Abstract
Background Physician specialists (PS) are often the type of healthcare provider initially contacted by an individual with low back pain (LBP). LBP clinical practice guidelines (CPG) recommend a stepped approach to management with an emphasis on first-line non-pharmaceutical and non-interventional services.
Objective Examine the association between the incorporation of CPG recommended first-line services, exposure to second- and third-line services and total episode cost for individuals with non-surgical LBP initially contacting a PS.
Design Retrospective observational study with identical design to previous study focused on primary care physicians.
Setting/Patients National sample of individuals with non-surgical LBP occurring in 2017-2019.
Measurements Independent variables were initial contact with a PS, and the timing of incorporation of five types of first-line services. Dependent measures included exposure to thirteen types of health care services and total episode cost.
Results 91,096 individuals were associated with 98,992 episodes of non-surgical LBP. 36.2% of the 33,277 PS initially contacted for an episode of LBP incorporated any first-line service at any time during an episode. A first-line service was provided in 24.0% of episodes with active care (19.5% of episodes), manual therapy (13.7%) and chiropractic manipulative therapy (6.5%) the most common. 7.3% of non-surgical LBP episodes included a first-line service within seven days of initial contact with a PS. These episodes were associated with a reduction in the use of prescription skeletal muscle relaxants (risk ratio 0.88) and opioids (0.55), spinal injections (0.84), and CT scans (0.71), with no impact on the use of prescription NSAIDs, radiography, or MRI scans. First-line services were associated with an increase in total episode cost at any time of incorporation with chiropractic manipulation associated with the lowest cost increase. Younger individuals from zip codes with higher adjusted gross income were more likely to receive a first-line service in the first seven days of an episode.
Limitations As a retrospective observational analysis of associations there are numerous potential confounders and limitations.
Conclusions For individuals with non-surgical LBP PS provide second- or third-line services more frequently and earlier than CPG recommended first-line services. There is an opportunity to improve concordance with LBP CPGs for individuals with LBP initially contacting a PS.
Introduction
The high prevalence and cost of low back pain (LBP) are well understood1-4 and considerable resources have been devoted to the development of high-quality LBP clinical practice guidelines (CPGs) that describe a stepped approach to management.5-7 Self-management, non-pharmacological and non-interventional services are recommended as first-line approaches for LBP without red flags of serious pathology.5-7 The timing of introduction of CPG recommended first-line services has been identified as being potentially important.8-17 Management of LBP that is not concordant with CPGs increases the risk of LBP transitioning from an acute to a chronic condition18 and is an important source of “low-value” care19-22, described as health care services generating cost without, or with minimal, beneficial impact on outcomes.23,24
The type of health care provider (HCP) initially contacted by an individual with LBP has been used as a method to evaluate variation in service utilization and cost outcomes.25-27 As specialists in the management of musculoskeletal conditions, it is not surprising that physician specialists (PS) like orthopedic surgeons, physical medicine and rehabilitation physicians, and pain management physicians are commonly consulted by individuals with LBP.27 Several barriers to PCP referral for CPG recommended first-line services have been identified.28-33 Compared to PCP management of LBP, the rate and timing of PS referral for first-line services is less well understood.
For individuals with non-surgical LBP initially contacting a PS, the aim of this retrospective, observational study was to examine the association between the timing of incorporation of active care (AC), manual therapy (MT), chiropractic manipulative therapy (CMT), osteopathic manipulative therapy (OMT), or acupuncture services, utilization of other healthcare services, and total cost. The hypothesis was physicians specializing in the management of LBP would be associated with a high degree of LBP CPG concordance and that early incorporation of one or more of these first-line services would be associated with lower rates of second- and third-line service use, and lower total episode cost.
Methods
Study design, population, setting and data sources
This is a retrospective cohort study of individuals initially contacting a PS for an episode of non-surgical LBP. The PS HCP category consisted of orthopedic surgeon, neurosurgeon, physical medicine and rehabilitation, pain management, neurology and rheumatology physician types.27 The study cohort was able to access all PS HCP types directly without a referral.
The study design was identical to a previous study involving primary care physicians (PCP).34 To facilitate a comparison between PCP and PS results, figure axis scales were kept constant. An enrollee database included de-identified enrollment records, and administrative claims data for individuals with LBP. De-identified HCP demographic information and professional licensure status was included in an HCP database. ZIP code level population race and ethnicity data was extracted from the US Census Bureau35, adjusted gross income (AGI) data from the Internal Revenue Service36 and socioeconomic status (SES) Area Deprivation Index (ADI) data, from the University of Wisconsin Neighborhood Atlas® database.37 An analytic database was created by linking these multiple databases.
With study data de-identified or a Limited Data Set in compliance with the Health Insurance Portability and Accountability Act (HIPAA) and customer requirements the UnitedHealth Group Office of Human Research Affairs determined that this study was exempt from Institutional Review Board review. The study was conducted and reported based on the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. (Supplement – STROBE Checklist).38
Like the identical PCP study34 the analysis was unable to control for numerous known, unknown, or unmeasurable confounders. In this analysis of associations, no attempt was made to compare results with the identical PCP study34, or to generate causal insights using potentially inadequate approaches such as propensity score matching39 to perform incomplete adjustment for typical confounders such as age, sex and non-LBP related co-morbidities.40,41 As an alternative, actual individual demographic and episodic characteristics and associations are reported for the timing of introduction of the first-line services analyzed in the study.
Unit of analysis and cohort selection
The cohort consisted of individuals with continuous medical and pharmacy insurance coverage during the entire study period who were aged 18 years and older with a complete episode of LBP commencing and ending during the calendar years 2017-2019. This timeframe was selected to follow the release of the American College of Physicians (ACP) LBP CPG5 in 2017 and before the influence of the COVID-19 epidemic on care patterns in early 2020.
Episode of care was selected as the unit of analysis. Episodes have been shown to be a valid way to organize administrative claims data associated with a condition.42 The Symmetry® Episode Treatment Groups® (ETG®) and Episode Risk Groups® (ERG®) version 9.5 methodologies and definitions were used to translate administrative claims data into episodes, which have been reported as a valid measurement for comparison of HCPs based on cost of care.43 A previous study found a low risk of misclassification bias associated with using episode of care as the unit of analysis.27
For this study complete episodes were defined as having at least 91-day pre- and 61-day post-episode clean periods during which no services were provided by any HCP for any LBP diagnosis. Excluded from the analysis were LBP episodes including a surgical procedure, or associated with diagnoses of malignant and non-malignant neoplasms, fractures and other spinal trauma, infection, congenital deformities and scoliosis, autoimmune disorders, osteoporosis, and advanced arthritis. These exclusions are particularly important for LBP episodes where a PS was initially contacted and were made to address a potential study limitation of individuals with more complex conditions confounding the analysis of timing of incorporation of first-line non-pharmacological and non-interventional services.
Variables
Data preprocessing, table generation, and initial analyses were performed using Python (Python Language Reference, Version 3.7.5., n.d.). A goodness of fit analysis was conducted using D’Agostino’s K-squared test. Non-normally distributed data are reported using the median and interquartile range (IQR).
The primary independent variables were initial contact with a PS, and the timing of incorporation of AC, MT, CMT, OMT, or acupuncture services. For LBP, these are the most frequently provided first-line services recommended by CPGs and covered by commercial insurance.27 Passive therapies, like ultrasound or electrical stimulation, were excluded from the definition of first-line services used in the analysis. For episodes initially contacting a PS the timing of incorporation of AC, MT, CMT, OMT, or acupuncture services was based on the number of days after the initial visit with a PS when these services were first billed by any type of HCP.
The primary dependent variable was the rate and timing of use of thirteen types of health care services segmented into first-, second-, and third-line service categories based on the ACP LBP CPG as a primary source for the designation.5 Secondary dependent variables included the total cost of care for all reimbursed services provided by any HCP during an episode, the number of different HCP seen during an episode, and episode duration measured in days. Total episode cost included costs associated with all services provided for an episode of LBP, including those not specifically identified in the thirteen categories used in the analyses. Costs for services for which an insurance claim was not submitted were not available. The episode duration was the number of days between the first and last date of service for each episode.
Bivariate analyses were performed comparing episode attributes associated with timing of introduction of AC, MT, CMT, OMT, or acupuncture services. For the bivariate analyses the reference baseline was episodes that did not include a specific first-line service. Fisher’s Exact test (p value of .001) was used for comparing the percent of episodes including a service, and Mann Whitney U test (p value of .001) was used for measures reported using median and IQR.
Using identical methods, all LBP tables, figures, and supplement items were replicated for non-surgical neck pain (NP) episodes where a PS was the initial HCP contacted. NP data are included as supplemental items using the same name as the corresponding LBP tables and figures.
Role of Funding Source
None
Results
The sample included 91,096 individuals, with a median age of 49 (Q1 38, Q3 57), and 54.9% females. These individuals were associated with 98,992 complete non-surgical LBP episodes involving 33,277 unique PS. There were $127,500,028 in reimbursed health care expenditures with a median total cost per episode of $389 (Q1 $138, Q3 $1,263). The median pre-episode clean period was 571 days (Q1 342, Q3 832). The median number of days between sequential episodes was 227 (Q1 126, Q3 364). The median post-episode clean period was 440 days (Q1 287, Q3 718) (Table 1). Individuals were from all 50 States and some U.S. territories. (Supplement - State) 76.0% of non-surgical LBP episodes did not include a first-line service at any time during an episode. For the 24.0% of episodes that included any first-line service at any time AC (19.5% of episodes), MT (13.7%) and CMT (6.5%) were most common. OMT (0.8%) and acupuncture (0.5%) were rarely provided. Individuals were more likely to receive radiography (41.3%), prescription NSAIDs (24.8%), MRI (22.2%), opioids (18.8%), skeletal muscle relaxants (16.2% of episodes), and spinal injections (14.0%) than all first-line services except AC. (Table 2)
Within the first 7 days of an episode 7.3% of episodes included a first-line service with AC (5.9% of episodes), MT (3.7%) and CMT (1.0%) being most common. When introduced in the first seven days of an episode, AC was associated with a significant reduction in exposure to prescription opioids (risk ratio 0.57), CT scans (0.76), spinal injections (0.87), and skeletal muscle relaxants (0.88). When introduced 8-14 days into an episode AC was associated with a significant reduction in exposure to prescription opioids (0.58), CT scans (0.58) and spinal injections (.088). At 15-28 days into an episode AC was associated with reductions in CT scans (0.75) prescription opioids (0.76). (Figure 1) MT was associated with similar reductions in exposure to prescription opioids. When introduced in the first seven days of an episode CMT was associated with significant reductions in CT scans (0.51), MRI (0.70), prescription opioids (0.83), and radiography (0.88). When introduced 8-14 days into an episode CMT was associated similar reductions in exposure to spinal injections (0.69), and MRI (0.81). (Figure 2) The small proportion of episodes including OMT or acupuncture at any time presented a barrier to identifying potential impact on second- and third-line service exposure. (Table 2) (Supplement – Risk Ratio)
For the 7.3% of episodes with a first-line service introduced in the first 7 days of an episode, and compared to episodes without a first-line service, individuals were younger (46 years old), with a lower ERG® risk score (1.7), from zip codes with lower deprivation (ADI 33), higher AGI (83,843), and with greater availability of a DC or PT. Among first-line services, acupuncture use was most strongly associated lower deprivation (ADI 14), higher AGI (125,155), and greater availability of LAcs (0.05). (Figure 3) (Table 3)
Total episode cost increased with the introduction of any first-line service, and progressively increased the later a first line service was introduced. CMT and OMT were associated with the smallest total episode cost increase. (Figure 4) (Table 3)
A statistical comparison of PS management of LBP and NP was not a study aim. Nevertheless, NP results are provided as supplemental items with results similar to LBP. LBP and NP cohort attributes were nearly identical except for cohort percent female where LBP was 54.9% and NP was 60.4%. (Table 1)(Supplement Table 1) The percent of episodes including any first-line service was 24.0% for LBP and 24.3% for NP. The first-line services most commonly incorporated during an episode was also similar; AC (19.5% for LBP, 19.3% for NP), MT (13.7%, 15.7%), CMT (6.5%, 6.7%), OMT (0.8%, 1.1%), and acupuncture (0.5%, 0.6%). (Table 2)(Supplement Table 2) Among first-line services CMT and OMT were associated the lowest total episode cost increase for both LBP and NP. (Table 3)(Supplement Table 3)
Discussion
As specialists in the management of musculoskeletal conditions it was hypothesized that PS HCP types like orthopedic surgeons, physical medicine and rehabilitation, and pain management physicians, when initially contacted by individuals with LBP, would be associated with a high degree of CPG concordance. Consequently, it was unexpected to find that PS incorporate CPG recommended non-pharmaceutical and non-interventional first-line services at any time in less than 25% of non-surgical LBP episodes, with less than 15% of episodes include a first-line service in the first fourteen days of an episode when the potential benefits of avoiding low-value second- and third-line services is highest. PS incorporate second- and third-line services more often than first-line services. These findings are nearly identical to an earlier study of PCP management of non-surgical LBP.27
As an observational retrospective cohort study of associations there are several limitations to consider. Perhaps most importantly a risk of selection bias is present due to the limited ability to control for individual preference for type of initial contact HCP, individual expectations or requests for specific health care services, and potentially meaningful differences in clinical complexity of individuals seeking treatment. These are particularly important potential limitations for individuals with LBP choosing to seek initial treatment from a PS. Following initial contact with a PS, and if not provided directly by the PS, it was not possible to differentiate whether first-line services were accessed by referral from the PS, by referral from another HCP, or accessed directing by the individual with LBP. These limitations were partially addressed by narrowing the study population by excluding LBP episodes associated with significant pathology and by focusing on only non-surgical LBP episodes.
While the cohort had continuous highly uniform commercial insurance coverage with quality and actuarial control measures applied to the processing of administrative claims data, data errors, variability in benefit plan design, variability in enrollee cost-sharing responsibility, and missing information were potential sources of confounding or bias. Even with the insurer HCP database being under continual validation it may have included errors or missing information regarding the identification of PS. Summarizing total episode cost has potential limitations associated with insurance coverage, nature of network participation, and alternative reimbursement models. While individuals from all 50 states and most US territories were included, providing a measure of generalizability, the cohort did not describe a U.S representative sample.
This study builds on and corroborates two earlier studies of the same underlying data. A low proportion of LBP episodes generally include timely incorporation of CPG recommended non-pharmacologic and non-interventional services.27 When initially contacted by an individual with LBP, PS are nearly identical to PCPs34 in that CPG recommended first-line services are infrequently incorporated, and if incorporated are typically later in an episode and in addition to second- and third-line services.
Administrative burden and the cost of CPG recommended first-line services have been identified as referral barriers.30,31 Like the previous identical PCP study34, this study found PS incorporate CPG recommended first-line services infrequently, and if incorporated it is after second- and third-line services, and resulted in higher total episode cost. The low rate of PS incorporation of CPG recommended first-line services may also reflect previously identified confounders like the time commitment to attend a visit, wait times to schedule a visit, transportation barriers, and a variety of individual characteristics.29,44
This study of PS management of LBP corroborates and expands on previous studies of PCP management of LBP that found earlier use of first-line services may be associated with a reduction in use of low value services and prescription pharmaceuticals, including opioids 13,14,34,45. The study finding that the benefits of early use of first-line services are most evident if initiated within seven days of initially contacting a PS corroborates previous similar findings 11,34, reinforces the principle that both the proportion of LBP episodes including, and timing of incorporation of, first-line services is important.
Conclusion
High quality LBP CPGs recommend a stepped approach in which favorable natural history, self-care, and non-pharmacological services are emphasized as first-line approaches for individuals without red flags of serious pathology. For individuals with non-surgical LBP initially contacting a PS this study reveals second- and third-line pharmaceutical, imaging, and interventional services are provided more commonly and earlier than CPG recommended first-line services. With PS being positioned in the delivery system as musculoskeletal and spine specialists, identifying and addressing root causes of the observed high rate of CPG discordant LBP care is important.
Data Availability
All data produced in the present work are contained in the manuscript
Acknowledgements
None
Footnotes
Declarations
Ethics approval and consent to participate Because the data was de-identified or a Limited Data Set in compliance with the Health Insurance Portability and Accountability Act and customer requirements, the UnitedHealth Group Office of Human Research Affairs determined that this study was exempt from Institutional Review Board review.
Consent for publication Not applicable
Availability of data and materials The data are proprietary and are not available for public use but, under certain conditions, may be made available to editors and their approved auditors under a data-use agreement to confirm the findings of the current study.
Competing interests At the time of manuscript submission DE and MZ are UnitedHealth Group employees and UNH stockholders. No other potential conflicts of interest or competing interests exist.
Funding None
List of Abbreviations
- LBP
- Low back pain
- US
- United States
- CPG
- Clinical practice guideline
- PCP
- Primary care provider
- PS
- Physician specialist
- DC
- Doctor of Chiropractic
- PT
- Physical Therapist
- HCP
- Health care provider
- IQR
- Interquartile range
- AGI
- Adjusted Gross Income
- ADI
- Area Deprivation Index
- STROBE
- Strengthening the Reporting of Observational Studies in Epidemiology
- ETG®
- Episode Treatment Group®
- ERG®
- Episode Risk Group®
- ACP
- American College of Physicians
- CMT
- Chiropractic manipulative treatment
- OMT
- Osteopathic manipulative treatment
- AC
- Active care
- MT
- Manual therapy