OMB_Part_B

Supporting Statement B

Survey of Eligible Users of the Data Bank

OMB Control No. 0915-XXXX

B. Collections of Information Employing Statistical Methods

1. Respondent Universe and Sampling Methods

The survey will be administered to eligible users of the National Practitioner Data Bank (NPDB). The respondent universe is comprised of both users and non-users that are eligible to report to the NPDB, query the NPDB, or both between January 1, 2010 and December 31, 2012. Respondents are hospitals, managed care organizations, physician group practices, state licensing boards, medical malpractice payers, professional societies, other smaller groups of government entities at both the federal and state level, and those who self-query the NPDB.

The NPDB receives and discloses reports on medical malpractice payments made on behalf of health care practitioners; as well as certain adverse actions taken against health care practitioners, health care entities, providers and suppliers. Depending on their provisional authority requirements, entities that meet the NPDB's definition of an "eligible entity" may use the NPDB to query, or to submit certain reports on health care practitioners, health care entities, providers, and suppliers. Some entities both report to and query the NPDB (Exhibit 1).

If an entity uses an authorized agent to perform NPDB reporting or querying activities on their behalf, for the purposes of this survey, both the entity and the agent will be considered registered users of the NPDB. Reporters are required to report certain information related to the professional competence and conduct of physicians, dentists other health care practitioners. Queriers are authorized to receive information on health care practitioners, health care entities, providers, or suppliers to assist in making employment, credentialing, certification, and licensing decisions.

Table A, The NPDB Overview, located in the appendix of the Supporting Statement A, summarizes the respondent universe of current eligible users that report to and query the NPDB. Additionally, a glossary of relevant terms is provided in Appendix A of Statement B.

The NPDB administrative data will provide a census of currently registered NPDB users as well as previously registered users who were not re-registered for the specified time frame. A stratified sample design will be used to capture non-registered non-users and self-queriers.

1. NPDB Registered Users

The NPDB registered users are defined as eligible users who were registered and queried, reported to, or accessed the NPDB using an authorized agent during the defined time frame. This group of respondents includes entities that submitted reports, performed traditional queries (a one-time query), enrolled in continuous query, and those that utilized an authorized agent to query or report on their behalf (see Exhibit 1). The respondent is defined as the entity. For each respondent in this group, the entity’s most recent registered user will be contacted to represent the entity.

2. NPDB Non-Users

The NPDB non-users are defined as entities that were eligible to query or report to the NPDB and did not do so during the defined time frame. The non-user group includes previously registered users who were not re-registered to utilize the NPDB, eligible users who never registered to utilize the NPDB, and registered users who did not submit a query or report during the specified time period. Thus, if an entity was a registered user during the time frame and did not query, report, or utilize an authorized agent to query or report on their behalf, the entity is a non-user.

3. NPDB Self-Queriers

The NPDB self-queriers are health care practitioners and practitioner organizations that submitted a query during the specified time frame to either verify their own status or to provide a copy of the results to a third party. Self-queriers were not included in previous NPDB surveys. While the majority of self-queriers are health care practitioners, this group of respondents may also include entities such medical service providers and medical suppliers.

Sampling Methods. The NPDB administrative data will provide the basis for the sampling frame used to implement the sampling methodology for all users identified as having ever registered in the NPDB. Non-users who never registered to utilize the NPDB will not be found in the NPDB administrative data. For this group of non-users, external sources of information will be used to build a stratified sampling frame for each identified subgroup. Self-queriers do not register; but, they are required to provide information to verify their identity when they submit a query. For the self-queriers, NPDB administrative data will provide the basis for the sample.

The NPDB administrative data contains the population of registered entities. Therefore a census survey will be conducted for each type of registered entity. A stratified sample will be applied to the remaining eligible users. For strata where the population size is sufficient, a stratified random sample that allows a +3 percent margin of error with 95% confidence will be employed allowing analysis of user perspectives across disparate groups.

1. NPDB Registered Users. Administrative data will be used to construct a census of eligible entities that are responsible for meeting specific querying or reporting requirements, sometimes both. Eligible entities that use an authorized agent to fulfill their reporting or querying requirements will be included in this census sample. The entity user survey component is provided in Appendix B.

• A census of entities nationwide that submitted a report will receive the user survey component by contacting the most recent registered user who completed a NPDB report.

• A census of entities nationwide that submitted a query will receive the user survey component by contacting the most recent registered user who completed a NPDB query. As described in Part A, queries may result in a “matched response” that can be the result of a one-time query or notification through continuous query.

2. NPDB Non-Users. Registered non-user sub-groups will be identified using administrative data to develop the census samples. A stratified sample of non-users who never registered to utilize the NPDB will be applied for a maximum sample size of 2,000. The entity non-user survey component is provided in Appendix B.

• A census sample of registered users that do not utilize an authorized agent to query or report on their behalf, and the entity themselves have not queried or reported in the defined time frame, will be asked to complete the non-user survey component.

• A census sample of previous users, those whose registration expired previous to and were not re-renewed during the defined time frame, will be asked to complete the non-user survey component.

• A stratified sample of users who never registered to access to NPDB, and who are known to be eligible to report to or query the NPDB will be asked to complete the non-user survey component.

3. NPDB Self-Queriers. NPDB administrative data will be used to develop a sample, stratified by State, of health care practitioners and practitioner organizations that submitted a query during the specified time frame to either verify their own status or to provide a copy of the results to a third party. The self-querier survey component is provided in Appendix C.

• A sample of approximately 150 practitioners nationwide that received a matched response will receive the self-query survey component.

• A sample of approximately 1,350 practitioners nationwide that did not receive a matched response or did not complete the process will receive the self-query survey component.

Based on this sampling methodology, the estimated initial sample size and the expected number of completed questionnaires within the NPDB population of eligible users are provided in Table 1. Based on the 2001 survey and a set of enhancements discussed later, an expected response rate of 80 percent is anticipated.

Table 1. Estimated Initial Sample Size and Expected Number of Completed Surveys for the Eligible User Survey

Respondents Type	Respondents Description	Initial Sample Size	Estimated Responses (80%)	Hours per Response	Total Burden (Hours)
NPDB Users	Reporters Queriers (Non-Matched Responses)	14,790	11,832	.333 (20 min)	3,940
	Queriers (Matched Responses)	2,210	1,768	.383 (23 min)	677
Non-Users	Ever registered	1,500	1,200	.133 (8 min)	160
	Never-registered	500	400	.10 (6min)	40
NPDB Self Queriers	Non-Matched Responses	1,350	1,080	.10 (6 min)	108
	Matched	150	120	.216 (13 min)	26
Total		20,500	16,400	-----------------	4,951

2. Data Collection Procedures

Data collection procedures were designed to maximize timely response, reduce burden to respondents, and promote accuracy and completeness of the responses. In preparation for implementing the survey and an effort to increase the response rate, the NPDB will release a web-based newsletter (see Exhibit 7) to inform the users of the NPDB, about the purpose of the survey and to disclose the expected release date.

The survey instruments are provided in Appendix B and C for each stratum, in addition to the Codebook in Appendix D. Email addresses are available for registered users and self-queriers of the NPDB. Eligible non-users that never registered in the NPDB will first be contacted by telephone to obtain an email address. Non-users, whose registration has expired, will be contacted by the email address under the old registration.

Once the email address has been collected, the sampled respondents will receive the introductory letter and a web link that enables the eligible users to complete a web-based survey from a NPDB email account. Screen shots are provided in Appendix B and Appendix C to provide examples of how the questions are displayed to the respondent.

The NPDB will contact registered users via the IQRS internal notification system utilized by the NPDB registered users’ email (Exhibit 3). This web interface currently provides notifications to the users and will therefore prevent the survey email from being categorized as SPAM. The initial email will alert the respondent to log into the IQRS system, similar to Exhibit 3. Once the user logs in, their work queue will contain an introductory letter (Exhibit 4) with an invitation to complete the online survey.

Non-users and self-queriers are not registered to utilize the IQRS; therefore, these respondents will receive an email from a NPDB/HRSA email address.

Users (see Exhibit 4), Non-users (see Exhibit 5) and self-queriers (see Exhibit 6) will receive the introductory letter via email that provides directions on how to complete the survey. Respondents may then log in to the web-based survey at their convenience to complete the questionnaire. The survey is designed to allow the respondent the ability to complete the survey on their own schedule in an effort to help maximize overall response rates.

Utilizing the internal web interface IQRS for users and NPDB internal email for non-registered users, we plan to send a series of email reminders to encourage participation and completion of all surveys. Respondents will have access to a helpline in case they encounter problems completing the survey, lose their log-in credentials, or have other questions.

Additionally, a series of follow-up reminder calls are planned. For the respondents that have yet to complete the survey after the planned series of telephone and email reminders, respondents will be provided the opportunity to complete the survey with a telephone interviewer during the final round of calls.

3. Methods to Maximize Response Rates.

In addition to the NPDB announcing the survey in an article within a web-based newsletter for users of the NPDB, the following four enhancements were incorporated to maximize response rates:

• Respondents will receive advance official notification of the survey from NPDB using IQRS as well as follow-up reminders through the NPDB internal notification system. See Exhibit 4 for the introductory letter that will be sent via email.

• Information regarding the survey will be made available on the NPDB web site. See Exhibit 7 for the Newsletter article announcing the survey.

• The data collection instruments have been reduced in size as compared to the 2008 survey instruments in an effort to reduce burden and increase response rates.

• The contractor will inform the respondent of the following incentive to encourage completing surveys and to increase response rates:

“One dollar for every completed survey will be donated to the Children's Inn, a non-profit providing a home-like environment for children receiving medical treatment at NIH (up to a maximum of $5,000 for all surveys).”

Dealing with Nonresponse. An evaluation of potential nonresponse bias will be conducted for entities in the user, non-user, and the self query sample (e.g., type of entity, geographic location) to make comparisons of responders and non-responders . This determines whether certain types of entities were less likely to respond to the survey than others and, if significant differences are found, measure the direction and significance of the impact on the results. We will perform these analyses midway through the study. If a certain entity type is under-represented at that time then extra effort will be targeted toward that group to increase response rates.

4. Test of Procedures or Methods.

The current questionnaire is a substantial revision of the instruments previously implemented in 2001 and 2008. The recently developed instruments were tested internally to assess survey length and comprehensibility. The component of the user survey and the non-users survey was revised to such an extent that the instruments required a pre-test on fewer than 10 respondents to assess estimates of burden, and clarity of the instructions and questions of each survey component. The component of the self-query survey is new; therefore, the instrument required a pre-test on fewer than 10 respondents.

5. Individuals Consulted on Statistical Aspects and Individuals Collecting or Analyzing the Data.

The current collection instruments contain questions regarding recent system enhancements. The survey and evaluation will be conducted by a small business, Cherry Tree Business and Statistical Consulting under the consultation of Dr. George McCabe of Purdue University. Dr. McCabe has provided consultant services for previous federal surveys. All suggestions from the consultants have been incorporated in the data collection instruments that are included in this package.

Page | 6