1 Introduction

With a prevalence of 68% amongst individuals over the age of 60, adult spinal deformity (ASD) encompasses a spectrum of anomalies producing spinal sagittal, coronal, and rotational malalignment.^1,2 Patients with ASD typically complain of low back discomfort, radiculopathy, impairment, and poor health-related quality of life (HRQoL).^3,4 Over the past ten years, corrective surgery has been the primary treatment for ASD that is refractory to conservative measures.⁵ For ASD, posterior spinal fusion (PSF) is the most often performed surgical method. Restoring the spine's natural alignment, reducing pain, and enhancing the patient's quality of life are the main objectives of surgery. The pursuit of the optimal course of treatment for this condition and the early diagnosis of ASD may be aided or hindered by socioeconomic factors. Thus, it is challenging to achieve equitable and just health results.⁶

Disparities in healthcare outcomes can manifest itself in a number of ways, including differences in final health outcomes, quality of care, social support, and access to care.⁶ In order to create policies that reduce these inequalities, a dedicated study is necessary. Like many fields in medicine, the healthcare disparities that patients experience after orthopaedic procedures, including PSF, are mostly impacted by social determinants of health.⁷ Hopefully, by identifying the demographic and socioeconomic factors that influence care for ASD, surgeons will be able to appropriately indicate patients for surgery, ensuring they are set up for optimal success. Nevertheless, studies have also shown that these traits have been markedly underreported in randomized controlled trials (RCTs) across a wide range of medical and surgical disciplines, including RCTs in orthopaedics ^8–12.

RCTs are often regarded as the most effective method for assessing the efficacy of medical treatments. Sociodemographic variables may act as potential confounding factors in RCTs since they influence health outcomes. If sociodemographic variables are not addressed, it is challenging to determine whether a study's outcome is exclusively the product of the experimental therapy. Therefore, adding these characteristics to RCTs will improve data quality and generalizability. The purpose of this study is to ascertain the rate of reporting of sociodemographic variables in RCTs looking into adult PSF. Similar to other studies in orthopaedic surgery, we hypothesized that PSF RCTs would significantly under-report these variables.

2 Methods

3 Results

32 studies were identified in the initial database search; four duplicates were eliminated. After screening using title and abstract, 19 more studies were excluded. The remaining nine studies were evaluated for eligibility in their entirety. Two of the nine studies were disqualified for lack of a full-length text and the remaining for discussing the wrong intervention. Five RCTs were chosen for inclusion (Fig. 1).

Close

Fig. 1 Flow diagram of the literature search and selection process, performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A total of 32 records were identified from database searches (PubMed = 20, Embase = 7, Scopus = 5). After removal of four duplicates, 28 studies were screened by title and abstract. Nineteen studies were excluded, and nine full-text articles were assessed for eligibility. Of these, four were excluded (abstract only = 2; wrong intervention = 2). Five randomized controlled trials were ultimately included in the final review.

Export to PPT

Three journals published the included RCTs between 2008 and 2020 ^13–17. Age and sex were among the demographic factors that were recorded in all included RCTs. No studies included race, ethnicity, labor status, education, or income. One of the five studies included insurance status (Table 1). In the included RCTs, sociodemographic factors were reported much less frequently (insurance status: 1/5) than demographic factors such as age (5/5, 100%), sex, or gender (5, 100%) (P < 0.001).

There was no significant difference in the reporting of any sociodemographic factors when looking at journal type (p = 0.66) and publishing year (p = 0.29) (Tables 2 and 3). Of the included studies, there was only one study that included any sociodemographic data.

4 Discussion

This systematic review found that age and sex are more frequently reported than other sociodemographic variables. All five included studies reported age and sex, with one study including insurance status. However, race, ethnicity, housing status, work status, employment status, income, social support, and education level were not reported in any of the included studies. The aforementioned underreporting highlights a broader trend in neglecting important demographic and socioeconomic variables that can significantly affect health outcomes. This review emphasizes the need to include these variables in future studies to enhance adult spinal deformity care.

In this review, age and sex were reported 100% of the time. Although studies have shown that age can significantly impact discharge outcomes, literature demonstrates minimal effect of age and sex on LOS, perioperative complications, and revision rates. Lovecchio et al. found no significant differences in age (p = 0.606) or gender (p = 0.397) between individuals with short and long lengths of stay (LOS) following PSF.¹⁸ Wang et al. conducted a retrospective study examining factors affecting perioperative complications in individuals with degenerative scoliosis.¹⁹ They reported no significant differences in age (p = 0.729) or sex (p = 0.459) between individuals with and without perioperative complications undergoing PSF. Daniels et al. found no differences in age (64.0 vs. 63.7, p = 0.787) or sex (110 vs. 105, p = 0.377) between individuals with ASD who experienced short operating room (OR) times compared to those with long OR times.²⁰ Pennicooke et al. reported that patients over the age of 70 were more likely to be discharged to an inpatient rehabilitation center (Odds Ratio [OR] = 0.978) than to home.²¹ Puvanesarajah et al. found no significant difference in postoperative revision rates between patients with ASD undergoing PSF when looking at age less than or greater to 75 years old.²² However, due to age-related effects on postoperative destinations, age reporting should be continued in RCTs. Age and sex are two elementary, yet important demographic variables with implications in patient pathophysiology and outcomes, thus necessitating their inclusion in every study.

Race and ethnicity have been shown to influence health outcomes. Elsamadicy et al. reported that Hispanic patients experienced significantly higher healthcare costs compared to White patients (univariate OR = 2.36 and multivariate OR = 1.84, p = 0.041) following open thoracic/thoracolumbar PSF.²³ In an additional retrospective review, Elsamadicy et al. examined narcotic consumption and healthcare resource utilization in geriatric patients, finding increased hospital costs for Hispanic patients and those categorized in the Other/Unknown group (OR = 2.17, p = 0.003 and OR = 1.71, p = 0.005, respectively) compared to non-Hispanic White patients.²⁴ Our systematic review did not identify any RCTs that reported participant race and ethnicity data. However, due to the financial implications of PSF and subsequent rehabilitation, the reporting of race and ethnicity should be emphasized to optimize outcomes for ASD patients undergoing PSF procedures. A potential avenue for future research could involve examining the impact of high healthcare costs on rehabilitation outcomes among different racial groups.

Insurance has been shown to significantly impact patient health outcomes. Elsamadicy et al. examined differences in Hospital Frailty Risk Scores (HFRS) across individuals with Medicare, Medicaid, private insurance, or other forms of insurance. Their analysis revealed that within the Intermediate-High HFRS group, 67.4% of individuals were insured by Medicare, while 25.7% had private insurance (p < 0.001).²³ Furthermore, the study highlighted higher mean ages within the intermediate-high HFRS cohort compared to the low HFRS cohort (65.4 vs. 60.5, p < 0.001), indicating that older individuals are more susceptible to higher frailty scores and associated adverse health outcomes.²³ These findings highlight the link between frailty and insurance status, particularly among Medicare recipients. For comparison, Elsamadicy et al. reported an insignificant relationship between having Medicaid and both low HFRS scores (5.8%) and intermediate-to-high HFRS scores (4.7%) (p = 0.365).²³ Therefore, incorporating insurance status into future studies is essential, particularly as private insurance may be associated with better outcomes typically due to lower out-of-pocket expenses, shorter wait times, and greater access to healthcare providers. Furthermore, they found a significant relationship between non-routine discharge (NRD) to an additional hospital, rehabilitation, or nursing facility and individuals with private/self-pay insurance. Additionally, Jain et al. highlighted the financial burden for Medicare patients undergoing PSF, reporting costs that increased substantially with complications and readmissions—$35,878 for multilevel PSF without complications, $88,648 for three-level PSF with readmission, and $117,215 for eight-level PSF with readmission.²⁵ Our systematic review identified only one study (20%) that accounted for insurance status. Given the heterogeneity of insurance providers and variability in coverage levels, it is critical for future research to document both patient insurance status and detailed plan information. This practice would enhance the precision of data interpretation and inform strategies to optimize patient care delivery.

The employment status of patients is a crucial variable to report, as it serves as a significant predictor of functional outcomes, rehabilitation needs, and quality of life improvements. Employment status is inconsistently reported in the literature, with our systematic review identifying no RCTs addressing this factor. Grubb et al. reported 22 employed patients, two housewives, and four disabled patients (mean age = 40.9 years) diagnosed with idiopathic scoliosis before undergoing posterior stabilization and fusion.²⁶ Conversely, Grubb et al. reported a different distribution in the degenerative scoliosis group, with three employed patients, two housewives, and 19 disabled or retired individuals.²⁶ Grubb et al. observed that patients in the degenerative scoliosis group, with a mean age of 63.3 years, showed a reduced desire to return to work post-posterior spinal fusion.²⁶ These findings highlight that different age groups may have varying post-operative goals, potentially influencing the type of procedure offered and the rehabilitation strategy. This highlights the importance of capturing preoperative employment status and examining return-to-work (RTW) outcomes, which reflect a patient's ability to reintegrate functionally and economically after surgery. For example, Guan et al. reported RTW differences based on surgical approach, with 6/17 (35%) patients returning to work after open transforaminal lumbar interbody fusion (TLIF), compared to 2/10 (20%) following minimally invasive TLIF (miTLIF).²⁷ Ogura et al. similarly emphasized the relevance of RTW in the ASD population, noting that patients frequently ask, “When will I be able to return to work?”.²⁸ They found that 67% of patients (125 out of 187) who worked prior to surgery returned to work after two years.²⁸ Although Ogura et al. reported a higher mean patient age (58.2 years) compared to Guan et al. (54 and 44 years in TLIF and miTLIF cohorts, respectively), they observed higher RTW rates—highlighting that factors beyond age and technique, such as occupation type and patient expectations, may influence outcomes.^27,28 Future research should incorporate more detailed variables, such as RTW timelines and qualitative RTW data, as prolonged absence from work can also lead to economic challenges for patients.

Socioeconomic indicators such as income, housing, and education were not included in any of the RCTs included in this review, despite their known influence on pre-, peri-, and post-operative outcomes. Ng et al. utilized the Area Deprivation Index (ADI), a higher score indicating worse socioeconomic status, and reported that PSF patients who were in the top 10% of ADI had higher odds of hospital readmission at 30 and 90 days (OR = 5.00, p < 0.001 and OR = 4.50, p < 0.001, respectively).²⁹ Although Ng et al.'s study did not specifically address ASD, its findings underscore the need to examine area-level socioeconomic disadvantage in spine surgery patients. Elsamadicy et al. reported no significant differences between income quartiles (0-25th, 26-50th, 51-75th, and 76-100th) and HFRS scores (p = 0.936).²³ These findings suggest that in this cohort, income along was not associated with frailty severity, though other SES factors may still play a role. The interplay of socioeconomic variables and differing patient characteristics, including diagnosis and treatment, clearly create numerous confounding factors. Future studies should control for these social determinants to better understand health outcomes.

5 Conclusion

This systematic review found that demographic variables, such as age and sex, are more frequently reported than sociodemographic variables. All five included studies reported age and sex, with only one study also considering insurance status. However, critical variables such as race, ethnicity, housing status, employment status, income, social support, and education level were not reported. The consistent omission of these variables, which impact health outcomes, underscores the need for their inclusion in future research to enhance ASD care. Notably, incorporating the Area Deprivation Index (ADI) is recommended, as it provides a comprehensive assessment of socioeconomic disparities and enables higher-quality research by identifying and addressing spinal health interventions in geographically underserved areas. Ultimately, including more specific sociodemographic variables in randomized controlled trials is essential for understanding and addressing healthcare disparities in ASD treatment outcomes.

Guardian/patient's consent

None Needed.

Ethical approval

This study is a systematic review of existing published data and does not involve human participants or identifiable personal data. Therefore, ethical approval was not required in accordance with institutional guidelines.

Credit author statements

Kenneth T. Nguyen: Conceptualization; Methodology; Formal analysis; Writing – Original Draft; Writing – Review & Editing.

Erin L. Brown: Investigation; Data curation; Writing – Original Draft; Writing – Review & Editing.

Bhavesh R. Koppala: Investigation; Data curation; Writing – Original Draft.

Daman P. Dhunna: Conceptualization; Methodology; Writing – Original Draft.

Neil Samant: Supervision; Writing – Original Draft; Writing – Review & Editing.

Hanna House: Supervision; Writing – Review & Editing.

Alexander Kuzma: Methodology; Supervision; Writing – Review & Editing.

Nikolas Baksh: Supervision; Writing – Review & Editing.

All authors approved the final manuscript and are accountable for all aspects of the work.

Funding statement

No funding was provided for the entirety of this research study.