Insurance Expansion and the Utilization of Inpatient Surgery: Evidence for a "Woodwork" Effect?

Closing the Socioeconomic Gap in Massachusetts: Trends in Total Hip Arthroplasty From 2013 to 2015

To extend insurance coverage to all residents, Massachusetts legislation expanded Medicaid eligibility and added new private insurance categories. To date, no one has analyzed the effect of these changes and compared recent trends in total hip arthroplasty (THA) utilization. Therefore, this study sought to update the current trends of THA utilization in Massachusetts from 2013 to 2015. The Massachusetts State Inpatient Database was queried for all patients who underwent primary THA between 2013 and 2015, and 30,308 patients were identified. Analyzed variables included age, sex, race, Charlson Comorbidity Index, median household income, primary payer, discharge disposition, length of stay, hospital charges, hospital costs, and complications. Categorical and continuous variables were assessed using chi-square analyses and analyses of variance, respectively. Between 2013 and 2015, annual THAs increased from 9361 to 10,562. Race did not vary significantly (P=.447), although an increase in patients using Medicaid and a decrease in patients using other insurance was observed (P<.001). Patients with an income quartile of 1 increased, whereas the number of THA patients in quartile 3 decreased (P<.001). There was a decrease in both hospital charges (P<.001) and costs (P<.001). Mean length of stay decreased (P<.001), and the number of patients with complications decreased (P<.001). Massachusetts has been successful in increasing access to THA procedures for low-income patients and increasing the number of patients who use Medicaid for THAs. The current delivery of health care in Massachusetts has shown improvement for its residents, serving as an example that other states can learn from. [Orthopedics. 2021;44(2):e167-e172.].

The Affordable Care Act: Implications for underserved populations with head & neck cancer

PURPOSE

This study was done to determine the direct impact implementation of the Affordable Care Act (ACA) on patients with Head and Neck Cancer (HNCA) in states that chose to expand Medicaid compared to in states that did not, as well as assess whether this impact varied among different demographic groups.

MATERIALS AND METHODS

The Surveillance, Epidemiology, and End Results (SEER) database was queried for cases of HNCA diagnosed from 2011 to 2014. Rates of uninsured status were compared before and after Medicaid expansion and contrasted between states that did and did not expand coverage, stratified by patient and tumor characteristics, and assessed via multivariate regression.

RESULTS

Overall rates of uninsured status (UR) were decreased by 63.08% in states that expanded coverage (ES) but only by 2.6% in states that did not (NS). In NS, there was an increase in proportion of black patients who were uninsured over the study period (13.7%, p = 0.077) whereas in ES, this proportion decreased by 73.3%. When stratified by primary site, patients with laryngeal cancer had the highest UR with an increase by 16.7% in NS and a decrease by 70.5% in ES. Multivariate analysis yielded predictors of uninsured status including residence in a NS, Hispanic ethnicity, and black race.

CONCLUSIONS

Implementation of the ACA resulted in expanded insurance coverage for patients diagnosed with HNCA concentrated mainly in states that expanded Medicaid coverage and for patients derived from vulnerable populations, including black and Hispanic patients. In states that did not expand Medicaid, vulnerable populations were disproportionately affected.

The Relationship Between Medicaid Coverage and Outcomes Following Total Knee Arthroplasty: A Systematic Review

BACKGROUND

Access to elective total knee arthroplasty is important in the treatment of end-stage arthritis, and numerous initiatives, including Medicaid expansion, have sought to improve patients' ability to undergo this procedure. However, despite this, the role of Medicaid insurance in patient outcomes remains unclear. The purpose of this study was to perform a systematic review of the literature to explore the relationship between preoperative Medicaid insurance status and outcomes following primary total knee arthroplasty.

METHODS

A systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to identify studies examining outcomes in patients who had Medicaid and were undergoing total knee arthroplasty. Studies including complex revision operations or less common indications for total knee arthroplasty were excluded. Data on insurance status, postoperative complications, length of stay, readmissions, and subsequent revision surgical procedures were collected for each article.

RESULTS

A total of 13 studies showing 6.18 million patients undergoing total knee arthroplasty were included in the qualitative synthesis. Seven analyses described an important association between Medicaid coverage and short-term readmissions, and 2 analyses showed a relationship between Medicaid and prolonged length of stay. However, the included studies did not describe a significant association between Medicaid and postoperative mortality or revision rates.

CONCLUSIONS

Patients with Medicaid undergoing total knee arthroplasty may be more likely to experience an increased length of stay and to be readmitted postoperatively. The unique factors associated with these patients may help to inform customized perioperative surveillance and optimization to improve outcomes in this group.

LEVEL OF EVIDENCE

Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.

Discretionary Surgery: A Comparison of Workers' Compensation and Commercial Insurance

BACKGROUND

Workers' compensation is intended for injuries that occur at work and is expected to be mostly for trauma and mostly nondiscretionary conditions. We tested the null hypothesis that there is no difference in the ratio of likely discretionary to likely nondiscretionary surgery between patients treated under workers' compensation compared with commercial insurance controlling for age, sex, and anatomical site for either traumatic or nontraumatic diagnoses.

METHODS

Using claims data from the Texas workers' compensation database and Truven Health commercial claims we classified International Statistical Classification of Diseases and Related Health Problems, Ninth Revision, Clinical Modification (ICD-9-CM) diagnoses and procedure codes as likely discretionary or likely nondiscretionary, and as traumatic or nontraumatic. Ratios of likely discretionary to likely nondiscretionary surgery were calculated and compared.

RESULTS

Among patients treated under workers' compensation, the ratio of likely discretionary to likely nondiscretionary surgery was significantly higher for traumatic diagnoses (0.57 [95% confidence interval, CI, = 0.56-0.61] vs 0.38 [95% CI = 0.37-0.40], P < .05) and significantly lower for nontraumatic diagnoses (9.4 [95% CI = 9.20-9.42] vs 13.2 [95% CI = 12.9-13.3], P < .05) compared with commercial insurance.

CONCLUSIONS

Workers' compensation often covers likely discretionary musculoskeletal surgery, and insurance type may influence treatment.

Is There a Woodwork Effect? Addressing a 200-Year Debate on the Impacts of Expanding Community-Based Services

In U.S. social welfare history, many have suggested that if benefits were too attractive, consumers would come out of the woodwork to take advantage of the opportunity. Clinical trials have provided evidence of the woodwork effect's existence, suggesting caution when expanding home- and community-based services (HCBS). However, it is unclear whether these studies are best suited to assess whether a system-level effect occurs. Using state and federal data tracking Ohio's long-term services and support (LTSS) system from 1995 to 2015, this paper examines changes in the utilization rates and expenditures of Medicaid LTSS to explore whether a woodwork effect occurred as Ohio moved to improve its LTSS system balance (80% Nursing Home [NH], 20% HCBS) to (49% Nursing Home [NH], 51% HCBS). After accounting for population growth of individuals older than 60 and those with two or more impairments in activities of daily living, there was no change in utilization rates of older people with severe disability (1995: 491 per 1000 population, 2015: 495 per 1000 population) or overall LTSS expenditures (1997: $2.7 million [in 2013 dollars], 2013: $2.9 million). Our results suggest that states can make significant strides in HCBS expansion without increasing the overall long-term services utilization rate.

A Preoperative Scale for Determining Surgical Readmission Risk After Total Knee Arthroplasty

BACKGROUND

Total knee arthroplasty (TKA) is one of the most common orthopaedic procedures performed in the U.S. The purpose of this study was to develop and verify a scale to preoperatively stratify a patient's risk of being readmitted to the hospital following a TKA.

METHODS

Discharge data on 433,638 patients from New York and California (derivation cohort) and 269,934 patients from Florida and Washington (validation cohort) who underwent TKA were collected from the State Inpatient Database, a part of the Healthcare Cost and Utilization Project of the Agency for Healthcare Research and Quality (2006 to 2011). Demographic and clinical characteristics of patients were abstracted and analyzed to develop the Readmission After Total Knee Arthroplasty (RATKA) Scale.

RESULTS

Overall 30-day readmission rates in the derivation and validation cohorts were 5.11% and 4.98%, respectively. The following factors were significantly associated with increased 30-day readmission rates in the derivation cohort: age of 41 to 50 years (odds ratio [OR] = 1.13), age of 71 to 80 years (OR = 1.21), age of 81 to 90 years (OR = 1.70), male sex (OR = 1.19), African-American race (OR = 1.37), "other" race/ethnicity (OR = 1.08), Medicaid payer (OR = 1.43), Medicare payer (OR = 1.27), anemia (OR = 1.19), chronic obstructive pulmonary disease (OR = 1.29), coagulopathy (OR = 1.22), congestive heart failure (OR = 1.64), diabetes (OR = 1.19), fluid and electrolyte disorder (OR = 1.25), hypertension (OR = 1.10), liver disease (OR = 1.27), renal failure (OR = 1.33), and rheumatoid arthritis (OR = 1.14). These factors were used to create the RATKA Scale. The RATKA score was then used to define 3 levels of risk for readmission: low (RATKA score of <13; 3.7% readmission rate), moderate (RATKA score of 13 to 16; 5.4% readmission rate), and high (RATKA score of >16; 7.6% readmission rate). The relative risk of readmission was 2.06 for the high-risk group compared with the low-risk group.

CONCLUSIONS

The RATKA Scale derived from patient data from the derivation cohort was reliably able to explain readmission variability after TKA for patients in the validation cohort at a rate of >95%. Models such as the RATKA Scale will enable identification of the risk of readmission following TKA based on a patient's risk profile prior to surgery.

LEVEL OF EVIDENCE

Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

Predictors of emergency ventral hernia repair: Targets to improve patient access and guide patient selection for elective repair

BACKGROUND

Emergency operations are associated with worse outcomes than elective operations. If not repaired electively, ventral hernias are at risk of strangulating and requiring emergency repair. We sought to identify patient- and hospital-level factors associated with emergency ventral hernia repair in a nationally representative, United States sample.

METHODS

We abstracted data from the 2003-2011 Nationwide Inpatient Sample for adults (≥18 years) who underwent inpatient ventral hernia repair. Our primary outcome was emergency repair. We assessed differences in patient- and hospital-level factors as possible predictors of emergency repair using multivariable logistic regression. We examined secondary outcomes (mortality, total hospital cost, duration of stay) using multivariable logistic and generalized linear (family gamma; link log) regression.

RESULTS

After weighting to the United States population, we included 453,161 adults (39.5% emergency). Independent predictors of emergency repair included payer status (uninsured: odds ratio 3.50, [3.10, 3.96]; Medicaid: 1.29 [1.20, 1.39] compared with private insurance), race/ethnicity (black: 1.77 [1.64, 1.92]; Hispanic: 1.44 [1.28, 1.61] compared with white), age (≥85 years: 2.23 [2.00, 2.47] compared with <45 years), and comorbidities (Charlson Comorbidity Index ≥3: 1.68 [1.56, 1.80] compared with 0). After risk-adjustment, emergency repair was associated with greater odds of in-hospital death, greater costs, and longer hospital stay.

CONCLUSION

Inpatient ventral hernia repairs are frequently performed emergently, with worse outcomes in this group. Independent predictors of emergency repair include factors that may limit access to and/or selection for an elective operation. These predictors provide targets for interventions to improve access to elective care and inform patient selection with the goal of improving patient outcomes.

Patient characteristics associated with treatment choice for pelvic organ prolapse and urinary incontinence

INTRODUCTION AND HYPOTHESIS

Women with pelvic organ prolapse (POP) and stress urinary incontinence (SUI) frequently undergo more than one treatment prior to settling on their final strategy. We hypothesize that women who are younger, with worse POP and SUI symptoms will desire and choose surgical treatment.

METHODS

A retrospective cohort study was performed over 1 year identifying new patients presenting with POP and/or SUI at a university hospital. Our aim was to determine patient desire for either surgical or conservative treatment, as well as the actual treatment chosen and received after the first visit and 1 year later. To identify predictors of choice, baseline demographic characteristics were obtained.

RESULTS

Of the 203 women who met the inclusion criteria, 44.3 % (90/203) desired surgery and 55.7 % (113/203) desired conservative treatment at their first visit. Women who desired surgery were more likely to be younger (p = 0.003), sexually active (p = 0.001), have more advanced prolapse (p = 0.006), and have more bothersome symptoms (p = 0.05). Of the women who desired surgery at their first visit, 12.2 % (11/90) actually chose conservative treatment. These women were less likely to be insured (p = 0.01). By 1 year, of the women who initially desired and subsequently chose conservative treatment, 26.5 % (30/113) had undergone surgery. The women who changed from conservative to surgical treatment were more likely to be younger (p = 0.01), non-White (p = 0.03), and sexually active (p = 0.04).

CONCLUSIONS

In this study, younger, sexually active women were more likely to either opt for surgery initially or to change their treatment plan from conservative to surgical.