The spine is one of the most common sites of metastatic disease, contributing significant morbidity and negatively impacting the quality of life of many cancer patients.1 Estimates for spine metastasis range from 50% to 70% of all patients with cancer, with these patients commonly experiencing pain, as well as disturbances in physical function and mental health.2,3 Considering both the widespread prevalence and life-altering sequelae of metastatic spine disease, it is important for physicians to assess and follow clinical outcomes that matter most to the patient. Patient-reported outcome measures (PROMs) are an increasingly used tool within the field of orthopedics that provide valuable insight into patients’ responses to treatment and progress in navigating a variety of health conditions.
Disease specific PROMs validated for spine patients include the Neck Disability Index (NDI), Oswestry Disability Index (ODI), Scoliosis Research Society (SRS) Questionnaires, and the Zuriq Claudication Questionnaire.4–8 In 2010, the Spine Oncology Study Group-Outcomes Questionnaire (SOSG-OQ) was designed to specifically measure the quality of life in patients with meta-static spine disease.9 This PROM was validated in 201510 and has since been recommended for use in clinical practice.11 However, the use of a general symptom-based PROM, such as Patient-Reported Outcomes Measurement Information System (PROMIS), may reduce both patient and physician burden and improve interdisciplinary care if shown to be concurrently valid. PROMIS was developed in 2004 under an initiative supported by the National Institutes of Health with the goal of capturing physical function (PF), pain interference (PI), and mental health domains across a variety of conditions.12 In addition to its ability to capture a wide range of disease states, the universal PROMIS instrument may allow for the evaluation of more complex patients with multiple comorbidities, who often cannot be adequately evaluated with spine-specific PROMs.8
In previous studies, PROMIS has been found to correlate with a number of commonly used spine relevant PROMs, including the NDI and ODI,4 as well as the SRS question-naires.5 In 2017, Pereira et al11 evaluated the ODI, NDI, PROMIS PF, PROMIS PI, EuroQol-5 Dimensions (EQ-5D), and the SOSG-OQ in patients with metastatic spine disease. SOSG-OQ was found to correlate strongly with PROMIS PF.11 However, PROMIS Depression was not studied. The aim of this study was to evaluate the PROMIS PF, PI, and Depression domains in comparison to the SOSG-OQ in patients with metastatic spine disease.
MATERIALS & METHODS
This study was approved by the Institutional Review Board at our institution. We prospectively enrolled metastatic spine tumor patients from January 1,2017 to July 2021 at a single, urban academic medical center with any primary cancer type. Patients were asked to complete the SOSG-OQ questionnaire in paper form during their visit. As part of our routine clinical workflow, patients received and completed the PROMIS domains on an Apple iPad (Apple, Cupertino, CA). Responses to both tools were reviewed retrospectively. Only patients who completed the SOSG-OQ and all three PROMIS domains were included in the analysis. Additionally, a retrospective chart review was conducted to obtain additional patient characteristics including age, gender, race, type of primary tumor and the treatment of metastatic spine disease.
The SOSG-OQ questionnaire consists of 27 items divided into six domains: Physical Function, Neurological Function, Pain, Mental Health, Social Function, and Posttherapy (Supplemental file, https://links.lww.com/BRS/B850). Each question is scored on a five-point scale, with higher scores typically indicating worse quality of life. However, questions 12, 15, 19, and 20 were scored inversely due to the reversed nature of the answers, as previously described.11 For example, in response to question 12, “how confident do you feel about your ability to manage your pain on your own?”, a four (“mostly confident”) would be scored as a two, whereas a one (“not confident at all”) would be scored as a five.11 The maximum number of points is 135. While the SOSG-OQ questionnaire is static in that the same 27 questions are given to each patient, PROMIS is completed using an iterative Computerized Adaptive Testing (CAT) system, based on item response theory that allows more flexibility and efficiency.13 Furthermore, PROMIS scores are normed to have a t-score of 50 with a standard deviation of 10.4 Higher scores in the PROMIS PF domain indicate better physical function, whereas lower scores on the PROMIS PI and PROMIS Depression domains indicate decreased pain and depressive symptoms, respectively.
Statistical Analysis
Spearman correlation (ρ) coefficients were calculated to assess concurrent validity. Following the correlation groupings of a previous PROMIS correlation study,14 we similarly defined the strength of correlations as follows: excellent (>0.7), excellent-good (0.61–0.70), good (0.4–0.6), and poor (<0.4). All statistical analyses were performed with Stata statistical software (v 16.1; StataCorp). Significance was set at P < 0.05.
RESULTS
Demographics
A total of 103 unique visits, representing 79 patients met our inclusion criteria. A majority were male (59; 57%), Caucasian (93; 90%), and the average age was 64-years-old (range: 34–87) (Table 1). Nine patients were excluded for missing one or more questions from the SOSG-OQ and another nine patients were excluded for not having PROMIS scores.
TABLE 1 -
Patient Characteristics
∗
| Demographic |
Mean [SD] [Range] or n [%] |
| Age |
64 (13) (34–87) |
| Gender |
| Male |
59 (57) |
| Female |
44 (43) |
| Race |
| White or Caucasian |
93 (90) |
| Black or African American |
10 (10) |
| Tumor type |
| Multiple myeloma |
28 (27) |
| Breast |
26 (25) |
| Prostate |
13 (13) |
| Renal cell |
10 (9.7) |
| Lung |
7 (6.8) |
| Colon |
5 (4.9) |
| Thyroid |
4 (3.9) |
| Large B cell lymphoma |
3 (2.9) |
| NonHodgkin lymphoma |
2 (1.9) |
| Esophageal |
2 (1.9) |
| Plasmacytoma |
1 (0.97) |
| Metastatic paraganglioma |
1 (0.97) |
| Metastatic spindle cell sarcoma |
1 (0.97) |
∗103 total visits, representing 79 unique patients.
Tumor Type
There were 13 different types of histologies reported, with multiple myeloma, breast cancer, and prostate cancer representing 28 (27%), 26 (25%), and 13 (13%), respectively. Additional cancers included renal cell carcinoma, lung cancer, colon cancer, thyroid cancer, large B-cell lymphoma, non-Hodgkin lymphoma, esophageal cancer, plasmacytoma, metastatic paraganglioma, and metastatic spindle cell sarcoma.
Treatments
There were five different treatment categories. Four patients (3.9%) underwent radiation therapy followed by surgery, seven (6.8%) underwent surgery only, 21 (20%) underwent surgery followed by radiation therapy, 46 (45%) only had radiation therapy, and 12 (12%) underwent cement augmentation (kyphoplasty) (Table 2).
TABLE 2 -
Treatments Summary
| Treatment |
n [%] |
| Prior radiation therapy followed by surgery |
4 (3.9) |
| Surgery only |
7 (6.8) |
| Surgery followed by radiation therapy |
21 (20) |
| Radiation therapy only |
46 (45) |
| Cement augmentation (kyphoplasty) |
12 (12) |
PROMIS and SOSG Scores
Several patients had multiple visits throughout the study period during which both PROMIS and SOSG-OQ questionnaire were completed, showing progression of scores over time. One patient who had surgery followed by radiation therapy had four total visits with PROMIS and SOSG-OQ scores (Figure 1). Following surgery on 5/6/2017, the patient's score trends showed overall improvement in PROMIS PF, PROMIS PI, and SOSG total when compared to baseline.
;)
Figure 1: SOSG total and PROMIS Physical Function (PF), Pain Interference (PI), and Depression (DEP) scores over four visits for a patient with metastatic lung cancer who underwent L3-pelvis posterolateral fusion and L5-S1 bilateral laminectomies on 5/6/17 (red arrow) followed by radiation therapy. Higher values are better for PROMIS PF. Lower values are better for SOSG total, PROMIS PI, and PROMIS Depression. PROMIS, Patient-Reported Outcomes Measurement Information System.
Average PROMIS PF, PI, and Depression for all 103 patient visits were 36.9 (range, 19–55), 50.27 (range, 39–76), and 50.27 (range, 34–77), respectively. Average SOSG-OQ scores were as follows: Physical Function, 9.66 (range: 4–20), Neurological Function, 6.53 (range: 4–16), Pain, 10.89 (range: 4–19), Mental Health, 9.77 (range: 4–19), and Social Function, 9.07 (range: 4–18).
PROMIS PF demonstrated excellent correlations with SOSG-OQ physical function (ρ = 0.78, P < 0.0001) but only good to excellent-good correlations with the remaining domains (ρ = 0.42–0.67, P < 0.0001) (Table 3). PROMIS PI showed excellent correlations with SOSG-OQ pain (ρ = 0.78, P < 0.0001), but only good to excellent-good correlations with the remaining domains (ρ = 0.54–0.65, P < 0.0001) (Table 3). PROMIS Depression showed an excellent correlation with SOSG-OQ mental health (ρ = 0.72, P < 0.0001) (Table 3), but only poor to good correlations with the remaining domains (ρ = 0.38–0.45, P < 0.0001) (Table 3).
TABLE 3 -
Spearman Correlation (ρ) Between
PROMIS &
SOSG-OQ Domains
|
PROMIS PF |
PROMIS PI |
PROMIS Depression |
| SOSG-OQ, Physical Function |
0.78 |
0.65 |
0.45 |
|
P value
|
<0.0001 |
<0.0001 |
<0.0001 |
| SOSG-OQ, Neurological Function |
0.67 |
0.54 |
0.38 |
|
P value |
<0.0001 |
<0.0001 |
<0.0001 |
| SOSG-OQ, Pain |
0.56 |
0.78 |
0.41 |
|
P value |
<0.0001 |
<0.0001 |
<0.0001 |
| SOSG-OQ, Mental Health |
0.53 |
0.60 |
0.72 |
|
P value |
<0.0001 |
<0.0001 |
<0.0001 |
| SOSG-OQ, Social Function |
0.42 |
0.50 |
0.40 |
|
P value |
<0.0001 |
<0.0001 |
<0.0001 |
| SOSG-OQ, Total |
0.71 |
0.78 |
0.58 |
|
P value |
<0.0001 |
<0.0001 |
<0.0001 |
PF indicates physical function; PI, pain interference; PROMIS, Patient-Reported Outcomes Measurement Information System; SOSG-OQ, Spine Oncology Study Group-Outcomes Questionnaire.
The total SOSG-OQ demonstrated an excellent correlation with PROMIS PI (ρ = 0.78, P < 0.0001) and PROMIS PF (ρ = 0.71, P < 0.0001), and a good correlation with PROMIS Depression (ρ = 0.58, P < 0.0001) (Table 3).
DISCUSSION
Disease specific PROMs and the more generic PROMIS instrument are increasingly used in spine and oncological patients.15–17 PROMIS provides insight into a patient's health across a variety of health conditions and may therefore capture information provided by various disease specific PROMs.18,19 If shown to be concurrently valid, PROMIS may adequately replace PROMs and reduce both patient and physician burden. Previous studies have found that PROMIS sufficiently captures information obtained through disease-specific PROMs.4,5,14–16 PROMIS is unique in that it utilizes an iterative CAT system, based on item response theory that improves flexibility, sensitivity, and efficiency.13 The aim of the current study was to evaluate the PROMIS PF, PI, and Depression domains in comparison to the SOSG-OQ in patients with metastatic spine disease.
Evaluating the concurrent validity of PROMIS domains with SOSG-OQ provides us with a better understanding of the information that is being captured by both tools. In the present study, PROMIS PF demonstrated excellent correlations with SOSG-OQ physical function but only good to excellent-good correlations with the remaining domains. PROMIS PI showed excellent correlations with SOSGOQ pain, but only good to excellent-good correlations with the remaining domains. PROMIS Depression showed an excellent correlation with SOSG-OQ mental health, but only poor to good correlations with the remaining domains. The total SOSG-OQ demonstrated an excellent correlation with PROMIS PI and PROMIS PF, and a good correlation with PROMIS Depression.
In previous studies, PROMIS has been found to correlate with commonly used spine relevant PROMs, including the NDI and ODI.4 Indeed, in a study of 80 unique visits from 51 patients with primary and metastatic spine tumors, Bernstein et al4 found strong correlations (1) between the PROMIS PI and ODI NDI in both primary and metastatic tumor patient subgroups (range: ρ = 0.75–0.86, P < 0.05), (2) between the PROMIS PF and ODI NDI among all patients (ρ = −0.75, P < 0.05) and in the metastatic disease subgroup (ρ = −0.78, P < 0.05), and (3) between the PROMIS Depression and ODI NDI in the primary tumor subgroup (ρ = 0.79, P < 0.05). In another study of 227 visits from 173 patients, Bernstein et al5 found a moderate to strong correlation between PROMIS PF mobility and SRS function/activity domains (range: ρ = 0.59–0.84, P < 0.01), a strong-moderate to strong correlation between PROMIS PI and SRS Pain domains (range: ρ = 0.68–0.83, P < 0.001), and a strong-moderate to strong correlation between PROMIS Depression and SRS mental health domains (range: ρ = 0.67–0.80, P < 0.001). Lastly, in 2017, Pereira et al11 evaluated the ODI, NDI, PROMIS PF, PROMIS PI, EQ-5D, and the SOSG-OQ in patients with metastatic spine disease and found that SOSG-OQ correlated strongly with PROMIS PF (ρ = 0.72, P < 0.05), whereas there was no significant correlation between SOSG-OQ and PROMIS PI (ρ = 0.62, P > 0.05).
Similar to the results of the abovementioned studies, our study findings suggestthatPROMIS captures similar clinical insight as a disease specific PROM, which, in our case, was the SOSG-OQ. The strong correlations found in the current study are understandable because the SOSG-OQ also has pain, physical function, and mental health sections, addressing similar questions to those found in PROMIS domains. Interestingly, while we found an excellent correlation between SOSG-OQ and PROMIS PI (ρ = 0.78, P < 0.0001), Pereira et al11 did not find any significant correlation between SOSG-OQ and the PROMIS PI (ρ = 0.62, P > 0.05). This may be explained by the small sample sizes as well as the different patient populations considered in the studies.
The spine is a common site for metastasis, contributing significant morbidity and negatively impacting the lives of many cancer patients. Thus, it is important for clinicians to assess outcomes that are meaningful to patients. These outcomes include physical function, the limiting impact that pain has on performing everyday activities, and mental health. PROMs are an important tool that aim to provide insight into these domains as patients are followed by their treatment teams. While disease specific PROMs including the NDI, ODI, SRS Questionnaire, and Zuriq Claudication Questionnaire exist, PROMIS may more efficiently provide such insight across multiple disease processes and therefore reduce administrative burden. As we found that PROMIS PF, PI, and Depression capture similar clinical insight as the SOSG-OQ, spine surgeons can consider using these PROMIS domains in lieu of the SOSG-OQ in metastatic spine tumor patients, thereby reducing the quantity of paperwork and related administrative burden. Indeed, the growing number of administrative tasks contribute to unnecessary costs to the U.S. healthcare system, physicians, and patients.20 The time needed to address administrative tasks including the logistics of overseeing multiple questionnaires may divert time from the actual care of patients. Further, increased response burden for patients can lead to overall lower response rates to questionnaires and thereby reduced data quality, which can lead to poorer quality of care down the line.21 Thus, it is important to study which questionnaires provide unique versus redundant data to provide the highest quality of care for patients.
Several limitations should be considered when interpreting our results. First, our patient sample consisted of patients from the spine center of a single large academic institution. With a limited sample size, the findings of this study are not necessarily representative of all patients with metastatic spine disease. Further, our patients represented 11 different types of primary tumors but were combined into one sample group. While a larger sample size would have allowed for sub-analyses amongst the 11 different primary tumor types, they were combined to ensure that the cohort sample size was sufficient to provide statistically significant correlations. Future studies may seek to increase the sample size or involve multiple centers for a more representative patient population and to further stratify the sample based on primary tumor etiology. Second, patients who did not complete the SOSG-OQ were excluded from the analysis. Whether they were excluded for missing responses to a couple of questions or for not answering either survey in its entirety, the exclusion of specific patients creates a potential for selection bias that must be considered when reviewing our study results. Third, we did not include the seven posttherapy questions of the SOSG-OQ in our analysis. These were excluded as many of the patients were undergoing therapy at the time of survey completion and did not find these questions to be relevant. Additional studies are needed to better understand the relationship between SOSG-OQ, PROMIS, and other frequently used PROMs (i.e., ODI, NDI) in the specific population of patients with metastatic spine disease.
In a previous study, Pereira et al11 evaluated the ODI, NDI, PROMIS PF, PROMIS PI, EQ-5D, and the SOSG-OQ in patients with metastatic spine disease and found a significant excellent correlation between PROMIS PF and SOSG-OQ. To our knowledge, this is the first study evaluating the PROMIS Depression domain, in addition to the PROMIS PF and PI, with the SOSG-OQ in metastatic spine disease patients. We found that PROMIS PF, PI, and Depression capture similar clinical insight as the SOSG-OQ. Thus, spine surgeons can consider using these PROMIS domains in lieu of the SOSG-OQ in metastatic spine tumor patients. This is important for broader, more interdisciplinary clinical care for this complex patient population, as PROMIS domains can also be used by oncologists and other health care providers in collaborative care.Key Points
- The SOSG-OQ was designed and validated for metastatic spine tumor patients.
- SOSG-OQ was strongly correlated with PROMIS PI (ρ = 0.78) and PROMIS PF (ρ = 0.71), and moderately correlated with PROMIS Depression (ρ = 0.58).
- PROMIS PF, PI, and Depression appear to capture similar clinical insight as the SOSG-OQ.
- Spine surgeons can consider using PROMIS domains in lieu of the SOSG-OQ in metastatic spine tumor patients.
Supplemental digital content is available for this article. Direct URL citations appearing in the printed text are provided in the HTML and PDF version of this article on the journal's Web site (www.spinejournal.com).
References
1. Rothrock RJ, Barzilai O, Reiner AS, et al. Survival trends after surgery for spinal metastatic tumors: 20-year cancer center experience.
Neurosurgery 2021; 88:402–412.
2. Ciftdemir M, Kaya M, Selcuk E, Yalniz E. Tumors of the spine.
World J Orthop 2016; 7:109–116.
3. Street J, Berven S, Fisher C, Ryken T. Health related quality of life assessment in metastatic disease of the spine: a systematic review.
Spine 2009; 34:S128–S134.
4. Bernstein DN, Greenstein AS, D’Amore T, Mesfin A. Do
PROMIS physical function,
pain interference, and
depression correlate to the Oswestry Disability Index and neck disability index in spine trauma patients?
Spine 2020; 45:764–769.
5. Bernstein DN, Papuga MO, Sanders JO, Rubery PT, Menga EN, Mesfin A. Evaluating the correlation and performance of
PROMIS to SRS questionnaires in adult and pediatric spinal deformity patients.
Spine Deform 2019; 7:118–124.
6. Rubery PT, Houck J, Mesfin A, Molinari R, Papuga MO. Preoperative patient reported outcomes measurement information system scores assist in predicting early postoperative success in lumbar discectomy.
Spine 2019; 44:325–333.
7. Papuga MO, Mesfin A, Molinari R, Rubery PT. Correlation of
PROMIS physical function and pain CAT instruments with Oswestry Disability Index and neck disability index in spine patients.
Spine 2016; 41:1153–1159.
8. Bernstein DN, Bakhsh W, Papuga MO, Menga EN, Rubery PT, Mesfin A. An evaluation of
PROMIS in patients with primary or metastatic spine tumors.
Spine 2019; 44:747–752.
9. Street J, Lenehan B, Berven S, Fisher C. Introducing a new health-related quality of life outcome tool for metastatic disease of the spine: content validation using the international classification of functioning, disability, and health; On behalf of the spine oncology study group.
Spine 2010; 35:1377–1386.
10. Janssen SJ, Teunis T, van Dijk E, et al. Validation of the Spine Oncology Study Group—Outcomes Questionnaire to assess quality of life in patients with
metastatic spine disease.
Spine J 2017; 17:768–776.
11. Paulino Pereira NR, Janssen SJ, Raskin KA, et al. Most efficient questionnaires to measure quality of life,
physical function, and pain in patients with
metastatic spine disease: a cross-sectional prospective survey study.
Spine J 2017; 17:953–961.
12. Cella D, Yount S, Rothrock N, et al. The Patient-Reported Outcomes Measurement Information System (
PROMIS): progress of an NIH roadmap cooperative group during its first two years.
Med Care 2007; 45:S3–S11.
13. National Institutes of Health. Patient-Reported Outcomes Measurement Information System (
PROMIS). June 2016. Available at:
https://commonfund.nih.gov/promis/overview. Accessed August 28, 2021.
14. Bernstein DN, Houck JR, Mahmood B, Hammert WC. Responsiveness of the
PROMIS and its concurrent validity with other region- and condition-specific PROMs in patients undergoing carpal tunnel release.
Clin Orthop Relat Res 2019; 477:2544–2551.
15. Matthys MB, Dempsey AM, Melisko ME, et al. Incorporation of Patient-Reported Outcomes Measurement Information System to assess quality of life among patients with breast cancer initiating care at an academic center.
Cancer 2021; 127:2342–2349.
16. Neal JW, Roy M, Bugos K, et al. Distress screening through Patient-Reported Outcomes Measurement Information System (
PROMIS) at an academic cancer center and network site: implementation of a hybrid model.
JCO Oncol Pract 2021; 17:e1688–e1697.
17. Sinha P, Wong AWK, Kallogjeri D, Piccirillo JF. Baseline cognition assessment among patients with oropharyngeal cancer using
PROMIS and NIH toolbox.
JAMA Otolaryngol Head Neck Surg 2018; 144:978–987.
18. Hancock KJ, Glass N, Anthony CA, et al. Performance of
PROMIS for healthy patients undergoing meniscal surgery.
J Bone Joint Surg Am 2017; 99:954–958.
19. Hancock KJ, Glass N, Anthony CA, et al.
PROMIS: a valid and efficient outcomes instrument for patients with ACL tears.
Knee Surg Sports Traumatol Arthrosc 2019; 27:100–104.
20. Erickson SM, Rockwern B, Koltov M, McLean RM. Medical Practice and Quality Committee of the American College of Physicians. Putting patients first by reducing administrative tasks in health care: a position paper of the American College of Physicians.
Ann Intern Med 2017; 166:659–661.
21. Rolstad S, Adler J, Rydén A. Response burden and questionnaire length: is shorter better? A review and meta-analysis. Value in Health. 2011; 14:1101–1108.
