Figures
Abstract
Background
Clinical guidelines (CGs) are popular for healthcare decision making but their acceptability and use by healthcare providers is influenced by numerous factors. Some of these factors are professional-related, such as knowledge and perceptions of and attitudes toward CGs in general. The aim of our study was to evaluate attitudes and perceptions of Spanish physicians towards CGs.
Methods
We coordinated six discussion groups with a total of 46 physicians. The participants were drawn from 12 medical specialties from both specialized and primary care. We recorded the sessions and transcribed the content verbatim. We analyzed the data using an approach based on the grounded theory.
Results
We identified two main constructs that defined the physicians' perceptions towards guidelines: knowledge and usefulness. “Knowledge” defined the theoretical meanings of guidelines, while “Usefulness” referred to the pragmatic approach to guidelines. These constructs were interrelated through a series of categories such as confidence, usability, accessibility, dissemination and formats.
Conclusions
In our study, the constructs that impacted most on physician's attitudes to clinical guidelines were knowledge and usefulness. The tension between the theoretical and the pragmatic constructs determined the attitudes and how physicians use guidelines. Groups developing guidelines should ask relevant clinical questions and develop implementable and context specific recommendations. Developers should be explicit and consistent in the development and presentation of recommendations.
Citation: Solà I, Carrasco JM, Díaz del Campo P, Gracia J, Orrego C, Martínez F, et al. (2014) Attitudes and Perceptions about Clinical Guidelines: A Qualitative Study with Spanish Physicians. PLoS ONE 9(2): e86065. https://doi.org/10.1371/journal.pone.0086065
Editor: Laxmaiah Manchikanti, University of Louisville, United States of America
Received: September 20, 2013; Accepted: December 9, 2013; Published: February 5, 2014
Copyright: © 2014 Solà et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This project has been funded by the Instituto de Salud Carlos III, co-financed by the European Regional Development Fund (PI08/90647). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The other authors have declared that no competing interests exist. Pablo Alonso-Coello and Ivan Solà are members of the GRADE working group. This does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials.
Background
Evidence-based medicine (EBM) has driven the use of tools for the critical synthesis of the literature, including systematic reviews and, more significantly, clinical guidelines (CGs). The former were recently defined as “statements that include recommendations intended to optimize patient care that are informed by a systematic review of evidence and an assessment of the benefits and harms of alternative care options” [1]. Noticeably, the definition highlights the necessity to support recommendations based on the results of rigorous systematic reviews to address specific research questions that minimize bias by using explicit methods [1], [2]. The last few years have seen some significant changes in the methods concerning the development of CGs, and methodological working groups such as GRADE have concentrated on providing transparent, systematic, methodologically-sophisticated guidance for each step in the formulation of recommendations [3].
Although concern has been raised over the trustworthiness of guidelines, several international initiatives have been put forward in order to develop credible guidelines. These endeavors, amongst other things, have targeted the vexing problem of guidelines and conflicts of interest, balancing the pros of having input from experts and assuring that conflicts of interest do not influence recommendations [4], [5]. At the same time, methodologists continue work to develop frameworks and tools to facilitate the development and dynamic updating of trustworthy guidelines [6].
Health professionals' attitudes and expectations concerning EBM and CGs share some similarities. These include aspects such as a generally positive attitude toward bringing rigor to clinical practice, the need to improve their skills and the difficulties in the implementation of the evidence in real life [7], [8]. In general, evidence shows physicians are highly satisfied with guidelines [9], [10]. In a Canadian survey, physicians considered CGs a useful educational tool to guide their practice and improve quality of care [11]. Nevertheless, the practical impact of these documents was of concern, particularly as to their application in individual patients, their role in resource saving, and their potential for a possible increase of lawsuits [11]. A meta-synthesis of qualitative studies recently showed that although GPs have a positive attitude to CGs, they do not follow them for a number of reasons, mostly related to the lack of applicability to the particular patient's needs [12]. The analysis showed that physicians' reasons to adhere to CGs or not differed depending on whether the guideline was prescriptive, that is, whether it promoted a certain type of behavior or treatment, or proscriptive, that is, if it discouraged certain treatments or behaviors [12].
In Spain, and despite the fact that CGs play a central role in the strategies to improve decision-making, several studies show that the implementation of CGs has certain drawbacks, such as the limited systematization and coordination of resources, the proliferation of guidelines that contrast with suboptimal updating, the redundancy in the subjects addressed, and biases in their methodology [13]. Other experiences have also stressed the perceived need of transparency through the process of development and implementation of guidelines [14].
Another factor that may imply a barrier to the use of CGs is the existence of multiple systems for the formulation of recommendations, which sometimes have important limitations [15]. This hinders understanding and communication. The lack of transparency in the criteria applied in evaluating evidence and grading of the recommendations, the unclear estimation of a benefit/risk ratio, and the absence of the grading of the relative importance of outcomes of interest are aspects that limit the credibility of recommendations and, hence, may limit implementation. There is little research evaluating how these systems may influence the perception among health professionals. Some authors have argued that the typical classification of the recommendations using letters, symbols or numbers has been scarcely evaluated [16], [17]. Finally, empirical evidence suggests that their wording could improve their implementation [18], [19]. For all the above-mentioned arguments, we undertook this study to explore knowledge, perceptions and attitudes of health care professionals towards CGs, grading systems, and factors contributing to their understanding, acceptance, and use.
Methods
Qualitative research based on discussion groups. We reported the complete details of the study protocol elsewhere [20]. This qualitative research method allows comprehensive approximations to the constitutive processes of reality and the perceptions of specific groups, an in-depth analysis of their subjectivity, and an understanding of participant's experiences and beliefs [21]–[23]. This study is part of a mixed methods research project which includes a nationwide survey among health professionals.
First, we organized six discussion groups in different geographical areas to reflect the regional differences in the organization of healthcare services in Spain. Enrolment of the 46 participating physicians from twelve medical specialties was carried out considering a purposive sample [24], [25]. Group composition was based on the level of care in which physicians provided professional assistance (specialized or primary care) and by the knowledge declared in evidence-based medicine. In the case of physicians who provide services in specialized care, a greater representativeness of clinical rather than surgical profiles was sought. Furthermore, to guarantee basic group homogeneity and intergroup heterogeneity, an appropriate age and sex distribution was considered. The first contact with informants was by letter or by telephone. After consent was granted, participants were reminded by SMS of the day when the discussion groups would take place. Table 1 outlines the main characteristics from participants in the discussion groups. The study has been approved by the Ethics Board of the Hospital de la Santa Creu i Sant Pau, Barcelona, Spain. Participants provided written informed consent to participate in this study.
We designed a protocol to guide the moderation and the role of the observers, and to homogenize the process between groups. Similarly, we prepared a background document with general topics to be addressed in the group interviews, and with suggestions of both open and concrete questions that could be of assistance or guidance for moderators. To explore participants' perception regarding the different systems of classification and grading of the evidence, we prepared illustrative materials containing recommendations formulated with different rating systems.
Discussion groups were moderated by members of the research team who had previous experience in that field, together with the collaboration of an observer, who collected field notes on anything that might be of interest for subsequent data analysis. No group interviews lasted more than ninety minutes. They were digitally audio-taped with the informants' consent and transcribed verbatim to obtain the final set of qualitative data for the analysis. The moderator initiated the discussion indirectly with an open-ended question (what comes to your mind when we speak about CGs?). During the session, the moderator assumed a minimally directive role, limiting intervention to prompting the group - if necessary - to address the topics in the script.
Data were analysed in a first step from a theoretical approach based on the grounded theory in which categories of analysis were defined. This was later complemented with a discursive sociological approach that analysed discursive positions, narrative configurations and semantic spaces. All the analyses were triangulated between several members of the research team, and discussed with those in charge of the moderation and observation in each group [26].
Results
Physicians participating in the discussion groups showed that their perception of CGs was based on two main constructs related to their knowledge of CGs (such as those aspects determining the attitude of professionals toward the theoretical concept of CGs) and the possible usefulness of CGs (pragmatic characteristics). Both discursive constructs were involved in the following categories that shaped the participants' attitudes toward CGs and were also determinants of their utilization (implementation of the CPG in the setting of daily practice): confidence (position of trust that professionals confer to the theoretical information found in the CPG), access and dissemination (health professionals' modes of contact with CGs and their recommendations), format (aspects related to the way information is presented), recommendations (CGs conclusions as indications aimed at directing clinical practice and decision-making), and levels of scientific evidence (conceptual criteria of classification of the recommendations obtained from studies based on clinical practice).
Theoretical conception and applicability to practice (Table 2)
The observed attitudes that define the knowledge construct that physicians have regarding CGs are based on a series of intellectual constructions on the characteristics of these documents, or the concepts supporting them. The main point in common when it comes to defining a CG, regardless of the methodological knowledge possessed by participants in the discussion groups in this respect, was that of the homogenizing function of CGs, which helps to standardize the practice, to organize assistance and, ultimately, to assist in clinical decision-making, assigning them an instrumental character (table 2).
Participants easily identified the theoretical concepts that define CGs, showing a positive perception of medical professionals toward CGs as tools based either on scientific literature, or on a consensus that should homogenize the clinical practice (table 2). We observed a disagreement between physicians' positive perception regarding CGs at a theoretical level and their applicability in practice. This disagreement stemmed from the identification of a series of barriers that limit this positive perception. Such barriers mainly corresponded to CGs that did not establish a process offering a clear solution and CGs that were not individually applicable to patients. However, they also corresponded to the vast quantity of information they gather, which hinders their comprehension and management in daily practice (table 2). Data show the link between theoretical knowledge on CGs (which defines its perception) and their practical knowledge and, ultimately, the possible use that can be made of them. Similarly, training was regarded as a facilitator of knowledge and use of CGs.
Usefulness, utilization and implementation of the recommendations (Table 3)
As to the attitudes towards CGs use, two differentiated behavioural patterns or demands were observed. First, there was the motivation of professionals to carry out an active search of documents that could facilitate their clinical practice. Second, professionals appeared to have a passive attitude, expecting the workplace organization (planners, managers, etc) to facilitate access (table 3).
CGs use in daily practice implies a process of implementation, involving dissemination and training in these documents. These two meanings constitute facilitating elements that take part in a positive feedback process, given that a correct dissemination and training contribute to a better knowledge and use of CGs (table 3).
The perceived usefulness of CGs was identified as the main category filling the gap between theoretical knowledge and their use in the clinical practice. Physicians who participated in the discussion groups had a positive opinion about CGs as these tools can provide standards from which they can make the best possible decision when facing situations of uncertainty (table 3), although they were not recognized as resolution tools for all the problems they have to face in clinical practice.
As for the CGs aim of reducing the variability in clinical practice, two opposing attitudes have been identified among participants in the discussion groups. On the one hand, CGs are perceived as useful tools for a correct approach of the most frequent conditions, and on the other hand, they are also perceived as assistance tools to address less common health issues of a lesser complexity (table 3).
Availability of several formats of a CG is identified as another possible determinant of its use. Implementation and use of CGs could be facilitated by having a summarized version with concise information on the most important recommendations for clinical practice, submitting written recommendations in plain language with no ambiguities, and gaining access to an electronic format that allows for frequent updating (table 3).
Another determinant in the use of CGs was confidence. Although the development and formulation process of recommendations is perceived as an important factor, participants in the discussion groups ascribed much more importance to those who develop and endorse the CGs. The fact that CGs developers can be expert methodologists or professionals not involved in clinical practice generates some rejection and limits their implementation. A factor that contributed most to the confidence perceived by health care professionals was authorship of recommendations, which could lead to a positive or negative attitude toward the use of a given CG. The involvement of the pharmaceutical industry in the development of CGs generates a certain attitude of suspicion, and even rejection to the recommendations included in CGs developed or financed by the same company (table 3).
Certain aspects related to the usefulness of CGs create barriers to their use. One barrier is the inability of CGs to adapt or take into account the characteristics of the particular setting in which their recommendations are intended to be used, thus limiting its applicability. Their use is also limited by the availability of several CGs to address a single health problem; this makes it difficult to assimilate their contents, and recommendations may even diverge. Another point to consider is that updating of CGs may not keep up with scientific advances in certain pathologies, possibly triggering scepticism toward some of these documents (table 3).
Levels of evidence and recommendations (Table 4)
The collected data show that the levels of scientific evidence and the recommendations are a common element between the two main categories built upon discourse analysis (knowledge and use of CGs), being located in a macro-discourse level. They appear in the theoretical plane (as one of the basic components in the development of CGs), but also as a category that could guarantee credibility and confidence in CGs, thus facilitating their use, dissemination, and access.
CGs methodological sections, with information such as tables synthesizing the evidence that back up the recommendations, were highly regarded by participants, since they are perceived as indicators of strictness, responsibility and transparency. However, although the inclusion of these sections in CGs adds value, it does not necessarily mean that users know how to handle them, since some knowledge of research methodology is needed to this end (table 4).
Although the moderators of the groups tried to explore participants' perception of the different ways in which levels of evidence and the strength of the recommendations were graded, participants showed little interest in the grading models used. In general, participants showed a preference for systems based on numbers and letters to classify the evidence and grade the strength of the recommendations (table 4). The lack of standardized classification systems unifying the many suggestions to date is responsible for considerable confusion among users who must move from one CG to the other, and is regarded as an important barrier that may condition CGs use.
Discussion
Our study identified two main constructs explaining how physicians perceive CGs, and how this perception impacts on the attitude towards them. On one hand, there is a conception at the level of knowledge of CGs (from a theoretical construction) and, on the other hand, there are a series of categories related to their usefulness (from a pragmatic construction). The categories identified in the texts from the discussion groups shape aspects that enable us to relate these two main constructs, acting both as facilitators in the implementation of this conceptual perception of CGs, and as barriers to their use.
Overall, health care professionals show a positive attitude toward CGs, regarding them as useful tools to standardize clinical practice and shared decision-making. However, a number of factors appear to hinder their practical use. These factors are related to aspects such as a limited knowledge and limited access to CGs, lack of confidence in developers, lack of manageable formats in daily clinical practice, and the absence of standardized classification systems of the levels of evidence and recommendations.
Some barriers to the use of CGs have also been identified around these two constructs. The main sceptical attitudes toward CGs are related, on one hand, to the often difficult applicability of the scientific literature that supports recommendations at the individual level [27]. On the other hand, our study shows that the questioning of CGs and their usefulness is also rooted in the distance that separates the theoretical content collected by CGs from their realization in the clinical practice. This is mainly due to a limited adaptation of CGs to their contexts of application.
Our study suggests that an important barrier to the use of CGs, and a factor of physicians' perception of their usefulness, is the disagreement between the theoretical content of CGs and the daily practice. The role played by health professionals as primary actors in decision-making in clinical practice can differ from the objectives of CGs [28]. This represents a barrier to the interest that professionals show towards them, since there is a conflict between the recommendations presented by a CG and physicians' needs and experiences. We also found in the literature the need to incorporate front-line clinicians in the development of recommendations to improve the representativeness of CGs [12], [14].
As in other studies [12], [27], [28], participants in our discussion groups expressed the need for commitment in the dissemination and implementation of CGs on the part of their organizational structures. Participants highlighted the need to involve healthcare professionals with expertise in the area covered by the guideline scope to avoid methodologically rigorous documents but with limited applicability. In our context, the influence of the institutions or companies promoting CGs, funding source, and its authors are vital to the confidence that health professionals place in a CG. In this regard, the fact that a CG can be backed up or financed by the pharmaceutical industry is perceived with fear given the risk that their content might be conditioned by profitable intentions [27]. These aspects are even more relevant to participants in the study than the rigor of methods that sustain recommendations.
Some of the barriers to the use of CGs have been associated with practical aspects in our study, as other authors have also pointed out [12]. They are realized as aspects such as the lack of time to read or evaluate these tools, and this usually implies a setback for an adequate knowledge of CGs.
Some participants in our study expressed the need to disseminate CGs better before their implementation [27]. This includes receiving adequate training about the CGs content and their implications for their practice. However, this need for training is in conflict with settings having a high workload pressure and the limited time available to this end.
Dissemination, form of access to CGs, and CGs format are the main determinants of their knowledge and use. Overall, the literature also identifies format as a facilitator for their use [14]. Our study has identified the brief and simple formats as the aspects most appreciated by professionals. Accessibility stands out as a new component that adds value to the format of a CG. In this regard, it is of note that our study reinforces the positive perceptions received by the electronic formats mentioned in previous studies [27].
Contrary to our expectations, the participants in the discussion groups did not devote much time discussing grading systems. This is due to the natural complexity of qualitative designs and specifically in group discussions. A researcher who moderates group discussions has less control over the data produced than in in-depth interviews or quantitative studies. Participants expressed that the lack of standardized classification systems could be responsible for considerable confusion among users and a potential barrier to the use of CGs. This topic clearly deserves further research. Other authors highlight the importance of standardizing these systems to find a balance between rigour and simplicity [29].
Our study highlights the involvement that CGs developer groups should have when developing CGs aimed at answering specific and relevant questions. They should count on all the actors implicated, and they should be explicit, consistent and strict in the process of development. Finally, it is important that CGs are accompanied by working tools that strengthen their utility, trying to facilitate their dissemination and applicability, and taking into account the need to adapt CG formats to the specific characteristics of the different user profiles.
Acknowledgments
To all the professionals who participated in the discussion groups. We are also grateful to Carolyn Newey (Language Support Service, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain) for her help editing the manuscript.
Author Contributions
Conceived and designed the experiments: IS PAC. Performed the experiments: IS JMC PDC JC CO FM AK IG EC IdG AL RR FS PV PAC. Analyzed the data: IS JMC. Wrote the paper: IS JMC. Designed the study and organized the groups: IS JMC PDC JC CO FM AK IG EC IdG AL RR FS PV PAC. Organized and ran the groups: IS JMC PAC PDC JG FM EC CO. Provided substantial feed-back and accepted the final version of the manuscript: IS JMC PDC JC CO FM AK IG EC IdG AL RR FS PV PAC.
References
- 1.
Institute of Medicine (IOM) (2011) Clinical Practice Guidelines We Can Trust. Washington, DC: The National Academies Press. 290 p.
- 2.
The Cochrane Collaboration (2011) Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 Available: www.cochrane-handbook.org.Accessed 2013 Dec 23.
- 3. Guyatt GH, Oxman AD, Schünemann HJ, Tugwell P, Knottnerus A (2011) GRADE guidelines: a new series of articles in the Journal of Clinical Epidemiology. J Clin Epidemiol 64: 380–382.
- 4. Hirsh J, Guyatt G (2009) Clinical experts or methodologists to write clinical guidelines? Lancet 374: 273–275.
- 5. Guyatt G, Akl EA, Hirsh J, Kearon C, Crowther M, et al. (2010) The vexing problem of guidelines and conflict of interest: a potential solution. Ann Intern Med 152: 738–741.
- 6. Vandvik PO, Brandt L, Alonso-Coello P, Treweek S, Akl EA, et al. (2013) Creating clinical practice guidelines we can trust, use, and share: a new era is imminent. Chest 144: 381–389.
- 7. Freeman AC, Sweeney K (2001) Why general practitioners do not implement evidence: qualitative study. British Medical Journal 323: 1100–2A.
- 8. Heselmans A, Donceel P, Aertgeerts B, Van de Velde S, Ramaekers D (2009) The attitude of Belgian social insurance physicians towards evidence-based practice and clinical practice guidelines. BMC Fam Pract 10: 64.
- 9. Farquhar CM, Kofa EW, Slutsky JR (2002) Clinicians' attitudes to clinical practice guidelines: a systematic review. Med J Aust 177: 502–506.
- 10. Heselmans A, Donceel P, Aertgeerts B, Van de Velde S, Ramaekers D (2009) The attitude of Belgian social insurance physicians towards evidence-based practice and clinical practice guidelines. BMC Fam Pract 10: 64.
- 11. Hayward RS, Guyatt GH, Moore KA, McKibbon KA, Carter AO (1997) Canadian physicians' attitudes about and preferences regarding clinical practice guidelines. CMAJ 156: 1715–1723.
- 12. Carlsen B, Glenton C, Pope C (2007) Thou shalt versus thou shalt not: a meta-synthesis of GPs' attitudes to clinical practice guidelines. Br J Gen Pract 57: 971–978.
- 13. Morillo García A, Pérez Lozano MJ, Calderón E, Guerra JA, Santos JM, et al. (2005) Factores asociados al uso de guías de práctica clínica en el manejo del ictus. Perspectivas del profesional. In XXI Congreso de la SADEMI, Jerez. Libro de comunicaciones 131.
- 14. Carlsen B, Norheim OF (2008) “What lies beneath it all?”–an interview study of GPs' attitudes to the use of guidelines. BMC Health Serv Res 8: 218.
- 15. Atkins D, Eccles M, Flottorp S, Guyatt GH, Henry D, et al. (2004) Systems for grading the quality of evidence and the strength of recommendations I: Critical appraisal of existing approaches. BMC Health Services Research 4: 38.
- 16. Schunemann HJ, Best D, Vist G, Oxman AD (2003) GRADE Working Group (2003) Letters, numbers, symbols and words: how to communicate grades of evidence and recommendations. CMAJ 169: 677–680.
- 17. Akl EA, Maroun N, Guyatt G, Oxman AD, Alonso-Coello P, et al. (2007) Symbols were superior to numbers for presenting strength of recommendations to healthcare consumers: a randomized trial. J Clin Epidemiol 60: 1298–1305.
- 18. Michie S, Lester K (2005) Words matter: increasing the implementation of clinical guidelines. Qual Saf Health Care 14: 367–370.
- 19. Michie S, Johnston M (2004) Changing clinical behaviour by making guidelines specific. BMJ 328: 343–345.
- 20. Kotzeva A, Solà I, Carrasco JM, Díaz del Campo P, Gracia FJ, et al. (2010) Perceptions and attitudes of clinicians in Spain toward clinical practice guidelines and grading systems: a protocol for a qualitative study and a national survey. BMC Health Serv Res 10: 328.
- 21.
Morgan DL, Krueger RA editors (1998) The Focus Group Kit. Thousand Oaks (CA): Sage Publications. 103 p.
- 22.
Green J, Thorogood N, editors(2009) Qualitative Methods for Health Research, 2nd Edition. London: SAGE Publications. 320 p.
- 23.
Pope C, Mays N, editors: Qualitative Research in Health Care (2000). London: BMJ books. 168 p.
- 24. Curtis S, Gesler W, Smith G, Washburn S (2000) Approaches to sampling and case selection in qualitative research: examples in the geography of health. Soc Sci Med 50: 1001–1014.
- 25. Tuckett AG (2004) Qualitative research sampling: the very real complexities. Nurse Res 12: 47–61.
- 26.
Phillips N, Hardy Ceditors (2002) Discourse Analysis. Investigating Processes of Social Construction. Thousand Oaks (CA): Sage Publications. 104 p.
- 27. Carlsen B, Norheim OF (2008) “What lies beneath it all?”–an interview study of GPs' attitudes to the use of guidelines. BMC Health Serv Res 8: 218.
- 28. Van der Weijden T, Légaré F, Boivin A, Burgers JS, van Veenendaal H, et al. (2010) How to integrate individual patient values and preferences in clinical practice guidelines? A research protocol. Implement Sci 5: 10.
- 29. Dahm P, Djulbegovic B (2011) The Australian ‘FORM’ approach to guideline development: the quest for the perfect system. BMC Med Res Methodol 11: 17.