Supporting Statement A for:
The National Survey of Physician Attitudes Regarding the Care of Cancer Survivors (SPARCCS) (NCI)
October 1, 2008
(Anticipated date submission will be sent to HHS)
Paul Han, M.D.,
Task Order Monitor
National Institutes of Health
National Cancer Institute
Division of Cancer Control and Population Sciences
Applied Research Program
Health Services and Economics Branch
EPN 4005; 6130 Executive
Boulevard
Bethesda, MD 20892-7344
Phone: 301-402-3362
Fax: 301-435-3710
E-mail: potoskya@mail.nih.gov
TABLE OF CONTENTS
A. Justification 1
A.1. Circumstances Making the Collection of Information Necessary 1
A.2. Purpose and Use of the Information 2
A.3. Use of Improved Information Technology and Burden Reduction 4
A.4. Efforts to Identify Duplication and Use of Similar Information 5
A.5. Involvement of Small Entities 6
A.6. Consequences of Collecting the Information Less Frequently 6
A.7. Special Circumstances Relating to the Guidelines of 5 CFR 1320.5 6
A.8. Comments in Response to the Federal Register Notice and Efforts to Consult Outside the Agency 7
A.9. Explanation of Any Payment or Gift to Respondents 7
A.10. Assurance of Confidentiality Provided to Respondents 8
A.11. Justification for Sensitive Questions 10
A.12. Estimates of Burden Hours and Costs 10
A.13. Estimates of Other Total Annual Cost Burden to Respondents or Record Keepers 13
A.14. Annualized Cost to the Federal Government 13
A.15. Explanation for Program Changes or Adjustments 14
A.16. Plans for Tabulation and Publication and Project Time Schedule 14
A.17. Reason(s) Display of OMB Expiration Date is Inappropriate 21
A.18. Exceptions to Certification for Paperwork Reduction Act Submissions 21
ATTACHMENTS
1 Medical Oncologist Instrument
2 Primary Care Physician Instrument
3 Questionnaire design consultants
4 Statement Regarding the Privacy Act Systems of Records
5 Westat & NCI IRB Approvals
6 NCI Cover Letter, Letters of Support & FAQs
7 Non-Response Follow-Up Letters
8 Primary Care Physician Script
9 Medical Oncologist Script
10 Follow-up Call Scripts
11 FAX and email
A. JUSTIFICATION
Today, it is estimated that one in three persons will experience cancer in their lifetime. Due to advances in treatment, more people are surviving cancer; ten million Americans are now living with a personal history of this disease. After the active phase of cancer treatment is complete and the immediate threat of mortality postponed, cancer continues to have profound implications for the survivor’s ongoing health and health care needs. One in five adult cancer survivors who worked before their diagnosis have cancer-related limitations in their ability to work 1 to 5 years later. Psychological distress, sexual dysfunction, infertility, weakened organ function, cosmetic changes, and limitations in mobility, communication, and cognition are among the problems faced by some cancer survivors (Hewitt, 2005).
Understanding the transition from active cancer treatment to post-treatment care is crucial to our country’s long-term health. While active treatment is typically handled by cancer specialists, health care provided after the active phase of treatment may be handled by the cancer specialist, a primary care physician (PCP), or both. Little is known about the transition from active cancer treatment to follow-up care, and there is little evidence regarding optimal follow-up care practices. There is consensus, however, from the Institute of Medicine (IOM) that the care of survivors is fragmented and poorly coordinated, and that considerable practice variation exists in survivorship care. In 2005 the IOM published a report that described the essential components of cancer survivorship care within a health care delivery system. They include the following: (a) prevent cancers and other late effects; (b) monitor for new and recurring cancers and assess medical and psychosocial late effects; (c) intervene for consequences of cancer and its treatment; and (d) coordinate between specialists and primary care providers to ensure that all survivors’ health needs are met (Hewitt, 2005).
As the premier institution for cancer research at the National Institutes of Health (NIH), the National Cancer Institute (NCI) is responsible for the National Cancer Program, which consists of (1) an expanded, intensified, and coordinated cancer research program encompassing the research programs conducted and supported by the Institute, and the related research programs of the other national research institutes, and (2) the other programs and activities of the Institute. NCI’s Division of Cancer Control and Population Sciences (DCCPS) aims to reduce risk, incidence, and deaths from cancer. Within DCCPS, the mission of the Office of Cancer Survivorship (OCS) is to enhance the quality and length of survival of all persons diagnosed with cancer and to minimize or stabilize adverse effects experienced during cancer survivorship. OCS conducts and supports research that both examines and addresses the long- and short-term physical, psychological, social, and economic effects of cancer and its treatment among pediatric and adult survivors of cancer and their families. Section 410 of the Public Health Service Act (42 USC § 285) authorizes the collection of information for the Survey on Physicians’ Attitudes Regarding the Care of Cancer Survivors.
In this document we provide supporting information for approval by the Office of Management and Budget (OMB) for survey data collection under the Paperwork Reduction Act. The primary method for data collection will be a self-administered instrument sent via Priority Mail to primary care physicians and to oncologists.
To achieve this purpose, two distinct survey instruments will be administered: one to 1,100 oncologists (Attachment 1) and another to 1,100 primary care physicians (PCPs) (Attachment 2). The rationale for the dual survey is to permit comparisons of the perceived roles, knowledge and practices of these two key provider groups with regard to follow-up care for cancer survivors. Coordination of care across these two groups is repeatedly articulated as a key facet of care for survivors of adult cancer. The surveys focus on the treatment of patients surviving two common cancers -- breast cancer and colon cancer.
In addition, in order to facilitate improvements in the quality of follow-up care, it is essential to understand gaps between current and ideal follow-up care practices and potential reasons for those gaps.
Key analytic topics include determining:
Estimates of the proportion of physicians that engage in specific practices to monitor patients for recurrence and to screen for late and long-term effects;
Estimates of the proportion of physicians that agree that certain practices are optimal or ideal;
Difference in PCPs versus oncologists regarding their implementations and perceptions of optimal, or ideal practices;
Differences between academic vs. community PCPs and academic vs. community medical oncologists regarding practices and approaches to survivor care; and
Key barriers and facilitators to the implementation of optimal practices related to the care of cancer survivors.
These data will inform the process of developing best practices or guidelines for quality survivors care; augment the data collected in other cancer survivorship studies such as the Cancer Care Outcomes Research and Surveillance Consortium (CanCORS) and the Cancer Research Network (CRN); and monitor the progress being made toward achieving NCI strategic goals of improving the quality of cancer care across the cancer control continuum and specifically with respect to cancer survivorship. Information based on study findings will be disseminated through published papers and conference presentations, and is expected to reach a range of audiences within the U.S. public health research community and health care system.
Data will be collected through a pencil and paper survey distributed to physicians by Priority Mail. Physicians will be given the option of completing the questionnaire over the telephone with an interviewer who will record answers on paper. Upon returning the questionnaire the data will be key entered into a database. That database will be maintained and stored on Westat’s secure server. Westat, the contractor that will conduct the survey, found in two recent surveys (OMB No. 0925-0562 the National Survey of Primary Care Physicians’ Recommendations and Practice for Breast, Cervical, Colorectal, and Lung Cancer Screening Program and OMB No. 0930-0262 Evaluation of Buprenorphine Waiver Program Waivered Physician Survey) that fewer than 5 percent of the sample used the telephone option. Notably, these surveys achieved response rates of 68 and 80 percent using pencil and paper methods. The NCI Privacy Impact Assessment officer will be contacted to assess the need for a Privacy Impact Assessment for this study.
NCI has determined that the planned data collection activities do not duplicate any other current data collection efforts. A review of literature reveals consensus that cancer survivors represent an ever growing, distinct, and key population with unique health challenges owing to the late or long-term effects of cancer treatment and / or the cancer disease process itself. Management of these adverse health consequences across primary care and oncologist specialties is likely to require modifications in existing models of care as well as education and training of providers regarding cancer survivorship. While issues in quality and continuity of care have been identified in specific studies, little is known at the national level about current practices, perceptions and knowledge of oncologists and primary care physicians (Aziz, 2007; Hewitt, Greenfield,& Stovall, 2005; Hewitt, Weiner &Simone, 2003; President's Cancer Panel 2005/2006; President's Cancer Panel 2003/2004; Hewitt, Simone, 1999; Institute of Medicine, 2007; Oeffinger, 2006; Aziz et al. 2006; Ganz, 2006). The SPARCCS survey will fill these critical and oft repeated gaps in knowledge.
Over the last several years, NCI has spoken to numerous experts in the field and has presented at and attended multiple conferences on cancer survivorship. In that time, conversations with other attendees has yielded no information about any data collection efforts comparable to survey data collection described in this document. No surveys exist at this national scale and none focus specifically on post-treatment survivorship practices relevant to breast and colon cancer.
Some physicians will be in private practice and thus may be considered small entities. The instrument has been designed to minimize burden, including an easy-to-read format, checkboxes for responses, and consistent response choices across questions. Time-to-complete is estimated at 20 minutes or less.
Funding has been allocated for a one-time survey of PCPs and oncologists. If this information is not collected, there will be no national information available about PCP and oncologists practices and attitudes regarding cancer survivors.
There is a special circumstance related to the collection of information in which the 5 CRF 1320.5 are not met. Physicians who do not respond to this voluntary, one-time survey within several weeks of receiving the survey are mailed a second survey and receive a follow up call. This is in contrast to the four weeks indicated by the Guidelines of 5 CRF 1320.5. In past experiences it is known that physicians’ offices are very busy and tend to either respond quickly to requests for information or not respond at all. A prompt second mailing and a follow up call capitalizes on the recency of the previous mailed contact and encourages a response. All other aspects of this data collection are consistent with the information collection guidelines in 5 CFR 1320.5.
A 60‑day Federal Register notice soliciting comments prior to initial submission to OMB was published in the Federal Register on July 31, 2008, Volume 73, Number 148, Page 44751. In response to the notice, there were no requests for information received.
The Oncologist and Primary Care Physician Questionnaires (Attachments 1 & 2, respectively) were developed with the consultation of several physicians and subject matter experts during the development period of the instrument. The names, addresses and affiliations of the consultants are listed in Attachment 3.
Payment for participating in an interview or survey is standard practice when seeking participation of professionals such as physicians. The incentive payment is an effective method of drawing physicians’ attention to the study and gaining their cooperation in completing the survey. It is not intended to be a payment for their time. Historically, physicians are one of the most difficult survey populations, partly because of the number of surveys they receive as well as the demands on their professional time, so incentives assume an even greater importance with this group.
NCI believes that in order to achieve an adequate response rate for this survey, it is essential to offer a modest incentive of $50. There is considerable evidence in the literature showing that the most effective way to increase response rates among professionals (particularly physicians) is by offering a monetary incentive. In a survey of physicians, Gunn and Rhodes (1981) found the response rate to an initial survey with no incentive was 58 percent, with a $25 incentive, 69 percent, and with a $50 incentive, 77 percent, with the difference between the $50 and the $25 incentive rate being statistically significant. Based on these results, they decided to offer the higher incentive ($50) to achieve the highest response rate possible. Many other studies confirm that a monetary incentive is the best way to achieve acceptable response rates from physicians. Similarly, a $50 incentive was used successfully in 2002 and 2004 by Westat to obtain response rates above 80% for two unrelated physician surveys (OMB No. 0930-0246 and 0930-0262).
We propose to use a modest incentive of $50 in order to maximize response rates. Some studies show that physicians may be becoming accustomed to the much greater monetary incentives ($100-$150) offered by others such as pharmaceutical companies. This shift in expectations is thought, in some cases, to render surveys without an incentive meaningless. For this reason, NCI feels that an incentive of at least $50 is essential to attract enough attention to the survey to achieve acceptable response rates. A $50 incentive for completing a Federally-sponsored survey about a subject of importance to public health should be high enough to communicate the importance of physicians’ responses to the survey and to gain their attention.
The Privacy Act applies to this data collection, though data will be collected by Westat, and data files delivered to NCI will exclude personal identifiers. The survey data will be analyzed in the aggregate and no individual respondents will be identified. A statement from the Privacy Act Officer at the NIH is found in Attachment 4. This data collection is covered by the NIH Privacy Act Systems of Record 09‑25‑0156, "Records of Participants in Programs and Respondents in Surveys Used to Evaluate Programs of the Public Health Service, HHS/PHS/NIH/OD".
Westat’s Institutional Review Board (IRB) reviewed the Primary Care and Oncologist Questionnaires and gave them expedited approval since the questions focus on non-sensitive issues related to the physician’s clinical opinion and practice, and there is low risk of breach of confidentiality (Attachment 5). The study has also been reviewed and approved by NCI’s IRB (Attachment 5).
Instructions on the survey, a Frequently Asked Questions document, and letters of support, found in Attachment 6, will provide background information and apprise the respondent of the following:
The survey is sponsored by the NCI, an agency of the Federal Government, with support from the American Cancer Society;
Survey data will be used to improve health experts’ understanding of optimal practices relevant to the follow up care and surveillance of post-treatment cancer survivors, and how such care is implemented in primary care and oncologists practices, with the ultimate goal of improving the quality of health care;
Information provided will be kept confidential, and will not be disclosed to anyone but the researchers conducting this study, except as otherwise required by law; and
Providing the information is voluntary, and there are no penalties for not responding to the information collection as a whole or to any particular questions.
Each physician’s survey will have a unique ID label at the top of the cover page. This number will be used as a unique record identifier during data entry. The data file containing physicians’ names and ID numbers will be maintained separately by Westat and used only for mailing the surveys and subsequent follow-up in the case of non-response. The non response follow-up letters can be found in Attachment 7.
A.11. Justification for Sensitive Questions
The physician survey will require 20 minutes to complete. (Copies of the instruments are included in Attachments 1 and 2). Only one response is requested from a total of 2,200 physicians. This includes 1,100 primary care physicians and 1,100 oncologists who may be working in for-profit or not-for-profit medical practices or clinics. The average burden hour per physician is 20 minutes, with the estimated total annual burden hours for all physicians estimated at 366. The total burden hours to administer the instrument for all physicians over the two-year field period of the study is estimated at 732.
After the second mailing, the offices of nonresponding physicians will receive follow-up reminder phone calls. These phone calls will again be conducted with the receptionist who answers the phone at the number the study has on record for the physician. Based on previous experience, about 70 percent of the mailed sample of about 3,150 (2,207 cases) will require telephone contact and will receive on average four contacts, resulting in about 4,412 calls annually. The average burden hour for the follow-up calls to the receptionist and office administrator is 5 minutes, with the estimated total annual burden hours for all follow-up calls estimated at 368. The total burden hours for all receptionist and office administrator follow-up calls over the two-year field period of the study is estimated at 736.
The total annualized cost to respondents is estimated at $39,049 per year as shown in Table A.12‑2. As data collection will span two years, the total cost for data collection will be $78,098. The total burden hours for this study over the two years will be 1808.
The hourly wage rate for physicians, receptionists and office administrators with responsibilities comparable to those in the target population is based on U.S. Department of Labor, Bureau of Labor Statistics data from the National Occupational Employment and Wage Estimates, May 2006; 24 October 2007 (U.S. Department of Labor) (accessed: 31 March & 17 June 2008).
A.12-1: Estimates of Annual Burden Hours |
|
|
|
|
||||||||
Type of Respondents |
Survey |
Number of Respondents |
Frequency of Response |
Average Time per Response (Minutes/Hour) |
Annual Burden Hours |
|
||||||
Receptionists |
Screener |
2,033 |
1 |
5/60 |
169 |
|
||||||
Family Practice |
PCP Instrument |
250 |
1 |
20/60 |
83 |
|
||||||
General Internists |
PCP Instrument |
250 |
1 |
20/60 |
83 |
|
||||||
OB/GYNs |
PCP Instrument |
50 |
1 |
20/60 |
17 |
|
||||||
Oncologists |
Oncology Instrument |
550 |
1 |
20/60 |
183 |
|
||||||
Receptionists & Administrators |
Follow-Up Phone Calls |
1,103 |
4 |
5/60 |
368 |
|
||||||
TOTAL |
|
4,236 |
|
904 |
|
|||||||
|
|
|
|
|
|
|||||||
|
|
|
|
|
|
|||||||
|
|
|
|
|
|
|||||||
A.12-2 Annualized Estimates of Burden Hours and Cost to Respondents |
|
|
||||||||||
Type of Respondents |
Survey |
Annual Burden Hours |
Hourly Wage |
Annual Respondent Cost |
|
|||||||
Receptionists |
Screener |
169 |
12.50 |
$2,118 |
|
|||||||
Family Practice Physicians |
PCP Instrument |
83 |
72.04 |
$6,003 |
|
|||||||
Internal Medicine Physicians |
PCP Instrument |
83 |
77.34 |
$6,445 |
|
|||||||
OB/GYN Physicians |
PCP Instrument |
17 |
85.60 |
$1,427 |
|
|||||||
Oncology Physicians |
Oncology Instrument |
183 |
91.67 |
$16,806 |
|
|||||||
Receptionists & Administrators |
Follow-Up Phone Calls |
368 |
17.00 |
$6,250 |
|
|||||||
TOTAL |
|
904 |
|
$39,049.21 |
|
|
||||||
There are no capital or start-up costs, and there are no costs to the respondents or record keepers for operation and maintenance of services.
The total cost to the Federal Government to conduct and analyze the physician survey is estimated to be $605,500 over the 2 year field period of the study. The budget includes the costs of survey design and development, all data collection and follow-up, incentive payments to physicians, and other miscellaneous costs such as supplies, expenses, and postage. Professional service time is included for study management, dissemination of the interviews to the physicians and receipt control, data management, data analysis, and overhead costs. Costs have been added for both the contractor and for NCI staff, who have developed the survey and will provide oversight of the contractor.
Of the total cost, $605,500 will be spent during the data collection phase over the course of 2 years as shown in Table A.14-1.
Table A.14 - 1 Estimates of Annualized Cost to the Government
Year |
Contractor Costs |
Estimated NCI Costs |
2009 |
$289,500 |
$13,250 |
2010 |
$289,500 |
$13,250 |
Annualized Cost |
$302,750 |
|
Total Cost for 2 Years |
$605,500 |
|
This is a new collection of information.
The current plan is to mail the survey in January 2009, if it has received OMB clearance by that date. Otherwise, the survey will be fielded as soon as possible after clearance has been received.
Table A.16 - 1 Tentative Survey Activity Timeline
Activity |
Date |
Cognitive Testing of Instruments |
October 2007 – May 2008 |
Pilot test Screeners |
June 2008 |
Initial calls to sampled physicians |
1 month after OMB approval |
Send survey package via FedEx |
2 months after OMB approval |
Send reminder letter and replacement survey via FedEx |
2-4 months after OMB approval |
Begin phone contacts to encourage response |
2-6 months after OMB approval |
Final survey non-response follow-up package sent |
5-7 months after OMB approval |
Final follow-up calls completed |
6-8 months after OMB approval |
Final surveys accepted |
8 months after OMB approval |
Data entry and data check completed |
9 months after OMB approval |
Initial analyses completed by Westat |
9-11 months after OMB approval |
Initial papers submitted for publication |
12 months after OMB approval |
Additional papers submitted for publication |
12-16 months after OMB approval |
Publication Plan
NCI plans to publish a number of papers based on the data collected. Some topics identified for papers include:
National Estimates of Physicians Performing follow-up care and surveillance of post-treatment Cancer Survivors;
Primary Care Physicians - approaches and practices with respect to cancer survivors follow up care and surveillance;
Medical oncologists - Approaches and practices with respect to cancer survivors follow up care and surveillance;
Primary care physicians and medical oncologists – Practices and approaches to care;
Factors influencing practices of physicians treating cancer survivors;
Specific practices to monitor patients for recurrence and to screen for late and long-term effects;
Optimal or ideal Practices – similarities and differences in perceptions between primary care physicians and medical oncs;
Academic and community primary care physicians – practice patterns, approaches to and perceptions of survivors’ care;
Academic and community medical oncologists – practice patterns, approaches to and perceptions of survivors’ care;
Key barriers and facilitators to the implementation of optimal practices related to the care of cancer survivors;
Care coordination and transitioning – continuity of care for survivors;
Models of care - similarities and differences in preferences across medical oncologist and primary care physicians; and
Knowledge of late and long-term adverse consequences of cancer and its treatment and training needs.
It is expected that two or three papers will be submitted by the end of 2010, with the remainder submitted for publication in early 2011.
This survey will obtain national data on PCPs’ and oncology physicians’ practices, perceptions, and knowledge related to the care of cancer survivors. There are three main types of analyses to be conducted for the study. The first involves descriptive data based on estimates of the proportion of physicians nationally who pursue certain practices, their perceptions regarding roles, transitioning of care, knowledge regarding survivorship surveillance per current recommendations, knowledge of late or long-term adverse consequences of cancer or its treatment, perceived barriers or facilitators to follow-up care, preferred models of care and practice characteristics affecting provision of care. All estimates will incorporate sample weights, described in detail in Section B.2, page 2.
The second type of analysis involves group comparisons involving primary care physicians who may practice differently or may report different attitudes in their treatment of cancer survivors when compared to medical oncologists. In this study, analyses involving primary care physicians involve physicians belonging to Family Practice (FP) or General Internal Medicine (IM) specialties.
The third type of analysis involves the identification of factors associated with physicians’ reports of their practices and attitudes. In the following sections we provide greater detail on these analyses, including examples of table shells.
Descriptions of Practices Related to the Care of Cancer Survivors. Table A.16 - 2 gives examples of practices and recommendations reported as estimated percentages at the national level and indicate how the results of such analyses might be presented. Physicians’ perceptions of screening effectiveness, recommended starting age and interval for screening, volume of patients screened, and perceptions of barriers are critical information that will be reported outright as weighted estimates. Weighting procedures are described in detail in Section B.2, page 2. Statistical comparisons between different screening procedures used for the same disease may be made using a simple chi-square test. Table A.16 - 2 shows how the results of such analyses may be presented.
Table A.16 - 2 EXAMPLE -- Percent of Primary Care Physicians Reporting Beliefs, Perceptions, Recommendations, Facilitators, and Barriers Regarding Cancer Survivors
|
n |
Estimated Percentage |
Very confident in knowledge of appropriate surveillance testing for Breast Cancer Colon Cancer
|
|
|
Strongly agree that PCPs have the skills necessary to initiate appropriate screening or diagnostic work-up to detect recurrent cancer, for survivors of… Breast Cancer Colon Cancer |
|
|
Report shared oncologist/PCP responsibility for evaluation of long-term effects of cancer treatment for… Breast Cancer Colon Cancer
Patients are often or almost always unable to pay (or lack insurance coverage) for follow-up care
|
|
|
Group Comparisons. There are four group comparisons planned for the study. The first involves comparisons of PCPs versus oncologists, the second involves comparisons of primary care physicians that are academically or community-based, third involves comparisons of medical oncologist physicians that are academically-based versus those that are community-based, and the fourth involves comparisons of Family Practice (FP) and Internal Medicine (IM) physicians. For all group analyses, estimates for the groups on a range of items will be examined using analysis of variance models to determine whether there are statistically significant differences between the groups.
(1) Comparisons of PCP and Oncologists. Physicians from two medical specialties (FP and IM) providing primary medical care for adults in the U.S. will be included in the survey. The PCPs will be compared to oncologists on a number of variables related to medical practices and opinions. In particular, the analyses will yield information regarding practices and attitudes related to physician willingness to refer or accept cancer survivors back into a primary care setting after active cancer treatment. Table A.16 – 3 below shows how findings may be displayed.
Table A.16 – 3 EXAMPLE -- Relationship between Specialty and Recommendations
|
|
Type of Specialty |
|||
|
|
Primary Care |
Oncologists |
||
|
|
n |
Estimated Percentage |
n |
Estimated Percentage |
Recommend mammograms every 2-3 years
|
|
|
|
||
Preferred treatment model -- oncologists have primary responsibility for follow-up care |
|
|
|
||
Receive/provide a summary of follow-up care |
|
|
|
||
(2) Comparisons of Academically-based Versus Community-based Physicians. Academically-based physicians are defined as physicians who provide clinical care within a university-affiliated system, or who teach medical students. Community physicians are defined as all other physicians. The attitudes and practices of physicians affiliated with academia will be compared with those of physicians working outside the academic arena. Because they have routine access to the most recent research findings, academically-based physicians are expected to be more likely to practice in accordance with expert recommendations. Table A.16‑4 shows how findings may be displayed.
Table A.16 – 4 EXAMPLE -- Relationship between Affiliation and Practices
|
|
Affiliation |
|||
|
|
Academic |
Community |
||
|
|
n |
Estimated Percentage |
n |
Estimated Percentage |
Recommend mammograms every 2-3 years
|
|
|
|
||
Receive/provide a summary of follow-up care |
|
|
|
||
(3) Comparisons of Academically-based Versus Community-based Medical Oncologists. Because oncology is a rapidly changing field, and the issues of continuity of care for cancer survivors has received increased focus in academia, academically-based oncologists are expected to be more open to sharing care with primary care physicians than oncologists based in the community. In addition, understanding differences between these two groups will indicate the degree to which best practice recommendations have diffused from the academic to the community setting. Table A.16-5 shows how findings may be displayed.
Table A.16 – 5 EXAMPLE -- Relationship between Affiliation and Practices
Oncologists Only
|
|
Affiliation |
|||
|
|
Academic |
Community |
||
|
|
n |
Estimated Percentage |
n |
Estimated Percentage |
Recommend mammograms every 2-3 years
|
|
|
|
||
Preferred treatment model -- oncologists have primary responsibility for follow-up care |
|
|
|
||
Receive/provide a summary of follow-up care |
|
|
|
||
(4) Comparisons of Family Practice and General Internal Medicine Physicians. FP physicians will be compared to IMs on a number of variables related to medical practices and opinions. In particular, the analyses will yield information regarding these physician’s practices and attitudes as they relate to the care of cancer survivors. Table A.16 - 6 shows how findings may be displayed.
Table A.16 – 6 EXAMPLE-- Comparison between General Internal Medicine and Family Practice Physicians
|
|
Specialty |
|||
|
|
General Internal Medicine |
Family Practice |
||
|
|
n |
Estimated Percentage |
n |
Estimated Percentage |
Recommend mammograms every 2-3 years
|
|
|
|
||
Preferred treatment model -- oncologists have primary responsibility for follow-up care |
|
|
|
||
Receive/provide a summary of follow-up care |
|
|
|
||
Factors related to survivor care practices and attitudes. To identify factors related to whether physicians engage in certain practices (some of which will be identified as optimal), a dichotomous variable will be created. Physicians implementing certain practices will be compared to physicians who do not on a number of different variables, including access to medical record information, confidence in follow-up knowledge, perceptions of PCP skill levels, number of cancer patients treated, practice influences, barriers, academic affiliation, practice and physician demographics, level of training, and specialty. Logistic regression modeling will help identify predictors of successful implementation. Chi-square analyses will be used to explore relationships.
Non-response. As described in Section B4 page 8, considerable effort will be expended to minimize rates of nonresponse. Analyses will be conducted on the characteristics of individuals who do not respond to the survey based on information available on the frame. For this study, variables reflecting both PCPs’ demographics and practice location will be assessed to identify those that characterize the propensity to respond. Logistic regression and/or the software package CHAID (a categorical search algorithm) will be used to identify variables to be incorporated into models for nonresponse adjustment. Westat statisticians have made extensive contributions to the literature on the effectiveness of such approaches (see Rizzo, Kalton, & Brick (1996) and Kalton (2003)). A report will be provided to NCI describing the findings of these evaluations, indicating subpopulations with higher levels of nonresponse and thus with a greater potential for contributing to any bias in estimates.
Dealing with issues of survey nonresponse is a standard part of the weighting effort associated with the estimation and analysis of the survey data. In developing sample weights, we will adjust the weights to reduce the potential for bias associated with nonresponse.
The expiration date for OMB approval will be displayed.
| File Type | application/msword |
| File Title | Supporting Statement for the |
| File Modified | 2008-09-18 |
| File Created | 2008-09-18 |