- Research article
- Open Access
- Open Peer Review
Identifying outcome-based indicators and developing a curriculum for a continuing medical education programme on rational prescribing using a modified Delphi process
© Esmaily et al; licensee BioMed Central Ltd. 2008
- Received: 11 December 2007
- Accepted: 30 May 2008
- Published: 30 May 2008
Continuing medical education (CME) is compulsory for physicians in Iran. Recent studies in Iran show that modifications of CME elements are necessary to improve the effectiveness of the educational programmes. Other studies point to an inappropriate, even irrational drug prescribing. Based on a needs assessment study regarding CME for general physicians in the East Azerbaijan province in Iran, rational prescribing practice was recognized as a high priority issue. Considering different educational methods, outcome-based education has been proposed as a suitable approach for CME. The purpose of the study was to obtain experts' consensus about appropriate educational outcomes of rational prescribing for general physicians in CME and developing curricular contents for this education.
The study consisted of two phases: The first phase was conducted using a two-round Delphi consensus process to identify the outcome-based educational indicators regarding rational prescribing for general physicians in primary care (GPs). In the second phase the agreed indicators were submitted to panels of experts for assessment and determination of content for a CME program in the field.
Twenty one learning outcomes were identified through a modified Delphi process. The indicators were used by the panels of experts and six educational topics were determined for the CME programme and the curricular content of each was defined. The topics were 1) Principles of prescription writing, 2) Adverse drug reactions, 3) Drug interactions, 4) Injections, 5) Antibiotic therapy, and 6) Anti-inflammatory agents therapy. One of the topics was not directly related to any outcome, raising a question about the need for a discussion on constructive alignment.
Consensus on learning outcomes was achieved and an educational guideline was designed. Before suggesting widespread use in the country the educational package should be tested in the CME context.
- Adverse Drug Reaction
- Continue Medical Education
- Rational Prescribe
- Panel Discussion
- Curricular Content
The need for continuing medical education (CME) as a part of a medical doctor's professional development has been recognized all over the world [1, 2]. Since 1991, CME has been compulsory for all Iranian physicians and other health professionals.
Traditionally, general physicians working in primary care as general practitioners (here referred to as GPs) have obtained their CME through lectures and written materials. These methods may increase their knowledge, but there is no clear evidence that their skills or performance are impacted [3–9].
Although outcome-based education (OBE) initially was proposed for undergraduate medical education and training , it is now increasingly recognized as a useful approach for CME . OBE can influence the entire process of education, i.e. decisions about the content, formulation of aims, educational strategies, the teaching methods, the assessment procedures and the educational environment .
Several advantages have been suggested for adopting an OBE model for medical education, such as helping to increase the relevance of the education to the future practice of medicine. OBE allows for a wide degree of participation in the curriculum design process as well as flexibility regarding the choice of educational strategies used. OBE appears to be acceptable to most teachers probably because the concepts of OBE are clear, easily understandable, and provide a robust framework for the curriculum [10, 12].
Recent studies of CME for physicians in Iran show that modifications of CME elements are necessary to improve the effectiveness of the educational programs [13, 14]. There are also convincing data showing excessive prescribing behaviour in Iran among GPs . Some studies point to an inappropriate, even irrational drug prescribing [16, 17]. According to an unpublished needs assessment study about CME for GPs in the East Azerbaijan province in Iran (2002), rational prescribing practice was recognized as a high priority issue.
Given the timeliness and importance of this issue in Iran, we are conducting a project aimed at evaluating the effectiveness of using OBE on GPs knowledge, skills, attitudes, and performance in a CME program on rational prescribing. No studies on OBE indicators for rational prescribing were found in the literature. The aim of this study was, therefore, to identify the appropriate outcomes and indicators for rational prescribing based on expert opinions and developing the curricular content in a CME program for GPs.
The study was conducted between October 2005 and July 2006 in three phases. The first phase consisted of two rounds of a Delphi process. In the second phase, panel discussion sessions were used to further clarify, reach consensus, and then prioritize the learning outcomes and indicators. Finally, an appointed team determined content for the CME programme.
The Delphi technique was originally conceived of as a way to obtain the opinion of experts without necessarily bringing them together face to face . This is what attracted us to the method. For our study, we used a modified version of the technique which also involved panel discussions with experts in the field of rational prescribing and CME. In the first two phases, the methodology was similar to what has been described previously [12, 19].
A questionnaire was designed consisting of potential learning outcomes identified from a range of sources: WHO (World Health Organization) prescribing indicators [20, 21], relevant documents [16, 17, 22, 23], and topics covered by CME programs on prescribing in Iran. A small group of experts (a pharmacologist, a pharmacist, and two medical specialists) assisted the research team in defining a preliminary list of 16 potential outcomes with appropriate measurement indicators for each one.
We selected 30 stakeholders who had a background from at least one of the following categories (some represented several categories): 1) seven experienced GPs, 2) four CME decision makers from Iranian medical science universities with a background in pharmacy, pharmacology, or health management and who work for the Ministry of Health and Medical Education (MOHME), 3) nine pharmacists, 4) seven pharmacologists, 5) six medical specialists. With the exception of the experienced GPs, all the participants were faculty members of four of the Universities of Medical Sciences in Iran as well as being CME trainers.
Delphi process and panel discussion
The questionnaire, with the 16 potential outcomes including a definition of each outcome (to avoid any misunderstanding), was sent out to the participants. The respondents were asked if there was a need to include the outcome in a programme on rational prescribing. They answered using a 5-point Likert scale (1 = totally agree, 2 = partly agree, 3 = uncertain, 4 = partly disagree and 5 = totally disagree). They were also asked to suggest any other potential learning outcomes, based on their expertise in the area.
The results of the returned questionnaires were assessed by the research team and a second questionnaire was developed for the next stage of the Delphi process. It consisted of the initial potential learning outcomes from round 1 and ten new learning outcomes that had been suggested by the experts. In this round, each member of the study was asked to determine which of the proposed outcomes should be included by using a 4-point scale (1 = totally agree, 2 = partly agree, 3 = partly disagree and 4 = totally disagree) to avoid uncertain answers.
The questionnaire responses were coded and frequencies were determined using Microsoft Excel 2003 . The results of the second stage of the Delphi process were submitted to a specially assigned panel of experts comprised of four CME decision-makers, four pharmacists or pharmacologists, four medical specialists, three GPs, one epidemiologist, and the first author. The task of the panel was to reach consensus on the final educational outcomes and indicators and to identify a suitable team of experts to develop the content.
Developing the content
The team responsible for content development was comprised of seven experienced CME trainers, all who had participated in the panel discussion, and the first author. They identified educational topics for an outcome-based CME program in rational prescribing informed by the list of outcomes and outcome indicators that were developed from the second Delphi round. To each of these topics a small group of experts and CME trainers was assigned, with the task of developing the curricular content for that topic.
The results from each of the small groups were distributed to all the other teams one week prior to a final review meeting. At this meeting, ideas about the curricular content were discussed and consensus was reached after a revision of some of the topics. The final curricular content formed the basis for the OBE educational program that was to be used in the CME program on Rational Prescribing.
Ethical approval for the study was received from the Ethical Vetting Committee of the Iranian Ministry of Health and Medical Education. Informed consent was given through agreeing to participate in the different phases of the study.
The Delphi process
Completed questionnaires were returned from 21 of the 30 participants in the first round. Nine of the participants (four medical specialists, two pharmacologists and three pharmacists) did not answer, despite the reminder we sent out one month later. These non-responders were contacted by e-mail and/or telephone; six replied that they were too busy to answer and three did not wish to participate as they saw no personal benefit.
Potential learning outcomes as assessed through the Delphi process
Potential outcomes for Rational Prescribing (round 1)
Agreement* (n = 21)
Agreement* (n = 20)
Upon completion of the course, a doctor will be able to write a prescription that includes or considers:
1. Date of the prescription
2. Name and age of the patient
3. Name and identification number of Iranian Medical Council of the prescriber**
4. Main complaint of the patient
5. Legible hand writing
6. Generic name of the drugs
7. Administration form of drugs
8. Strength of the drugs and dose frequency
9. Adequate duration of treatment
10. Latin abbreviation of terminology in drug use order
11. Appropriate number of drugs
12. To consider homogeneity of prescription per individual (all drugs prescribed pertain to the same individual)
13. Not prescribing drugs with the same pharmacological effect
14. Not prescribing drugs which have negative interactions with each other
15. Appropriate number and amount of antibiotics
16. Appropriate number of injections
Suggested potential outcomes (added in round 2)
1. Contact telephone number of the prescriber
2. Refill information
3. Initial diagnosis
4. Time and manner of drug use
5. Necessary precautions
6. Necessary notifications about signals to continue or stop drug use
7. Notification about side effects of drugs in the prescription
8. Name of the foods which have negative interactions on drug efficacy and the treatment process
9. Appropriate no. of NSAID (Non-steroidal anti-inflammatory) drugs
10. Appropriate number of vitamins
The panel discussion
Rational prescribing outcomes and indicators for GP's CME programmes as agreed by an expert panel
Upon completion of the course, a participant should be able to:
1. Write the date of the prescription
1. Date of the prescription
2. Write the name of the patient
2. Name of the patient
3. Write the age of the patient
3. Age of the patient
4. Write the name and identification number of the Iranian Medical Council for the prescriber
4. Name and identification number of Iranian Medical Council of the prescriber
5. Write the contact telephone number of prescriber
5. Contact telephone number of prescriber
6. Write refill information
6. Refill information
7. Write the prescription clearly
7. Legible hand writing
8. Write the generic name of the drugs
8. Generic names of drugs
9. Write the administration form of the drugs
9. Administration form of drugs
10. Write the strength of the drugs (dose and dose frequency)
10. Strength of the drugs
11. Write the duration of treatment
11. Treatment duration
12. Use the Latin abbreviation of terminology in drug use order
12. Use Latin abbreviation terminology
13. Write the time and manner of drug use
13. Time and manner of drug use
14. Write the necessary precautions
14. Necessary precautions
15. Prescribe the appropriate number of drugs
15. Appropriate number of drugs
16. Consider homogeneity of prescription per individual (all drugs prescribed pertain to the same individual)
16. Homogeneity of prescriptions
17. Avoid prescribing drugs with the same pharmacological effect
17. Number of drugs in the same pharmacological group
18. Avoid prescribing drugs which have negative interactions with each other
18. Number of interactive drugs per prescription
19. Prescribe the appropriate number and amount of antibiotics
19. Number and amount of antibiotics per prescription and proportion of antibiotics prescribed
20. Prescribe the appropriate number of injections
20. Number of injections per prescription and proportion of injections prescribed
21. Prescribe the appropriate number of Anti-Inflammatory Agents [Corticosteroids and Non-steroidal anti-inflammatory drugs (NSAID)]
21. Appropriate number of Anti-Inflammatory Agents [Corticosteroids and Non-steroidal anti-inflammatory drugs (NSAID)]
In the third phase, the team responsible for content development divided the 21 learning outcomes into six main areas for the educational programme: 1) Principles of prescription writing, 2) Adverse drug reactions (ADR), 3) Drug interactions, 4) Injections, 5) Antibiotic therapy, and 6) Anti-inflammatory agents therapy
Curricular content for the programme on rational prescribing as developed by teams of CME trainers.
Topics for the rational prescribing curriculum
1-Principles of prescription writing (1–16)*
5-Antibiotic therapy (19)
History of prescription writing
Value of taking culture samples for infections
Classification of drugs
Assessment of infectious organisms
Definition and format
Importance of host factors in selection of
Elements of prescription writing
Adherence to correct indications
Mistakes and errors in prescription writing
Selection of antibacterial drug(s)
Important factors for choosing form, dose and
course of antibiotics
Importance of switching antibiotics based on
2-Adverse reactions to drugs
culture and antibiogram results
Pharmacology of antibiotic groups:
Exaggeration of an intended pharmacologic
action of the drugs
Toxicity unrelated to a drug's primary
- Cytotoxic Reactions
6- Anti-inflammatory agents therapy (21)
- Immunologic Mechanisms
Diagnosis and treatment of adverse drug
Emphasis on reducing injections
3- Drug Interactions (17, 18)
Important mechanisms of drug interactions
Common drug interactions in the practice of
B) Non steroidal anti inflammatory drugs
4- Injections (20)
Consideration of real needs for prescribing
Mechanism of injections
Indications for injections
Important factors in prescribing injections
Prevalence of prescribing injections in the
world and in Iran
The Delphi technique was chosen in order to distil the opinions of several experts and reach consensus without geographical constraints  and with heterogenic groups . A panel of selected experts has been used in other Delphi studies to identify prescribing errors in order to increase the validity of the findings [27, 28]. We used a modified version with panel discussions and small working groups in order to better validate the results, minimize ambiguity, and further develop program content. The process yielded twenty-one outcomes and their related indicators. These included all but one of the initially proposed indicators and half of the additionally suggested ones (the indicator for name and age was divided into two separate ones to make it easier to assess). Based on discussions about the outcomes, six topic areas were identified and curricular content for each was determined by an expert team based on these outcomes. The content was then validated by all the participating experts.
We began the Delphi process by creating a preliminary list of potential learning outcomes through a literature review and then made use of experts' opinions to develop and prioritize these outcomes. These experts were involved in CME either as decision makers or trainers. GPs representing the target audience were also included as a way of improving the relevancy of the programme as has been described previously . In these ways, we aimed to bridge the gap between the actual learning needs of individual practitioners and the educational content that is considered to meet assumed needs, a problem that has been described in the educational literature [29, 30].
Two of the learning outcomes which were suggested by the experts during the Delphi process were excluded during the panel discussions. Due to a concern that the CME programme as a whole would be rejected if individual parts were considered to be too controversial, the outcome "writing main complaints of the patient in the prescription" was rejected. The panellists thought that it would be too great a departure from the current work practice of physicians. Even though the over prescribing of vitamins is thought to be a problem in Iran, the outcome, "appropriate number of vitamins", was rejected since it would be difficult to assess as most vitamins are not reimbursed by insurance organizations, the registries of which provide the best way to assess physician prescribing behaviour in Iran.
The panel added one outcome related to the prescribing of corticosteroids as it had been reported that corticosteroid drugs had been prescribed more than any other drug group in East Azerbaijan Province during the last year (yearly report of Rational Prescribing Committee, Tabriz University of Medical Sciences, Tabriz, Iran; 2005). The panel also emphasized the importance of the learning outcomes related to instructions on how to administer medications as there is evidence that there is a large number of prescriptions that do not include directions for patient use . These were two examples, where outcomes from the original Delphi process were further modified or built upon with the modified Delphi process.
A comparison of the outcomes with the curricular content of the course that was developed reveals that while all of the outcomes are covered in the course, the entire course is not covered by the outcomes as the topic "adverse reactions to drugs" is not directly related to an outcome. ADR can be the result of irrational prescribing. Thus, knowledge about the consequences of irrational prescribing can motivate a change in participants' behaviour. As such, ADR can be considered integral to a course on rational prescribing. However, if the course was supposed to be designed using an OBE approach, why wasn't ADR identified as an outcome? Possible explanations are on the one hand that tradition prevailed and on the other that the specific construction of the task for the Delphi rounds had a decisive influence.
Regarding the first explanation, a review of the topics that were included in the curricula shows that these topics were taught in other CME courses on prescribing. Since the team of experts responsible for content development all had taught according to the earlier curricula, tradition might have prevailed despite receiving the finalized list of outcomes and an explanation about how they were derived.
The other more distinct and plausible explanation for why the initial Delphi process did not elicit a topic that content developers deemed essential could be due to the formulation of the instructions for the Delphi process. The task was to identify what should be included or considered when the doctor writes a prescription. We believe this to be the main reason for not suggesting ADR as an outcome indicator as information on ADR cannot be written directly in the prescription. However, since ADR has to be considered before a particular drug is chosen, content developers included this topic in the curriculum. While it would have been possible to develop direct outcomes and indicators for ADR, ADR can be seen to indirectly be part of and important in achieving some of the other outcomes, such as avoiding prescribing drugs with negative interactions or similar effects or prescribing injections too frequently.
A final observation was that the constructive alignment approach , where outcomes are tied together with content and assessment methods, was not expressly discussed during the third phase of content development. This could mean that content developers were simply unaware of the need to explicitly link outcomes to course content and could thus benefit from more knowledge about outcome-based education and constructive alignment prior to designing courses.
Another question concerns the quality of the outcomes and the indicators that were formulated. Outcomes can be analyzed based on how specific they are, if they are measurable, if they cover the domains of knowledge, skills, and attitudes, and if they are of a manageable number . Most of the outcomes meet these criteria. However, the word "appropriate" (found in outcomes 15 and 19–21) is harder to measure, although it would be possible to measure if linked to clear guidelines. According to the ideas behind outcome-based education and constructive alignment, the validated indicators could form the basis for assessment of GPs' performance and provide them with feedback on the effectiveness of their learning. This could be pursued in further research.
The modified Delphi process was effective in developing outcomes and indicators that were then used in the development of content for a course in rational prescribing. The indicators can be used to assess what participants have learned during the course as well as prescribing behaviour after the course. To improve the constructive alignment of the course, course developers should understand that outcomes need to be explicitly related to course curricula.
We would like to thank all members of the panels and the participating experts in this study, Professor Ivan Silver and Associate Professor Eskandar Fathi Azar for their valuable comments, the Educational Development Center of Tabriz University of Medical Sciences for logistical and financial support, the National Public Health Management Center in Tabriz for financial support and Dr. Hassan Aghajani Deputy of Pharmacy, Tabriz University of Medical Sciences for help in providing useful information. The study was also financially supported by Ministry of Health and Medical Education (MOHME) of Iran This funding source had no influence in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.
- Holm HA: Quality issues in continuing medical education. BMJ (Clinical research ed. 1998, 316 (7131): 621-624.View ArticleGoogle Scholar
- Winzenberg T, Higginbotham N: Factors affecting the intention of providers to deliver more effective continuing medical education to general practitioners: a pilot study. BMC medical education. 2003, 3: 11-10.1186/1472-6920-3-11.View ArticleGoogle Scholar
- Davis DA, Thomson MA, Oxman AD, Haynes RB: Changing physician performance. A systematic review of the effect of continuing medical education strategies. JAMA. 1995, 274 (9): 700-705. 10.1001/jama.274.9.700.View ArticleGoogle Scholar
- Davis D, O'Brien MA, Freemantle N, Wolf FM, Mazmanian P, Taylor-Vaisey A: Impact of formal continuing medical education: do conferences, workshops, rounds, and other traditional continuing education activities change physician behavior or health care outcomes?. JAMA. 1999, 282 (9): 867-874. 10.1001/jama.282.9.867.View ArticleGoogle Scholar
- Davis DA, Thomson MA, Oxman AD, Haynes RB: Evidence for the effectiveness of CME. A review of 50 randomized controlled trials. JAMA. 1992, 268 (9): 1111-1117. 10.1001/jama.268.9.1111.View ArticleGoogle Scholar
- Thomson O'Brien MA, Freemantle N, Oxman AD, Wolf F, Davis DA, Herrin J: Continuing education meetings and workshops: effects on professional practice and health care outcomes. Cochrane database of systematic reviews (Online). 2001, CD003030.Google Scholar
- Pohlmann G: Continuing medical education in Wisconsin: current status and future directions. WMJ. 2007, 106 (1): 20-26.Google Scholar
- Grol R: Changing physicians' competence and performance: finding the balance between the individual and the organization. The Journal of continuing education in the health professions. 2002, 22 (4): 244-251. 10.1002/chp.1340220409.View ArticleGoogle Scholar
- Gercenshtein L, Fogelman Y, Yaphe J: Increasing the satisfaction of general practitioners with continuing medical education programs: a method for quality improvement through increasing teacher-learner interaction. BMC family practice. 2002, 3: 15-10.1186/1471-2296-3-15.View ArticleGoogle Scholar
- Harden RM: AMEE Guide No. 14: Outcome-based education: Part 1-An introduction to outcome-based education. Medical teacher. 1999, 21 (1): 7-14. 10.1080/01421599979969.View ArticleGoogle Scholar
- Harden RM: Developments in outcome-based education. Medical teacher. 2002, 24 (2): 117-120. 10.1080/01421590220120669.View ArticleGoogle Scholar
- Paterson Davenport LA, Hesketh EA, Macpherson SG, Harden RM: Exit learning outcomes for the PRHO year: an evidence base for informed decisions. Medical education. 2004, 38 (1): 67-80. 10.1111/j.1365-2923.2004.01736.x.View ArticleGoogle Scholar
- Amini A, Kargarmaher MH, Hatami Sadabadi F, Salami H: Determining CME needs of general physicians in East Azerbijan province. Iranian Journal of Medical Education. 2002, 1 (8): 12-13.Google Scholar
- Shirazi M: Assessing the gap between current and desirable needs in TUMS CME Unit: participants' viewpoints . Iranian Journal of Medical Education. 2004, 5 (1): 17-22.Google Scholar
- World Health Organization: Essential Drugs Monitor. 2003, Geneva, SwitzerlandGoogle Scholar
- Moghadamnia AA, Mirbolooki MR, Aghili MB: General practitioner prescribing patterns in Babol city, Islamic Republic of Iran. Eastern Mediterranean health journal = La revue de sante de la Mediterranee orientale = al-Majallah al-sihhiyah li-sharq al-mutawassit. 2002, 8 (4-5): 550-555.Google Scholar
- Cheraghali AM, Nikfar S, Behmanesh Y, Rahimi V, Habibipour F, Tirdad R, Asadi A, Bahrami A: Evaluation of availability, accessibility and prescribing pattern of medicines in the Islamic Republic of Iran. Eastern Mediterranean health journal = La revue de sante de la Mediterranee orientale = al-Majallah al-sihhiyah li-sharq al-mutawassit. 2004, 10 (3): 406-415.Google Scholar
- Varkevisser CM: Designing and Conducting Health System Research Projects. Health System Research Training Series. 1991, Health Sciences Division of the International Development Research Centre,Ottawa,Canada, 2 .Google Scholar
- Stewart J, O'Halloran C, Harrigan P, Spencer JA, Barton JR, Singleton SJ: Identifying appropriate tasks for the preregistration year: modified Delphi technique. BMJ (Clinical research ed. 1999, 319 (7204): 224-229.View ArticleGoogle Scholar
- World Health Organization: How to investigate drug use in health facilities, Selected drug use indicators. Edited by: . 1993, Geneva, SwitzerlandGoogle Scholar
- World Health Organization: Guide to Good Prescribing, A practical manual. 1994, Geneva, SwitzerlandGoogle Scholar
- Katzung BG: Basic and clinical Pharmacology. 2004, MC-Graw Hill, 2: Chapter 66-10Google Scholar
- Gholamreza S, Meimandi MS: The quality of prescribing in general practice in Kerman, Iran. International journal of health care quality assurance incorporating Leadership in health services. 2005, 18 (4-5): 353-360.View ArticleGoogle Scholar
- Redmond W: Excel 2003 . 2003, Microsoft Corporation 2002Google Scholar
- Jones J, Hunter D: Consensus methods for medical and health services research. BMJ (Clinical research ed. 1995, 311 (7001): 376-380.View ArticleGoogle Scholar
- Walley T, Webb DJ: Developing a core curriculum in clinical pharmacology and therapeutics: a Delphi study. British journal of clinical pharmacology. 1997, 44 (2): 167-170. 10.1046/j.1365-2125.1997.00669.x.View ArticleGoogle Scholar
- Dean B, Barber N, Schachter M: What is a prescribing error?. Qual Health Care. 2000, 9 (4): 232-237. 10.1136/qhc.9.4.232.View ArticleGoogle Scholar
- Ghaleb MA, Barber N, Dean Franklin B, Wong IC: What constitutes a prescribing error in paediatrics?. Quality & safety in health care. 2005, 14 (5): 352-357. 10.1136/qshc.2005.013797.View ArticleGoogle Scholar
- Norman GR, Shannon SI, Marrin ML: The need for needs assessment in continuing medical education. BMJ (Clinical research ed. 2004, 328 (7446): 999-1001. 10.1136/bmj.328.7446.999.View ArticleGoogle Scholar
- Grant J: Learning needs assessment: assessing the need. BMJ (Clinical research ed. 2002, 324 (7330): 156-159. 10.1136/bmj.324.7330.156.View ArticleGoogle Scholar
- Biggs JB: Teaching for quality learning at university: what the student does. 2003, Phildelphia, Pa., Society for Research into Higher Education : Open University Press, 2ndGoogle Scholar
- Kern DE: Curriculum development for medical education: a six step approach. 1998, Baltimore, Johns Hopkins University Press, 178.Google Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1472-6920/8/33/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.