Health disparities in infants with hypertrophic pyloric stenosis

The American Journal of Surgery (2017) 214, 329-335

Health disparities in infants with hypertrophic pyloric stenosis Alexander Feliz, M.D.a,*, Janette L. Holub, M.D., M.P.H.b, Nima Azarakhsh, M.D.b, Marielena Bachier-Rodriguez, M.D.b, Kate B. Savoie, M.D.b a

Departments of Surgery and Pediatrics, bDepartment of Surgery, University of Tennessee Health Science Center, Memphis, TN, USA

KEYWORDS: Hypertrophic pyloric stenosis; Race; Ethnicity; Socioeconomic status; Health disparities

Abstract BACKGROUND: This study investigates whether health disparities exist in infants with hypertrophic pyloric stenosis (HPS), to identify factors affecting definitive treatment, and if more morbidity occurs. METHODS: A 6-year retrospective analysis was performed on infants with HPS. Analysis of variance was used to evaluate the impact of socioeconomic factors on disease severity and hospitalization. General linear models were used to assess the impact of risk factors on the outcomes. RESULTS: There were a total of 584 infants. African-American’s had lower serum chloride (P , .001), higher bicarbonate (P 5 .001), and sodium levels (P 5 .006), adding to longer hospitalization than whites (P 5 .03). Uninsured infants had lower sodium and chloride (P , .001) and higher bicarbonate (P , .001), resulting in a longer time to operation (P 5 .05) than privately insured infants. In multivariable analyses, African-American’s were associated with chloride (P 5 .002) and higher bicarbonate (P 5 .009), and uninsured status remained significantly associated with all electrolyte abnormalities. CONCLUSIONS: African-American and poorly insured infants with HPS had greater risk of metabolic derangements. This required more time to correct dehydration and electrolytes, adding to longer hospitalizations. Ó 2016 Published by Elsevier Inc.

There were no relevant financial relationships or any sources of support in the form of grants, equipment, or drugs. The authors declare no conflicts of interest. Clinical Trial Registration: None. Dr. Feliz supervised the conceptualization and design of the study, drafting of the manuscript, and approved the final manuscript as submitted. Dr. Holub conceptualized and designed the study, drafted the initial manuscript, and approved the final manuscript as submitted. Dr. Azaraksh designed the data collection instruments, coordinated and supervised data collection, critically reviewed the manuscript, and approved the final manuscript as submitted. Dr. Bachier-Rodriguez assisted in the drafting, editing, and approval of the final manuscript. Dr. Savoie assisted in the drafting, editing, and approval of the final manuscript. * Corresponding author. Tel.: 11 (901) 287-6031; fax: 11 (901) 287-4434. E-mail address: [email protected] Manuscript received April 4, 2016; revised manuscript June 22, 2016 0002-9610/$ - see front matter Ó 2016 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.amjsurg.2016.07.009

One of the original Healthy People initiatives, and ‘‘overarching goals’’ of Healthy People 2020,1 is ‘‘Maternal, Infant, and Child Health’’ which focuses on achieving health equity, eliminating disparities, and improving the health of all groups in the United States. Social determinants have great influence on infant health. Racial and ethnic disparities in infant mortality exist, particularly for African-American infants. Understanding how social determinant factors contribute to risk of illness and mortality is often complicated, due to the high correlation between many of these factors. Studying how these factors affect diseases of infancy may help add clarity to the care of infants at risk.

330 Currently, there is a paucity of work examining whether social determinants affect care in infants with hypertrophic pyloric stenosis (HPS), and few studies in the literature focus on the presenting factors and resultant potential morbidity for pediatric surgical diseases.2,3 It has been suggested that disease models for the assessment of barriers to access to health care should have no or few clear modifiable risk factors and be fairly uniform across race, ethnicity, and socioeconomic status.4,5 HPS represents an excellent model to evaluate disparities in infant health, care utilization, and outcomes due to its relative frequency, natural disease progression, and standardized treatment. The prevalence of HPS ranges from 1.5 to 4.0 per 1,000 live births in white infants, with less prevalence in African-American and Asian children, making it one of the most common surgical diseases in infancy.6–9 Although a male predominance has been well reported, no clear modifiable risk factor has been identified.10,11 Infants with HPS present with progressive, projectile emesis from gastric outlet obstruction due to hypertrophy of the pyloric musculature. Operative treatment is curative; however, surgical intervention is preceded by correction of acid-base and fluid derangements present at the time of diagnosis. Delays in presentation lead to greater metabolic derangements, placing these infants at risk for requiring higher acuity of care and longer hospitalizations. Once stabilized, pyloromyotomy allows for complete resolution of the obstruction, and most infants are discharged 24 to 48 hours after their operation.12 The present study seeks to determine if health disparities exist in the presentation of infants with HPS and to identify factors that may delay definitive surgical treatment. We hypothesize that socioeconomic determinants of health, such as race/ethnicity and insurance status, affect the presentation of infants with HPS, resulting in a subset of infants presenting more acutely ill, requiring greater care, and longer hospitalization before definitive surgical intervention.

Methods This is a retrospective cohort study performed after Institutional Review Board approval by the University of Tennessee Health Science Center. A chart review of the medical records of infants who underwent pyloromyotomy for HPS (International Classification of Diseases - version 9 [ICD-9]: 750.5, 537.0) during a 5-year period (2007 to 2012), at Le Bonheur Children’s Hospital, a tertiary freestanding pediatric hospital, was performed; all infants with a confirmed intraoperative diagnosis of HPS were included in the final analysis. The primary outcome of illness acuity at time of admission was assessed by the presence of electrolyte abnormalities (specifically, serum chloride, bicarbonate, sodium, and potassium levels) and time to operation (ie, length of hospitalization before definitive surgical

The American Journal of Surgery, Vol 214, No 2, August 2017 treatment). Increased morbidity was defined as an increased total length of hospital stay, compared with the control group. The diagnosis of pyloric stenosis is made with the assistance of ultrasonography, which has been readily available at our institution, in most infants. Infants are taken for operative intervention once they are appropriately resuscitated, which is defined as having adequate urine output and normalized electrolytes, according to institutional standard reference values: chloride greater than 100 meq/L and serum bicarbonate level less than 30 mmol/L. At our institution, volume resuscitation is based on a standardized protocol. The infants receive 1 or 2 10 mL/kg boluses of isotonic fluid in the emergency department to establish urine output. They then receive a standard resuscitative fluid at a rate of 150 mL/kg/day. Laboratory values are checked each morning, and operative plan is decided depending on these values. Hence, greater volume requirements require greater time length of stay before their definitive operative therapy. Patient demographic information included gender, race/ ethnicity, and insurance status. Patient illness characteristics included days of symptomatic emesis before index hospitalization and admission weight. Descriptive statistics including mean 6 standard deviation were computed to summarize continuous data. Categorical data were analyzed using the Fisher’s exact test. Some data were missing on the following variables: weight (1), serum bicarbonate (11), serum chloride (9), serum potassium (9), serum sodium (12), and duration of symptoms (4). One-way analysis of variance models, using the Tukey–Kramer method to adjust for multiple comparisons, were employed to test differences in means of each outcome for the race/ethnicity groups and insurance status. A P value less than .05 was considered to be statistically significant. A generalized linear model was performed to assess the impact of risk factors (patient demographics and illness characteristics as defined previously) on the outcomes electrolyte abnormalities, time to operation, and total length of hospital stay. Variables with a P value less than .05 were retained in the final model based on the stepwise model selection procedure. Statistical analyses were performed using SAS, version 9.3 (SAS Institute, Cary, NC).

Results A total of 584 patients with HPS were identified during the study period (Table 1). The mean age and weight were 6.9 6 8.6 weeks and 3.9 6 .8 kg, respectively. There were 339 (58.1%) white, 188 (32.2%) African-American, and 57 (9.8%) Latino infants. There were 199 (34.1%) infants with private insurance, 362 (62.0%) with public insurance, and 23 (3.9%) with no insurance. Of white infants, 142 (41.8%) had private insurance, 190 (56%) had public insurance, and only 7 (2%) were uninsured. In contrast, only 41

A. Feliz et al. Table 1

Health disparities in pyloric stenosis

331

Patient race/ethnicity and insurance status N (%)

Demographic

Private

Public

None

Total

White African-American Latino Total

142 41 16 199

190 136 36 362

7 11 5 23

339 (58.1) 188 (32.2) 57 (9.8) 584

(24.3) (7.0) (2.7) (34.1)

(21.8%) of African-American infants had private insurance, whereas 136 (72.3%) had public insurance and 11 (5.9%) were uninsured. Most Latino infants 36 (63.2%) had public insurance and 5 (8.8%) were uninsured; 16 (28.1%) had private insurance. Of the 584 patients, 528 (90.4%) infants had imaging to assist in the diagnosis. Most infants (n 5 393, 67.3%) underwent laparoscopic pyloromyotomy. Admission to the pediatric intensive care admission was uncommon (n 5 20, 3.4%), and was not statistically different between race/ ethnic groups (P 5 .79) or insurance status (P 5 .16). One infant in the entire group had a wound infection. One infant died at home 3 days after discharge due to cardiopulmonary arrest while sleeping in the same bed as adults.

Bivariate analyses There were no significant differences in mean age at presentation, gender, or mean duration of symptoms between race/ethnicity groups (Table 2). African-American infants were smaller than Latino infants (3.8 6 .83 vs 4.2 6 1.0 kg, P 5 .01) but not statistically smaller than white patients (P 5 .27). African-American infants had significantly lower serum chloride (95.0 6 10.0 vs 98.6 6 8.9 mmol/L, P , .001) and sodium levels (135.3 6 3.6 vs 136.2 6 3 mmol/L, P 5 .006) and presented with higher bicarbonate levels (30.6 6 6.2 vs 28.7 6 5.7 mmol/L, P 5 .001) when compared with white patients. Time from admission to pyloromyotomy was Table 2

(32.5) (23.3) (6.2) (62.0)

(1.2) (1.88) (.9) (3.9)

significantly longer for African-American patients when compared with whites (1.5 6 1.0 vs 1.2 6 .9 days, P , .001). Similarly, total length of hospitalization was longer for African-American infants when compared with white infants (3.1 6 2.5 vs 2.5 6 2.3 days, P 5 .03). When comparing African-American patients to the Latino patients, African-American infants had greater alterations of serum bicarbonate (30.6 6 6.2 vs 28.5 6 5.1 mmol/L, P 5 .04). There were no differences in time to operation or total length of hospitalization for the Latino and AfricanAmerican infants. There were no statistically significant differences in serum electrolyte levels or characteristics of hospitalization between white and Latino infants. There were no significant differences in mean age at presentation, weight, gender, or mean duration of symptoms, for each insurance status comparison (Table 3). Infants with public insurance had lower serum chloride (96.9 6 9.0 vs 99.0 6 8.8 mmol/L, P 5 .03) and higher bicarbonate (29.6 6 5.7 vs 28.3 6 5.2 mmol/L, P 5 .04) levels at admission compared with privately insured infants. Time to operation (1.4 6 1.0 vs 1.1 6 .9 days, P 5 .01) and total length of stay (2.9 6 2.7 vs 2.4 6 1.2 days, P 5 .03) were longer for publicly insured infants compared with those with private insurance. Uninsured patients presented with lower sodium (133.4 6 4.0 vs 135.8 6 3.2 mmol/L, P 5 .002), lower chloride (89.9 6 14.6 vs 96.9 6 9.0 mmol/L, P 5 .001), and higher bicarbonate (33.5 6 10.6 vs 29.6 6 5.7 mmol/L, P 5 .005) levels than publicly insured patients. This difference was exaggerated when comparing the uninsured patients to the

Bivariate analysis by race/ethnicity

Variable Age (weeks) Weight (kg) Gender male, n (%) Duration of symptoms (days) Chloride (mmol/L) Bicarbonate (mmol/L) Sodium (mmol/L) Potassium (mmol/L) Time to operation (days) Length of Stay (days)

White (n 5 339)

African-American (n 5 188)

Latino (n 5 57)

7.1 6 9.4 3.9 6 .8 283 (83.5) 10.9 6 9.9

7.1 6 8.1 3.8 6 .83 146 (77.7) 12.3 6 11.2

5.2 6 2.9 4.2 6 1.0 47 (82.5) 10.1 6 9.2

98.6 28.7 136.2 4.7 1.2 2.5

6 6 6 6 6 6

8.9 5.7 3.0 .8 .9 2.3

95.0 30.6 135.3 4.6 1.5 3.1

6 6 6 6 6 6

10.0 6.2 3.6 .93 1.0 2.5

Data are presented as mean 6 standard deviation unless stated. *P value.

97.7 28.5 136.3 4.6 1.2 2.6

6 6 6 6 6 6

White vs African-American*

7.9 5.1 3.2 1.0 .7 1.2

White vs Latino*

African-American vs Latino*

.99 .27 .25 .28

.28 .09

.33 .01

.85

.32

,.001 .001 .006 .37 ,.001 .028

.77 .98 .93 .56 .82 .98

.12 .04 .07 .99 .18 .35

332

The American Journal of Surgery, Vol 214, No 2, August 2017

Table 3

Bivariate analysis by insurance status

Variable

Private (n 5 199)

Public (n 5 362)

None (n 5 23)

Private vs public*

Age (weeks) Weight (kg) Gender male, n (%) Duration of symptoms (days) Chloride (mmol/L) Bicarbonate (mmol/L) Sodium (mmol/L) Potassium (mmol/L) Time to operation (days) Length of stay (days)

7.3 6 10.5 4.0 6 .7 167 (83.9) 10.9 6 10.0

6.6 6 7.1 3.9 6 .9 294 (81.2) 11.5 6 10.4

8.5 6 10.8 3.9 6 .6 15 (65.2) 11.7 6 9.8

.64 .56 .09 .82

.80 .89

.56 .10

.94

.99

6 6 6 6 6 6

96.9 1 9.0 29.6 6 5.7 135.8 6 3.2 4.6 6 .9 1.4 6 1.0 2.9 6 2.7

89.9 33.5 133.4 4.3 1.6 3.0

6 6 6 6 6 6

.03 .04 .37 .15 .01 .03

,.001 ,.001 ,.001 .08 .05 .38

.001 .005 .002 .32 .43 .95

99 28.3 136.2 4.7 1.1 2.4

8.8 5.2 3.1 .8 .9 1.2

14.6 10.6 4.0 1.2 1.0 2.3

Private vs none*

Public vs none*

Data are presented as mean 6 standard deviation. *P-value.

privately insured patients; uninsured patients presented with lower serum sodium (133.4 6 4.0 vs 136.2 6 3.1 mmol/L, P , .001), lower chloride (89.9 6 14.6 vs 99.0 6 8.8 mmol/L, P , .001), and higher bicarbonate (33.5 6 10.6 vs 28.3 6 5.2 mmol/L, P , .001) levels than privately insured patients. Although the uninsured patients did not have a statistically different time to operation or length of hospitalization when compared with publicly insured infants, they did have a significantly longer time to operation when compared with privately insured infants (1.6 6 1.0 vs 1.1 6 .9 days, P 5 .05). In addition, the uninsured patients had longer total length of stay (3.0 6 2.3 days) than privately insured infants (2.4 6 1.2 days), but the difference was not statistically significant (P 5 .38).

Multivariable analyses Multivariable analyses were performed to assess the impact of each risk factor on the electrolytes at admission, time to operation, and length of hospital stay using White

Table 4

and private insurance as the respective reference groups (Table 4). A generalized linear model was performed to assess the impact of risk factors (patient demographics and illness characteristics as defined previously) on the outcomes electrolyte abnormalities, time to operation, and total length of hospital stay. Variables with a P value less than .05 were retained in the final model based on the stepwise model selection procedure. African-American race, underinsured status, weight, and duration of symptoms were independently associated with chloride and bicarbonate levels at admission and time to operation. Based on the parameter estimates (b), African-American infants were more likely to have lower chloride levels (b 5 22.54, P 5 .002), higher bicarbonate levels (b 5 1.39, P 5 .009), and a longer time to operation (b 5 .25, P 5 .004) when compared with white infants. Electrolyte levels and time to operation among Latino infants were not significantly different from whites. Infants with public insurance were more likely to have a longer time to operation (b 5 .18, P 5 .03) than infants

Multivariable analyses of characteristics Parameter estimates (b)

Variables N Race/ethnicity White African-American Latino Insurance Private Public None Weight (Kg) Duration of Symptoms (days)

Chloride (mmol/L)

Bicarbonate (mmol/L)

Sodium (mmol/L)

Time to operation (days)

Length of stay (days)

570

568

567

578

578

Reference 22.54 (.002) 2.47 (.71)

Reference 1.39 (.009) 2.47 (.57)

—— —— ——

Reference .25 (.004) .11 (.41)

—— —— ——

Reference 21.17 (.14) 27.88 (,.001) 2.15 (,.001) 2.22 (,.001)

Reference .833 (.10) 4.70 (,.001) 21.07 (,.001) .09 (,.001)

Reference 2.26 (.36) 22.63 (,.001) .54 (,.001) 2.07 (,.001)

Reference .18 (.03) .39 (.06) 2.27 (,.001) .01 (.001)

—— —— —— 2.27 (,.001) .01 (.001)

Data are presented as parameter estimate b (P value). White and private insurance are reference groups for race/ethnicity and insurance status, respectively. ‘‘——’’ indicates variables those did not meet criteria to be retained in the final model.

A. Feliz et al.

Health disparities in pyloric stenosis

with private insurance. Infants without insurance were more likely to have lower serum sodium levels (b 5 22.63, P , .001), lower serum chloride levels (b 5 27.88, P , .001), and higher serum bicarbonate levels (b 5 4.7, P , .001) than infants with private insurance. Insurance status, weight, and duration of symptoms were significant predictors of sodium levels at admission.

Comments HPS provides an excellent model for exploring health disparities for a number of reasons. Although administration of erythromycin and azithromycin has been shown to be associated with the development of HPS, there is no known biologic or environmental predisposition to the development of the disease.13–15 In addition, it follows a predictable course, its treatment is standardized (pyloromyotomy), and the severity of illness at presentation can easily be measured by electrolyte disturbances. Furthermore, infants with lower weights and longer preoperative time for fluid resuscitation are known to be at risk for prolonged hospitalization.16 This study demonstrates that social determinants of health, namely race/ethnicity and insurance status, play a significant role in disease severity at the time of presentation to medical care for these infants. Although more common in white infants, a relatively common incidence of pyloric stenosis with differences in presentation in African-American infants has been documented for many years. In 1966, the first published study focusing on African-American infants showed that the incidence of HPS at a single institution that primarily treated African-American patients was .92 per 1,000 live births17; more than half of their infants had a history of low birth weight or prematurity. In a retrospective analysis of infants who underwent pyloromyotomy by a single pediatric surgical group at either a public hospital or a university hospital, Finkelstein et al2 found that the groups of infants differed significantly in racial composition and insurance status. The public hospital had no white infants and only a small percent had private insurance. Comparatively, the private hospital had a group that was mainly white with private insurance. Despite similarities in the medical condition of the groups at presentation, the authors found longer preoperative times and a greater total length of stay for infants at the public hospital compared with a private university teaching facility. Their analysis determined that differences resulted from delayed diagnostic imaging at the public hospital, suggesting a disparity in the availability of health care services between the 2 hospitals. Our study is the first to show that there are health disparities in illness severity in infants with HPS. AfricanAmerican infants and infants without private insurance are at greater risk of presenting to definitive medical care with metabolic derangements. A greater requirement for

333 resuscitation is reflected in longer preoperative length of stay, leading to a longer total length of stay. All the infants are placed on our standard resuscitation protocol, which reduces variance in care and time to operation. The current analysis does not explain the causative factors for differences in electrolyte abnormalities, which lead to differences in preoperative hospital length of stay among the studied racial and insurance status groups. There were no significant differences in the duration of symptoms between our groups of infants although African-American and underinsured infants presented with a greater severity of illness. The duration of symptoms was a self-reported variable and is thus subject to parental interpretation. The retrospective nature of this study precludes the ability to further decipher the causes for this lack of statistical difference. Whether similar reports between groups are due to under-reporting, a failure of symptom recognition, differences in milk or feeding practices, or other factors is difficult to assess. All infants in the study population were cared for by a single pediatric surgical group with a standardized management strategy, thus minimizing the impact of practice variability in management. Over the study period, there was a transition from open to laparoscopic pyloromyotomy; however, once the infants arrived at our hospital they all received equitable care. This study adds to the growing knowledge that social determinants of health, such as race/ethnicity and health insurance status, which can be used as a surrogate for socioeconomic status, play a significant role in placing certain infants at risk for illness. Todd et al18 documented significantly higher rates of hospital admission, inhospital morbidity and mortality, and hospital charges for children with public or no insurance when compared with children with private health insurance. The authors theorize that these findings are the result of decreased access to medical care. Healthy People 20201 describes geographic location as a source of health disparities. Using sequential spatial analyses to investigate child health status and mapping, Goldhagen 19 found that regional residence in the United States was a powerful predictor of disparate child health outcomes. Living in the southern region of the United States was the best predictor of poor health outcomes for children. The percent of African-Americans, the undereducated and underemployed in the population were also predictors of child poor health outcomes. Eudy20 highlights the pervasive nature of health disparities in a longitudinal secondary data analysis in 2008. Model testing of descriptive statistics over a 23-year period demonstrated that an increased percentage of AfricanAmerican population and increased prevalence of poverty resulted in persistently increased infant mortality rates in the Lower Mississippi Delta despite improvements in regional disparities due to public health initiatives. Our tertiary free-standing children’s hospital is situated in the Mississippi Delta region and our data reinforce the

334 existence of underlying disparities in the health of minority, Medicaid-insured, and uninsured infants and the need for continued efforts to address this public health issue. The Affordable Care Act offers great opportunity to improve the health of populations at risk and reduce the burden of health disparities. The benefits for children at risk are provided by an expansion of Medicaid to a national eligibility floor of 138% federal poverty level ($26,344 for a family of 3 in 2012).21 These expansions should help reduce variability in access to health coverage and improve coverage for low- and moderate-income populations. These expansions are particularly significant for people of color, who make up a disproportionate share of the uninsured and of low-income populations. Unfortunately, families in our region are presently not able to take advantage of these programs.22 The states of Tennessee and Mississippi have not adopted the Medicaid expansion. Arkansas did not adopt the Medicaid expansion but has approved Section 1,115 waivers, which allow for poor families to apply for benefits in the health care exchanges. Low-income families in this region have not been able to benefit from expanded access to care provided by the Affordable Care Act. As a retrospective single institution study, the presented data do have their limitations. Hospital medical records are not designed, this particular research and available information allows for some missing data. Our institution is the only children’s hospital in a region, ie, geographically unique in the United States. We have a significantly higher proportion of African-American infants in our patient population (32.2%) than the national average, but this is proportional to our local population. According to the Shelby County, Tennessee demographics in the United States Census 201023 African-Americans comprise 52.1%, whereas whites comprise 40.6% and Latinos 5.2% of the population. Our patient demographics and findings may be unique to our region, but they highlight the existence of regional disparities in the United States. Despite this study’s limitations, our findings highlight the importance of recognizing that, although national data may suggest a broad trend toward equivalent care for a given condition, the unique characteristics of any individual institution or system may highlight a region with particular needs.

Conclusion This is the first study to show that health disparities exist in infants with HPS. African-American and infants without private insurance are at greater risk of presenting to definitive medical care later in their illness course and with metabolic derangements; this requires more time to correct dehydration and electrolyte derangements, resulting in longer hospitalizations. From these results, we surmise that efforts should be made in educating minorities and those without private insurance on the normal symptoms of

The American Journal of Surgery, Vol 214, No 2, August 2017 reflux and when to seek additional medical attention; furthermore, educational efforts should be geared toward primary care physicians to increase earlier diagnosis of these patients. In summary, the present study provides further support that significant health disparities exist for minority infants and infants without private insurance infants. Future investigation on the causative factors contributing to severity of illness would provide opportunities to improve the health status for at-risk infants.

References 1. Healthy People 2020. Available at: http://www.healthypeople.gov/ 2020/about/default.aspx. Accessed June 26, 2014. 2. Finkelstein JB, Stamell EF, Zilbert NR, et al. Management and outcomes for children with pyloric stenosis stratified by hospital type. J Surg Res 2010;158:6–9. 3. Kelley-Quon LI, Tseng CH, Jen HC, et al. Hospital type predicts surgical complications for infants with hypertrophic pyloric stenosis. Am Surg 2012;78:1079–82. 4. Guagliardo MF, Teach SJ, Huang ZJ, et al. Racial and ethnic disparities in pediatric appendicitis rupture rate. Acad Emerg Med 2003; 10:1218–27. 5. 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. 6. To T, Wajja A, Wales PW, et al. Population demographic indicators associated with incidence of pyloric stenosis. Arch Pediatr Adolesc Med 2005;159:520–5. 7. Habbick BF, To T. Incidence of infantile hypertrophic pyloric stenosis in Saskatchewan, 1970-85. CMAJ 1989;140:395–8. 8. Hedback G, Abrahamsson K, Husberg B, et al. The epidemiology of infantile hypertrophic pyloric stenosis in Sweden 1987-96. Arch Dis Child 2001;85:379–81. 9. Schechter R, Torfs CP, Bateson TF. The epidemiology of infantile hypertrophic pyloric stenosis. Paediatr Perinat Epidemiol 1997;11: 407–27. 10. Mitchell LE, Risch N. The genetics of infantile hypertrophic pyloric stenosis. A reanalysis. Am J Dis Child 1993;147:1203–11. 11. Wang J, Waller DK, Hwang LY, et al. Prevalence of infantile hypertrophic pyloric stenosis in Texas, 1999-2002. Birth Defects Res A Clin Mol Teratol 2008;82:763–7. 12. Carpenter RO, Schaffer RL, Maeso CE, et al. Postoperative ad lib feeding for hypertrophic pyloric stenosis. J Pediatr Surg 1999;34: 959–61. 13. Eberly MD, Eide MB, Thompson JL, et al. Azithromycin in early infancy and pyloric stenosis. Pediatrics 2015;135:483–8. 14. Friedman DS, Curtis CR, Schauer SL, et al. Surveillance for transmission and antibiotic adverse events among neonates and adults exposed to a healthcare worker with pertussis. Infect Control Hosp Epidemiol 2004;25:967–73. 15. Honein MA, Paulozzi LJ, Himelright IM, et al. Infantile hypertrophic pyloric stenosis after pertussis prophylaxis with erythromcyin: a case review and cohort study. Lancet 1999;354:2101–5. 16. Lee SL, Stark R. Can patient factors predict early discharge after pyloromyotomy? Perm J 2011;15:44–6. 17. Hara S, Crump EP, Parker WG. Pyloric stenosis. The incidence of infantile hypertrophic pyloric stenosis in the Negro infant. J Natl Med Assoc 1966;58:250–3. 18. Todd J, Armon C, Griggs A, et al. Increased rates of morbidity, mortality, and charges for hospitalized children with public or no health insurance as compared with children with private insurance in Colorado and the United States. Pediatrics 2006;118: 577–85.

A. Feliz et al.

Health disparities in pyloric stenosis

19. Goldhagen J, Remo R, Bryant 3rd T, et al. The health status of southern children: a neglected regional disparity. Pediatrics 2005;116:e746–53. 20. Eudy RL. Infant mortality in the Lower Mississippi Delta: geography, poverty and race. Matern Child Health J 2009;13:806–13. 21. Kaiser Family Foundation. Disparities in Health and Health Care: Five Key Questions and Answers. Available at: http://kff.org/disparities-policy/factsheet/key-facts-on-health-coverage-for-low/. Accessed September 1, 2015.

335 22. Kaiser Family Foundation. Status of State Action on the Medicaid Expansion Decision. Available at: http://kff.org/health-reform/ state-indicator/state-activity-around-expanding-medicaid-underthe-affordable-care-act/. Accessed December 1, 2015. 23. Fact Finder Census Shelby County, Tennessee. Available at: http:// factfinder2.census.gov/faces/tableservices/jsf/pages/productview. xhtml?src5bkmk. Accessed June 26, 2014.