Decision-Making Quality in Parents Considering Adenotonsillectomy or Tympanostomy Tube Insertion for Their Children

Key Points

Question  What is the level of decisional conflict and decisional regret experienced by parents considering elective pediatric otolaryngology surgeries for their children, and are they related to perceptions of shared decision making?

Findings  In this prospective cohort study, many parents experienced clinically significant decisional conflict, while very few had significant decisional regret. Parent ratings of shared decision making were significantly negatively correlated with decisional conflict and regret levels, and parents with clinically significant decisional conflict had significantly higher decisional regret.

Meaning  Parents who perceived themselves to be more involved in the decision-making process had lower levels of uncertainty and regret about their decision.

Importance  Shared decision making is a process in which clinicians and patients make health care decisions in a collaborative manner using the most up-to-date evidence, while considering patient values and preferences. Shared decision making is thought to have a positive influence on the decision-making process in medicine.

Objective  To describe the level of decisional conflict and decisional regret experienced by parents considering surgery for their children and to determine relations among decisional conflict, decisional regret, and shared decision making.

Design, Setting, and Participants  A prospective cohort study was conducted at an academic pediatric otolaryngology clinic. Participants included 126 parents of children younger than 6 years who underwent consultation for adenotonsillectomy or tympanostomy tube insertion.

Main Outcomes and Measures  Parent participants completed the Shared Decision Making Questionnaire–Parent version, Decisional Conflict Scale (DCS), and Decisional Regret Scale (DRS). Surgeons completed the Shared Decision Making Questionnaire–Physician version.

Results  This study included 126 parents; 102 women (mean [SD] age, 33.2 [5.1] years) and 24 men (mean [SD] age, 35.6 [6.3] years). Overall, 34 parents (26%) reported clinically significant decisional conflict. Only 1 parent experienced moderate to strong decisional regret; 28 parents (43.7%) had mild decisional regret. Both parent and physician ratings of shared decision making were significantly negatively correlated with total DCS scores. Parent SDM-Q-9 and total DCS scores were significantly negatively correlated (r_s[118] = −0.582; P < .001). Similarly, physician SDM-Q-Doc and total DCS scores were also significantly negatively correlated (r_s[118] = −0.221; P = .04). Only parent ratings of shared decision making were significantly negatively correlated with total DRS scores (r_s[63] = −0.254; P = .045). Those parents with clinically significant decisional conflict had significantly higher DRS scores (P = .02).

Conclusions and Relevance  Many parents experienced significant decisional conflict when making decisions about their child’s elective surgical treatment. Parents who perceived themselves as being more involved in the decision-making process reported less decisional conflict and decisional regret. Future research should explore the influence of decision quality on health outcomes and develop methods to improve shared decision making.

For many medical conditions, there can be multiple treatment options, and evidence-based assessments often fail to identity one treatment as clearly superior. When considering pediatric conditions, children can often improve over time and can “grow out” of the medical condition.¹ Therefore, parents contemplating elective surgical procedures for their children may have difficulty deciding on a treatment course. This uncertainty, referred to as decisional conflict, is frequently accompanied by negative consequences, such as emotional distress and delays in decision making.² A decisional outcome related to decisional conflict is decisional regret. Decisional regret has been associated with postoperative complications,^3,4 and research suggests that it may affect future medical decision making.⁵ Decisional outcomes have rarely been studied in pediatrics, despite the innate challenges parents face making proxy decisions for their children.^6,7 The few studies to date have demonstrated that parents do struggle with medical decisions for their children.^8-10

Taking a shared approach has been suggested as a potential strategy to improve quality of decision making in medicine.¹¹ Shared decision making describes an approach in which patients and health care providers collaborate to formulate a treatment decision that is based on the most up-to-date evidence, while considering the values and decision needs of the patient.¹² Previous research indicates that parents may not be as involved in decision making for their children as they would like.^10,13,14 Moreover, increased parental perceptions of shared decision making has been related to improved decisional outcomes.^10,15 Specifically, a pilot study¹⁰ conducted by our group showed that parents who thought themselves to be more involved in the decision-making process reported significantly lower levels of decisional conflict. However, due to the small sample size, some of the important baseline and contextual factors (eg, provider influences) could not be assessed with respect to decisional conflict. Decisional regret, an important measure of decision quality, was also not measured in this study.

The objectives of the current study were to describe outcomes related to decision making in parents considering elective pediatric otolaryngology surgeries and whether these outcomes were related to perceptions of shared decision making.

This study was part of a larger mixed-method prospective cohort study that included video recording of pediatric otolaryngology consultations. The video recordings were analyzed and reported elsewhere.¹⁶

Parents of children presenting to the pediatric otolaryngology clinic at a tertiary academic hospital in Eastern Canada with either recurrent and/or chronic pharyngotonsillitis, sleep disordered breathing, or recurrent acute and/or chronic otitis media were prospectively enrolled. Parents were considering the surgical options of adenotonsillectomy, tonsillectomy, or tympanostomy tube insertion, respectively. Participants were excluded if they were not fluent in English or not legal guardians of the child. When more than 1 parent was present, only the mother was included because they participated most commonly and previous research has mainly focused on mothers.⁶

For health care providers, 3 fellowship-trained pediatric otolaryngologists were included.

Local institutional review board approval was obtained. Parents gave consent to participate in this study before undergoing the surgical consultation. Parents completed the Demographic Form, Decisional Conflict Scale (DCS), and Shared Decision Making Questionnaire–Patient Version (SDM-Q-9) after the consultation. They also completed the Decisional Regret Scale (DRS) via a telephone call 10 to 16 months after the consultation.

Physicians completed the Shared Decision Making Questionnaire (SDM-Q-Doc) after the consultation.

This form was used to obtain demographic information and whether the family had any previous surgical experience.

The DCS assessed uncertainty that the parents experienced pertaining to their child’s treatment decision.¹⁷ It is composed of 16 items rated on a Likert scale from strongly agree to strongly disagree. Sample items include: I am clear about what benefits matter most to me; I feel sure about what to choose; and My decision shows what is important to me. The DCS is a validated measure that has been administered to parents of pediatric patients in other medical settings.^5,10 Previous research has shown that a total DCS score of 25 or higher is suggestive of clinically significant decisional conflict.¹⁷

The SDM-Q-9 assessed parent perceptions of shared decision making. It is composed of 9 items rated on a Likert scale from completely disagree to completely agree.¹⁸ Sample items include: My child’s doctor and I selected a treatment option together; My child’s doctor made it clear that a decision had to be made; and My child’s doctor asked me which treatment option I prefer. The questionnaire yields scores that range from 0 (no shared decision making) to 100 (extremely high level of shared decision making). The SDM-Q-9 is a validated scale that has shown high reliability in several medical settings.¹⁸

The SDM-Q-Doc assessed surgeon perceptions of shared decision making. It is composed of 9 items rated on a Likert scale from completely disagree to completely agree.¹⁹ Sample items include: I wanted to know exactly from my patient how he/she wants to be involved in making the decision; I told my patient that there are different options for treating his/her medical condition; and My patient and I thoroughly weighed the different treatment options. The SDM-Q-Doc is a validated measure that has previously demonstrated high reliability.¹⁹

The DRS assessed regret that the parents had about their child’s treatment decision. It is composed of 5 items rated on a Likert scale from strongly agree to strongly disagree.²⁰ Sample items include: It was the right decision; I regret the choice that was made; and I would go for the same choice if I had to do it over again. The scores range from 0 (no regret) to 100 (extremely high regret). Based on previous research, DRS scores above 26 imply the presence of moderate to strong decisional regret, while DRS scores between 1 and 25 indicate mild regret.²⁰ The DRS is a validated scale that has been used in other surgical settings.^4,15

The data analysis was conducted with SPSS Version 17 (IBM Corp).

All measures (DCS, SDM-Q-9, SDM-Q-Doc, and DRS) were not normally distributed; therefore, nonparametric tests were used. Mann-Whitney U (2 groups) and Kruskal-Wallis (3 groups) tests were used to examine DCS and DRS scores by baseline factors, such as surgery type and which surgeon was involved in the consultation. Three bivariate Spearman ρ correlations were used to examine the relation between perceptions of shared decision making, decisional conflict, and decisional regret. The first correlation was conducted between SDM-Q-9 and DCS, second between SDM-Q-Doc and DCS, and third between DCS and DRS.

A sample size of 52 participants would provide 80% power to detect a correlation coefficient of 0.38 between decisional conflict and shared decision making.²¹ To account for multiple possible predictors and to facilitate multivariate analysis with correction and potential attrition or incomplete survey responses, 131 participants were recruited.

One hundred twenty-six caregivers were enrolled. Sixty-nine (55%) elected to proceed with surgery, while 57 (45%) decided to monitor their child’s symptoms (watchful waiting group). Participant information is presented in Table 1.

The 3 surgeons were all fellowship-trained pediatric otolaryngologists (1 female; age range, 37-47 years).

The Cronbach α (internal consistency) for the DCS was 0.97, indicating excellent scale reliability.²² The median total DCS score was 9.38 (standard error [SE], 1.37; range, 0-98.44; interquartile range [IQR], 25.00). Thirty-four parents (27.2%) had DCS scores greater than or equal to 25, which indicates clinically significant decisional conflict (Table 2).

There were no differences in total DCS scores by surgery type, previous surgical experience (for the index child or other children in the family), consulting surgeon, and parent education level. Parents who decided not to proceed with surgery had significantly higher levels DCS scores than parents who decided to proceed with surgery (mean difference, 7.32; 95% CI, 4.75-9.85; Mann-Whitney U P < .001). Of the 64 parents who decided to proceed with surgery, 13 (20%) had significant decisional conflict. Of the 54 parents who decided to monitor their child’s symptoms, 21 (39%) had significant DCS scores.

Parent SDM-Q-9 and total DCS scores were significantly negatively correlated (r_s[118] = −0.582; P < .001). Similarly, physician SDM-Q-Doc and total DCS scores were also significantly negatively correlated (r_s[118] = −0.221; P = .04).

The Cronbach α for the DRS was 0.81, indicating good reliability.²² Of the 126 initial participants, 97 consented to be contacted for follow-up and 13 of those had their telephone numbers disconnected. A total of 64 parents completed the DRS (76% response rate). There were no differences between those who did and did not complete the DRS in terms of demographics, previous surgical experience, decision type, and consulting surgeon. Also, there were no differences in DCS, SDM-Q-9, and SDM-Q-Doc scores between DRS responders and nonresponders (Table 3).

The median total DRS score was 0.00 (SE, 1.18; range, 0-35.00; IQR, 15.00). Thirty-five parents (54.7%) had no regret, 28 parents (43.7%) had mild decisional regret, while 1 parent (1.6%) reported moderate to strong decisional regret (DRS score ≥26). There were no differences in total DRS scores by surgery type, previous surgical experience, consulting surgeon, parent education level, and decision to proceed with surgery. However, the DRS scores were higher in parents who experienced postoperative complications (mean difference, 8.88; 95% CI, −1.45-16.22; Mann-Whitney U P = .04).

Parent SDM-Q-9 and DRS were significantly negatively correlated (r_s[63] = −0.254; P = .045). However, physician SDM-Q-Doc and DRS were not significantly correlated (r_s[64] = −0.169; P = .18). The total DCS and DRS scores were also not significantly correlated (r_s[63] = 0.226; P = .08). Yet, when assessing DCS as a dichotomous variable (ie, those with DCS scores ≥25), there was a significant relation to DRS (Mann-Whitney U P = .02; effect size = 0.294).

Many pediatric conditions can improve over time without any significant intervention.¹ Although surgery may be helpful for these conditions, they are not without risks. Furthermore, there is a lack of unequivocal evidence for some treatment options, including surgery over watchful waiting or less invasive management options.^1,23 Hence, parents can be at risk of experiencing distress when considering surgery for their children. In the current study, over 25% of parents had clinically significant decisional conflict. Similar levels of parental decisional conflict have been reported in studies pertaining to elective pediatric surgeries.¹⁰ Given the high prevalence of decisional conflict, there is a need to better understand the etiology of this negative outcome and methods to reduce it.

There was no significant difference in decisional conflict scores across different surgical procedures. Therefore, despite the notion that tympanostomy tube insertion is less invasive than adenotonsillectomy, parents did not report less decisional conflict when considering the former surgery. Likewise, previous surgical experience for any child in the family did not predict lower levels of decisional conflict, suggesting that experience and familiarity with prior surgical decisions is not associated with a decrease in uncertainty about making later treatment decisions. All together, these findings indicate that regardless of the surgical procedure itself (minor surgery vs more invasive surgery) or the previous experience of undergoing a similar process, making proxy health care decisions can bring about uncertainty for some parents.

Parents’ decisional conflict was not related to which surgeon was involved in the consultation visit. This was a novel finding that could not be determined in our previous study due to the smaller sample size.¹⁰ Hence, although physicians have varied interaction styles and even dispense different content and amount of information when discussing the same surgical procedure,²⁴ it did not significantly impact parents’ level of decisional conflict. The implication may be that health care provider influences, at least in this relatively homogeneous sample of providers, are not as imperative as parents’ perceptions of shared decision making when considering decisional conflict as a measure of decision quality. Furthermore, parents also have their own decision-making style and history of interactions with the medical system that could be more important than physician interaction style when considering perceptions of shared decision making and decisional outcomes.

Another novel finding in this study was that parents who elected not to have surgery (ie, watchful waiting group) had significantly higher decisional conflict than parents who decided to proceed with surgery. The expectation may be that those parents who were more uncertain or conflicted about their decision tend to select watchful waiting rather than surgery. It is also possible that the decision to watch and wait may be more conflicting for parents; that is, they may feel more insecure about an unclear course of watchful waiting vs a certain course of surgery. These findings highlight that parents who choose watchful waiting are potentially at more risk of struggling with decisions about their child’s health care and require close follow-up.

Decisional regret is another negative outcome that can result from poor quality decision making. To date, only 2 other studies^4,15 have assessed the issue of decisional regret from the perspective of parents. In both studies, moderate to strong decisional regret was rarely reported. The level decisional conflict was also significantly positively correlated with decisional regret in both studies,^4,15 and 1 study also found a significant correlation between complications and decisional regret.⁴ Similar findings were observed in the current study with the majority of parents reporting no decisional regret and only 1 parent reporting moderate and strong decisional regret. Moreover, when DCS was considered as a dichotomous variable, the DRS was significantly different. That is, those parents with clinically significant decisional conflict had significantly higher DRS scores when compared with those parents without clinically significant decisional conflict. Unsurprisingly, parents who experienced postoperative complications had significantly higher DRS scores.

The significance of mild decisional regret is unclear. Although it was a common finding in this study and previous studies, the practical implication is unknown because mild decisional regret can result from the decision being made in the setting of imprecise or ambiguous information.²⁰ That is, mild decisional regret can be an expected finding in most medical decisions because there are many conditions in which there is a lack of unequivocal evidence for 1 treatment option. In contrast, moderate to strong decisional regret may affect future decision making in medicine. For example, regret has been associated with negative actions in studies pertaining to business and marketing.²⁵ Therefore, decisional regret in medicine may lead to actions, such as formal complaints or litigations, taken toward the physician or health care organization.²⁶ Since decisional regret research in medicine is in its infancy, more studies are required to understand the implications and methods to reduce this phenomenon.

Shared decision making has been shown to improve decisional outcomes.^2,27 A recent systematic review²⁸ showed that shared decision making improves quality of medical decisions and decreases decisional conflict in patients considering elective surgery. In the current study, parent ratings of shared decision making were significantly negatively correlated with total DCS and DRS scores. That is, parents who thought themselves to be more involved in the decision-making process had less decisional conflict and less decisional regret. These findings are consistent with our pilot study and also research in other areas.^10,13,25,29 Interestingly, physician ratings of shared decision making were only significantly correlated with DCS but not DRS scores. This indicates that parent perceptions of shared decision making is different than physician perceptions of the decision-making process, and parent perceptions are more likely to influence decisional outcomes.^10,15

The potential benefit of shared decision making may extend to level of health care systems or organizations because it has been shown to improve quality of care delivered and reduce unnecessary variation in care and costs across different health care regions. Specifically, shared decision making with the use of most up-to-date evidence and decision support tools (eg, decision aids) have been shown to reduce elective surgery rates, thereby decreasing the health care system costs.³⁰ In the United States, Section 3506 of the Patient Protection and Affordable Care Act specifically encourages the use of shared decision making in medicine to improve quality of care and reduce costs.³¹ Unfortunately, the practice of shared decision making has not been widely adopted in the clinical setting. Clearly, more effort is required to advance shared decision making in otolaryngology and beyond.

There are several limitations in this study. First, we included only a small number of health care providers, thus representing a restricted range of potential clinician influences. In addition, surgeon 1 had higher decisional conflict and regret scores reported by parents. However, there were no statistically significant differences in DCS and DRS scores between all of the surgeons and the confidence intervals overlapped. The second limitation pertains to the study population. Specifically, the study participants were relatively well educated and homogenous in terms of ethnicity (ie, mostly white). Also, not all participants completed the DRS; yet, there were no significant differences those who did and did not respond. Therefore, the current results may not be generalizable in other settings. Lastly, we did not measure the severity of disease or presentation, which may have influenced the level of decisional conflict and regret reported by the parents. This could be a major limitation since the decision to proceed with surgery may be easier for those with more severe diseases.³² However, previous research suggests that severity of presentation (or invasiveness of intervention) does not always lead to higher levels of decisional conflict or regret.^10,16 Also, perceptions of disease severity is not always an accurate indicator of actual disease severity.³³

To address these issues, future studies should include other centers and examine long-term outcomes of decisional conflict and decisional regret, including knowledge about the procedure, satisfaction, and other measures of decision quality. Future research should also measure decisional outcomes in relation to disease severity and clinical stratification. Finally, decision support tools, such as decision aids, should be developed to improve the decision-making process for parents and clinicians.

There was a subset of parents who experienced clinically significant decisional conflict when making decisions about their child’s health care. Although causation cannot be implied, the parents who perceived themselves to be more involved in the decision-making process reported lower levels of decisional conflict and regret. Efforts should, therefore, be made to increase parent involvement in the decision-making process when discussing elective surgery for their children. The development and use of decision support tools should be explored in otolaryngology as they have been found to improve decision quality.

Corresponding Author: Paul Hong, MD, FRCSC, IWK Health Centre, 5850/5920 University Ave, PO Box 9700, Halifax, NS B3K 6R8, Canada (paul.hong@iwk.nshealth.ca).

Accepted for Publication: September 3, 2016.

Published Online: December 8, 2016. doi:10.1001/jamaoto.2016.3365

Author Contributions: Dr Hong had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Hong, Ritchie, Chorney.

Acquisition, analysis, or interpretation of data: All Authors.

Drafting of the manuscript: Hong, Chorney.

Critical revision of the manuscript for important intellectual content: All Authors.

Statistical analysis: Maguire, Purcell, Ritchie, Chorney.

Obtained funding: Hong, Ritchie, Chorney.

Administrative, technical, or material support: Hong, Maguire, Ritchie.

Study supervision: Hong, Chorney.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.

Funding/Support: This study was supported by research grants from the Canadian Institutes of Health Research, Nova Scotia Health Research Foundation, and Dalhousie Department of Surgery awarded to Drs Hong and Chorney.

Role of the Funder/Sponsor: The funders/sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: We wish to express our thanks to all participants in this study, and clinic nurses (Lynn Cavanagh, RN; and Cathy Beaton-Campbell, RN) who helped with recruitment. Mss Cavanagh and Beaton-Campbell were not compensated for their contributions.