Journal of Pediatric Surgery 49 (2014) 777–781
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Determinants of appendicitis outcomes in Canadian children Li Hsia Alicia Cheong, Sherif Emil ⁎ Division of Pediatric General and Thoracic Surgery, The Montreal Children's Hospital, McGill University Health Centre
a r t i c l e
i n f o
Article history: Received 9 February 2014 Accepted 13 February 2014 Key words: Appendicitis Outcomes Determinants Children Pediatric Canada
a b s t r a c t Background: Outcomes of appendicitis may be influenced by access to healthcare. We investigated the determinants of pediatric appendicitis outcomes in the single-payer Canadian healthcare system. Methods: Children coded for urgent appendectomy by the Canadian Institute of Health Information during the period 2004–2010 were analyzed. Misdiagnosis rate, perforated appendicitis rate, and hospital stay were the outcomes studied. Analyzed variables included age, gender, domicile, socioeconomic status, surgeon’s specialty, hospital type, region, and operative approach. Logistic regression analysis was used to examine associations, and a quintile regression model examined the effect on median hospital stay. Results: 41,702 patients were studied. A higher rate of perforated appendicitis was associated with lower age [OR 2.66], male gender [OR 1.18], pediatric surgeon [OR 1.25], and treatment outside the Maritimes. A higher rate of misdiagnosis was associated with lower age [OR 1.53], female gender [OR 2.29], non-children’s hospital [OR 1.33], and western Canada [OR 1.22]. A significantly longer hospital stay was associated with open appendectomy, pediatric surgeon, and the Territories for simple appendicitis, and open appendectomy, pediatric surgeon, children’s hospital, and the Maritimes for perforated appendicitis. Conclusions: In Canada, outcomes of pediatric appendicitis are associated with regional and treatment-level factors. Rural domicile and socioeconomic status do not affect outcomes. © 2014 Elsevier Inc. All rights reserved.
Appendicitis is the most common pediatric surgical emergency, and appendectomy is the most common urgent pediatric surgical operation. The post-operative outcomes of appendicitis are largely dependent on the severity of the disease at presentation. Perforated appendicitis is associated with prolonged hospital stay and higher rates of postoperative complications [1,2]. In addition, perforation and misdiagnosis rates in appendicitis are considered outcomes that reflect access to appropriate level healthcare. Previous studies from the United States have shown that patient-level factors, such as socioeconomic status, insurance status, and race, influence the outcomes of children with appendicitis [3,4]. The effects of race and socioeconomic status vanish when equal access to care is provided [5,6]. In Canada, all children are insured under a single-payer system which theoretically provides equal access. In a recent comparison of pediatric appendicitis outcomes between the United States and Canada, we showed that the outcomes of Canadian children generally fell in between US children with public, and those with private, insurance [7]. In the present study, we analyzed a large national Canadian database to elucidate what determinants, within the context of a universal health care system, influence the outcomes of pediatric appendicitis.
⁎ Corresponding author at: Division of Pediatric General and Thoracic Surgery, The Montreal Children's Hospital, 2300 Tupper, C-818, Montreal, Quebec H3H 1P3, Canada. Tel.:+1 514 412 4497; fax: +1 514 412 4289. E-mail address: [email protected] (S. Emil). http://dx.doi.org/10.1016/j.jpedsurg.2014.02.074 0022-3468/© 2014 Elsevier Inc. All rights reserved.
1. Methods 1.1. Data source Canadian data for the years 2004–2010 were obtained from the Discharge Abstract Database (DAD), managed by the Canadian Institute for Health Information (CIHI). DAD contains mandatory discharge data from all hospitals in Canada, except those in Quebec. Data were coded using ICD-10-CA. The database includes diagnoses, patient demographics, treating physician information, and hospital characteristics. The study was approved by the McGill University Faculty of Medicine Institutional Review Board (A06-E53-13A).
1.2. Study population All patients less than 18 years old with a valid procedure code for appendectomy were candidates for the study population. Patients had to have one of the following three procedure codes: (i) Appendectomy (1.NV.89.DA, 1.NV.89.LA), (ii) Drainage of appendiceal abscess (1.NV.52) AND secondary code of (i), or (iii) Abdominal or pelvic drainage (1.OT.52, 1.SQ.52) AND secondary code of (i) or (ii). Patients with no primary or secondary procedure code of appendectomy, patients who had incidental appendectomy, and patients with an elective admission, or unrelated primary diagnostic code were excluded. Demographic data gathered for all patients included age, gender, patient’s domicile, socioeconomic status, and geographic
778
L.H.A. Cheong, S. Emil / Journal of Pediatric Surgery 49 (2014) 777–781
region (excluding Quebec). Data on surgical approach, treating surgeon’s specialty, and type of treating hospital were also obtained. 1.3. Outcomes Three main outcomes of interest were analyzed: normal appendix rate, perforated appendicitis rate, and the total length of hospital stay. Normal appendix was defined as an appendectomy procedure code in combination with a diagnostic code other than appendicitis. Perforated appendicitis was defined as a procedure code of appendectomy and a primary diagnostic code of acute appendicitis with peritonitis or abscess. Hospital stay was recorded in days. There was no reliable information on post-operative complications such as abscess, phlegmon, wound infection, or readmission. Secondary procedures (e.g. abscess drainage) were recorded without specification of whether they occurred at the same time or subsequent to the primary procedure. Eight potential determinants of outcomes were studied: gender, age group (≤ 5, 6–11 and 12–17 years), socioeconomic status (5 categories from lowest to highest using median income for postal code), domicile (rural vs urban), Canadian region (Maritimes, Ontario, Western Canada, Territories), hospital type (children’s versus nonchildren’s), surgeon’s specialty (pediatric versus general), and operative approach (laparoscopic versus open). 1.4. Statistical analysis Database files were downloaded to an SPSS 20.0 for Windows file (SPSS, Chicago, IL). A multivariate analysis of the eight potential outcome determinants was performed. A Chi-square test was used for comparing nominal responses, a Cochran–Armitage Trend Test was used to test for trend when the feature was ordinal (e.g., age category), and the Wilcoxon Rank Sum Test was used for numerical responses. All analyses were carried out using a two-sided test at an alpha level of 5% unless otherwise specified. No formal adjustments were made for the multiplicity of inferences for multiple outcomes. Statistical analyses including descriptive statistics and logistic regression were performed with SAS (Version 9.2, SAS Institute, Cary, NC, USA). Quantile regression was used in the analysis of length of hospital stay to account for the leftward skew in this outcome. 2. Results During the study period, 47,840 patients were considered potential subjects for the study. Of those, 41,405 met the inclusion criteria (Fig. 1). Age, gender, region, and hospital type were available for all patients. Several subsets of patients were removed from specific analyses: 5.1% received treatment from neither a pediatric nor a general surgeon; 16.2% did not have a clearly identified domicile status; 1.9% had a non-valid socioeconomic status. The demographics of the study population are shown in Table 1. In the entire cohort, the rates of normal appendix, non-perforated appendicitis, and perforated appendicitis were 6.2%, 66.6%, and 27.2%, respectively. Table 2 outlines the distribution of the various patient-level and treatmentlevel factors studied by category of appendicitis. Mmultivariate logistic regression analysis of determinants of the misdiagnosis rate (Table 3) showed that children were more likely to be misdiagnosed if they were females (OR 2.33, p b 0.0001), in the youngest age group (OR 1.51, p b 0.0001), received treatment at nonchildren’s hospitals (OR 1.42, p = 0.01), or were from western Canada (OR 1.21, p b 0.02). Domicile, socioeconomic status, and surgeon’s specialty did not significantly influence the likelihood of misdiagnosis. Multivariate logistic regression analysis of determinants of the perforation rate (Table 4) showed that children were more likely to have perforated appendicitis if they were males (OR 1.09, p b 0.0001), younger [(≤5 years: OR 2.91, p b 0.0001), (6–11: OR 1.19, p b .0001)], resided outside maritime Canada, and treated by pediatric surgeons
Fig. 1. Flowchart outlining how study cohort was chosen.
(OR 1.25, p = 0.0003). Domicile, socioeconomic status, and type of treating hospital did not significantly influence the likelihood of perforated appendicitis.
Table 1 Characteristics of study population. Gender Male Female Age (years) 0–5 6–11 12–17 Geographic region Ontario West Maritime Territories Socioeconomic Status Lowest 2nd lowest Middle 2nd highest Highest Domicile Rural Urban Not available Type of treating hospital Non-children's Children's Specialty of treating surgeon General surgeon Pediatric surgeon Other Surgical appendectomy approach Open Laparoscopic Values reported in %.
59.0 41.0 4.3 35.2 60.5 47.6 43.0 8.7 0.7 18.9 18.5 20.0 20.1 20.5 13.7 70.1 16.2 75.0 25.0 72.5 22.4 5.1 52.7 47.3
L.H.A. Cheong, S. Emil / Journal of Pediatric Surgery 49 (2014) 777–781 Table 2 Distribution by type of appendicitis. Patient- & Hospital-level Factors Gender Male Female Age (%) 0–5 years old 6–11 years old 12–17 years old Type of treating hospital Non-children's Children's Type of treating surgeon Pediatric surgeon General surgeon Surgical approach Open Laparoscopic Geographic region Ontario West Maritime Territories Domicile Rural Urban Socioeconomic status Lowest 2nd lowest Middle 2nd highest Highest
Normal appendicitis
Acute appendicitis
Perforated appendicitis
1176 (4.8) 1780 (10.4)
16452 (67.0) 11080 (64.6)
6911 (28.2) 4303 (25.1)
177 (10.0) 754 (5.2) 1663 (6.6)
757 (42.8) 9580 (65.7) 17225 (68.7)
836 (47.2) 4243 (29.1) 6170 (24.6)
2043 (6.6) 551 (5.3)
20999 (67.6) 6563 (63.3)
8001 (25.8) 3248 (31.3)
476 (5.1) 1841 (6.1)
5805 (62.7) 20447 (68.1)
2974 (32.1) 7720 (25.7)
1266 (5.8) 1328 (6.8)
14306 (65.5) 13256 (67.7)
6255 (28.7) 4994 (25.6)
1062 (5.4) 1282 (7.2) 220 (6.1) 30 (10.0)
13570 11174 2640 178
(68.9) (62.8) (73.1) (59.1)
5060 (25.7) 5345 (30.0) 751 (20.8) 93 (31.0)
351 (6.2) 1839 (6.3)
3747 (66.2) 19323 (66.6)
1562 (27.6) 7869 (27.1)
464 516 513 478 570
(5.9) (6.7) (6.2) (5.7) (6.7)
5213 5104 5560 5531 5621
(66.6) (66.6) (67.2) (66.3) (66.2)
2149 (27.5) 2046 (26.7) 2205 (26.6) 2337 (28.0) 2295 (27.0)
Values reported as n (%).
Multivariate quantile regression analysis indicated that, compared to those with non-perforated appendicitis, median hospital stay was longer in children with perforated appendicitis (4 vs 2 days). In those with non-perforated appendicitis, longer hospital stays were found in children who had open surgery (estimate + 1 day, p b 0.0001) and were treated by pediatric surgeons (estimate + 1 day, p b 0.0001). Children from Ontario, western Canada and the Maritimes had shorter hospital stays (estimates − 1 day, p b 0.0001), versus those in the Territories. Among children with perforated appendicitis, longer hospital stays were found in children who had open surgery (estimate + 1 day, p b 0.0001) and were treated by pediatric surgeons (estimate +1 day, p b 0.0001). Children from the Maritimes had
779
slightly longer hospital stays (estimate 0.19 day, p b 0.0001). A shorter hospital stay was found for children treated at non-children’s hospital (estimate − 1 day, p b 0.0001), and those from Ontario (estimate − 0.81 day, p b 0.0001) and western Canada (estimate − 0.81 day, p b 0.0001). The age of children did affect length of hospital stay in nonperforated and perforated appendicitis but only to a negligible extent (estimate 0, p b 0.0001 in both instances). 3. Discussion Acute appendicitis is the most common surgical illness in children, and appendectomy is one of the most common surgical procedures performed in this age group. Any improvement in outcomes, even if small, may therefore have a profound effect on health care resource utilization and cost. An understanding of the determinants of these outcomes is the first step towards any attempts at improving them. Most of the population-based studies concerning pediatric appendicitis have come from the United States, a multi-payer health care system with inherent disparities in health care access due to differences in race, socioeconomic status, or insurance status. These disparities have correlated with outcomes of pediatric appendicitis, as well as multiple other diseases [3,4,8–10]. In a plethora of US studies looking at medical and surgical disease outcomes, disparities have often been identified [11,12]. For example, a recent study of adults admitted for acute cholangitis demonstrated that non-Caucasian patients, as well as patient insured by US Medicare or Medicaid, were more likely to die, experience a prolonged hospital stay, and incur higher hospital charges [8]. Canada has a single-payer health care system that covers all citizens and legal residents. The literature on disparities in health care in Canada has not been as extensive as that of the US, and has yielded a more mixed picture. For example, a study comparing access to breast cancer care in Ontario and California showed essentially no Canadian disparities based on socioeconomic status, while definite disparities existed in California [13]. A recent review of the Canadian literature on access to cancer care identified income as having a strong influence on screening, while age and geography were the most influential predictors of access to treatment and end-of-life care [14]. There are very few similar studies pertaining to pediatric surgical outcomes. Bratu et al. analyzed twenty years of data on pediatric appendicitis in Manitoba, and found that lower rural or urban socioeconomic status and northern area of residence were significantly associated with a
Table 3 Multivariate logistic regression: rate of misdiagnosis. Variables Gender Age (%)
Type of treating hospital Type of treating surgeon Geographic region
Domicile Socioeconomic status
Male Female (reference) 0–5 years old 6–11 years old 12–17 years old (reference) Non-children's Children's (reference) Pediatric surgeon General surgeon (reference) Ontario West Territories Maritime (reference) Rural Urban (reference) Lowest 2nd lowest Middle 2nd highest Highest (reference)
OR
95% CI
P
0.43 (−) 1.51 0.79 (−) 1.42 (−) 1.07 (−) 0.91 1.21 1.51 (−) 1.02 (−) 0.81 0.95 0.91 0.82 (−)
0.39–0.48 (−) 1.2–1.84 0.71–0.87 (−) 1.13–1.79 (−) 0.84–1.36 (−) 0.77–1.09 1.02–1.44 0.93–2.46 (−) 0.90–1.16 (−) 0.70–0.94 0.82–1.09 0.79–1.04 0.71–0.94 (−)
b0.0001 b0.0001 b0.0001 0.003 0.57 0.30 0.03 0.10 0.72 .06 0.23 0.78 0.06
780
L.H.A. Cheong, S. Emil / Journal of Pediatric Surgery 49 (2014) 777–781
Table 4 Multivariate logistic regression: rate of perforated appendicitis. Variables Gender Age (%)
Type of treating hospital Type of treating surgeon Geographic region
Domicile Socioeconomic status
Male Female (reference) 0–5 years old 6–11 years old 12–17 years old (reference) Non-children's Children's (reference) Pediatric surgeon General surgeon (reference) Ontario West Territories Maritime (reference) Rural Urban (reference) Lowest 2nd lowest Middle 2nd highest Highest (reference)
OR
(95% Wald CI)
1.09 (−) 2.91 1.19 (−) 1.04 (−) 1.25 (−) 1.31 1.70 1.82 (−) 1.04 (−) 1.02 1.01 0.98 1.04 (−)
1.04–1.15 (−) 2.59–3.27 1.13–1.25 (−) 0.92–1.17 (−) 1.11–1.41 (−) 1.18–1.45 1.52–1.88 1.35–2.47 (−) 0.97–1.10 (−) 0.94–1.11 0.94–1.10 0.91–1.06 0.96–1.12 (−)
P 0.0006 b0.0001 b0.0001 0.52 0.0003 b0.0001 b0.0001 b0.0001 0.27 0.65 0.89 0.23 0.31
higher perforation rate [15]. In population-based studies in the province of Ontario, Langer et al. showed lower duodenal perforation rates after pyloromyotomy and a lower appendicitis misdiagnosis rate, when children were cared for by pediatric surgeons [16,17]. We recently reported results of a population-based comparison of children with appendicitis in the United States and Canada, looking at some of the same outcomes examined here [7]. In general, the outcomes of Canadian children fell in between US children with private, and those with public or no insurance. The exception was the youngest age group, where Canadian children had better outcomes versus US children in all insurance categories [7]. We conducted the present study to further investigate possible outcome determinants in Canadian children. We obtained all the information from the Canadian Institute of Health Information that may potentially influence outcomes. In addition to investigating possible determinants of misdiagnosis and perforation rates, we also investigated length of stay, adding the surgical approach as a potential factor. Our findings reinforced some well-known and previously reported determinants of appendicitis outcomes. Females and the youngest patients were more likely to be misdiagnosed. Perforated patients were more likely to be in the youngest age group and be treated by pediatric surgeons [16,18]. However, Langer’s findings in Ontario of a higher rate of misdiagnosis by general surgeons, were not reproduced in this larger study from most parts of the country [16]. Rather, a higher risk of misdiagnosis was found if the child was treated at a non-children’s hospital. This may implicate the overall experience of the institution and other physicians involved in the diagnosis of appendicitis, such as emergency physicians and radiologists, rather than the treating surgeon. The slightly higher rates of misdiagnosis and perforation in western Canada may relate to greater distances between patients and children’s hospitals or pediatric surgical centres in the western provinces. The lack of association of misdiagnosis and perforation with domicile and income in our study is perhaps the most significant finding. It lends strong support to the notion that a single-payer system provides better access, and negates the deleterious effects of rural residence or lowsocioeconomic status on outcomes. We were surprised to find that a slight majority of children were still receiving open appendectomies, despite reported advantages of the laparoscopic approach [19–21]. This was associated with a longer hospital stay both for simple and perforated appendicitis in multivariate analysis, and certainly presents a potential area for improvement. Hospital stay was also longer for children with the entire spectrum of appendicitis treated by pediatric surgeons. This was previously reported in Ontario, and was difficult to explain since
this longer hospital stay was not associated with increases in complications or re-admissions. Pediatric surgeons may keep patients in the hospital until strict discharge criteria are met. Clinical pathways for appendicitis treatment have been shown to slightly prolong hospital stay, while decreasing complications [1]. Our study is limited by issues common to this type of outcomes research. Accuracy of the data is determined by coders in each hospital who extract the data retrospectively from medical charts. While standardized coding minimizes discrepancy and variability in data collection and reporting, errors can occur at any point from collection to submission and distribution. Data were not included from Quebec, Canada’s second largest province, representing approximately 25% of the nation’s population. There is more than one regulatory body that oversees healthcare data from Quebec, and approval from the Commission d'accès à l'information du Québec (CAI) is necessary before data can be released. Given the complexity of obtaining uniform information from more than one regulatory body in Quebec, information for children in that province was not obtained. Data regarding other important appendicitis outcomes, such as wound infections, intra-abdominal infections, need for additional procedures, and re-admissions were not adequately provided in the discharge abstract database. Despite these limitations, we have reported the first large Canadian population-based study investigating the outcome determinants in children with appendicitis. Within the Canadian single-payer, universal health care system, socioeconomic status and rural domicile do not affect outcomes or hospital stay. Nevertheless, opportunities for outcomes improvement exist. These include increasing the rate of laparoscopic appendectomies, exploring ways to decrease misdiagnosis rates in non-children’s hospitals, and investigating clinical pathways that may shorten hospital stay without increasing the complication rate. Acknowledgment We would like to thank Dr. Xianming Tan of the Biostatistics Core Facility, McGill University Health Centre Research Institute for providing assistance with the statistical analyses. References [1] Emil S, Taylor M, Ndiforchu F, et al. What are the true advantages of a pediatric appendicitis clinical pathway? Am Surg 2006;72:885–9. [2] Emil S, Laberge JM, Mikhail P, et al. Appendicitis in children: a ten-year update of therapeutic recommendations. J Pediatr Surg 2003;38:236–42. [3] 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. [4] 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. [5] Lee SL, Shekherdimian S, Chiu VY. Effect of race and socioeconomic status in the treatment of appendicitis in patients with equal health care access. Arch Surg 2011;146:156–61. [6] Lee SL, Shekherdimian S, Chiu VY, et al. Perforated appendicitis in children: equal access to care eliminates racial and socioeconomic disparities. J Pediatr Surg 2010;45:1203–7. [7] Cheong LHA, Emil S. Outcomes of pediatric appendicitis: an international comparison of the United States and Canada. JAMA Surg 2014;149:50-55. [8] McNabb-Baltar J, Trinh QD, Barkun AN. Disparities in outcomes following admission for cholangitis. PloS One 2013;8:e59487. [9] Nathan H, Frederick W, Choti MA, et al. Racial disparity in surgical mortality after major hepatectomy. J Am Coll Surg 2008;207:312–9. [10] Anger JT, Rodriguez LV, Wang Q, et al. Racial disparities in the surgical management of stress incontinence among female Medicare beneficiaries. J Urol 2007;177:1846–50. [11] Haider AH, Scott VK, Rehman KA, et al. Racial disparities in surgical care and outcomes in the United States: a comprehensive review of patient, provider, and systemic factors. J Am Coll Surg 2013;216:482–92. [12] Morris AM, Rhoads KF, Stain SC, et al. Understanding racial disparities in cancer treatment and outcomes. J Am Coll Surg 2010;211:105–13. [13] Gorey KM, Luginaah IN, Holowaty EJ, et al. Wait times for surgical and adjuvant radiation treatment of breast cancer in Canada and the United
L.H.A. Cheong, S. Emil / Journal of Pediatric Surgery 49 (2014) 777–781
[14] [15] [16] [17]
States: greater socioeconomic inequity in America. Clin Invest Med 2009;32: E239–49. Maddison AR, Asada Y, Urquhart R. Inequity in access to cancer care: a review of the Canadian literature. Cancer Causes Control 2011;22:359–66. Bratu I, Martens PJ, Leslie WD, et al. Pediatric appendicitis rupture rate: disparities despite universal health care. J Pediatr Surg 2008;43:1964–9. Somme S, To T, Langer JC. Effect of subspecialty training on outcome after pediatric appendectomy. J Pediatr Surg 2007;42:221–6. Langer JC, To T. Does pediatric surgical specialty training affect outcome after Ramstedt pyloromyotomy? A population-based study. Pediatrics 2004;113:1342–7.
781
[18] Emil SG, Taylor MB. Appendicitis in children treated by pediatric versus general surgeons. J Am Coll Surg 2007;204:34–9. [19] Esposito C, Calvo Al, Castagnetti M, et al. Open versus laparoscopic appendectomy in the pediatric population: a literature review and analysis of complications. J Laparoendosc Adv Surg Tech A 2012;22:834–9. [20] Masoomi H, Mills S, Dolich MO, et al. Comparison of outcomes of laparoscopic versus open appendectomy in children: data from the Nationwide Inpatient Sample (NIS), 2006–2008. World J Surg 2012:573–8. [21] Aziz O, Athanasiou T, Tekkis PP, et al. Laparoscopic versus open appendectomy in children: a meta-analysis. Ann Surg 2006;243:17–27.
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Determinants of appendicitis outcomes in Canadian children Li Hsia Alicia Cheong, Sherif Emil ⁎ Division of Pediatric General and Thoracic Surgery, The Montreal Children's Hospital, McGill University Health Centre
a r t i c l e
i n f o
Article history: Received 9 February 2014 Accepted 13 February 2014 Key words: Appendicitis Outcomes Determinants Children Pediatric Canada
a b s t r a c t Background: Outcomes of appendicitis may be influenced by access to healthcare. We investigated the determinants of pediatric appendicitis outcomes in the single-payer Canadian healthcare system. Methods: Children coded for urgent appendectomy by the Canadian Institute of Health Information during the period 2004–2010 were analyzed. Misdiagnosis rate, perforated appendicitis rate, and hospital stay were the outcomes studied. Analyzed variables included age, gender, domicile, socioeconomic status, surgeon’s specialty, hospital type, region, and operative approach. Logistic regression analysis was used to examine associations, and a quintile regression model examined the effect on median hospital stay. Results: 41,702 patients were studied. A higher rate of perforated appendicitis was associated with lower age [OR 2.66], male gender [OR 1.18], pediatric surgeon [OR 1.25], and treatment outside the Maritimes. A higher rate of misdiagnosis was associated with lower age [OR 1.53], female gender [OR 2.29], non-children’s hospital [OR 1.33], and western Canada [OR 1.22]. A significantly longer hospital stay was associated with open appendectomy, pediatric surgeon, and the Territories for simple appendicitis, and open appendectomy, pediatric surgeon, children’s hospital, and the Maritimes for perforated appendicitis. Conclusions: In Canada, outcomes of pediatric appendicitis are associated with regional and treatment-level factors. Rural domicile and socioeconomic status do not affect outcomes. © 2014 Elsevier Inc. All rights reserved.
Appendicitis is the most common pediatric surgical emergency, and appendectomy is the most common urgent pediatric surgical operation. The post-operative outcomes of appendicitis are largely dependent on the severity of the disease at presentation. Perforated appendicitis is associated with prolonged hospital stay and higher rates of postoperative complications [1,2]. In addition, perforation and misdiagnosis rates in appendicitis are considered outcomes that reflect access to appropriate level healthcare. Previous studies from the United States have shown that patient-level factors, such as socioeconomic status, insurance status, and race, influence the outcomes of children with appendicitis [3,4]. The effects of race and socioeconomic status vanish when equal access to care is provided [5,6]. In Canada, all children are insured under a single-payer system which theoretically provides equal access. In a recent comparison of pediatric appendicitis outcomes between the United States and Canada, we showed that the outcomes of Canadian children generally fell in between US children with public, and those with private, insurance [7]. In the present study, we analyzed a large national Canadian database to elucidate what determinants, within the context of a universal health care system, influence the outcomes of pediatric appendicitis.
⁎ Corresponding author at: Division of Pediatric General and Thoracic Surgery, The Montreal Children's Hospital, 2300 Tupper, C-818, Montreal, Quebec H3H 1P3, Canada. Tel.:+1 514 412 4497; fax: +1 514 412 4289. E-mail address: [email protected] (S. Emil). http://dx.doi.org/10.1016/j.jpedsurg.2014.02.074 0022-3468/© 2014 Elsevier Inc. All rights reserved.
1. Methods 1.1. Data source Canadian data for the years 2004–2010 were obtained from the Discharge Abstract Database (DAD), managed by the Canadian Institute for Health Information (CIHI). DAD contains mandatory discharge data from all hospitals in Canada, except those in Quebec. Data were coded using ICD-10-CA. The database includes diagnoses, patient demographics, treating physician information, and hospital characteristics. The study was approved by the McGill University Faculty of Medicine Institutional Review Board (A06-E53-13A).
1.2. Study population All patients less than 18 years old with a valid procedure code for appendectomy were candidates for the study population. Patients had to have one of the following three procedure codes: (i) Appendectomy (1.NV.89.DA, 1.NV.89.LA), (ii) Drainage of appendiceal abscess (1.NV.52) AND secondary code of (i), or (iii) Abdominal or pelvic drainage (1.OT.52, 1.SQ.52) AND secondary code of (i) or (ii). Patients with no primary or secondary procedure code of appendectomy, patients who had incidental appendectomy, and patients with an elective admission, or unrelated primary diagnostic code were excluded. Demographic data gathered for all patients included age, gender, patient’s domicile, socioeconomic status, and geographic
778
L.H.A. Cheong, S. Emil / Journal of Pediatric Surgery 49 (2014) 777–781
region (excluding Quebec). Data on surgical approach, treating surgeon’s specialty, and type of treating hospital were also obtained. 1.3. Outcomes Three main outcomes of interest were analyzed: normal appendix rate, perforated appendicitis rate, and the total length of hospital stay. Normal appendix was defined as an appendectomy procedure code in combination with a diagnostic code other than appendicitis. Perforated appendicitis was defined as a procedure code of appendectomy and a primary diagnostic code of acute appendicitis with peritonitis or abscess. Hospital stay was recorded in days. There was no reliable information on post-operative complications such as abscess, phlegmon, wound infection, or readmission. Secondary procedures (e.g. abscess drainage) were recorded without specification of whether they occurred at the same time or subsequent to the primary procedure. Eight potential determinants of outcomes were studied: gender, age group (≤ 5, 6–11 and 12–17 years), socioeconomic status (5 categories from lowest to highest using median income for postal code), domicile (rural vs urban), Canadian region (Maritimes, Ontario, Western Canada, Territories), hospital type (children’s versus nonchildren’s), surgeon’s specialty (pediatric versus general), and operative approach (laparoscopic versus open). 1.4. Statistical analysis Database files were downloaded to an SPSS 20.0 for Windows file (SPSS, Chicago, IL). A multivariate analysis of the eight potential outcome determinants was performed. A Chi-square test was used for comparing nominal responses, a Cochran–Armitage Trend Test was used to test for trend when the feature was ordinal (e.g., age category), and the Wilcoxon Rank Sum Test was used for numerical responses. All analyses were carried out using a two-sided test at an alpha level of 5% unless otherwise specified. No formal adjustments were made for the multiplicity of inferences for multiple outcomes. Statistical analyses including descriptive statistics and logistic regression were performed with SAS (Version 9.2, SAS Institute, Cary, NC, USA). Quantile regression was used in the analysis of length of hospital stay to account for the leftward skew in this outcome. 2. Results During the study period, 47,840 patients were considered potential subjects for the study. Of those, 41,405 met the inclusion criteria (Fig. 1). Age, gender, region, and hospital type were available for all patients. Several subsets of patients were removed from specific analyses: 5.1% received treatment from neither a pediatric nor a general surgeon; 16.2% did not have a clearly identified domicile status; 1.9% had a non-valid socioeconomic status. The demographics of the study population are shown in Table 1. In the entire cohort, the rates of normal appendix, non-perforated appendicitis, and perforated appendicitis were 6.2%, 66.6%, and 27.2%, respectively. Table 2 outlines the distribution of the various patient-level and treatmentlevel factors studied by category of appendicitis. Mmultivariate logistic regression analysis of determinants of the misdiagnosis rate (Table 3) showed that children were more likely to be misdiagnosed if they were females (OR 2.33, p b 0.0001), in the youngest age group (OR 1.51, p b 0.0001), received treatment at nonchildren’s hospitals (OR 1.42, p = 0.01), or were from western Canada (OR 1.21, p b 0.02). Domicile, socioeconomic status, and surgeon’s specialty did not significantly influence the likelihood of misdiagnosis. Multivariate logistic regression analysis of determinants of the perforation rate (Table 4) showed that children were more likely to have perforated appendicitis if they were males (OR 1.09, p b 0.0001), younger [(≤5 years: OR 2.91, p b 0.0001), (6–11: OR 1.19, p b .0001)], resided outside maritime Canada, and treated by pediatric surgeons
Fig. 1. Flowchart outlining how study cohort was chosen.
(OR 1.25, p = 0.0003). Domicile, socioeconomic status, and type of treating hospital did not significantly influence the likelihood of perforated appendicitis.
Table 1 Characteristics of study population. Gender Male Female Age (years) 0–5 6–11 12–17 Geographic region Ontario West Maritime Territories Socioeconomic Status Lowest 2nd lowest Middle 2nd highest Highest Domicile Rural Urban Not available Type of treating hospital Non-children's Children's Specialty of treating surgeon General surgeon Pediatric surgeon Other Surgical appendectomy approach Open Laparoscopic Values reported in %.
59.0 41.0 4.3 35.2 60.5 47.6 43.0 8.7 0.7 18.9 18.5 20.0 20.1 20.5 13.7 70.1 16.2 75.0 25.0 72.5 22.4 5.1 52.7 47.3
L.H.A. Cheong, S. Emil / Journal of Pediatric Surgery 49 (2014) 777–781 Table 2 Distribution by type of appendicitis. Patient- & Hospital-level Factors Gender Male Female Age (%) 0–5 years old 6–11 years old 12–17 years old Type of treating hospital Non-children's Children's Type of treating surgeon Pediatric surgeon General surgeon Surgical approach Open Laparoscopic Geographic region Ontario West Maritime Territories Domicile Rural Urban Socioeconomic status Lowest 2nd lowest Middle 2nd highest Highest
Normal appendicitis
Acute appendicitis
Perforated appendicitis
1176 (4.8) 1780 (10.4)
16452 (67.0) 11080 (64.6)
6911 (28.2) 4303 (25.1)
177 (10.0) 754 (5.2) 1663 (6.6)
757 (42.8) 9580 (65.7) 17225 (68.7)
836 (47.2) 4243 (29.1) 6170 (24.6)
2043 (6.6) 551 (5.3)
20999 (67.6) 6563 (63.3)
8001 (25.8) 3248 (31.3)
476 (5.1) 1841 (6.1)
5805 (62.7) 20447 (68.1)
2974 (32.1) 7720 (25.7)
1266 (5.8) 1328 (6.8)
14306 (65.5) 13256 (67.7)
6255 (28.7) 4994 (25.6)
1062 (5.4) 1282 (7.2) 220 (6.1) 30 (10.0)
13570 11174 2640 178
(68.9) (62.8) (73.1) (59.1)
5060 (25.7) 5345 (30.0) 751 (20.8) 93 (31.0)
351 (6.2) 1839 (6.3)
3747 (66.2) 19323 (66.6)
1562 (27.6) 7869 (27.1)
464 516 513 478 570
(5.9) (6.7) (6.2) (5.7) (6.7)
5213 5104 5560 5531 5621
(66.6) (66.6) (67.2) (66.3) (66.2)
2149 (27.5) 2046 (26.7) 2205 (26.6) 2337 (28.0) 2295 (27.0)
Values reported as n (%).
Multivariate quantile regression analysis indicated that, compared to those with non-perforated appendicitis, median hospital stay was longer in children with perforated appendicitis (4 vs 2 days). In those with non-perforated appendicitis, longer hospital stays were found in children who had open surgery (estimate + 1 day, p b 0.0001) and were treated by pediatric surgeons (estimate + 1 day, p b 0.0001). Children from Ontario, western Canada and the Maritimes had shorter hospital stays (estimates − 1 day, p b 0.0001), versus those in the Territories. Among children with perforated appendicitis, longer hospital stays were found in children who had open surgery (estimate + 1 day, p b 0.0001) and were treated by pediatric surgeons (estimate +1 day, p b 0.0001). Children from the Maritimes had
779
slightly longer hospital stays (estimate 0.19 day, p b 0.0001). A shorter hospital stay was found for children treated at non-children’s hospital (estimate − 1 day, p b 0.0001), and those from Ontario (estimate − 0.81 day, p b 0.0001) and western Canada (estimate − 0.81 day, p b 0.0001). The age of children did affect length of hospital stay in nonperforated and perforated appendicitis but only to a negligible extent (estimate 0, p b 0.0001 in both instances). 3. Discussion Acute appendicitis is the most common surgical illness in children, and appendectomy is one of the most common surgical procedures performed in this age group. Any improvement in outcomes, even if small, may therefore have a profound effect on health care resource utilization and cost. An understanding of the determinants of these outcomes is the first step towards any attempts at improving them. Most of the population-based studies concerning pediatric appendicitis have come from the United States, a multi-payer health care system with inherent disparities in health care access due to differences in race, socioeconomic status, or insurance status. These disparities have correlated with outcomes of pediatric appendicitis, as well as multiple other diseases [3,4,8–10]. In a plethora of US studies looking at medical and surgical disease outcomes, disparities have often been identified [11,12]. For example, a recent study of adults admitted for acute cholangitis demonstrated that non-Caucasian patients, as well as patient insured by US Medicare or Medicaid, were more likely to die, experience a prolonged hospital stay, and incur higher hospital charges [8]. Canada has a single-payer health care system that covers all citizens and legal residents. The literature on disparities in health care in Canada has not been as extensive as that of the US, and has yielded a more mixed picture. For example, a study comparing access to breast cancer care in Ontario and California showed essentially no Canadian disparities based on socioeconomic status, while definite disparities existed in California [13]. A recent review of the Canadian literature on access to cancer care identified income as having a strong influence on screening, while age and geography were the most influential predictors of access to treatment and end-of-life care [14]. There are very few similar studies pertaining to pediatric surgical outcomes. Bratu et al. analyzed twenty years of data on pediatric appendicitis in Manitoba, and found that lower rural or urban socioeconomic status and northern area of residence were significantly associated with a
Table 3 Multivariate logistic regression: rate of misdiagnosis. Variables Gender Age (%)
Type of treating hospital Type of treating surgeon Geographic region
Domicile Socioeconomic status
Male Female (reference) 0–5 years old 6–11 years old 12–17 years old (reference) Non-children's Children's (reference) Pediatric surgeon General surgeon (reference) Ontario West Territories Maritime (reference) Rural Urban (reference) Lowest 2nd lowest Middle 2nd highest Highest (reference)
OR
95% CI
P
0.43 (−) 1.51 0.79 (−) 1.42 (−) 1.07 (−) 0.91 1.21 1.51 (−) 1.02 (−) 0.81 0.95 0.91 0.82 (−)
0.39–0.48 (−) 1.2–1.84 0.71–0.87 (−) 1.13–1.79 (−) 0.84–1.36 (−) 0.77–1.09 1.02–1.44 0.93–2.46 (−) 0.90–1.16 (−) 0.70–0.94 0.82–1.09 0.79–1.04 0.71–0.94 (−)
b0.0001 b0.0001 b0.0001 0.003 0.57 0.30 0.03 0.10 0.72 .06 0.23 0.78 0.06
780
L.H.A. Cheong, S. Emil / Journal of Pediatric Surgery 49 (2014) 777–781
Table 4 Multivariate logistic regression: rate of perforated appendicitis. Variables Gender Age (%)
Type of treating hospital Type of treating surgeon Geographic region
Domicile Socioeconomic status
Male Female (reference) 0–5 years old 6–11 years old 12–17 years old (reference) Non-children's Children's (reference) Pediatric surgeon General surgeon (reference) Ontario West Territories Maritime (reference) Rural Urban (reference) Lowest 2nd lowest Middle 2nd highest Highest (reference)
OR
(95% Wald CI)
1.09 (−) 2.91 1.19 (−) 1.04 (−) 1.25 (−) 1.31 1.70 1.82 (−) 1.04 (−) 1.02 1.01 0.98 1.04 (−)
1.04–1.15 (−) 2.59–3.27 1.13–1.25 (−) 0.92–1.17 (−) 1.11–1.41 (−) 1.18–1.45 1.52–1.88 1.35–2.47 (−) 0.97–1.10 (−) 0.94–1.11 0.94–1.10 0.91–1.06 0.96–1.12 (−)
P 0.0006 b0.0001 b0.0001 0.52 0.0003 b0.0001 b0.0001 b0.0001 0.27 0.65 0.89 0.23 0.31
higher perforation rate [15]. In population-based studies in the province of Ontario, Langer et al. showed lower duodenal perforation rates after pyloromyotomy and a lower appendicitis misdiagnosis rate, when children were cared for by pediatric surgeons [16,17]. We recently reported results of a population-based comparison of children with appendicitis in the United States and Canada, looking at some of the same outcomes examined here [7]. In general, the outcomes of Canadian children fell in between US children with private, and those with public or no insurance. The exception was the youngest age group, where Canadian children had better outcomes versus US children in all insurance categories [7]. We conducted the present study to further investigate possible outcome determinants in Canadian children. We obtained all the information from the Canadian Institute of Health Information that may potentially influence outcomes. In addition to investigating possible determinants of misdiagnosis and perforation rates, we also investigated length of stay, adding the surgical approach as a potential factor. Our findings reinforced some well-known and previously reported determinants of appendicitis outcomes. Females and the youngest patients were more likely to be misdiagnosed. Perforated patients were more likely to be in the youngest age group and be treated by pediatric surgeons [16,18]. However, Langer’s findings in Ontario of a higher rate of misdiagnosis by general surgeons, were not reproduced in this larger study from most parts of the country [16]. Rather, a higher risk of misdiagnosis was found if the child was treated at a non-children’s hospital. This may implicate the overall experience of the institution and other physicians involved in the diagnosis of appendicitis, such as emergency physicians and radiologists, rather than the treating surgeon. The slightly higher rates of misdiagnosis and perforation in western Canada may relate to greater distances between patients and children’s hospitals or pediatric surgical centres in the western provinces. The lack of association of misdiagnosis and perforation with domicile and income in our study is perhaps the most significant finding. It lends strong support to the notion that a single-payer system provides better access, and negates the deleterious effects of rural residence or lowsocioeconomic status on outcomes. We were surprised to find that a slight majority of children were still receiving open appendectomies, despite reported advantages of the laparoscopic approach [19–21]. This was associated with a longer hospital stay both for simple and perforated appendicitis in multivariate analysis, and certainly presents a potential area for improvement. Hospital stay was also longer for children with the entire spectrum of appendicitis treated by pediatric surgeons. This was previously reported in Ontario, and was difficult to explain since
this longer hospital stay was not associated with increases in complications or re-admissions. Pediatric surgeons may keep patients in the hospital until strict discharge criteria are met. Clinical pathways for appendicitis treatment have been shown to slightly prolong hospital stay, while decreasing complications [1]. Our study is limited by issues common to this type of outcomes research. Accuracy of the data is determined by coders in each hospital who extract the data retrospectively from medical charts. While standardized coding minimizes discrepancy and variability in data collection and reporting, errors can occur at any point from collection to submission and distribution. Data were not included from Quebec, Canada’s second largest province, representing approximately 25% of the nation’s population. There is more than one regulatory body that oversees healthcare data from Quebec, and approval from the Commission d'accès à l'information du Québec (CAI) is necessary before data can be released. Given the complexity of obtaining uniform information from more than one regulatory body in Quebec, information for children in that province was not obtained. Data regarding other important appendicitis outcomes, such as wound infections, intra-abdominal infections, need for additional procedures, and re-admissions were not adequately provided in the discharge abstract database. Despite these limitations, we have reported the first large Canadian population-based study investigating the outcome determinants in children with appendicitis. Within the Canadian single-payer, universal health care system, socioeconomic status and rural domicile do not affect outcomes or hospital stay. Nevertheless, opportunities for outcomes improvement exist. These include increasing the rate of laparoscopic appendectomies, exploring ways to decrease misdiagnosis rates in non-children’s hospitals, and investigating clinical pathways that may shorten hospital stay without increasing the complication rate. Acknowledgment We would like to thank Dr. Xianming Tan of the Biostatistics Core Facility, McGill University Health Centre Research Institute for providing assistance with the statistical analyses. References [1] Emil S, Taylor M, Ndiforchu F, et al. What are the true advantages of a pediatric appendicitis clinical pathway? Am Surg 2006;72:885–9. [2] Emil S, Laberge JM, Mikhail P, et al. Appendicitis in children: a ten-year update of therapeutic recommendations. J Pediatr Surg 2003;38:236–42. [3] 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. [4] 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. [5] Lee SL, Shekherdimian S, Chiu VY. Effect of race and socioeconomic status in the treatment of appendicitis in patients with equal health care access. Arch Surg 2011;146:156–61. [6] Lee SL, Shekherdimian S, Chiu VY, et al. Perforated appendicitis in children: equal access to care eliminates racial and socioeconomic disparities. J Pediatr Surg 2010;45:1203–7. [7] Cheong LHA, Emil S. Outcomes of pediatric appendicitis: an international comparison of the United States and Canada. JAMA Surg 2014;149:50-55. [8] McNabb-Baltar J, Trinh QD, Barkun AN. Disparities in outcomes following admission for cholangitis. PloS One 2013;8:e59487. [9] Nathan H, Frederick W, Choti MA, et al. Racial disparity in surgical mortality after major hepatectomy. J Am Coll Surg 2008;207:312–9. [10] Anger JT, Rodriguez LV, Wang Q, et al. Racial disparities in the surgical management of stress incontinence among female Medicare beneficiaries. J Urol 2007;177:1846–50. [11] Haider AH, Scott VK, Rehman KA, et al. Racial disparities in surgical care and outcomes in the United States: a comprehensive review of patient, provider, and systemic factors. J Am Coll Surg 2013;216:482–92. [12] Morris AM, Rhoads KF, Stain SC, et al. Understanding racial disparities in cancer treatment and outcomes. J Am Coll Surg 2010;211:105–13. [13] Gorey KM, Luginaah IN, Holowaty EJ, et al. Wait times for surgical and adjuvant radiation treatment of breast cancer in Canada and the United
L.H.A. Cheong, S. Emil / Journal of Pediatric Surgery 49 (2014) 777–781
[14] [15] [16] [17]
States: greater socioeconomic inequity in America. Clin Invest Med 2009;32: E239–49. Maddison AR, Asada Y, Urquhart R. Inequity in access to cancer care: a review of the Canadian literature. Cancer Causes Control 2011;22:359–66. Bratu I, Martens PJ, Leslie WD, et al. Pediatric appendicitis rupture rate: disparities despite universal health care. J Pediatr Surg 2008;43:1964–9. Somme S, To T, Langer JC. Effect of subspecialty training on outcome after pediatric appendectomy. J Pediatr Surg 2007;42:221–6. Langer JC, To T. Does pediatric surgical specialty training affect outcome after Ramstedt pyloromyotomy? A population-based study. Pediatrics 2004;113:1342–7.
781
[18] Emil SG, Taylor MB. Appendicitis in children treated by pediatric versus general surgeons. J Am Coll Surg 2007;204:34–9. [19] Esposito C, Calvo Al, Castagnetti M, et al. Open versus laparoscopic appendectomy in the pediatric population: a literature review and analysis of complications. J Laparoendosc Adv Surg Tech A 2012;22:834–9. [20] Masoomi H, Mills S, Dolich MO, et al. Comparison of outcomes of laparoscopic versus open appendectomy in children: data from the Nationwide Inpatient Sample (NIS), 2006–2008. World J Surg 2012:573–8. [21] Aziz O, Athanasiou T, Tekkis PP, et al. Laparoscopic versus open appendectomy in children: a meta-analysis. Ann Surg 2006;243:17–27.
