Top Tips for Conducting Global Online Research

1
Top Tips for Conducting Global Online Research
PBIRG May 8, 2005Presented byElys Roberts

2
Why This Session Should Be Useful For You

• Attending this session will provide
• An understanding of the key differences between
  the US and OUS for online physician market
  research
• A knowledge of the pitfalls to avoid when
  conducting global online research
• An understanding of the factors that you can
  influence and those you cant when conducting
  online research
• Tips for conducting better global online
  physician research

• You will see the contributions of some of the
  great American Presidents and British Prime
  Ministers to the field of global online
  research...

3
Agenda

• Broad principles/concepts
• Group work
• Landscape analysis
• Including results of Medefield pharma client
  survey
• Key differences US vs. Outside the US (OUS)
• Top tips
• Questions and summary

4
Broad Principles

• 1. Span of control
• 2. Fielding effectiveness
• Wastefulness or art of being frugal
• Increased importance of sample/fieldwork
• 3. Duty of care

5
Span of Control
Language
Universe size
Quotas
Questionnaire length
Specialty types
Internet penetration
Holidays
Respondent modem speeds
Screeners
Sample size
6
Span of Control

• OSOC Outside your Span of Control
• WISOC With-In your Span of Control

• OSOC
• Universe size
• Specialty types
• Language
• Holidays
• Internet penetration
• Modem speeds

• WISOC
• Sample size
• Questionnaire length
• Screeners
• Quotas

7
Fielding Effectiveness
8
Landscape Analysis
"Progress might have been all right once, but it
has gone on far too long" Ogden Nash (Poet,
1902-71)
9
Landscape Analysis

• ATT pioneered CATI in 1971
• Online research in pharma from around 1996
• Took off in the US around 01-02
• Response rate for phone interviews dropping
  alarmingly
• Internet now the predominant pharma data
  collection method in the US
• EU and other areas still lagging
• But at a point where the US was around 2002

10
Landscape Analysis
11
Medefield Pharma Client Survey Share of
Studies Personally Commissioned/Used by MRers
All MRers who commission/conduct/use at least
1quantitative MR study based on interviews with
physicians during previous year (Total includes
other)
All US-based MRers with International
responsibilities
Q.6 During 2003, how many studies with physicians
did you personally commission/conduct/use where
each of the following methodologies were used?
12
Group WorkIn tables All even number
groups What are the key differences between
conducting online physician research globally,
compared with conducting the same study in the
US? All odd number groups List the
challenges/ issues you and your colleagues face
when conducting online physician research
globally? 10 minutes discussion be ready to
share your answers

13
Comfort Level in Commissioning/Conducting
Physician Internet Research
Comfortable
Not at all
Extremely
US (mean5.46)
Canada (mean5.11)
UK (mean4.73)
Germany (mean4.55)
France (mean4.43)
Italy (mean4.01)
Spain (mean3.95)
Japan (mean3.83)
Mexico (mean2.97)
Brazil (mean2.97)
Base All MRers (146) 1-7 scale
Q. 13 How comfortable are you with
commissioning/conducting physician studies that
use the Internet for data collection in each of
the following countries?
14
Perceived Credibility of Data Collection Methods
Increase decrease from 2004
Mean score - credibility
Base All MRers (146) 1-7 scale
Q.8 Please rate each physician data collection
methodology according to how credible you
perceive it to be in terms of producing reliable
information for business decision-making in the
pharmaceutical industry
15
Importance of Specific Issues in Restricting
International Internet Physician Studies
Important
Not at all
Extremely
Penetration of the Internet among physicians
internationally needs to be higher (mean5.53)
Internet technology used by physicians
internationally needs to be more
up-to-date(mean5.40)
There needs to be better availability of robust
online physician respondent communities/panels.
(mean5.22)
Physician preferences most would prefer to
participate in research by other means
(mean4.71)
It is difficult to motivate physicians to respond
to Internet surveys. (mean4.62)
of respondents
Base All MRers (146)
Q. 14 Thinking about physicians studies in major
markets (France, Germany, Italy, Spain, UK,
Canada, US, Japan, Mexico, Brazil) how important
do you think the following issues are in
restricting international Internet physician
studies?
16
Technology General Population Internet
Penetration
InternetWorldStats.com (data from Nielsen Net
Ratings Computer Industry Almanac)
17
Technology Physician Internet Penetration

• US
• 68 of the general population are online
• 96 of physicians

• EU - of physicians online

• Implications

OSOC

• Minimal tech-bias with MDs (be more wary with
  patients)
• Technology barrier is less of a hurdle than
  many think

Boston Consulting Group, European Physicians
and the Internet
InternetWorldStats.com
18
Technology General Connectivity

• Internet Traffic Report

ASIA
54
85
US
81
EUROPE
S. AMERICA
64
79
AUSTRALIA

• Implications

OSOC

• Visuals simple small (gif/jpg)
• Video clips short if at all

InternetTrafficReport.com Monitors flow of
data around world - higher score equals faster,
more reliable connections (April 12, 2005)
19
Survey Vital Statistics - Length

• Most physicians in Europe still use 56K Modem
• Same study takes longer in the EU vs. OUS
• And outside the EU, differences can be even
  greater

Advertised Study Time

• Implications

WISOC

• Dont assume survey length based on a US study

Review of previous studies from Medefields
Healthcare Advisory Board, 04-05
20
Survey Vital Statistics

• Wasnt it John F. Kennedy who said
• Ask not what your can do
  for you ask what you can do for your
  

• On the Web, respondents are just one click away
  from doing something else
• As MRs, it is our job to ensure that they dont
  exercise that option!

• Implications

• Balance research objectives against respondent
  experience to achieve fielding effectiveness

21
Survey Vital Statistics Length Drop-Out

• OUS, drop-out rate increases sharply as survey
  length increases

• Implications

• He can compress the most words into the
  smallest ideas
• of any man I ever met (Abraham Lincoln)
• Length of survey has greater significance/impact
  OUS

Review of previous studies from Medefields
Healthcare Advisory Board 04-05
22
Survey Vital Statistics

• Internet is the last true source of co-operative
  respondents
• As a research community we need to take
  collective responsibility for protecting this
  resource. This can include
• Ensure survey invite is accurate
• Limit number of screening questions
• Limit survey length
• Limit boring, repetitive tasks
• Limit open ends (and dont place near start of
  survey)
• Where possible provide results/feedback to
  respondents

• The price of greatness is responsibility
  (Winston Churchill)
• We all have a duty of care towards our
  respondents

23
Doctor Universe Comparisons

• US versus EU etc
• Size matters
• Sample size to universe ratio

• Within the EU
• Some surprising data
• E.g. Italy vs. UK
• Practicing physicians

• Sample sizes often need to be smaller than US
• Monadic-design difficult, except for GPs

24
Specialty Categorization

• Europe
• IMs
• Hospital vs. Office
• Italian specialties
• Japan
• Vague specialty categorization
• e.g Endos, Oncologists
• GPs

• Implications

OSOC

• What we know for the US doesnt necessarily
  cross-over
• Consider an agency that understands the global
  pharma market

25
Specialty Type Included in Most Recent
Quantitative Internet MR Study
of Respondents
Base All MRers who personally
commissioned/conducted/used internet research
quantitative studies (100)
Q.11b What physician types were included in the
study?
26
Incidence/Screeners

• Overly stringent screening criteria on top of
  small universe sizes
• Result sample burn-through
• Key watch-outs
• Sheer of screener questions
• of Rx written
• of time in patient care
• of years in clinical practice

• Implications

WISOC

• 1 WISOC issue for global online research
• Pre-tests or small initial sample deployment
• Be frugal with sample

27
Country specific/cultural issues
"We really do have a lot in common with America
these days, except of course the language" Oscar
Wilde (English Playwright, 1854-1900)
28
Country-Specific/Cultural Issues

• Yes, it is true southern Europe does close down
  in August!
• Northern Europe vs. Southern Europe
• Can Italy be used as a surrogate for Spain?
• Questionnaire content/language
• Generic vs. brand name
• Demographics e.g. ethnicity, salary
• Error messages

• Implications

OSOC

• Consider each country individually
• Consider an agency with broad global knowledge

29
System of Healthcare
Private Health
Public Health

• Every EU country has a different healthcare
  system
• Pricing questions can get particularly complex
• EU physicians can be price-conscious without
  having a clear understanding of what the costs
  are
• Government price negotiation e.g. NICE

• Implications

OSOC

• Take input from local affiliates wherever
  possible
• If unsure, pretest survey prior to field

30
Privacy Laws/Data Protection

• Spamming
• Makes up approx. 50 of mail server traffic
• Spamming laws similar in the US to EU
• Unsolicited email recruitment of potential
  respondents
• However, this does not apply to panels
• It is acceptable to e-mail
• Those who have given prior agreement
• Those you have a previous relationship with

• Implications

OSOC

• Consider agency with reliable credible
  respondent source

Implementing the Directive on Privacy
Electronic Communication (2002/58/EC) 12/03
31
Privacy Laws/Data Protection

• EU e.g.
• ESOMAR
• MRS
• EphMRA

• US e.g.
• CASRO
• PBIRG

• EU may have individual country requirements
• EU all have own guidelines, but trying to bring
  together
• But, EU not harmonized with the US

• Implications

• Info. on privacy laws posted on the Websites of
  the societies listed
• Consider an agency that understands privacy
  issues of Internet research

32
Privacy Laws/Data Protection

• Identifying respondents
• Here the EU and US diverge with the EU being
  tougher

• Implications

OSOC

• Greater complexity OUS, so care must be taken
  with target lists respondent data

33
Secondary Data

• Physician level Rx data almost non-existent in
  EU
• Unique physician ID numbers
• Equivalent of ME not widely-used or known in EU
• Target lists
• Matching c. 80 US just c.10 EU
• Often maintained by sales reps
• Quality very variable
• Target list matches very difficult
• Quotas
• If going to use at all, need to be allocated from
  response to screeners, not secondary data

• Implications

OSOC

• Less data available to target specific groups
• Be careful not to be wasteful with sample

34
Availability of Panels/Physician Lists

• US physician MR panels prevalent
• OUS very few effective panels/respondent
  communities
• No AMA equivalent willing to rent-out access
• Building panel OUS much more difficult
• Panels variable size, quality and response rate
• Phone-to-web inefficient/ response rates dropping
• Third party lists less common and often dubious
  quality

• Implications

WISOC

• Range of options more limited
• If using lists, confirm physicians are opted-in
• Focus on achievable sample

35
Putting It All Together

• 10 languages
• 4 client contacts

• The Heart-sink study
• 9 countries
• 9 different questionnaires
• Tough screener

• So
• One base questionnaire OR treat each country as
  a separate study
• Ensure final paper copy survey is the final
  version
• 1 conduit for changes
• Impact of changes to survey after
  programming/after translations can be multiplied
  by the number of countries

• Implications

• 2 WISOC issue for global online research
• Potentially huge cost/time implications

36
Reason for Choosing Method of Data Collection
Base 94 multiple mentions allowed
37
Reason for Choosing Method of Data Collection
Total of Mentions
Base 94
Base 65
Base 69
38
Top Tips for Conducting Global Online Research
39
Questions and Summary

• Questions
• Global online surveys
• More complex than US
• Have to be more flexible (adapt to OSOC flexible
  with WISOC)
• Consider fielding effectiveness as much as you
  QTC
• But can reduce logistical headaches with online
  reearch
• Global online physician research is a growing,
  viable option for many research studies

40
Contact Details Elys Roberts Managing
Director Office 1 212 220 0880
x223 E-mail elys.roberts_at_medefield.com URL
www.medefield.com 24 West 40th Street Suite
950 New York, NY 10018
41
Back Up Slides
42
Sample ProfileMarket researchers
Region
Area of responsibility
Base All MRers Calculated numbers
Q.2 Does your area of responsibility  Focus
primarily on a single domestic market.Focus
primarily on international/global market.Involve
both domestic and international/global markets.
43
Sample Profile Market researchers
Base All MRers (146)
D1 In which country are you based?