Journal Logo

Research Paper

Effect of back problems on healthcare utilization and costs in Ontario, Canada: a population-based matched cohort study

Wong, Jessica J.a,b,*; Côté, Pierrea,b,c,d; Tricco, Andrea C.a,d,e; Watson, Tristanf; Rosella, Laura C.a,f,g

Author Information
doi: 10.1097/j.pain.0000000000002239

1. Introduction

Low back pain (LBP) is the leading cause of years lived with disability globally.27 Global years lived with disability for LBP were 42.5 million in 1990 and increased by 53% to 64.9 million in 2017.61 Approximately 80% of people experience at least one episode of LBP during their lifetime, and 20% of Canadians have back problems at any given time.15,34,56 The global point prevalence of LBP was 7.8% in 2017, affecting 577 million people at any given time.61 Back problems have led to considerable disability, functional limitations, and lost productivity worldwide.13,14,24,27,45

Back problems are associated with high healthcare utilization and costs, with LBP ranked as the fifth most common reason for all physician visits in the United States.18–20,23 The pooled prevalence of healthcare utilization among individuals with LBP in the general population was 56% (95% confidence interval [CI] 45-67).9 In the United States, healthcare spending for back and neck pain was an estimated $87.6 billion US dollars (USD) in 2013, which was the third highest after diabetes and ischemic heart disease.20 Healthcare spending for back and neck pain increased $57.2 billion USD over 18 years, representing the second highest increase in healthcare spending after diabetes.20

Few studies have comprehensively quantified the burden of back problems at the health system level (eg, physician visits and hospitalizations) in Canada, particularly with approaches that account for comorbidities and a wide range of potential confounders. A cross-sectional study reported 1.6 million outpatient physician visits for spinal conditions and $264 million Canadian dollars (CAD) in total costs for spine-related care among adults in 2013 to 2014.40 Moreover, studies are needed to determine per-person incremental costs for back problems in the population, which are preferred over cost-of-illness approaches to guide decision makers.8,57 Incremental costs represent additional costs from a disease and cost savings if the condition was appropriately managed or resolved.8,57 An incremental physician cost of $96.25 was reported for back pain among adults in Ontario in 1994,31 but this study was limited to a short time frame (1994-1995) and physician visits. More recent, comprehensive, high-quality estimates to quantify the health and economic burden of back problems will provide critical information to guide health services delivery and monitoring, economic models, and strategies for healthcare improvements.

To address these knowledge gaps, linking population health surveys with health administrative data is a unique opportunity to build a population-based cohort of individuals with back problems within a single-payer health system. The Canadian Community Health Survey (CCHS) captures self-reported back problems and overcomes the limitations of coding back problems in administrative data.60 Data from the CCHS are representative of the community-dwelling Canadian population aged 12 years and older.49 This data linkage captures all medical encounters and direct person-level healthcare costs, allowing for comprehensive estimates in health and economic burden generalizable to the entire population.

The objective was to assess the effect of self-reported back problems compared with no self-reported back problems on healthcare utilization and costs in a population-based sample of Ontario adults in a single-payer health system.

2. Methods

We conducted a dynamic population-based matched cohort study of Ontario adult respondents of the CCHS to examine healthcare utilization and costs associated with back problems. We reported this study according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.52 Additional details on methodology are available in the published study protocol.59 This project has been approved by the Health Sciences Research Ethics Board at the University of Toronto.

2.1. Study sample

We included all Ontario respondents of at least one of 5 CCHS cycles (cycle 2.1 [2003-2004], cycle 3.1 [2005-2006], 2007/2008, 2009/2010, and 2011/2012) aged 18 years or older at the time of the survey interview. We excluded respondents who could not be linked with health administrative databases or had a death date preceding the CCHS interview date. The linkage rate between CCHS and health administrative databases ranged from 81% to 85% (ie, 2003 [83%], 2005 [85%], 2007 [83%], 2009 [83%], and 2011 [81%]). We only used data from the first survey for respondents of multiple CCHS cycles (<1% of respondents excluded).

Ontario is the largest province by population (∼14.3 million in 2018) in Canada, and the most ethnically diverse province with more than 200 ethnicities represented.47 Many healthcare services are publicly funded in Ontario, including family physician and specialist visits and most basic and emergency healthcare services (eg, surgery and hospital stays).36 These services are paid through the government-run provincial health insurance plan, which is the Ontario Health Insurance Plan (OHIP).

2.2. Data sources

Data from the CCHS were individually linked to individual-level healthcare utilization data from health administrative databases. These datasets were linked using unique encoded identifiers and analyzed at ICES. ICES is an independent, non-profit research institute whose legal status under Ontario′s health information privacy law allows it to collect and analyze health care and demographic data, without consent, for health system evaluation and improvement. The CCHS is a cross-sectional survey administered by Statistics Canada that collects data on the distribution of health determinants, outcomes, and healthcare use across Canada.49 The CCHS uses a multistage sampling survey design to target Canadians aged 12 years and older living in private dwellings and excludes persons living in institutions (eg, long-term care or complex continuing care facilities), full-time members of the Canadian Forces, and persons living on-reserve and other First Nations settlements.49 The CCHS uses 3 sampling frames to generate survey participants: (1) area frame, which consists of a selection of dwellings from Statistics Canada's Labour Force Survey sampling frame; (2) list frame, which consists of a list of telephone numbers from the Canada phone directory; and (3) random digit dialing, which is used to supplement the sample in 4 health regions.49 We restricted the sample to respondents aged 18 years and older to focus on adults with back problems. Starting in 2001, the CCHS collected data from a sample of respondents every 2 years until 2007, after which CCHS data were collected annually.49 The CCHS data are representative of 98% of the Canadian population aged 12 years and older living in private dwellings at national and provincial levels, with response rates ranging from 67% to 81% (ie, 2003 [81%], 2005 [79%], 2007/2008 [78%], 2009/2010 [72%], and 2011/2012 [67%]).49 Detailed survey methodology is described elsewhere.48

We used health administrative data from OHIP, Canadian Institute for Health Information (CIHI) Discharge Abstract Database and Same-Day Surgeries, and National Ambulatory Care Reporting System to capture physician billings, emergency department visits, and hospitalizations. The OHIP covers all Ontario residents, including all CCHS respondents, as a single-payer health insurance system. These data cover all healthcare providers who can claim OHIP (eg, physicians and laboratories) and include service codes, dates of service, and associated diagnosis.29 The CIHI Discharge Abstract Database and Same-Day Surgeries collect demographic, administrative, and clinical data on hospital discharges and same-day surgeries, which are received from acute care facilities, health/regional authority, or ministry of health depending on the province. National Ambulatory Care Reporting System captures data on all hospital-based and community-based ambulatory care collected from specific facilities, regional health authorities, and ministries of health.

2.3. Exposure

The exposure of self-reported back problems was obtained from the CCHS question: “Do you have back problems, excluding fibromyalgia and arthritis?” Individuals who responded yes to this question were classified as having self-reported back problems. This CCHS question refers to “conditions diagnosed by a health professional and are expected to last or have already lasted 6 months or more.” Previous studies have used this definition of self-reported back problems.1,10,16,34,37,60

2.4. Outcomes

Outcomes of interest were cause-specific and all-cause healthcare utilization and healthcare costs, from CCHS interview date to March 31, 2018 as end of study period or death date. The duration of follow-up ranged 6 to 15 years given the dynamic nature of the cohort. One visit was counted as one claim per patient per service day per physician for OHIP data. Cause-specific visits were calculated based on billing or procedural codes related to back problems in regions spanning from the costal margin to the inferior gluteal folds or procedural codes for imaging of the spinal region (ie, spinal radiographs, computed tomography, and magnetic resonance imaging) (Appendix I, available at http://links.lww.com/PAIN/B310). International Classification of Diseases-10 codes for LBP-related physician billing and hospital visits included M47, M48, M51, M53, M54, M99, and S33, with similar International Classification of Diseases-9 codes for LBP. These billing and procedural codes were informed by previous studies.12,22,60 All-cause healthcare utilization included all physician visits, emergency department visits, and hospitalizations.

Total healthcare spending in Canadian dollars, adjusted to 2018, was calculated using a person-centred costing approach to linked health administrative databases.58 This methodology uses an algorithm to compute costs accrued by each person based on healthcare visits covered by the Ministry of Health and Long-Term Care after the CCHS interview date. Costs were calculated from the perspective of the Ontario Ministry of Health and Long-Term Care, which represents the healthcare payer. Previous studies successfully applied these methods to estimate attributable costs for other conditions.11,42,43,58 Specifically, healthcare costs were estimated using validated algorithms at ICES.58 The costing methodology computes cumulative individual-level healthcare costs for all publicly funded health system encounters over time. The methodology focuses on the formal component of direct healthcare costs and therefore excludes copayments, costs associated with caregivers, private insurance, overheads and capital expenditures, and community-level services where an individual's health card number is not tracked.

The established costing methodology at ICES allocates healthcare costs to individual patients by (1) identifying each individual's healthcare encounters and (2) assigning unit costs/prices to services used during the encounter.58 Patient encounters are generally grouped into episodes and visits or claims. Costs for inpatient hospital-based episodes are computed by multiplying Resource Intensity Weights with cost per weighted case. Resource Intensity Weights are a measure of the amount of hospital resources used during the encounter (eg, administration, staff, supplies, drugs, technology, and equipment). For episodes such as complex continuing care, utilization measures and unit costs based on weighted days are used. The CIHI developed the methods for calculating utilization weights and unit costs for the episodes of care. For visits or claims, costs are determined at the time of utilization. These include costs for long-term care (fixed per diem costs based on government payment rates), physician costs (claims submitted to OHIP and capitation payments for primary care physicians), drug costs (costs for prescription drugs dispensed to individuals eligible for publicly funded drug coverage), home care costs (visit costs based on service type as well as case management and administration costs), and assistive devices (reimbursements through the Assistive Devices Program).

2.5. Potential confounders

The following variables were considered potential confounders of the association between back problems and healthcare utilization and costs, as informed by previous literature.24,25,30,38,51 These include (1) sociodemographic factors: age (years), sex (male or female), location of residence (urban or rural), household income (lowest to highest quintiles), education (less than secondary, secondary graduate, or more than secondary), immigrant status (immigrant or Canadian-born); (2) health-related or behavioural factors: self-reported factors: smoking status (former/current smoker or never smoker), alcohol consumption (heavy/moderate drinker or light/never drinker), physical activity status (active/moderately active or inactive), body mass index (normal, overweight, or obese), self-rated general health (excellent/very good/good, fair, or poor); and (3) comorbidities (taken from health administrative data before survey date): ACG System Aggregated Diagnosis Groups (ADGs) using The Johns Hopkins ACG System Version 10.0.1 (Johns Hopkins HealthCare, LLC; https://www.hopkinsacg.org/), which have been validated among adults in Ontario,6 health conditions using health administrative database algorithms (ie, diabetes, hypertension, congestive heart failure, chronic obstructive pulmonary disease, dementia, stroke, and coronary artery disease).17,21,28,46,54

3. Analysis

We used a survey-weighted logistic regression model that included the aforementioned confounders and CCHS cycles to estimate a propensity score for the probability of having back problems compared to not having back problems. We created a propensity score-matched cohort (hard matched on sex) using a nearest-neighbor 1:1-greedy matching algorithm to match participants in the exposed and unexposed groups based on the logit of the propensity score, with a caliper width of 0.2 times the standard deviation.3,5 We assessed the balance of each baseline covariate between matched exposed and unexposed groups using standardized differences, with differences of <0.1 (ie, <10%) suggesting good balance.2 After propensity score matching, negative binomial regression was used to model the association between back problems and rate of healthcare visits to compute rate ratios (RRs) and 95% CI, stratified by sex. For each subject, the numerator of the rate was the number of healthcare visits over their follow-up period and the denominator was the follow-up duration, with an offset term to account for varying follow-up. We also modeled differences in healthcare costs adjusted to 2018 Canadian dollars using linear (log transformed) models.32 Analyses were stratified by sex because healthcare utilization patterns for back problems, such as frequency and type of visits, likely differ according to sex.35

All estimates incorporated the CCHS survey weights, and variance calculations were based on bootstrap weights with balanced repeated replication.4 We used a pooled approach to combine CCHS cycles, which increases sample size and statistical power.53 To calculate the population-level burden of back problems, we applied the CCHS weighted sample prevalence and rate differences in healthcare utilization or incremental costs of back problems to the 2019 Ontario population.50 All costs were adjusted to 2018 Canadian dollars, and the annual exchange rate for 1 Canadian dollar was $0.77 USD in 2018.7 Analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC) and Stata/MP 15.1 for Unix (StataCorp, College Station, TX).

3.1. Sensitivity analyses

We conducted sensitivity analyses to assess the potential impact of misclassification and residual confounding on study results. First, we conducted separate analyses to define back problems using both self-report and diagnostic information to assess the potential impact of misclassification of the exposure. Specifically, we conducted analyses with adults who self-reported back problems and also had at least 1 healthcare visit for LBP within 1 year before the CCHS interview date. Moreover, in separate analyses, we included at least 1 healthcare visit related to thoracic or rib pain in addition to LBP codes within 1 year before the CCHS interview to further assess for potential misclassification of the exposure (Appendix I, available at http://links.lww.com/PAIN/B310). We also added thoracic and rib pain codes when defining cause-specific healthcare utilization to broaden the outcome. Second, we conducted a quantitative bias analysis to assess the potential impact of residual confounding from unmeasured or unknown confounders. This analysis estimated the extent to which these confounding variables may explain some or all of the reported association between the exposure and outcome.33

We also conducted a number of analyses to inform the generalizability of results. First, we conducted separate analyses with CCHS data on ethnicity (ie, visible minority and white) included in the propensity score matching. Second, we conducted a sensitivity analysis with opioid use as the outcome in a subset of the population because of data availability to inform the generalizability of results. This analysis was conducted in respondents of the 2011/2012 CCHS cycle followed to March 31, 2018 for having claims for a prescribed opioid (ie, opioid group) in the Narcotic Monitoring System data. The Narcotic Monitoring System captures all prescriptions for monitored drugs dispensed from community pharmacies in Ontario, excluding those filled in hospitals or prisons.

4. Results

The CCHS data had 168,074 respondents from 5 combined CCHS cycles 2003 to 2012 (Appendix II, available at http://links.lww.com/PAIN/B310). A total of 17,537 respondents were excluded because of ineligibility or having missing exposure data (0.1%). Of the 150,537 respondents used for analysis, 36,806 had self-reported back problems. After matching, there were 36,806 pairs of respondents (21,054 pairs for women and 15,752 pairs for men) with and without self-reported back problems.

Before matching, adults with back problems were older (mean age 51 vs 45 years) and had higher average ADGs scores (mean 7 vs 5), with standardized differences ≥0.1 (Table 1). A higher proportion of respondents with back problems were obese, physically inactive, former or current smokers, self-rated their general health as fair or poor, or had at least 1 chronic disease compared to those without back problems, with standardized differences ≥0.1. After matching, the mean age in both groups was 51 years and the mean ADGs score was 7. All characteristics across groups achieved standardized differences of less than 0.1.

Table 1 - Baseline characteristics (weighted) of (1) pool of adults with and without back problems and (2) propensity score-matched cohort, pooled participants surveyed from 2003 to 2012 and followed up to 2018, Canadian Community Health Survey, Ontario, Canada.*
Variable Adults with back problems, n = 36,806 Adults without back problems, n = 113,731 Absolute standardized difference Variance ratio Adults with back problems (propensity score-matched cohort), n = 36,806 Adults without back problems (propensity score-matched cohort), n = 36,806 Absolute standardized difference Variance ratio
Hard match variable
 Female sex (%) 53.91 50.45 0.06 0.99 53.91 53.43 <0.01 1.00
Propensity score variables
 Age at index date in years, mean (median) 50.86 (49.79) 44.66 (42.56) 0.32 0.82 50.86 (49.79) 51.21 (50.32) 0.05 0.88
Location of residence
 Rural (%) 16.56 13.82 0.04 1.06 16.56 16.47 0.01 0.99
Income quintile (%)
 1 (lowest) 17.00 13.69 0.14 1.30 17.00 16.93 0.03 1.06
 2 15.16 15.13 0.04 1.09 15.16 15.04 0.02 1.04
 3 18.05 16.89 0.01 1.01 18.05 17.96 0.02 0.97
 4 19.77 20.75 0.05 0.93 19.77 19.62 0.01 0.98
 5 (highest) 20.19 22.91 0.12 0.84 20.19 20.87 0.03 0.96
 Unknown 9.83 10.62 0.01 0.98 9.83 9.57 0.01 1.04
Education (%)
 Less than secondary 8.81 5.36 0.16 1.44 8.81 8.67 0.01 1.02
 Secondary graduate 11.82 10.15 0.03 1.07 11.82 12.12 0.01 0.98
 More than secondary 73.92 78.52 0.13 1.12 73.92 74.11 <0.01 1.00
 Unknown 5.45 5.97 0.02 0.90 5.45 5.09 0.01 1.04
Immigrant status (%)
 Immigrant 28.83 32.98 0.04 0.94 28.83 28.69 0.02 1.03
 Canadian-born 69.73 65.25 0.05 0.93 69.73 69.92 0.02 1.03
 Unknown 1.44 1.78 0.02 0.80 1.44 1.39 <0.01 1.05
Body mass index (%)
 Obese, ≥30 kg/m2 21.07 15.01 0.15 1.24 21.07 21.08 0.02 1.03
 Overweight, 25-29.9 kg/m2 34.76 32.29 0.04 1.03 34.76 34.55 <0.01 1.00
 Normal weight, 18.5-24.9 kg/m2 37.75 45.60 0.15 0.93 37.75 37.85 0.02 0.99
 Unknown 6.42 7.10 0.02 0.95 6.42 6.52 0.01 0.97
Physical activity (%)
 Active/moderately active 43.49 50.96 0.17 0.99 43.49 43.14 0.01 1.00
 Inactive 54.01 46.94 0.17 1.00 54.01 54.51 0.02 1.00
 Unknown 2.49 2.10 0.02 1.17 2.49 2.35 <0.01 0.98
Alcohol consumption (%)
 Heavy/moderate drinker 28.69 28.80 0.05 0.96 28.69 28.60 0.02 0.98
 Light/never drinker 69.70 69.89 0.05 0.95 69.70 69.84 0.02 0.98
 Unknown 1.61 1.31 0.01 1.12 1.62 1.56 <0.01 0.98
Smoking status (%)
 Former or current smoker 52.10 40.32 0.20 0.99 52.10 52.83 <0.01 1.00
 Never smoker 44.20 56.04 0.21 0.96 44.20 43.55 <0.01 1.00
 Unknown 3.70 3.64 0.01 1.04 3.70 3.62 <0.01 1.00
Self-rated general health (%)
 Excellent/very good/good 75.44 90.95 0.44 2.08 75.44 76.48 0.04 1.04
 Fair/poor 24.46 8.98 0.44 2.09 24.46 23.38 0.04 1.04
 Unknown 0.09 0.07 0.02 1.93 0.09 0.14 0.01 1.50
 Chronic disease(s) (%) 44.57 32.58 0.24 1.04 44.57 44.58 0.01 1.00
 ADGs score, mean (median) 7.33 (3.62) 4.53 (1.51) 0.27 1.14 7.33 (3.62) 7.23 (3.25) 0.01 0.95
CCHS cycle (%)
 2003-2004 19.27 18.82 0.01 1.01 19.27 19.40 0.01 0.99
 2005-2006 19.07 19.51 0.02 0.97 19.07 19.44 <0.01 0.99
 2007-2008 21.36 19.69 0.05 1.07 21.36 21.42 0.01 1.01
 2009-2010 20.75 20.58 0.01 0.99 20.75 20.66 0.01 0.98
 2011-2012 19.54 21.40 0.03 0.95 19.54 19.08 0.01 1.02
ADGs, Aggregated Diagnosis Groups; CCHS, Canadian Community Health Survey.
*Data were derived from the Ontario components of Canadian Community Health Survey (2003-2012) linked to health administrative databases. All estimates were weighted using Canadian Community Health Survey sampling weights to provide population estimates.

4.1. Healthcare utilization

The mean number of cause-specific visits per person-year was higher among adults with back problems than those of propensity score-matched adults without back problems (0.46 vs 0.22 in women; 0.41 vs 0.18 in men) (Table 2). The mean number of all-cause physician visits per person-year was also higher in adults with back problems (13.29 vs 11.86 in women; 10.24 vs 9.66 in men). Compared with propensity score-matched adults without back problems, adults with back problems had approximately 2 times the rate of cause-specific visits (RRwomen 2.06, 95% CI 1.88-2.25; RRmen 2.32, 95% CI 2.04-2.64) (Table 2). Women with back problems had an additional 0.24 (95% CI 0.21-0.27) cause-specific visits per person-year than women without back problems, which corresponded to an annual burden of 323,000 cause-specific visits in Ontario in 2019 (Table 3). Men with back problems had an additional 0.23 (95% CI 0.20-0.26) cause-specific visits per person-year than men without back problems. This corresponded to a burden of 267,000 cause-specific visits among men in Ontario annually.

Table 2 - Rate ratios (RRs) and rate differences (RDs) for healthcare utilization and healthcare costs (adjusted to 2018 Canadian dollars) in adults with back problems compared with propensity score-matched adults without back problems, pooled participants surveyed from 2003 to 2012 and followed up to 2018, Canadian Community Health Survey, Ontario, Canada.*
Adults with back problems n = 21,054 females; n = 15,752 males Propensity score-matched adults without back problems n = 21,054 females; n = 15,752 males Effect estimate, 95% CI
Cause-specific visits (number of visits per person-year)
 Women 0.46 (95% CI 0.43 to 0.48) 0.22 (95% CI 0.21 to 0.24) RR 2.06 (95% CI 1.88 to 2.25)
RD 0.24 (95% CI 0.21 to 0.27)
 Men 0.41 (95% CI 0.38 to 0.44) 0.18 (95% CI 0.16 to 0.20) RR 2.32 (95% CI 2.04 to 2.64)
RD 0.23 (95% CI 0.20 to 0.26)
All-cause physician visits (number of visits per person-year)
 Women 13.29 (95% CI 13.03 to 13.55) 11.86 (95% CI 11.58 to 12.15) RR 1.12 (95% CI 1.09 to 1.16)
RD 1.43 (95% CI 1.04 to 1.82)
 Men 10.24 (95% CI 9.95 to 10.53) 9.66 (95% CI 9.38 to 9.95) RR 1.10 (95% CI 1.05 to 1.14)
RD 0.58 (95% CI 0.17 to 0.99)
All-cause ED visits (number of visits per person-year)
 Women 0.55 (95% CI 0.54 to 0.57) 0.49 (95% CI 0.47 to 0.51) RR 1.15 (95% CI 1.09 to 1.20)
RD 0.06 (95% CI 0.03 to 0.09)
 Men 0.49 (95% CI 0.47 to 0.50) 0.45 (95% CI 0.43 to 0.48) RR 1.08 (95% CI 1.02 to 1.15)
RD 0.04 (95% CI 0.01 to 0.07)
All-cause hospitalizations (number of visits per person-year)
 Women 0.32 (95% CI 0.31 to 0.33) 0.29 (95% CI 0.28 to 0.30) RR 1.12 (95% CI 1.07 to 1.17)
RD 0.03 (95% CI 0.02 to 0.04)
 Men 0.31 (95% CI 0.30 to 0.32) 0.30 (95% CI 0.29 to 0.32) RR 1.03 (95% CI 0.97 to 1.10)
RD 0.01 (95% CI −0.01 to 0.03)
Healthcare costs, $CAD (adjusted to 2018)
 Women 23,232.79 (95% CI 22,569.84 to 23,915.45) 19,155.11 (95% CI 18,580.30 to 19,747.73) 1.21 (95% CI 1.16 to 1.27)
 Men 14,848.17 (95% CI 14,236.22 to 15,486.42) 12,786.48 (95% CI 12,269.21 to 13,325.58) 1.16 (95% CI 1.09 to 1.23)
CI, confidence interval; ED, emergency department; RD, rate difference; RR, rate ratio.
*Estimates based on Canadian Community Health Survey sampling weights, and variance estimates based on bootstrap weights computed using balanced repeated replication.
Estimates based on linear (log transformed) regression models.

Table 3 - Annual burden in healthcare utilization and healthcare costs (adjusted to 2018 Canadian dollars) related to back problems extrapolated to adults aged 18 years and older in Ontario in 2019.*
t Ontario population (2019)
Women Men
Cause-specific visits 322,971 (95% CI 282,600 to 363,342) 266,540 (95% CI 231,774 to 301,306)
All-cause physician visits 1,924,368 (95% CI 1,399,540 to 2,449,196) 672,144 (95% CI 197,008 to 1,147,281)
All-cause ED visits 80,743 (95% CI 40,371 to 121,114) 46,355 (95% CI 11,589 to 81,121)
All-cause hospitalizations 40,371 (95% CI 26,914 to 53,828) 11,589 (95% CI −11,589 to 34,766)
Healthcare costs, $CAD (adjusted to 2018) $531,556,242 (95% CI $378,145,073 to $684,967,410) $227,138,385 (95% CI $108,933,715 to $347,660,794)
CAD, Canadian dollars; CI, confidence interval; ED, emergency department visits.
*Applied Canadian Community Health Survey weighted sample prevalence and rate differences in healthcare utilization or incremental costs of back problems to the 2019 population for Ontario.51

Adults with back problems had higher rates of all-cause healthcare utilization than adults without back problems (Table 2). Adults with back problems had 1.1 times the rate of all-cause physician visits than those without back problems (RRwomen 1.12, 95% CI 1.09-1.16; RRmen 1.10, 95% CI 1.05-1.14). Compared with those without back problems, women and men with back problems, respectively, had an additional 1.43 and 0.58 all-cause physician visits per person-year, which corresponded to an annual burden of 1.9 million and 672,000 all-cause physician visits in Ontario (Table 3). Adults with back problems also had approximately 1.1 times the rate of all-cause emergency department visits than those without back problems (RRwomen 1.15, 95% CI 1.09-1.20; RRmen 1.08, 95% CI 1.02-1.15) (Table 2). For all-cause hospitalizations, women with back problems had 1.1 times the rate than women without back problems (RRwomen 1.12, 95% CI 1.07-1.17), whereas no differences in rates among men were found (RRmen 1.03, 95% CI 0.97-1.10) (Table 2).

4.2. Healthcare costs

Compared with propensity score-matched adults without back problems, adults with back problems had approximately 1.2 times the healthcare costs (women: 1.21, 95% CI 1.16-1.27; men: 1.16, 95% CI 1.09-1.23). Incremental annual costs per person were higher in adults with back problems than those without, with an incremental annual cost of $395 CAD (95% CI $281-$509) in women and $196 CAD (95% CI $94-$300) in men (Table 4). At the population level, this corresponded to an annual burden of $531.6 million CAD for women and $227.1 million CAD for men in Ontario (Table 3).

Table 4 - Total and annual healthcare costs adjusted to 2018 Canadian dollars ($CAD) in adults with back problems compared with propensity score-matched adults without back problems, pooled participants surveyed from 2003 to 2012 and followed up to 2018, Canadian Community Health Survey, Ontario, Canada.*
Women Men
Costs in adults with back problems, n = 21,054 Costs in propensity score-matched adults without back problems, n = 21,054 Incremental costs Costs in adults with back problems, n = 15,752 Costs in propensity score-matched adults without back problems, n = 15,752 Incremental costs
Median IQR Median IQR Median difference 95% CI Median IQR Median IQR Median difference 95% CI
Total costs, $CAD 22,776 10,003-54,746 18,667 7,810-47,266 4109 3080-5144 14,410 5,143-44,384 12,334 4,102-41,615 2076 1055-3093
Annual costs, $CAD 2251 1042-5887 1856 804-5166 395 281-509 1481 531-4699 1285 426-4709 196 94-300
Total costs by age at index date (years), $CAD
 18-34 14,755 6,806-26,246 11,894 4,779-23,535 2861 1544-4196 5034 2,142-11,767 3757 1644-9163 1277 721-1835
 35-49 13,914 7,025-29,323 10,854 5,347-22,413 3060 1920-4206 8400 3,846-21,623 6945 2,924-15,906 1455 521-2391
 50-64 23,361 11,202-53,652 18,249 8,030-41,860 5112 3571-6675 20,485 8,426-51,753 17,919 6,610-47,080 2566 89-5047
 65-74 50,803 24,345-103,462 45,373 21,750-99,864 5430 600-10,268 52,050 23,376-106,888 46,976 21,140-100,636 5074 2,270-12,404
 ≥75 81,553 38,076-160,284 75,893 31,792-155,574 5660 2,676-14,026 81,120 34,985-149,400 64,970 30,654-130,285 16,150 9,004-23,374
CAD, Canadian dollars; CI, confidence interval; IQR, interquartile range.
*Estimates were weighted using Canadian Community Health Survey sampling weights to provide population estimates.

4.3. Sensitivity analyses

We conducted a sensitivity analysis using combined self-reported and diagnostic information to define back problems as a more specific definition, to reduce the potential impact of nondifferential misclassification of the exposure. In this analysis, we observed stronger associations between back problems and healthcare utilization and costs than using only self-reported data (Appendix IIIa and IIIb, available at http://links.lww.com/PAIN/B310). Compared with propensity score-matched adults without back problems, adults with back problems had at least 3 times the rate of cause-specific visits (RRwomen 3.03, 95% CI 2.67-3.44; RRmen 4.60, 95% CI 3.98-5.31) and 1.2 times the rate of all-cause physician visits (RRwomen 1.25, 95% CI 1.18-1.32; RRmen 1.22, 95% CI 1.12-1.34) (Appendix IIIb, available at http://links.lww.com/PAIN/B310). Compared with adults without back problems, adults with back problems had about 1.4 times the rate of all-cause emergency department visits and 1.4 times the rate of all-cause hospitalizations. Adults with back problems also had about 1.7 times the healthcare costs than those without back problems (women: 1.68, 95% CI 1.52-1.85; men: 1.65, 95% CI 1.43-1.90).

We also conducted a sensitivity analysis by further incorporating diagnostic codes for thoracic and rib pain in the lookback window to define the exposure. In this analysis, few individuals were added to the sample (ie, 14 additional individuals with the exposure in the propensity score-matched cohort) (Appendix IV, available at http://links.lww.com/PAIN/B310). The results were similar to the previous sensitivity analysis that combined self-reported and diagnostic information to define back problems [cause-specific utilization (RRwomen 2.97, 95% CI 2.59-3.40; RRmen 4.04, 95% CI 3.44-4.74), all-cause physician visits (RRwomen 1.21, 95% CI 1.14-1.28; RRmen 1.14, 95% CI 1.05-1.25), all-cause emergency department visits (RRwomen 1.27, 95% CI 1.16-1.39; RRmen 1.09, 95% CI 0.96-1.23), all-cause hospitalization (RRwomen 1.27, 95% CI 1.18-1.36; RRmen 1.10, 95% CI 0.97-1.24), and costs (women: 1.37, 95% CI 1.26-1.49; men: 1.31, 95% CI 1.15-1.50)]. When incorporating additional thoracic and rib pain codes to define cause-specific utilization, the associations remained unchanged from the primary analysis.

We found similar results to the primary analysis when ethnicity was included in the propensity score matching. In this analysis, compared with adults without back problems, adults with back problems had about 2 times the rate of cause-specific visits (RRwomen 2.06, 95% CI 1.89-2.26; RRmen 2.31, 95% CI 2.03-2.63), 1.1 times the rate of all-cause physician visits (RRwomen 1.13, 95% CI 1.09-1.16; RRmen 1.09, 95% CI 1.04-1.13), 1.1 times the rate of all-cause emergency department visits (RRwomen 1.15, 1.10-1.21; RRmen 1.10, 95% CI 1.05-1.16), 1.1 times the rate of all-cause hospitalizations (RRwomen 1.12, 95% CI 1.06-1.17; RRmen 1.06, 95% CI 1.01-1.12), and 1.2 times the costs (women: 1.20, 95% CI 1.15-1.25; men: 1.22, 95% 1.15-1.30) (Appendix V, available at http://links.lww.com/PAIN/B310). In addition, back problems were associated with opioid use. Compared with propensity score-matched adults without back problems, adults with back problems had at least 2 times the risk of opioid prescriptions (RRwomen 2.37, 95% CI 1.25-4.47; RRmen 2.35, 95% CI 0.96-5.73) (Appendix VI, available at http://links.lww.com/PAIN/B310).

Based on our quantitative bias analysis, unmeasured confounding (eg, allied health care with chiropractic or physiotherapy) attenuates the association slightly, but a strong association between back problems and healthcare utilization remains (RRwomen 1.40 and RRmen 1.49) (Appendix VII, available at http://links.lww.com/PAIN/B310).

5. Discussion

We found that adults with back problems had higher rates of healthcare utilization and costs than propensity score-matched adults without back problems. Overall, adults with back problems had approximately 2 times the rate of cause-specific visits, 1.1 times the rate of all-cause physician visits, 1.1 times the rate of all-cause emergency department visits, and 1.2 times the healthcare costs than those without back problems. Compared with those without back problems, higher rates of all-cause hospitalizations were found among women with back problems. Incremental annual costs per person were also higher in adults with back problems than those without. The incremental annual cost was $395 (95% CI $281-$509) in women and $196 (95% CI $94-$300) in men. This corresponded to an annual burden (adjusted to 2018 CAD) of $531.6 million for women and $227.1 million for men within a single-payer health system, representing a substantial economic burden provincially. When extrapolated to the general adult population nationally in 2019,50,55 this corresponds to an annual burden of $1.36 billion CAD for women and $589 million for men in Canada, and an annual burden of $8.90 billion USD for women and $3.76 billion USD for men in the United States.

Our findings build on previous findings on healthcare utilization and costs for back problems. Rampersaud et al.40 reported 1.6 million outpatient physician visits for spinal conditions among Ontario adults in 2013 to 2014. We reported an estimated 1 million cause-specific visits in Ontario in 2019, which suggests that back problems are responsible for a large proportion of spine care visits. The Economic Burden of Illnesses in Canada 2010 reported direct costs, which includes health expenditure and formal caregiving costs, for a range of illnesses.39 Based on this report, the costliest illnesses in Canada were diseases of the digestive system ($19.2 billion), injuries ($13.5 billion), diseases of the circulatory system ($13.1 billion), mental disorders ($10.5 billion), and musculoskeletal diseases ($6.8 billion).39 The conditions considered under the category of musculoskeletal diseases included hip and knee arthritis, rheumatoid arthritis, systemic connective tissue disorders, osteoporosis, and intervertebral and soft tissue disorders, of which back problems would be included. In light of this, our findings suggest that the economic burden of back problems contributes to a considerable proportion of annual costs for all musculoskeletal diseases in Canada. Our study also advances our knowledge of incremental annual costs for back problems because it is higher than previously estimated. Iron et al.31 reported an incremental physician cost of $96.25 for back pain among Ontario adults in 1994. Our incremental annual costs were $395 (95% CI $281-$509) in women and $196 (95% CI $94-$300) in men, which are higher when considering costs across all major sectors of healthcare spending for back problems.

5.1. Strengths and limitations

There are several strengths and unique contributions of our study. First, CCHS data are a unique source of population data on back problems, which was lacking given the challenges with coding back problems in administrative databases.60 The CCHS data are representative of 98% of the community-dwelling Canadian population aged 12 years and older.49 We also used a more specific definition of back problems by incorporating diagnostic information in our sensitivity analysis. Second, each CCHS respondent was linked individually and deterministically to population-based health administrative databases.41 This data linkage allowed us to capture all medical encounters, including physician visits and hospitalizations in the publicly funded single-payer system of Ontario, providing comprehensive utilization estimates generalizable to the population. Third, the costing methodology used direct person-level healthcare cost data to generate total healthcare spending. This serves as a comprehensive estimate of costs across all major sectors and represents actual costs to the healthcare payer instead of cost projections as performed in previous studies or costing approaches limited by recall. Finally, we used rigorous methods to develop a propensity score-matched cohort to closely match adults with and without self-reported back problems on a wide range of potential confounders to more accurately estimate the direct health system costs associated with back problems. Although propensity score matching does not address unmeasured confounders that may lead to differences in costs and utilization, we conducted a quantitative bias analysis to estimate the extent to which unmeasured confounders may explain the reported association between back problems and healthcare utilization.33 Our quantitative bias analysis suggests that unmeasured confounding attenuates the association slightly; however, a strong association between back problems and healthcare utilization remains.

Our study has limitations. First, CCHS and administrative data were only linked for those who agreed to linkage; however, the linkage rate was very high at 81% to 85%. Previous analyses found coverage rates of linkage between CCHS and administrative data to be adequate for individuals aged 12 to 74 years and similar between males and females.44 Although coverage rates were lower for individuals aged 75 years and older, this was primarily due to residents of institutions who were excluded from our cohort (ie, excluded from the CCHS sampling frame) and thus unlikely to impact results. In addition, we accounted for any minor differences in our analysis by applying survey weights provided by Statistics Canada, which adjust for nonparticipation in the survey and linkage. Second, because CCHS captures self-reported data, measurement error may arise due to social desirability bias or problems with recall. However, a prevalence of 21% for back problems based on self-reported data is similar to the global prevalence of 20% for chronic LBP reported in a systematic review, suggesting unlikely under-reporting or poor recall in our study.26 Moreover, we used a more specific definition of back problems by combining self-report with diagnostic information in our sensitivity analysis. This sensitivity analysis suggests that our estimates likely underestimate the association between back problems and healthcare utilization and costs. This is likely because our sensitivity analysis reduced the impact of nondifferential misclassification of back problems. Third, billing and procedural codes for back problems in health administrative data exclude services not covered by OHIP, such as allied healthcare utilization (eg, chiropractic care and community-based physiotherapy). Our study assesses direct costs to the healthcare payer and does not include indirect costs, likely underestimating the economic burden of back problems. We also considered other healthcare utilization outside of the provincial health insurance plan of OHIP (eg, chiropractic and physiotherapy paid through extended health insurance, workers' compensation, or auto insurance) as an unmeasured confounder in our quantitative bias analysis. Finally, the CCHS sampling frame includes individuals living in private dwellings only, and thus results may not be generalizable to other populations (eg, persons living in institutions or on reserve and other First Nations settlements).

6. Conclusion

Adults with back problems have higher cause-specific and all-cause healthcare utilization and costs than adults without back problems. Our study provides comprehensive estimates for healthcare utilization and incremental costs for back problems in Ontario that account for a wide range of confounders. Our findings will guide policy and decision makers by informing healthcare planning, monitoring of health system burden, and future research for back problems. Importantly, the comprehensive cost estimates can serve as high-quality reference data for future cost-effectiveness and cost-utility analyses. Given the substantial health and economic burden, new strategies to reduce the healthcare utilization and costs associated with back problems are warranted.

Conflict of interest statement

The authors have no conflicts of interest to declare.

Ethics approval: This project received ethics approval from the Health Sciences Research Ethics Board at the University of Toronto.

Appendix A. Supplemental digital content

Supplemental digital content associated with this article can be found online at http://links.lww.com/PAIN/B310.

Acknowledgments

Funding for this study was supported by the Canada Research Chair held by Dr L.C. Rosella. Dr J.J. Wong is funded by the Canadian Institutes of Health Research Frederick Banting and Charles Best Canada Graduate Scholarships Doctoral Award and the tuition assistance program at the Canadian Memorial Chiropractic College. Dr L.C. Rosella is funded by a Tier 2 Canada Research Chair in Population Health Analytics. Professor P. Côté is funded by a Tier 2 Canada Research Chair in Disability Prevention and Rehabilitation. Dr A.C. Tricco is funded by a Tier 2 Canada Research Chair in Knowledge Synthesis.

Author Contributions: J.J. Wong: conceptualization, methodology, formal analysis, and writing—original draft, review, and editing; P. Côté: methodology, supervision, and writing—review and editing; A.C. Tricco: methodology, supervision, and writing—review and editing; T. Watson: data curation and writing—review and editing; L.C. Rosella: conceptualization, methodology, supervision, and writing—review and editing.

This study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC). Parts of this material are based on data and information compiled and provided by MOH and the Canadian Institute for Health Information (CIHI). The analyses, conclusions, opinions, and statements expressed herein are solely those of the authors, and do not reflect those of the funding or data sources; no endorsement is intended or should be inferred. Parts of this material are based on data and/or information compiled and provided by CIHI. However, the analyses, conclusions, opinions and statements expressed in the material are those of the authors, and not necessarily those of CIHI.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

The dataset from this study is held securely in coded form at ICES. While legal data sharing agreements between ICES and data providers (eg, healthcare organizations and government) prohibit ICES from making the dataset publicly available, access may be granted to those who meet prespecified criteria for confidential access, available at https://www.ices.on.ca/DAS (email [email protected]). The full dataset creation plan and underlying analytic code are available from the authors upon request, understanding that the computer programs may rely on coding templates or macros that are unique to ICES and are therefore either inaccessible or may require modification.

References

[1]. Alkherayf F, Agbi C. Cigarette smoking and chronic low back pain in the adult population. Clin Invest Med 2009;32:E360–367.
[2]. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009;28:3083–107.
[3]. Austin PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat 2011;10:150–61.
[4]. Austin PC, Jembere N, Chiu M. Propensity score matching and complex surveys. Stat Methods Med Res 2018;27:1240–57.
[5]. Austin PC, Small DS. The use of bootstrapping when using propensity-score matching without replacement: a simulation study. Stat Med 2014;33:4306–19.
[6]. Austin PC, Walraven C. The mortality risk score and the ADG score: two points-based scoring systems for the Johns Hopkins aggregated diagnosis groups to predict mortality in a general adult population cohort in Ontario, Canada. Med Care 2011;49:940–7.
[7]. Bank of Canada. Annual Exchange Rates. 2018. Available at: https://www.bankofcanada.ca/rates/exchange/annual-average-exchange-rates/. Accessed September 28, 2020.
[8]. Barlow WE. Overview of methods to estimate the medical costs of cancer. Med Care 2009;47(7 suppl 1):S33–36.
[9]. Beyera GK, O'Brien J. Health-care utilisation for low back pain: a systematic review and meta-analysis of population-based observational studies. Rheumatol Int 2019;39:1663–79.
[10]. Bielecky A, Chen C, Ibrahim S, Beaton DE, Mustard CA, Smith PM. The impact of co-morbid mental and physical disorders on presenteeism. Scand J Work Environ Health 2015;41:554–64.
[11]. Bilandzic A, Rosella L. The cost of diabetes in Canada over 10 years: applying attributable health care costs to a diabetes incidence prediction model. Health Promot Chronic Dis Prev Can 2017;37:49–53.
[12]. Bouck Z, Pendrith C, Chen XK, Frood J, Reason B, Khan T, Costante A, Kirkham K, Born K, Levinson W, Bhatia RS. Measuring the frequency and variation of unnecessary care across Canada. BMC Health Serv Res 2019;19:446.
[13]. Carey TS, Evans A, Hadler N, Kalsbeek W, McLaughlin C, Fryer J. Care-seeking among individuals with chronic low back pain. Spine (Phila Pa 1976) 1995;20:312–17.
[14]. Carey TS, Evans AT, Hadler NM, Lieberman G, Kalsbeek WD, Jackman AM, Fryer JG, McNutt RA. Acute severe low back pain. A population-based study of prevalence and care-seeking. Spine (Phila Pa 1976) 1996;21:339–44.
[15]. Cassidy JD, Carroll LJ, Côté P. The Saskatchewan health and back pain survey. The prevalence of low back pain and related disability in Saskatchewan adults. Spine (Phila Pa 1976) 1998;23:1860–6; discussion 1867.
[16]. Cassidy T, Fortin A, Kaczmer S, Shumaker JTL, Szeto J, Madill SJ. Relationship between back pain and urinary incontinence in the Canadian population. Phys Ther 2017;97:449–54.
[17]. Chen H, Kwong JC, Copes R, Villeneuve PJ, Goldberg MS, Ally SL, Weichenthal S, van Donkelaar A, Jerrett M, Martin RV, Brook JR, Kopp A, Burnett RT. Cohort profile: the ONtario population health and environment cohort (ONPHEC). Int J Epidemiol 2016;46:405–405j.
[18]. Côté P, Cassidy JD, Carroll L. The treatment of neck and low back pain: who seeks care? who goes where? Med Care 2001;39:956–67.
[19]. Cypress BK. Characteristics of physician visits for back symptoms: a national perspective. Am J Public Health 1983;73:389–95.
[20]. Dieleman JL, Baral R, Birger M, Bui AL, Bulchis A, Chapin A, Hamavid H, Horst C, Johnson EK, Joseph J, Lavado R, Lomsadze L, Reynolds A, Squires E, Campbell M, DeCenso B, Dicker D, Flaxman AD, Gabert R, Highfill T, Naghavi M, Nightingale N, Templin T, Tobias MI, Vos T, Murray CJ. US spending on personal health care and public health. JAMA 2016;316:2627–46.
[21]. Gershon AS, Wang C, Guan J, Vasilevska-Ristovska J, Cicutto L, To T. Identifying individuals with physician diagnosed COPD in health administrative databases. COPD 2009;6:388–94.
[22]. Gunz AC, Canizares M, Mackay C, Badley EM. Magnitude of impact and healthcare use for musculoskeletal disorders in the paediatric: a population-based study. BMC Musculoskelet Disord 2012;13:98.
[23]. Hart LG, Deyo RA, Cherkin DC. Physician office visits for low back pain. Frequency, clinical evaluation, and treatment patterns from a U.S. national survey. Spine (Phila Pa 1976) 1995;20:11–19.
[24]. Hartvigsen J, Hancock MJ, Kongsted A, Louw Q, Ferreira ML, Genevay S, Hoy D, Karppinen J, Pransky G, Sieper J, Smeets RJ, Underwood M. What low back pain is and why we need to pay attention. Lancet 2018;391:2356–67.
[25]. Hirsch O, Strauch K, Held H, Redaelli M, Chenot JF, Leonhardt C, Keller S, Baum E, Pfingsten M, Hildebrandt J, Basler HD, Kochen MM, Donner-Banzhoff N, Becker A. Low back pain patient subgroups in primary care: pain characteristics, psychosocial determinants, and health care utilization. Clin J Pain 2014;30:1023–32.
[26]. Hoy D, Bain C, Williams G, March L, Brooks P, Blyth F, Woolf A, Vos T, Buchbinder R. A systematic review of the global prevalence of low back pain. Arthritis Rheum 2012;64:2028–37.
[27]. Hoy D, March L, Brooks P, Blyth F, Woolf A, Bain C, Williams G, Smith E, Vos T, Barendregt J, Murray C, Burstein R, Buchbinder R. The global burden of low back pain: estimates from the Global Burden of Disease 2010 study. Ann Rheum Dis 2014;73:968–74.
[28]. Hux JE, Ivis F, Flintoft V, Bica A. Diabetes in Ontario: determination of prevalence and incidence using a validated administrative data algorithm. Diabetes Care 2002;25:512–16.
[29]. ICES. Data Dictionary. OHIP Library, 2019. Available at: https://datadictionary.ices.on.ca/Applications/DataDictionary/Library.aspx?Library=OHIP. Accessed April 1, 2019.
[30]. Iles RA, Davidson M, Taylor NF. Psychosocial predictors of failure to return to work in non-chronic non-specific low back pain: a systematic review. Occup Environ Med 2008;65:507–17.
[31]. Iron KS, Manuel DG, Williams J. Using a linked data set to determine the factors associated with utilization and costs of family physician services in Ontario: effects of self-reported chronic conditions. Chronic Dis Can 2003;24:124–32.
[32]. Khan NA, Quan H, Bugar JM, Lemaire JB, Brant R, Ghali WA. Association of postoperative complications with hospital costs and length of stay in a tertiary care center. J Gen Intern Med 2006;21:177–80.
[33]. Lash TL, Fox MP, Fink AK. Applying quantitative bias analysis to epidemiologic data. New York: Springer Science & Business Media, 2011.
[34]. Lim KL, Jacobs P, Klarenbach S. A population-based analysis of healthcare utilization of persons with back disorders: results from the Canadian Community Health Survey 2000-2001. Spine (Phila Pa 1976) 2006;31:212–18.
[35]. Mortimer M, Ahlberg G. To seek or not to seek? Care-seeking behaviour among people with low-back pain. Scand J Public Health 2003;31:194–203.
[36]. Ontario Ministry of Health and Long-term Care. Understanding health care in Ontario, 2012. Available at: http://www.health.gov.on.ca/en/ministry/hc_system/. Accessed April 1, 2019.
[37]. Patten SB, Williams JV, Wang J. Mental disorders in a population sample with musculoskeletal disorders. BMC Musculoskelet Disord 2006;7:37.
[38]. Pinheiro MB, Ferreira ML, Refshauge K, Maher CG, Ordonana JR, Andrade TB, Tsathas A, Ferreira PH. Symptoms of depression as a prognostic factor for low back pain: a systematic review. Spine J 2016;16:105–16.
[39]. Public Health Agency of Canada. Economic Burden of Illness in Canada. 2010. Available at: https://www.canada.ca/content/dam/phac-aspc/documents/services/publications/science-research/economic-burden-illness-canada-2010/economic-burden-illness-canada-2010.pdf.
[40]. Rampersaud YR, Power JD, Perruccio AV, Paterson JM, Veillette C, Coyte PC, Badley EM, Mahomed NN. Healthcare utilization and costs for spinal conditions in Ontario, Canada - opportunities for funding high-value care: a retrospective cohort study. Spine J 2020;20:874–81.
[41]. Roos LL, Wajda A. Record linkage strategies. Part I: estimating information and evaluating approaches. Methods Inf Med 1991;30:117–23.
[42]. Rosella LC, Fitzpatrick T, Wodchis WP, Calzavara A, Manson H, Goel V. High-cost health care users in Ontario, Canada: demographic, socio-economic, and health status characteristics. BMC Health Serv Res 2014;14:532.
[43]. Rosella LC, Lebenbaum M, Fitzpatrick T, O'Reilly D, Wang J, Booth GL, Stukel TA, Wodchis WP. Impact of diabetes on healthcare costs in a population-based cohort: a cost analysis. Diabet Med 2016;33:395–403.
[44]. Rotermann M. Evaluation of the coverage of linked Canadian Community Health Survey and hospital inpatient records. Health Rep 2009;20:45–51.
[45]. Schofield DJ, Shrestha RN, Passey ME, Earnest A, Fletcher SL. Chronic disease and labour force participation among older Australians. Med J Aust 2008;189:447–50.
[46]. Schultz SE, Rothwell DM, Chen Z, Tu K. Identifying cases of congestive heart failure from administrative data: a validation study using primary care patient records. Chronic Dis Inj Can 2013;33:160–6.
[47]. Statistics Canada. Canada at a Glance 2018: Population. 2018. Available at: https://www150.statcan.gc.ca/n1/pub/12-581-x/2018000/pop-eng.htm?HPA=1. Accessed April 1, 2019.
[48]. Statistics Canada. Canadian Community Health Survey 2003: User Guide for the Public Use Microdata File. Ottawa: Statistics Canada, 2005; (Catalogue no. 82M0013GPE).
[49]. Statistics Canada. Canadian Community Health Survey (CCHS)—Annual component. Available at: http://www23.statcan.gc.ca/imdb-bmdi/document/3226_D7_T9_V8-eng.htm Accessed October 20, 2018.
[50]. Statistics Canada. Population estimates on July 1st, by age and sex. 2019. Available at: https://www150.statcan.gc.ca/t1/tbl1/en/cv.action?pid=1710000501. Accessed August 25, 2020.
[51]. Steenstra IA, Munhall C, Irvin E, Oranye N, Passmore S, Van Eerd D, Mahood Q, Hogg-Johnson S. Systematic review of prognostic factors for return to work in workers with sub acute and chronic low back pain. J Occup Rehabil 2017;27:369–81.
[52]. STROBE Statement. Strengthening the reporting of observational studies in epidemiology: STROBE checklists. 2009. Available at: https://www.strobe-statement.org/?id=available-checklists. Accessed August 3, 2020.
[53]. Thomas S, Wannell B. Combining cycles of the Canadian community health survey. Health Rep 2009;20:53–8.
[54]. Tu K, Chen Z, Lipscombe LL. Prevalence and incidence of hypertension from 1995 to 2005: a population-based study. CMAJ 2008;178:1429–35.
[55]. United States Census Bureau. National Population by Characteristics: 2010-2019, 2020. Available at: https://www.census.gov/data/tables/time-series/demo/popest/2010s-national-detail.html. Accessed October 20, 2020.
[56]. Walker BF. The prevalence of low back pain: a systematic review of the literature from 1966 to 1998. J Spinal Disord 2000;13:205–17.
[57]. Ward MM, Javitz HS, Smith WM, Bakst A. A comparison of three approaches for attributing hospitalizations to specific diseases in cost analyses. Int J Technol Assess Health Care 2000;16:125–36.
[58]. Wodchis W, Bushmeneva K, Nikitovic M, McKillop I. Guidelines on Person level Costing Using Administrative Databases in Ontario. Toronto: Health System Performance Research Network, 2013.
[59]. Wong JJ, Côté P, Tricco AC, Rosella LC. Examining the effects of low back pain and mental health symptoms on healthcare utilisation and costs: a protocol for a population-based cohort study. BMJ Open 2019;9:e031749.
[60]. Wong JJ, Côté P, Tricco AC, Watson T, Rosella LC. Assessing the validity of health administrative data compared to population health survey data for the measurement of low back pain. PAIN 2020;162:219.
[61]. Wu A, March L, Zheng X, Huang J, Wang X, Zhao J, Blyth FM, Smith E, Buchbinder R, Hoy D. Global low back pain prevalence and years lived with disability from 1990 to 2017: estimates from the Global Burden of Disease Study 2017. Ann Transl Med 2020;8:299.
Keywords:

Back pain; Health care utilization; Costs; Health system; Cohort study

Supplemental Digital Content

© 2021 International Association for the Study of Pain