Abstract
Study Design
A retrospective cohort study.
Purpose
To evaluate the association between preoperative opioid use and discharge disposition following major spine surgery and between discharge disposition and opioid availability through 1 year postoperatively.
Overview of Literature
Preoperative opioid use is prevalent in spine surgery and is associated with larger postoperative opioid consumption, longer hospitalizations, increased healthcare expenses, and greater risk of surgical revision. However, whether preoperative opioid use is associated with discharge disposition following major spine surgery, which may serve as an indicator of postoperative functional recovery, remains unclear.
Methods
This retrospective population-based cohort study incorporated comprehensive prescription opioid information for 2223 adults (age ≥18 years) undergoing spine surgery in Olmsted County, Minnesota, between January 1, 2005, and December 31, 2016. Multivariable models were employed to assess the relationships among preoperative opioid exposures, postoperative opioid exposures, and discharge disposition (home, inpatient rehabilitation facility [IRF], and skilled nursing facility [SNF]).
Results
A total of 2,223 adults were included with the following preoperative opioid availability: none (778 [35.0%]), short term (1,118 [50.3%]), episodic (227 [10.2%]), and long term (100 [4.5%]). Discharge dispositions were home (1,984 [89.2%]), IRF (94 [4.2%]), and SNF (145 [6.5%]). Compared with patients with no preoperative opioid availability, those with short-term or episodic opioid availability are less likely to be discharged to an IRF (odds ratio [OR], 0.56; 95% confidence interval [CI], 0.36–0.87; p=0.010). Patients with long-term opioid availability had significantly increased odds of SNF discharge (OR, 2.93; 95% CI, 1.39–6.17; p=0.005). At 1-year follow-up, patients discharged to IRF had an increased likelihood of long-term postoperative opioid availability compared with those discharged home (OR, 12.49; 95% CI, 4.84–32.24; p<0.001).
Conclusions
Preoperative opioid prescribing was associated with post-hospitalization discharge disposition, which in turn was associated with opioid prescribing patterns 1 year postoperatively. Assessing opioid prescribing trends preoperatively may guide discussions regarding anticipated discharge disposition following spine surgery.
Keywords: Spine, Patient discharge, Opioid: Postoperative period, Surgery
GRAPHICAL ABSTRACT
Introduction
In the last decade, opioid expenditures for spine-related pain have surged by 660% [1–3], and spine surgeries were often accompanied by prevalent prescriptions of opioids postoperatively [4–10]. Preoperative opioid use is associated with increased postoperative morbidity, prolonged postoperative opioid use, and protracted recovery, which contribute to suboptimal spine surgery outcomes [11–14]. In addition, discharge disposition is associated with outcomes and expenses following spine surgery [15,16], and nonhome discharges incur greater expenses and higher complication rates [11,12].
Considering the growing emphasis on value-based healthcare, identifying clinical factors influencing discharge disposition becomes crucial. A study identified several factors (e.g., age, mobility, marital status, and duration of hospital stay) associated with discharge disposition [13]. However, data on the association between preoperative opioid use and postoperative discharge disposition following major spine surgery are limited.
The primary aim of this study was to investigate the relationships between preoperative opioid use and postoperative discharge disposition following major spine surgery. The secondary aim was to explore whether discharge disposition is linked to postoperative opioid use through the first year postoperatively to promote shared decision-making regarding anticipated postoperative outcomes and potentially reducing post-discharge outpatient consumption of opioids.
Materials and Methods
This population-based cohort study employed the Rochester Epidemiology Project (REP), a unique medical record linkage system that collects and links healthcare records across healthcare institutions that serve southeastern Minnesota [14,17]. The study was approved by the Institutional Review Board of the Mayo Clinic (IRB approval no., 17-010593) and Olmsted Medical Center (IRB approval no., 002-OMC-18) Rochester, Minnesota, which waived the requirement for written informed consent because of minimal risk to patients. The Strengthening the Reporting of Observational Studies in Epidemiology guidelines were employed in the design and conduct of this study [18].
Study population
All adult patients (age ≥18 years) residing in 11 counties included in the REP, who underwent major spine surgery between January 1, 2005, and December 31, 2016, were identified based on the Current Procedural Terminology codes, as validated in our previous work [19]. Patients who denied research authorization for medical record use in accordance with state law (MN Statute 144.295) and patients who died during index hospitalization were excluded. For patients who underwent multiple procedures during the study period, only the first qualifying procedure was analyzed. To ensure adequate access to preoperative prescription data, patients who had not lived in the study area for a minimum of 180 days before surgery were excluded.
Prescription information and opioid use definitions
All outpatient prescriptions for opioid analgesics given to the study cohort from 180 days up to 1 year following the index surgical procedure were obtained from the REP prescription database, which provides comprehensive medication prescription data for REP participants [14]. Prescriptions were classified based on the medication, dosage, and route in accordance with previous work [19]. The calculated number of days of opioid use for each prescription was based on the premise that patients consumed the medication at the highest prescribed rate. Opioid prescriptions were quantified by calculating the average daily oral morphine milligram equivalents based on the maximum prescribed rate for each unique prescription, following the opioid calculation tool of the Centers for Disease Control and Prevention [20].
Definitions of prescription opioid availability
The preoperative prescription opioid availability was quantified in accordance with the Consolidated Standards of Reporting Trials (CONSORT) definitions [21]: (1) none: patients with no opioid prescriptions within 180 days preoperatively; (2) short term: patients with a prescribing period of 90 days within the 180 days preoperatively; (3) long term: patients with a prescribing period of ≥90 days and either >10 prescriptions during this period or opioids available for >120 days based on the maximum prescribed rate; and (4) episodic: patients with a prescribing period of >90 days but who did not meet the criteria for long-term availability.
In addition, postoperative availability of prescribed opioids was defined using CONSORT definitions based on opioid prescription data between 181 and 365 days after hospital discharge, categorizing patients into similar groups: none, short-term, episodic, or long-term use. For the simplicity of analysis and interpretation, short-term and episodic opioid availabilities were condensed into a single opioid category.
Outcomes
The primary outcome of interest was post-index hospitalization discharge disposition following major spine surgery: home (with or without home health assistance), inpatient rehabilitation facility (IRF), or skilled nursing facility (SNF). SNF primarily focuses on providing skilled care, including medical and therapy services, daily (i.e., 5–7 days a week a minimum of 60 minutes per day) [22]. In contrast, IRF offers a more intensive and coordinated interdisciplinary approach, requiring at least 3 hours of therapy daily, 5 days per week, involving multiple therapy disciplines, physician supervision, and rehabilitation nursing care [23]. In addition, patients admitted to IRF must meet specific criteria, including any postoperative neurological deficits, spinal cord injury, or preexisting medical comorbidities. The second outcome of interest was to evaluate whether discharge disposition (exposure) is associated with postoperative opioid availability in accordance with the CONSORT definitions through 1 year postoperatively.
Statistical methods
Patient characteristics, preoperative and operative information, discharge location, and postoperative outcomes were summarized in a tabular form. Continuous variables were presented as median (25th and 75th percentiles) and compared using the Kruskal-Wallis rank-sum tests, whereas categorical variables were presented as frequency (percentage) and compared using Pearson’s chi-square tests. Multivariable generalized logit models were utilized to evaluate the relationship between (1) preoperative opioid availability and discharge location (SNF and IRF versus home) and (2) discharge location and opioid availability at 1 year post-discharge (short term/episodic and long term versus none). All models were adjusted for age, sex, Charlson comorbidity index, surgery date, and preoperative opioid availability. For the relationship between discharge location and postoperative opioid availability, the analysis was also adjusted for surgical duration as a marker of procedural complexity (underlying spinal pathology). Results were reported as odds ratios (ORs) with 95% confidence intervals (CIs) and associated p-values. Interaction analyses assessed the interplay between age and preoperative opioid use regarding discharge location and between age and discharge location concerning opioid availability at 1 year. All analyses were performed using SAS ver. 9.4M5 (SAS Institute Inc., Cary, NC, USA). All p-values <0.05 were considered significant in the primary analyses, and p<0.1 were considered potentially meaningful in the interaction analyses.
Results
A total of 2,223 patients were enrolled (Fig. 1), including 1,984 patients (89.2%) discharged to their homes, 94 (4.2%) to IRF, and 145 (6.5%) to SNF. A significant variation was noted in patient age across discharge dispositions, including median (Q1–Q3) age of 53 (42–65) years for home, 62 (49–74) years for IRF, and 75 (67–81) years for SNF (p<0.001) (Table 1). Furthermore, patients discharged to SNF were predominantly female (56% versus 45% and 33% for home and IRF, respectively), had the highest comorbidity burden (Charlson comorbidity index 4 [2–7] versus 1 [0–2] and 3 [2–6] for home and IRF, respectively), and the greatest prevalence of comorbid depression, anxiety, alcohol use, and illicit drug use. In addition, the anesthesia duration, surgery duration, instrumentation rates, and hospital length of stay differed across discharge dispositions (p<0.001 for all comparisons) and tended to be greater in patients discharged to an IRF and SNF than in those discharged to home.
Fig. 1.
Flowchart of patient selection. Inclusion criteria: spine surgery at Mayo Clinic from January 1, 2005, through December 31, 2016, age >18 years old, and residency in the 11-county region. Exclusions: withheld medical record authorization, mortality during hospitalization, residency <180 days pre-surgery.
Table 1.
Patient, procedural, and hospitalization characteristics according to discharge disposition
| Characteristic | Home (N=1,984) | IRF (N=94) | SNF (N=145) | p-value |
|---|---|---|---|---|
| Age (yr) | 53 (42–65) | 62 (49–74) | 75 (67–81) | <0.001 |
| Gender | 0.002 | |||
| Male | 1,087 (55) | 63 (67) | 64 (44) | |
| Female | 897 (45) | 31 (33) | 81 (56) | |
| Race (White) | 1,833 (92) | 90 (96) | 136 (94) | 0.41 |
| ASA PS (n=2,150) | <0.001 | |||
| 1–2 | 1,422 (74) | 28 (31) | 46 (33) | |
| 3–4 | 498 (26) | 63 (69) | 93 (67) | |
| Charlson comorbidity index | 1 (0–2) | 3 (2–6) | 4 (2–7) | <0.001 |
| Depression | 740 (37) | 30 (32) | 76 (52) | <0.001 |
| Anxiety | 436 (22) | 16 (17) | 45 (31) | 0.018 |
| Tobacco abuse | 641 (32) | 36 (38) | 52 (36) | 0.35 |
| Alcoholism | 191 (10) | 11 (12) | 21 (14) | 0.15 |
| Illicit drug use | 122 (6) | 8 (9) | 14 (10) | 0.18 |
| Anesthesia duration (mo) | 219 (178–288) | 366 (247–452) | 322 (236–429) | <0.001 |
| Surgery duration (mo) | 141 (106–196) | 219 (156–352) | 218 (138–297) | <0.001 |
| Instrumentation | 487 (25) | 57 (61) | 75 (52) | <0.001 |
| Hospital length of stay (day) | 1 (1–3) | 6 (4–10) | 5 (4–7) | <0.001 |
Values are presented as number (%) for categorical variables and compared using Pearson’s chi-square tests and median (interquartile range) for continuous variables compared using Kruskal-Wallis rank-sum tests.
IRF, inpatient rehabilitation facility; SNF, skilled nursing facility; ASA PS, American Society of Anesthesiologists Physical Status.
Preoperative opioid availability and discharge disposal
Among patients who had no or had short-term or episodic preoperative opioid availability, the majority were discharged home (88% and 90%, respectively), with only 6% in either group discharged to an SNF (Supplement 1). Moreover, 6% and 3% of the patients who had no and had short-term or episodic preoperative opioid availability were discharged to an IRF, respectively. Those with long-term preoperative opioid availability experienced lower rates of home discharge (80%), low rates of IRF discharge (4%), and higher rates of SNF discharge (16%).
In the multivariable analyses, short-term or episodic preoperative opioid availability was not associated with discharge to an SNF (OR, 1.26; 95% CI, 0.83–1.91; p=0.282) but was associated with reduced odds of discharge to an IRF (OR, 0.56; 95% CI, 0.36–0.87; p=0.010) compared with those with no preoperative opioid availability (Table 2). Long-term preoperative opioid availability was associated with increased odds of being discharged to an SNF (OR, 2.93; 95% CI, 1.39–6.17; p=0.005) but was not associated with discharge to an IRF (OR, 0.61; 95% CI, 0.21–1.79; p=0.365).
Table 2.
Estimated association between preoperative opioid availability and discharge disposition
| Exposure variable | Discharge disposition outcome | |||
|---|---|---|---|---|
|
| ||||
| Discharged to inpatient rehabilitation | Discharged to skilled nursing facility | |||
|
|
|
|||
| OR (95% CI) | p-value | OR (95% CI) | p-value | |
| Preoperative opioid availability | ||||
|
| ||||
| None | Reference | Reference | ||
|
| ||||
| Short-term or episodic | 0.56 (0.36–0.87) | 0.010 | 1.26 (0.83–1.91) | 0.282 |
|
| ||||
| Long term | 0.61 (0.21–1.79) | 0.365 | 2.93 (1.39–6.17) | 0.005 |
Results are from multivariable generalized logistic regression. Multiplicative increase in odds for the given outcome vs. discharge to home with or without home health care associated with the given preoperative opioid availability group and corresponding p-values are presented. Covariates included are age (spline with 2 degrees of freedom), sex, Charlson comorbidity index (log), date of surgery, and preoperative opioid availability.
OR, odds ratio; CI, confidence interval.
The effects of preoperative opioid availability on the discharge location differed according to age (interaction p-value=0.007). Short-term or episodic availability was associated with lower odds of IRF discharge in younger patients (OR, 0.29; 95% CI, 0.15–0.57; p<0.001 at age 45 years), whereas it had no significant effect in older patients (OR, 0.75; 95% CI, 0.41–1.35; p=0.332 at age 65 years) (Table 3). Long-term preoperative availability was associated with greater odds of being discharged to an SNF in late- to middle-aged and older patients but not in younger patients (OR, 6.26; 95% CI, 2.31–16.98; p<0.001; OR, 5.64; 95% CI, 1.80–17.60; p=0.003; and OR, 1.47; 0.14–15.43; p=0.748 at ages 65, 55, and 45 years, respectively).
Table 3.
Interaction of age on preoperative opioid and discharge disposition relationships
| Exposure variable | Discharge disposition outcome | |||
|---|---|---|---|---|
|
| ||||
| Discharged to inpatient rehabilitation | Discharged to skilled nursing facility | |||
|
|
|
|||
| OR (95% CI) | p-value | OR (95% CI) | p-value | |
| Preoperative opioid availability | ||||
|
| ||||
| None | Reference | Reference | ||
|
| ||||
| Short-term or episodic | ||||
|
| ||||
| Age 45 yr | 0.29 (0.15–0.57) | <0.001 | 0.39 (0.10–1.49) | 0.167 |
|
| ||||
| Age 55 yr | 0.48 (0.24–0.97) | 0.040 | 1.15 (0.52–2.58) | 0.727 |
|
| ||||
| Age 65 yr | 0.75 (0.41–1.35) | 0.332 | 1.75 (0.92–3.33) | 0.089 |
|
| ||||
| Long term | ||||
|
| ||||
| Age 45 yr | 0.16 (0.01–4.03) | 0.266 | 1.47 (0.14–15.43) | 0.748 |
|
| ||||
| Age 55 yr | 0.89 (0.19–4.21) | 0.879 | 5.64 (1.80–17.60) | 0.003 |
|
| ||||
| Age 65 yr | 1.55 (0.37–6.55) | 0.552 | 6.26 (2.31–16.98) | <0.001 |
Results are from multivariable generalized logistic regression. Multiplicative increase in odds for the given outcome vs. discharge to home with or without home health care associated with the given age by preoperative opioid availability group combination and corresponding p-values are presented. Covariates included are age (spline with 2 degrees of freedom), sex, Charlson comorbidity index (log), date of surgery, and preoperative opioid availability.
OR, odds ratio; CI, confidence interval.
Discharge disposition and opioid availability 1-year postoperatively
A total of 2,158 participants (97.1%) remained in the population-based sample until 12 months postoperatively. Discharge to an IRF was not associated with short-term or episodic availability 12 months postoperatively (OR, 1.60; 95% CI, 0.95–2.68; p=0.077) (Table 4) but was associated with long-term availability (OR, 12.49; 95% CI, 4.84–32.24; p<0.001). Discharge to an SNF was associated with neither short-term/episodic opioid availability (OR, 0.87; 95% CI, 0.55–1.38; p=0.544) nor long-term use (OR, 2.28; 95% CI, 0.83–6.25; p=0.110) 12 months postoperatively.
Table 4.
Estimated association between discharge disposition and postoperative opioid availability at 1 year
| Exposure variable | Opioid availability outcome | |||
|---|---|---|---|---|
|
| ||||
| Short-term/episodic availability | Long-term availability | |||
|
|
|
|||
| OR (95% CI) | p-value | OR (95% CI) | p-value | |
| Discharge disposition | ||||
|
| ||||
| Home | Reference | Reference | ||
|
| ||||
| Inpatient rehabilitation | 1.60 (0.95–2.68) | 0.077 | 12.49 (4.84–32.24) | <0.001 |
|
| ||||
| Skilled nursing facility | 0.87 (0.55–1.38) | 0.544 | 2.28 (0.83–6.25) | 0.110 |
Only patients under follow-up and residing in the catchment area 1 year following hospital discharge are included in the analysis (n=2,148). Results are from multivariable generalized logistic regression. Covariates include age, sex, Charlson comorbidity index (log), date of surgery, surgery duration, and preoperative opioid availability. Multiplicative increase in odds for the given outcome vs. no opioid availability associated with the given discharge disposition and corresponding p-values are presented.
OR, odds ratio; CI, confidence interval.
Potentially meaningful interactions were found by age (interaction p-value=0.053) such that being discharged to an IRF was associated with short-term or episodic availability in younger and early middle-aged patients but not in older patients (Table 5). Being discharged to an IRF was associated with an increased risk of long-term opioid availability at 1 year in all age groups (p<0.001). Discharge to an SNF was similarly associated with an increased risk for long-term opioid availability, and effects were most pronounced in younger patients (p-value=0.008, ages 45 and 55 years).
Table 5.
Interactions of age on discharge disposition and 1-year postoperative opioid availability relationships
| Exposure variable | 1-Year opioid availability | |||
|---|---|---|---|---|
|
| ||||
| Short-term/episodic | Long term | |||
|
|
|
|||
| OR (95% CI) | p-value | OR (95% CI) | p-value | |
| Discharge disposition | ||||
|
| ||||
| Home | Reference | Reference | ||
|
| ||||
| Inpatient rehabilitation | ||||
|
| ||||
| Age 45 yr | 2.35 (1.27–4.35) | 0.007 | 16.97 (5.24–55.00) | <0.001 |
|
| ||||
| Age 55 yr | 1.74 (1.02–2.96) | 0.041 | 13.60 (5.14–36.03) | <0.001 |
|
| ||||
| Age 65 yr | 1.29 (0.73–2.30) | 0.382 | 10.90 (4.10–29.02) | <0.001 |
|
| ||||
| Skilled nursing facility | ||||
|
| ||||
| Age 45 yr | 0.59 (0.15–2.35) | 0.459 | 10.82 (1.87–62.72) | 0.008 |
|
| ||||
| Age 55 yr | 0.67 (0.25–1.76) | 0.414 | 5.33 (1.54–18.37) | 0.008 |
|
| ||||
| Age 65 yr | 0.75 (0.41–1.39) | 0.362 | 2.62 (0.97–7.11) | 0.059 |
Results are from multivariable generalized logistic regression. Multiplicative increase in odds for the given outcome vs. No opioid availability associated with the given age by discharge disposition combination and corresponding p-values are presented. Covariates included are age, sex, Charlson comorbidity index (log), date of surgery, surgery duration, and preoperative opioid availability.
OR, odds ratio; CI, confidence interval.
Discussion
In this population-based cohort study of >2,000 adults who underwent major spine surgery with comprehensive longitudinal follow-up, greater preoperative opioid availability was linked with nonhome discharge disposition. Specifically, those with long-term preoperative opioid availability were more likely to be discharged to an SNF, and those with short-term or episodic preoperative opioid availability were less likely to experience IRF discharge. Importantly, the relationships between preoperative opioid availability and discharge location differed by age, such that greater preoperative opioid availability was consistently associated with nonhome discharges in older adults than in younger adults who underwent major spine surgery. Nonhome discharges were also associated with long-term opioid availability at 1 year, a finding that was most pronounced in younger adults. These findings shed light on the complex relationships between opioid prescribing and discharge outcomes following spine surgery.
This study was built upon previous investigations evaluating the relationships between prescription opioids and postsurgical outcomes in spine surgery. For example, preoperative opioid use has been associated with increased disability, prolonged hospital stays, surgical complications, diminished functional outcomes, delayed return to work, and compromised overall health and well-being [7–9]. Furthermore, the preoperative opioid prescribing rate (65% with any degree of availability preceding surgery) aligns with previous research reporting rates ranging from 20% to >70% [24]. Distinct from our previous work, we focused on the relationships between preoperative prescription of opioids and discharge disposition, important information for preoperative planning and patient-centered shared decision-making regarding optimal management of spine pathology and expected surgery outcomes.
Approximately 90% of patients with no preoperative opioid availability were discharged home, and a similar rate of home discharge was observed for those with short-term or episodic preoperative availability. Conversely, 20% of those with long-term opioid use experienced nonhome discharge, mostly to an SNF, with multivariable analyses confirming a nearly threefold increase in the odds for SNF discharge in the presence of long-term preoperative opioid availability. This study aligns with the prevailing national trend, which indicates a rising tendency to refer older and medically compromised patients to an SNF for postacute care [25]. Numerous factors may drive this relationship. For example, patients with either no or short-term/episodic preoperative opioid availability may have lower pain severity and/or pain duration preoperatively, less complex surgical procedures, greater preoperative functional capacity, or higher levels of social support to facilitate home discharge. Conversely, patients with long-term preoperative opioid use may represent those with long-standing chronic pain, greater medical and mental health comorbidity burden, and baseline functional limitations, which may prevent home discharge. Interestingly, a previous study, albeit not specifically limited to spine surgery, highlighted that 70% of the patients discharged to an SNF from an acute care hospital are prescribed an opioid on discharge [26]. These results highlight the complex interactions between prescription opioid availability and discharge disposition, and those with long-term preoperative opioid exposures face unique postoperative care needs, as reflected by the higher incidence of SNF discharges.
As regards IRF discharge, patients with short-term or episodic preoperative opioid use had a paradoxically lower chance of being discharged to an IRF than those without preoperative opioid availability. Despite the unclear mechanisms for this relationship, the qualification for IRF discharge is based on the individualized rehabilitation potential of the patients, which in turn incorporates the assessments of factors such as pre-morbid functional status, rehabilitation diagnosis and comorbid conditions, level of family and caregiver support, and home living arrangements. Preexisting opioid use is not considered an exclusion criterion for IRF admission. Patients who had preoperative opioid availability may experience a greater prevalence of chronic pain or other related conditions, which may be perceived by clinicians as factors that could interfere with their potential to tolerate the intensity of acute inpatient rehabilitation services. Although rehabilitation is also provided in SNF environments, rehabilitation is not as rigid in scheduling, allowing greater flexibility for those with pain that may be more difficult to manage. Future work is needed to understand the potential barriers to IRF discharge in those with preexisting opioid use undergoing major spine surgery and may benefit from IRF services.
Although previous studies have explored risk factors for prolonged opioid use following spine surgery [6–9,27–29], this study uniquely addresses how discharge disposition may influence longer-term opioid prescribing trends. To that end, patients discharged to nonhome locations experienced higher opioid availability rates 1 year postoperatively, even after adjustment for preoperative opioid exposures and other confounding variables. For example, compared with patients discharged home, those discharged to an IRF had a greater than 12-fold increase in the odds of long-term prescription opioid availability 1 year postoperatively. These findings were observed across all patient age groups but were most prominent among younger adults, with similar but less pronounced findings observed for those discharged to SNF. There are several potential reasons. First, patients discharged from the hospital to an IRF may have undergone more complex spine surgeries, which may be accompanied by increased prescription opioid usage. Second, the higher frequency and dosing of rehabilitation services in IRF may lead to longer-term opioid use. The average length of stay in acute IFs in Olmsted County, Minnesota, is <2 weeks, which may limit the complete full taper of opioids before the expected discharge date. Third, opioid prescribing responsibilities following discharge from an IRF usually transition back to the patient’s surgical team or primary care provider. These hand-offs in opioid prescribing represent substantial gaps in postsurgical care, which may perpetuate longer-term opioid prescribing [30]. Finally, not all patients were discharged from the IRF to home, such that some patients may transfer to SNF where opioid prescribing may be extended.
This study has several important limitations. First, similar to nearly all observational studies of opioid prescriptions, the study relies on opioid prescription data to categorize patients into opioid exposure groups based on prescription opioid availability. However, opioid availability may not uniformly reflect opioid consumption. Second, despite the use of a population-based cohort design with nearly comprehensive follow-up, a few patients (<3%) moved out of the study area or died during the 1-year follow-up. Third, the relationships highlighted in this study represent associations and should be considered hypothesis-generating rather than causal. The potential for residual confounding remains despite prespecified multivariable adjustment. Fourth, the data are representative of residents (and opioid prescribers) in the north–central part of the United States, which does not account for variability across institutions in prescribing practices and criteria employed for discharge disposition planning. Finally, time is an important variable where patterns of opioid prescribing may have shifted during 2005–2016 and the 1-year follow-up period.
Conclusions
Preoperative opioid availability is associated with discharge disposition, with short-term or episodic availability associated with a lower likelihood of IRF discharge and long-term availability associated with a higher chance of SNF discharge. Interestingly, patients discharged to an IRF had substantially higher rates of developing a long-term pattern of opioid prescribing 1 year after surgery than those discharged home. The mechanisms driving these relationships are unclear and warrant further investigations. Given these findings, the assessment of opioid prescribing trends before surgery may guide discussions regarding the anticipated discharge disposition following spine surgery, which may be indicative of longer-term opioid prescribing patterns.
Key Points
Assessing opioid prescribing trends preoperatively could provide information for discussions regarding anticipated discharge disposition and potential long-term opioid prescribing patterns postoperatively.
Patients with long-term preoperative opioid availability are more likely to be discharged to a skilled nursing facility, whereas those with short-term or episodic opioid use are less likely to be discharged to an inpatient rehabilitation facility.
Compared with younger adults who underwent major spine surgery, older adults were consistently more likely to experience nonhome discharges with greater preoperative opioid availability.
Patients discharged to nonhome locations had higher rates of opioid availability 1 year postoperatively.
Footnotes
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
Funding
This work was supported by grant K23AG070113 to Dr. Warner through the National Institute on Aging (NIA).
Acknowledgments
This study used the resources of the Rochester Epidemiology Project (REP) medical records-linkage system, which is supported by the National Institute on Aging (NIA; AG 058738), by the Mayo Clinic Research Committee, and by fees paid annually by REP users. The content of this article is solely the responsibility of the authors and does not represent the official views of the National Institutes of Health (NIH) or the Mayo Clinic.
Author Contributions
SRW helped with formal analysis, wrote the original draft, and reviewed & edited the final manuscript. ACH assisted with formal analysis, data curation, wrote the original draft, and reviewed & edited the formal manuscript. EB, MDM, XS, and WMH analyzed and interpreted the data and finalized the manuscript. NSW conceptualized the design, performed the formal analysis, analyzed and interpreted the data, wrote the original draft, and finalized the manuscript.
Supplementary Materials
Supplementary materials can be available from https://doi.org/10.31616/asj.2024.0414.
Supplement 1. Preoperative opioid availability and discharge disposition.
References
- 1.Walid MS, Hyer L, Ajjan M, Barth AC, Robinson JS., Jr Prevalence of opioid dependence in spine surgery patients and correlation with length of stay. J Opioid Manag. 2007;3:127–8. 130–2. doi: 10.5055/jom.2007.0050. [DOI] [PubMed] [Google Scholar]
- 2.Deyo RA, Von Korff M, Duhrkoop D. Opioids for low back pain. BMJ. 2015;350:g6380. doi: 10.1136/bmj.g6380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ahn J, Bohl DD, Tabaraee E, Aboushaala K, Elboghdady IM, Singh K. Preoperative narcotic utilization: accuracy of patient self-reporting and its association with postoperative narcotic consumption. J Neurosurg Spine. 2016;24:206–14. doi: 10.3171/2015.3.SPINE141300. [DOI] [PubMed] [Google Scholar]
- 4.Bicket MC, White E, Pronovost PJ, Wu CL, Yaster M, Alexander GC. Opioid oversupply after joint and spine surgery: a prospective cohort study. Anesth Analg. 2019;128:358–64. doi: 10.1213/ANE.0000000000003364. [DOI] [PubMed] [Google Scholar]
- 5.Schoenfeld AJ, Nwosu K, Jiang W, et al. Risk factors for prolonged opioid use following spine surgery, and the association with surgical intensity, among opioid-naive patients. J Bone Joint Surg Am. 2017;99:1247–52. doi: 10.2106/JBJS.16.01075. [DOI] [PubMed] [Google Scholar]
- 6.Sharma M, Ugiliweneza B, Aljuboori Z, Nuno MA, Drazin D, Boakye M. Factors predicting opioid dependence in patients undergoing surgery for degenerative spondylolisthesis: analysis from the MarketScan databases. J Neurosurg Spine. 2018;29:271–8. doi: 10.3171/2018.1.SPINE171258. [DOI] [PubMed] [Google Scholar]
- 7.Schoenfeld AJ, Belmont PJ, Jr, Blucher JA, et al. Sustained preoperative opioid use is a predictor of continued use following spine surgery. J Bone Joint Surg Am. 2018;100:914–21. doi: 10.2106/JBJS.17.00862. [DOI] [PubMed] [Google Scholar]
- 8.Pugely AJ, Bedard NA, Kalakoti P, et al. Opioid use following cervical spine surgery: trends and factors associated with long-term use. Spine J. 2018;18:1974–81. doi: 10.1016/j.spinee.2018.03.018. [DOI] [PubMed] [Google Scholar]
- 9.Lawrence JT, London N, Bohlman HH, Chin KR. Preoperative narcotic use as a predictor of clinical outcome: results following anterior cervical arthrodesis. Spine (Phila Pa 1976) 2008;33:2074–8. doi: 10.1097/BRS.0b013e3181809f07. [DOI] [PubMed] [Google Scholar]
- 10.Laufenberg-Feldmann R, Kappis B, Mauff S, Schmidtmann I, Ferner M. Prevalence of pain 6 months after surgery: a prospective observational study. BMC Anesthesiol. 2016;16:91. doi: 10.1186/s12871-016-0261-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Heligman JL. The effect of a discharge disposition algorithm on patient outcomes and satisfaction. Orthop Nurs. 2021;40:125–33. doi: 10.1097/NOR.0000000000000753. [DOI] [PubMed] [Google Scholar]
- 12.Eastlack RK, Ledesma JB, Tran S, et al. Home versus rehabilitation: factors that influence disposition after minimally invasive surgery in adult spinal deformity surgery. World Neurosurg. 2018;118:e610–5. doi: 10.1016/j.wneu.2018.06.249. [DOI] [PubMed] [Google Scholar]
- 13.Kanaan SF, Yeh HW, Waitman RL, Burton DC, Arnold PM, Sharma NK. Predicting discharge placement and health care needs after lumbar spine laminectomy. J Allied Health. 2014;43:88–97. [PMC free article] [PubMed] [Google Scholar]
- 14.Rocca WA, Yawn BP, St Sauver JL, Grossardt BR, Melton LJ., 3rd History of the Rochester Epidemiology Project: half a century of medical records linkage in a US population. Mayo Clin Proc. 2012;87:1202–13. doi: 10.1016/j.mayocp.2012.08.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Passias PG, Poorman GW, Bortz CA, et al. Predictors of adverse discharge disposition in adult spinal deformity and associated costs. Spine J. 2018;18:1845–52. doi: 10.1016/j.spinee.2018.03.022. [DOI] [PubMed] [Google Scholar]
- 16.McCarthy IM, Hostin RA, Ames CP, et al. Total hospital costs of surgical treatment for adult spinal deformity: an extended follow-up study. Spine J. 2014;14:2326–33. doi: 10.1016/j.spinee.2014.01.032. [DOI] [PubMed] [Google Scholar]
- 17.Melton LJ., 3rd History of the Rochester Epidemiology Project. Mayo Clin Proc. 1996;71:266–74. doi: 10.4065/71.3.266. [DOI] [PubMed] [Google Scholar]
- 18.von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61:344–9. doi: 10.1016/j.jclinepi.2007.11.008. [DOI] [PubMed] [Google Scholar]
- 19.Warner NS, Habermann EB, Hooten WM, et al. Association between spine surgery and availability of opioid medication. JAMA Netw Open. 2020;3:e208974. doi: 10.1001/jamanetworkopen.2020.8974. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Centers for Disease Control and Prevention . Opioid overdose: guideline overview [Internet] Atlanta (GA): Centers for Disease Control and Prevention; 2019. [cited 2024 Sep 20]. Available from: https://www.cdc.gov/drugoverdose/prescribing/guideline.html#anchor_1561563251 . [Google Scholar]
- 21.Von Korff M, Saunders K, Thomas Ray G, et al. De facto long-term opioid therapy for noncancer pain. Clin J Pain. 2008;24:521–7. doi: 10.1097/AJP.0b013e318169d03b. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Centers for Medicare & Medicaid Services . Medicare coverage of skilled nursing facility care [Internet] Baltimore (MD): Centers for Medicare & Medicaid Services; 2022. [cited 2024 Sep 20]. Available from: https://www.medicare.gov/Pubs/pdf/10153-Medicare-Skilled-Nursing-Facility-Care.pdf . [Google Scholar]
- 23.Centers for Medicare & Medicaid Services . Medicare Learning Network. Medicare provider compliance tips (MLN4824456) [Internet] Baltimore (MD): Centers for Medicare & Medicaid Services; c2024. [cited 2024 Sep 20]. Available from: https://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network-MLN/MLNProducts/medicare-provider-compliance-tips/medicare-provider-compliance-tips.html#IRF . [Google Scholar]
- 24.Yerneni K, Nichols N, Abecassis ZA, Karras CL, Tan LA. Preoperative opioid use and clinical outcomes in spine surgery: a systematic review. Neurosurgery. 2020;86:E490–507. doi: 10.1093/neuros/nyaa050. [DOI] [PubMed] [Google Scholar]
- 25.Rodriguez A, Edelen MO, McMullen TL, et al. Who receives post-acute care?: characteristics of national population and field test sample. J Am Geriatr Soc. 2022;70:991–1000. doi: 10.1111/jgs.17649. [DOI] [PubMed] [Google Scholar]
- 26.Hubsky AR, Noble BN, Hartung DM, Tjia J, Lapane KL, Furuno JP. Opioid prescribing on discharge to skilled nursing facilities. Pharmacoepidemiol Drug Saf. 2020;29:1183–8. doi: 10.1002/pds.5075. [DOI] [PubMed] [Google Scholar]
- 27.Jain N, Brock JL, Phillips FM, Weaver T, Khan SN. Chronic preoperative opioid use is a risk factor for increased complications, resource use, and costs after cervical fusion. Spine J. 2018;18:1989–98. doi: 10.1016/j.spinee.2018.03.015. [DOI] [PubMed] [Google Scholar]
- 28.Cho J, Place K, Salstrand R, et al. Developing a predictive tool for hospital discharge disposition of patients poststroke with 30-day readmission validation. Stroke Res Treat. 2021;2021:5546766. doi: 10.1155/2021/5546766. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Patel PR, Bechmann S. Discharge planning [Internet] Treasure Island (FL): StatPearls Publishing; 2023. [cited 2024 Sep 20]. Available from: https://www.ncbi.nlm.nih.gov/books/NBK557819/ [PubMed] [Google Scholar]
- 30.Warner NS, Finnie D, Warner DO, et al. The system is broken: a qualitative assessment of opioid prescribing practices after spine surgery. Mayo Clin Proc. 2020;95:1906–15. doi: 10.1016/j.mayocp.2020.02.027. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplement 1. Preoperative opioid availability and discharge disposition.