Abstract
Objective:
To quantitatively evaluate parental preferences for the various treatments for vesicoureteral reflux using crowd-sourced best-worst scaling, a novel technique in urologic preference estimation.
Methods:
Preference data were collected from a community sample of parents via two best-worst scaling survey instruments published to Amazon’s Mechanical Turk online community. Attributes and attribute levels were selected following extensive review of the reflux literature. Respondents completed an object case best-worst scaling exercise to prioritize general aspects of reflux treatments and multi-profile case best-worst scaling to elicit their preferences for the specific differences in reflux treatments. Data were analyzed using multinomial logistic regression. Results from the object-case provided probability scaled values (PSV) that reflected the order of importance of attributes.
Results:
We analyzed data for 248 and 228 respondents for object and multi-profile case BWS, respectively. When prioritizing general aspects of reflux treatment, effectiveness (PSV=20.37), risk of future urinary tract infection (PSV=14.85) and complication rate (PSV=14.55) were most important to parents. Societal cost (PSV=1.41), length of hospitalization (PSV=1.09), and cosmesis (PSV=0.91) were least important. Parents perceived no difference in preference for the cosmetic outcome of open versus minimally invasive surgery (p=0.791). Bundling attribute preference weights, parents in our study would choose open surgery 74.9% of the time.
Conclusions:
High treatment effectiveness was the most important and preferred attribute to parents. Alternatively, cost and cosmesis were among the least important. Our findings serve to inform shared parent-physician decision-making for vesicoureteral reflux.
Keywords: Vesico-ureteral reflux, pediatrics, patient preference
INTRODUCTION
Vesicoureteral reflux (VUR) is a common, morbid pediatric urologic condition. VUR affects 1-10% of all children and carries a substantial economic burden of more than $100 million per year in the United States alone.1 For many children, VUR has four reasonable treatment options: open ureteral reimplantation, minimally invasive (MIS) ureteral reimplantation, endoscopic injection of a sub/intraureteral bulking agent, or conservative therapy with continuous antibiotic prophylaxis with watchful waiting. In making treatment decisions for their children, parents must make value judgments that weigh treatment attributes such as effectiveness, complication rate, cost, and cosmetic outcome. However, studies have shown that the treating hospital, rather than parental preferences or clinical factors, is the primary driver of VUR treatment choice.2
Best-worst scaling (BWS) is a validated multi-characteristic preference evaluation method that has been increasingly used in health-preference research.3 By asking respondents to rank the extremes of their preferences, repeated BWS questions enable a full-ranking of the alternatives presented. Object case BWS asks respondents to evaluate general characteristics of treatment, called ‘attributes,’ (e.g., cost or risk of complications) without linking these characteristics to specific interventions. Alternatively, multi-profile case BWS estimates how aspects of an experimentally-designed treatment influence respondents’ relative preference for the overall treatment. The multi-profile case more closely mirrors clinical decision-making processes faced by patients by requiring respondents to evaluate tradeoffs between competing attributes.4
The objective of this study was to evaluate parental preferences regarding VUR treatment using object case and multi-profile case BWS. Our aim was to first use the object case to define the characteristics of VUR treatments that parents identify as most important. Then, we used the multi-profile case to evaluate how parents evaluate tradeoffs in treatment characteristics when actually choosing a treatment. Our secondary aim was to use the results from the multi-profile case BWS to estimate relative preferences for current VUR treatments. We hypothesized that treatment effectiveness and treatment complication rate would be the most important aspects of treatment.
METHODS
Object Case BWS
An object case BWS instrument was developed to identify respondents’ relative preferences for 13 treatment aspects referred to as ‘attributes’ (Supplementary Table 1). We presented respondents with 14 BWS questions, each including only a subset of four attributes (Supplementary Figure 1). The 4-attribute subset for each BWS question varied, forcing each parent to evaluate tradeoffs across all 13 attributes. Attributes were chosen based on clinical experience and prior qualitative and quantitative studies.5–8 In each question, respondents were asked to select the most and least important attribute in choosing a VUR treatment for their child. The attributes included in each BWS question were prepared according to an experimental design using Sawtooth MaxDiff. The experimental design was appropriately balanced and orthogonal, meaning each attribute appeared and co-appeared with each other attribute an equal number of times.
Data were analyzed using a multinomial logistic (MNL) regression model. MNL coefficients represent the intensity of importance for each attribute relative to an omitted attribute (long-term treatment failure). Because logit results do not convey an absolute measure of relative importance, but rather differences in importance across attributes, we transformed the coefficients into probability scaled values (PSVs), which allow relative comparisons across attributes.9 PSVs sum to 100 and represent the likelihood that a given attribute was chosen as ‘best’ in the experiment; PSVs are also ratio-scaled, i.e., an attribute with a PSV=20 is twice as preferred as an attribute with PSV=10. Wald Chi-squared tests were used to test MNL coefficient significance relative to zero and to one another.
Multi-profile Case BWS
The multi-profile case experiment investigated parental preferences in the context of experimentally-constructed VUR treatments. Parents were presented with ten BWS questions asking them to select the best overall treatment for their child, just as they would in a clinical setting, and also the worst treatment (Supplementary Figure 2). Each VUR treatment profile was described by six treatment attributes (effectiveness, societal cost, treatment complication rate, cosmesis/invasiveness, doctor recommendation, and hospitalization length if required), which were chosen based on those attributes typically presented to parents making treatment decisions in the clinic (Supplementary Figure 2).
The multi-profile case asked respondents to evaluate VUR treatment profiles defined in terms of the way treatments fulfill those aspects (‘attribute levels’). Attribute levels were selected to cover clinically-related values for the four treatment alternatives typically offered to parents based on extensive literature review (Table 1).10–22 The levels for attributes varied across treatment profiles following a balanced, orthogonal experimental design with known statistical properties developed using the Sawtooth choice-based conjoint tool.
Table 1.
Multi-profile case BWS attribute levels.
| Attribute | Levels |
|---|---|
| Effectiveness | 53%, 77%, 88%, 97% |
| Treatment complication rate | <1%, 4%, 10%, 13% |
| Societal cost | $1343, $6365, $7059, $9556 |
| Length of hospitalization | No visit to hospital, Outpatient hospitalization (no overnight), 1 day, 2 days |
| Doctor recommendation as best treatment | Yes, No |
| Cosmesis/invasiveness | No scar and no invasive procedure, No scar and endoscopic intervention with anesthesia, Minimally-invasive surgery with anesthesia and small scars (with image), Open surgery with anesthesia and large scar (with image) |
Results from the MNL model are log-odds, also called preference weights, indicating how the utility associated with a treatment changes as the levels of an attribute change.23 Differences in preference weight estimates for the levels in an attribute indicate the importance of changing a treatment attribute from one level to another. The difference between the most and least preferred levels of an attribute indicates the overall relative importance of the attribute, given the range of levels considered in the study.23
Predicted choice probabilities were calculated by adding the preference weights for the attribute levels that corresponded to each clinically available VUR treatment. Using the MNL probability density function we calculated the proportion of average parent respondents who would select each treatment profile, given the relative utility of each profile. 2-sided t-tests were used to test differences between coefficients in the MNL model. All analyses were performed using Sawtooth Lighthouse Studio 9.5.2 and SAS 9.4.
Survey Instrument
We provided each respondent with an introduction to VUR, its treatments, and possible complications using text, images, and video designed by a board-certified pediatric urologist. Parents were required to answer several basic attention questions during the course of the survey. We introduced the BWS questions by describing each treatment attribute and the range of attribute levels. Parents were first introduced to several example BWS questions of increasing complexity to ensure that they understood how to answer the questions. Finally, we collected basic demographic information. Survey instruments were completed by a pilot group of 15 parents on the Amazon Mechanical Turk (MTurk) online work interface and feedback was incorporated.
Data Source
We published the two final BWS survey instruments to the Amazon MTurk. MTurk offers a crowd-sourced, US national sample that provides results equally as reliable as traditional methods and has been validated in VUR experiments by our group.24–27 Members of the MTurk community who were over 18 years old, self-reported parents, and had a >95% approval rating were eligible to participate. Respondents were compensated $1 to complete each survey. MTurk account-specific and IP address restrictions ensured only one response per parent per survey. Respondents were excluded from analysis if they completed the survey instrument below a reasonable time threshold (4 minutes), as we considered these respondents to be non-attentive.
RESULTS
Demographics
In total, 248 and 228 parents completed the object case and multi-profile case BWS survey instruments, respectively, after excluding a total of 26 parents who did not meet response time threshold (Table 2). Overall, most parents were female (71.6%), married (74.2%), 30-44-year-old (59.5%), non-Hispanic whites (77.1%) who had no prior experience with VUR (81.7%). Most had an annual household income between $25,000 and $74,999 (57.1%) and completed at least a university degree (54.0%).
Table 2.
Parent respondent demographic characteristics.
| Characteristic | Response | Object Case (%) | Multi-profile Case (%) | Overall % | ||
|---|---|---|---|---|---|---|
| Sex | ||||||
| Male | 70 | (28.2) | 65 | (28.5) | 28.4 | |
| Female | 178 | (71.8) | 163 | (71.5) | 71.6 | |
| Race | ||||||
| White Non-Hispanic | 189 | (76.2) | 178 | (78.1) | 77.1 | |
| White Hispanic | 21 | (8.5) | 10 | (4.4) | 6.5 | |
| Black Non-Hispanic | 18 | (7.3) | 14 | (6.1) | 6.7 | |
| Black Hispanic | 2 | (0.8) | 4 | (1.8) | 1.3 | |
| Asian | 12 | (4.8) | 13 | (5.7) | 5.3 | |
| American Indian or Native Alaskan | 0 | (0) | 1 | (0.4) | 0.2 | |
| Native Hawaiian or Pacific Islander | 1 | (0.4) | 1 | (0.4) | 0.4 | |
| Multi-racial/Other | 5 | (2.0) | 7 | (3.1) | 2.5 | |
| Age | ||||||
| 18-29 | 55 | (22.2) | 41 | (18.0) | 20.2 | |
| 30-44 | 147 | (59.3) | 136 | (59.6) | 59.5 | |
| 45-60 | 40 | (16.1) | 41 | (18.0) | 17.0 | |
| 60 and older | 6 | (2.4) | 10 | (4.4) | 3.4 | |
| Annual household income | ||||||
| Less than $25,000 | 23 | (9.3) | 28 | (12.3) | 10.7 | |
| $25,000 to $49,999 | 77 | (31.0) | 64 | (28.1) | 29.6 | |
| $50,000 to $74,999 | 71 | (28.6) | 60 | (26.3) | 27.5 | |
| $75,000 to $99,999 | 41 | (16.5) | 45 | (19.7) | 18.1 | |
| $100,000 and greater | 36 | (14.5) | 31 | (13.6) | 14.1 | |
| Highest education completed | ||||||
| Less than high school | 0 | (0) | 1 | (0.4) | 0.2 | |
| High school/GRD | 61 | (24.6) | 54 | (23.7) | 24.2 | |
| Technical/trade school | 52 | (21.0) | 51 | (22.4) | 21.6 | |
| University degree | 93 | (37.5) | 87 | (38.2) | 37.8 | |
| Advanced masters/doctoral degree | 42 | (16.9) | 35 | (15.4) | 16.2 | |
| Marital Status | ||||||
| Single, never married | 34 | (13.7) | 32 | (14.0) | 13.9 | |
| Married | 183 | (73.8) | 170 | (74.6) | 74.2 | |
| Divorced, single | 25 | (10.1) | 18 | (7.9) | 9.0 | |
| Divorced, remarried | 1 | (0.4) | 0 | (0.0) | 0.2 | |
| Separated | 5 | (2.0) | 8 | (3.50) | 2.7 | |
| Region of Residence | ||||||
| Northeast | 45 | (18.1) | 38 | (16.7) | 17.4 | |
| South | 93 | (37.5) | 96 | (42.1) | 39.7 | |
| Midwest | 59 | (23.8) | 47 | (20.6) | 22.3 | |
| West | 51 | (20.6) | 47 | (20.6) | 20.6 | |
| Metropolitan Status | ||||||
| Urban | 53 | (21.4) | 48 | (21.1) | 21.2 | |
| Suburban | 119 | (48.0) | 116 | (50.9) | 49.4 | |
| Rural | 76 | (30.6) | 64 | (28.1) | 29.4 | |
| Previous experience with VUR? | ||||||
| Yes | 45 | (18.1) | 42 | (18.4) | 18.3 | |
| No | 203 | (81.9) | 186 | (81.6) | 81.7 | |
| Prior overnight hospitalization? | ||||||
| Yes | 219 | (88.3) | 205 | (89.9) | 89.1 | |
| No | 29 | (11.7) | 23 | (10.1) | 10.9 | |
| Prior surgery? | ||||||
| Yes | 187 | (75.4) | 174 | (76.3) | 75.8 | |
| No | 61 | (24.6) | 54 | (23.7) | 24.2 | |
Object Case BWS
Parents ranked treatment outcome measures, such as effectiveness of curing reflux (PSV 20.4), risk of future urinary tract infection (UTI) after treatment (PSV 14.9), risk of long-term treatment failure (PSV 14.6), and treatment complication rate (PSV 14.2) as the most important attributes (Table 3). Treatment invasiveness (PSV 7.6) and parent perception of their doctor’s reputation (PSV 7.2) were also relatively important attributes. Treatment effectiveness was ~20 times more important than the least important attributes (cosmesis, length of hospitalization, societal cost) (all p<01).
Table 3.
Object case BWS results.
| Attribute | Logit Coefficient | 95% CI** | Probability Scaled Value |
|---|---|---|---|
| Effectiveness in curing reflux | 1.313 | (1.140, 1.486) | 20.37 |
| Risk of future UTIs post-treatment | 0.054 | (−0.104, 0.212) | 14.85 |
| Risk of long-term treatment failure | 0.00* | N/A | 14.55 |
| Treatment complication rate | −0.056 | (−0.218, 0.105) | 14.22 |
| Invasiveness | −1.207 | (−1.370, −1.044) | 7.59 |
| Perceived doctor reputation | −1.224 | (−1.388, −1.060) | 7.50 |
| Doctor recommends as best treatment | −1.295 | (−1.461, −1.130) | 7.14 |
| Risk of antibiotic resistance | −1.387 | (−1.551, −1.223) | 6.69 |
| Delay to treatment | −2.813 | (−2.986, −2.640) | 2.04 |
| Daily burden of care post-treatment | −3.047 | (−3.223, −2.871) | 1.64 |
| Societal cost | −3.208 | (−3.385, −3.031) | 1.41 |
| Length of hospitalization | −3.485 | (−3.663, −3.306) | 1.09 |
| Cosmetic appearance | −3.670 | (−3.850, −3.489) | 0.91 |
Object case multinomial logistic regression model.
Reference coefficient in model.
95% CI- 95% confidence interval.
When responses were stratified by sex, women ranked invasiveness 3 spots higher, antibiotic resistance 3 spots lower, and delay to treatment 2 spots higher in importance when compared to men. However, the rank order of the four most important attributes was the same for both sexes. Parents with university and advanced degrees had similar preferences to those with less education, other than placing less importance on risk of future UTI (down 2 spots) and perceived doctor reputation (down 3 spots). Similarly, prior experience with VUR and income level did not change the rank order of the most important attributes.
Multi-profile Case BWS
Preference weights estimated by the MNL model followed the expected preference pattern within each attribute given the clinical impact of each attribute level, i.e. higher complication rates had smaller preference weights than lower complication rates (Table 4). Preference weights were also roughly linear in continuous variables like effectiveness (Supplementary Figure 3). The preference weight of cosmesis/invasiveness for endoscopic injection was significantly higher than those for open surgery and MIS (each p<0.05), while the preference weights for open surgery and MIS were no different from one another (p=0.791). Consistent with findings in the object case experiment, treatment effectiveness as a whole had the highest relative importance (50.7%) among the included attributes; complication rate (20.9%), doctor recommendation (12.7%), cosmesis/invasiveness (6.2%), length of stay (6.0%), and societal cost (3.5%) were relatively less important.
Table 4.
Multiprofile case BWS results.
| Attribute | Relative Importance (%)* | Level | Logit Preference Weight | 95% CI** | p-value |
|---|---|---|---|---|---|
| Effectiveness | 50.7 | ||||
| 53 percent | −1.754 | (−1.847, −1.660) | <.001 | ||
| 77 percent | −0.275 | (−0.341, −0.209) | <.001 | ||
| 88 percent | 0.552 | (0.481, 0.624) | <.001 | ||
| 97 percent | 1.476 | (1.395, 1.558) | <.001 | ||
| Societal cost | 3.5 | ||||
| $1,343 | 0.155 | (0.083, 0.227) | <.001 | ||
| $6,365 | −0.027 | (−.100, 0.047) | 0.307 | ||
| $7,059 | −0.034 | (−0.105, 0.038) | 0.260 | ||
| $9,556 | −0.094 | (−0.169, −0.019) | 0.019 | ||
| All complication rate | 20.9 | ||||
| <1 percent | 0.684 | (0.608, 0.760) | <.001 | ||
| 4 percent | 0.251 | (0.179, 0.323) | <.001 | ||
| 10 percent | −0.283 | (−0.356, −0.209) | <.001 | ||
| 13 percent | −0.652 | (−0.728, −0.577) | <.001 | ||
| Length of hospital stay | 6 | ||||
| No visit to hospital | 0.170 | (0.097, 0.242) | <.001 | ||
| Same day outpatient hospitalization | 0.039 | (−0.034, 0.111) | 0.228 | ||
| 1 day (overnight, home next day) | −0.041 | (−0.113, 0.031) | 0.212 | ||
| 2 days | −0.168 | (−0.237, −0.098) | <.001 | ||
| Doctor recommends as best treatment | 12.7 | ||||
| Yes | 0.402 | (0.359, 0.445) | <.001 | ||
| No | −0.402 | (−0.445, −0.359) | <.001 | ||
| Cosmetic appearance and Invasiveness | 6.2 | ||||
| No scar and no invasive procedure | 0.239 | (0.166, 0.311) | <.001 | ||
| No scar and endoscopic intervention with anesthesia | 0.058 | (−0.012, 0.128) | 0.105 | ||
| MIS with anesthesia and small scars | −0.152 | (−0.223, −0.080) | <.001 | ||
| Open surgery with anesthesia and large scar | −0.145 | (−0.218, −0.073) | <.001 |
Multi-profile case BWS multinomial logistic regression (MNL) model constructed using leveled attributes. Model coefficients are preference weights.
Constructed using expanded MNL model including interaction terms.
95% CI - 95% confidence interval. Significance testing conducted using 2-sided t-tests.
The estimated predicted choice probabilities (Supplementary Table 2) suggest that an average parent from our sample would select a treatment with characteristics similar to open surgery 74.9% of the time, compared to MIS, endoscopic injection, and antibiotic prophylaxis 9.6%, 8.9%, and 6.6% of the time, respectively.
DISCUSSION
Consistent with our hypothesis, we found that effectiveness in curing VUR and treatment complication rate were by far the most important treatment attributes, ~20 times more important than cosmesis or societal cost. Unsurprisingly then, we identified that parents associated the most highly effective and least complication-prone treatments with the greatest utility. Combining these preferences revealed that the average parent, when presented with information similar to that which we provided in this study, would choose open surgery in nearly 75% of similar clinical situations.
These quantitative findings are largely consistent with the previous largely qualitative literature examining parental preferences in VUR treatment. Of the 13 primary parental preference themes identified in parent interviews by Tran et al., the most commonly considered treatment themes were efficacy, future UTI prevention, daily care burden, antibiotic resistance, and invasiveness.5 Similar to our findings that cosmesis and cost are relatively less important attributes, only 19% of parents interviewed mentioned post-treatment cosmesis; treatment cost did not emerge as a primary theme.5
Parental preference for success rate over cosmesis has been previously documented in pediatric urology. Barbosa et al. noted that when parents were told that either robotic or open reimplantation was more effective, parents chose the ‘more effective’ option about 90% of the time.28 Interestingly, they also observed that parents preferred robotic scars to open scars, whereas we and others find no significant difference in scar preference or even preference for open cosmesis.7, 28
For other parents, invasiveness seems to influence treatment choice. Callaghan et al. found that success rate was “most important” in 90.4% of VUR cases treated by a successful open surgery, but in only 52% of cases treated by a successful endoscopic intervention.8 Instead, degree of invasiveness was “most important” in 84% of those cases. The authors speculate that the competing preference for effectiveness versus invasiveness was the factor driving this behavior but were unable to definitively identify the granular attributes that drove the holistic treatment choices they observed.8
Although parents clearly identified treatment effectiveness as the most important and desirable attribute in our study, translating this finding to a clinical recommendation is less straightforward. The difficulty arises primarily from variation in reported success rates for VUR treatments, a topic frequently debated over the last two decades in pediatric urology. While some institutions regularly report success rates for endoscopic injection and MIS >90%, others report rates far lower.10, 21 Classically thought to be a result of surgeon experience, this variation is likely also due to differences in patient follow-up and definitions of treatment success. Variation in success rates may explain why our overall treatment choice prediction deviates from observational data for endoscopic injection in recent clinical practice (~35% vs. 8.9% model), but is more consistent for open surgery (~60% vs. 74.9% model), which has a greater consensus success rate.29 With such variation in institutional VUR treatment experience, it is perhaps not surprising that institutional preferences, not clinical factors, have been identified as the primary determinant of VUR treatment choice.2, 30
Regardless, these experiments have allowed us to develop a preference model that might be applied to as a decision-aid for parents facing a VUR treatment decision. We intend for parents to complete this brief BWS exercise prior to seeing their child’s urologist. Then, the urologist can use the revealed preferences to better target face-to-face shared decision-making. Individual surgeons could incorporate their own success rates and surgical outcomes, further customizing the model. This future application of our model bridges a theoretical crowd-sourced study to direct patient care in the clinic.
The results of our study must be interpreted in the context of several limitations. First, we conducted our study using a sample from MTurk rather than parents of children with VUR. If actual parents have inherently different preference structures than those who receive only VUR instruction, then our results may not be generalizable to specific clinical scenarios. Our sensitivity analysis showed that parents in our study who had prior VUR knowledge had largely analogous preferences compared to those who had no prior experience. Further, in areas of more direct overlap our study finds agreement with previous research that sampled parents if children with VUR.5, 8 Nonetheless, our intended future application of similar BWS techniques to evaluate preferences in parents of children with VUR would increase the external validity of this research.
In addition, we presented parent respondents with the concept of societal cost in our survey instrument as opposed to out-of-pocket cost. Though we directly acknowledged to our survey respondents that societal cost was a potentially confusing concept, we did not receive any comments from respondents about the difficulty of considering societal cost. Besides interpreting societal cost appropriately, parents may have interpreted societal costs as out-of-pocket costs, leading our models to overestimate the effect of societal cost since societal costs are necessarily greater than out-of-pocket costs. Though the unimportance of cost we observed was noted in prior studies, we were still surprised by its magnitude, especially in a healthcare climate that is increasingly cost-conscious.5
Last, we presented parents with a general case of VUR, without in-depth discussion of clinical factors like VUR grade. We recognize that clinical factors impact prognosis, treatment effectiveness, complication risk, and treatment recommendation. Future investigation for specific grades of VUR with actual patients will inform the specific value of preference research in the clinical setting.
CONCLUSIONS
High treatment effectiveness was the most desirable treatment attribute to parents choosing a treatment for VUR, while complication rate and risk of long-term treatment failure were also relatively important. Cosmesis and societal cost were among the least important. Future research will determine how targeted discussion that aligns with parental preferences might improve shared decision-making for VUR treatment.
Supplementary Material
Acknowledgments
Support/Financial Disclosures
Funding Source: Mr. Dionise is supported in part by a Urology Care Foundation Summer Medical Student Fellowship from the American Urological Association (AUA). Dr. Routh is supported in part by grant K08-DK100534 from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). Neither funding source had any role in the collection, analysis, and interpretation of data; in the writing of the manuscript; or in the decision to submit the manuscript for publication.
Financial Disclosure: The authors have no financial relationships relevant to this article to disclose.
This study was granted approval by the institutional internal review board as an exempt study of full internal review board review on the basis that it involved no protected health information.
Abbreviation Key:
- VUR
vesicoureteral reflux
- MIS
minimally invasive surgery
- BWS
best-worst scaling
- MTurk
mechanical Turk
- MNL
multinomial logit
- PSV
probability scaled value
- UTI
urinary tract infection
- CBC
choice-based conjoint
- CI
confidence intervall
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of Interest: The authors have no conflicts of interest relevant to this article to disclose.
REFERENCES
- 1.Pohl HG, Joyce GF, Wise M, Cilento BG Jr Vesicoureteral reflux and ureteroceles. J Urol. 2007;177:1659–1666. [DOI] [PubMed] [Google Scholar]
- 2.Routh JC, Nelson CP, Graham DA, Lieu TA. Variation in surgical management of vesicoureteral reflux: influence of hospital and patient factors. Pediatrics. 2010;125:e446–451. [DOI] [PubMed] [Google Scholar]
- 3.Flynn TN, Louviere JJ, Peters TJ, Coast J. Best–worst scaling: what it can do for health care research and howto do it. J Health Econ. 2007;26:171–189. [DOI] [PubMed] [Google Scholar]
- 4.Chrzan K, Golovashkina N. An empirical test of six stated importance measures. Int J Mark Res. 2006;48:717–740. [Google Scholar]
- 5.Tran GN, Bodapati AV, Routh JC, Saigal CS, Copp HL. Parental Preference Assessment for Vesicoureteral Reflux Management in Children. J Urol. 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Krill AJ, Pohl HG, Belman AB, Skoog SJ, Snodgrass WT, Rushton HG. Parental preferences in the management of vesicoureteral reflux. J Urol. 2011;186:2040–2044. [DOI] [PubMed] [Google Scholar]
- 7.Garcia-Roig ML, Travers C, McCracken C, Cerwinka W, Kirsch JM, Kirsch AJ. Surgical Scar Location Preference for Pediatric Kidney and Pelvic Surgery: A Crowdsourced Survey. J Urol. 2017;197:911–919. [DOI] [PubMed] [Google Scholar]
- 8.Callaghan K, Gray E, Caldamone A, Ellsworth P. Factors involved in parental decision making for surgical correction of vesicoureteral reflux. J Urol. 2008;180:701–705; discussion 705-706. [DOI] [PubMed] [Google Scholar]
- 9.Yuan Z, Levitan B, Burton P, Poulos C, Brett Hauber A, Berlin JA. Relative importance of benefits and risks associated with antithrombotic therapies for acute coronary syndrome: patient and physician perspectives. Curr Med Res Opin. 2014;30:1733–1741. [DOI] [PubMed] [Google Scholar]
- 10.Routh JC, Inman BA, Reinberg Y. Dextranomer/hyaluronic acid for pediatric vesicoureteral reflux: systematic review. Pediatrics. 2010;125:1010–1019. [DOI] [PubMed] [Google Scholar]
- 11.Investigators RT, Hoberman A, Greenfield SP, et al. Antimicrobial prophylaxis for children with vesicoureteral reflux. N Engl J Med. 2014;370:2367–2376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wang HH, Tejwani R, Cannon GM Jr, Gargollo PC, Wiener JS, Routh JC. Open versus minimally invasive ureteroneocystostomy: A population-level analysis. J Pediatr Urol. 2016;12:232 e231–236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wang HS, Tejwani R, Wolf S, Wiener JS, Routh JC. Readmissions, unplanned emergency room visits, and surgical retreatment rates after anti-reflux procedures. J Pediatr Urol. 2017;13:507.e501–507.e507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wang HH, Gbadegesin RA, Foreman JW, et al. Efficacy of antibiotic prophylaxis in children with vesicoureteral reflux: systematic review and meta-analysis. J Urol. 2015;193:963–969. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Benoit RM, Peele PB, Cannon GM Jr., Docimo SG. The cost-effectiveness of dextranomer/hyaluronic acid copolymer for the management of vesicoureteral reflux. 2. Reflux correction at the time of diagnosis as a substitute for traditional management. J Urol. 2006;176:2649–2653; discussion 2653. [DOI] [PubMed] [Google Scholar]
- 16.Kurtz MP, Leow JJ, Varda BK, et al. Robotic versus open pediatric ureteral reimplantation: Costs and complications from a nationwide sample. J Pediatr Urol. 2016;12:408.e401–408.e406. [DOI] [PubMed] [Google Scholar]
- 17.Kasturi S, Sehgal SS, Christman MS, Lambert SM, Casale P. Prospective longterm analysis of nerve-sparing extravesical robotic-assisted laparoscopic ureteral reimplantation. Urology. 2012;79:680–683. [DOI] [PubMed] [Google Scholar]
- 18.Akhavan A, Avery D, Lendvay TS. Robot-assisted extravesical ureteral reimplantation: outcomes and conclusions from 78 ureters. J Pediatr Urol. 2014;10:864–868.24642080 [Google Scholar]
- 19.Grimsby GM, Dwyer ME, Jacobs MA, et al. Multi-institutional review of outcomes of robot-assisted laparoscopic extravesical ureteral reimplantation. J Urol. 2015;193:1791–1795. [DOI] [PubMed] [Google Scholar]
- 20.Smith RP, Oliver JL, Peters CA. Pediatric robotic extravesical ureteral reimplantation: comparison with open surgery. J Urol. 2011;185:1876–1881. [DOI] [PubMed] [Google Scholar]
- 21.Gundeti MS, Boysen WR, Shah A. Robot-assisted Laparoscopic Extravesical Ureteral Reimplantation: Technique Modifications Contribute to Optimized Outcomes. European urology. 2016;70:818–823. [DOI] [PubMed] [Google Scholar]
- 22.Herz D, Fuchs M, Todd A, McLeod D, Smith J. Robot-assisted laparoscopic extravesical ureteral reimplant: A critical look at surgical outcomes. J Pediatr Urol. 2016;12:402.e401–402.e409. [DOI] [PubMed] [Google Scholar]
- 23.Hauber AB, Gonzalez JM, Groothuis-Oudshoorn CG, et al. Statistical Methods for the Analysis of Discrete Choice Experiments: A Report of the ISPOR Conjoint Analysis Good Research Practices Task Force. Value Health. 2016;19:300–315. [DOI] [PubMed] [Google Scholar]
- 24.Lloyd JC, Yen T, Pietrobon R, et al. Estimating utility values for vesicoureteral reflux in the general public using an online tool. J Pediatr Urol. 2014;10:1026–1031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Nelson CP, Routh JC, Logvinenko T, et al. Utility scores for vesicoureteral reflux and anti-reflux surgery. J Pediatr Urol. 2015;11:177–182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Berinsky AJ, Huber GA, Lenz GS. Evaluating Online Labor Markets for Experimental Research: Amazon.com’s Mechanical Turk. Political Analysis. 2012;20:351–368. [Google Scholar]
- 27.Horton JJ, Rand DG, Zeckhauser RJ. The online laboratory: conducting experiments in a real labor market. Experimental Economics. 2011;14:399–425. [Google Scholar]
- 28.Barbosa JA, Barayan G, Gridley CM, et al. Parent and patient perceptions of robotic vs open urological surgery scars in children. J Urol. 2013;190:244–250. [DOI] [PubMed] [Google Scholar]
- 29.Michael Garcia-Roig CT, McCracken Courtney E, Kirsch Andrew J. National Trends in Management Of Primary Vesicoureteral Reflux in Children. The Societies for Pediatric Urology Fall Congress. Montreal, CA: 2017. [DOI] [PubMed] [Google Scholar]
- 30.Lee OT, Durbin-Johnson B, Kurzrock EA. Physician preference is a major factor in management of vesicoureteral reflux. Pediatric nephrology (Berlin, Germany). 2015;30:131–138. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.