Supporting Statement Part B
The respondent universe includes individuals at health care payers (health insurance companies, managed care organizations, and pharmacy benefits providers), health care providers (hospitals and large and small physician practices for this study), and vendors (health IT developers, clearinghouses, and third-party billers) who are responsible for managing the planning and implementation of their organization’s ICD-10 transition. The total universe is unknown; estimates of the larger universes containing individuals responsible for planning and implementation of ICD-10 transitions are shown in Table B.1.
Table B.1: Estimated Universe and Sample Sizes
Population |
Size |
Sample size |
Providers |
4,000,000 |
400 |
Payers |
57,000 |
100 |
Vendors (HIT developers, clearinghouses, third-party billers) |
6,900 |
100 |
To ensure that information is obtained from all sectors needing to address the ICD-10 transition, the sample will be stratified based on the role of the respondent’s organization: health care payer, health care provider, or vendors to payers or providers. The goal for each fielding is to obtain 400 completed surveys with representatives of health care providers (the largest and most diverse group) and 100 each with payers and vendors.
All samples will be convenience samples from established panels that have been constructed using association, state licensing, and publication data on the populations of interest. It would be cost-prohibitive to construct probability samples of such low-incidence populations. Every nth eligible panel member will be selected to achieve desired sample sizes. Given that the purpose of the survey is for directional guidance in education and communication efforts, we believe a convenience sample is appropriate.
More information about the panels to be used for this project is as follows:
This project will utilize multi-mode health care panels. The Ipsos Healthcare Professional panel is comprised of hard-to-reach targets (such as hospital executives and administrators, nurses, and pharmacists), and includes more than 250 subspecialties. Individuals can be reached via different approaches, such as email, fax, direct mail, and phone. By using a multi-mode approach for this project, we combine postal mail and email to recruit the right individuals to participate in this research.
The panels that will be used have purchased and licensed key association and governmental databases that verify essentials like a physician’s practicing status. These verification resources include DEA number and AMA ME number to help to ensure validity.
Recruitment to these panels is also done by a range of techniques: email invitation, affiliate networks, online recruitment, router assignment, and others. We use Captcha software to eliminate automatic registration (bots) of surveys, and also check for suspicious IP addresses to prevent them from joining the panels.
Ipsos panel management protocols include a series of legitimacy checks as well, including name/address matching using external databases, digital fingerprinting, and TrueSample. Ipsos utilizes inbuilt criteria such as the exclusion of individuals who repeatedly fail to respond to survey invitations, as well as those who exhibit undesirable survey behavior (e.g., inconsistent response, straight-lining, speeding through).
More information about likely sources of samples per audience segment is described as follows:
The majority or all completes for group 1 (provider community) will likely come from the Ipsos online e-mail panel
The majority or all completes for group 2 (payer community) to come from direct mail-to-web
The majority or all completes for group 3 (HIT developers, clearinghouses, and third-party billers) to come from direct mail-to-web
Upon OMB approval, the survey instrument will be programmed into the survey vendor’s web hosting servers. The survey vendor, Ipsos, a leading market research firm, and contractor staff will then review the online version for accuracy and test it to ensure all skip patterns function as intended, items permit all valid responses and exclude invalid ones, items intended to accept multiple responses do so, etc.
The survey vendor will then commence data collection by sending mail and email invitations to sample members. The mail samples will be sent survey invitations via traditional mail that ask potential respondents to go to the specific website and enter a code to complete the survey. Email samples will be emailed a survey invitation containing a link to the survey. (See Appendix C.)
The in-panel survey completion rates for projects of this type range from 60%-80%, depending to some extent on the particular subpopulation groups being included in the survey as well as the incentive structure and reminder mechanisms. While this rate cannot be guaranteed, we are aiming for approximately 80%.
The survey vendor will track responses on a daily basis and provide the contractor and CMS with periodic updates. Up to two follow-up emails will be sent; the first, approximately five days after the initial invitation is emailed; the second, approximately five days after that. No follow-up mailings will be sent to sample members initially contacted by postal mail, as the fieldwork period is fairly short. There is not sufficient time to send out a postal reminder without essentially sending it a day or two after the first invitation, which some potential participants would view negatively as overly aggressive.
As this is a convenience sample, there is no response rate for this research. Instead, we will provide a participation rate, per recommendations from the American Association for Public Opinion Research (AAPOR) Task Force (2010) and ISO 26362 (2009). Regardless of this, a number of methods of maximizing response will be employed in line with industry standards of best practice. The email and mail invitations to participate in the survey, the introductory survey language, and the survey itself are designed to facilitate rapid understanding and promote compliance with and completion of the survey. The survey vendor utilizes best practices derived from mail and online surveys in its approach to initial contacts and all follow-on email reminders. The importance of the survey and the reasons each individual should participate will be emphasized in all contacts with sample members; the body of research around response has demonstrated that respondent understanding of the reasons for and importance of a survey are important factors in their likelihood of completing the survey.
The estimated time involved to complete the survey is stated in the invitation so that respondents will realize that it will take only a small amount of time. Invitations also succinctly describe procedures for maintaining the privacy of respondents (i.e., identities of individuals will not be released, identifying information will be stored separately from the survey responses, and all information collected will be analyzed in the aggregate). Reminders for survey completion will be sent only to those who have not completed it and will be worded in such a way that potential participants do not feel pressured.
Consistent with the calculations approved by AAPOR, participation rates for this study will be calculated as: “the number of respondents who have provided a usable response divided by the total number of initial personal invitations requesting participation” (per AAPOR guidelines on Non-Probability Internet Panels).
The sampling and data collection procedures are used on an ongoing basis by the survey vendor and are refined and improved as needed. The survey questionnaire previously fielded in 2011 under OMB Approval No. 0938-1149 was pretested by several trade associations that represent the three audiences (providers, vendors, and payers). The organizations fielded the questionnaire among their members during a March-April 2011 time frame. No fielding difficulties were reported then or when the approved survey was fielded in November-December 2011. The information obtained through the readiness assessment survey fielded in November-December 2011 has been valuable in informing CMS outreach and education efforts.
CMS’s contractor, Ketchum, will contract with Ipsos to conduct the sampling, data collection and analysis for this study. Ipsos’ efforts will be overseen by Dr. Clifford Young, Managing Director, Public Sector, Ipsos Public Affairs, USA. Dr. Young is a survey statistician and methodologist, trained in statistics and survey methods (MA and PhD) at the University of Chicago, and in survey sampling at the University of Michigan. He is expert in sampling and methodological design, and has led on more than 30 full samples and worked as lead survey methodologist on more than 50. Ipsos’s Meghann Jones, Senior Research Manager, and Neale El-Dash, Director, will also lead this project. Ketchum’s Christina Nicols, Vice President and Director of Research, will prepare a report of findings based on Ipsos data tables. Several Ipsos staff members with appropriate training are also available as needed to program the questionnaire, perform statistical programming, prepare tables and summary statistics for reports, and assist in interpretation of the results of the quantitative analysis.
Rosali Topper, Health Insurance Specialist, of CMS will serve as the Technical Monitor and the federal agency personnel responsible for receiving and approving all contract deliverables. Ms. Topper’s phone number is (410) 786-7260.
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | WILLIAM PARHAM |
| File Modified | 0000-00-00 |
| File Created | 2021-01-29 |