Content uploaded by Peter J Schulz
Author content
All content in this area was uploaded by Peter J Schulz on Sep 04, 2015
Content may be subject to copyright.
Studies in Communication Sciences 11/2 (2011) 99–130
Peter J. Schu lz, Mar i a cai ata zufferey & uwe ha rtung*
First check the Internet, then see the Doctor:
How many Patients do it, and who are they?
This study reports results from a survey of patients in physicians’ practices in
two different Swiss metropolitan areas, Berne and Lugano. The main focus of
the survey was health information behavior, especially with regard to the Inter-
net, and patients’ use of the Internet for finding information on the condition
they’re seeking medical advice for. About one-fourth of patients were found to
be condition-informed by the Internet at the time of consultation. This is largely
independent of gender, but more frequent in the younger age groups and the
better educated. Patients can be divided into information-behavior types such
as health information seekers, chance finders, overlookers, and avoiders. Use of
the Internet is predictably related to these types, as is the use of other media for
health information with regard to patients’ present conditions. This suggests a
need to examine the relationship of Internet information-seeking behavior with
other relevant behaviors. There are hints indicating that patients who were better
reached by traditional information channels are also those who turn more often
to the Internet for health information. However all in all, evidence for this is
rather weak.
Keywords: health communication, health information exchange in doctor-patient
consultation, health information seeking behavior on the internet.
*
Università della Svizzera italiana, peter.schulz@usi.ch, maria.caiata.zufferey@usi.ch,
uwe.hartung@usi.ch
100 SCH UL Z, CA IATA ZU FFE RE Y & H ARTU NG
1. Introduction
In recent years, research has addressed medical patients’ use of the Inter-
net to acquire information related to medical conditions they may have.
Of particular concern was the issue of whether and how patients’ Internet
use affects the relationship between patients and health professionals. A
recent review of such studies (McMullen 2006) shows that patients do
not consider the Internet a replacement for consultations with a physician,
but rather as a supplement. They often turn to the Internet after consul-
tation or diagnosis, especially serious diagnoses such as cancer (Chen &
Siu 2001; Jenkins et al. 2001). Patients as well as physicians appear to
react with mixed feelings to opportunities offered by the Internet. As to
the consequences for the patient-physician relationship, patients using the
Internet to acquire information on their condition can be expected to play
a more active role, and may be less easily satisfied with the information
offered by health professionals (Anderson 2004).
The Internet is obviously a very different medium than the traditional
mass media. Many medical patients, however, brought information about
their medical problems to doctors’ offices long before the advent of the
Internet. Such information may be derived from earlier experiences, the
experiences of family and friends, hearsay, newspapers, magazines, televi-
sion, diverse fictional media like television entertainment or pulp fiction,
advertising, and so on. Some of this information may have been correct
and helpful, though surely not all. Patients differ in the amount of infor-
mation they bring and in the relative shares of correct and faulty informa-
tion they consume with regard to their medical problems. The Internet
may have improved patients’ opportunities to acquire information, but it
has not created the phenomenon of patients arriving to medical consulta-
tions with information they’ve previously acquired via the Internet. From
the physicians’ side, the information that patients bring to their offices can
serve both positive and negative functions. It can be seen as positive when
a doctor is able to build upon information to give a patient the necessary
facts for making an informed decision. It can be seen as a problem when a
patient holds so much false information that a physician must spend valu-
able time to counter the false conceptions, or when a doctor is perceived as
less credible by the patient when his or her expectations are disappointed.
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 101
Within the last decade, the Internet has pervaded society to such a
degree that it now reaches a large majority of the people living in indus-
trialized countries (European Commission 2010). Research on Internet
use shows that medical and health information are among the most com-
monly explored topics by users. (Jeannot et al. 2004; Madden & Fox
2006; Dolan, Iredale & Ameen 2004; Spadaro 2003). This has brought
patients to physicians’ offices who have found information on their
symptoms or illnesses on the Web. Some patients mention their Internet-
acquired knowledge in consultation, often without being able to specify
the website where the information comes from, or to assess the reliability
of this information (Fox & Rainie 2002; Diaz et al. 2005).
The Internet has not just increased the medical information available
to patients, it has also offered a new quality to information seeking and
the information patients may acquire. There is undoubtedly a great deal
of reliable information available on the Web, and it can be discovered
with greater ease than searching through libraries or medical publica-
tions, for example. Also, when sought, useable information can be found
on the Web much faster, and with much less expenditure in terms of time
and money than it takes, for instance, to find and pick up an informa-
tion leaflet on a particular disease. But then, not all medical information
on the Web is good. Chatrooms may perpetuate medical myths, and the
vested interests of pharmaceutical companies have created sites where the
underlying persuasive intent is not always easily recognizable.
The specific qualities of medical information drawn from the Inter-
net notwithstanding, any assessment of the consequences of information
obtained by patients should take the full media menu into account. As a
first step to such consideration, this paper seeks to define who the Internet-
informed patients of two regions in Switzerland are in terms of their com-
munication skills, their general orientation towards health information, and
their communicative behavior. Data come from two samples of patients in
medical practices in two cantons of Switzerland, the Italian-speaking Ticino
(N = 649) and the German-speaking Berne (N = 325). Patients were handed a
self-administered questionnaire upon entering the office. The main purpose
of the questionnaire was to recruit participants for a qualitative study of the
consultation. The answers in the recruitment questionnaire, however, are
also of interest. They are the basis of the analyses presented here.
102
2. State of Research in the Field
It is a paradox of the information age that the impact of patients’ unprec-
edented access to information on the doctor-patient relationship remains
poorly understood. Many of the articles that explore patients’ informa-
tion-seeking behaviors are editorials or opinion pieces, are focused on
issues of technology, or have examined very broad topics. There have been
relatively few studies of how such information empowers patients (Cline &
Haynes 2001; Powell & Clarke 2002; Powell et al. 2003), studies related
to the impact of Internet-based information sources on patient decision-
making (Gerber & Eiser 2001), or studies exploring the evolution of the
doctor-patient relationship in the information age (Anderson et al. 2003).
A few articles have focused on the Internet as a source of information, and
these provide overviews of the effects or possible effects of the Internet on
the doctor-patient relationship (Ball & Lillis 2001; Cline 2003; Eysen-
bach et al. 1999; Kassirer 2000; Levy & Strombeck 2002).
It is nevertheless clear that a significant number of patients turn to the
Internet to get health information (Houston & Allison 2002; Murray et
al. 2003a, b; Helft et al. 2003; Diaz et al. 2002; Dolan et al. 2004) and
this number is likely to increase (O’Connor & Johanson 2000). Patients
frequently report that the quality of the information they found was
good (Murray et al. 2003b; Hardey 1999; Diaz et al. 2002; O’Connor &
Johanson 2000; Houston & Allison 2002; Gordon et al. 2002; Hellawell
et al. 2000; Akerkar et al. 2005), but according to physicians, patients
had difficulties in interpreting the information (Wilson 1999; Potts &
Wyatt 2002; Chen & Siu 2001; Helft et al. 2003; Tann et al. 2003). A
further consideration is that research states that health professionals rarely
advise patients in Internet search strategies (Olson et al. 2003; O’Connor
& Johanson 2002; Gordon et al. 2002; Diaz et al. 2005). Several studies
suggest that information derived from the Internet can have a negative
impact on the doctor-patient relationship (Murray et al. 2003a, b; Wilson
1999; Potts & Wyatt 2002; Chen & Siu 2001; Helft et al. 2003; Tann
et al. 2003; Pautler et al. 2001; Hardey 1999), especially when it leads to
misinformation or patient confusion (Helft et al. 2003), or when the phy-
sician does not have the necessary communication skills to deal with such
patients or feels challenged (Murray et al. 2003a, b). It is clear that despite
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 103
the growth of this important information medium, neither patients nor
physicians feel well-positioned to deal with the growing – and often con-
tradictory – amount of health information on the Internet (Murray et
al. 2003a, 2003b; Chen & Siu 2001; Tann et al. 2003; Diaz et al. 2002;
O’Connor & Johanson 2000; Houston & Allison 2002).
Several articles highlight the specific challenges for health care in the
information age (Coiera 1996; Cotten 2001; Jadad 1999; Mullner 2002).
There are articles that provide advice to physicians for dealing with Inter-
net-informed patients (Pemberton & Goldblatt 1998), while another
focused specifically on the effects of the Internet on cancer patients
(Eysenbach 2003). A few articles examined the content of patients’
online information searching (Anderson et al. 2003) and the positive and
negative effects of Internet information on the doctor-patient relation-
ship (Johnson & Ramaprasad 2000). While it is true that the impact of
Internet information on the doctor-patient relationship is a relatively new
subject of study, patients have more and older information sources avail-
able to them (print or broadcast media, information from other people).
Articles that highlight Internet information seeking rarely examine the
question of other information sources.
In recent years, a relatively small number of studies on the impact of
Internet-derived information in general practice have been published. A
cross-sectional survey among 1,050 U.S physicians showed that 44 % of
the subjects found that patients bringing information from the Internet
influences the physician-patient relationship, most of them (38 % of all)
in a positive sense, yet 8 % in a negative sense (Murray et al. 2003a).
The quality of the Internet information was thought to be crucial to the
nature of its impact. Jones et al. (2001) interviewed clinicians on the issue
of well-informed patients. They concluded that clinicians should adapt
and strengthen their communication skills in order to meet the patient’s
need for information. Another issue in recent research is the potential
benefit of the Internet for particular subgroups of the population. A lack
of necessary material and educational resources will prevent many people
from using the Internet for health matters. As a consequence, the social
patterning of Internet use is likely to reinforce social inequality in the
distribution of health literacy (Kickbusch 2001). On a primary care level,
this indicates the need for a particularly active information style suitable
104
for patients from underprivileged social backgrounds (McKinstry 2000;
Davis 2004) in order to compensate for the relative lack of information
sources outside medical practice.
Only very few studies looked at the percentage and features of patients
who brought information acquired from the Web when showing up for
consultation. In a study conducted in April 2001, Gordon et al. (2002)
found that, among the patients (N = 138) in a Scottish rheumatology
clinic who had ever searched the Internet for health information (about
one-fourth of the sample), almost all had also done so for their rheuma-
tology-related disease. Ross et al. (2000) found that one in 10 patients
with Internet access had used the Internet to find out more about the
problem that brought them to a genitourinary clinic in England in 1999.
Knowledge on how many and which patients come to physicians bearing
Internet-derived information has been scarce so far.
Most of the studies on the impact of Internet information on the
doctor-patient relationship have been carried out in Anglo-Saxon coun-
tries while only a few studies looked at related issues in Switzerland. One
of them (Coulter & Magee 2003) was carried out in eight European
countries. It found that 10 % of the respondents (the result is based on
a random sample of 1,000 persons per country; people were interviewed
by telephone in July 2002) saw the Internet as a source of information
about new treatments. This figure, however, masks considerable variation
between countries, between age groups and between people with different
levels of education. The proportion who sought health information from
the Internet had risen to more than 15 % in Sweden, Switzerland and the
UK, and nearly one in five people with a university degree saw the Inter-
net as a useful source of health information (Coulter & Magee 2003).
The second study examined the need for health information within
the population of the Swiss canton Vaud (Ammann 2000). This popula-
tion survey (n = 613, aged between 18 and 74 years), showed that 18 %
of respondents had accessed health-related information on the Internet
(4.6 % often, 7.5 % from time to time and 5.7 % rarely). Compared to
other media such as the press, television, medical journals and radio, the
Internet was the least consulted, but this figure has surely changed since.
When researchers considered only individuals who had actively been
searching for health information in the last 12 months (47 % of the initial
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 105
sample), Internet use was higher. In fact, 25 % said they had used the
Internet. However, the major finding of the study was that people from
Vaud rated doctors as the preferred source of health information.
The third study (Koller et al. 2001) looked at Swiss German primary
care physicians’ Internet use for medical purposes (n = 1103). The study
found that access to the Internet was widespread (75 % of respondents
reported having access to the Internet) but only 14 % of respondents
reported regularly finding useful medical information on the Internet. In
addition, to solve patient-specific problems, most practitioners said they
consulted textbooks or colleagues rather than the Internet.
A fourth study (Jeannot et al. 2004) investigated information-seeking
on the Web by 1,604 patients of eight medical practices in 2001, with
both rural and urban areas included. It found that every tenth patient had
used the Internet for medical information-seeking, again a figure that is
sure to be higher today.
In summary, Internet use among the Swiss population has increased
rapidly in recent years. Swiss patients do use the Internet for health infor-
mation but view the physician as being the most valuable source of infor-
mation (Ammann 2000). Finally, Swiss physicians seemed skeptical about
the Internet some years ago as they reported they had not found useful
information related to medical practices on the Web (Koller et al. 2001).
3. Method
The following results are based on a survey of patients in 21 different
practices in the Lugano metropolitan area of Ticino, Switzerland, and
19 different practices in the Berne metropolitan area. The Lugano prac-
tices were run by six internists/general practitioners, five gynaecologists,
two orthopaedic surgeons, two oncologists, two urologists, one endo-
crinologist, one rheumatologist, one vascular surgeon and one allergist.
The Berne physicians included seven internists, five general practitioners,
three dermatologists, two orthopaedic surgeons, one oral surgeon, and
one urologist. Data was gathered in three waves, from March–May 2006,
in October and November 2006 and from May–September 2007 in
Lugano, and from June–August 2006, November 2006–February 2007
and June–August 2007 in Berne.
106
Questionnaires were handed out to patients on particular days agreed
upon with the physician in advance. Respondents were asked to fill in the
questionnaires while they were on the premises, and the questionnaires
were collected there. They asked about media and Internet use, seeking
of health information, trust in different media, self-perceived health com-
munication competence, familiarity with the doctor whose advice was
sought, and whether the respondent came for a chronic or acute problem.
Socio-demographic variables included gender, age, and education. All in
all, there were 28 questions in the longest version of the questionnaire.
There was some variation as to the questions based on field periods and
sites. This paper reports results only for questions that were asked in both
sites and all three waves. Participation was voluntary.
Some results are presented in three steps, corresponding to two filter
questions asked before the crucial question about holding Internet-derived
information on one’s present health problem: (1) Internet experience (rather
than access to the Internet), and if that was answered in the affirmative,
(2) experience with health information on the Internet, and again in case
of a positive answer, (3) the encounter of specific information there that
was related to the present visit with the physician. The three questions
appeared as follows: “Have you ever used the Internet, that is: Have you
ever sat down at a computer and logged in to the Internet?” Those who
said yes or did not answer were directed to the question, “Have you ever
searched the Internet for health information or come across such informa-
tion by accident? By health information we mean information on illness
but also on prevention and improvement of health.” Again, those who
said yes or gave no answer were directed to the third and crucial question,
“Have you ever searched for (or come across) health information that has
to do with the reason why you are seeing a doctor today?”
As this was a write-in survey, many inconsistent answers were given.
Recoding policy was to ignore second and third filter replies when a person
had said “no” to the first and second filters. Hence, a “no” to the question
of general Internet experience overruled any reply a respondent may have
given to the subsequent question about searching health information on
the Web. If, however, the first and second filters were unanswered, but the
second or third answered, a “yes” was substituted in the earlier filter ques-
tion. Which means if someone had not given an answer to the question
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 107
of whether s/he had ever searched for health information, but said yes to
the subsequent question about specific health information, s/he was also
coded “yes” in the earlier filter.
4. Results
The Berne and the Lugano sample were not similar with respect to socio-
demographics and other basic variables. There were many more men in
the Berne sample, and more people of the age group of 60 and older. More
patients in Berne visited their doctors because of chronic disease, and they
were more familiar with particular doctors, that is: the share of those indi-
cating they had seen this doctor many times before was larger. It cannot
be decided whether such differences reflect discrepancies in the two cities’
populations, in their culture, in customary health care behavior, in the
different mix of medical expertise the doctors had, or in any other meth-
odological difference. We will therefore deal with the two research sites
separately, for the most part.
The analysis will first address how many patients present themselves to
their physician after having first checked the Internet for information on
their condition. Then three steps follow which aim at describing who the
condition-informed patients are, first in terms of socio-demographic vari-
ables, second in terms of general orientation towards health information,
and third in terms of their reachability by traditional health communica-
tion channels. The third step (reachability) considers four aspects, health
communication skills, trust in health information, use of health-related
media content, and general media use.
4.1. Share of Internet-Informed Patients in General
Generally, three in four patients indicated they had had some experi-
ence with the Internet. Among those, roughly two-thirds said they had
ever searched or incidentally found health information on the Internet
before. Based on all respondents, this comes down to a share of slightly
above 50 %. Among these, more than every second had searched or found
information on the Internet related to the condition behind the visit to
the doctor at the time of the survey. Based on all respondents again, the
108
share is 28 %, with only marginal differences between the Berne data
and the Lugano data. As Internet use is similar in other (urban) regions
in Switzerland, physicians can expect that every fourth of their patients
has encountered Internet information related to the reason for seeking
medical advice before entering the practice. Table 1 shows the results for
all three steps and the similarity of the research sites in overview.
Information from the Internet may also be transmitted indirectly to
patients. A question tapped this by asking, “Has someone else at any time
given you information from the Internet on the problem why you are
seeing a physician today?” In Berne, the share of patients who answered
in the affirmative was 8 %, in Lugano 21 %. This indicates more fre-
quent interpersonal communication of health information derived from
the Web in Ticino, as compared to Berne. The people who have received
indirect information may also have received direct information by logging
on to the Web themselves. This is the case, to a large degree. However,
the percentage of patients who have only received Internet information
on their present health problem indirectly (that is by other people) comes
down to 9 % in Lugano and only 3 % in Berne. This means that in Luga-
no’s medical practices, there are 28 % of patients who hold some informa-
tion on their present problem that they have gathered from the Internet
themselves, in addition to 9 % of patients who have been given informa-
tion by family and friends. Thus, 37 % of the Lugano patients have had
contact (directly or via other people) with Internet information on their
present health problem prior to their visits to the doctor. Due to less fre-
quent indirect contact, the figure for Berne is 32 %.
4.2. Sociodemographic Structure of Internet-informed Patients
Generally, men use the Internet more intensely than women, but women
are more interested in health issues than men. What do these general
observations come down to when Internet-informed patients are consid-
ered? Higher Internet use among men appearsin our Lugano data, but
not in Berne. Contact with health information on the Internet is more
frequent among women, both in relative (based on those with Internet
experience) and in absolute terms. This is true for both Berne and Ticino
patients. Among those who had contact with Internet health informa-
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 109
Table 1: Share of Internet-informed Patients in Two Cantons (in percent)
Ticino
(n = 649)
Berne
(n = 325)
Total
(n = 974)
Have never been on the Internet 23 23 23
Have been on the Internet 76 72 75
No reply 1 4 2
Among those with Internet experience or no reply
Never had contact with
health information on the
Internet
21 19 20
Had contact with health
information on the Internet
53 53 53
No reply 2–1
Among those with contact to health information on the Internet
Never had contact with
information applying
to present health problem
24 24 24
Had contact with infor-
mation applying to
present health problem
28 29 28
No reply 1 1 1
Total 100 76 53 99 72 54 100 74 53
Differences between cantons significant only with regard to the higher number
of “No replies” in Berne to the question about Internet experience. Survey among
patients in physicians’ practices. Question wording see text.
tion, men had informed themselves more often than women on their
present health problems, again in both cantons. The three steps work in
different directions: Men are more likely to use the Internet (at least in
Ticino), among Internet users, women are more likely to encounter health
information on the Internet, and among those with past encounters with
health information, men are more likely to have checked on their present
health problem. Interestingly, the different attributes of the steps make
the share of Internet-informed patients roughly equal: 27 % of the male
and 30 % of the female patients in the Lugano practices had consulted
the Internet before they showed up in their doctors’ offices. For Berne, the
numbers are 31 % of men and 28 % of women (Table 2).
110
Similar analyses were run for age and education. For age, the antic-
ipated differences emerge: The younger the patients, the more they’ve
used the Internet, which means they’ve more frequently encountered
health information online, and the share of those who come to a physi-
cian’s practice with some knowledge about the medical issue at question
derived from the Internet is larger. Educational differences also surface in
the expected direction. Patients who have obtained information from the
Internet on their present health problem make up 40 % of those with ter-
tiary education (in both Lugano and Berne), 29 % and 26 % (in Lugano
and Berne) of those with secondary education, and only 11 % and 19 %
of those with obligatory education only (or less).
Table 2: Share of Internet-informed Patients among Men and Women in
Two Cantons
Ticino Berne
Men
(n = 146)
Women
(n = 293)
Men
(n = 153)
Women
(n = 168)
Have never been on the Internet 17 29 24 24
Have been on the Internet 83 70 73 73
No reply –13???
Among those with Internet experience or no reply
Never had contact with
health information on the
Internet
34 16 24 16
Had contact with health
information
on the Internet
46 53 50 56
No reply 3234
Among those with contact to health information on the Internet
Never had contact with
information applying to
present health problem
18 26 20 28
Had contact with
information applying to
present health problem
27 27 29 27
No reply 4 2 3 5
Total 100 83 49 100 71 55 100 77 52 ??? 76 60
Survey among patients in physicians’ practices. Question wording see text.
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 111
4.3. Internet-informed Patients’ General Orientations towards Health
Communication
Checking the Internet before one has an appointment with a doctor is a
behavior likely to be related to other health communication behaviors.
One way to address such relationships is to bring general orientations
toward health information into the picture. Orientations can be concep-
tualized as follows: People may react differently to health-related content
in the mass media. Some will actively seek such content, and we would
expect that among those, the share of patients showing up for a consulta-
tion with Internet-acquired knowledge on their minds could be especially
large. Other people will not seek health information, but will inadvert-
ently encounter it on occasion. A third group may be prone to overlook-
ing health-related content, and still others may even try to avoid it. The
orientation towards health-related media content is strongest in the first
group and decreases over the second and the third to the lowest level in
the fourth. We would expect that health-related media use decreases for
the four groups in a similar pattern.
The typology of orientations to health-related content is based on a
question asking respondents to select the most appropriate among four
items: “I actively seek out health information in the media;” “I come
across health information in the media every once in a while;” “I never
notice health information in the media;” and “I try to avoid health infor-
mation in the media as much as I can.” These people are defined as seekers,
chance finders, overlookers and avoiders.
In Lugano, three-fourths of the patients (74 %) were chance finders,
choosing the item “I come across health information in the media every
now and then.” Nine percent said they tried to avoid information (avoi-
ders), another 9 % never noticed it (overlookers), and 8 % indicated
they actively sought health information (seekers). The results for Berne
patients are similar, except that there were fewer overlookers (3 %) and
more chance finders. As would be expected, the typology is by and large
unrelated to general media use and to measures of health communication
skills (see below). It is, however, related to a question addressing use of
health-related content in magazines. The question was: “Many magazines
print articles on health and health problems. What would you say, how
112
many such articles do you read in magazines in a typical week?” Seekers
among Lugano patients (n = 29) read 2.8 health articles in a week, chance
finders 1.6 (n = 286), overlookers 0.8 (n = 23), and avoiders (n = 30) still
read 0.6 health articles in a typical week. The difference between the
seekers and the chance finders is significant (t = 3.414, df = 313, p < 001),
and so is the difference between chance finders and overlookers (t = 2.474,
df = 307, p < .05) and chance finders and avoiders (t = 3.405, df = 314,
p < .001). Among Berne patients, similar results appear, but are thwarted
by a low n = 8 for overlookers there. Yet in Berne seekers also read more
magazine articles on health (2.8; n = 23) than chance finders (1.5, n = 217;
t = 3.422, df = 238, p < .001), overlookers (0.5, n = 8; t = 2.372, df = 29,
p < .05f) and avoiders (1.0, n = 24; t = 2.666, df = 45, p < .05), but the dif-
ferences between chance finders, overlookers and avoiders are not sig-
nificant. Health information seekers read more health-related magazine
articles than health information chance finders, who read more than over-
lookers and avoiders, with the latter two differences failing to achieve
significance in the Berne data set.
How do the types relate to possessing Internet-acquired information
at the time of consultation with a physician? About every second health
information seeker shows up for consultation with Internet-gained infor-
mation. This share sinks to under 30 % for the chance finders, and to
about 10 % for the overlookers. Interestingly, more avoiders than over-
lookers appear to turn to the Web for specific health information. Figure
1 shows the precise results for the two cantons. The percentages are part
of 4x4 cross-tabulations, which show highly significant differences for
both cantons (Ticino: χ2 = 31.875, df = 9, p < .001; Berne: χ2 = 26.213,
df = 9, p < .01).
Health information seekers do not only turn to the Internet for infor-
mation on their present condition, but to other media as well. The related
question asked: “In the last two weeks, have you heard or read or watched
anything in any medium about the health problem that you are seeing the
doctor about today?” A positive answer was given by 32 % of the Ticino
and 59 % of the Berne information seekers. For chance finders, the per-
centages are clearly lower, and lower still for overlookers. Again, health
information avoiders appear to be more easily reached than overlookers.
The percentages are based on 4x3 cross-tabulations, which show highly
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 113
significant overall differences (Ticino: χ2 = 17.471, df = 6, p < .01, Berne
χ2 = 49.644, df = 6, p < .001).
The latter results suggest, and it is indeed true, that Internet-informed
patients at consultation are also informed about on their present condition
by other media. This invokes the old observation that people who use a
particular medium for information, tend to use other information media
as well (Lazarsfeld et al. 1944; De Waal & Schoenbach 2010). The same
can be formulated for entertainment media. This further suggests a need to
look at other communication behavior, orientations and abilities and their
correlation with being Internet-informed about one’s present condition.
4.4. Internet-informed Patients’ Reachability by Traditional Health
Communication Channels
Every new information medium creates hope that it will help to improve
the knowledge of those who have not been well-served by the existing
media or communication structures, or who have not communicated well
in the past. In our context, the emergence of eHealth (i.e., the Internet and
other interactive technologies) is widely hoped “to enable health improve-
ment and health care services,” for “traditionally underserved popula-
tions” in addition to others (Ahern, Kreslake & Phalen 2006; see also
Kreps & Neuhauser 2010, Wyatt & Sullivan 2005). Often such hopes
are destroyed when it turns out that the new medium is mostly used
by those who were easily reached by the existing media and channels.
The hope is that the previously difficult-to-reach may benefit most from
new information media. The disappointment comes when it turns out
that it is mostly the people who were easily reached by older media who
will also benefit from newly introduced media. The latter may be consid-
ered a variant of the knowledge gap hypothesis (Donohue, Tichenor &
Olien 1970) in communication science: When new information is offered
by the media, it will be taken up more intensely by people who already
know much, while those who know less, absorb a smaller proportion of
the new information. Thus both groups will be better informed, but the
knowledgeable more so. This creates a wider gap between the (relatively)
knowledgeable and the (relatively) ignorant. For the Internet, evidence of
the “digital divide” (i.e., inequality in Internet or computer use) resonates
114
with inequalities in health care. Groups who face difficulties in acquiring
good health care (disabled persons, rural residents, disadvantaged racial
minorities) were found to use the Internet less frequently than other
groups with easier access to health care (Wang, Bennett & Probst 2011).
The same can be said of persons with low socio-economic status (Kontos,
Bennett & Viswanath 2007). It has also been pointed out that benefitting
from the promises of eHealth requires particular abilities (called eHealth
literacy) that many people do not have (Norman & Skinner 2006).
Our data does not allow for the formal testing of these hypotheses for
the case of health information on the Internet. But they provide some
hints as to which of the two general hypotheses is more valid, the hope to
reach previously uninformed patients or the disappointing finding that
health information on the Internet reaches mostly those who are capable
of making use of traditional information sources. Four aspects of reach-
ability by traditional channels are considered:
– health communication skills
– trust in health information in traditional media (newspapers, maga-
zines, radio, television)
– use of health information in traditional media (magazines)1
– general use of traditional media (newspaper, magazines, television).
If the Internet is able to reach the previously difficult-to-reach, lower
parameter values on these aspects should go along with higher Internet
use for health information. The idea is: People with lower health commu-
nication skills will not be in a good position to acquire health information
by way of face-to-face communication. They will have a higher demand
for more health information, and therefore turn in higher proportion to
the new medium of the Internet. Alternately: People who do not trust
traditional media will have a high demand for reliable information, and
therefore turn to the new source of the Internet. For general media use,
the argument could run this way: As all media contain health informa-
tion, those with lower media use will encounter such information less
often than people with higher levels of media use. They will therefore
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
1 Magazines are the only medium available for comparison as health information use
in television, radio and newspapers was not asked about.
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 115
have a higher demand for such information, and consequently turn to the
Internet for it.
Conversely, the hypothesis derived from the disappointment of high
expectations with regard to new information media would mean, pro-
vided the data available in our study, that high levels of Internet use for
health information (both general and specifically for the present health
problem) in patients would go along with higher parameter values on the
four aspects.
These questions were tested, again separately for the two cities. In
all ensuing analyses, there are five comparisons: (1) Internet-experienced
patients vs. Internet-inexperienced; (2) among the Internet-experienced
those holding general health information from the Internet vs. those who
do not; (3) among the Internet-health-informed those having contacted
information on their present condition on the Internet vs. those who have
not. The first three comparisons approach information behavior step
by step. Comparison (4) is the gross total comparison of those who are
Internet-informed about health in general vs. all who are not (no experi-
ence with the Internet or never encountered health information there).
Comparison (5) is between those patients who have encountered Internet
information on their present condition and those who have not, regardless
of the reason (no experience with the Internet or having never encoun-
tered general or specific, condition-related health information there). All
comparisons are of scale means of the communication variables (skills,
trust, use of health-related content, general traditional media use), with
t-tests computed for the significance of differences.
Health communication skills – understood in a broad and not tech-
nical sense as the capacity of basic comprehension and expression skills
in the field of health information (Schulz & Nakamoto, 2006) – were
measured with four items: “When you read the instruction leaflet that
comes with medicaments, is it your impression that these leaflets are
easy to understand?” “When you have a chance to talk quietly with your
doctor, how well do you succeed in telling the doctor about your health
problem?” “When you do not understand what your doctor said, or when
you have doubts about it, is it easy or difficult for you to ask that the
doctor explain?” and “When you have a health problem, how well do
you normally succeed in explaining to your friends and family what the
116
problem is?” The question about the leaflet offered five response options,
the other questions offered four. Items were recoded so that high numbers
indicate high skills. For a summary analysis, an index was formed from
these items. The 4-item questions were recoded 0, 4, 8 and 12, and the
5-item question 0, 3, 6, 9, and 12. Scale averages were computed for index
formation. Respondents with missing values were left out. Factor analysis
over all four items resulted in a single-factor solution, but a low Cron-
bach’s α (.51), which could be increased to .57 if the leaflet item was
left out. We will therefore report results for the items separately, but also
include an index formed of the other three items.
Data indicate that in Lugano people with higher self-ascribed health
communication skills are more likely to have used the Internet. All five
Table 3: Communication Skills and Use of the Internet for Health
Information (General and Specific) among Lugano Patients
All Ever used
the Internet
(1)
Among (1):
Ever used
the Internet
for health
information
(2)
Among (2):
Ever used
the Internet
for infor-
mation on
present
condition
(3)
Among all:
Ever used
the Internet
for health
informa-
tion
(4)
Among
all: Ever
used the
Internet for
informa-
tion on
present
condition
(5)
Skills index 9.30 Yes 9.41* Yes 9.44 No 9.70* Yes 9.44+No 9.34
No 8.92 No 9.37 Yes 9.26 No 9.14 Yes 9.26
Can relate
trouble to doctor
9.42 Yes 9.54* Yes 9.64 No 9.89 Yes 9.64* Yes 9.49
No 9.01 No 9.32 Yes 9.49 No 9.17 No 9.41
Can relate trouble
to relatives
8.65 Yes 8.79* No 8.93 No 8.97 Yes 8.77 No 8.68
No 8.17 Yes 8.77 Yes 8.60 No 8.53 Yes 8.60
Easy to ask
doctor questions
9.68 Yes 9.79+Yes 9.84 No
10.25*
Yes 9.84 No 9.76
No 9.30 No 9.70 Yes 9.55 No 9.49 Yes 9.55
Leaflet easy
to understand
8.01 Yes 8.17** Yes 8.24 No 8.26 Yes 8.24* Yes 8.17
No 7.47 No 8.05 Yes 8.17 No 7.76 No 7.93
High scale values indicate higher skills, values run from 0 to 12. Yes/No refers to the
column entries. Yes > No suggests higher skills go along with higher Internet use.
No > Yes suggests the opposite. + p < .10; * p < .05; ** p < .01; *** p < .001
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 117
Table 4: Communication Skills and Use of the Internet for Health
Information (General and Specific) among Berne Patients, including
Summary of Lugano and Berne Results
All Ever used
the Internet
(1)
Among (1):
Ever used
the Internet
for health
information
(2)
Among (2):
Ever used
the Internet
for infor-
mation on
present
condition
(3)
Among all:
Ever used
the Internet
for health
informa-
tion
(4)
Among
all: Ever
used the
Internet
for infor-
mation on
present
condition
(5)
Skills index 9.32 No 9.46 No 9.40 Yes 9.23 No 9.46 No 9.36
Yes 9.26 Yes 9.23 No 9.23 Yes 9.23 Yes 9.23
Can relate
trouble to doctor
9.75 No 10.05 Yes 9.66 Yes 9.70 No 9.86 No 9.76
Yes 9.63 No 9.57 No 9.58 Yes 9.66 Yes 9.70
Can relate
trouble to relatives
8.50 No 8.76 Yes 8.53 No 8.84 Yes 8.53 No 8.59
Yes 8.44 No 8.20 Yes 8.27 No 8.47 Yes 8.27
Easy to ask
doctor questions
9.65 Yes 9.67 No 10.45* Yes 9.62 No 9.96+No 9.67
No 9.55 Yes 9.42 No 9.19 Yes 9.42 Yes 9.62
Leaflet easy
to understand
7.32 Yes 7.52 Yes 7.61 No 7.92 Yes 7.61+Yes 7.40
No 6.88 No 7.25 Yes 7.40 No 6.97 No 7.30
Yes > No 7 7 3
c) 7 3
No > Yes 3
c) 3 7 3
c) 7
Yes > No at p < .10 5
a) – – 4 –
No > Yes at p < .10 – 1 2
b) 1 –
Difference not
significant
5 9 8 5 10
High scale values indicate higher skills, values run from 0 to 12. Yes/No refers to the col-
umn entries. Yes > No suggests higher skills go along with higher Internet use. No > Yes
suggests the opposite. Summary rows refer to both Berne and Lugano (Table 3) values.
a)
All in Lugano;
b)
Both in Lugano;
c)
All in Berne; + p < .10; * p < .05; ** p < .01; *** p < .001
scale averages show significant differences (at least on the p < .10-level) in
this direction. In Berne no such differences appear (Columns 1 in Tables
3 and 4). Among those who have ever used the Internet, using it for health
information seems unrelated to health communication skills. The major-
ity of differences are in the direction of those with higher skills using the
118
Internet more for health information, but the only significant difference
is in the other direction (Column 2 in Tables 3 and 4). Among those who
have ever used the Internet for health information, using it for informa-
tion on the present condition also seems unrelated to health communi-
cation skills. The majority of comparisons and the only two significant
differences (both from Lugano) suggest rather that it is the less skilled
communicators who have used the Internet to learn about their present
medical condition (Column 3 in Tables 3 and 4). If general and specific
Internet health information users are compared with all others (Columns
4 and 5 in Tables 3 and 4), the differences that appear primarily reflect the
very first observation made above: that Lugano patients with high health
communication skills are more likely to have experience with the Web.
Tables 3 and 4 show all the scale averages and the significance levels of
the pairwise comparisons, separately for the two research sites as well as a
summary of both in the Berne table. If all 50 comparisons are considered,
27 (9 of them significant at least at p < .10) show a difference that indicates
persons with higher health communication skills use the Internet more for
health information, while 23 (4 of them significant) suggest the opposite.
Trust in a message or a source can be considered a condition for the
successful conveying of information. The concept, classically divided
into concepts of credibility/trustworthiness and perceived competence/
expertise (Hovland, Janis & Kelley 1953; Hovland & Weiss 1953), can
be defined as “the generalized expectancy that a message received is true
and reliable and that the communicator demonstrates competence and
honesty by conveying accurate, objective, and complete information”
(Renn & Levine 1991: 179). With this concept in mind, respondents’
trust in health information published in the media was measured with the
question: “How strongly do you normally trust health information in the
media? Please tick the corresponding answer in this table.” The table men-
tioned five forms of media (television, radio, newspaper, magazines, and
the Internet) in the rows. The columns showed a 4-digit verbal scale of
always, mostly, sometimes yes – sometimes no, and hardly ever. The scale
values were highly correlated, and a factor analysis revealed a single-factor
solution explaining 61 % of the variance. Cronbach’s α for a scale of all five
items was at .838 and hardly changed when the Internet item was left out.
For an index of trust in traditional media, the four values except Internet
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 119
were averaged, omitting cases with missing values. The index ranged from
1 to 4, with high values indicating high trust. The mean was 2.29.
Trust in health information in traditional media was positively related to
the use of the Internet among Berne patients, but not to the use of the Inter-
net for health information in general or specifically for the current problem.
But the Internet users’ higher trust surfaceswhen health information users
are compared with all patients (last two rows in Table 5). Among Lugano
patients, no relationship between trust in health information provided by
the traditional media and Internet use was found, neither for Internet use in
general, nor for its use for general or specific health information.
The use of health information in traditional media was measured only
for magazines. The question appeared as, “Many magazines publish arti-
cles on health and health problems. How many such articles do you read
in a typical week?” Open answers were averaged for the ensuing analysis.
Table 5: Trust in Health Information in Traditional Media and Use of
Internet for Health Information
Trust in media health
information among
Lugano patients
Trust in media health
information among
Berne patients
Ever used the Internet (1) No 2.33 t = -0.010
df = 454
p = .992
Yes 2.28** t = 2.844
df = 236
p = .005
Yes 2.33 No 1.98
Among (1): Ever used the
Internet for health informa-
tion (2)
No 2.38 t = -0.979
df = 371
p = .328
Yes 2.31 t = 1.145
df = 192
p = .254
Yes 2.32 No 2.19
Among (2): Ever used the
Internet for information on
present condition (3)
No 2.34 t = -0.634
df = 279
p = .527
Yes 2.32 t = 0.436
df = 148
p = .664
Yes 2.31 No 2.28
Among all: Ever used the
Internet for health informa-
tion (4)
No 2.35 t = -0.633
df = 449
p = .527
Yes 2.31** t = 2.691
df = 240
p = .008
Yes 2.32 No 2.08
Among all: Ever used the
Internet for information on
present condition (5)
No 2.35 t = -0.770
df = 450
p = .442
Yes 2.32+t = 1.816
df = 241
p = .071
Yes 2.31 No 2.17
High scale values indicate higher trust, values run from 1 to 4. Yes/No refers to the row
entries. Yes > No suggests higher trust in traditional media goes along with higher
Internet use. No > Yes suggests the opposite.
+ p < .10; * p < .05; ** p < .01; *** p < .001
120
The results are clear: A high consumption of health-related articles in
magazines has no relationship with Internet use in general, but in the
step-by-step analysis, there is a relationship with the use of Internet health
information in general appearing in both cantons, and with having seen
information pertaining to the condition which prompted respondents to
visit a doctor, showing up in Berne (Table 6).
Finally, general media use was taken into consideration. The ques-
tion asked for frequency of newspaper reading (everyday, almost every
day, some time each week, some time each month, hardly ever or never),
number of magazines read in a typical week (open) and hours of television
use on a weekday. The answers to the newspaper items were recoded so that
they reflect the number of days a newspaper is read in a week. The results
are clear: Newspaper reading is unrelated to Internet use in general, and
unrelated to being health-informed or condition-informed by the Internet
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
Table 6: Use of Health Information in Magazines and Use of Internet
for Health Information
Use of health informa-
tion in magazines among
Lugano patients
Use of health information
in magazines among
Berne patients
Ever used the Internet (1) Yes 1.59 t = 0.150
df = 373
p = .881
Yes 1.58 t = 1.035
df = 266
p = .302
No 1.55 No 1.31
Among (1): Ever used
the Internet for health
information (2)
Yes 1.73* t = 2.394
df = 302
p = .017
Yes 1.74* t = 1.996
df = 204
p = .047
No 1.17 No 1.14
Among (2): Ever used the
Internet for information on
present condition (3)
Yes 1.74 t = 0.017
df = 229
p = .987
Yes 1.93 t = 1.436
df = 152
p = .153
No 1.74 No 1.46
Among all: Ever used
the Internet for health
information (4)
Yes 1.73* t = 1.990
df = 373
p = .047
Yes 1.74* T = 2.013
df = 270
p = .045
No 1.36 No 1.30
Among all: Ever used the
Internet for information on
present condition (5)
Yes 1.74 t = 1.180
df = 373
p = .239
Yes 1.93* t = 2.525
df = 270
p = .012
No 1.51 No 1.35
High scale values indicate higher number of articles read. Yes/No refers to the row entries.
Yes > No suggests higher magazine use goes along with higher Internet use. No > Yes
suggests the opposite.
+ p < .10; * p < .05; ** p < .01; *** p < .001
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 121
in both Lugano and Berne. There is some indication that avid magazine
readers among the Berne patients are more likely to be Internet-informed
on health in general, in addition to their present condition. And heavy
television use among patients in both Berne and Lugano corresponds with
a lower probability of Internet use in general, or for health-related matters,
or for the individual’s present condition (Tables 7 and 8). The effect for
television use is to a large degree created by age: Older people watch more
television and use the Internet less than younger people.
To summarize: There is more evidence that health information on the
Internet is sought or accidentally encountered more often by people who
are also reachable for health information over traditional channels. This is
Table 7: General Media Use and Use of Internet for Health Information
among Lugano Patients
Number of days per
week that a newspa-
per is read
Number of
magazines looked at
per week
Hours per day
spent watching
television
Ever used the
Internet (1) Yes 4.93 t = 0.101
df = 568
p = .920
No 2.51 t = -0.975
df = 504
p = .330
No 2.29*** t = -5.126
df = 573
p < .001
No 4.90 Yes 2.26 Yes 1.65
Among (1): Ever
used the Internet
for health infor-
mation (2)
Yes 5.03 t = 1.109
df = 440
p = .268
Yes 2.32 t = 0.951
df = 397
p = .342
No 1.81+t = -1.743
df = 438
p = .082
No 4.76 No 2.09 Yes 1.59
Among (2): Ever
used the Internet
for information
on present condi-
tion (3)
Yes 5.16 t = 0.952
df = 308
p = .342
No 2.38 t = -0.319
df = 285
p = .750
No 1.65
t = -0.845
df = 308
p = .399
No 4.91 Yes 2.29 Yes 1.53
Among all: Ever
used the Internet
for health infor-
mation (4)
Yes 5.03 t = 0.946
df = 567
p = .344
Yes 2,32 t = 0.075
df = 505
p = .940
No 2.06***
t = -4.480
df = 569
p < .001
No 4.84 No 2.31 Yes 1.59
Among all: Ever
used the Internet
for information
on present condi-
tion (5)
Yes 5.16 t = 1.454
df = 566
p = .146
No 2.33 t = -0.178
df = 504
p = .859
No 1.90***
t = -3.213
df = 572
p = .001
No 4.84 Yes 2.29 Yes 1.53
Yes/No refers to the row entries. Yes > No suggests higher media use goes along with high-
er Internet use. No > Yes suggests the opposite.
+ p < .10; * p < .05; ** p < .01; *** p < .001
122 SCH ULZ, CA IATA ZU FFE RE Y & HA RTU NG
the case for Lugano patients with higher health communication skills (as
far as general Internet use is concerned), for Berne patients who put high
trust in health information in media other than the Internet, again mostly
for Internet use in general, and for both Berne and Lugano patients who
read many magazine articles on health subjects were also found more likely
to be health-informed (and in Berne also condition-informed) by the Inter-
net. Correlations in the other direction, indicating that people difficult to
reach due to lower communication skills, lower trust in health informa-
tion, lower use of such information and lower media use are more likely to
use the Internet for health information also appear at places, but consist-
ently only for television use, and these are very likely to be spurious.
Table 8: General Media Use and Use of Internet for Health Information
among Berne Patients
Number of days per
week that a newspa-
per is read
Number of
magazines looked at
per week
Hours per day
spent watching
television
Ever used the
Internet (1) Yes 5.63 t = 0.829
df = 309
p = .408
No 1.60
t = -0.006
df = 273
p = .996
No 1.97** t = -3.100
df = 306
p = .002
No 5.41 Yes 1.60 Yes 1.46
Among (1): Ever
used the Internet
for health infor-
mation (2)
No 5.82 t = -0.893
df = 232
p = .373
Yes
1.72*
t = 2.097
df = 211
p = .037
No 1.49 t = -0.209
df = 231
p = .835
Yes 5.56 No 1.23 Yes 1.45
Among (2): Ever
used the Internet
for information
on present condi-
tion (3)
No 5.66 t = -0.582
df = 169
p = .562
Yes 1.91 t = 1.604
df = 157
p = .111
Yes 1.50 t = 0.453
df = 169
p = .651
Yes 5.49 No 1.51 No 1.42
Among all: Ever
used the Internet
for health infor-
mation (4)
No 5.66 t = -0.434
df = 322
p = .664
Yes 1.72 t = 1.406
df = 278
p = .161
No 1.81* t = -2.493
df = 316
p = .013
Yes 5.56 No 1.46 Yes 1,45
Among all: Ever
used the Internet
for information
on present condi-
tion (5)
No 5.67 t = -0.697
df = 322
p = .486
Yes 1.91* t = 2.185
df = 278
p = .030
No 1.67 t = -1.055
df = 316
p = .292
Yes 5.49 No 1.48 Yes 1.50
Yes/No refers to the row entries. Yes > No suggests higher media use goes along with high-
er Internet use. No > Yes suggests the opposite.
+ p < .10; * p < .05; ** p < .01; *** p < .001
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 123
4.5. Summary of Results
– Three in four patients indicated they had had some experience with
the Internet. Among those, roughly two-thirds (or about 50 % of the
total sample) said they had ever searched or incidentally found health
information on the Internet before. Among these, roughly every second
had searched or found information on the Internet that was related to
the motive of seeing a doctor at the time of the survey. Based on all
respondents again, the share is 28 %.
– Due to a higher frequency of personal communication, the additional
share of people to whom Internet-derived health information was
transmitted was higher in Lugano than in Berne.
– Men use the Internet more frequently, but women search more often for
health information, and male patients for information on their present
health-related problem. This creates roughly equal shares of men and
women who show up for consultation with some prior contact to Inter-
net information pertaining to their problem.
– As can be expected, Internet-derived information is much more
common for younger and better-educated groups.
– Three-fourths of patients can be typified as chance finders of health
information in the media. The remainder is equally distributed
(roughly) between types defined as seekers, overlookers and avoiders.
– Seekers more often turn to the Internet for general or specific health
information, followed by chance finders, avoiders and overlookers. For
other media content, a very similar pattern emerges.
– Patients who can more easily be reached by health messages on tradi-
tional channels also turn to the Internet, either generally or specifically,
for health information. The scattered evidence on this is related to
health communication skills, trust in health information in traditional
media, and use of health information in traditional media. All these
forces seem at times to increase patients’ likelihood to have encoun-
tered general or specific health information on the Web. Evidence for
the reverse relationship (that Internet health information goes along
with low skill, low trust in and low use of traditional media for health
information) was weaker.
124
5. Discussion
Most existing results on the share of Internet-informed patients can hardly
be directly compared with ours, due to different research sites, different
dates of data collection, and different medical conditions in the patients
who were interviewed (Gordon et al. 2002; Ross et al. 2000). The com-
parison of our results with those by Jeannot et al. (2004), however, at least
suggest that between 2001 (Jeannot et al.’s time of data collection) and
2006/07 (our data) the rate of having consulted the Web before a visit to
one’s doctor has considerably increased. The problems often associated
with patients’ bringing Internet-based knowledge to medical consulta-
tion (Murray et al. 2003a, 2003b; Wilson 1999; Potts & Wyatt 2002;
Chen & Siu 2001; Helft et al. 2003; Tann et al. 2003; Pautler et al. 2001;
Hardey 1999; Johnson & Ramaprasad 2000) may have increased to a
similar degree, but so may have the benefits of Internet information for the
doctor-patient relationship (Murray et al. 2003a). Analysis of the quali-
tative part of this study showed that most of the interviewed physicians
appreciated discussing Internet-derived health information with patients,
but also that misleading interpretations by patients and views contrary to
those held by the physician caused conflicts during consultations (Som-
merhalder et al. 2009). The qualitative analysis of the physicians’ inter-
views further shows that physicians respond to Internet-informed patients
in four different ways: by resisting online information (resistance), by
repairing online information (repairing), by constructing a shared reality
with the patient starting from the online information (co-construction),
and by enhancing online information (enhancement). Physicians opt for
a particular communicative strategy on the basis of their conception of
medical information for lay people through the Internet and on the basis
of their interpretation of the specific communicative context, that is, their
appraisal of the patient’s health literacy, the relevance of the online infor-
mation to be discussed, and their own communicative efficacy (Caiata
Zufferey et al., 2010; Caiata-Zufferey & Schulz, under review).
Our results show that for the younger age groups and the better edu-
cated it is no longer a small minority that appear in physicians’ practices
with some Internet-acquired knowledge on their present health problem.
To be sure, it is a minority still, but the numbers are approaching 50 %.
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 125
Among the elderly and those with lower levels of formal schooling, there are
fewer patients with Internet-acquired information on their present problem,
but among these groups, physicians can expect one or two in ten patients
to arrive with such information. As the younger and better educated age
cohorts will get older and will thus seek medical advice more often than
they do today, we can expect that within a generation, checking the Web
first and visiting a doctor afterward will be the rule rather than the excep-
tion, provided that the Internet-oriented young generation of today will not
lose its penchant for the Web, and also provided that health information on
the Internet will continue to prosper. As communication skills growmore
important in medical training, the ways of dealing with patients’ knowl-
edge derived from the Internet should receive some attention, too.
Information seeking is an activity that has become much easier to engage
in since the advent of the Internet, especially when specific information is
sought. Most health information sought by patients is very specific, and
any attempt to find such information in traditional mass media (news-
papers, magazines, radio, television) would in most cases be completely
futile. Practically the only ways to actively seek specific health information
in the pre-Internet days were to visit a library, go to a bookstore or ask an
expert. The Internet has made active information seeking much easier,
and it yields more diverse information and information easier to process
than could have been imagined in pre-Internet days. It is therefore not
surprising at all that, as our results show, patients who describe themselves
as habitual active seekers of health information more often than others
consult the Web before seeing their doctor. What is surprising, however,
is that roughly every third chance finder of health information as well as
every third health information avoider also checked the Internet before
seeing their physician. That means that health information is found (and
possibly also actively sought) by patients who still do not consider them-
selves active health information seekers. This can be interpreted as indica-
tion of the fact that the Internet as an easily available information-seeking
medium in the field of health is not yet as salient in everyday thinking and
behavior as might be expected. Positive experience with Internet-derived
health knowledge (within and without medical consultation) will likely
augment this salience and further increase the number of patients who
present themselves at their doctor after having consulted the Web.
126
Finally, the plotting of hopes (that the use of health information from
the Web might balance out social inequality based in traditional ways of
informing people about health) against a disappointment perspective (that
holds that the Internet as a new medium will benefit those who already ben-
efitted from older media) did not yield much support for the hope perspec-
tive. The analyses presented here provide more support for the assumption
that it is mostly patients who can more easily be reached by health messages
on traditional channels who also turn to the Internet, either generally or spe-
cifically, for health information. In other words: There is more evidence for
the disappointment than for the hope interpretation sketched above. This
is true especially for hopes that a social divide in health care can be bridged
by the Internet. Rather we have to assume that health-related aspects of the
digital divide will persevere, an expectation that has also been noted in the
literature (Wang, Bennett & Probst 2011; Kontos, Bennett & Viswanath
2007; Norman & Skinner 2006). The finding implies that in whatever way
the possibility of checking the Internet before one sees one’s doctor will
change health care in general and medical consultation in particular, this
will not likely to be similar for all types of patients. Increasing gaps between
well-informed, misinformed and uninformed patients will pose new chal-
lenges to physicians, who may have to face increasing problems in assessing
their patients’ state of knowledge on their condition. But caution needs to
be applied here as the results are inconsistent when the two research sites are
compared, as evidence to the contrary also appears, and as no differences
emerge in many comparisons. The link between Internet health communi-
cation behavior (both general and specific) with reachability by traditional
channels is weak at best. But still, there is little grounds for hope that those
patients will benefit from the Internet who have so far been disadvantaged
in their chances to acquire health information.
References
ahern, D.K.; KreSlaKe, J.M. & Phalen, J.M. (2006). What is eHealth (6): Perspectives
on the Evolution of eHealth Research. Journal of Medical Internet Research [Online]
8/1: e4 (last retrieved on 15.03.2011, from: http://www.jmir.org/2006/1/e4/).
aKer Ka r, S.M.; Kanit Ka r, M. & Bichile, l.S. (2005). Use of the Internet as a Re-
source of Health Information by Patients: A Clinic-based Study in the Indian Pop-
ulation. Journal of Postgraduate Medicine 51/2: 116–118.
SCH UL Z, CA IATA ZUFFERE Y & H ARTU NG
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 127
aMMann, y. (2000). Rapport sur les principaux résultats du sondage: santé et informa-
tion. Sanimedia. Service Cantonal de Recherche et d’Information Statistiques du
Canton de Vaud (SCRIS). Lausanne: SCRIS.
anDerSon, J.g. (2004). Consumers of E-health: Patterns of Use and Barriers. Social
Science Computer Review 22: 242–248.
anDerSon, J.g.; rainey, M.r. & eySenBach, g. (2003). The Impact of Cyber-health-
care on the Physician-patient Relationship. Journal of Medical Systems 27/1: 67–84.
Ball, M.J. & lilliS, J. (2001). E-health: Transforming the Physician/patient Relation-
ship. International Journal of Medical Informatics 61/1: 1–10.
caiata zuffer ey, M. et al. (2010). Online Health Information Seeking in the Con-
text of the Medical Consultation in Switzerland. Qualitative Health Research 20/8:
1050–1061.
caiata zuffe rey, M. & Schul z, P.J. (under review). Physicians’ Communication
Strategies in Interacting with Internet-informed Patients. Results from a Qualita-
tive Study.
chen, X. & Siu, l.l. (2001). Impact of the Media and the Internet on Oncology:
Survey of Cancer Patients and Oncologists in Canada. Journal of Clinical Oncol-
ogy 19: 4291–4297.
cline , r.J.w. (2003). At the Intersection of Micro and Macro: Opportunities and
Challenges for Physician–patient Communication Research. Patient Education and
Counseling 50/1: 13–16.
cline, r.J.w. & hayneS, K.M. (2001). Consumer Health Information Seeking on the
Internet: The State of the Art. Health Education Research 16/6: 671–692.
coiera, e. (1996). The Internet’s Challenge to Health Care Provision. British Medical
Journal 312: 3–4.
cotten, S.r. (2001). Implications of Internet Technology for Medical Sociology in
the New Millennium. Sociological Spectrum 21/3: 319– 40.
coulter, a. & Magee, h. (eds.). (2003). The European Patient of the Future. Maid-
enhead: Open University Press.
DaviS, t.c. & wolf, M.S. (2004). Health Literacy: Implications for Family Medicine.
Family Medicine 36(8): 595–98.
Diaz , J.a. et al. (2002). Patients’ Use of the Internet for Medical Information. Journal
of General Internal Medicine 17/3: 180–185.
Diaz , J.a. et al. (2005). What Types of Internet Guidance do Patients want from their
Physicians? Journal of General Internal Medicine 20/8: 683–685.
Dolan, g. et al. (2004). Consumer Use of the Internet for Health Information: A
Survey of Primary Care Patients. International Journal of Consumer Studies 28/2:
147–153.
euroPean coMMiSSion (2010). E-Communications Haushaltsumfrage: Bericht. Spe-
zial Eurobarometer 335 (last retrieved on 15.03.2011, from: http://ec.europa.eu/
public_opinion/archives/ebs/ebs_335_de.pdf).
EySenBach, g. (2003). The Impact of the Internet on Cancer Outcomes. CA: A Cancer
Journal for Clinicians 53: 356–371.
128 SCH ULZ, CA IATA ZU FFE RE Y & HA RTU NG
eySenBach, g.; ryoung, S.e. & DiePgen, t.l. (1999). Shopping around the Internet
Today and Tomorrow: Towards the Millennium of Cybermedicine. British Medical
Journal 319: 1294–1298.
foX, S. & rainie, l. (2002). Vital Decisions: How Internet Users decide what Infor-
mation to trust when they or their Loved Ones are Sick: Technical Report from the
Pew Internet and American Life Project. Washington, DC: Pew Research Center.
gerBer, B.S. & eiSer, a.r. (2001).The Patient-physician Relationship in the Inter-
net Age: Future Prospects and the Research Agenda. Journal of Medical Internet
Research [Online] 3(2): e15 (last retrieved on 10.01.2011, from: http://www.jmir.
org/2001/2/e15).
gorDon, M.M.; caPell, h.a. & MaDhoK, r. (2002). The Use of the Internet as a Re-
source for Health Information among Patients attending a Rheumatology Clinic.
Rheumatology 41: 1402–1405.
harDey, M. (1999). Doctor in the House: The Internet as a Source of Lay Health
Knowledge and the Challenge to Expertise. Sociology of Health & Illness 21/6:
820–835.
helft, P.r.; hluBocKy, f. & Daugherty, c.K. (2003). American Oncologists’ Views
of Internet Use by Cancer Patients: A Mail Survey of American Society of Clinical
Oncology Members. Journal of Clinical Oncology 21/5: 942–947.
hellawell, g.o. et al. (2000). Urology and the Internet: An Evaluation of Internet
Use by Urology Patients and Information available on Urological Topic. British
Journal of Urology International 86: 191–194.
houSton, t.K. & alliSon, J.J. (2002). Users of Internet Health Information: Differ-
ences by Health Status. Journal of Medical Internet Research [Online] 4/2: e7 (last
retrieved on 10.01.2011, from: http://www.jmir.org/2002/2/e7).
HovlanD, c.i.; JaniS, i.l. & Kelley, h.h. (1953). Communication and Persuasion.
New Haven, CT: Yale University Press.
hovlanD, c.i. & weiSS, w. (1953). The Influence of Source Credibility on Commu-
nication Effectiveness. Public Opinion Quarterly 15: 635–650.
JaDaD, a.r. (1999). Promoting Partnerships: Challenges for the Internet Age. British
Medical Journal 319: 761–764.
Jeannot, J.g. et al. (2004). Patient Use of the Internet for Health Care Information in
Switzerland. Swiss Medical Weekly 134: 307–312.
JenKinS, v.; fallowfielD, l. & Saul, J. (2001). Information Needs of Patients with
Cancer: Results from a Large Study in UK Cancer Centres. British Journal of Can-
cer 84: 48–51.
JohnSon, g.l. & raMa PraS aD, a. (2000). Patient-physician Relationship in the Infor-
mation Age. Marketing Health Services 20/1: 20–27.
JoneS, r.B. et al. (2001). The Clinician’s Role in Meeting Patient Information Needs:
Suggested Learning Outcomes. Medical Education 35: 565–571.
KaSSir er, J.P. (2000). Patients, Physicians and the Internet. Health Affairs 19/6: 115–123.
KicKBuSh, i.S. (2001). Health Literacy: Addressing the Health and Education Divide.
Health Promotion International 16: 289–297.
FIR ST CHE CK T HE INTER NET, TH EN SEE T HE DOC TOR 129
Koller, M. et al. (2001). Use of the Internet by Medical Doctors in Switzerland. Swiss
Medical Weekly 131: 251–254.
KontoS, e.z.; Bennett, g.g. & viSwanath, K. (2007). Barriers and Facilitators to
Home Computer and Internet Use among Urban Novice Computer Users of Low
Socioeconomic Position. Journal of Medical Internet Research [Online] 9/4: e31 (last
retrieved on 15.03.2011, from: http://www.jmir.org/2007/4/e31).
KrePS, g.l. & neuhauSer, l. (2010). New Directions in eHealth Communication:
Opportunities and Challenges. Patient Education and Counseling 78/3: 329–336.
laza rSfelD, P.f.; BerelSon, B. & gauDet, h. (1944). The People’s Choice: How the Vot-
er makes up his Mind in a Presidential Election. New York: Duell, Sloan and Pierce.
levy, J.a. & StroMBecK, r. (2002). Health Benefits and Risks of the Internet. Journal
of Medical Systems 26/6: 495–510.
MaDDen, M. & foX, S. (2006). Finding Answers Online in Sickness and in Health.
Pew Internet & American Life Project Report (last retrieved on 17.11.2010, from:
http://pewinternet.com/~/media/Files/Reports/2006/PIP_Health_Decisions_
2006.pdf.pdf).
McKinStry, B. (2000). Do Patients wish to be involved in Decision Making in the
Consultation? A Cross Sectional Survey with Video Vignettes. British Medical Jour-
nal 321: 867–871.
McMullan, M. (2006). Patients using the Internet to obtain Health Information:
How this affects Patient-health Professional Relationship. Patient Education and
Counseling 63: 24–28.
Mullner, r.M. (2002). The Internet and Healthcare: Opportunities and Challenges.
Journal of Medical Systems 26/6: 491–493.
Murr ay, e. et al. (2003a). The Impact of Health Information on the Internet on
Health Care and the Physician-patient Relationship: National U.S. Survey among
1,050 U.S. Physicians. Journal of Medical Internet Research [Online] 5/3: e17 (last
retrieved on 10.01.2011, from: http://www.jmir.org/2003/3/e17).
Murr ay, e. et al. (2003b). The Impact of Health Information on the Internet on the
Physician-patient Relationship: Patient Perceptions. Archives of Internal Medicine
163/14: 1727–1734.
norMa n, c.D & SKinner, h.a. (2006). eHealth literacy: Essential SKill S for con-
SuMer health in a networKeD worlD. Journal of Medical Internet Research [Online]
8/2: e9 (last retrieved on 15.03.2011, from: http://www.jmir.org/2006/2/e9).
o’connor, J.B. & JohanSon, J.f. (2000). Use of the Web for Medical Information by
a Gastroenterology Clinic Population. Journal of the American Medical Association
284/15: 1962–1964.
Pautler, S.e. et al. (2001). Use of the Internet for Self-education by Patients with
Prostate Cancer. Urology 57/2: 230–233.
PeMBerton, P.J. & golDBl att, J. (1998). The Internet and the Changing Roles of
Doctors, Patients and Families. The Medical Journal of Australia [Online] 169/11-
12: 594–595 (last retrieved on 10.01.2011, from: http://ww w.mja.com.au/public/
issues/xmas98/pemberton/pemberton.html).
130 SCH UL Z, CA IATA ZU FFE RE Y & H ARTU NG
PottS, h.w.w. & wyatt, J.c. (2002). Survey of Doctors’ Experience of Patients using
the Internet. Journal of Medical Internet Research [Online] 4/1: e5 (last retrieved on
10.01.2011, from: http://www.jmir.org/2002/1/e5).
Powell, J. & cl arKe, a. (2002). The WWW of the World Wide Web: who, what,
and why? Journal of Medical Internet Research [Online] 4/1: e4 (last retrieved on
10.01.2011, from: http://www.jmir.org/2002/1/e4).
Powell, J.; Darvell , M. & gr ay, J. (2003). The Doctor, the Patient and the World
Wide Web: How the Internet is Changing Healthcare. Journal of the Royal Society
of Medicine 96: 74–76.
renn, o. & le vine, D. (1991). Credibility and Trust in Risk Communication. In:
r.e. K aSPerSon, & P.J.M. Stall en (eds.). Communicating Risks to the Public:
International Perspectives. Dordrecht: Kluwer: 175–218.
roSS, J. et al. (2000). How much Interest is the Internet to Patients? Sexually Transmit-
ted Infections 76/5: 393–394.
Schulz, P.J. & na Ka Moto, K. (2005). Emerging Themes in Health Literacy. Special
Issue on Health Literacy. Studies in Communication Sciences 5: 1–10.
SoMMer ha lDer, K. et al. (2009). Internet Information and Medical Consultations:
Experiences from Patients’ and Physicians’ Perspectives. Patient Education and
Counseling 77: 266–271.
SPaDaro, r. (2003). European Union Citizens and Sources of Information about
Health. Eurobarometer 58.0 Special Surveys Report. The European Opinion
Research Group (last retrieved on 15.03.2011, from: http://ec.europa.eu/public_
opinion/archives/ebs/ebs_179_en.pdf).
tichenor, P.J.; Donohu e, g.a. & olien, c.n. (1970). Mass Media Flow and Dif-
ferential Growth in Knowledge. Public Opinion Quarterly 34: 159–170.
waal, e. De & SchoenBach, K. (2010). News Sites’ Position in the Mediascape: Uses,
Evaluations and Media Displacement Effects over Time. New Media Society 12:
477–496.
wang, J.y.; Ben nett, K. & ProBSt, J. (2011). Subdividing the Digital Divide: Dif-
ferences in Internet Access and Use among Rural Residents with Medical Limita-
tions. Journal of Medical Internet Research [Online] 13/1: e25 (last retrieved on
15.03.2011, from: http://www.jmir.org/2011/1/e25).
wilSon, S.M. (1999). Impact of the Internet on Primary Care Staff in Glasgow. Jour-
nal of Medical Internet Research [Online] 1/2: e7 (last retrieved on 10.01.2011, from:
http://www.jmir.org/1999/2/e7).
wyatt, J.c. & Sullivan, f. (2005). eHealth and the Future: Promise or Peril. British
Medical Journal 331/7529: 1391–1393.
Submitted: 14 June 2010; Resubmitted: 19 January 2011; Resubmitted: 25 February
2011; Accepted: 17 March 2011; Refereed anonymously.
