What are parents really thinking? Using conjoint analysis to understand family priorities in pediatric orthopaedics
Location
LSU Health NO Center for Advanced Learning and Simulation (CALS)
Document Type
Event
Start Date
20-6-2026 8:15 AM
End Date
20-6-2026 8:45 AM
Publication Date
June 2026
Description
Introduction: Many pediatric orthopaedic treatment decisions involve trade-offs, but limited data exists on what families at the population-level actually prioritize when choosing between reasonable surgical options. We present two studies that used choice-based conjoint analysis to better understand family preferences in two areas of pediatric orthopaedics: adolescent medial patellofemoral ligament reconstruction (MPFL-R) and early-onset scoliosis (EOS) patients called “tweeners” in which both growth-sparing treatment and early definitive posterior spinal fusion may be reasonable options. We hypothesized that family preferences could be measured at the population level using choice-based conjoint analysis.
Methods: Two separate choice-based conjoint analysis studies were performed using Sawtooth Software (Lighthouse Studio Version 9.2.0) with participants recruited through Prolific. Institutional review board approval was obtained for both studies. In the MPFL-R study, anonymous participants were parents with children aged 13 to 17 years old (n=496). In the EOS “tweener” study, anonymous participants were parents with children aged 8 to 9 years (n=432). Participants in both studies chose between repeated hypothetical surgical scenarios with varying attributes based on the procedures being studied. Statistical analysis included Hierarchical Bayes estimation, Root Likelihood filtering, unpaired t-tests, one-way ANOVA, Brown-Forsythe and Welch-type ANOVA, Dunnett correction for post hoc comparisons, and Sawtooth simulator analysis.
Results: In the MPFL-R study, parents prioritized surgical success/avoiding revision surgery most highly, with an average importance score of 48.8% +/- 16.8%, followed by co-pay at 21.5% +/-14.1%, incision size at 17.4% +/- 14.6%, and time to return to sport at 12.3% +/- 7.3%. In the EOS “tweener” study, parents prioritized chance of complications most highly with a relative importance value of 36.9 +/- 15.2, followed by total number of surgeries 22.3 +/- 11.2, curve correction 21.0 +/- 12.8, increase in spinal height 11.8 +/- 9.1, and duration of treatment 8 +/- 5.4.
Discussion: Across two separate pediatric orthopaedic scenarios, parents consistently prioritized avoiding complications and repeat surgery. Importantly, family preferences varied by income level, parent gender, and child gender, reinforcing that shared decision-making should be individualized rather than uniform. Limitations include the use of anonymous crowdsourced participants rather than actual patients and families actively facing these surgical decisions, the hypothetical nature of the survey scenarios, and the fact that population-level preferences may not apply to every individual family. Our two studies provide a model for quantifying family preferences at the population level using choice-based conjoint analysis.
Significance/Clinical Relevance: Understanding how parents at the population level weigh priorities and balance trade-offs around complex surgical treatment options can aid pediatric orthopaedic surgeons in shared decision-making conversations in the clinic.
Recommended Citation
Rajendra, Ravi MD, "What are parents really thinking? Using conjoint analysis to understand family priorities in pediatric orthopaedics" (2026). Dept. of Orthopaedics: Robert D. D’Ambrosia Lectureship & Research Day. 5.
https://digitalscholar.lsuhsc.edu/ortho_rd/2026/chief/5
What are parents really thinking? Using conjoint analysis to understand family priorities in pediatric orthopaedics
LSU Health NO Center for Advanced Learning and Simulation (CALS)
Introduction: Many pediatric orthopaedic treatment decisions involve trade-offs, but limited data exists on what families at the population-level actually prioritize when choosing between reasonable surgical options. We present two studies that used choice-based conjoint analysis to better understand family preferences in two areas of pediatric orthopaedics: adolescent medial patellofemoral ligament reconstruction (MPFL-R) and early-onset scoliosis (EOS) patients called “tweeners” in which both growth-sparing treatment and early definitive posterior spinal fusion may be reasonable options. We hypothesized that family preferences could be measured at the population level using choice-based conjoint analysis.
Methods: Two separate choice-based conjoint analysis studies were performed using Sawtooth Software (Lighthouse Studio Version 9.2.0) with participants recruited through Prolific. Institutional review board approval was obtained for both studies. In the MPFL-R study, anonymous participants were parents with children aged 13 to 17 years old (n=496). In the EOS “tweener” study, anonymous participants were parents with children aged 8 to 9 years (n=432). Participants in both studies chose between repeated hypothetical surgical scenarios with varying attributes based on the procedures being studied. Statistical analysis included Hierarchical Bayes estimation, Root Likelihood filtering, unpaired t-tests, one-way ANOVA, Brown-Forsythe and Welch-type ANOVA, Dunnett correction for post hoc comparisons, and Sawtooth simulator analysis.
Results: In the MPFL-R study, parents prioritized surgical success/avoiding revision surgery most highly, with an average importance score of 48.8% +/- 16.8%, followed by co-pay at 21.5% +/-14.1%, incision size at 17.4% +/- 14.6%, and time to return to sport at 12.3% +/- 7.3%. In the EOS “tweener” study, parents prioritized chance of complications most highly with a relative importance value of 36.9 +/- 15.2, followed by total number of surgeries 22.3 +/- 11.2, curve correction 21.0 +/- 12.8, increase in spinal height 11.8 +/- 9.1, and duration of treatment 8 +/- 5.4.
Discussion: Across two separate pediatric orthopaedic scenarios, parents consistently prioritized avoiding complications and repeat surgery. Importantly, family preferences varied by income level, parent gender, and child gender, reinforcing that shared decision-making should be individualized rather than uniform. Limitations include the use of anonymous crowdsourced participants rather than actual patients and families actively facing these surgical decisions, the hypothetical nature of the survey scenarios, and the fact that population-level preferences may not apply to every individual family. Our two studies provide a model for quantifying family preferences at the population level using choice-based conjoint analysis.
Significance/Clinical Relevance: Understanding how parents at the population level weigh priorities and balance trade-offs around complex surgical treatment options can aid pediatric orthopaedic surgeons in shared decision-making conversations in the clinic.