OUP user menu

A Survey of Patients' Use of the Internet for Chronic Pain-Related Information

Tomas B. Corcoran MB, BCh, BAO, MRCPI, FCARCSI, MD, FJFICM-ANZCA, Fran Haigh BSc, MB, BCh, FRCA, Amanda Seabrook, Stephan A. Schug MD FANZCA FFPMANZCA
DOI: http://dx.doi.org/10.1111/j.1526-4637.2010.00817.x 512-517 First published online: 1 April 2010


Objective. To investigate the use of the Internet as a pain information seeking tool among a population of patients attending a chronic pain clinic.

Methods. A bespoke self-completing questionnaire was given to 150 patients attending 17 consecutive chronic pain clinics at The Royal Perth Hospital during August and September 2007.

Results. One hundred twenty-two completed surveys were received, a response rate of 81%. Only 23.8% of the patients had used the Internet to access pain-related health care information. There was no gender difference between those who did and did not access the Internet for information. Age group, highest educational level attained, and the availability of Internet access were all significantly associated with the use of the Internet to search for pain-related information. 41.4% described the information they found as useful, 6.9% found it frightening and 10.3% found it confusing. Forty-four percent wanted more information to be available on the Internet while only 6.9% planned to discuss their findings with their doctor.

Conclusions. When compared with other studies about patient information-seeking behavior, a smaller than expected percentage of patients attending chronic pain clinics in Perth used the Internet to search for information about pain. There are a variety of reasons for this that would suggest that health care professionals should not be complacent but seek to maximize the potential of the Internet to inform our patients by advising them how and where to look for relevant information.

  • Pain
  • Chronic Pain
  • Internet
  • Patient Information


Internet-search behavior among patients and the quality of the available information is a concern for many fields of medicine, particularly in relation to how this may affect doctor–patient relationships [1]. A total of 4.5% of all Internet searches are for health-related information [2], and with an ever expanding population of Internet users the number of health related Websites accessed will inevitably increase. The adult population in developed countries have 40–60% Internet access [3] and approximately 40% of patients or carers who attend outpatient clinics use the Internet to access health-related information [3,4]. The quality of such information, its accessibility, and interpretation by patients, remains a topic of unresolved debate [3–8].

Chronic pain is a significant health care burden, afflicting 17% of males and 20% of females in the Australian population at some stage in their lives [9]. Patients with such chronic illnesses are more likely to source information related to their condition on the Internet [10]. According to previous studies, individuals who access health care related Internet sites are more likely to be female and have specific health-reasons (diagnosed with a new health problem, ongoing medical condition, prescribed new medication or treatment) [7]. Whether such retrieved information influences patient decision-making and illness perception remains undetermined [4]. For certain chronic pain conditions, use of the Internet as an information management tool has reported some success [11], but the available information, as with other specialties, is poor [5,12,13]. Generally, the information gained fails to meet patients' expectations [14,15]. Nevertheless, 39% of patients attending a pain clinic sought pain-related information on the Internet [16]. English-speaking patients would be expected to have a higher prevalence of Internet searching for medical information [17]. We performed a survey of patients, who are predominantly English speaking, attending our chronic pain clinic. We sought to examine how patients use the Internet to search for and retrieve information about chronic pain so that we could build a profile of the types of topics searched for and the attitudes of the patients to the retrieved information. This information, we believe, is important in determining how we might promote self-education to patients via targeted browsing of quality information.


The Pain Medicine Centre in Royal Perth Hospital is a tertiary referral center at a University teaching hospital. The Centre sees around 3,500 outpatients per year; of these appointments around 500 are new case referrals. The institutional ethics committee waived the need for formal consent for this study, although this was implied by voluntary participation and completion of the questionnaire. At 17 consecutive pain clinics during August and September 2007, a bespoke questionnaire (see Appendix) was distributed to patients to complete while waiting and was collected at the end of clinic. The questionnaire was designed to be a simple one-page leaflet, tick-box based and requiring a maximum of 10 minutes for completion. Recorded data included demographics, patterns of Internet usage, pain-condition related information, and specific topics searched for on the Internet. Categorical data were analyzed using Pearson's chi-square coefficient. Multivariate logistic regression analysis was employed to examine factors that may prove predictive of Internet usage. All factors that were significant on univariate analysis were entered into the regression procedure. A P value of <0.05 was considered significant and all statistical tests were performed with SPSS for Windows (version 15, SPSS Inc, Chicago, IL).


Of 150 attendees, 122 completed surveys were received, a response rate of 81%. Of this group, 29 (23.8%) patients used the Internet to access pain-related health care information. There was no gender difference between those who did and did not access the Internet for such information. However, age group, highest educational level attained and the availability of Internet access were all significantly associated with the use of the Internet to access pain-related information on univariate analysis (Table 1).

View this table:
Table 1

Patient demographics

Entire Group
n = 122
No Search
n = 93
Internet Search
n = 29
P *
Gender 0.38
  M64 (52.5)50 (53.8)14 (48.3)
  F58 (47.5)43 (46.2)15 (51.7)
Age group 0.006
  21–3511 (9)4 (4.3)7 (24.1)
  36–5032 (26.2)23 (24.7)9 (31)
  51–6541 (33.7)34 (36.6)7 (24.1)
  >6538 (31.1)32 (34.4)6 (20.7)
Highest educational level 0.001
  Nonsecondary49 (40.2)41 (44.1)8 (27.6)
  Secondary25 (20.5)18 (19.4)7 (24.1)
  Tafe11 (9)8 (8.6)3 (10.3)
  Tertiary/University13 (10.7)4 (4.3)9 (31)
  Unspecified24 (19.7)22 (23.7)2 (6.9)
Computer access
  Home60 (49.2)36 (38.7)24 (82.8)<0.001
  Work13 (10.7)7 (7.5)6 (20.7) 0.054
  Home or work62 (50.8)38 (40.9)24 (82.8)<0.001
  Any access65 (53)39 (41.9)26 (89.7)<0.001
First visit17 (13.9)13 (14)4 (13.8)0.624
Internet pain search29 (23.8)
  • * P values refer to Pearson's chi-square.

Nonsecondary education was over-represented among those who did not access the Internet; however, 24 (19.7%) patients did not specify their highest educational level attained. The patient population self-reported a broad range of chronic pain conditions, with chronic back pain being over-represented among those who used the Internet to access pain-related information (Table 2). Thirty-five patients (28.7%) did not list their pain condition diagnosis and this was taken to be unknown. Thirty-seven (30.3%) patients listed only the severity of their condition as a diagnosis. Thus, 72 patients (59%) were recorded as having an unspecified diagnosis. The most frequently researched topics in relation to the patients' pain condition were “exercise,”“depression,”“physiotherapy,” and “surgery” (Table 3).

View this table:
Table 2

Diagnostic groups*

Entire Group
n = 122
Internet Search
n = 29
No Search
n = 93
Back pain21 (17.2)8 (28)13 (14)
Nerve pain11 (9)2 (6.9)9 (9.7)
Osteoarthritis5 (4.1)1 (3.4)4 (4.3)
CRPS5 (4.1)4 (13.8)1 (1.1)
Neck pain3 (2.5)03 (3.2)
Occipital neuralgia3 (2.5)2 (6.9)1 (1.1)
Fibromyalgia2 (1.6)02 (2.2)
Unspecified72 (59)12 (41.4)60 (64.5)
  Unknown35 (28.7)6 (20.7)29 (31.2)
  Severe33 (27)6 (20.7)27 (29)
  Moderate4 (3.3)04 (4.3)
  • * P = 0.029; refer to overall Pearson's chi-square.

View this table:
Table 3

Topics examined by patients

Topicsn = 29
Mood disorders11
Drug therapy10
Coping strategies 9
Acupuncture 8
Anxiety 8
Herbal medicine 8
Chiropractic 7
Loss of control 7
Occupational therapy 7
Drug addiction 6
Sexual dysfunction 6
Cognitive behavioral therapy 5
Isolation 5
Job dissatisfaction 5
Reflexology 5
Relationships 5
Suicide 5
Electrical stimulation 3
Morphine pump 1
Spinal cord stimulation 1
Somatization 1

Of those patients who researched pain on the Internet, 12 (41.4%) described the information as useful, two (6.9%) found it frightening, and three (10.3%) found it confusing. Thirteen (44%) wanted more information about chronic pain to be available on the Internet. Only two (6.9%) respondents planned to discuss the information they had found with their doctor.

Of the 122 patients who responded to the survey, 65 (53%) had Internet access available to them. For 62 (50.8%) patients this access was either at home or at work, for a further three individuals this facility was limited to a public site. In the examination of the relationship between Internet usage and demographic and disease variables, logistic regression analysis was used after initial univariate analyses. In the modeling process, all potential prognostic variables were included initially but variables with a P value of >0.25 were eliminated stepwise. The only factor that significantly predicted patients Internet usage was the availability of Internet access at home; patients with home Internet access being more likely than those without to research their pain condition (OR 6.329 [1.85–21.5], P = 0.003). The Hosmer-Lemeshow C χ2 statistic for the goodness of fit of the Model was 7.25 (P = 0.509) (Table 4).

View this table:
Table 4

Logistic regression model for prediction of Internet usage to research pain

ModelVariables in the Multivariate Analysis Model*Adjusted Odds Ratio
(95% CI)
Home Internet access6.3(1.85–21.5)0.003
  • * The Hosmer-Lemeshow C χ2 statistic for the goodness of fit of model was 7.25 (P = 0.509).


This is the first time the information-seeking behavior on the Internet of a chronic pain population has been studied in Australia. The key finding of the study was that only 23.8% of our patients have made a search on the Internet for chronic pain-related information. In 2007, 39% of a Dutch pain clinic population had made such a search [16]. Previous studies have shown that the percentage of patients who search online for medical information is higher in English speaking countries than in non-English speaking countries [17]; and given that Australia is an English-speaking country, one would have predicted a higher search rate among our population. This was not what we found. The only comparable Australian data for Internet search behavior among clinic patients is for oncology patients. Newnham et al. found 42% of patients attending their oncology clinic had used the Internet to search for information about their condition [18].

The surprisingly small percentage of our patients using the Internet to seek pain-related information raises a number of questions. If Internet access and education are a significant factor, then perhaps the result is skewed by the demographic of patients that the Royal Perth Hospital serves: it has a higher proportion of ethnic immigrants, aboriginal and inner city citizens than other Perth hospitals. A total of 60.5% of the sample population left school at or before the age of 18 and 47% stated they had no access to the Internet. This is significantly different from the general population in Australia where in December 2007, 75.9% of the population were estimated to have access to the Internet [19]. The aforementioned Dutch and Australian studies both surveyed a patient population with a significantly higher rate of Internet access than our study group [16,18].

There may be other reasons why so few of our patients sought pain-related information on the Internet. It is possible that patients attending the Royal Perth Hospital Chronic Pain Clinic feel well informed by the service, and as such, do not feel the need to search elsewhere for supplementary information. Only 13.0% of respondents were on their first visit to the pain clinic indicating a significant proportion had experienced the service previously. However, it has been shown that even patients who feel they have received adequate information from their clinician still tend to look elsewhere, most frequently the Internet, for further information [18]. A significant majority of our patients were unable to name their pain condition diagnosis or listed only a measure of severity of their condition as their diagnosis. An individual who is unclear as to what they are suffering from is not going to be able to make an effective search about their condition on the Internet. This may be a particular problem in the field of chronic pain where diagnoses can be less clear-cut than in other fields of medicine. It is vital that we are aware of this lack of clarity with our patients and are able to appropriately educate them as to their diagnosis. A better understanding of their condition will enable those who wish to seek further for information to be able to do so. It may have been that as a consequence of this lack of clarity, and also perhaps other factors peculiar to chronic pain patients, that as a group they are less likely to seek information online—this is an issue that warrants further study.

In our study population, the profile of the most likely patient to conduct an Internet search for pain-related information was young, highly educated, and had Internet access. This is in line with findings from previous work [4,20]. We did not however find a gender difference in information seeking behavior as seen in other studies [4,16]. It may be that our sample size was too small to elicit such a difference.

It has previously been established that health care professionals have little faith in the quality of the information available on the Internet [17,21,22]. While practitioners have tended in the past to fear the Internet-informed patient, adopting a paternalistic approach to information, there is evidence that this dynamic is subtly changing [23]. The impact that the Internet is having on the doctor-patient relationship is now becoming clear [1]. Patients do not see it as an alternative—they most frequently use it to confirm the professionals advice and to seek more information [24]. Although the retrieved information can be overwhelming, it is also believed to empower patients as participants in their care [25–27].

A total of 41.4% of respondents in our survey who had searched for chronic pain information said they believed the material they had found was useful. As these individuals represent the more highly educated among the sample community, this figure should perhaps cause some alarm, particularly as such a small number of respondents (6.9%) planned to discuss their findings with their doctor. This is in contrast to de Boer et al. who reported that 46% of their patients had discussed the information they found on the Internet with their doctor [16]. Cultural differences might be thought to explain this significantly higher level. However, looking again at the Australian data and the oncology patients surveyed by Newnham et al., 31% of patients had discussed the information found with their general practitioner and 43% with their treating specialist [18]. Why are so few of our patients approaching us or their own local doctor? Do we need to make ourselves more accessible? As depression was the second most common topic to be searched for by our survey patients, perhaps this is a factor that influences their unwillingness to engage us in dialog? Should we be proactive and instigate a discussion about Internet searches and information with our patients, a so-called Internet prescription [28]? Given the empowerment that results from Internet searching, this might be a particularly useful behavioral psychology tool to assist in the care of these patients.

The Internet is clearly an important tool with the potential to broaden information dissemination, improve health care delivery, and ultimately have a positive effect on patient outcome. We should not be complacent about the small proportion of people actively seeking information on the Internet, but continue our efforts to maximize the utility of this tool. We need to ensure that our patients are clear about what their diagnosis is so that they are empowered to help themselves. Furthermore, and as previously suggested by Gilliam et al. [29], we should include in our treatment plan to advice our patients on how and where to look for good quality information relevant to their condition. This would be a truly collaborative patient-centered approach to care of our chronic pain patients.


Embedded Image


View Abstract