Telephone Sampling in Epidemiologic Research: To Reap the Benefits, Avoid the Pitfalls

Abstract

Recently, the use of telephone sampling methods in epidemiology has been sharply increasing. Properly applied, these methods provide powerful tools. Improperly applied, they may produce invalid results. This review covers many points to which the investigator should be alert. An underlying theme is that bias in studies that use telephone sampling can potentially spring from many sources and should be avoided wherever feasible. In epidemiology, there are two main uses of telephone sampling--in general surveys (cross-sectional studies) and in case-control studies. For the former, the principles differ little from those for general surveys in other fields. For the latter, most of the same principles apply, but case-control studies also have their own unique aspects. In this review, several topics receive detailed treatment. Valid combinations of area code and prefix can be found through careful processing of a file that is available commercially. Three options that can be used singly or in any combination provide broadened adaptability for the Mitofsky-Waksberg method of random digit dialing. Bias can be thwarted by certain steps in the interviewing and by weighting. For population-based and then center-based case-control studies, a scheme that samples controls from household censuses and avoids usual problems is offered.

Impact of COPD in North America and Europe in 2000: subjects' perspective of Confronting COPD International Survey

“…It is quite comparable with the prevalence estimates obtained in NHANES III [32], which estimated diagnosed COPD to have a prevalence of 3.1%, asthma of 2.7% and undiagnosed airflow obstruction of 12.0%. The limitations of random-digit dialling have been discussed elsewhere [24]. One limitation is the availability of telephone lines.…”

“…Recent telephone coverage reports indicate that the percentage of households with a telephone in 1999 wasw94% in the countries included in the survey, with the exception of Germany (89.3%) [23]. The sampling method used telephone numbers (listed and unlisted) randomly generated by a truncated listassisted random-digit dialling sampling procedure [24]. Participation was voluntary, confidential and anonymous.…”

Design, methods and demographics from phase I of Alberta's Tomorrow Project cohort: a prospective cohort profile

“…In wave 1, attempts were made to recruit a second household member; this practice was discontinued in subsequent waves because the assumption of random sampling in random digit dialling was not maintained. 16 Using a standard script, the random digit dialling team explained the rationale for Alberta's Tomorrow Project and checked eligibility against 4 criteria: age between 35 and 69 years; no previous personal history of cancer, other than nonmelanoma skin cancer; plans to reside in Alberta for at least 1 year; and ability to complete written questionnaires in English. Interested people were sent a cover letter, consent form, study information booklet, Health and Lifestyle Questionnaire and measuring tape (for anthropometric measurements).…”

“…20 These questionnaires had either been adapted (diet) or specifically developed and tested for reliability and validity (physical activity) for use in this cohort. Nonrespondents to the diet and activity questionnaires were sent a reminder postcard 6 weeks after the initial mailing and a replacement package after 16 23 lifetime shift work and environmental exposures, including the built environment 24 and types of occupations (employment for ≥ 6 mo). Current residence, lifetime residential history (street address, city or town, country, postal code, dates of occupancy of lifetime residences ≥ 1 yr), and birth places of participants, their parents and grandparents were also captured (Table 1).…”

Public Health Surveillance of Low-Frequency Populations

“…The frame of an RDD study is households or persons with telephones, which may underrepresent the lowest income families of greatest interest. It is also necessary to account for the number of phone numbers that a person has to ensure that all persons in the frame have an equal probability of being sampled (73).…”