Back to Annual Meeting

Back to Annual Meeting

APHA Scientific Session and Event Listing

Recorded presentation

Fatma Shebl, MD, MS, MHS¹, Annette Hopkins, RN MS¹, Faye Grillo, RN, MS², Ebenezer Israel, MD MPH¹, Diane M. Dwyer, MD³, Carmela Groves, RN MS³, Min Zhan, PhD¹, Jeanetta Churchill, MS⁴, Chinnadurai Devadason, MD, DPH², and Eileen K. Steinberger, MD MS¹. (1) Department of Epidemiology and Preventive Medicine, University of Maryland, Baltimore, Howard Hall, 660 West Redwood Street, Baltimore, MD 21201, 4107676967, fshebl@dhmh.state.md.us, (2) Department of Health - Charles County, P.O. Box 1050, White Plains, MD 20695, (3) Center for Cancer Surveillance and Control, Maryland Department of Health and Mental Hygiene, 201 West Preston Street, Baltimore, MD 21201, (4) Urban Spatial Analytics Program, University of Pennsylvania, Meyerson Hall, Room G-12, 210 South 34th Street, Philadelphia, PA 19104-6311

Background: Population-based landline telephone interview surveys are possibly biased due to inclusion of only those with landline telephones. Bias may be more serious in low-income populations. Methods: We conducted face-to-face interviews with 502 low-income adults, >18 years old, in rural Charles County, Maryland. We compared estimates from those with and without landline phones; for those variables that were significantly different, the difference between whole sample estimates and estimates from those with landline phones were calculated to quantify the magnitude of phone coverage bias. Results: Of 499 respondents, 80 (16%) did not have landline phones. Significant differences were found between those with and without landline phones. The absolute differences between percentages from the entire sample and percentages made including only those with landline phones ranged from 0 to 6.5 percentage points. The observed percentages for the entire sample versus percents based on those with landline phones were: African American (71.9 vs. 75.1); folic acid consumption; (30.8 vs. 33.4) having health insurance (67.4 vs. 70.0); having routine medical check-up within the last year (73.9 vs. 74.9); current smokers (35.6 vs. 32.1); current alcohol drinkers (25.6 vs. 23.6); having age-appropriate colorectal cancer screening (54.2 vs. 59.2); having mammograms (88.5 vs. 90.7); and having prostate cancer screening (53.1 vs. 59.6). Conclusion: Our data suggest there is some bias in estimates from low-income households with landline surveys. Special surveys are needed to obtain risk information from excluded subpopulations.

Learning Objectives:

• 1- To describe possible bias in landline telephone surveys among low income populations
• 2- Discuss implications for telephone survey research

Keywords: Survey, Methodology

Presenting author's disclosure statement:

Any relevant financial relationships? No

Recorded presentation