In the landmark 2001 report, “Crossing the Quality Chasm,” the Institute of Medicine (IOM) identified that effective patient-physician communication and information sharing were critical to high-quality patient-centered care [ 1].
These aspects of the clinical encounters are essential for pre-surgical decision making between the patient and the surgeon, particularly in the face of major, high-risk operations, in which complications can be serious.
In the field of surgery, the informed consent process has traditionally dominated preoperative communication between patients and surgeons. Experts have questioned whether informed consent alone is adequate for decision making [2-4]. The consent process can be unidirectional and focus on revealing the risks, benefits, and alternatives of a procedure with little patient involvement [5-7].
On the other hand, shared decision making, in a more collaborative approach to communication, that incorporates the patient’s values and preferences, while increasing their role in the decision [8]. The process aims to ensure patient understanding, align their preferences with their treatment choices, and improve patient outcomes [9,10]. Informed decision making is a key component of the shared process that focuses on how physicians encourage patient participation and provide them with information to make a decision about their treatment [4,11-13].
Despite the critical importance of effective communication and information sharing before surgery, few studies have examined the extent to which surgeons engage in informed decision making, and none have uniquely addressed how that process adequately incorporates to informed consent.
Therefore, the objectives of this study were to quantify informed consent and informed decision making in patient-surgeon consultations, and to examine the intersection of the two processes. To do this, the authors focused on high-risk procedures, because those conversations were more likely to have high levels of informed consent and informed decision making.
Additionally, it is an area where patient preferences and participation are critical. The ability to objectively evaluate these aspects of communication is crucial to developing quality standards and facilitating competency-based education and certification.
| Methods |
> Data collection and study population
This study was a secondary analysis of 90 audio recordings of patient-surgeon consultations involving the decision to perform major, high-risk surgery, collected as part of a prospective, multi-institutional study that qualitatively explored the content of the conversations [14 ,fifteen].
Participants were recruited from 3 sites in the United States and Canada: University of Wisconsin, Madison, WI ; Brigahm and Women’s Hospital, Boston, MA ; and University of Toronto, Ontario, Canada . Each consultation was recorded, transcribed verbatim, and deidentified.
Additional details on participant enrollment have been published previously [14,15]. Approval was obtained from the Institutional Review Boards of the 3 participating institutions. All surgeons, patients and associates, present during the conversation, gave informed consent for their participation.
> Measurement of results
Quantitative analysis of the content of each transcript was performed to measure the degree to which surgeons achieved informed consent and informed decision making. Two measures were used: a novel informed consent scale, based on the principles of the American College of Surgeons (ACS), and a well-established, validated informed decision-making scale, developed by Braddock et al. [4,11-13,16].
> Informed consent
To measure informed consent, the research team developed a novel scale with 11 unique items, in conjunction with an expert in medical and surgical ethics, based on the definition described in the ACS Statement of Principles (available at: http:/ /www.facs.org/fellows_info/statements/stonprin.html#anchor171960).
According to the Statement, “The surgeon’s discussion of informed consent should include: (1) the nature of the disease and (2) the natural consequences of not treating it, (3) nature of the proposed operation, including (4 ) estimated risk of mortality and morbidity, (5) most commonly known complications, which should be described and discussed. The patient must: (6) understand the risks as well as (7) the benefits of the proposed operation. The discussion should include: (8) discussion of what to expect during hospitalization and (9) post-hospital convalescence, and (10) alternative forms of treatment, including (11) non-surgical techniques. Those 11 elements capture the entire spectrum of informed consent communication and avoid overlap between elements.
> Informed decision making
To measure informed decision making, the Braddock scale was used, which has 9 items, including: (1) discussion of the patient’s role in the decision making, (2) impact of the decision on the patient’s daily life, (3) the clinical problem or nature of the decision, (4) alternative treatments, (5) potential benefits and risks of each alternative, (6) uncertainties surrounding the decision, (7) patient understanding, (8) patient preferences, and (9) provide the opportunity to involve other trusted people, such as family or friends [4,11,12].
> Transcript scoring
To ensure systematic measurement, two of the authors (SCP and JO) developed a codebook, after reviewing 5 representative transcripts. The codebook included the name of each element, its definition, and example quotes from the transcripts.
For the elements of informed decision making, example quotes were compared with published literature to ensure consistency and comparability of the results obtained. Throughout the scoring process, the codebook was iteratively revised to clarify definitions and ensure interrater reliability.
Five researchers with diverse backgrounds in general surgery (SCP, CB), critical care/trauma (AI), public health (EW), and psychology (BH), scored the 90 transcripts. Diversity improved the analysis and reduced any potential disciplinary bias. The research team was trained as a group to practice analyzing the transcripts until achieving interrater reliability of k > 0.65. Each team member then independently scored a subset of transcripts using the NVivo program (QRS International), and continued to meet at regular intervals to discuss issues and resolve them by consensus. The entire team obtained 25% of all transcripts at the same time to assess interrater reliability, which was excellent (k = 0.91).
> Discussion levels
To quantify the extent to which surgeons discussed informed consent and informed decision making, the 3-point scale described by Braddock et al was applied. [13].
For each conversation, items were assigned 0, 1, or 2 points if discussion of the item was absent (0), partial (1), or complete (2). Items were “partially discussed” and assigned 1 point if they were mentioned only briefly, without exchange between surgeon and patient. To be “fully discussed” and assigned 2 points, there needed to be a verbal exchange between surgeon and patient about the item.
To be consistent with the Braddock scale and previously published studies, points were assigned only if the discussion of the informed consent or informed decision-making element had been initiated by the surgeon [13,17]. Because this scoring method is a limitation of the Braddock system, discussion of items initiated by the patient or a family member, caregiver, or friend was tracked separately.
The primary outcome for each conversation was the sum of the points assigned for each scale. For informed consent, the score (CI-22) was from 0 to 22, and for informed decision making (TDI-18), it was from 0 to 18.
Higher scores indicated a greater extent, by the surgeon, of the informed consent discussion or informed decision making. The relative completeness of the conversations was also assessed using 3 subscales, as described by Braddock [4,18].
These subscales corresponded to reaching a basic (TDI-Min), intermediate (TDI-Int), or complex (TDI-Complex) level of informed decision making. To assess informed consent and facilitate comparisons, the authors (SCP, JO, and MS) reached consensus on appropriate items to include in 3 similar scales, which represented minimal (CI-Min), classic (CI-Classic), and maximum (CI-Max) level of informed consent. The definition of “classic” was based on the discussion of risks, benefits and alternatives.
> Statistical analysis
Standard descriptive statistics were used to summarize the data. To compare informed consent and informed decision making, CI-22 and TDI-18 scores were calculated, as well as the proportion of total items achieved with each scale. The proportion of discussions that reached basic (CI/TDI-Min), intermediate (CI-Classic/TDI-Int), and complex/maximum (CI/TDI-Complex) levels for informed consent and decision making was also compared. informed.
Furthermore, with bivariate analyzes the impact of the following variables on informed consent and informed decision making was evaluated: whether the decision was for surgery or alternative management, location of the surgeon, specialty of the surgeon, whether only 1 option was offered , if the patient had 1 or more consultations, and which consultation in the series was the one captured, if the family was present in the discussion, and the patient’s demographic data.
Fisher’s exact test was used to analyze categorical variables, t test for continuous variables, and analysis of variance (ANOVA) for continuous variables between 3 or more groups. Statistical significance was established at p < 0.05. All analyzes were carried out using the GraphPad QuickCalcs program (GraphPad Software).
| Results |
90 patient-surgeon conversations were analyzed about the decision to perform major, high-risk operations, which included cardiac (n = 38), thoracic (n = 13), neurological (n = 21), vascular (n = 9), and oncological (n = 9). Most patients (n = 73; 81.1%) had a single preoperative consultation, although 4 patients (4.4%) had 2 preoperative consultations that were recorded and stitched together, and 13 patients (14.4% ) had 2 preoperative consultations of which only 1 was recorded and analyzed.
Surgeons most frequently discussed common complications (90.0%), the nature of the disease (85.6%), and the nature of the proposed operation (82.2%). Explicit discussion of alternative treatments occurred in 46.7% of consultations. Infrequently, surgeons specifically assessed patients’ understanding of the risks (4.4%), and benefits (0%) of surgery, and unusually (13.3%) described the natural consequences of not doing treatment.
The vast majority of conversations included elements of informed decision making that overlap with informed consent – the “pros” and “cons” of surgery (92.2%), and common complications (90.0). %) – but less frequently they discussed the nature of the decision (60.0%). Surgeons assessed patients’ overall understanding (58.9%), although these were often closed-ended, yes/no questions such as "Does this make sense to you?" Surgeons infrequently discussed the impact of the decision on the patient’s daily life (14.4%), explored the patient’s preference (12.2%), or assessed their desire to receive information from trusted others (1.1). %) [9].
Comparison of the extent to which surgeons discussed elements of informed consent or informed decision making revealed that surgeons were more likely to discuss elements related to obtaining informed consent (48.9% vs 37 .9%, respectively, p < 0.0001).
Although surgeons were more likely to discuss informed consent, the proportion of conversations that met the minimum or basic criteria for informed consent and informed decision making was similar, 33.3% and 25.6%, respectively. ( p = 0.33). Similarly, there were no significant differences in the proportion of discussions that met the criteria for classic/intermediate and maximum/complex, consent, or decision making. Patients and/or their family members initiated discussion of an element of informed consent or informed decision making in 30% and 26% of the discussions, respectively.
Since not all patients underwent surgery, additional analyzes examined surgeon-patient communication, based on the outcome of the consultation. When a definitive plan was developed during the consultation, there was no difference in the extent to which surgeons discussed aspects of informed consent (mean CI-22 score: plan developed 9.3 ± 3.0 vs no plan 9.0 ± 4.9; p = 0.77). However, when a plan was not developed, surgeons tended to discuss more elements of informed decision making (mean TDI-18 score: plan developed 5.6 ± 2.9 vs. no plan 7.0 ± 3.5; p = 0.06), probably due to the need for more deliberation.
In contrast, if it was decided not to perform surgical management, no treatment, or additional testing, surgeons were more likely to discuss aspects of informed decision making, including decision uncertainty (50.0% vs. 15.0%). .3%; p < 0.01) and alternative treatment options (72.2% vs 35.6%; p < 0.01). Surgeons also tended to discuss patient preferences more frequently (22.2% vs 6.8%; p = 0.08). However, no difference was observed in the mean TDI-18 score in the consultations where the decision was to continue with surgery, compared to non-surgical management (5.5 ± 3.0 vs 5.8 ± 2.8; p = 0.70, respectively.
In addition to evaluating the impact of the outcome of the consultations on informed consent and informed decision making, the influence of the number of consultations between the patient and the surgeon was assessed, specifically if: (1) there had only been 1 consultation; (2) 2 consultations occurred but only the second was scored; and (3) 2 consultations occurred and both were scored together.
For informed consent, the mean CI-22 score increased when 2 patient-surgeon consultations occurred, although this finding was not significant (9.1 ± 3.6 vs 8.5 ± 3.3 vs 12.5 ± 3.3, p = 0.15, respectively). This trend was also present for the degree of informed decision making (mean TDI-18 score: 6.21 ± 3.1 ± 4.2 ± 3.0 vs 7.5 ± 2.4; p = 0.06, respectively).
When the degree to which patient and other factors affected the discussions was assessed, no overall differences in informed consent or informed decision making were observed based on patient race, sex, history of prior surgery, presence of family. in the consultation, and location. While no overall gender differences were observed, surgeons were more likely to discuss the benefits of the operation with women who chose surgery (64% vs 32.4%; p = 0.02).
Patients with a high level of health literacy (defined as rarely or never needing help with written materials), and with a level of education (LEE) greater than secondary school (ES), had discussions that included more elements of consent informed. Importantly, there were no differences in the level of informed decision making based on a high level of health literacy or NDE (mean IQ-22 for high level of health literacy or NDE 8.8 ± 3.3 vs low 5. 3 ± 3.9; p = 0.01; for NDE greater than ES 9.3 ± 3.0 vs ES or less 7.1 ± 3.7; p = 0.008).
Finally, when surgeon specialty was analyzed, a similar degree of informed consent and informed decision making was observed among surgeons. However, this examination also showed significant variability between surgeons and within each surgeon.
For informed consent, the mean minimum and maximum CI-22 scores for individual surgeons ranged from 3.0 ± 1.4 to 11.8 ±2.5. Furthermore, the mean difference between the minimum and maximum score for each surgeon was 7.9 ± 3.1. Similarly, the mean minimum and maximum scores for the TDI-18 ranged from 3.0 ± 1.4 to 9.0 ± 4.2, and the mean range of the difference was 7.9 ± 3.1.
| Discussion |
Adequate communication between patients and surgeons before surgery is essential for patient-centered care, especially for those undergoing major, high-risk procedures [1-3]. The IOM states that transparent communication should involve eliciting patient needs, preferences, and values; share knowledge and evidence; and place decision control in the hands of the patient, particularly when making complex decisions, such as those related to surgery [1].
Because informed consent is a long-standing medical practice that was developed as a legal requirement, obtaining consent is not inherently patient-centered and has not had a significant focus in medical education [5,19,20]. As such, the classic informed consent model may miss opportunities to involve patients more deeply in treatment decision-making, and fail to take into account the specific concerns of each patient’s life [9].
The IOM recommends shared decision making as an ideal patient-centered model necessary to achieve high-quality care. However, shared decision-making may not adequately encompass the critical process of informed consent [1].
In this study, we evaluated only the adequacy of informed consent, as defined by the ACS, and the extent of necessary information shared with patients to fully participate in the shared decision-making process [4,11- 13]. Not surprisingly, surgeons performed best in the areas inherent to obtaining consent: discussing the risks, benefits, and alternatives to surgery, and assessing patient understanding.
Surgeons performed less well in the areas of decision making, particularly eliciting patient preference, discussing the patient’s role in the decision, recognizing uncertainty about how to proceed, assessing desire to receiving information from trusted friends or family, and exploring the impact of the decision on the patient’s daily life [12].
Adding these facets of communication to high-risk preoperative surgical consultations will likely improve the degree to which surgeons and their patients participate in shared decision-making.
Training surgeons to communicate more thoroughly about decisions and obtain patient preferences is one avenue to improve that process [21]. Surgeons can add simple questions, such as: “What do you expect from the treatment regarding your condition?” or “Thinking about this decision, what is the most important aspect to consider for you?” [22].
Additional incorporation of patient-centered informational interventions, such as decision aids, will also facilitate information exchange and the shared decision-making process. However, decision aids, in and of themselves, are not sufficient to support shared decision making [23,24]. While there is concern that such interventions may take longer, previous studies show that more comprehensive informed decision making does not, or only minimally, affect total consultation time [13].
While one may be tempted to focus on the absolute informed consent and informed decision-making scores that the authors have reported, and conclude that the surgeons have suboptimal consent and decision-making skills, the results should be interpreted in the light of day. context of previous research, and the limitations of the study. These results are similar to those of previously published analyzes of preoperative communication between patients and surgeons before elective general, orthopedic, or vascular surgery procedures [4,13,18].
In these studies, the proportion of surgeons who fully discussed all elements of shared decision making ranged between 0% and 9%, which is comparable to the finding of 4.4% in this work. An analysis of cardiologists discussing percutaneous coronary intervention similarly found that cardiologists fully discussed elements for informed decision making in 3% of cases [17].
The analysis of the proportion of surgeons and cardiologists who met, in these previous studies, the minimum criteria for informed decision making, was from 22% to 57%, which is also similar to the results of this study (25, 6%).
The results of these studies further showed that the communication elements most and least discussed by surgeons were also similar to the findings presented. Furthermore, the number of decision-making elements increased, in a study with vascular patients, when they had more than 1 consultation with the surgeon before making a decision [18].
However, unique to this study is the focus on informed consent. The significant variability that was found in the communication of informed consent and informed decision making among professionals is also comparable with previous results.
Taken together, the results of this work and those of other studies suggest that there is room to improve consent and informed decision-making skills for surgeons. These communication skills are often not taught during training, although efforts are being made to improve education in that area [6,13,25]. Formal education and assessment of these skills will help improve shared decision making and quality of care.
This study has limitations that must be considered when interpreting the results. First, the measurement that assessed informed consent on a new scale that is not validated. Although the scale was sensitive to the outcome of the decision (surgical vs. nonsurgical management), more studies are needed to evaluate the validity of that scale and its ability to accurately and reliably assess true achievement of informed consent.
The scale may also omit consent elements covered during clinical activity, outside of the audio-recorded surgeon-patient consultation. Additionally, conversations were analyzed before high-risk surgeries, where patients and surgeons may have a higher level of acceptance, which could reduce the number of decision-making and consent elements communicated. Likewise, it can be argued that the decision to undergo high-risk surgery may not be sensitive to patient preferences.
The scales used in this study also measure only the surgeon-initiated elements of informed consent and informed decision making, possibly missing additional patient-initiated discussion.
Finally, the selection of the surgeon and the knowledge that the conversation was being recorded can bias the results, although we would expect the scores to increase rather than decrease. However, these or similar tools are useful for research and skill assessment, particularly in assessing communication skills over time. The results also suggest that there is significant room for improvement, which is similar to findings in other settings.
| Conclusions |
Decision making before high-risk surgery is a complex process that requires mutual understanding between surgeons and patients. Providing patients with adequate information, so that they participate effectively in decision-making, is crucial to the process. This study, and others conducted in more elective settings, surgical and nonsurgical, suggest that current processes for informed consent and informed decision making do not meet the IOM standard.
The development and validation of measurements that evaluate the quality of informed consent and informed decision making will help ensure the improvement of these processes. Implementation of those measurements will need to occur without additional burden on clinicians, and evaluate actual communication rather than documentation.
Educational programs for surgeons in training and practicing surgeons promise to improve informed consent and decision-making skills. Such training programs would bridge the gap between clinician-centered decision-making tools, such as risk calculators, and patient-facing materials, such as decision aids, to ensure appropriate communication, elicitation of preferences and truly shared decisions.
