A Comparison of Los Angeles County Health Survey Outcomes After Transitioning From Telephone to a Primarily Web-Based, Self-Administration Data Collection Mode.

Abstract

Context: As response rates to health surveys conducted by telephone continue to fall while data collection costs rise, practitioners are increasingly transitioning to address-based sample (ABS) designs with a self-administered, mail contact data collection mode.

Objective: To compare differences in key health indicators produced from both the Adult Survey and Child Survey of the Los Angeles County Health Survey (LACHS), which transitioned from a telephone to primarily self-administered mode in the 2022-2023 administration.

Design: Weighted survey estimates from the 2015, 2018, and 2022-2023 LACHS administrations are compared to investigate differences that may be attributable to LACHS' recent mode transition.

Setting: All survey data were collected of residents of Los Angeles County between 2015 and 2023.

Main outcome measures: Response rates and key health indicators produced from the Los Angeles County Health Survey.

Results: Introducing the self-administration mode led to modest response rate increases of approximately 10% to 16% for the Adult Survey and from 10% to 14% in the Child Survey. Key health indicator differences are mixed, but generally larger in magnitude across the mode transition, and also generally larger for the Adult Survey relative to the Child Survey.

Conclusions: Transitioning a population health survey from a telephone mode to a primarily self-administration mode using an ABS design comes with tradeoffs. Increased response rates and a greater ability to target lower-level geographies and other population domains of interest may be offset by mode effects that cannot be compensated for by weighting adjustments.