A Review of Decision Aids for Patients Considering More Than One Type of Invasive Treatment

Abstract

With continuous advances in medicine, patients are faced with several medical or surgical treatment options for their health conditions. Decision aids may be useful in helping patients navigate these options and choose based on their goals and values. We reviewed the literature to identify decision aids and better understand the effect on patient decision-making. We identified 107 decision aids designed to help patients make decisions between medical treatment or screening options; 39 decision aids were used to help patients choose between a medical and surgical treatment, and five were identified that aided patients in deciding between a major open surgical procedure and a less invasive option. Many of the decision aids were used to help patients decide between prostate, colorectal, and breast cancer screening or treatment options. Although most decision aids were not associated with a significant effect on the actual decision made, they were largely associated with increased patient knowledge, decreased decisional conflict, more accurate perception of risks, increased satisfaction with their decision, and no increase in anxiety surrounding their decision. These data identify a gap in use of decision aids in surgical decision-making and highlight the potential to help surgical patients make value-based, knowledgeable decisions regarding their treatment.

Introduction

Making decisions between competing medical treatment options is difficult for both patients and providers.¹ Historically, a paternalistic approach has been applied, whereby patients deferred the decision to the treating provider. However, this approach fails to consider patient preferences, values, and belief systems.² Efforts to engage patients in making their own medical decisions rely on empowering patients with education and the necessary tools,³ as patients desire more information about treatment options and want to participate in the decision-making process.^4–6 As the number of possible treatment options has increased exponentially, patient involvement in decision-making has simultaneously become more important and more challenging. For patients choosing between two treatments, especially surgical options, providers must ensure that patients correctly estimate both benefits and harms of any particular treatment.⁷

Decision aids have emerged as an effective means of providing information to patients, improving their knowledge, decreasing their decisional conflict, and standardizing the transmission of information from provider to patient.^8,9 Despite these documented benefits, there has been relatively few applications of decision aids to surgical practice.¹⁰ Current decision aids that do incorporate surgical options primarily focus on the choice between surgery and watchful waiting.⁹ However, there are situations in which the decision to be made is one surgical option versus another. For instance, a patient with large abdominal aortic aneurysms (AAA) must consider the benefits and harms of undergoing an endovascular repair versus an open surgical repair.¹¹ The patient must weight these two options within the context of other comorbid conditions that may be present, and they must understand how their comorbidities may impact the postoperative course.¹² Decision aids directed at patients facing these challenging situations may allow for better alignment of patient preferences and expectations with the surgical treatment delivered.

The objective of this article was to review published medical and surgical decision aids to highlight the disparity in decision aids used in surgical versus medical decision-making and to identify the utility in using decision aids to improve patient outcomes with the goal of using this information to inform future practice in surgical decision-making. This was done by analyzing three mutually exclusive groups of decision aids, those considering medical (nonsurgical) decisions, those considering medical versus surgical interventions, and those considering two or more surgical interventions.

Methods

Study identification

We reviewed all randomized-controlled trials (RCTs) that studied the use of decision aids in medical decision-making as identified by the Cochrane collaboration in their systematic reviews published in 2014 and 2017 (n = 133).^13,14 We then performed a MEDLINE search using the terms “decision aid” and “surg” to ensure that we included any additional decision aids evaluated that were not included in the Cochrane review. The search was limited to RCTs from 2015 to 2017 that included use of a decision aid or decision tool. This search resulted in 18 additional studies not included in either Cochrane review.

We categorized the decision aids into three distinct groups based on the type of decision they were designed for: (1) choosing between medical (nonsurgical) options, (2) choosing between a medical and a surgical option, (3) choosing between a minimally invasive and major surgical option, and (4) choosing between a minimally invasive and major surgical option for patients with complex comorbidities, defined as choosing between nonelective surgical treatment options for patients with severe comorbidities that factor into decision-making.

Study outcomes assessed

We evaluated all outcomes measured in each of the studies. We paid specific attention to studies that measured the effect of the decision aid on the ultimate decision that was made by the patient, as well as the effect on patient-centered factors surrounding the decision. We categorized these patient factors into five domains: (1) patient knowledge of the condition or decision, (2) patient perception of risks related to the decision, (3) patient decisional conflict, (4) patient satisfaction with the decision aid and/or decision made, (5) and patient anxiety related to the decision. These five domains were defined by two of the study authors (KL, JC) in concordance with the study outcomes assessed in the majority of the research papers included in this review. These were adjudicated by review of the remaining authors.

Results

Summary of studies of decision aids

We identified 151 studies that evaluated the use of a decision aid for patients facing a treatment choice. We categorized these into three groups: the first group consisted of 107 studies that evaluated decision aids for patients choosing between two medical treatments or screening options, the second group consisted of 39 studies that evaluated decision aids for patients choosing between a medical and surgical treatment option, and the third group had five studies that examined the utility of decision aids to assist patients in choosing between two surgical options (Fig. 1). We did not identify any decision aids that were designed to help patients with complex comorbid conditions choose between two types of surgical interventions.

Fig. 1 -. Study flow diagram.

We further categorized each type of decision aid by medical condition (Table 1). Among decision aids designed for choosing between medical decisions, most were designed for patients with either screening for or genetic testing for the following conditions: prostate conditions (n = 18),^15–32 cardiovascular disease (n = 11),^33–43 colorectal cancer (n = 14),^44–57 breast cancer (n = 13),^58–70 endocrine disorders (n = 11),^1,71–80 and women’s health conditions (n = 10).^81–90 Few studies assessed decision aids for medical decisions in other categories such as screening for congenital disorders,^91–97 infectious diseases,^98–102 orthopedic conditions,^103–106 diabetes,^107–112 psychiatric conditions,^113–115 neurologic conditions,^116,117 dental health conditions,^118,119 and health behaviors.¹²⁰ Among those designed for choosing between a medical and surgical interventions, most were targeted toward patients with breast cancer (n = 9),^121–129 orthopedic conditions (n = 9),^130–138 prostate conditions (n = 8),^139–146 and women’s health conditions (n = 6),^147–152 whereas few assessed decisions regarding cardiovascular conditions,^153–155 vasectomy,¹⁵⁶ obesity,¹⁵⁷ pulmonary conditions,¹⁵⁸ and end-of-life care.¹⁵⁹ Among the five decision aids that were designed to help patients choose between two surgical options, three were targeted toward patients with breast cancer,^3,160,161 one was designed for patients with prostate cancer,¹⁶² and one was for patients with appendicitis.¹⁶³

Table 1 -.

Number of decision aids addressing each condition.

Condition type	Medical versus medical (n = 107)	Medical versus surgical (n = 39)	Minimally invasive versus major operations (n = 5)
Prostate conditions	18	8	1
Cardiovascular disease	11	3	-
Colorectal cancer	14	-	-
Breast cancer	13	9	3
Women’s health	10	6	-
Endocrine disorders	11	-	-
Congenital disorders	7	-	-
Infectious diseases	5	-	-
Orthopedic conditions	4	9	-
Diabetes	6	-	-
Psychiatric conditions	3	-	-
Neurologic conditions	2	-	-
Dental conditions	2	-	-
Health behaviors	1	-	-
Vasectomy	-	1	-
Acute appendicitis	-	-	1
Obesity	-	1	-
Pulmonary disorders	-	1	-
End-of-life care	-	1	-

Domains assessed in the decision aids by decision aid type

We categorized studies that assessed the efficacy of the decision aids in the five aforementioned patient-centered domains (knowledge, perception of risks, decisional conflict, satisfaction, and anxiety related to the decision, Tables 2–4, Fig. 2).

Table 2 -.

All studies that evaluated decision aids designed for deciding between two or more medical (nonsurgery) treatments/screening options, by year and outcomes measured.

Medical versus medical decisions (n = 107)
Condition	Decision	Authors	Outcomes measured
			Knowledge	Perceived risk	Decisionalconflict	Satisfaction	Anxiety	Other
Breast cancer	Genetic testing	Lerman, 1997	√	√
		Green, 2001	√
		Green, 2004	√	√	√	√	√
		Miller, 2005	√	√
		Schwartz, 2001	√	√
		Wakefield, 2008a	√
		Wakefield, 2008b	√
	Mammogram	Mathieu, 2007	√		√		√
		Mathieu, 2010	√
		Baena-Canada, 2015	√
	Tamoxifen treatment	Fagerlin, 2011		√	√
		Ozanne, 2007	√
	Chemotherapy	Whelan, 2003	√			√
Colorectal cancer	Screening	Dolan, 2002	√		√
		Lewis, 2010						√
		Miller, 2011						√
		Pignone, 2000						√
		Ruffin, 2007						√
		Schroy, 2011	√			√
		Smith, 2010	√		√		√
		Steckelberg, 2011	√
		Trevena, 2008	√
		Wakefield, 2008	√		√
		Wolf, 2000		√
		Ferron, 2015						√
		Schroy, 2016				√
	Chemotherapy	Leighl, 2011	√		√		√
Congenital disorders	Prenatal testing	Bekker, 2004	√	√	√		√
		Bjorklund, 2012	√
		Kuppermann, 2009	√	√	√	√
		Kuppermann, 2014	√	√	√
		Nagle, 2008	√		√		√
		Hunter, 2005	√		√	√	√
		Leung, 2004						√
Dental conditions	Dental treatment	Johnson, 2006	√			√	√
		Kupke, 2003	√			√
Diabetes	Medication	Mann D, 2010		√	√
		Weymiller, 2007	√
		Mullan, 2009	√			√
		Mathers, 2012	√	√	√
	Screening	Mann E, 2010	√
	Lifestyle modifications	Marteau, 2010						√
Health behaviors	Smoking cessation	Warner, 2015						√
Cardiovascular disease	Stress testing	Hess, 2012	√
	Prevention of cardiovascular disease	Lalonde, 2006	√	√	√	√
		Sheridan, 2006						√
		Sheridan, 2011						√
	Preoperative autologous blood donation	Laupacis, 2006	√	√		√
	Antithrombotic therapy in atrial fibrillation	Man-Song-Hing, 1999	√		√	√
		McAlister, 2005						√
		Thomson, 2007						√
		Fraenkel, 2012	√		√
	Hypertension treatment	Montgomery, 2003						√
	Angiography access	Schwalm, 2012	√		√
Endocrine disorders	Hormone replacement therapy	Deschamps, 2004			√	√
		Dodin, 2001	√				√
		Legare, 2003			√
		McBride, 2002		√
		Murray, 2001			√		√
		O’Connor, 1998	√	√	√
		O’Connor, 1999		√
		Rostom, 2002						√
		Rothert, 1997						√
		Schapira, 2007	√		√	√
	Thyroid cancer	Sawka, 2012	√		√
Infectious diseases	Vaccinations	Chambers, 2012						√
		Clancy, 1988						√
		Shourie, 2013			√
	Antibiotics for respiratory infections	Legare, 2011			√
		Legare, 2012						√
Neurologic conditions	Feeding in dementia	Hanson, 2011	√		√
	Immunotherapy for multiple sclerosis	Kasper, 2008						√
Orthopedic conditions	Osteoarthritis	Fraenkel, 2007						√
	Osteoporosis	Montori, 2011		√
		LeBlanc, 2015	√	√	√
	Bisphosphonates use	Oakley, 2006						√
Prostate conditions	Prostate cancer screening	Allen, 2010	√		√
		Evans, 2010	√		√		√
		Frosch, 2008	√		√
		Gattellari, 2003	√		√
		Gattellari, 2005	√		√
		Krist, 2007	√		√
		Lepore, 2012	√		√		√
		Myers, 2005						√
		Myers, 2011	√		√
		Partin, 2004	√
		Rubel, 2010	√		√		√
		Schapira, 2000	√
		Taylor, 2006	√		√	√
		Volk, 1999	√
		Volk, 2008			√
		Watson, 2006	√
		Williams, 2013	√		√
		Wolf, 1996						√
Psychiatric conditions	Treatment of schizophrenia	Hamann, 2006						√
	Treatment of depression	Loh, 2007				√
	Treatment of PTSD	Mott, 2014						√
Women’s health	Contraceptives after abortion	Langston, 2010						√
	Natural health products in menopause	Legare, 2008	√		√
	Management of abnormal cervical smear	McCaffery, 2010					√
	Delivery post caesarian section	Montgomery, 2007			√
		Eden, 2014	√		√
	Management of breech presentation	Nassar, 2007	√		√	√	√
	Management of menorrhagia	Protheroe, 2007	√		√		√
	Labor analgesia	Raynes-Greenow, 2010	√		√		√
	Ovarian cancer risk management	Tiller, 2006	√		√	√	√
	Number of embryos implanted in IVF	Van Peperstraten, 2010	√				√

Table 4 -.

All studies that evaluated decision aids designed for deciding between two surgical treatment options, by year and outcomes measured.

Minimally invasive uersus major operations (n = 5)
Condition	Decision	Authors	Outcomes measured
			Knowledge	Perceived risk	Decisional conflict	Satisfaction	Anxiety	Other
Breast cancer	Breast-conserving surgery uersus total mastectomy	Goel, 2001	√		√		√
		Jibaja-Weiss, 2011	√			√
		Whelan, 2004	√		√	√
Acute appendicitis	Open uersus laparoscopic appendectomy	Russell, 2015						√
Prostate conditions	Radical prostatectomy uersus orchiectomy uersus medical management	Auvinen, 2004						√

Fig. 2 -. Impact of decision aids on five patient-centered domains: (A) decisional conflict, (B) knowledge, (C) anxiety, (D) satisfaction, and (E) perception of risk. (Color version of figure is available online.).

Domain 1-knowledge

The effect of the decision aid on patient knowledge was assessed in 58% (87/151) of all studied decision aids. The majority reported a significant increase in patient knowledge related to choosing between medical decisions (89%, 58/65). This finding of enhanced patient knowledge was also noted in studies that examined decision aids for medical versus surgical decision (89%, 17/19) and for those designed to assist in deciding between two surgical options (67%, 2/3).

Domain 2-perception of risk

The perception of risk was assessed by 14% (21/151) of all studied decision aids. Studies that assessed this domain found that patients had a more accurate perception of the risks associated with their medical treatment after decision aid use (83%, 15/18). This finding was consistent among studies that assessed risk perception in medical versus surgical decision-making (67%, 2/3). No studies that evaluated decision aids for deciding between two surgical interventions examined patient risk perception.

Domain 3-ecisional conflict

Decisional conflict was assessed by 43% (65/151) of studies. Most decision aids designed for medical decisions reported significantly decreased patient decisional conflict (64%, 30/47), whereas the remaining studies reported no significant effect (36%, 17/47). Decreased decisional conflict was also found among studies that evaluated decision aids designed for choosing between a medical versus surgical option (63%, 10/16) and those designed for choosing between two surgical options (100%, 2/2).

Domain 4-patient satisfaction

Few studies assessed patient satisfaction with the decision-making process (22%, 33/151). However, most studies assessing this outcome found that patients using the medical decision aid had a higher level of satisfaction with the decision-making process after using the aid (67%, 12/18). Patients who were choosing between a medical versus surgical intervention (46%, 6/13) or choosing between two types of surgical interventions (50%, 1/2) also reported significantly increased patient satisfaction.

Domain 5-patient anxiety

Patient anxiety was assessed by only 19% (28/151) of studies. All studies showed either a significant decrease or no significant increase in patient anxiety surrounding the decision-making process or the decision itself. All decision aids that were designed for choosing between two types of medical decisions demonstrated decreased (5%, 1/19) or no significant increase (95%, 18/19) in patient anxiety. For decision aids that were designed for choosing between a medical and surgical intervention, many studies were associated with a significant decrease (25%, 2/8) or no significant increase (63%, 5/8), whereas one study showed a significant increase in patient anxiety after using the decision aid. Only one study assessed patient anxiety related to a decision aid designed for choosing between two surgical interventions, and that study found no significant increase in patient anxiety.

Impact of the decision aid on study outcomes by decision aid type

We next identified studies that assessed the impact of the decision aid on the actual decision made by the patient (Table 5). Of the 78% (83/107) of studies that assessed impact of the decision aid on the medical decisions made by the patient, 20% (17/83) demonstrated that the decision aid was associated with greater utilization or preference of one of the medical options; 22% (18/83) were associated with decreased utilization or preference; and 58% (48/83) of the decision aids were not associated with a significant impact on the decision made by the patient.

Table 5 -.

Number studies that assessed the impact of the decision aid on the decision.

Study characteristic	Medical versus medical (n = 107)	Medical versus surgical (n = 39)	Minimally invasive versus major operations (n = 5)
Evaluated the decision, n (%)	83 (78)	26 (67)	5 (100.0)
Greater utilization/preference for medical option (or minimally invasive option), n (%)	17 (20)	3 (12)	1 (20)
Lower utilization of medical option (or preference for major operation), n (%)	18 (22)	4(15)	3(60)
No impact on decision	48 (58)	19 (73)	1 (20)

Findings were similar among decision aids designed for choosing between a medical versus surgical intervention. Of the 67% (26/39) of studies that assessed the impact of the decision aid on the decision made, 73% (19/26) documented that the decision aid had no impact on the ultimate choice made by the patient, and 12% (3/26) and 15% (4/26) of studies were associated with increased preference for medical versus increased preference for surgical option, respectively.

All five studies that used decision aids to assist patients in choosing between two surgical interventions assessed the impact of the aid on the decision made by the patient. Almost all of these studies reported that the decision aid had a significant impact on which surgical intervention was chosen by the patient, with one favoring minimally invasive surgery, and three favoring major open surgery, and the remaining study reported no significant impact.

Discussion

After reviewing 151 randomized-controlled trials on the use of decision aids to assist patients in choosing between medical and surgical treatments, we identified a gap in the use of decision aids for surgical decision-making. The majority of decision aids were designed to assist patients choose between two or more types medical interventions (n = 107), with a smaller group aimed at assisting patients choose between a medical and surgical intervention (n = 39). Only five decision aids were identified that were specifically designed for patients choosing between two surgical options, namely deciding between a major operation and a minimally invasive procedure. However, no decision aids were available to assist patients with complex comorbid conditions attempting to navigate the choice between competing surgical treatment options. We believe that this gap represents an opportunity to develop decision aids for this particular group of patients, as they may benefit substantially from additional help in making challenging treatment decisions.

Many of the decision aids that we identified focused on decisions surrounding cancer screening and care. Specifically, genetic testing for breast cancer and screening for prostate cancer or colorectal cancer were among the most commonly studied medical decision aids.^{15–32,44–70} In addition, many of the decision aids designed for patients choosing between medical and surgical treatments also focused on oncologic decisions such as radiation versus radical prostatectomy for men with prostate cancer or mastectomy versus watchful waiting for women with genetic risk factors for breast cancer.^{121–129,139–146} This focus on cancer care was also present among decision aids designed for deciding between two surgical options.^3,160–162 Multiple studies sought to help with the decision between breast-conserving surgery versus total mastectomy for women with breast cancer.^3,160–162 Decision aids in cancer care help patients make difficult decisions where a “right” answer often does not exist, and therefore choices must be governed by individualized patient values. Similarly, there are difficult surgical treatment decisions to make with patients who have complex comorbidities. The added challenge of comorbid conditions and the resultant balance between risk and benefit for two surgical interventions must incorporate patient-specific values for the most appropriate treatment choice to be made.

Recent evidence has shown that many difficult medical decisions are deemed “preference-sensitive,” and the ultimate choice depends on each patient’s feelings, values, and their perceptions of the risks and benefits of each treatment option.² Many of the decision aids identified in this review were associated with significantly increased patient knowledge, increased satisfaction with the decision and decision-making process, a more accurate perception of the risks, no difference in patient anxiety, and decreased decisional conflict. Interestingly, most of the decision aids were not associated with a significant effect on the actual treatment decision made by the patient. While it seems evident that decision aids may provide knowledge to patients, it still remains unclear exactly what influences decisions for patients facing surgery. Moreover, challenges remain as to how surgeons can best help patients align their decisions with their personal values. We advocate for the utility of decision aids in helping patients with complex comorbidities choose between surgical treatment options that more closely align with their personal values to ultimately improve patient satisfaction and outcomes.

It is worth noting, however, that of the five surgical decision aids, three demonstrated that patients were more likely to choose a major operation over a minimally invasive procedure. Specifically, one study showed that women were more likely to choose mastectomy over breast-conserving surgery,³ another demonstrated that parents were more likely to choose open appendectomy than laparoscopic surgery for their children,¹⁶³ and men were more likely to choose radical prostatectomy over orchiectomy and medical management for prostate cancer after decision aid use.¹⁶² Conversely, the two remaining studies found that women were more likely to choose breast-conserving surgery over mastectomy¹⁶¹ or demonstrated no difference in breast cancer treatment choice.¹⁶⁰ With these data, we conjecture that decision aids may be useful in elucidating factors that drive patients to choose certain treatments, such as financial concerns, surgical outcomes, recurrence risk, or aesthetic or functional concerns.

Our study has limitations. Although we included a broad array of studies, it is possible that some decision aids may have been missed by our search strategy. In addition, differences in the outcomes evaluated precluded the ability to perform formal meta-analyses and express measures of heterogeneity. The small sample size of surgical decision aids (n = 5) in our review limits our ability to draw significant generalizations. However, we believe that this small number highlights the opportunity to increase the utilization of decision aids in surgical practice. Furthermore, we believe that the data gained from this review can provide insight into the utility of decision aids in future surgical practice and may improve patient outcomes by promoting decision-making in congruence with patient values.

Conclusions

Decision aids are a commonly utilized tool in helping patients make the best decisions about their health care, and these tools have been used most widely in studies examining two or more medical treatments and those examining medical versus surgical treatments. However, decision aids designed to assist patients in choosing between competing surgical interventions remain less common and represents an important pathway forward in helping patients make the best quality decisions related to their surgical care.

Table 3 -.

All studies that evaluated decision aids designed for deciding between a medical versus surgical treatment options, by year and outcomes measured.

Medical versus surgical decisions (n = 39)
Condition	Decision	Authors	Outcomes measured
			Knowledge	Perceived risk	Decisional conflict	Satisfaction	Anxiety	Other
Breast cancer	Breast reconstruction	Heller, 2008	√			√	√
		Lam, 2013			√
		Causarano, 2015			√	√
		Luan, 2016			√		√
	Prophylactic mastectomy	Schwartz, 2009					√
		Van Roosmalen, 2004	√	√		√
	Treatment options	Street, 1995	√
		Vodermaier, 2009	√		√	√
		Lam, 2014	√		√		√
Cardiovascular disease	Repair of AAA	Knops, 2014	√		√	√	√
	Coronary	Bernstein, 1998	√			√
	revascularization
		Morgan, 2000	√			√
End-of-life care	Curative intent surgery versus noncurative intent treatment	Schubart, 2015						√
Obesity	Bariatric surgery	Arterburn, 2011	√		√
Orthopedic conditions	Management of osteoarthritis	De Achaval, 2012			√
		Veroff, 2013						√
		Bozic, 2013						√
		Stacey, 2014	√
		Stacey, 2015						√
		Vina, 2016						√
	Elective spinal	Deyo, 2000				√
	surgery
	Treatment for spinal	Kearing, 2016	√			√
	stenosis
	Repair of humerus fractures	Hageman, 2015						√
Prostate conditions	Management of BPH	Barry, 1997	√			√
		Murray, 2001			√		√
	Management of prostate cancer	Berry, 2013			√
		Chabrera, 2015	√		√	√
		Davison, 1997					√
		Huber, 2013	√			√	√
		Van Tol-Geerdink, 2013						√
		Van Tol-Geerdink, 2016						√
Pulmonary disorders	Lung transplant in cystic fibrosis	Vandemheen, 2009	√	√	√
Women’s health and vasectomy	Hysterectomy or medical management of menorrhagia	Kennedy, 2002						√
		Vuorma, 2003						√
	C-section uersus vaginal delivery	Shorten, 2005	√		√
	Uterine fibroid treatment	Solberg, 2010	√			√
	Pregnancy termination methods	Wong, 2006	√	√	√
	Treatment of pelvic organ prolapse	Brazell, 2015			√
	Vasectomy	Labrecque, 2010	√		√