Skip to Main Content
University of Detroit Mercy logo

Health Operations Management

evidence based medicine pyramid
Page contains:
  • information for constructing PICOT questions
  • best study designs for question types (i.e. intervention, diagnosis, prognosis, etc.)
  • evidence based practice filters in PubMed
  • levels (hierarchy) of evidence information
  • tutorials to assist with identifying evidence based practice articles and identifying individual (single) research studies
  • resources for analyzing research studies

Evidence based Practice Research for Administration or Operations Management

Sample Areas for Research:

Informing Policy and Procedures:
  • policies and procedures related to patient care
  • implementing electronic health record systems
  • optimizing patient flow
  • emergency response and disaster preparedness
Improving Administrative Processes:
  • optimize administrative workflows
  • appointment scheduling systems (e.g. research on patient preferences and appointment access)
  • effective patient communication strategies
  • staff scheduling or shift optimization
  • resource allocation research (e.g. inventory management, bed management, capacity planning)
  • design and layout of waiting rooms
  • analytics and decision support
Enhancing Patient Education:
  • up-to-date information about their health conditions and treatment options

Using a PICO Question for Searching the Literature

PICO/T is a way to format a research question. It can also get you started with keywords for your literature search.

 

P = Population of interest (consider age, gender, race, ethnicity, disease process, comorbidities)

I = Intervention (exposure to disease, prognostic factor A, risk behaviors, treatment, what do you want to do for this population? what could be done better?)

C = Comparison of interest (no comparison, placebo, prognostic factor B, absence of risk behavior, other treatments - "gold standard")

O = Outcome of interest (what result are you looking for? risk of disease, rate of occurrence of adverse outcomes like illness, comorbidity, or death)

T = Time (how long does it take to demonstrate an outcome? how long are participants observed? - if relevant to your question)

 
Non-Clinical Examples:

1. For healthcare organizations (P), does the adoption of a value-based care model (I) compared to maintaining a fee-for-service payment model (C) lead to improved patient outcomes (O)?

  • initial keywords: value-based care model, fee-for-service, patient outcomes

2. For patients visiting an outpatient specialty clinic (P), does the implementation of a patient-centered scheduling system (I) compared to traditional block scheduling methods (C) lead to reduced patient wait times (O)?

  • initial keywords: patient centered scheduling, block scheduling, wait time

3. For front-desk staff at a multi-specialty outpatient clinic (P), does the implementation of a comprehensive customer service training program (I) compared to existing, unstandardized, and optional training (C) lead to improved patient satisfaction scores (O)?

  • initial keywords: customer service training, patient satisfaction

 

Clinical Examples

1. P = menopausal women;  I = cranberry juice; C = no cranberry juice; O = incidence of UTI

In menopausal women, does drinking 1 cup of cranberry juice daily versus not drinking any cranberry juice lower the incidence of urinary tract infections?

Initial keywords: menopause, cranberry juice, urinary tract infections

2. P = adults with arthritis; I = tomatoes; C = No comparison; O = increased joint pain

Do tomatoes worsen joint pain in adults with arthritis?

Initial keywords: tomatoes, arthritis, pain

Reference:

Godshall, M. (2016). Fast facts for evidence-based practice in nursing (2nd ed.). Springer Publishing Company. DOI 10.1891/9780826194077

Best Study Designs for Clinical Question Types

Matching questions types  to study design

EBP Filters in PubMed

Methods for Limiting to Evidence Based Practice:

 

1. Filter by Article Types / Publication Types: (left column - see PubMed filters handout on Find Articles tab)

Clinical Study: automatically includes Clinical Trial, Controlled Clinical Trial, Randomized Clinical Trial, Observational Study

Comparative Study = comparison of outcomes, results, etc. for different techniques, therapeutic approaches or other inputs.

Meta-Analysis = (multi-study) - studies using a quantitative method of combining the results of independent studies and synthesizing summaries and conclusions which may be used to evaluate therapeutic effectiveness. 

Multicenter Study = a study conducted by several institutions

Systematic Reviews = (multi-study) - review of primary literature that attempts to identify, appraise, and synthesize all empirical evidence that meets specified eligibility criteria to answer a given research question. Systematic Reviews are NOT the same as Literature Reviews. Literature Reviews are NOT EBP.

Evaluation Study = studies determining the effectiveness of processes, personnel, and equipment

Validation Study = research using processes by which the reliability and relevance of a procedure for a specific purpose are established

Practice Guidelines: most are EBP, but some are still opinion-based. You will need to review the guideline to ensure it is EBP.


2. Include study type as a search term: (e.g. xxxx AND epidemiologic studies)

Epidemiologic Studies - Use for studies that examine associations, hypothesized causal relations. They are usually concerned with identifying or measuring the effects of risk factors or exposures.

  • Search term automatically includes: Cohort Studies, Longitudinal Studies, Prospective/Retrospective Studies, Case Control Studies, Cross Sectional Studies.

Qualitative Research - can also use the following search terms to capture the various types of qualitative studies:

  • focus groups OR phenomenology OR ethnography OR nursing methodology research OR narration OR thematic analysis 

Quasi-experimental designsare ideal for healthcare administration and quality improvement research because they can evaluate a large-scale organizational change that can't be ethically or practically randomized. Can also use the following subject term

  • Quasi-experimental OR "Non-Randomized Controlled Trials as Topic"[Mesh]

Interrupted time series (ITS) design - involves collecting data on key outcomes (e.g., average wait times) at multiple time points before and after the new protocol is implemented. This allows researchers to analyze trends and see if the intervention caused a significant change in the level of the outcomes. Can also use the following subject term

  • "Interrupted Time Series Analysis"[Mesh]

"Costs and Cost Analysis"[Mesh] - absolute, comparative, or differential costs pertaining to services, institutions, resources, etc., or the analysis and study of these costs.

 

Sample non-clinical study designs


Identifying Evidence based Practice and Single Studies

How do you know if an article is evidenced based practice
See the tutorial below
Need to identify single / individual studies from systematic reviews and meta-analysis? 
See the tutorial below

Resources for Analyzing and Appraising a Research Study

Decorative