ACCESS TO HEALTHCARE INSURANCE INCREASES THE RATES OF SURGERY FOR DIVERTICULITIS

Abstract

OBJECTIVE:

The goal of this study was to examine the effect of the Affordable Care Act Medicaid expansion on rates of hospitalization and surgery for diverticulitis.

STUDY SETTINGS:

Data were obtained from the Healthcare Cost and Utilization Project State Inpatient Databases from 2010 to 2014.

STUDY DESIGN:

Retrospective cohort study analyzing adult patients undergoing surgery for diverticulitis in the expansion and nonexpansion states, pre (2010–2013) and post (2014) Medicaid expansion.

FINDINGS:

There were a total of 159,419 patients in our cohort analysis. 75,575 (49%) in expansion states and 81,844 (51%) in non-expansion states. In multivariable Poisson regression, the rate of surgical procedures for diverticular disease increased among Medicaid patients (IRR 1.80; p<.01) whereas surgery rates in self-pay patients decreased (IRR 0.67; p<.01) in expansion states compared to non-expansion states.

CONCLUSIONS:

In states that expanded Medicaid coverage under the Affordable Care Act, the rate of surgery for diverticular disease in Medicaid patients increased. Therefore, legislation that increases healthcare access may increase the utilization of surgical care for diverticular disease.

INTRODUCTION

The Affordable Care Act (ACA) helped increase access to healthcare beginning in 2014 by the following mechanisms: it created subsidized insurance exchanges so that private insurance would be more affordable, it eliminated discrimination by insurers based on pre-existing conditions, and it expanded the Medicaid program.¹ The goal of Medicaid expansion was to extend eligibility to adult citizens who were less than 65 years old with an annual income below 138 percent of the federal poverty level.¹ It was initially intended to be available in all states; however, the final ruling by the US Supreme Court resulted in Medicaid expansion being voluntary for individual states.^1,2

Multiple studies have evaluated what happens when the individual cost of healthcare is decreased, and healthcare is thereby made more accessible. The RAND Healthcare Insurance Experiment and the Oregon Medicaid study are examples often cited.^3,4 Although somewhat oversimplified, the body of literature including both of these studies essentially showed that when the barrier of cost is reduced or removed for the individual, utilization of healthcare resources increases.^3–7 It remains controversial whether the increased use of healthcare resources leads to improved health; however, more recent studies have found insurance expansion is associated with improved access to primary care and reduction in rates of uninsured hospitalized patients.^1,8–10

With the understanding that the ACA Medicaid expansion fundamentally improved access by lowering the cost barrier to healthcare insurance for low-income adults, we sought to determine if there were differences observed after implementation in the care for one of the most common conditions colorectal surgeons treat: diverticulitis. Thus, the goal of this study was to examine the effect that the ACA Medicaid expansion had on rates of diverticulitis associated hospitalization or surgery in the various insurance categories (i.e., private, Medicaid, or self-pay).

MATERIALS & METHODS

Data Source

Patients were identified using the Healthcare Cost and Utilization Project (HCUP), State Inpatient Databases (SID) for the following states: Washington (WA), Iowa (IA), New York (NY), Maryland (MD), Florida (FL), and North Carolina (NC). Among these IA, WA, NY, and MD had opted to participate in the Medicaid expansion and implemented it in 2014 while FL and NC had opted to not participate in the expansion. Twenty four states initially declined to expand Medicaid eligibility in 2014, and 14 states still have not expanded eligibility as of 2019.² The expansion and non-expansion states were identified from the previously published literature. (2) HCUP is an administrative dataset composed of a family of healthcare databases developed through a Federal-State-Industry partnership sponsored by the Agency for Healthcare Research and Quality (AHRQ). Each SID captures all inpatient discharges from stays at non-federal facilities for the respective state, regardless of primary payer.

Study Population

The analysis cohort included adult patients between the ages of 18 and 64 years who were discharged from hospitals in expansion and non-expansion states in the SIDs between 2010 and 2014. The cohort included patients with the following inclusion criteria: (1) adult patients between the ages of 18 and 64 years; (2) patients discharged from hospitals in expansion and non-expansion states included in the SID between 2010 and 2014; (3) and underwent surgery for diverticulitis. International Classification of Diseases, Ninth Revision (ICD-9) diagnosis and procedure codes were used to identify patients undergoing surgical resection for Diverticulitis. (Supplemental Table. 1) Patients younger than 18 years of age, older than 64 years and on Medicare were excluded because they were not eligible for the ACA’s Medicaid expansion. Patient informed consent was not required for this study.

Outcomes

The primary outcome of interest was the rate of diverticulitis related surgeries and admissions pre- and post-ACA Medicaid expansion. A difference-in-difference (DID) analysis was used to estimate the effect of the ACA expansion on diverticulitis surgery rates and admissions for patients on Medicaid or self-pay compared to those with private insurance. For the analysis, we first calculated surgery and admission rates in the expansion states (IA, NY, WA, and MD) and non-expansion states (FL and NC), before (2010–2013) and after (2014) the expansion, and then calculated the difference between the change in rates in expansion states and change in non-expansion states.

Statistical Analysis

This analysis was a retrospective observational study of patients undergoing care from 2010 to 2014. Baseline characteristics are presented as means and standard deviations, medians, and interquartile ranges (IQR), or counts and percentages. Unadjusted comparisons of two or more proportions between pre- and post-expansion were performed using a chi-squared test, and continuous variables were compared using t-tests or Wilcoxon rank sum tests as appropriate.

We used a quasi-experimental, difference in difference (DID) design to compare utilization of surgical procedures in Medicaid expansion vs. non-expansion states pre- (2010–2013) and post- (2014) Medicaid expansion. The DID statistical approach is often used to evaluate the changes that occur following policy changes^11,12 and has been described in previous publications.^10,13

To test the significance of the impact of the ACA expansion, we used a multivariable Poisson regression to evaluate the risk-adjusted association of Medicaid insurance and outcomes of interest on expansion versus non-expansion states. For our model we created indicator variables for (1) time (before expansion (2010–2013) versus after expansion (2014)), (2) expansion status (expansion states (IA, NY, WA, and MD) versus non-expansion (NC and FL) states), and (3) insurance status (Medicaid or self-pay versus private insurance). The triple interaction term was the DID estimator, and its coefficient reflected the association in ACA Medicaid expansion in states that expanded Medicaid insurance compared to those states that opted from not expanding and the outcomes of interest. For presentation, the incidence rate ratio (IRR) was calculated and can be interpreted as the relative percentage point change in our outcomes of interest associated with the ACA Medicaid expansion. Models were adjusted for patient demographics and comorbidities. We weighted our models using state-level Census data to account for population growth and migration during the study period.¹⁴

To evaluate that there was no significant change between groups relative to one another before the ACA Medicaid expansion we conducted a sensitivity analysis to verify parallel trends in our outcomes of interest before reform. Our analysis used the models previously described and tests the significance of the triple interaction term to verify that there was no change before the ACA Medicaid expansion. All analyses were performed using STATA 14 software (College Station, TX). This study was evaluated and approved by the Institutional Review Board at Loyola University Chicago.

RESULTS

Demographics of Diverticulitis Hospital Admissions

A total of 159,419 patients with a diagnosis of diverticulitis met our inclusion and exclusion criteria: 77, 575 (49%) patients from states that expanded Medicaid and 81,844 (51%) from states that did not expand Medicaid. Comparing the four years prior (2010–2013) to the year after the ACA Medicaid expansion was implemented (2014), univariate analysis of the patients hospitalized for diverticulitis revealed several trends. In 2014, the states who expanded Medicaid demonstrated an increase in the proportion of patients being hospitalized for diverticulitis covered by Medicaid insurance (26% vs. 19%, p = 0.001) and a decrease in those with private insurance (69% vs. 73%, p= 0.001) or who were self-pay (5% vs. 8%, p = 0.001). In the states that did not expand Medicaid, no difference was seen in the proportion of patients being hospitalized for diverticulitis with Medicaid (14% vs. 14%, p = 0.83) after the ACA was enacted; however, there was an increase in the proportion of those with private insurance (68% vs. 66%, p = 0.001) and a decrease in the number of self-pay (19% vs. 21%, p = 0.001). In both groups of states, there was no significant difference in age, gender, comorbidities, and race/ethnicity, before or after the expansion in patients hospitalized with a diagnosis of diverticulitis. Our univariate analysis reveals that those states that expanded Medicaid insurance had an increase in the proportion of patients receiving healthcare which may be associated with improved access. (Table 1)

Table 1.

Demographics for diverticulitis admissions in expansion and non-expansion states Pre and Post ACA

	Expansion States					Nonexpansion States
	Pre (2010–2013)		Post (2014)		p	Pre (2010–2013)		Post (2010–2013)		p
No. Patients, n (%)	62,621	81%	14,954	19%		64,395	79%	17,449	21%
Age, mean (sd)	50	9.5	50	9.4	0.05	50	9.5	50	9.5	0.07
Female, n (%)	31,014	49.5%	7,253	48.5%	0.02	32,369	50.3%	8,695	49.8%	0.31
Elixhauser, mean (sd)	2	1.5	2	1.6	<.01	2	1.5	2	1.6	0.001
Insurance Type, n (%)
Medicaid	11,746	18.8%	3,848	25.7%	<.01	8,862	13.8%	2,390	13.7%	0.83
Private	45,729	73.0%	10,368	69.3%	<.01	42,236	65.6%	11,831	67.8%	0.001
Self-pay/No Charge	5,146	8.2%	738	4.9%	<.01	13,297	20.6%	3,228	18.5%	0.01
Race, n (%)
White	42,922	68.5%	10,303	68.9%	0.03	44,341	68.9%	12,013	68.8%	0.60
Black	7,894	12.6%	1,806	12.1%		6,983	10.8%	1,932	11.1%
Hispanic	7,290	11.6%	1,686	11.3%		11,628	18.1%	3,097	17.7%
Other	4,515	7.2%	1,159	7.8%		1,443	2.2%	407	2.3%

Pre-ACA period is from 2010 to 2013; Post-ACA is 2014. Expansion states include Maryland, Massachusetts, New-York, and Iowa. Non-expansion states include Florida and North Carolina. Excluded patients <19 & >64 years of age and Medicare patients.

Demographics of Patients Undergoing Surgery for Diverticulitis

The demographics and characteristics of patients undergoing surgery for diverticulitis in expansion and non-expansion states are summarized in Table 2. There was a total of 35,159 patients who had diverticulitis and underwent surgery: 17,488 (50%) in the expansion states and 17,671 (50%) in the non-expansion states. Comparing the four years prior (2010–2013) to the year after the ACA was enacted (2014), univariate analysis of patients having surgery for diverticulitis also revealed several trends. In 2014, the states who expanded Medicaid demonstrated a higher proportion of patients having surgery for diverticulitis who had Medicaid insurance (20% vs. 13%, p = 0.001) and a decrease in the proportions with private insurance (78% vs. 82%, p = 0.001) and self-pay (3% vs. 5%, p = 0.001). These results would suggest that increasing access to healthcare insurance may lead to patients receiving surgical care that otherwise they may have not been able to afford. (Table 2)

Table 2.

Demographics of patients undergoing surgery for diverticulitis in expansion and non-expansion states Pre and Post ACA

	Expansion States					Nonexpansion States
	Pre (2010–2013)		Post (2014)		p	Pre (2010–2013)		Post (2014)		p
No. Patients, n (%)	13,998	80%	3,490	20%		13,949	79%	3,722	21%
Age, mean (sd)	51	8.9	51	8.8	0.18	51	9.0	51	9.0	0.36
Female, n (%)	6,419	45.9%	1,568	44.9%	0.25	6,367	45.6%	1,706	45.8%	0.84
Elixhauser, mean (sd)	2	1.5	2	1.6	<.01	2	1.6	2	1.6	0.001
Insurance Type, n (%)
Medicaid	1,762	12.6%	691	19.8%	<.01	1,311	9.4%	325	8.7%	0.21
Private	11,532	82.4%	2,707	77.6%	<.01	10,763	77.2%	2,981	80.1%	0.001
Self-pay/No Charge	704	5.0%	92	2.6%	<.01	1,875	13.4%	416	11.2%	0.001
Race, n (%)
White	11,088	79.2%	2,759	79.1%	0.76	10,866	77.9%	2,866	77.0%	0.20
Black	1,026	7.3%	263	7.5%		1,028	7.4%	260	7.0%
Hispanic	1,035	7.4%	269	7.7%		1,805	12.9%	530	14.2%
Other	849	6.1%	199	5.7%		250	1.8%	66	1.8%

Pre-ACA period is from 2010 to 2013; Post-ACA is 2014. Expansion states include Maryland, Massachusetts, New York, and Iowa. Non-expansion states include Florida and North Carolina. Excluded patients <19 & >64 years of age and Medicare patients.

In the states that did not expand Medicaid, no difference was seen in the proportion of patients having surgery for diverticulitis covered by Medicaid (9% vs. 9%, p = 0.21) after the ACA was enacted; however, there was an increase in the proportion of those with private insurance (80% vs. 77%, p = 0.001) and a decrease in self-pay/no-charge (11% vs. 13%, p = 0.001). It’s unclear why states that did not expand insurance have an increased proportion of privately insured patients. This could secondary to reduced cost of purchasing private insurance or private employers offering improved health benefits to their employees. In both groups of states, there was no significant difference in age, gender, comorbidities, and race/ethnicity, before or after the expansion in patients undergoing surgery for diverticulitis. (Table 2)

Diverticulitis Related Admission Rates Before and After ACA

We evaluated the changes in the rates of admission for diverticulitis for patients before and after the ACA Medicaid expansion to better understand the impact of insurance expansion on diverticulitis related admission. The risk-adjusted rates of diverticulitis related admission before and after Medicaid expansion are summarized in Table 3. We found a seven-point unadjusted difference in the rate of patients with a diverticulitis diagnosis being admitted with Medicaid insurance after the expansion in those states that expanded Medicaid compared to the non-expansion states. In our risk-adjusted analysis, adjusted for age, gender, race/ethnicity, and Charlson comorbidity index, we found that the seven-point difference translated into a 45% (IRR 1.45, p = 0.001) increased rate of diverticulitis patients being insured by Medicaid in expansion states compared to non-expansions states. In expansion states, there was a decrease in the number of patients with private insurance by 3% (IRR 0.97, p = 0.04) and self-pay/no-charge by 29% (IRR 0.71, p = 0.001) compared to non-expansions states. (Table 3)

Table 3.

Risk-Adjusted rates of admissions pre and post-ACA by insurance type

	Pre	Post	∆ in Rate	DID	IRR*	P value
Medicaid Insurance
Expansion	16.07	20.19	4.11	7.09	1.45	0.001
Non-expansion	20.77	17.80	−2.97
Private Insurance
Expansion	19.11	17.21	−1.90	−2.75	0.97	0.04
Non-expansion	28.20	29.05	0.85
Self-Pay
Expansion	5.34	3.72	−1.62	−2.22	0.71	0.001
Non-expansion	10.05	10.65	0.60

Adjusted for age, gender, race, and comorbidities. Pre-ACA period is from 2010 to 2013; Post-ACA is 2014. Difference in difference (DID); incidence rate ratio (IRR). Rate is per every 100,000 people.

Diverticulitis Related Surgeries for Diverticulitis Before and After ACA

To better understand the effect of insurance expansion on diverticulitis related surgeries in expansion states compared to non-expansion states by admission type, we evaluated the changes in rates of diverticulitis surgeries before and after the ACA Medicaid expansion. We hypothesized that improved access might lead to an increase in the utilization of surgery for patients with diverticulitis. The risk-adjusted rates of diverticulitis related surgeries before and after Medicaid expansion are summarized in Table 4. In our risk-adjusted analysis, adjusted for age, gender, race/ethnicity, and Charlson comorbidity index, following the implementation of the ACA, there was a 80% increase in Medicaid (IRR 1.80, p = 0.001) and a 33% decrease in self-pay (IRR 0.67, p = 0.001) surgeries in states that expanded Medicaid compared to those that decided not to expand. There was no change in the rate of privately insured patients (IRR 0.96, p = 0.14) undergoing surgery in expansion states compared to those that did not expand Medicaid. (Table 4)

Table 4.

Risk-adjusted rates of surgery pre and post-ACA by insurance type

	Pre	Post	∆ in Rate	DID	IRR*	P value
Medicaid Insurance
Expansion	15.00	17.96	2.96	4.15	1.80	0.001
Non-expansion	14.79	13.60	−1.20
Private Insurance
Expansion	25.22	26.11	0.89	1.18	0.96	0.14
Non-expansion	25.48	25.20	−0.29
Self-Pay
Expansion	13.68	12.47	−1.21	0.00	0.67	0.001
Non-expansion	14.10	12.89	−1.21

Adjusted for age, gender, race, and comorbidities. Pre-ACA period is from 2010 to 2013; Post-ACA is 2014. Difference in difference (DID); incidence rate ratio (IRR). Rate is per every 100 admissions.

In a sensitivity analysis, we examined whether there were any significant changes between groups relative to one another before Medicaid expansion occurred in 2014. We found that there was no differential trend in surgery rates for diverticulitis between Medicaid (p = 0.11) and private (p = 0.12) insurance patients in expansion and the non-expansion states before the Medicaid expansion.

DISCUSSION

The ACA introduced insurance exchanges, and, in participating states, expansion of Medicaid in 2014.¹⁵ In general, studies have found that the ACA led to increased rates of insurance coverage, access to primary care and medications, and healthcare affordability; however, little is known about its impact on the treatment of diverticular disease.^9,16,17 In this study, we sought to examine how the ACA Medicaid expansion may have impacted the insurance status of those who were being hospitalized or having surgery for diverticular disease. To do this, we looked at the population of patients being treated for diverticulitis in two groups of states (expansion and non-expansion) at two different periods (four years before and one year after Medicaid expansion).

We had several interesting findings. When looking at hospital admissions for a diagnosis of diverticulitis, we found that both groups of states were similar in that their proportions of self-pay patients decreased in 2014 compared to the four years prior. The difference between the two groups was in what happened with the proportions of the other two insurance categories. Specifically, the expansion group’s decrease in self-pay coincided with an increase in Medicaid and a decrease in private insurance. The non-expansion group’s decrease in self-pay coincided with no change in Medicaid and an increase in the privately insured.

The fact that the expansion states showed an increase in those with Medicaid is intuitive; however, the simultaneous decrease in the proportion of privately insured is not. It could be a coincidence, but it raises the possibility that because the Medicaid opportunity was available, it not only drew in some self-pay patients as it was intended, but also some patients that were previously insured by their employer became eligible for Medicaid through the expansion and siphoned off patients who otherwise would have opted for private insurance. Future studies are needed using more granular state level longitudinal data to evaluate whether patients were transitioning from insurance group or there was an increase in the rate of patients who never had access to insurance.

As previously mentioned, the non-expansion group also showed a decrease in self-pay; but, unlike the expansion states, a corresponding increase was not seen in Medicaid. An increase was seen, however, in the privately insured. It is possible that because there was not a Medicaid expansion option in these states, more of these patients used the newly created health insurance exchanges purchase private plans. More than 1.2 million people combined in Florida and North Carolina enrolled in the ACA private insurance Exchanges in 2014.¹⁸

When we examined Medicaid patients alone and compared their rates of risk-adjusted hospital admissions between the two groups of states, we identified another interesting trend. Compared to the non-expansion states, the 2014 rates of admission for Medicaid patients in expansion states was greater. When we examined how the rate of surgery changed over time to identify if the increase in admissions led to an increase in surgical care amongst the various insurance categories, we found that the expansion states had an 80% increase in the rate of diverticulitis surgery done on Medicaid patients. Thus, our data suggest that after Medicaid was expanded, more diverticulitis related hospitalizations and surgery occurred on Medicaid patients than in prior years. Although the incidence of diverticulitis may have stayed the same. Our results showed there was a change in the rate of hospital admissions (give fluctuation in state-level populations during the time period) and surgeries in states that expanded Medicaid compared to those states that decided to opt out of the expansion.

These findings are consistent with literature previously identifying an association between insurance expansion and access to medical and surgical care.^10,19 In a previous study from our group, we found that the ACA Medicaid expansion was associated with an increase in lung and colorectal cancer surgeries in states that expanded Medicaid compared to non-expansion states.¹⁰ A study by Zerhouni et al. found that overall screening in expansion states increased compared to non-expansion states with the effect of increasing screening for low-income patients.²⁰

Our observation of an increased rate of hospitalization and surgery for diverticulitis in Medicaid patients in expansion states could have varying implications. On the one hand, it could represent the relief of pent up need resulting from increased access to care. In other words, pre-ACA, some self-pay patients may have been living with bad diverticular disease because treatment was unaffordable. Then, when Medicaid was expanded, and they became eligible, they obtained insurance and sought treatment. This hypothesis is supported in the literature for other surgery-related diagnoses such as thyroid, lung, and gastrointestinal cancers.^10,13,21,22

From an economics standpoint, our study raises the issue of the price elasticity of demand for diverticulitis care. It is possible that in the expansion states, the Medicaid expansion acted like a price reduction for care and resulted in increased demand. One would expect to see elastic behavior occur more in elective preventative services without many risks; but perhaps hospitalization and surgery for diverticulitis, even with its inherent risks (i.e., colostomy), behaves, from an economic standpoint, with more elasticity than would be apparent on the surface. On the other hand, the increased rates of hospitalization and surgery for diverticulitis in Medicaid patients could reflect a phenomenon where now that patients were newly insured the medical community itself was more willing to treat them. In other words, it is possible that the newly obtained Medicaid may have played a role in “inducing” treatment by the medical community to some extent.

The importance of this study is that it adds to the narrative describing what may happen to healthcare utilization when access is increased. Increasing access to care may have the unintended effect of inducement to treat.^23,24 This study also reveals that there may be a shift in the proportion of patients that go from being eligible for private insurance but with insurance expansion they can cross over to government subsidized insurance. Diverticulitis, in both the acute and elective setting, is one of those surgical conditions where there is variability regarding when operations are offered by surgeons.²⁵ Also, future studies are needed to determine the effect of Medicaid expansion on the costs of care and the financial implications for states who participated in Medicaid expansion. Also future studies are needed examining the current barrier uninsured patients have accessing affordable healthcare insurance.²⁶

There are several limitations to this study which have been discussed in greater detail in prior publications.^10,27 It is a retrospective review of an administrative dataset and cannot definitively control or adjust for the selection biases that are inherent in these datasets. The administrative dataset does not contain detailed clinical information on many physiologic and pathologic measures which may contribute to patient selection for different interventions and determine outcomes. HCUP does not include surgeon specific information; therefore, we cannot determine the subspecialty of the surgeon performing the procedure or factors that led to the surgeon’s decision making. Another limitation of our study is that the dataset that we used only includes inpatient encounters. We are not able to evaluate those patients that were evaluated in the emergency department but were never admitted. Finally, the study was limited to a few states and years (2010–2014); therefore, more recent rates of utilization were not examined and cannot be extrapolated to the entire US population.

CONCLUSION

The Affordable Care Act helped increased access to healthcare eliminating discrimination by insurers based on pre-existing conditions, subsidized insurance exchanges and expanding Medicaid. There was limited data on the impact of ACA Medicaid expansion had on rates of diverticulitis related hospitalizations or surgery. We found in our study that in states that expanded Medicaid coverage under the ACA, there was an increase in the rate of Medicaid patients undergoing medical treatment and surgery for diverticulitis compared to non-expansion states. This suggests that increased access to care results in increased utilization of healthcare resources even when it comes to surgery for diverticulitis. From an economics standpoint, our study raises the issue of the price elasticity of demand for diverticulitis care. It is possible that in the expansion states, the Medicaid expansion acted like a price reduction for care and resulted in increased demand. On the other hand, the increased rates of hospitalization and surgery for diverticulitis in Medicaid patients could reflect a phenomenon where now that patients were newly insured the medical community itself was more willing to treat them.