Marginal Effects – Quantifying the Effect of Changes in Risk Factors in Logistic Regression Models

Details

• Personal Author:
  Dowd BE ; Maciejewski ML ; Norton EC
• Description:
  Marginal effects can be used to express how the predicted probability of a binary outcome changes with a change in a risk factor. For example, how does 1-year mortality risk change with a 1-year increase in age or for a patient with diabetes compared with a patient without diabetes? This approach can make the results more easily understood. Marginal effects often are reported with logistic regression analyses to communicate and quantify the incremental risk associated with each factor. In a 2013 article in JAMA Psychiatry, Cummings et al studied factors that predicted access to outpatient mental health facilities that accept Medicaid. Their main outcome had 3 categories, which were labeled "no access," "some access," and "good access." An ordered logistic regression model was developed and results were presented as the change in the probability of each outcome for a change in certain demographic factors. [...] Cummings et al described how changes in 4 county-level characteristics would change the predicted probability of either having no access or having good access to mental health outpatient treatment facilities that accept Medicaid (see Table 2 in the article). For example, an increase of 31 percentage points in the fraction of the county population living in a rural community (the standard deviation of that variable) would on average increase the probability of no access to mental health care by 27.9 percentage points (baseline risk = 34.8%) but would also increase the probability of good access by 3.4 percentage points (baseline risk = 20.2%), holding the effect of other explanatory variables constant. Such a change in rural population therefore would decrease the probability of some access, the third possible outcome, by 31.3 percentage points (27.9 + 3.4). Marginal effects are a useful way to describe the average effect of changes in explanatory variables on the change in the probability of outcomes in logistic regression and other nonlinear models. Marginal effects provide a direct and easily interpreted answer to the research question of interest. [Description provided by NIOSH]
• Subjects:
  Demography Mental Health Nervous System Occupational Health Risk Assessment Risk Factors Safety
• Keywords:
  Demographic Characteristics Health Care Mental Health Models Risk Analysis Risk Factors Statistical Analysis
• ISSN:
  0098-7484
• Document Type:
  Text
• Funding:
  Grant
• Genre:
  Journal Article
• Place as Subject:
  Massachusetts ; Michigan ; Minnesota ; North Carolina ; OSHA Region 1 ; OSHA Region 4 ; OSHA Region 5
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Volume:
  321
• Issue:
  13
• NIOSHTIC Number:
  nn:20064232
• Citation:
  JAMA 2019 Apr; 321(13):1304-1305
• Contact Point Address:
  Matthew L. Maciejewski, PhD, Durham Center of Innovation to Accelerate Discovery and Practice Transformation (ADAPT), Durham Veterans Affairs Health Care System, 508 Fulton St, Ste 600, Durham, NC 27705
• Email:
  matthew.maciejewski@va.gov
• Federal Fiscal Year:
  2019
• Performing Organization:
  University of Minnesota Twin Cities
• Peer Reviewed:
  True
• Start Date:
  20050701
• Source Full Name:
  Journal of the American Medical Association
• End Date:
  20250630
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:b3012557d4ec8933e2cb4e7c9c2aeec93caffc00bd7ede1b8f9f5d5ef5eabe4cb1e223257a96cfc6c092643e903ef4e5bad5decd23f7a093baf40f4d0710eaf1
• Download URL:
  https://stacks.cdc.gov/view/cdc/222733/cdc_222733_DS1.pdf
• File Type:
  [PDF - 50.03 KB ]