Search tips
Search criteria 


Logo of jamiaAlertsAuthor InstructionsSubmitAboutJAMIA - The Journal of the American Medical Informatics Association
J Am Med Inform Assoc. 2012 Nov-Dec; 19(6): 960–964.
PMCID: PMC3534450

Physicians who use social media and other internet-based communication technologies


The demographic and practice-related characteristics of physicians who use social networking websites, portable devices to access the internet, email to communicate with patients, podcasts, widgets, RSS feeds, and blogging were investigated. Logistic regression was used to analyze a survey of US primary care physicians, pediatricians, obstetrician/gynecologists, and dermatologists (N=1750). Reported technology use during the last 6 months ranged from 80.6% using a portable device to access the internet to 12.9% writing a blog. The most consistent predictors of use were being male, being younger, and having teaching hospital privileges. Physician specialty, practice setting, years in practice, average number of patients treated per week, and number of physicians in practice were found to be inconsistently associated or unassociated with use of the technologies examined. Demographic characteristics, rather than practice-related characteristics, were more consistent predictors of physician use of seven internet-based communication technologies with varying levels of uptake.

Keywords: Internet, web 2.0, physicians, communication, health communication, cancer


Use of internet-based communication technologies among US physicians is widespread (table 1) and may improve care delivery. Physician use of personal digital assistants (PDAs) was associated with more rapid provider responses, fewer medication prescription errors, and improved data management and documentation.6 7 Email communication between physicians and patients was found to increase access to care and patients' perception of care quality.8 Similarly, industry research found that a third of physicians have changed a patient's treatment as a result of an internet search.9 Anecdotal accounts of enhanced patient care resulting from treatment innovations shared between physicians on social networking websites have also been published.10–13

Table 1
Selected internet-based communication technologies and usage rates among US adults and physicians

Little is known about the characteristics of physicians who use internet-based communication technologies. Systematic reviews7 14 found that physician use of PDAs was associated with being male, younger, and a medical resident, and working in a large or hospital-based practice. Use of email to communicate with patients was more common among physicians who practiced primary care (as compared with specialists), had graduated from a US or Canadian medical school, worked in the western USA, and worked in large group practices, academic settings, or hospital-based practices.8

We investigated the demographic and practice-related characteristics of US physicians who used seven internet-based communication technologies—social networking websites, portable devices to access the internet, email communication to patients, podcasts, widgets, RSS feeds, and blogging.


We analyzed data from the 2009 DocStyles survey which included 1750 physicians in five specialties—family medicine (n=609), internal medicine (n=391), pediatrics (n=250), obstetrics/gynecology (n=250), and dermatology (n=250). DocStyles is an annual web-based survey which investigates the attitudes and clinical practices of US physicians and allied health professionals. Analysis of 2009 DocStyles data was exempted from institutional review board approval, as personal identifiers were not included in the dataset provided to investigators.

Recruitment and sampling

The sample for the 2009 DocStyles survey was drawn from the Epocrates Honors Panel, which included 156 000 US physicians. When physicians enrolled in the Epocrates Honors Panel, their identities were verified by comparing first name, last name, date of birth, medical school, and graduation date against the American Medical Association's (AMA) Physician Masterfile, an inventory of licensed US physicians which includes both AMA members and non-members.

To ensure that adequate representation of targeted provider specialties was obtained, the 2009 DocStyles survey utilized quota sampling, a form of non-probability sampling in which the proportions of selected participant characteristics within the sample are deliberately set.15–17 The 2009 sampling quotas were set to reach 1000 primary care physicians (including family physicians and internists), 250 pediatricians, 250 obstetrician/gynecologists, and 250 dermatologists. In July 2009, email invitations to take part in the survey were sent to 2325 primary care physicians, 500 pediatricians, 500 obstetrician/gynecologists, and 562 dermatologists who were randomly selected to match the proportions for age, gender, and region of the AMA Physician Masterfile.

Physicians were eligible to take part in the survey if they practiced in the USA; treated at least 10 patients a week; worked in an individual, group, hospital, or clinic setting; and had practiced medicine for at least 3 years. Physicians who completed the survey were paid an honorarium of US$55–US$95 (depending on specialty).

Once a sampling quota was filled, the survey website locked out additional respondents in that specialty. As a result, response rates were capped at around 50% for each specialty. Analysis conducted by Porter Novelli (Washington, DC) found that participants were demographically comparable with physicians in the AMA Masterfile on gender, average age, and average years in practice.18


Participants’ use of various internet-based communication technologies was assessed by asking, ‘In the past 6 months, have you done any of the following?’ with seven technologies listed—(1) ‘used a portable device, such as an iPod, cell phone, personal digital assistant (PDA), smartphone, etc, to download information from the internet,’ (2) ‘blogged (wrote in an online diary),’ (3) ‘used a social network site, such as Sermo, Twitter, LinkedIn, UpToDate, or a similar site,’ (4) ‘used widgets (online applications built by one website that can be displayed on another website),’ (5) ‘used email to communicate with patients,’ (6) ‘downloaded podcasts (digital audio or video files),’ and (7) ‘subscribed to RSS (Really Simple Syndication) feeds of frequently read websites in order to receive notification of content updates.’ For each technology listed, respondents clicked either ‘yes’ or ‘no.’


Percentages of physicians using each of the technologies studied were calculated. Next, univariate logistic regression models were run with use of the various technologies (used during the last 6 months versus not used) as the dependent variables paired with each covariate of interest—specialty, gender, age, years in practice, teaching hospital privileges, practice setting, average number of patients treated per week, and number of physicians in practice. Covariates were categorized as shown in table 2.

Table 2
Participant characteristics and use of selected internet-based communication technologies (N=1750)

The covariates found to be significant (p≤0.05, indicating an OR which is significantly different from 1.0) in the univariate models were included in multivariate, forward-stepwise logistic regression models. The effect size interpretations suggested by Ferguson19 and Hopkins20 were used to gauge the practical significance of the observed ORs. All analyses were conducted using IBM SPSS Statistics V.19.0 software.


Participant characteristics

The majority of participants were male (67.9%) and worked in group practices (63.7%) (table 2). The distribution of participants across specialties reflected the sampling quotas.

Descriptive analysis

Reported technology use during the last 6 months ranged from 80.6% using a portable device to access the internet to 12.9% writing a blog (table 2). Use of the remaining technologies in descending order was social networking websites (59.1%), email to communicate with patients (49.0%), podcasts (41.4%), widgets (22.0%), and RSS feeds (19.1%).

Univariate and multivariate logistic regression models

In the univariate models, average number of patients treated per week was the only covariate tested that was not a statistically significant predictor (p<0.05) of use for one or more of the technologies modeled (results not shown). Thus, the multivariate models included age, gender, teaching hospital privileges, years in practice, specialty, practice setting, and number of physicians in practice. Of the covariates included in the multivariate models, number of physicians in practice was the only covariate which did not remain a significant predictor (p<0.05) of use for one or more of the technologies modeled (table 3). Evaluation of the practical significance19 20 of predictive variables found no covariates were associated with a large effect size (OR ≥4.0) and age was the only covariate associated with a moderate effect size (OR 3.0–3.9)—physicians who were younger than 35 years old had higher odds of using social networking websites compared with physicians who were aged 55 years or older.

Table 3
Physician characteristics significantly associated with use of selected internet-based communication technologies


Across the technologies modeled, the most consistent predictors of use were being male (associated with use of six technologies), younger age (associated with use of three technologies), and having privileges at a teaching hospital (associated with use of three technologies). The consistency of predictors, particularly in the case of gender, is interesting, given that use ranged from near saturation (80.6% reported having accessed the internet through a portable device) to fledging (12.9% reported having written a blog).

A variety of practice-related covariates were found to be inconsistently associated or unassociated with reported technology use. Fewer years in practice was only associated with writing blogs. The most predictive categories of physician specialty and practice setting varied by the technology modeled. The number of physicians in practice and the average number of patients treated per week were not significantly associated with use of any of the technologies examined in the adjusted models.

Our findings are somewhat consistent with prior peer-reviewed studies which are limited to physician characteristics associated with use of PDAs and email to communicate with patients. In previous studies, male physicians were found to be more likely to use PDAs,14 and in the current study, males had higher odds of reporting that they accessed the internet through portable devices (the closest comparable measure). Earlier studies of physician–patient email communication8 also included some of the covariates tested in this study and found that physicians who specialized in family practice and internal medicine, had privileges at a teaching hospital, and practiced in a larger group setting were more likely to communicate with patients via email. Of these covariates, having privileges at a teaching hospital was the only one in the current study to be significantly associated with use of email to communicate with patients.

In the general adult population, gender and age are also associated with use of internet-based communication technologies. Consistent with our results, men more often reported owning smartphones than women,1 and more adult male internet users reported downloading podcasts than females.21 However, age trends in the general population did not always match our results. Age has been found to be inversely associated with ownership of smartphones,1 use of social networking websites,5 downloading podcasts,5 and blogging.5 In our study, younger age was significantly associated with only two of these technologies, using portable devices to access the internet (which is comparable with smartphone ownership) and social networking websites. While the odds of using the technologies examined consistently decreased as age increased in the general population, this was not always the case among physicians in our study. For instance, physicians aged 35–44 years old had higher odds of reporting use of portable devices to access the internet and widgets than those younger than 35 years of age. These differences are not surprising given that physician use of internet-based communication technologies typically far exceeds mainstream use (table 1). In addition, physicians systematically differ from the general adult population: they have higher educational attainment and incomes; they are more likely to be male; and 26 years is typically the minimum age, given the training requirements.

In order to guide the development of communication initiatives targeting physicians, we evaluated the practical significance of predictive variables. It should be emphasized that the threshold for practical significance (in this case OR ≥3.0) is typically set much higher than statistical significance.19 20 With the exception of physicians under 35 years of age having moderately higher odds of using social networking websites, technology use among physicians in our study was uniform for practical purposes. Thus, communication initiatives involving internet-based technologies have the potential to reach a broad range of physicians, and messages should be designed accordingly.

The present study has several limitations. First, the use of quota sampling diminishes the generalizability of our findings, and the usage rates reported here should not be interpreted as national estimates of technology use. Next, aggregate use of technology was of particular interest to the Centers for Disease Control and Prevention's (CDC) Inside Knowledge: Get the Facts About Gynecologic Cancer campaign22 which conducted this study primarily to inform development of communication strategies to reach healthcare providers. Thus, the technology use studied was not limited to a professional context. Also, the survey items were brief and did not capture all uses associated with the listed examples. For instance, UpToDate is more commonly used for information retrieval than social networking. Finally, the study measures did not allow for the differentiation of frequent and infrequent users, as participants were only asked if they had used the listed technologies during the last 6 months.

Our results may guide the development of provider-targeted communication initiatives as well as investigations of the clinical implications of physician use of the technologies studied. However, physician use of some internet-based communication technologies has expanded rapidly,2 23 and the characteristics of users may change as uptake increases and technologies evolve. Despite the often dynamic nature of physician technology adoption, we found several consistent predictors of use across seven technologies with varying levels of uptake. The extent to which this consistency translates into longevity for our results will require further observation over time.


Contributed by

Contributors: All authors contributed sufficiently to this report to be included as authors, and all those who are qualified to be authors are listed as authors.

Funding: The Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Cancer Prevention and Control provided funding for this study.

Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

Competing interests: None.

Ethics approval: Analysis of data used in this study was exempted from institutional review board approval, as personal identifiers were not included in the dataset provided to investigators.

Provenance and peer review: Not commissioned; externally peer reviewed.


1. Smith A. 35% of American adults own a smartphone. Pew Research Center 2011. (accessed 13 Feb 2012).
2. Manhattan Research Taking the Pulse v10.0. Mobihealth news, 2010. (accessed 5 Oct 2011).
3. Madden M, Zickuhr K. 65% of online adults use social networking sites. Pew Research Center 2011. (accessed 14 Feb 2012).
4. Modahl M, Tompsett L, Moorhead T. Doctors, Patients & Social Media. QuantiaMD, 2011. (accessed 13 Feb 2012).
5. Zichuhr K. Generations 2010. Pew Research Center. 2010. (accessed 14 Feb 2012).
6. Progmet M, Georgiou A, Westbrook JI. The impact of mobile handheld technology on hospital physicians' work practices and patient care: a systematic review. J Am Med Inform Assoc 2009;16:792–801 [PMC free article] [PubMed]
7. Lindquist AM, Johansson PE, Petersson GI, et al. The use of the personal digital assistant (PDA) among personnel and students in health care: a review. J Med Internet Res 2008;10:e31 (accessed 6 Oct 2011). [PMC free article] [PubMed]
8. Ye J, Rust G, Fry-Johnson Y, et al. E-mail in patient-provider communication: a systematic review. Patient Educ Couns 2010;80:266–73 [PubMed]
9. Google/Hall & Partners Connecting With Physicians Online: Searching For Answers. 2009. (accessed 1 Dec 2011).
10. Kim J. 85% of physicians polled believe patient care improved through use of Sermo & social media. 2010., (accessed 5 Dec 2011).
11. Rethlefsen M, Segovis C. Medicine in the Era of Web 2.0. Minnesota Medicine, 2009. (accessed 5 Dec 2011).
12. McGuire S. The cutting edge. Med Market Media 2008:55–6 (accessed 15 May 2012).
13. Vascellaro JE. Social networking goes professional. Wall Street J 2007:D1 (accessed 5 Oct 2011).
14. Garritty C, El Emam K. Who's using PDAs? Estimates of PDA use by health care providers: a systematic review of surveys. J Med Internet Res 2006;8:e7 (accessed 6 Oct 2011). [PMC free article] [PubMed]
15. Im EO, Chee W. Quota sampling in internet research: practical issues. Comput Inform Nurs 2011;29:381–5 [PubMed]
16. Morrow KM, Vargas S, Rosen RK, et al. The utility of non-proportional quota sampling for recruiting at-risk women for microbicide research. AIDS Behav 2007;11:586–95 [PMC free article] [PubMed]
17. Cumming RG. Is probability sampling always better? A comparison of results from a quota and a probability sample survey. Community Health Stud 1990;14:132–7 [PubMed]
18. Porter Novelli DocStyles 2009 Methods. Washington D.C: Porter Novelli, 2009
19. Ferguson CJ. An effect size primer: a guide for clinicians and researchers professional psychology: research and practice. Professional Psychol: Res Pract 2009;40:532–8
20. Hopkins WG. A New View Of Statistics. 2011. (accessed 5 Oct 2011).
21. Madden M, Jones S. Podcast downloading 2008. Pew research Center. 2008. (accessed 15 Feb 2012).
22. Rim SH, Polonec L, Stewart SL, et al. A national initiative for women and healthcare providers: CDC's inside knowledge: get the facts about gynecologic cancer campaign. J Womens Health (Larchmt) 2011;20:1579–85 [PubMed]

Articles from Journal of the American Medical Informatics Association : JAMIA are provided here courtesy of American Medical Informatics Association