Skip to main content
. 2022 May 2;5(5):e229661. doi: 10.1001/jamanetworkopen.2022.9661

Table 2. Results From Multivariable Difference-in-Differences Analyses Comparing Patients Undergoing SG vs RYGB Between 2008 and 2016 Across Categories of Ambulatory Care Costs and Usea.

Cost category Postoperative year 1, SG (n = 3049) vs RYGB (n = 3251) Postoperative year 2, SG (n = 1952) vs RYGB (n = 2173) Postoperative year 3, SG (n = 1054) vs RYGB (n = 1151) Postoperative year 4, SG (n = 514) vs RYGB (n = 643)
Absolute difference (95% CI), $b Relative difference, % (95% CI)c Absolute difference (95% CI), $b Relative difference, % (95% CI)c Absolute difference (95% CI), $b Relative difference, % (95% CI)c Absolute difference (95% CI), $b Relative difference, % (95% CI)c
Total ambulatory care costs −421.9 (−1334.1 to 490.4) −5.3 (−16.1 to 5.6) −642.9 (−1643.4 to 357.5) −8.4 (−20.5 to 3.7) −138.9 (−1273.2 to 995.4) −2.0 (−17.8 to 13.9) 382.8 (−1366.0 to 2131.6) 5.1 (−19.2 to 29.5)
Prescription drug costs 36.1 (−252.0 to 324.3) 1.7 (−11.9 to 15.2) 173.3 (−80.1 to 426.7) 9.7 (−5.6 to 25.1) 140.2 (−283.0 to 563.3) 7.1 (−15.6 to 29.9) 852.8 (395.6 to 1310.0)d 52.4 (17.8 to 87.0)e
Office visit costs −36.7 (−103.5 to 30.1) −4.1 (−11.2 to 3.1) −14.4 (−85.4 to 56.5) −1.7 (−10.1 to 6.6) −2.0 (−87.1 to 83.1) −0.3 (−11.1 to 10.6) −43.8 (−176.1 to 88.5) −5.3 (−20.5 to 10.0)
Laboratory costs −118.9 (−220.2 to −17.5)f −13.9 (−24.2 to −3.6)e −107.6 (−197.2 to −18.1)f −15.9 (−27.4 to −4.5)e −106.7 (−244.5 to 31.1) −18.0 (−37.5 to 1.4) −4.0 (−116.7 to 108.7) −0.8 (−23.6 to 22.0)
Radiology costs −4.6 (−97.7 to 88.5) −0.7 (−15.5 to 14.1) −63.6 (−189.7 to 62.6) −9.2 (−26.0 to 7.6) 6.7 (−121.7 to 135.0) 1.1 (−20.3 to 22.5) 22.5 (−150.8 to 195.7) 3.4 (−23.7 to 30.6)
All other outpatient costs −264.7 (−945.4 to 416.1) −7.7 (−26.1 to 10.7) −600.9 (−1432.8 to 231.0) −16.4 (−35.7 to 3.0) −164.6 (−1103.8 to 774.6) −5.1 (−33.2 to 22.9) −537.1 (−1900.9 to 826.7) −13.9 (−45.1 to 17.2)

Abbreviations: RYGB, Roux-en-Y gastric bypass; SG, sleeve gastrectomy.

a

Difference-in-differences analyses adjusting for all matched variables were used to generate between-group estimates for change in each outcome, at each time period, relative to a preoperative baseline period spanning months −24 to −13 before the index surgical date. Year −2 was selected as the preoperative baseline for comparison because the year immediately before surgery represents a very high utilization time, owing to preoperative workup.

b

Absolute differences in each period refer to the estimated actual change in costs among patients undergoing SG minus those among patients undergoing RYGB in that segment vs baseline (months −24 to −13), accounting for all other periods.

c

Relative differences in each period refer to the estimated relative difference between SG and RYGB groups vs baseline, accounting for all other periods.

d

P < .001.

e

P < .01.

f

P < .05.