Composite endpoint of reduction from baseline in VAS back pain score by at least 20 mm, lack of device-related serious adverse events, absence of neurologic worsening and absence of surgical re-intervention. Note that the primary endpoint analysis is **intent to treat**, meaning that an outcome (success or failure) is assigned to all subjects randomized and treated. Subjects who withdrew early were deemed study failures.

A modified intent-to-treat approach was used.

Improvement in SI joint pain VAS score of greater than or equal to 20 points, at post-operative & NSM visit after 1 month. The Visual Analog Scale (VAS) is a 100 mm line on which the subject indicates their level of pain. 0 = no pain. 100 = worst imaginable pain. Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

2 iFuse and 1 NSM subjects did not complete the pain score.

Improvement in SI joint pain VAS score of greater than or equal to 20 points, at post-operative & NSM visits after 3 months. The Visual Analog Scale (VAS) is a 100 mm line on which the subject indicates their level of pain. 0 = no pain. 100 = worst imaginable pain. Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

2 iFuse and 3 NSM subjects did not complete the 3-month pain score.

Improvement in SI joint pain VAS score of greater than or equal to 20 points, at post-operative & NSM visits after 6 months. The Visual Analog Scale (VAS) is a 100 mm line on which the subject indicates their level of pain. 0 = no pain. 100 = worst imaginable pain. Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

1 iFuse and 3 NSM subjects did not complete the 6-month pain score.

Improvement in SI joint pain VAS score of greater than or equal to 20 points compared to baseline. The Visual Analog Scale (VAS) is a 100 mm line on which the subject indicates their level of pain. 0 = no pain. 100 = worst imaginable pain. Note that secondary endpoint analysis is based on available data only. Subjects in the NSM group who crossed over are considered failures for this endpoint by definition.

2 iFuse subjects did not complete the 12-month pain score. Analysis of 12-month scores in the NSM group is not valid because of the high crossover rate.

Improvement in SI joint pain VAS score of greater than or equal to 20 points compared to baseline. The Visual Analog Scale (VAS) is a 100 mm line on which the subject indicates their level of pain. 0 = no pain. 100 = worst imaginable pain. Note that secondary endpoint analysis is based on available data only. Subjects in the NSM group who crossed over are considered failures for this endpoint by definition.

12 iFuse subjects did not complete the 24-month pain score. Analysis of 24-month scores in the NSM group is not valid because of the high crossover rate.

Improvement in ODI score of greater than or equal to 15 points, at month 1.

Oswestry Disability Index is a validated measure of disability related to low back pain. There are 10 sections, each with a score between 0-5. Scores are expressed on a percent basis without using the percent term. Scores range from 0 (no disability) to 100 (completely disabled).

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

2 iFuse and 1 NSM subjects did not complete ODI at 1 month

Improvement in ODI score of greater than or equal to 15 points, at month 3.

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

2 iFuse subjects and 3 NSM subjects did not complete ODI at month 3.

Improvement in ODI score of greater than or equal to 15 points, at post-operative visits. 6 month visit.

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

1 iFuse and 2 NSM subjects did not complete ODI at month 6.

Improvement in ODI score of greater than or equal to 15 points compared to baseline. Oswestry Disability Index is a validated measure of disability related to low back pain. Subjects in the NSM group who crossed over are considered failures for this endpoint by definition.

2 iFuse subjects did not complete ODI at month 12. ODI analysis is not valid in NSM group at 12 months because of high crossover.

Improvement in ODI score of greater than or equal to 15 points compared to baseline. Oswestry Disability Index is a validated measure of disability related to low back pain. Subjects in the NSM group who crossed over are considered failures for this endpoint by definition.

The Count of participants = number of subjects that had a threshold change in ODI score of greater than or equal to 15 points. Overall number of participants = number of subjects analyzed. Of the 102, a total of 13 iFuse Implant subjects exited the study and 1 did not complete the ODI survey. By 6 months, 39 NSM subjects crossed over to surgical.

Improvement in quality of life (QOL) as measured by SF-36 PCS (Physical Component) at post-operative / NSM visits. The Short Form 36 health survey (SF-36) is a 36-item patient reported health questionnaire to measure quality of life across 8 domains. The PCS is the Physical Component Summary score. PCS is normed, so that "normal" scores are 50 +- 10. Higher scores indicate higher quality of life; lower scores indicate lower quality of life. SF-36 PCS (NBS, 2009 norms) ranges from 5 (minimum value, poor physical function) to 80 (maximum, excellent clinical function).

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

2 iFuse and 2 NSM subjects did not complete SF-36 at month 6.

Improvement in quality of life (QOL) as measured by SF-36 PCS (Physical Component) at post-operative / NSM visits. The Short Form 36 health survey (SF-36) is a 36-item patient reported health questionnaire to measure quality of life across 8 domains. The PCS is the Physical Component Summary score. PCS is normed, so that "normal" scores are 50 +- 10. Higher scores indicate higher quality of life; lower scores indicate lower quality of life. SF-36 PCS (NBS, 2009 norms) ranges from 5 (minimum value, poor physical function) to 80 (maximum, excellent clinical function).

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

A total of 2 iFuse subjects had exited the study before making it to their 12 month visit. This outcome is not evaluated in the NSM group at month 12 because high crossover to surgical treatment prevents valid analysis.

Improvement in quality of life (QOL) as measured by SF-36 PCS (Physical Component) at post-operative / NSM visits. The Short Form 36 health survey (SF-36) is a 36-item patient reported health questionnaire to measure quality of life across 8 domains. The PCS is the Physical Component Summary score. PCS is normed, so that "normal" scores are 50 +- 10. Higher scores indicate higher quality of life; lower scores indicate lower quality of life. SF-36 PCS (NBS, 2009 norms) ranges from 5 (minimum value, poor physical function) to 80 (maximum, excellent clinical function).

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

Change in QOL score at 24 months. 89 participants had QOL information at 24 months. Analysis of the NSM cohort is not done because high crossover to surgery prevents valid analysis.

Improvement in quality of life (QOL) as measured by EQ-5D (EuroQol-5D) at post-operative visits. EQ-5D is a five-question broad quality of life measure that can be combined into a single index and represents the time trade-off (TTO) utility of current health. A score of 0 would = worst imaginable health, while a score of 1.0 would be best imaginable health.

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

Results show improvement in score from baseline.

Improvement in quality of life (QOL) as measured by EQ-5D (EuroQol-5D) at post-operative visits. EQ-5D is a five-question broad quality of life measure that can be combined into a single index and represents the time trade-off (TTO) utility of current health. A score of 0 would = worst imaginable health, while a score of 1.0 would be best imaginable health.

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

2 iFuse subjects did not complete the survey at month 12. Due to high crossover rate to surgery, analysis in the NSM group is not valid.

Improvement in quality of life (QOL) as measured by EQ-5D (EuroQol-5D) at post-operative visits. EQ-5D is a five-question broad quality of life measure that can be combined into a single index and represents the time trade-off (TTO) utility of current health. A score of 0 would = worst imaginable health, while a score of 1.0 would be best imaginable health.

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

13 iFuse subjects did not complete the 24-month assessment. Analysis was not done for the NSM group at 24 months due to the high crossover rate.

Time to full ambulation among those without full ambulation at baseline.

60 days was the median of time to full ambulation for the iFuse implant System arm.

Number of days to full ambulation among those without full ambulation at baseline. At baseline only 13 in the surgical arm were not Ambulatory without assistance and for the NSM group, it was only 5 at baseline.

Proportion of non-working (due to back pain or other reasons) subjects who return to work

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

Patients not working at baseline due to back or other pain. Note this is a subset of the entire population.

Non-working subjects (due to back pain or other reasons) who return to work

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

Patients not working at baseline due to back or other pain

Non-working subjects who return to work

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

Patients not working at baseline due to back or other pain

Non-working subjects who return to work

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

Patients not working at baseline due to back or other pain

Non-working subjects who return to work

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

Patients not working at baseline due to back or other pain

Non-working subjects who return to work

Note that secondary endpoint analysis is based on available data only. No imputation of missing scores was prespecified in the protocol.

Patients not working at baseline due to back or other pain

Any event meeting ISO 14155 definition for serious adverse event at following time points: during procedure (if randomized to iFuse), hospital discharge (if iFuse, typically 1-2 days), and 1, 3, 6, 12, 18 and 24 months after randomization.

All treated study subjects.