment against which the intervention is measured.
   • Examples: Multi-bed ward, centralized nursing station, mirrored patient room layout, standard static fluorescent lighting (4000K constant), standard acoustic ceiling tiles, windowless room or wall-facing room.
4. Measurable Outcome (O): The objective clinical, operational, financial, or experiential endpoint targeted for improvement.
   • Examples: Central-line-associated bloodstream infection (CLABSI) incidence, unassisted patient fall rate per 1,000 patient days, average nurse travel distance per shift, medication administration errors (MAEs), HCAHPS nighttime noise scores, delirium duration (days).

Formulating an EBD Search Question: Practical Example

• Clinical Challenge: An elderly inpatient rehabilitation unit suffers high fall rates during night hours.
• Adapted PICO Question: In inpatient rehabilitation facilities housing geriatric patients [S], does the installation of automated motion-sensor pathway night lighting [E], compared to conventional manual wall switches [C], reduce unassisted patient falls and fall-related fractures [O]?

---

Primary Academic Databases for Healthcare Design

An evidence-based design team should search multiple indexed databases to capture the full spectrum of healthcare design research. General search engines are a useful starting point—Google Scholar in particular—but they return a mix of research, marketing, and press releases, so every result needs the same screening, and indexed databases provide more complete, structured coverage.

1. PubMed / MEDLINE

• Host: United States National Library of Medicine (NLM) at the National Institutes of Health (NIH).
• Scope: Over 35 million citations across medicine, nursing, clinical epidemiology, and health services research.
• EBD Value: The definitive database for clinical outcome studies (e.g., healthcare-associated infections, medical errors, surgical wound healing, circadian photobiology).
• Search Strength: Utilizes Medical Subject Headings (MeSH)—a standardized controlled vocabulary that indexes articles regardless of variations in author terminology (e.g., searching the MeSH term "Hospital Design and Construction" captures facilities, architecture, room design, and layout).

2. CINAHL (Cumulative Index to Nursing and Allied Health Literature)

• Host: EBSCOhost.
• Scope: The authoritative bibliographic database for nursing, physical therapy, occupational health, and allied health professions.
• EBD Value: Unmatched coverage of bedside care delivery, nurse walking distances, physical fatigue, musculoskeletal patient-handling injuries, medication preparation distractions, and patient comfort.
• Search Strength: Highly specific indexing of nursing workflows and operational environments.

3. PsycINFO

• Host: American Psychological Association (APA).
• Scope: Over 5 million records covering psychology, behavioral science, cognitive neuroscience, and environmental psychology.
• EBD Value: Exceptional resource for psychiatric hospital design, dementia and memory care environments, pediatric sensory processing, positive distractions, stress recovery, and psycho-neuro-immunology.

4. Web of Science & Scopus

• Scope: Massive multidisciplinary citation indexing databases covering sciences, social sciences, arts, and humanities.
• EBD Value: Crucial for forward and backward citation tracking. Researchers can locate a landmark study (such as Roger Ulrich's 1984 Science paper) and instantly track every subsequent study that has cited it over the past four decades, mapping the evolutionary tree of an environmental design theory.

5. The Center for Health Design (CHD) Knowledge Repository

• Host: The Center for Health Design (healthdesign.org).
• Scope: A curated, searchable collection of healthcare design research, including key point summaries for many studies.
• EBD Value: Pre-filters the literature for healthcare design relevance, saving screening time.

Database Comparison Matrix

---

Advanced Search Mechanics: Boolean Logic and Syntax

Navigating academic databases requires mastering Boolean operators and advanced search syntax. Because healthcare and design professionals use different words to describe identical concepts, search strings must be carefully constructed to avoid omitting critical evidence.

1. The Core Boolean Operators

• AND: Narrows the search. All terms must appear in the citation.
  • Example: "intensive care" AND acoustics AND delirium (Retrieves only papers containing all three concepts).
• OR: Broadens the search. Connects synonyms, alternative spellings, and related terms. Essential for capturing architectural synonyms.
  • Example: daylight OR sunlight OR "natural light" OR fenestration (Retrieves papers containing any of these terms).
• NOT: Excludes specific terms. Must be used with extreme caution because it risks inadvertently eliminating relevant papers.
  • Example: acoustics AND hospital NOT neonatal (Excludes NICU acoustic studies, but also excludes adult studies that mention neonatal units in passing).

2. Nesting with Parentheses

When combining AND and OR operators, parentheses dictate the mathematical order of operations. Omitting parentheses causes database search engines to execute operators unpredictably, corrupting search results.

• Proper Nested Query: ("intensive care" OR ICU) AND (acoustics OR noise OR "sound levels") AND (sleep OR delirium) (This query tells the database: Find articles that have at least one term from the setting group, AND at least one term from the environmental group, AND at least one term from the outcome group.)

3. Truncation and Wildcards

Truncation allows searching for the root of a word and all its variable endings using an asterisk (*):

• pediatr* retrieves: pediatric, pediatrics, pediatrician, paediatric.
• light* retrieves: light, lights, lighting.
• fall* retrieves: fall, falls, falling.

4. Phrase Searching with Quotation Marks

Enclosing multi-word terms in quotation marks forces the database to search for the exact adjacent phrase rather than treating words as separate terms:

• Searching "decentralized nurse station" searches for the complete phrase.
• Searching without quotes (decentralized nurse station) retrieves papers discussing decentralized heating systems that happen to mention a nursing station in a hospital.

---

Gray Literature: Value, Risks, and Critical Appraisal

Not all relevant built-environment evidence is published in traditional indexed academic journals. In healthcare design, a substantial volume of data exists as gray literature—information produced by government agencies, academic institutions, professional associations, and commercial corporations that is not published through commercial academic publishing channels.

Typologies of Gray Literature in EBD

• Academic Theses and Dissertations: Masters and doctoral research containing extensive primary field data, spatial floor plans, and exhaustive bibliographies.
• Government and Agency Technical Monographs: In-depth publications from NIOSH, AHRQ, CDC, and the Veterans Health Administration (VHA), detailing facility guidelines and safety engineering.
• Architectural Firm White Papers & Case Studies: High-quality project profiles, post-occupancy walkthroughs, and spatial programming methodologies.
• Manufacturer-Sponsored Research Studies: Technical laboratory reports on acoustic ceiling absorption coefficients, antimicrobial copper touch surfaces, slip-resistant flooring durability, or tunable LED spectral power distributions.
• Conference Proceedings & Presentations: Emerging findings presented at the Healthcare Design Conference (HCD), Environmental Design Research Association (EDRA), or European Healthcare Design Congress.

The Dual Nature of Gray Literature: Strengths vs. Risks

┌────────────────────────────────────────┐   ┌────────────────────────────────────────┐
│      STRENGTHS OF GRAY LITERATURE      │   │        RISKS OF GRAY LITERATURE        │
├────────────────────────────────────────┤   ├────────────────────────────────────────┤
│ • Mitigates Publication Bias (captures │   │ • Lack of Rigorous Blind Peer Review   │
│   null or negative study findings)     │   │   (methodology may be flawed)          │
│ • Timeliness (captures emerging trends │   │ • Potential Commercial Bias & Conflict │
│   without 2-year journal lag)          │   │   of Interest (vendor-funded claims)   │
│ • Rich Physical Details (detailed      │   │ • Limited Methodological Transparency  │
│   blueprints, MEP specs, photography)  │   │   (omitted sample sizes or statistics) │
└────────────────────────────────────────┘   └────────────────────────────────────────┘

[!CAUTION]

EXAM TRAP: The Status of Manufacturer White Papers

A common scenario involves a building product manufacturer that provides an impressive white paper claiming their product drastically improves clinical outcomes (e.g., "Our antimicrobial wall cladding reduces hospital-acquired infections by 50%").

Rule of Thumb: Manufacturer white papers are gray literature. While they provide useful technical specifications, their outcome claims must be treated with professional skepticism due to inherent commercial conflicts of interest and lack of independent peer review. EBD decisions must never rest solely on manufacturer marketing materials without independent, peer-reviewed corroboration.