Let us not forget telescoping as a major risk of telephone surveys
Laura Cunha Rodrigues, London School of HYgiene and Tropical Medicine
4 September 2006
Mairin Boland et al(1) discuss aspects of telephone surveys in which morbidity data are collected by asking subjects whether they experienced an incidence disease over a period of time. They identify as potential limitations of this approach lack of representativeness, low response rates, and costs. They do not mention forward telescoping, the well documented distortion of results caused by a tendency to over report recent events.(2)
When asked to recall whether they have experienced morbidity in the recent past (for example whether they had diarrhoea in the previous 3 weeks) people tend to disrespect boundaries of time. There is a body of research showing very consistently that over relatively short periods, up to a few months, the bias is towards over reporting, with events that happened before the period reported as having happened during the period.(2) Underreporting, backward telescoping, has been described rarely, usually when the period investigated covers several years.(3) To investigate a disease which is not routinely recorded in the health system (for example diarrhoea that did not lead to a health service consultation) the most common way of preventing telescoping is to ask about occurrence over a very short period of time (one or two days) or to use a form of concurrent active surveillance, like diaries.
An example is provided by the Study of Intestinal Infectious Disease (IID) in England.(4) The study estimated the frequency of diarrhoea in the community (including episodes that may not have resulted in a general practitioner consultation) by using active surveillance. Participants mailed a postcard to the research centre every week indicating whether or not they had diarrhoea. As the study included a question on recalled morbidity at the baseline survey, it was possible to compare both rates: the rate estimated by reported morbidity in the previous 3 weeks was markedly higher (55/100 person years), nearly 3 times the rate estimated by the active surveillance (19.4/100 person years). A second phase of the IID Study, recently commissioned by the Foods Standard Agency to estimate the frequency of gastrointestinal illness 10 years after the fist study, will compare rates from a prospective surveillance and from telephone surveys using different periods of recall.
Telescoping is a feature of estimates of levels of morbidity based on reported past disease, be it from telephone surveys or personal interviews; but in their assessment of the advantages and disadvantages of these two approaches the authors failed to consider the importance of this bias. Given the increasing popularity of the use of telephone surveys to estimate morbidity, with an apparent lack of concern for telescoping, we feel it is important to remind the reader that one of the potential limitations of the approach is the risk of getting the wrong result!
1. Boland M, Sweeney MR, Scallan E, Harrington M, Staines A. Emerging advantages and drawbacks of telephone surveying in public health research in Ireland and the U.K BMC Public Health 2006, 6:208
2. Armstrong B, Whote E and Saracci R. Principles of exposure measurement in epidemiology. Oxford University press, oxfrd, UK, 1992. Pages 24 and 199.
3. Prohaska V, Brown NR, Belli RF. Forward telescoping: the question matters.
Memory. 1998 Jul;6(4):455-65
4. Wheeler JG, Sethi D,Cowden JM,Wall PG, Rodrigues LC,Tompkins DS,Hudson MJ, Roderick PJ Study of infectious intestinal disease in England: rates in the community, presenting to general practice, and reported to national surveillance. The Infectious Intestinal Disease Study Executive. BMJ. 1999 Apr 17;318(7190):1046-50
Laura C Rodrigues*, Clarence Tam*, Grace Lordan+, Charles Normand+, Donal Martin+, Craig Higgins*
Let us not forget telescoping as a major risk of telephone surveys
4 September 2006
Mairin Boland et al(1) discuss aspects of telephone surveys in which morbidity data are collected by asking subjects whether they experienced an incidence disease over a period of time. They identify as potential limitations of this approach lack of representativeness, low response rates, and costs. They do not mention forward telescoping, the well documented distortion of results caused by a tendency to over report recent events.(2)
When asked to recall whether they have experienced morbidity in the recent past (for example whether they had diarrhoea in the previous 3 weeks) people tend to disrespect boundaries of time. There is a body of research showing very consistently that over relatively short periods, up to a few months, the bias is towards over reporting, with events that happened before the period reported as having happened during the period.(2) Underreporting, backward telescoping, has been described rarely, usually when the period investigated covers several years.(3) To investigate a disease which is not routinely recorded in the health system (for example diarrhoea that did not lead to a health service consultation) the most common way of preventing telescoping is to ask about occurrence over a very short period of time (one or two days) or to use a form of concurrent active surveillance, like diaries.
An example is provided by the Study of Intestinal Infectious Disease (IID) in England.(4) The study estimated the frequency of diarrhoea in the community (including episodes that may not have resulted in a general practitioner consultation) by using active surveillance. Participants mailed a postcard to the research centre every week indicating whether or not they had diarrhoea. As the study included a question on recalled morbidity at the baseline survey, it was possible to compare both rates: the rate estimated by reported morbidity in the previous 3 weeks was markedly higher (55/100 person years), nearly 3 times the rate estimated by the active surveillance (19.4/100 person years). A second phase of the IID Study, recently commissioned by the Foods Standard Agency to estimate the frequency of gastrointestinal illness 10 years after the fist study, will compare rates from a prospective surveillance and from telephone surveys using different periods of recall.
Telescoping is a feature of estimates of levels of morbidity based on reported past disease, be it from telephone surveys or personal interviews; but in their assessment of the advantages and disadvantages of these two approaches the authors failed to consider the importance of this bias. Given the increasing popularity of the use of telephone surveys to estimate morbidity, with an apparent lack of concern for telescoping, we feel it is important to remind the reader that one of the potential limitations of the approach is the risk of getting the wrong result!
1. Boland M, Sweeney MR, Scallan E, Harrington M, Staines A. Emerging advantages and drawbacks of telephone surveying in public health research in Ireland and the U.K BMC Public Health 2006, 6:208
2. Armstrong B, Whote E and Saracci R. Principles of exposure measurement in epidemiology. Oxford University press, oxfrd, UK, 1992. Pages 24 and 199.
3. Prohaska V, Brown NR, Belli RF. Forward telescoping: the question matters.
Memory. 1998 Jul;6(4):455-65
4. Wheeler JG, Sethi D,Cowden JM,Wall PG, Rodrigues LC,Tompkins DS,Hudson MJ, Roderick PJ Study of infectious intestinal disease in England: rates in the community, presenting to general practice, and reported to national surveillance. The Infectious Intestinal Disease Study Executive. BMJ. 1999 Apr 17;318(7190):1046-50
Laura C Rodrigues*, Clarence Tam*, Grace Lordan+, Charles Normand+, Donal Martin+, Craig Higgins*
*London School of Hygiene and Tropical Medicine
+Trinity College, Dublin.
Competing interests
No competing interest