Using the Web
for
Surveys of Medical Providers
Vasudha Narayanan
Presented at
Third International Conference on Establishment Surveys
June 21, 2007
1
Acknowledgements
Co-Authors
 Stephanie Fry, Jennifer Crafts
& Pamela Giambo from Westat
Project Officers
 David Clark, William Taylor, Gladys Valentin
& Mei Wang from Centers for Medicare &
Medicaid Services
This material was prepared by Westat Inc., under contract with the
Centers for Medicare & Medicaid Services (CMS), an agency of the U.S.
Department of Health and Human Services. The contents presented do
not necessarily reflect CMS policy
2
Overview
•
•
•
•
•
•
•
About the Surveys
Survey Methods
Questionnaire Design Issues
Sampling and Contacting Providers
Completes by Mode
Factors Affecting Web Response
Conclusions
3
About MCPSS & QIO Survey
• Two Surveys for Centers for Medicare &
Medicaid Services (CMS)
– QIO Provider Survey (2004)
– MCPSS (2006 & 2007)
•
•
•
•
National Sample of Medicare Providers
Satisfaction Surveys
Approx. 15 minute survey
Multi-Mode Data Collection
4
Survey Methods
• Web with telephone follow-up; paper on
request
• Telephone and paper surveys to:
– Increase response rates
– Reduce potential bias
• Importance of:
– Screening
– High quality mail outs
– Professional Interviewers
5
Survey Methods (cont.)
• Field Period (12-16 weeks)
– Screener
– Notification of web survey
– Reminder letters/e-mails
– Hard copy questionnaire mailing/fax
– Telephone non-response
– Response rates (AAPOR-RR3)
• QIO: 90%
• MCPSS: 65%
6
Questionnaire Design
Usability Testing
• Purpose of Testing
– Ease of navigation
– User satisfaction and perceived burden
• Profile of Testers
• Response mode bias
7
Questionnaire Design
Usability Testing Issues
• Allow for multiple respondents
– Clarity of instructions to “designate a
colleague”
• Clarity of instruction for submitting survey
• Ability to skip sections and questions
8
Elements of the Final Design
• Allow multiple respondents for each
survey
• Track section-level designees
• Track section-level status
• Crisp and easy to navigate web survey
• Seamless link between CATI & Web
9
10
11
Completes by Mode
Web
Mail/Fax* Telephone Completes
2004
32.2%
21.9%
45.9%
17,628
2006
40.1%
1.7%
58.1%
16,121
2007
55.1%
0.9%
43.9%
17,747
*non-response mail out of hard-copy survey in 2004
• Highest Rates of Web response – Hospitals
• Lowest Rates of Web response – Physicians
12
Factors Affecting Web Response
• Percentage Web access
– 93% physician practices (2007)
– 97% other providers (2007)
Hypotheses:
• Ease of Web access at work station
• Ease of survey website navigation
• Ability to obtain help when needed
13
Conclusions
• Feasibility of a Web surveys no longer an
issue
• Necessary factors for survey management
and achieving high response rates
– telephone non-response follow up essential
– Frequent interface between data collection
modes
– Allow the respondent maximum flexibility
while maintaining survey integrity
14
Contacting the Authors
Vasudha Narayanan
([email protected])
Stephanie Fry ([email protected])
Jennifer Crafts ([email protected])
Pamela Giambo ([email protected])
15
Thank you
16
Descargar

Using Medicare Files as Sampling frames for a Survey of