The use of information technologies promises to improve the overall healthcare system in many ways 12. For example, electronic health records (EHRs) and personal digital assistants (PDAs) have been linked to improvements in overall quality in pediatric and adult populations34, a decrease in medical errors 5678, enhanced financial performance 910, improved physician satisfaction 11, and better clinical outcomes 1213. Other computer applications that can improve the healthcare experience for all parties include emailing between physicians and patients 1415 (or their family members 16) as well as Internet connectivity for online access to reference material and health information 1718.

The overall literature has been growing with reports regarding the use of information technology (IT) in the ambulatory setting. For example, studies of adult populations have demonstrated that EHR systems are available in an estimated 12.9 to 23%1819 of practices serving adults. Moreover, PDA usage among physicians at work has been growing. Currently, estimates of PDA use are 35% among pediatricians 13 with higher rates among younger doctors 20, residents 21 and medical students. However, despite increasing national attention to the computerization of healthcare in ambulatory settings, several large studies examining IT usage among physicians did not include, or did not separately analyze, pediatricians 181922. Moreover, no study has specifically examined IT utilization among those physicians whose practices see a significant proportion of the approximately 75 million children and adolescents in the U.S.

Along with the increased national attention to IT, a strategic plan has been developed by policymakers and healthcare leaders in the U.S. 23. The plan, in part, calls for the adoption of EHRs and will require the participation of all physicians, including those who routinely treat children. However, little is known about the current use of IT, especially EHR and PDAs among physicians who treat children. As such, the purpose of the current study is to examine IT use among child health providers (CHPs) in Florida. In the current study CHPs include general pediatricians, family physicians (who treat a significant number of children) and pediatric sub-specialists.

Given the gaps in the literature, we were interested to know if differences existed between these groups and whether or not their collective utilization differed from those physicians who routinely treat adults. Moreover, given the importance of specific EHR capabilities in promoting overall quality and patient safety for children, we examined differences in the availability of these key functions (e.g., clinical notes, medication lists, electronic order entry, etc.) among CHPs and other physician EHR users. Lastly, we also present data on the barriers to EHR use among these physicians. Given that no previous study has specifically examined a broad group of CHPs, the results of the current study may serve as a benchmark for those interested in charting the progress over time that pediatricians and other CHPs are making toward the widespread adoption of EHRs.

In the present study, we specifically focus on child health providers and compare them with all other physicians that participated in a large scale study designed to assess the level of IT use in Florida. Child health providers were defined as general pediatricians, pediatric specialists, and family physicians who self-reported a practice composition of at least 20% children less than 18 years of age.

During the spring of 2005 a survey (see additional file 1), along with a cover letter, was mailed to 14,921 physicians practicing in the ambulatory setting. This number included all allopathic and osteopathic primary care physicians and a 25% stratified random sample of specialists with a clear (e.g., not in training) and active Florida medical license. The office address for these physicians was obtained from the current list maintained by the State Department of Health (DOH) for licensure purposes. Because of the nature of the overall study, hospital-based physicians were excluded (e.g., radiologists, pathologists, anesthesiologists, and emergency physicians).

The survey was designed to capture information regarding general IT use and included more in-depth questions regarding EHR use in the office practice location. The survey also included questions about the barriers to EHR use. Those who indicated personally and routinely using EHR were further asked to select from a list those EHR functions that were available to them. This list of EHR functions was derived from the Institute of Medicine's list of standard and desirable EHR functions 2.

Surveys were tracked by a six-digit identification code. Four weeks after the initial mailing, non-respondents were mailed a second cover letter and survey reiterating our interest in their participation. Efforts were made to obtain updated information on addresses from surveys that were returned due to unknown or changed addresses. Those indicating that they were no longer seeing patients were excluded. The questionnaires were mailed back to, and processed by, the Florida State University Survey Research Laboratory where the data was entered into a computer database and subjected to data verification and cross-check methodologies. Additionally, the project received approval from the human subjects committees at the Florida State University and the University of South Florida.

In the context of significant increases in public and private sector attention and resources devoted to promoting the adoption of EHRs and other IT platforms, it is critical to understand where we stand currently in the uptake of these technologies among those serving children. The present study is, to our knowledge, the first of its kind to demonstrate on a statewide basis what many have reported anecdotally or in single site studies – namely that general pediatricians are significantly slower to incorporate EHRs into their office practice than other physicians. It also provides the first quantitative estimate of the proportion of physicians who have access to essential functionalities including those specific to pediatric patients 27 (e.g., growth charts and weight based dosing).

There are many factors that likely contribute to the slow rate of adoption among general pediatricians. Compared to family physicians, general pediatricians in our sample were significantly more likely to practice in solo or small practices (data not shown), which has been shown to be associated with slower adoption of EHRs 1819. Pediatric sub-specialists did not differ significantly from general pediatricians with respect to practice size configurations. However, after controlling for practice size, and other factors, pediatricians were still less likely to be EHR users than both family physicians and pediatric sub-specialists. A possible explanation may be related to the initial startup and ongoing maintenance costs of EHRs to physician practices 2829. Given that general pediatricians have the lowest median incomes of all physicians 30 and often rely heavily on Medicaid reimbursements which are well below those of Medicare, they may not be able to overcome many of the financial burdens necessary to adopt EHR. In our sample, general pediatricians rated financial barriers as a major barrier to EHR significantly more often than other CHPs and were significantly more likely to report that there practice was comprised of a higher percentage of Medicaid patients (data not shown). Nevertheless, while financial barriers are clearly sizable, other important barriers exist as well. For example, the lack of time needed to acquire and implement and EHR system, which was among the top barriers for all CHPs, is particularly problematic for general pediatricians who indicated this barrier most frequently. Overall, a better understanding of adoption barriers among child health providers is warranted. It should be noted, that our group has utilized the data representing all the physicians from the current study (not just CHPs), in a more in-depth analysis examining barriers to EHR systems by physicians. That study, which is forthcoming in a separate journal, examined barriers among physicians of differing adoption intentions 31. Similarly, pediatric and health quality researchers should focus on this issue of barriers to greatly benefit the national debate on EHR use.

Much of the patient safety and quality improvement benefits associated with EHR use are attributable to key functions available in many systems. These functions include, but are not limited to, clinical decision support, preventive service reminders, electronic order and prescription entry, allergy and medication lists, and electronic connection to other sources of clinical data (i.e., pharmacy or laboratory). EHR systems lacking some or more of these important functions may not provide the same overall patient safety and quality related benefits. In the present study, we find that general pediatricians are significantly less likely to routinely use an EHR system. Moreover, when an EHR system was present, general pediatricians and pediatric sub-specialists were significantly less likely than other doctors, including family physicians, to report the presence of key patient safety function such as electronic prescription and order entry, allergy lists, and connections to information from local pharmacies.

Given that EHRs and these key functionalities hold significant promise for improving the quality and safety of health care, the implication of our findings is that the quality and safety of care for children and adolescents may not improve as quickly as for other populations unless specific attention is paid to the needs of child health providers and child health care. It should be noted that many EHR vendors may not have designed their systems around pediatric needs. If so, present functions in EHR systems may be more difficult to use in pediatric patients.

The information in the present study may be limited in several ways. The generalizability of our research may be limited by the cross-sectional nature of this single state study. Additionally, we recognize that the potentially low response rate may be a limiting factor. However, after employing established methodologies to detect bias, we found no evidence of response bias in our sample 32. Nevertheless, the possibility always exists that physicians who participated in our study by completing the six-page questionnaire were more interested in IT than those who chose not to participate. If that were the case, our findings may be an overestimate of overall adoption rates. However, it is important to note that our response rate is comparable to published studies utilizing survey methodologies with physicians 2533. In addition, Florida may differ from other states in ways that could plausibly affect IT adoption. For example, the Governor established a health information technology advisory board in 2003, which has met around the state with key stakeholders, and may have raised the visibility and focus on IT over the last two years.

We hope our study will support action on the part of all those interested in promoting the adoption of IT to improve the quality and safety of care for children. In the last two years alone, a number of new groups have formed nationally to promote the adoption of EHRs. Child health representatives have been involved in some, but not all of these activities. In particular, we hope that these data will be used to support the following actions:

1. Advocacy for an increase in the proportion of federal resources dedicated to EHR adoption that is child specific. Despite continued efforts to the contrary, a decreasing proportion of the AHRQ budget includes a focus of children. Several bills are currently before Congress to address IT and these should include specific provisions to help child health providers adopt EHRs.

2. Increased representation by child health experts in national IT and IT standards organizations – thanks to the efforts of groups such as the American Academy of Pediatrics, the Child Health Corporation of America and the National Initiative for Children's Healthcare Quality, the voice of pediatrics is now being heard much more often at venues where IT is being discussed. This needs to continue and grow.

3. Increased attention by vendors to building pediatric specific functionalities into their products, such as weight based dosing and growth charting; and

4. Increased involvement by practicing child health providers in on-going efforts to reward the use of clinical IT in pay-for-performance programs and to design ways to overcome the non-financial barriers as well.

It is our hope that if this study were to be repeated in five years, we would find that we have closed the EHR-divide between pediatrics and all other specialties. This will not happen without concerted action on the part of child health providers and the organizations that represent them.

The author(s) declare that they have no competing interests.

NM, RGB, and LS developed the survey instrument. NM conducted the statistical analyses and wrote the first draft of the manuscript. NM, RGB and LS drafted subsequent versions of the manuscript. All author helped interpret the results of the analyses, and all authors read and approved the final manuscript.