Nonresponsive Bias: An Analysis of Data From A Study of the Economic Consequences of Cancer Survivorship

2.50
Hdl Handle:
http://hdl.handle.net/10755/165400
Category:
Abstract
Type:
Presentation
Title:
Nonresponsive Bias: An Analysis of Data From A Study of the Economic Consequences of Cancer Survivorship
Author(s):
Davitt, Margaret; Polomano, R.; Short, P.
Author Details:
Margaret Davitt, Pennsylvania State University, The Milton S. Hershey Medical Center, Hershey, Pennsylvania, USA, email: mdavitt@psu.edu; R. Polomano; P. Short
Abstract:
Non-participation bias, referred to as non-response bias or non-response error, involves recruitment of samples with preferential selection of subjects that can potentially jeopardize the interpretation of results from survey research because samples do not adequately reflect the population. In a multi-site study of cancer survivors, “The Economic Consequences of Cancer Survival", (PI - Short, 1998), non-participation bias is analyzed to elucidate the characteristics of non-responders. One thousand eligible subjects (cancer diagnosis 1 to 5 yr prior, < Stage IV for solid tumors, age 25 to 60 yr) were identified through tumor registries from 4 sites, contacted by mail to participate, and interviewed by phone, if consented, to determine the impact of cancer on economic state. We tested the hypotheses that variations in responder characteristics (N=318; response rate 29%) for age, gender, marital status, type and stage of cancer, and ability to acknowledge a prior diagnosis of cancer were not statistically different from those declining participation (N=777). No differences were found between age groups of responders and non-responders, except for a higher percentage of non-responders were age 55-62 yr (Chi-square, p<0.05). Using Student t-tests for independent samples, statistical equivalence was established between male responders (36.5%) and non-responders (41.1%) and female responders (63.5%) and non-responders (58.8%). An analysis of gender by marital status showed a higher (Chi-square, p<0.05) percentage of unmarried males (10.3%) did not participate compared to unmarried males who participated (6.3%). For women with breast cancer, greater numbers participated than those who did not. Of those with unknown cancer staging, a higher percentage (p<0.05) was in the non-responder group (6.1%) compared to responders (3.1%). Chi-square analysis identified overall statistically significant difference in participation (p<0.01). Based on the results, disparities in sample characteristics between survey responders and non-responders were evident. Persons of older age, 55-62 yr, were more likely to decline participation, which may reflect a lower level of interest and perceived importance of economic-related issues. Investigators must identify barriers to survey participation and risks for non-participation. Strategies can be employed to optimize participation of eligible participants and statistical adjustments can be performed to control for non-participation bias.
Repository Posting Date:
27-Oct-2011
Date of Publication:
27-Oct-2011
Conference Date:
2003
Conference Name:
28th Annual Oncology Nursing Society Congress
Conference Host:
Oncology Nursing Society
Conference Location:
Denver, Colorado, USA
Note:
This is an abstract-only submission. If the author has submitted a full-text item based on this abstract, you may find it by browsing the Virginia Henderson Global Nursing e-Repository by author. If author contact information is available in this abstract, please feel free to contact him or her with your queries regarding this submission. Alternatively, please contact the conference host, journal, or publisher (according to the circumstance) for further details regarding this item. If a citation is listed in this record, the item has been published and is available via open-access avenues or a journal/database subscription. Contact your library for assistance in obtaining the as-published article.

Full metadata record

DC FieldValue Language
dc.type.categoryAbstracten_US
dc.typePresentationen_GB
dc.titleNonresponsive Bias: An Analysis of Data From A Study of the Economic Consequences of Cancer Survivorshipen_GB
dc.contributor.authorDavitt, Margareten_US
dc.contributor.authorPolomano, R.en_US
dc.contributor.authorShort, P.en_US
dc.author.detailsMargaret Davitt, Pennsylvania State University, The Milton S. Hershey Medical Center, Hershey, Pennsylvania, USA, email: mdavitt@psu.edu; R. Polomano; P. Shorten_US
dc.identifier.urihttp://hdl.handle.net/10755/165400-
dc.description.abstractNon-participation bias, referred to as non-response bias or non-response error, involves recruitment of samples with preferential selection of subjects that can potentially jeopardize the interpretation of results from survey research because samples do not adequately reflect the population. In a multi-site study of cancer survivors, &ldquo;The Economic Consequences of Cancer Survival&quot;, (PI - Short, 1998), non-participation bias is analyzed to elucidate the characteristics of non-responders. One thousand eligible subjects (cancer diagnosis 1 to 5 yr prior, &lt; Stage IV for solid tumors, age 25 to 60 yr) were identified through tumor registries from 4 sites, contacted by mail to participate, and interviewed by phone, if consented, to determine the impact of cancer on economic state. We tested the hypotheses that variations in responder characteristics (N=318; response rate 29%) for age, gender, marital status, type and stage of cancer, and ability to acknowledge a prior diagnosis of cancer were not statistically different from those declining participation (N=777). No differences were found between age groups of responders and non-responders, except for a higher percentage of non-responders were age 55-62 yr (Chi-square, p&lt;0.05). Using Student t-tests for independent samples, statistical equivalence was established between male responders (36.5%) and non-responders (41.1%) and female responders (63.5%) and non-responders (58.8%). An analysis of gender by marital status showed a higher (Chi-square, p&lt;0.05) percentage of unmarried males (10.3%) did not participate compared to unmarried males who participated (6.3%). For women with breast cancer, greater numbers participated than those who did not. Of those with unknown cancer staging, a higher percentage (p&lt;0.05) was in the non-responder group (6.1%) compared to responders (3.1%). Chi-square analysis identified overall statistically significant difference in participation (p&lt;0.01). Based on the results, disparities in sample characteristics between survey responders and non-responders were evident. Persons of older age, 55-62 yr, were more likely to decline participation, which may reflect a lower level of interest and perceived importance of economic-related issues. Investigators must identify barriers to survey participation and risks for non-participation. Strategies can be employed to optimize participation of eligible participants and statistical adjustments can be performed to control for non-participation bias.en_GB
dc.date.available2011-10-27T12:17:52Z-
dc.date.issued2011-10-27en_GB
dc.date.accessioned2011-10-27T12:17:52Z-
dc.conference.date2003en_US
dc.conference.name28th Annual Oncology Nursing Society Congressen_US
dc.conference.hostOncology Nursing Societyen_US
dc.conference.locationDenver, Colorado, USAen_US
dc.description.noteThis is an abstract-only submission. If the author has submitted a full-text item based on this abstract, you may find it by browsing the Virginia Henderson Global Nursing e-Repository by author. If author contact information is available in this abstract, please feel free to contact him or her with your queries regarding this submission. Alternatively, please contact the conference host, journal, or publisher (according to the circumstance) for further details regarding this item. If a citation is listed in this record, the item has been published and is available via open-access avenues or a journal/database subscription. Contact your library for assistance in obtaining the as-published article.-
All Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated.