a b s t r a c t

Objective: We sought to identify barriers and delays in care associated with the increased prevalence of Perforated appendicitis among Colorado’s pediatric Medicaid population.

Methods: We conducted a retrospective cohort study of all cases of Pediatric appendicitis, which had Colorado Medicaid from 2007 to 2008 using descriptive statistics, bivariate analysis, and multivariable logistic regression. Results: Of the 479 appendicitis cases, 42.6% were perforated. In both the bivariate and multivariate analysis, perforated cases did not significantly differ from nonperforated cases with respect to sex, rurality of residence, or race with the exception of Black race in the multivariate model. Perforated cases were more likely to be younger, have been enrolled in Medicaid for less than 6 months, have seen a provider within 5 days of their diagnosis, and have been transferred to another hospital for treatment.

Conclusions: The high prevalence of perforated appendicitis in Colorado children with Medicaid coverage is not associated with race or physical proximity to care but may be associated with the duration of Medicaid coverage, which highlights the importance of establishing medical homes to direct patients on where and how to seek care.

(C) 2013

Introduction

Appendicitis is the most common surgical emergency in children. Without timely treatment, acute appendicitis progresses to appendi- ceal perforation resulting in increased morbidity, mortality, hospital length of stay, and subsequent increased Costs of care. Time to definitive treatment has been shown to be the most predictive indicator of perforation [1-4]. Consequently, acute appendicitis is an ideal disease for analyzing disparities in access and quality of care that contribute to perforation.

The prevalence of perforated appendicitis in children has been stable for the past 25 years, averaging 31.5% (range, 30%-32.5%) [4-7]. Multiple studies have shown an association between prevalence of perforation and Insurance type [4-9]. Published prevalence of perforated appendicitis in children with Medicaid coverage averages 39.1% (range, 35.6%-43%), whereas prevalence in children with private insurance averages 28.8% (range, 25.7%-30.9%) [4-7]. In the

? Funding sources/disclosures: N/A.

?? Prior presentations: Poster Presentation at 2011 American College of Emergency Physicians Scientific Assembly.

* Corresponding author.

E-mail address: [email protected] (S.A. Trent).

state of Colorado, prevalence of perforated appendicitis in children with Medicaid coverage has consistently been higher than 40% (J Zerzan, oral communication, July 2010). The goal of this study was to identify potential factors associated with perforated appendicitis in Colorado’s pediatric Medicaid population.

Methods

Study data and sample

We extracted data from Colorado’s Medicaid Claims database to conduct a retrospective cohort study on all children (ages 0-17 years) with acute appendicitis from January 1, 2008, through December 31, 2009. All data used in this study were extracted from the Medicaid Management Information System by a statistical analyst at the Colorado Department of Health Policy and Financing. The data were then de-identified before use by the authors of this study. Claims data from all children age 0 to 17 years with International Classification of Diseases, Ninth Revision, codes indicating a diagnosis of appendicitis from January 1, 2008, through December 31, 2009 were obtained. All claims data for each child from time of discharge to 3 months prior were included. Age, sex, race/ethnicity, and county of residence at time of discharge were taken from information on the claim at the

0735-6757/$ – see front matter (C) 2013 http://dx.doi.org/10.1016/j.ajem.2012.09.009

470 S.A. Trent et al. / American Journal of Emergency Medicine 31 (2013) 469–472

time of adjudication. Data on duration of Medicaid coverage were taken from all eligibility spans that began on or before the first date of service of the appendectomy claim. No cases were excluded.

Study variables

The main outcome variable was appendiceal perforation at time of discharge. The International Classification of Diseases, Ninth Revision, codes [10] were used to differentiate nonperforated appendicitis (540.9 and 541) from perforated appendicitis (540 and 540.1). Independent variables included age, sex, race/ethnicity, rurality of residence, duration of Medicaid coverage, preceding physician visit, and transfer.

Patient demographics

Age in years was included as a continuous variable given the linear relationship between decreasing age and appendiceal perforation in the pediatric population [1,2,5,6,11]. Sex was included as a categorical variable given the mixed association between sex and perforation in the pediatric population [6,11-13]. Race/ethnicity was also included as a categorical variable given the mixed association between race/ ethnicity and appendiceal perforation in the pediatric population. [5,6,8,11,12,14,15] In this study, data on race/ethnicity were self- reported and categorized as Hispanic, white, black, Asian, American Indian, Native Hawaiian/Pacific Islander, other, and unknown. Given their small numbers, Asian, American Indian, Native Hawaiian/Pacific Islander, and other were combined into a single “other” group. Race/ ethnicity was unknown for 80 patients (16.7%), and it was only variable with missing data.

Rurality of residence

Rurality of residence was included as a categorical variable related to physical access to care. County of residence for each patient was differentiated into rural and urban counties as designated by the Colorado Rural Health Center [16]. Frontier and rural counties were both designated as rural.

Duration of Medicaid coverage

Duration of Medicaid coverage was included as a continuous variables related to perceived access to care. We hypothesized that there may be an association between duration of Medicaid coverage and prevalence of perforated appendicitis, such that those patients who had Medicaid coverage longer would be less likely to perforate because they would have a better understanding of how to use their Medicaid coverage and may have established a primary care physician or a medical home. Consequently, these patients and their families would be more likely to know where to seek timely care and be less concerned about financial implications of seeking acute care.

Preceding physician visit

Preceding physician visit was included as a categorical variable related to potential Delays in diagnosis and definitive treatment. A preceding physician visit indicates that a patient had seen a provider within 5 days of their diagnosis of appendicitis. All visits to emergency departments, urgent cares, and outpatient clinics were included. Visits related to mental health, dental, and eye care were excluded given that it could not be determined if these patients saw a physician. Preceding physician visits were manually coded by one author, and a percentage was coded by another author. Interrater reliability was then calculated using Cohen’s ? and was perfect (1.0).

Transfer

Transfer was included as a categorical variable related to potential delays in definitive treatment. Transfer indicates that a patient was diagnosed and treated at different hospitals. Like preceding physician visits, transfer was manually coded by one author, and a percentage

was coded by another author. Interrater reliability was then calculated using Cohen’s ? and was good (0.72). Interrater reliability was perfect (1.0) for the preceding physician visit cases and good (0.72) for the transfer cases.

Socioeconomic status

socioeconomic status is associated with access to care as it relates to ability to pay for care. Additional data on SES, however, were not obtained given that all children covered by Medicaid in the state of Colorado have similar SES (b 133% Federal poverty level (FPL) for 0-5 years of age and b 100% FPL for 6-19 years of age). Moreover, no copayment is required for emergency care, hospital- izations, or emergency surgeries. Consequently, direct costs should not be a factor in accessing care for this cohort.

Statistical analysis

Descriptive statistics were calculated for all variables. Continuous data were reported as medians with interquartile ranges (IQRs) and categorical data as percentages with 95% confidence intervals (CIs). Bivariate statistical testing was performed, where appropriate, using the Wilcoxon rank sum test for continuous data or Fisher exact test for categorical data. Multivariable logistic regression analysis was used to model the associations between patient demographics (age, sex, and race/ethnicity), rurality of residency, duration of Medicaid coverage, preceding physician visit, transfer, and perforation at the time of discharge. Missing values were handled using multiple imputation (MI) to minimize bias and preserve study power. We used a Markov Chain Monte Carlo approach for MI with generation of 10 imputed data sets, each analyzed independently and combined using Rubin’s rules to appropriately account for variance within and between data sets. The MI model included age (0% missing), sex (0%), race/ethnicity (16.7%), county (0%), duration of Medicaid (0%), preceding physician visit (0%), and transfer (0%). All data were maintained in Microsoft Excel (Microsoft Corporation, Redmond, WA) and analyzed using SAS Version 9.2 (SAS Institute, Inc, Cary, NC).

Results

Of the 479 cases of acute appendicitis during the study period, 203 cases were perforated appendicitis (42.4%). The median age was 10 (IQR, 6-14), and 61.4% (95% CI, 56.9-65.6) were male. In addition,

52.6% (95% CI, 48.1-57.0) were Hispanic, 22.3% (95% CI, 18.8-26.3)

were white, 2.9% (95% CI, 1.7-4.8) were black, 5.4% (95% CI, 3.7-7.8)

were other, and 16.7% (95% CI, 13.6-20.3) were unknown (Table 1).

In the unadjusted bivariate analysis, perforation was significantly associated with age, duration of Medicaid coverage, preceding provider visit, transfer, and length of stay (Table 2). Children with perforated appendicitis were significantly more likely than children with nonperforated appendicitis to be younger than 5 years (odds ratio [OR], 5.38; 95% CI, 3.01-9.63), have been in enrolled in Medicaid for less than 6 months (OR, 1.90; 95% CI, 1.30-2.79), have seen a physician within 5 days of their diagnosis of appendicitis (OR, 3.10; 95% CI, 1.79-5.38), have been transferred to another hospital for surgery (OR, 2.49; 95% CI, 1.34-4.60), and have stayed in the hospital for more than 2 days following their diagnosis (OR, 36.18; 95% CI, 18.29-71.59). After removing cases where the race was unknown, racial minorities were found to be more likely to perforate than whites, although not statistically significant (OR, 1.28; 95% CI, 0.81-2.04).

In the multivariate model, perforation was again significantly associated with age, duration of Medicaid coverage, preceding physician visit, and transfer as well as black race (Table 3). Younger patients were more likely (OR, 0.88; 95% CI, 0.84-0.92) to perforate than older patients. Patients with less than 6 months of Medicaid coverage less were more likely (OR, 2.3; 95% CI, 1.51-3.51) to

S.A. Trent et al. / American Journal of Emergency Medicine 31 (2013) 469–472 471

Table 1 Unadjusted patient characteristics
	Total % (95% CI)	Nonperforated % (95% CI)	Perforated % (95% CI)
Patient sample (%)	479	276 (57.6)	203 (42.4)
Age Median (IQR)	10 (6-14)	11 (7.75-14)	8 (4-12)
0-4	14.6 (11.7-18.1)	24.3 (15.7-35.6)	75.7 (64.5-84.2)
5-9	34.9 (30.7-39.2)	56.3 (48.7-63.6)	43.7 (36.4-51.3)
10-14	33.8 (29.7-38.2)	63.0 (55.3-70.0)	37.0 (30.0-44.7)
15-17	16.7 (13.6-20.3)	78.8 (68.6-86.3)	21.2 (13.7-31.4)
Sex Male	61.4 (56.9-65.6)	59.2 (53.5-64.6)	40.8 (35.3-46.5)
Female	38.6 (34.4-43.1)	55.1 (47.9-62.1)	44.9 (37.9-52.1)
Race Hispanic	52.6 (48.1-57.0)	61.1 (55.0-66.9)	38.9 (33.1-45.0)
White	22.3 (18.8-26.3)	65.4 (56.0-73.8)	34.6 (26.2-44.0)
Black	2.9 (1.7-4.8)	35.7 (16.3-61.2)	64.3 (38.8-83.7)
Other	5.4 (3.7-7.8)	57.7 (38.9-74.5)	42.3 (25.5-61.0)
Unknown	16.7 (13.6-20.3)	40.0 (30.0-50.9)	60.0 (49.0-70.0)
Residence Urban	81.4 (77.7-84.6)	57.0 (52.0-61.7)	43.0 (38.2-48.0)
Rural Duration of Medicaid	18.5 (15.3-22.3)	60.7 (50.3-70.2)	39.3 (29.8-49.7)
Median (IQR)	394 (58-1046.5)	519 (132.75-1204.75)	322 (34.5-861)
0-180 d	34.4 (30.3-38.8)	47.3 (39.8-54.9)	52.7 (45.1-60.2)
N 180 d	65.5 (61.2-69.7)	63.1 (57.6-68.2)	36.9 (31.8-42.4)
Preceding physician visit	13.6 (10.8-16.9)	33.8 (23.5-46.0)	66.2 (54.0-76.5)
Transfer	10.0 (7.6-13.0)	37.5 (25.2-51.6)	62.5 (48.4-74.8)
Length of stay Median (IQR)	3 (2-5)	2 (2-3)	6 (4-8)
1-2 d	39.7 (35.4-44.1)	94.7 (90.6-97.1)	5.3 (2.9-9.4)
3-4 d	27.6 (23.8-31.7)	66.7 (58.3-74.1)	33.3 (25.9-41.8)
5-6 d	16.1 (13.1-19.6)	7.8 (3.6-15.9)	92.2 (84.0-96.4)
7-8 d	10.2 (7.8-13.3)	0.0 (0.0-7.3)	100 (92.7-100)
9-10 d	2.7 (1.6-4.6)	7.7 (1.4-33.3)	92.3 (66.7-98.6)
N 10 d	3.7 (2.4-5.9)	5.6 (0.99-25.8)	94.4 (74.2-99.0)

perforate than patients with longer durations of coverage. Patients who had seen a provider within 5 days of their diagnosis of appendicitis were more likely (OR, 2.97; 95% CI, 1.64-5.4) to perforate than patients who had not. Those patients who were transferred were found to have higher prevalence of perforation (OR, 2.19; 95% CI, 1.13- 4.27). Lastly, black patients were also found to be more likely (OR, 3.84; 95% CI, 1.13-13.02) than white patients to perforate.

Table 2

Bivariate analysis

Discussion

More than 7 years has passed since the last study on perforated appendicitis was published, and yet, disparities continue with a much higher prevalence of perforation in the Medicaid population. With Medicaid budgets already stretched thin and the coming increase in Medicaid coverage, it is important to recognize that this disparity gap still exists and accounts for significant direct costs to Medicaid budgets as well as indirect costs to patients and their family. If the disparity gap was eliminated in the state of Colorado, an additional

$2148 per patient treated in an urban Colorado hospital would be

Total n (%)

Nonperforated n (%)

Perforated n (%)

OR (95% CI)

saved in direct costs, a yearly saving of $162 050, without accounting for the indirect direct costs to families of having a child in the hospital.

Age 5.38 (3.01-9.64)

b 5	70 (14.6)	17 (24.3)	53 (75.7)
>= 5	409 (85.4)	259 (63.3)	150 (36.7)
Sex				1.18 (0.81-1.71)
Male	294 (61.4)	174 (59.2)	120 (40.8)

Table 3

Multivariate analysis

Adjusted OR 95% CI

Female	185 (38.6)	102 (55.1)	83 (44.9)
Race (excluding				1.28 (0.81-2.04)	Age	0.88	(0.84-0.92)
unknown)					Sex
White	107 (26.8)	70 (65.4)	37 (34.6)		Male	1.00	–
Minority race	292 (73.2)	174 (59.6)	118 (40.4)		Female	0.79	(0.53-1.19)
Residence				0.86 (0.54-1.37)	Race
Urban	390 (81.4)	222 (57.0)	168 (43.0)		Hispanic	1.01	(0.59-1.74)
Rural	89 (18.5)	54 (60.7)	35 (39.3)		White	1.00	–
Duration of				1.90 (1.30-2.79)	Black	3.84	(1.13-13.02)
Medicaid					Other	1.58	(0.64-3.87)
<= 180 d	165 (34.4)	78 (47.3)	87 (52.7)		County
N 180 d	314 (65.6)	198 (63.1)	116 (36.9)		Urban	1.00	–
Preceding	65 (13.6)	22 (33.8)	43 (66.2)	3.10 (1.79-5.38)	Rural	1.17	(0.69-1.99)
physician visit					Duration of Medicaid
Transfer	48 (10.0)	18 (37.5)	30 (62.5)	2.49 (1.34-4.60)	0-180 d	2.30	(1.51-3.51)
Length of stay				36.18 (18.29-71.59)	>= 181 d	1.00	–
<= 2 d	190 (40.0)	180 (94.7)	10 (5.3)		Preceding physician visit	2.97	(1.64-5.4)
N 2 d	289 (60.0)	96 (33.2)	193 (66.8)		Transfer	2.19	(1.13-4.27)

472 S.A. Trent et al. / American Journal of Emergency Medicine 31 (2013) 469–472

The reasons for this continued disparity are poorly understood, and further study is needed. Patient factors, which are difficult to assess from large administrative databases, likely play a significant role. Our study is consistent with other studies showing an association between younger age and appendiceal perforation [1,2,5,6,11]. Younger patients may be more difficult to communicate with and examine and can also present atypically. Although 1 study showed an association between male sex and perforation [6], our study is consistent with the majority of studies that do not show an association between perforation and sex [11-13]. Multiple studies have evaluated racial and Ethnic disparities and their associations with appendiceal perforation, and most studies have found little association [5,6,8,14]. However, 2 larger studies have found evidence of racial and ethnic disparities in appendiceal Perforation rates [11,12,15]. In the multivariate analysis in our study, blacks were the only race associated with increased odds of perforation. The diminished association between race and appendiceal perforation in this study may be reflective of the inherent similarities of socioeco- nomic status and payer source in this cohort.

In addition to the demographic characteristics in this cohort, we analyzed four additional factors that could represent delays or barriers to prompt care. Rurality of residence was not associated with increased prevalence of perforated appendicitis in this cohort indicating that physical proximity to a hospital did not impact a patient’s ability to seek care when needed. Duration of Medicaid coverage was significantly associated with perforated appendicitis. The median numbers of days of Medicaid coverage in the nonperfo- rated group is 519 days, suggesting a potential association of a minimum eligibility period of nearly a year and a half is necessary for patients to know how to obtain care with their coverage. Once patients present to a medical facility, however, the ready availability of diagnostic technology should improve diagnosis, and yet, in this population, 13.6% of the all patients had seen a provider before their diagnosis of appendicitis suggesting a possible missed appendicitis. Of those possibly missed, 66.2% were perforated. Finally, there may be delays in getting to the operating room that may contribute, as suggested by the increased prevalence of perforation in patients who were transferred to another hospital. However, patients who were transferred may have been more ill or may have been transferred due to a lack of capability of the facility to care for these Pediatric cases, but this would be expected to be similar in both Medicaid and commercial coverage and likely does not explain the gap.

4.1. Limitations

The use of administrative data has recognized limitations for this study, but it is the only available data source for analyzing this cohort. Given that time to definitive treatment is the most predictive indicator of perforation, time to presentation, diagnosis, and treat- ment are important factors for establishing where delays and barriers exist. We are unable to establish these definitive times with only administrative data. Moreover, demographic data regarding race/ ethnicity were self-reported and not required. Consequently, race/ ethnicity data were unknown for 16.7% of this cohort. We used multiple imputation to account for this and include it in the

multivariate analysis, however. Additional information such as immigration status and English language skills, which affect accul- turation and familiarity with the health care system, can also not be assessed with these data. Lastly, we were unable to compare the prevalence of perforated appendicitis in the Colorado pediatric population with different Insurance types as the available data to answer this question are limited in its ability to identify the true payor and thus do not accurately reflect the prevalence of perforated appendicitis in pediatric patients in Colorado with private insurance, Medicaid, and no insurance.

Conclusion

policy makers need to work both with patients and providers to decrease this preventable complication, which both increases patient morbidity and mortality as well as patient and system costs. Improving Primary care access to patients with Medicaid coverage and establishing medical homes through accountable care organiza- tions at the time coverage is established will likely point patients and their families in the right direction as to where and how to seek care. Continued work to decrease the prevalence of perforated appendicitis will not only improve the quality of patient care but will also account for significant cost savings.

References

1. Korner H, Sondenaa K, Soreide JA, et al. Incidence of acute nonperforated and perforated appendicitis: age-specific and sex-specific analysis. Word J Surg 1997;21(3):313-7.
2. Brender JD, Marcuse EK, Koepsell TD, et al. Childhood appendicitis: factors associated with perforation. Pediatrics 1985;76(2):301-6.
3. Cappendijk VC, Hazebroek FWJ. The impact of diagnostic delay on the course of acute appendicitis. Arch Dis Child 2000;83:64-6.
4. O’Toole SJ, Karamanoukian HL, Allen JE, et al. Insurance-related differences in the presentation of pediatric appendicitis. J Pediatr Surg 1996;31(8):1032-4.
5. Bratton SL, Haberkern CM, Waldhausen JHT. Acute appendicitis risks of complications: age and Medicaid insurance. Pediatrics 2000;106:75-8.
6. Gadomski A, Jenkins P. Ruptured appendicitis among children as an indicator of access to care. Health Serv Res 2001;36(1 Pt 1):129-42.
7. Smink DS, Fishman SJ, Kleinman K, et al. Effects of race, insurance status and hospital volume on perforated appendicitis in children. Pediatrics 2005;115: 920-5.
8. Bravemen P, Schaaf VM, Egerter S, et al. Insurance-related differences in risk of ruptured appendix. N Engl J Med 1994;331:444-9.
9. Krajewski SA, Hameed SM, Smink DS, et al. Access to emergency operative care: a comparative study between the Canadian and American health care systems. Surgery 2009;146:300-7.
10. World Health Organization. International Classification of Diseases. Ninth Revision (ICD-9). 2012. Available at: http://www.who.int/classifications/icd/en.
11. Jablonski KA, Guagliardo MF. Pediatric appendicitis rupture rate: a national indicator of disparities in healthcare access. population health Metrics 2005;3:4.
12. Guagliardo MF, Teach SJ, Huang ZJ, Chamberlain JM, et al. Racial and ethnic disparities in pediatric appendicitis rupture rate. Acad Emerg Med 2003;10: 1218-27.
13. Ponsky TA, Huang ZJ, Kittle K, et al. Hospital- and patient-level characteristics and the risk of appendiceal rupture and Negative appendectomy in children. JAMA 2004;292:1977-82.
14. Buckley RG, Distefan J, Gubler KD, Slymen D. The risk of appendiceal rupture based on hospital admission source. Acad Emerg Med 1999;6:596-601.
15. Kokoska ER, Bird TM, Robbins JM, Smith SD, et al. racial disparities in the management of pediatric appendicitis. J Surg Res 2007;137(1):83-8.
16. www.coruralhealth.org/programs/rhc/documents/rhcbycounty.pdf.