Association Between High-Deductible Health Plans and Hernia Acuity

Key Points

Question  Is enrollment in a high-deductible health plan (HDHP) associated with higher odds of presenting with an incarcerated or strangulated abdominal wall or groin hernia?

Findings  In this cohort study that included 83 281 patients from a large commercial insurance claims database, the study team found that those enrolled in individual HDHP coverage were at 6% higher odds of presenting with an incarcerated or strangulated hernia.

Meaning  Among patients with groin and ventral hernias, enrollment in an HDHP may be associated with delays in surgical care that may result in complex disease presentation.

Importance  About half of people younger than 65 years with private insurance are enrolled in a high-deductible health plan (HDHP). While these plans entail substantially higher out-of-pocket costs for patients with chronic medical conditions who require ongoing care, their effect on patients undergoing surgery who require acute care is poorly understood. It is plausible that higher out-of-pocket costs may lead to delays in care and more complex surgical conditions.

Objective  To determine the association between enrollment in HDHPs and presentation with incarcerated or strangulated hernia.

Design, Setting, and Participants  This retrospective cohort analysis included privately insured patients aged 18 to 63 years from a large commercial insurance claims database who underwent a ventral or groin hernia operation from January 2016 through June 2019 and classified their coverage as either a traditional health plan or an HDHP per the Internal Revenue Service’s definition. Multivariable regression, adjusting for demographic and clinical covariates, was used to examine the association between enrollment in an HDHP and the primary outcome of presentation with an incarcerated or strangulated hernia.

Exposures  Traditional health plan vs HDHP.

Main Outcomes and Measures  Presence of an incarcerated or strangulated hernia per International Statistical Classification of Diseases and Related Health Problems, Tenth Revision diagnosis codes.

Results  Among 83 281 patients (71.9% men and 28.1% women; mean [SD] age, 48.7 [10.9] years) who underwent hernia surgery, 27 477 (33.0%) were enrolled in an HDHP and 21 876 (26.2%) had a hernia that was coded as incarcerated or strangulated. The mean annual deductible was considerably higher for those in the HDHP group than their traditional health plan counterparts (unadjusted mean [SD], $3635 [$2094] vs $705 [$737]; adjusted, −$2931; P < .001). Patients in the HDHP group were more likely to present with an incarcerated or strangulated hernia (adjusted odds ratio, 1.07; 95% CI, 1.03-1.11; P < .001).

Conclusions and Relevance  In this cohort study, enrollment in an HDHP was associated with higher odds of presenting with an incarcerated or strangulated hernia, which is more likely to require emergency surgery that precludes medical optimization. These data suggest that, among patients with groin and ventral hernias, enrollment in an HDHP may be associated with delays in surgical care that result in complex disease presentation.

More than 50% of private sector employees with both family and single coverage were enrolled in a high-deductible health plan (HDHP) in 2019.¹ Such plans present substantially higher out-of-pocket costs for patients with chronic medical conditions.² Furthermore, enrollment in HDHPs has been associated with delayed care for multiple medical conditions, including macrovascular complications of diabetes and breast cancer management.^3,4 Among families whose members have chronic conditions, enrollment in an HDHP has been associated with cost-related delay or forgone care for both adults and children compared with enrollment in a traditional plan.^5,6 While the effect of these plans on chronic medical conditions has been studied extensively, to our knowledge, their effect on acute surgical conditions has not been studied.

In contrast with chronic medical conditions, surgical conditions are treated with short-term episodic care. When treatment of a surgical condition is delayed, it may become more complex and ultimately require an emergency operation that does not allow for preoperative medical optimization or procedural planning. For a groin or ventral hernia, an exceedingly common surgical condition, an elective operation has the potential to reduce serious hernia-related morbidity and mortality. Moreover, while a watchful waiting approach is appropriate at times for asymptomatic hernias, symptomatic hernias and all groin hernias in women should be repaired given the risk of major morbidity.^7,8

To explore the issue of potential delays in surgical care and enrollment in HDHPs, we examined the association between HDHPs and presentation with an incarcerated (ie, obstructed) or strangulated (ie, impending or frank gangrene) hernia using a large private insurance claims database. We hypothesized that patients who were enrolled in HDHPs were more likely to present with incarcerated or strangulated hernias, indicating potential delay in an elective, simple hernia repair.

We performed a retrospective cohort analysis of commercially insured patients who underwent a groin or ventral hernia repair surgery from January 2016 through June 2019. We drew our study population from Optum Clinformatics Data Mart, a deidentified commercial insurance claims database. Optum contains outpatient, inpatient, and pharmaceutical claims data from 1993 to 2021 for more than 111 million enrollees. We used Current Procedural Terminology codes to identify patients who underwent a ventral or groin hernia operation from January 2016 through June 2019 (Figure). We selected repair of ventral abdominal wall and inguinal hernias because they have potential to become an emergency (ie, incarceration and strangulation) when care is delayed. This study was deemed exempt from review by the Stanford University Institutional Review Board owing to the use of deidentified data.

The primary outcome was presence of a complicated hernia, determined by the hernia diagnosis code (ie, incarcerated or strangulated) International Statistical Classification of Diseases and Related Health Problems, Tenth Revision. Enrollment in an HDHP, the primary independent variable, was calculated for patients in both family plans and individual plans. We identified all family members of the patient using their unique family identification number, then summed each family member’s deductible spending to find each patient’s total deductible for the year of the index operation. The sample was restricted to enrollees who underwent hernia surgery in a given year, defined as January 1 through December 31, and this approach was based on the assumption that the expense associated with the surgery would exceed the deductible for most patients. As a result, deductible spending is a proxy for the deductible facing the patient. We then stratified patients into a traditional group and a high-deductible group based on the year of their operation, enrollment in a family plan, and the Internal Revenue Service (IRS) deductible cutoff. The IRS defines an HDHP as a plan with an annual deductible of $1300 or higher for individuals or $2600 or higher for families in 2016 and 2017, and an annual deductible of $1350 or higher for individuals or $2700 or higher for families in 2018 and 2019.^9-12 We excluded the top and bottom 0.5 percentile of costs to remove outliers (879 observations). Patients who were dichotomized to the traditional group and had costs that did not exceed their deductible amount were also excluded, because there was potential for deductible group misclassification (133 observations). Additionally, we excluded 86 patients whose deductibles were a negative amount, as this was likely misclassification or clerical error.

As a secondary analysis, the study team examined the association between presence of a complicated hernia and deductible amounts among patients enrolled in individual coverage. For patients with individual coverage, the study team also distinguished between 3 deductible levels among the high-deductible group. The deductible amount was determined by assigning patients into tiers based on their annual deductible and IRS HDHP coverage for the year of their index operation. Tier 1 included patients whose index operation occurred from 2016 to 2017 and annual deductible was $1300 to $2600, and those whose index operation occurred from 2018 to 2019 and annual deductible was $1350 to $2700. Tier 2 included patients whose index operation occurred from 2016 to 2017 and annual deductible was $2600.01 to $3900, and patients whose index operation occurred from 2018 to 2019 and deductible was $2700.01 to $4050. Tier 3 included patients whose index operation occurred from 2016 to 2017 and deductible was $3900.01 or higher, and those whose index operation occurred from 2018 to 2019 and deductible was $4050.01 or higher. The study team then tested for equivalence between tiers with Wald tests.

In an additional secondary analysis, to determine whether the presence of a health savings account (HSA) mitigated the association between HDHP and complex hernia presentation, the study team repeated the main analysis without individuals who were enrolled in an HDHP but did not have an HSA. The group without a HDHP plan remained the same.

In all analyses, odds ratios (ORs) were adjusted for age, sex, race, state, Charlson Comorbidity Index score, and hernia type using multivariate logistic regression. The study team also controlled for family vs individual coverage for the full sample analysis. Missing/unknown data for control variables were considered a separate category for each variable, and no patient was missing data for the dependent variable. The study team defined level of significance as α = .05, and all P values were 2-tailed. All data were analyzed using Stata, version 16.1 (StataCorp).

A total of 83 281 patients in the cohort underwent a groin or ventral hernia operation during the study period (71.9% men and 28.1% women; mean [SD] age, 48.7 [10.9] years). Among these, 55 804 (67.0%) were enrolled in a traditional (low-deductible) health plan and 27 477 (33.0%) were enrolled in an HDHP. The mean annual deductible was considerably higher for those in the HDHP group than their traditional health plan counterparts (unadjusted mean [SD], $3635 [$2094] vs $705 [$737]; adjusted, −$2931; P < .001). A total of 98.0% of patients had costs that exceeded deductible amounts. Patients who were enrolled in an HDHP were considerably younger (47.6 [11.0] vs 49.3 [10.7] years; P < .001), were more likely to be enrolled in individual coverage (16 657 of 27 477 [60.6%] vs 21 675 of 55 804 [38.8%]; P < .001), and had fewer comorbid conditions (Charlson Comorbidity Index score 3 or higher, 10 810 of 27 292 [39.6%] vs 27 946 of 55 753 [50.1%]; P < .001) compared with patients enrolled in a traditional plan (Table 1).

Overall, 26.7% (n = 7348) of patients in the HDHP group presented with a hernia coded as an incarcerated or strangulated (ie, complicated) hernia compared with 26.0% (n = 14 519) in the traditional plan group (adjusted OR [AOR], 1.06; 95% CI, 1.03-1.11; P < .001) (Table 2). Among those enrolled in individual coverage, the odds of presenting with a complicated hernia were statistically significantly higher among those enrolled in an HDHP than those enrolled in a traditional plan (AOR, 1.09; 95% CI, 1.04-1.15; P < .001). Among patients enrolled in family coverage, there was no statistically significant difference between HDHP and traditional plan enrollees in frequency of presentation with an incarcerated or strangulated hernia (AOR, 1.05; 95% CI, 0.99-1.10; P = .10).

Among patients enrolled in individual HDHPs, those with the highest deductible amounts (ie, tier 3) were at higher odds of presenting with an incarcerated or strangulated hernia (AOR, 1.10; 95% CI, 1.03-1.19; P = .02) relative to patients with traditional plans. Patients in tier 1 had the next highest odds of presenting with an incarcerated or strangulated hernia (AOR, 1.07; 95% CI, 0.98-1.15; P = .16), followed by patients in tier 2 (AOR, 1.06; 95% CI, 1.01-1.14; P = .02) (Table 3). Wald test for equivalence revealed that tier 1 was equivalent to tier 2 (χ² = 0.69), tier 1 was equivalent to tier 3 (χ² = 0.47), tier 2 was equivalent to tier 3 (χ² = 0.78), and no tiers were equivalent to patients in the traditional plan group (χ² = 0.02, 0.02, and 0.01 for tiers 1, 2, and 3, respectively).

Enrollment in an HDHP without having an HSA was associated with higher odds of presenting with an incarcerated or strangulated hernia (AOR, 1.10; 95% CI, 1.05-1.15) (Table 4). This was true for patients in both individual plans (AOR, 1.12; 95% CI, 1.00-1.18) and family plans (AOR, 1.08; 95% CI, 1.00-1.15). By contrast, patients enrolled in an HDHP who did have an HSA were no more likely than patients in a traditional plan to present with an incarcerated or strangulated hernia (AOR, 1.00; 95% CI, 0.95-1.06).

In this study of 2 common conditions, abdominal wall and groin hernias, the study team found that enrollment in an individual HDHP was associated with higher odds of presenting with an incarcerated or strangulated hernia, regardless of magnitude of high-deductible amount. Enrollment in HDHP family coverage was not associated with a statistically significant increase in odds of incarcerated or strangulated hernia presentation. While expectant management is appropriate for most asymptomatic hernias, repair is indicated for all groin hernias in women and for symptomatic hernias in surgically fit patients to reduce the risk of incarceration or strangulation.^13,14 The results suggest that cost-related delay in care may have deleterious effects on hernia presentation.

To our knowledge, this study is one of the first to examine the association between enrollment in an HDHP and presentation with an urgent surgical condition. In sensitivity analyses, the study team found a positive association between the deductible amount and odds of strangulated or incarcerated hernia presentation, though the ORs of each tier were not statistically different from each other. However, patients in each high-deductible tier had greater odds of complications than those who were enrolled in traditional plans, indicating that primary results were not skewed by patients enrolled in catastrophic deductible plans. The modest but statistically significant effect size is supported by what is to our knowledge the only other current publication to examine HDHP and urgent surgical conditions (diverticulitis and appendicitis).¹⁵

In 2019, more than half of employees at both small and large firms were enrolled in an HDHP.¹ Proponents of these plans argue that they decrease waste and alleviate some of the financial burden of higher premiums, particularly when coupled with a tax-free savings account. These findings partially support this because patients enrolled in individual-coverage HDHPs presented with more advanced disease requiring emergency intervention, but those with HSAs did not. That is, HDHPs may be beneficial for individuals who can afford to self-insure with the HSA mechanism but not for those unable to set aside money for unexpected health expenses. Until the effects of these plans on acute surgical conditions are better understood, it is important that employers and enrollees are well informed regarding the implications of choosing a lower premium in exchange for higher out-of-pocket costs. Future research regarding HDHP and acute surgical conditions should focus on understanding why we are observing these patterns and identifying whether they extend to other general surgery conditions.

While this study does not provide direct causal evidence that the results are solely owing to delay in care, this interpretation is consistent with other evidence from the literature. Many studies have found evidence for delay and forgone care secondary to out-of-pocket costs for other common conditions. Wharam et al³ found that patients with diabetes waited 1.5 months longer to seek care for their first major symptom of macrovascular diabetic complications and were delayed more than 3 months for their first procedure-based treatment when compared with a matched cohort in a traditional plan. Forgone care has been demonstrated with bariatric surgery use. Chhabra et al¹⁶ found that patients who were enrolled in lower-cost sharing plans (preferred provider organizations) underwent bariatric surgery at a higher rate (20 operations/100 000 enrollees) than those enrolled in HDHPs (12.1 operations/100 000 enrollees). While results of the current study are consistent with those of Chhabra et al, they provide stronger evidence of poor outcomes associated with delay for an acute surgical condition. The study team notes, however, that hernia operations have the potential to become an emergency in the setting of delay,^7,8 whereas bariatric surgery will always be performed in the elective setting.

In addition, research indicating that delay in hernia repair is associated with incarceration and strangulation supports this interpretation. Hair and colleagues¹⁷ found that cumulative probability of inguinal hernia irreducibility increased from 6.5% (95% CI, 4%-9%) at 12 months to 30% (95% CI, 18%-42%) at 10 years, and a European study¹⁴ found that the cumulative probability of groin hernia strangulation was 2.8% (95% CI, 1.1%-4.5%) at 3 months and 4.5% (95% CI, 1.3%-7.7%) at 2 years.

The study team also found that patients enrolled in HDHP family coverage were not at a statistically significant higher odds of presenting with a complicated or strangulated hernia. Between 52% and 58% of families in the US have multiple incomes,¹⁸ and family HDHPs often have higher caps on tax-free savings accounts, both of which can offset high out-of-pocket costs. It is possible that most patients aged 18 to 26 years in this analysis were still enrolled on their parents’ coverage and may have had financial support that offsets out-of-pocket costs.

This study was subject to the limitations inherent to administrative claims data, including potential for misclassification and clerical error. Because the study team was unable to determine the time duration between date of diagnosis and date of repair, we used presentation with an incarcerated or strangulated hernia as a proxy for delay in seeking care. While extensive evidence in the literature supports a causal relationship between delayed care and incarcerated and strangulated hernias, this is not always the case. Furthermore, the administrative claims database lacks information that would improve patient stratification, such as length of symptoms, imaging findings, and detailed demographic information. Additionally, the study team was only able to observe whether an individual had an HSA, but not if the account was funded or used.

As with all retrospective studies, the data may have additionally been subject to selection bias. Some patients with multiple comorbidities may have not undergone repair because they were poor operative candidates. These patients may have been more likely to present with an incarcerated or strangulated hernia not because there was delay in seeking care secondary to medical costs but because of poor surgical fitness in the elective setting. While we controlled for comorbid disease and age in the multivariate analysis, the potential for residual confounding remains. Additionally, for patients in family plans, the study team used the aggregate annual deductible amount for the household. Individual deductible amounts could only be derived for patients with individual coverage.

This cohort study found that enrollment in an HDHP among patients undergoing surgery with individual coverage was associated with higher odds of presenting with an incarcerated or strangulated hernia. While the clinical significance of the increase in the likelihood of a complication at an individual patient level may be small, hernia operations are very common procedures. When considering the vast numbers of these operations annually, the costs quickly become substantial. While the findings do not provide direct evidence that these results are solely owing to delay in seeking care, HDHPs have been associated with delay in care and decreased health care use in other clinical contexts.^3,4,15 To better understand the effect of cost sharing on patients undergoing surgery, it would be prudent to examine whether these patterns are consistent across other general surgery conditions in which cost-related delay in care has an effect on patient outcomes.

Accepted for Publication: November 6, 2021.

Published Online: February 13, 2022. doi:10.1001/jamasurg.2021.7567

Corresponding Author: Arden M. Morris, MD, MPH, S-SPIRE Center, Surgery Department MC5552, Stanford University School of Medicine, 3145 Porter Dr, Wing C0, Stanford, CA 94304 (ammsurg@stanford.edu).

Author Contributions: Drs Yelorda and Morris had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Yelorda, Rose, Bundorf, Morris.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Yelorda, Rose, Muhammad, Morris.

Critical revision of the manuscript for important intellectual content: Yelorda, Rose, Bundorf, Morris.

Statistical analysis: Yelorda, Rose, Bundorf.

Administrative, technical, or material support: Muhammad.

Supervision: Rose, Bundorf, Morris.

Conflict of Interest Disclosures: Financial and material support was provided to Dr Yelorda by the US Department of Health and Human Services, Agency for Healthcare Research and Quality (5T32HS26128-2). Dr Rose reports grants from the US Department of Veterans Affairs during the conduct of the study. No other disclosures were reported.

Meeting Presentation: This paper was presented at the Pacific Coast Surgical Association Annual Meeting; February 13, 2022; Maui, Hawaii.