- Email: [email protected]
Increased Distance From the Tertiary Cardiac Center Is Associated With Worse 30-Day Outcomes After Cardiac Operations
Increased Distance From the Tertiary Cardiac Center Is Associated With Worse 30-Day Outcomes After Cardiac Operations Claudia L. Cote, BS, Shubhi Singh, BS, Alexandra M. Yip, MS, Joshua Murray, MS, Jeffrey B. MacLeod, BS, Sohrab Lutchmedial, MD, Craig D. Brown, MD, Rand Forgie, MD, Marc P. Pelletier, MD, MS, and Ansar Hassan, MD, PhD Dalhousie Medicine New Brunswick; Cardiovascular Research New Brunswick, New Brunswick Heart Center, Saint John Regional Hospital; and Horizon Health Network, Saint John, New Brunswick, Canada
Background. Numerous studies have examined the effect of geographic place of residence on access to cardiovascular care, but few have examined their effect on outcomes after cardiac operations. This study examined the effect of geographic place of residence on in-hospital and 30-day outcomes after cardiac operations. Methods. We performed a retrospective analysis of all patients undergoing nonemergency cardiac operations at a single institution between April 2004 and March 2011. Geographic place of residence was defined as the driving distance from the patient’s home to the tertiary cardiac care center divided into the following categories: 0 to 50 km, 50 to 100 km, 100 to 150 km, 150 to 200 km, 200 to 250 km, and more than 250 km. Multivariable logistic regression was used to determine the independent effect of driving distance on in-hospital and 30-day outcomes. Results. The final study population included 4,493 patients, of whom 3,897 (86.7%) had 30-day follow-up.
After adjusting for differences among patient groups, no consistent relationship existed between distance and in-hospital outcomes. However, increased distance beyond 100 km was significantly associated with a greater risk of adverse outcomes at 30 days (0 to 50 km: referent; 50 to 100 km: odds ratio, 1.16 [95% confidence interval, 0.83 to 1.62]; 100 to 150 km: 1.32 [1.05 to 1.65], 150 to 200 km: 1.68 [1.33 to 2.11], 200 to 250 km: 1.41 [1.06 to 1.88], and >250 km: 1.30 [1.04 to 1.63]). Conclusions. Patients who live at an increased distance from the tertiary cardiac care center are more likely to have worse 30-day outcomes after cardiac operations. Further study is required to determine the mechanisms underlying this relationship and how such inequalities may be minimized.
L
Patients and Methods
ower socioeconomic status and remote geographic place of residence have been associated with an increased prevalence of cardiovascular disease risk factors [1, 2], decreased access to cardiac care after acute myocardial infarction [3, 4], and inferior overall access to cardiac operations [5–7] and percutaneous coronary interventions [8, 9]. However, less is known about the effect of these nonclinical determinants of health on outcomes after cardiac operations. Although lower income [10] and socioeconomic status [11, 12] have been identified as independent predictors of reduced survival in patients undergoing cardiac operations, few studies to date have examined the effect of geographic place of residence on postsurgical outcomes. The purpose of the study was to examine the effect of geography on in-hospital and 30-day morbidity and mortality after cardiac operations at a single institution.
(Ann Thorac Surg 2015;-:-–-) Ó 2015 by The Society of Thoracic Surgeons
The Horizon Health Network Research Ethics Board approved the protocol for this study. All patients undergoing nonemergency cardiac operations between April 2004 and March 2011 at the New Brunswick Heart Center (NBHC), the sole tertiary cardiac care facility providing cardiac operations for the province of New Brunswick, Canada, were identified through the NBHC Cardiac Surgery Registry. The NBHC Cardiac Surgery Registry is a detailed clinical database that prospectively captures preoperative, intraoperative, and postoperative data on all patients undergoing cardiac operations at NBHC as well as 30-day follow-up data for most patients. For the purposes of this study, the data were restricted to residents of the province of New Brunswick covered by the provincial health insurance plan. Patients with a
Accepted for publication May 14, 2015. Address correspondence to Dr Hassan, Department of Cardiac Surgery, New Brunswick Heart Center, Saint John Regional Hospital, 400 University Ave, PO Box 2100, Saint John, NB E2L 4L2, Canada; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
The Appendix can be viewed in the online version of this article [http://dx.doi.org/10.1016/j.athoracsur.2015. 05.058] on http://www.annalsthoracicsurgery.org.
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.05.058
2
COTE ET AL GEOGRAPHY AND OUTCOMES IN CARDIAC OPERATIONS
Ann Thorac Surg 2015;-:-–-
postoperative hospital length of stay of greater than 30 days were excluded. Data for the following baseline clinical characteristics were collected: age (70 years), sex, body mass index, smoking history, diabetes mellitus, dyslipidemia, renal failure, hypertension, chronic obstructive pulmonary disease, peripheral vascular disease, cerebrovascular disease, recent myocardial infarction (21 days), atrial fibrillation, New York Heart Association Functional Classification, decreased left ventricular ejection fraction (<0.40), urgency status (urgent vs elective), previous cardiac operation, and length of wait from the time of cardiac catheterization to the time of the operation. Intraoperative variables of interest included procedure type, intraaortic balloon pump use, presence of inotropes upon transfer from the operating room to the intensive care unit, cardiopulmonary bypass time, and aortic crossclamp time.
following driving distance categories: within 50 km, 50 to 100 km, 100 to 150 km, 150 to 200, 200 to 250 km, or more than 250 km.
Socioeconomic Status The Statistics Canada’s Postal Code Conversion File [14] was used to match each patient’s residential postal code to a patient-specific dissemination area (DA), the smallest geographic unit for which Canadian Census information was available. DA-level socioeconomic information was derived from the 2006 Canada Census and was quantified using a DA-level deprivation index that consists of material and social components [12, 15, 16]. The material components include the proportion of people with no high school diploma, percentage employment, and the average income, all among those aged 15 years and older. The social components include the proportion of individuals aged 15 years and older living alone, the proportion of individuals who were separated, divorced, or widowed, and the proportion of single-parent families. The regional version of the deprivation index was used to assign a socioeconomic quintile rank to each patient comparing him or her with other residents of Atlantic Canada.
Distance Each patient’s driving distance from his or her home to the tertiary cardiac care center was calculated by submitting the residential postal code to a MapQuest (Denver, CO) application programming interface [13]. This information was used to assign patients to one of the Table 1. Geographic Distribution of Baseline Characteristics Characteristica Age 70 years Female sex Body mass index, kg/m2 <25 25–30 30 Smoking history Diabetes Dyslipidemia Renal failure Hypertension COPD Peripheral vascular disease Cerebrovascular disease Previous MI 21 days Atrial fibrillation Congestive heart failure NYHA classification I–II III–IV LVEF <0.40 Urgent status Previous cardiac operation Length of wait for operation, d a
50 km (n ¼ 813)
50–100 km (n ¼ 278)
100–150 km (n ¼ 1,028)
150–200 km (n ¼ 863)
200–250 km (n ¼ 405)
>250 km (n ¼ 1,106)
33.7 23.2
36.3 22.3
36.5 25.8
38.0 26.0
36.0 26.4
35.5 24.4
21.5 44.2 34.3 55.4 32.0 70.2 3.4 66.3 8.2 4.9 7.5 20.0 8.7 17.6
16.5 44.2 39.2 55.4 33.8 72.3 2.2 68.0 8.6 6.5 7.6 23.4 10.4 15.8
24.2 42.7 33.1 50.0 31.0 74.4 2.4 66.3 9.5 5.3 8.4 22.1 9.1 13.1
21.1 45.7 33.3 50.4 28.7 76.4 2.2 64.8 8.8 7.4 9.4 20.4 9.6 14.0
23.2 41.0 35.8 53.3 30.4 70.1 2.2 67.2 8.6 8.1 9.4 21.7 9.1 13.6
24.6 42.9 32.5 50.8 31.1 78.5 4.0 71.7 8.6 7.1 9.2 17.7 8.3 14.9
42.6 57.4 11.3 55.8 3.2 40 62
43.5 56.5 10.1 59.7 2.5 37 56
46.7 53.3 10.6 52.1 3.0 44 60
47.6 52.4 10.0 54.6 2.1 42 59
48.4 51.6 8.4 56.3 3.2 43 60
45.1 54.9 9.2 53.3 2.3 51 67
p Value 0.61 0.58 0.15
0.14 0.62 0.0003 0.11 0.02 0.96 0.08 0.67 0.12 0.87 0.13 0.24
0.57 0.21 0.60 0.01
Categoric data are presented as the percentage and continuous data as mean standard deviation.
COPD ¼ chronic obstructive pulmonary disease; Heart Association.
LVEF ¼ left ventricular ejection fraction;
MI ¼ myocardial infarction;
NYHA ¼ New York
Ann Thorac Surg 2015;-:-–-
3
COTE ET AL GEOGRAPHY AND OUTCOMES IN CARDIAC OPERATIONS
Table 2. Geographic Distribution of Intraoperative Characteristics Characteristica Procedure CABG Valve CABG þ valve Other CABG valve IABP insertion None Preoperative Intraoperative Postoperative Inotropes leaving OR CPB used CPB time, min AXC time, min a
50 km (n ¼ 813)
50–100 km (n ¼ 278)
100–150 km (n ¼ 1,028)
150–200 km (n ¼ 863)
200–250 km (n ¼ 405)
>250 km (n ¼ 1,106)
66.9 11.4 10.6 11.1
66.9 13.7 10.8 8.6
67.2 12.1 9.8 10.9
66.0 11.6 11.8 10.5
62.0 13.3 12.6 12.1
71.0 10.6 10.0 8.4
0.28
97.3 1.8 0.7 0.1 11.6 98.2 111 46 82 40
97.1 2.5 0.4 0.0 12.6 98.6 108 43 80 38
97.2 1.8 0.9 0.2 11.1 98.4 111 47 82 39
97.7 1.5 0.8 0.0 11.1 99.0 111 44 83 39
96.8 2.2 0.7 0.2 8.4 99.3 111 49 81 42
97.3 2.0 0.6 0.1 9.0 98.7 107 45 79 37
0.99
p Value
0.19 0.61 0.29 0.53
Categoric data are presented as the percentage and continuous data as mean standard deviation.
AXC ¼ aortic cross-clamp; OR ¼ operating room.
CABG ¼ coronary artery bypass grafting;
Outcomes In-hospital postoperative outcomes of interest included death, reoperation for any cause, prolonged ventilation exceeding 24 hours, surgical site infection, permanent stroke, renal failure, atrial fibrillation, a composite of these outcomes, and hospital length of stay. The 30-day outcomes consisted of any complication occurring after the point of initial discharge from the hospital but within 30 days of the surgical procedure. Complications that occurred before discharge were not included in the 30-day outcomes. The 30-day outcomes of interest consisted of all-cause mortality, pulmonary, infectious, neurologic, renal, valvular, and vascular complications, atrial fibrillation, readmission for any cause, and a composite of these outcomes (a detailed list of the 30-day outcomes of interest is available in Appendix A.)
Statistical Analysis Baseline characteristics, intraoperative variables, and rates of in-hospital and 30-day adverse outcomes were compared across distance categories using the KruskalWallis and analysis of variance test for continuous variables and the c2 and Fisher exact tests for categoric variables. Multivariable logistic regression was performed to determine the effect of driving distance category on likelihood of in-hospital and 30-day adverse outcomes after adjusting for differences in baseline characteristics, socioeconomic status, and procedure type. Statistical analyses were performed using SAS 9.2 software (SAS Institute Inc, Cary, NC). P values of less than 0.05 were considered significant.
Results The final study population included 4,493 patients (1,117 female [24.9%]) with a mean age of 64.9 10.5 years. They
CPB ¼ cardiopulmonary bypass;
IABP ¼ intraaortic balloon pump;
were distributed across the distance categories in the following manner: within 50 km, 813 (18.1%); 50 to 100 km, 278 (6.2%); 100 to 150 km, 1,028 (22.9%); 150 to 200 km, 863 (19.2%); 200 to 250 km, 405 (9.0%); and more than 250 km, 1,106 (24.6%). Thirty-day follow-up was available for 3,897 patients (86.7%). Regarding baseline characteristics, rates of dyslipidemia and hypertension varied with distance, with the highest rates found in patients more than 250 km from the tertiary cardiac care center (Table 1). Patients at a greater distance also experienced longer waiting times for their operation. No differences in intraoperative variables were noted (Table 2). Significant variation in socioeconomic status was found across distance categories, with 27.7% of patients living within 50 km of the tertiary cardiac care center assigned to the first quintile of material deprivation (most privileged) compared with only 3.0% of patients living more than 250 km (Fig 1a). Similarly, 19.2% of patients living within 50 km of the tertiary cardiac care center were assigned to the first quintile of social deprivation (most privileged) compared with only 8.8% of patients living more than 250 km (Fig 1b). During the course of the study period, 83 patients (1.8%) died in-hospital and 2,074 (46.1%) experienced an in-hospital complication. Of the 3897 patients who had 30-day follow-up, 1,180 (30.3%) experienced a 30-day complication, including death, 594 (15.2%) were readmitted to the hospital, and 1,190 (30.5%) experienced a composite 30-day adverse outcome consisting of any complication or readmission, or both. An increased distance from the tertiary cardiac care center had no effect on in-hospital morbidity or death (Table 3) but was associated with higher 30-day rates of any complication or readmission, or both (Table 4).
4
COTE ET AL GEOGRAPHY AND OUTCOMES IN CARDIAC OPERATIONS
Ann Thorac Surg 2015;-:-–-
Comment
Fig 1. Geographic distribution of (a) material and (b) social deprivation.
After adjustment for baseline characteristics, socioeconomic status, and intraoperative variables, increased driving distance did not have an independent effect on rates of in-hospital adverse outcomes with the exception of patients who resided beyond 250 km, for whom rates of in-hospital complications appeared significantly lower. However, increased distance beyond 100 km was consistently associated with a greater adjusted risk of composite 30-day adverse outcomes, with the highest risk found in patients who resided between 150 and 200 from NBHC (Table 5).
This study found no consistent relationship between distance and in-hospital outcomes. However, increased distance beyond 100 km from NBHC was uniformly associated with a higher rate of composite 30-day adverse outcomes after cardiac operations. Although lower income [10] and socioeconomic status [11, 12] have been associated with worse outcomes after cardiac operations, few studies have examined the effect of geographic place of residence on postoperative outcomes. In a prospective evaluation of risk factors for readmission after cardiac operations from 2001, Ferraris and colleagues [17] demonstrated that distance from the patient’s home to the treating hospital was a predictor for 30-day readmission. The findings of this study and those of Ferraris and colleagues from more than a decade ago suggest that geography continues to be important in determining 30-day outcomes after cardiac operations. Despite the presence a universal health care system in Canada, there exists an inequitable distribution of specialized care services, with a higher proportion of specialized services for the treatment of congestive heart failure and acute myocardial infarction found in urban areas compared with rural or remote areas [18–20]. The effect of increased distance on 30-day outcomes as seen in this study may have been the result of a difference in the type and level of follow-up care received by patients who lived at a further distance from the tertiary cardiac care center. In addition, where physician readmission practices have been shown to differ amongst specialties [21, 22], the possibility exists that physicians looking after cardiac surgical patients in more remote areas may have had an altered threshold for readmission. Beyond these differences in physician services, access to exercise-based cardiac rehabilitation after cardiac operations, which has been shown to reduce cholesterol levels, systolic blood pressure, self-reported smoking rates, and death [23], may have been reduced among patients living at an increased distance. Further research into the varying level of postoperative follow-up, differing
Table 3. Geographic Distribution of In-Hospital Outcomes In-Hospital Outcomesa
50 km (n ¼ 813)
50–100 km (n ¼ 278)
100–150 km (n ¼ 1028)
150–200 km (n ¼ 863)
200–250 km (n ¼ 405)
>250 km (n ¼ 1,106)
p Value
Death Reoperation Prolonged ventilation >24 h Wound infection Stroke Renal failure Atrial fibrillation In-hospital compositeb Hospital length of stay, d
2.6 2.7 4.1 3.2 3.3 10.7 38.5 48.7 5 (4–7)
2.5 1.8 3.2 4.7 1.1 11.9 37.8 47.1 5 (4–7)
2.4 2.9 4.0 2 1.8 10.9 37.6 47.9 5 (4–7)
1.4 1.6 3.2 1.5 1.6 8.9 39.5 45.1 5 (4–6)
1.2 1.7 3.5 1.2 1.7 9.9 36.8 45.2 5 (4–7)
1.2 3.1 3.2 1.9 1.0 9.6 36 43.7 5 (4–6)
0.09 0.25 0.85 0.007 0.007 0.6 0.71 0.24 <0.001
a Categoric data are presented as the percentage and continuous data as median (interquartile range). complication or death before discharge.
b
In-hospital composite outcome includes any
Ann Thorac Surg 2015;-:-–-
5
COTE ET AL GEOGRAPHY AND OUTCOMES IN CARDIAC OPERATIONS
Table 4. Geographic Distribution of 30-Day Adverse Outcomes 30-Day Outcome, % Death Pulmonary complication Infection Neurologic complication Renal complication Valvular complication Vascular complication Atrial fibrillation Any complication/death Readmission 30-day compositea a
50 km (n ¼ 685)
50–100 km (n ¼ 238)
100–150 km (n ¼ 885)
150–200 km (n ¼ 768)
200–250 km (n ¼ 350)
>250 km (n ¼ 971)
p Value
0.3 5.1 7.3 0.4 0.2 0 0.2 4.4 24.7 11.4 25.1
1.3 5.0 9.2 0.8 0.4 0 0 6.7 28.2 11.8 28.2
0.5 4.9 9.6 1.5 0.1 0.5 0.1 5.8 30.8 15.8 31.0
0.1 6.9 10.6 0.7 0.3 0.1 0.1 5.1 35.3 17.4 35.5
0.3 6.9 7.7 0.6 0.3 0 0 3.1 31.4 15.7 32.0
0.5 8.3 7.1 0.8 0.3 0.2 0.2 3.5 29.9 16.4 30.1
0.27 0.03 0.09 0.3 0.9 0.3 0.94 0.08 0.001 0.01 0.001
Composite 30-day adverse outcome includes any complication, death, or readmission within 30 days after discharge.
thresholds for readmission, and unequal access to services, such as postoperative cardiac rehabilitation, is needed to determine their role in driving the relationship between increased distance and 30-day adverse outcomes. Patient-specific behavior may have also contributed to the differences seen in 30-day outcomes across distance categories. For instance, patients living in differing parts of the province may have been more likely to wait for more severe exacerbations of symptoms before seeking medical attention. Inconsistencies in health care–seeking behaviors have been seen between patients in rural and urban populations, where patients from rural populations tend to underuse specialized physician services, whereas those from urban populations may overuse them [24]. Furthermore, patients from remote parts of the province may have been less willing or were less able to have been managed as an outpatient due to issues of increased distance from their local hospital or inadequate social and financial support, or both. The cost of getting to a health care center has been previously identified as a deterrent for rural populations seeking care [25]. Although this study adjusted for differences in socioeconomic status in terms of material and social deprivation, the cost of travel to receive health care
was not taken into account. A closer look at the direct and indirect costs for patients and their families is needed to establish whether patients at increased distances delay seeking attention for financial reasons. This study is not without limitations. This study did not use individual patient-level data in determining socioeconomic status; rather, it was derived from population data using DA of residence to derive a social and material deprivation index at an aggregate level. Although data at the individual level would potentially be a more accurate measure of socioeconomic status, the deprivation index is a validated tool for estimating socioeconomic status [15]. In conclusion, increased distance from the tertiary cardiac care center was not consistently associated with differences in in-hospital outcomes after cardiac operations. However, distance was uniformly associated with worse 30-day outcomes. These results highlight the important disparities that exist in patient outcomes after cardiac operations based on their geographic place of residence. Although this association may have a clinical basis, further study is required to determine the roles that factors, such as variable physician follow-up and differences in patient behavior, play in promoting this relationship.
Table 5. Logistic Regression of In-Hospital and Composite 30-Day Adverse Outcome on Baseline Clinical Factors and Distance Categories In-Hospital Composite Distance From NBHC 50 km 50–100 km 100–150 km 150–200 km 200–250 km >250 km CI ¼ confidence interval;
30-Day Composite
OR (95% CI)
p Value
OR (95% CI)
p Value
1.00 (reference) 0.92 (0.69–1.24) 0.97 (0.80–1.18) 0.83 (0.68–1.02) 0.86 (0.66–1.11) 0.79 (0.65–0.96)
0.58 0.74 0.08 0.24 0.02
1.00 (reference) 1.16 (0.83–1.62) 1.32 (1.05–1.65) 1.68 (1.33–2.11) 1.41 (1.06–1.88) 1.30 (1.04–1.63)
0.4 0.02 <0.0001 0.02 0.02
NBHC ¼ New Brunswick Heart Center;
OR ¼ odds ratio.
6
COTE ET AL GEOGRAPHY AND OUTCOMES IN CARDIAC OPERATIONS
The authors gratefully acknowledge financial support from the New Brunswick Health Research Foundation and the Heart and Stroke Foundation of New Brunswick. As well, the authors wish to recognize the contribution of Lorelei Churchill, Health & Business Analytics–Decision Support, Horizon Health Network New Brunswick, to the completion of this study.
Ann Thorac Surg 2015;-:-–-
12.
13. 14.
References 1. Fang R, Kmetic A, Millar J, Drasic L. Disparities in chronic disease among Canada’s low-income populations. Prev Chronic Dis 2009;6:A115. 2. Lee DS, Chiu M, Manuel DG, et al. Trends in risk factors for cardiovascular disease in Canada: temporal, sociodemographic and geographic factors. CMAJ 2009;181: E55–66. 3. Alter DA, Naylor CD, Austin P, Tu JV. Effects of socioeconomic status on access to invasive cardiac procedures and on mortality after acute myocardial infarction. N Engl J Med 1999;341:1359–67. 4. Diex Roux AV, Merkin SS, Arnett D, et al. Neighborhood of residence and incidence of coronary heart disease. N Engl J Med 2001;345:99–106. 5. Gregory PM, Malka ES, Kostis JB, Wilson AC, Arora JK, Rhoads GG. Impact of geographic proximity to cardiac revascularization services on service utilization. Med Care 2000;38:45–57. 6. Seidel JE, Ghali WA, Faris PD, et al. Geographical location of residence and uniformity of access to cardiac revascularization services after catheterization. Can J Cardiol 2004;20: 517–23. 7. Vanasse A, Niyonsenga T, Courteau J, Hemiari A. Access to myocardial revascularization procedures: closing the gap with time? BMC Public Health 2006;6:60. 8. Hartford K, Roos LL, Walld R. Regional variation in angiography, coronary artery bypass surgery, and percutaneous transluminal coronary angioplasty in Manitoba, 1987 to 1992: the funnel effect. Med Care 1998;36:1022–32. 9. Hassan A, Pearce NJ, Mathers J, Veugelers PJ, Hirsch GM, Cox JL. The effect of place of residence on access to invasive cardiac services following acute myocardial infarction. Can J Cardiol 2009;25:207–12. 10. Boscarino JA, Chang J. Survival after coronary artery bypass graft surgery and community socioeconomic status: clinical and research implications. Med Care 1999;37:210–6. 11. Ancona C, Agabiti N, Forastiere F, et al. Coronary artery bypass graft surgery: socioeconomic inequalities in access
15. 16.
17. 18. 19. 20. 21. 22. 23.
24.
25.
and in 30 day mortality. A population-based study in Rome, italy. J Epidemiol Community Health 2000;54:930–5. Blais C, Hamel D, Rinfret S. Impact of socioeconomic deprivation and area of residence on access to coronary revascularization and mortality after a first acute myocardial infarction in Qu ebec. Can J Cardiol 2012;28:169–77. Roy A, Na Y. Batch production of driving distances and times using SASÒ and Web Map APIs. Paper 091-2012, SAS Global Forum 2012. Postal codes conversion file (PCCF). Reference guide, catalogue no. 92-153-g, no 02. Ottawa: Minister of Industry, July 2011. Pampalon R, Hamel D, Gamache P, Raymond G. A deprivation index for health planning in Canada. Chronic Dis Can 2009;29:178–91. e publique du Qu ebec (INSPQ). Institut national de sant Deprivation index. Equivalence table between the dissemination/enumeration areas and the different geographic territories, version: March 3, 2010. http://www2.Inspq.Qc.Ca/ santescope/indicedefavoeng.Asp?Noindd¼9&lg¼en. Ferraris VA, Ferraris SP, Harmon RC, Evans BD. Risk factors for early hospital readmission after cardiac operations. J Thorac Cardiovasc Surg 2001;122:278–86. Pitblado JR, Pong RW. Geographic distribution of physicians in Canada. Sudbury, ON: Centre for Rural and Northern Health Research, Laurentian University; 1999. Pong RW, Pitblado JR. Geographic distribution of physicians in Canada: beyond how many and where. Ottawa: Canadian Institute for Health Information; 2005. Tu K, Gong Y, Austin PC, Jaakimanian L, Tu JV. An overview of the types of physicians treating acute cardiac conditions in Canada. Can J Cardiol 2004;20:282–91. Roos NP. Hospitalization style of physicians in Manitoba: the disturbing lack of logic in medical practice. Health Serv Res 1992;27:361. Roos NP, Flowerdew G, Wajda A, Tate RB. Variations in physicians’ hospitalization practices: a population-based study in Manitoba, Canada. Am J Public Health 1986;76:45–51. Taylor RS, Brown A, Ebrahim S, et al. Exercise-based rehabilitation for patients with coronary heart disease: systematic review and meta-analysis of randomized controlled trials. Am J Med 2004;116:682–92. Vanasse A, Courteau J, Cohen A, Orzanco M, Drouin C. Rural-urban disparities in the management and health issues of chronic diseases in Quebec (Canada) in the early 2000s. Rural Remote Health 2010;10:1–15. Harju BL, Wuensch KL, Kuhl EA, Cross NJ. Comparison of rural and urban residents’ implicit and explicit attitudes related to seeking medical care. J Rural Health 2006;22: 359–63.
Background. Numerous studies have examined the effect of geographic place of residence on access to cardiovascular care, but few have examined their effect on outcomes after cardiac operations. This study examined the effect of geographic place of residence on in-hospital and 30-day outcomes after cardiac operations. Methods. We performed a retrospective analysis of all patients undergoing nonemergency cardiac operations at a single institution between April 2004 and March 2011. Geographic place of residence was defined as the driving distance from the patient’s home to the tertiary cardiac care center divided into the following categories: 0 to 50 km, 50 to 100 km, 100 to 150 km, 150 to 200 km, 200 to 250 km, and more than 250 km. Multivariable logistic regression was used to determine the independent effect of driving distance on in-hospital and 30-day outcomes. Results. The final study population included 4,493 patients, of whom 3,897 (86.7%) had 30-day follow-up.
After adjusting for differences among patient groups, no consistent relationship existed between distance and in-hospital outcomes. However, increased distance beyond 100 km was significantly associated with a greater risk of adverse outcomes at 30 days (0 to 50 km: referent; 50 to 100 km: odds ratio, 1.16 [95% confidence interval, 0.83 to 1.62]; 100 to 150 km: 1.32 [1.05 to 1.65], 150 to 200 km: 1.68 [1.33 to 2.11], 200 to 250 km: 1.41 [1.06 to 1.88], and >250 km: 1.30 [1.04 to 1.63]). Conclusions. Patients who live at an increased distance from the tertiary cardiac care center are more likely to have worse 30-day outcomes after cardiac operations. Further study is required to determine the mechanisms underlying this relationship and how such inequalities may be minimized.
L
Patients and Methods
ower socioeconomic status and remote geographic place of residence have been associated with an increased prevalence of cardiovascular disease risk factors [1, 2], decreased access to cardiac care after acute myocardial infarction [3, 4], and inferior overall access to cardiac operations [5–7] and percutaneous coronary interventions [8, 9]. However, less is known about the effect of these nonclinical determinants of health on outcomes after cardiac operations. Although lower income [10] and socioeconomic status [11, 12] have been identified as independent predictors of reduced survival in patients undergoing cardiac operations, few studies to date have examined the effect of geographic place of residence on postsurgical outcomes. The purpose of the study was to examine the effect of geography on in-hospital and 30-day morbidity and mortality after cardiac operations at a single institution.
(Ann Thorac Surg 2015;-:-–-) Ó 2015 by The Society of Thoracic Surgeons
The Horizon Health Network Research Ethics Board approved the protocol for this study. All patients undergoing nonemergency cardiac operations between April 2004 and March 2011 at the New Brunswick Heart Center (NBHC), the sole tertiary cardiac care facility providing cardiac operations for the province of New Brunswick, Canada, were identified through the NBHC Cardiac Surgery Registry. The NBHC Cardiac Surgery Registry is a detailed clinical database that prospectively captures preoperative, intraoperative, and postoperative data on all patients undergoing cardiac operations at NBHC as well as 30-day follow-up data for most patients. For the purposes of this study, the data were restricted to residents of the province of New Brunswick covered by the provincial health insurance plan. Patients with a
Accepted for publication May 14, 2015. Address correspondence to Dr Hassan, Department of Cardiac Surgery, New Brunswick Heart Center, Saint John Regional Hospital, 400 University Ave, PO Box 2100, Saint John, NB E2L 4L2, Canada; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
The Appendix can be viewed in the online version of this article [http://dx.doi.org/10.1016/j.athoracsur.2015. 05.058] on http://www.annalsthoracicsurgery.org.
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.05.058
2
COTE ET AL GEOGRAPHY AND OUTCOMES IN CARDIAC OPERATIONS
Ann Thorac Surg 2015;-:-–-
postoperative hospital length of stay of greater than 30 days were excluded. Data for the following baseline clinical characteristics were collected: age (70 years), sex, body mass index, smoking history, diabetes mellitus, dyslipidemia, renal failure, hypertension, chronic obstructive pulmonary disease, peripheral vascular disease, cerebrovascular disease, recent myocardial infarction (21 days), atrial fibrillation, New York Heart Association Functional Classification, decreased left ventricular ejection fraction (<0.40), urgency status (urgent vs elective), previous cardiac operation, and length of wait from the time of cardiac catheterization to the time of the operation. Intraoperative variables of interest included procedure type, intraaortic balloon pump use, presence of inotropes upon transfer from the operating room to the intensive care unit, cardiopulmonary bypass time, and aortic crossclamp time.
following driving distance categories: within 50 km, 50 to 100 km, 100 to 150 km, 150 to 200, 200 to 250 km, or more than 250 km.
Socioeconomic Status The Statistics Canada’s Postal Code Conversion File [14] was used to match each patient’s residential postal code to a patient-specific dissemination area (DA), the smallest geographic unit for which Canadian Census information was available. DA-level socioeconomic information was derived from the 2006 Canada Census and was quantified using a DA-level deprivation index that consists of material and social components [12, 15, 16]. The material components include the proportion of people with no high school diploma, percentage employment, and the average income, all among those aged 15 years and older. The social components include the proportion of individuals aged 15 years and older living alone, the proportion of individuals who were separated, divorced, or widowed, and the proportion of single-parent families. The regional version of the deprivation index was used to assign a socioeconomic quintile rank to each patient comparing him or her with other residents of Atlantic Canada.
Distance Each patient’s driving distance from his or her home to the tertiary cardiac care center was calculated by submitting the residential postal code to a MapQuest (Denver, CO) application programming interface [13]. This information was used to assign patients to one of the Table 1. Geographic Distribution of Baseline Characteristics Characteristica Age 70 years Female sex Body mass index, kg/m2 <25 25–30 30 Smoking history Diabetes Dyslipidemia Renal failure Hypertension COPD Peripheral vascular disease Cerebrovascular disease Previous MI 21 days Atrial fibrillation Congestive heart failure NYHA classification I–II III–IV LVEF <0.40 Urgent status Previous cardiac operation Length of wait for operation, d a
50 km (n ¼ 813)
50–100 km (n ¼ 278)
100–150 km (n ¼ 1,028)
150–200 km (n ¼ 863)
200–250 km (n ¼ 405)
>250 km (n ¼ 1,106)
33.7 23.2
36.3 22.3
36.5 25.8
38.0 26.0
36.0 26.4
35.5 24.4
21.5 44.2 34.3 55.4 32.0 70.2 3.4 66.3 8.2 4.9 7.5 20.0 8.7 17.6
16.5 44.2 39.2 55.4 33.8 72.3 2.2 68.0 8.6 6.5 7.6 23.4 10.4 15.8
24.2 42.7 33.1 50.0 31.0 74.4 2.4 66.3 9.5 5.3 8.4 22.1 9.1 13.1
21.1 45.7 33.3 50.4 28.7 76.4 2.2 64.8 8.8 7.4 9.4 20.4 9.6 14.0
23.2 41.0 35.8 53.3 30.4 70.1 2.2 67.2 8.6 8.1 9.4 21.7 9.1 13.6
24.6 42.9 32.5 50.8 31.1 78.5 4.0 71.7 8.6 7.1 9.2 17.7 8.3 14.9
42.6 57.4 11.3 55.8 3.2 40 62
43.5 56.5 10.1 59.7 2.5 37 56
46.7 53.3 10.6 52.1 3.0 44 60
47.6 52.4 10.0 54.6 2.1 42 59
48.4 51.6 8.4 56.3 3.2 43 60
45.1 54.9 9.2 53.3 2.3 51 67
p Value 0.61 0.58 0.15
0.14 0.62 0.0003 0.11 0.02 0.96 0.08 0.67 0.12 0.87 0.13 0.24
0.57 0.21 0.60 0.01
Categoric data are presented as the percentage and continuous data as mean standard deviation.
COPD ¼ chronic obstructive pulmonary disease; Heart Association.
LVEF ¼ left ventricular ejection fraction;
MI ¼ myocardial infarction;
NYHA ¼ New York
Ann Thorac Surg 2015;-:-–-
3
COTE ET AL GEOGRAPHY AND OUTCOMES IN CARDIAC OPERATIONS
Table 2. Geographic Distribution of Intraoperative Characteristics Characteristica Procedure CABG Valve CABG þ valve Other CABG valve IABP insertion None Preoperative Intraoperative Postoperative Inotropes leaving OR CPB used CPB time, min AXC time, min a
50 km (n ¼ 813)
50–100 km (n ¼ 278)
100–150 km (n ¼ 1,028)
150–200 km (n ¼ 863)
200–250 km (n ¼ 405)
>250 km (n ¼ 1,106)
66.9 11.4 10.6 11.1
66.9 13.7 10.8 8.6
67.2 12.1 9.8 10.9
66.0 11.6 11.8 10.5
62.0 13.3 12.6 12.1
71.0 10.6 10.0 8.4
0.28
97.3 1.8 0.7 0.1 11.6 98.2 111 46 82 40
97.1 2.5 0.4 0.0 12.6 98.6 108 43 80 38
97.2 1.8 0.9 0.2 11.1 98.4 111 47 82 39
97.7 1.5 0.8 0.0 11.1 99.0 111 44 83 39
96.8 2.2 0.7 0.2 8.4 99.3 111 49 81 42
97.3 2.0 0.6 0.1 9.0 98.7 107 45 79 37
0.99
p Value
0.19 0.61 0.29 0.53
Categoric data are presented as the percentage and continuous data as mean standard deviation.
AXC ¼ aortic cross-clamp; OR ¼ operating room.
CABG ¼ coronary artery bypass grafting;
Outcomes In-hospital postoperative outcomes of interest included death, reoperation for any cause, prolonged ventilation exceeding 24 hours, surgical site infection, permanent stroke, renal failure, atrial fibrillation, a composite of these outcomes, and hospital length of stay. The 30-day outcomes consisted of any complication occurring after the point of initial discharge from the hospital but within 30 days of the surgical procedure. Complications that occurred before discharge were not included in the 30-day outcomes. The 30-day outcomes of interest consisted of all-cause mortality, pulmonary, infectious, neurologic, renal, valvular, and vascular complications, atrial fibrillation, readmission for any cause, and a composite of these outcomes (a detailed list of the 30-day outcomes of interest is available in Appendix A.)
Statistical Analysis Baseline characteristics, intraoperative variables, and rates of in-hospital and 30-day adverse outcomes were compared across distance categories using the KruskalWallis and analysis of variance test for continuous variables and the c2 and Fisher exact tests for categoric variables. Multivariable logistic regression was performed to determine the effect of driving distance category on likelihood of in-hospital and 30-day adverse outcomes after adjusting for differences in baseline characteristics, socioeconomic status, and procedure type. Statistical analyses were performed using SAS 9.2 software (SAS Institute Inc, Cary, NC). P values of less than 0.05 were considered significant.
Results The final study population included 4,493 patients (1,117 female [24.9%]) with a mean age of 64.9 10.5 years. They
CPB ¼ cardiopulmonary bypass;
IABP ¼ intraaortic balloon pump;
were distributed across the distance categories in the following manner: within 50 km, 813 (18.1%); 50 to 100 km, 278 (6.2%); 100 to 150 km, 1,028 (22.9%); 150 to 200 km, 863 (19.2%); 200 to 250 km, 405 (9.0%); and more than 250 km, 1,106 (24.6%). Thirty-day follow-up was available for 3,897 patients (86.7%). Regarding baseline characteristics, rates of dyslipidemia and hypertension varied with distance, with the highest rates found in patients more than 250 km from the tertiary cardiac care center (Table 1). Patients at a greater distance also experienced longer waiting times for their operation. No differences in intraoperative variables were noted (Table 2). Significant variation in socioeconomic status was found across distance categories, with 27.7% of patients living within 50 km of the tertiary cardiac care center assigned to the first quintile of material deprivation (most privileged) compared with only 3.0% of patients living more than 250 km (Fig 1a). Similarly, 19.2% of patients living within 50 km of the tertiary cardiac care center were assigned to the first quintile of social deprivation (most privileged) compared with only 8.8% of patients living more than 250 km (Fig 1b). During the course of the study period, 83 patients (1.8%) died in-hospital and 2,074 (46.1%) experienced an in-hospital complication. Of the 3897 patients who had 30-day follow-up, 1,180 (30.3%) experienced a 30-day complication, including death, 594 (15.2%) were readmitted to the hospital, and 1,190 (30.5%) experienced a composite 30-day adverse outcome consisting of any complication or readmission, or both. An increased distance from the tertiary cardiac care center had no effect on in-hospital morbidity or death (Table 3) but was associated with higher 30-day rates of any complication or readmission, or both (Table 4).
4
COTE ET AL GEOGRAPHY AND OUTCOMES IN CARDIAC OPERATIONS
Ann Thorac Surg 2015;-:-–-
Comment
Fig 1. Geographic distribution of (a) material and (b) social deprivation.
After adjustment for baseline characteristics, socioeconomic status, and intraoperative variables, increased driving distance did not have an independent effect on rates of in-hospital adverse outcomes with the exception of patients who resided beyond 250 km, for whom rates of in-hospital complications appeared significantly lower. However, increased distance beyond 100 km was consistently associated with a greater adjusted risk of composite 30-day adverse outcomes, with the highest risk found in patients who resided between 150 and 200 from NBHC (Table 5).
This study found no consistent relationship between distance and in-hospital outcomes. However, increased distance beyond 100 km from NBHC was uniformly associated with a higher rate of composite 30-day adverse outcomes after cardiac operations. Although lower income [10] and socioeconomic status [11, 12] have been associated with worse outcomes after cardiac operations, few studies have examined the effect of geographic place of residence on postoperative outcomes. In a prospective evaluation of risk factors for readmission after cardiac operations from 2001, Ferraris and colleagues [17] demonstrated that distance from the patient’s home to the treating hospital was a predictor for 30-day readmission. The findings of this study and those of Ferraris and colleagues from more than a decade ago suggest that geography continues to be important in determining 30-day outcomes after cardiac operations. Despite the presence a universal health care system in Canada, there exists an inequitable distribution of specialized care services, with a higher proportion of specialized services for the treatment of congestive heart failure and acute myocardial infarction found in urban areas compared with rural or remote areas [18–20]. The effect of increased distance on 30-day outcomes as seen in this study may have been the result of a difference in the type and level of follow-up care received by patients who lived at a further distance from the tertiary cardiac care center. In addition, where physician readmission practices have been shown to differ amongst specialties [21, 22], the possibility exists that physicians looking after cardiac surgical patients in more remote areas may have had an altered threshold for readmission. Beyond these differences in physician services, access to exercise-based cardiac rehabilitation after cardiac operations, which has been shown to reduce cholesterol levels, systolic blood pressure, self-reported smoking rates, and death [23], may have been reduced among patients living at an increased distance. Further research into the varying level of postoperative follow-up, differing
Table 3. Geographic Distribution of In-Hospital Outcomes In-Hospital Outcomesa
50 km (n ¼ 813)
50–100 km (n ¼ 278)
100–150 km (n ¼ 1028)
150–200 km (n ¼ 863)
200–250 km (n ¼ 405)
>250 km (n ¼ 1,106)
p Value
Death Reoperation Prolonged ventilation >24 h Wound infection Stroke Renal failure Atrial fibrillation In-hospital compositeb Hospital length of stay, d
2.6 2.7 4.1 3.2 3.3 10.7 38.5 48.7 5 (4–7)
2.5 1.8 3.2 4.7 1.1 11.9 37.8 47.1 5 (4–7)
2.4 2.9 4.0 2 1.8 10.9 37.6 47.9 5 (4–7)
1.4 1.6 3.2 1.5 1.6 8.9 39.5 45.1 5 (4–6)
1.2 1.7 3.5 1.2 1.7 9.9 36.8 45.2 5 (4–7)
1.2 3.1 3.2 1.9 1.0 9.6 36 43.7 5 (4–6)
0.09 0.25 0.85 0.007 0.007 0.6 0.71 0.24 <0.001
a Categoric data are presented as the percentage and continuous data as median (interquartile range). complication or death before discharge.
b
In-hospital composite outcome includes any
Ann Thorac Surg 2015;-:-–-
5
COTE ET AL GEOGRAPHY AND OUTCOMES IN CARDIAC OPERATIONS
Table 4. Geographic Distribution of 30-Day Adverse Outcomes 30-Day Outcome, % Death Pulmonary complication Infection Neurologic complication Renal complication Valvular complication Vascular complication Atrial fibrillation Any complication/death Readmission 30-day compositea a
50 km (n ¼ 685)
50–100 km (n ¼ 238)
100–150 km (n ¼ 885)
150–200 km (n ¼ 768)
200–250 km (n ¼ 350)
>250 km (n ¼ 971)
p Value
0.3 5.1 7.3 0.4 0.2 0 0.2 4.4 24.7 11.4 25.1
1.3 5.0 9.2 0.8 0.4 0 0 6.7 28.2 11.8 28.2
0.5 4.9 9.6 1.5 0.1 0.5 0.1 5.8 30.8 15.8 31.0
0.1 6.9 10.6 0.7 0.3 0.1 0.1 5.1 35.3 17.4 35.5
0.3 6.9 7.7 0.6 0.3 0 0 3.1 31.4 15.7 32.0
0.5 8.3 7.1 0.8 0.3 0.2 0.2 3.5 29.9 16.4 30.1
0.27 0.03 0.09 0.3 0.9 0.3 0.94 0.08 0.001 0.01 0.001
Composite 30-day adverse outcome includes any complication, death, or readmission within 30 days after discharge.
thresholds for readmission, and unequal access to services, such as postoperative cardiac rehabilitation, is needed to determine their role in driving the relationship between increased distance and 30-day adverse outcomes. Patient-specific behavior may have also contributed to the differences seen in 30-day outcomes across distance categories. For instance, patients living in differing parts of the province may have been more likely to wait for more severe exacerbations of symptoms before seeking medical attention. Inconsistencies in health care–seeking behaviors have been seen between patients in rural and urban populations, where patients from rural populations tend to underuse specialized physician services, whereas those from urban populations may overuse them [24]. Furthermore, patients from remote parts of the province may have been less willing or were less able to have been managed as an outpatient due to issues of increased distance from their local hospital or inadequate social and financial support, or both. The cost of getting to a health care center has been previously identified as a deterrent for rural populations seeking care [25]. Although this study adjusted for differences in socioeconomic status in terms of material and social deprivation, the cost of travel to receive health care
was not taken into account. A closer look at the direct and indirect costs for patients and their families is needed to establish whether patients at increased distances delay seeking attention for financial reasons. This study is not without limitations. This study did not use individual patient-level data in determining socioeconomic status; rather, it was derived from population data using DA of residence to derive a social and material deprivation index at an aggregate level. Although data at the individual level would potentially be a more accurate measure of socioeconomic status, the deprivation index is a validated tool for estimating socioeconomic status [15]. In conclusion, increased distance from the tertiary cardiac care center was not consistently associated with differences in in-hospital outcomes after cardiac operations. However, distance was uniformly associated with worse 30-day outcomes. These results highlight the important disparities that exist in patient outcomes after cardiac operations based on their geographic place of residence. Although this association may have a clinical basis, further study is required to determine the roles that factors, such as variable physician follow-up and differences in patient behavior, play in promoting this relationship.
Table 5. Logistic Regression of In-Hospital and Composite 30-Day Adverse Outcome on Baseline Clinical Factors and Distance Categories In-Hospital Composite Distance From NBHC 50 km 50–100 km 100–150 km 150–200 km 200–250 km >250 km CI ¼ confidence interval;
30-Day Composite
OR (95% CI)
p Value
OR (95% CI)
p Value
1.00 (reference) 0.92 (0.69–1.24) 0.97 (0.80–1.18) 0.83 (0.68–1.02) 0.86 (0.66–1.11) 0.79 (0.65–0.96)
0.58 0.74 0.08 0.24 0.02
1.00 (reference) 1.16 (0.83–1.62) 1.32 (1.05–1.65) 1.68 (1.33–2.11) 1.41 (1.06–1.88) 1.30 (1.04–1.63)
0.4 0.02 <0.0001 0.02 0.02
NBHC ¼ New Brunswick Heart Center;
OR ¼ odds ratio.
6
COTE ET AL GEOGRAPHY AND OUTCOMES IN CARDIAC OPERATIONS
The authors gratefully acknowledge financial support from the New Brunswick Health Research Foundation and the Heart and Stroke Foundation of New Brunswick. As well, the authors wish to recognize the contribution of Lorelei Churchill, Health & Business Analytics–Decision Support, Horizon Health Network New Brunswick, to the completion of this study.
Ann Thorac Surg 2015;-:-–-
12.
13. 14.
References 1. Fang R, Kmetic A, Millar J, Drasic L. Disparities in chronic disease among Canada’s low-income populations. Prev Chronic Dis 2009;6:A115. 2. Lee DS, Chiu M, Manuel DG, et al. Trends in risk factors for cardiovascular disease in Canada: temporal, sociodemographic and geographic factors. CMAJ 2009;181: E55–66. 3. Alter DA, Naylor CD, Austin P, Tu JV. Effects of socioeconomic status on access to invasive cardiac procedures and on mortality after acute myocardial infarction. N Engl J Med 1999;341:1359–67. 4. Diex Roux AV, Merkin SS, Arnett D, et al. Neighborhood of residence and incidence of coronary heart disease. N Engl J Med 2001;345:99–106. 5. Gregory PM, Malka ES, Kostis JB, Wilson AC, Arora JK, Rhoads GG. Impact of geographic proximity to cardiac revascularization services on service utilization. Med Care 2000;38:45–57. 6. Seidel JE, Ghali WA, Faris PD, et al. Geographical location of residence and uniformity of access to cardiac revascularization services after catheterization. Can J Cardiol 2004;20: 517–23. 7. Vanasse A, Niyonsenga T, Courteau J, Hemiari A. Access to myocardial revascularization procedures: closing the gap with time? BMC Public Health 2006;6:60. 8. Hartford K, Roos LL, Walld R. Regional variation in angiography, coronary artery bypass surgery, and percutaneous transluminal coronary angioplasty in Manitoba, 1987 to 1992: the funnel effect. Med Care 1998;36:1022–32. 9. Hassan A, Pearce NJ, Mathers J, Veugelers PJ, Hirsch GM, Cox JL. The effect of place of residence on access to invasive cardiac services following acute myocardial infarction. Can J Cardiol 2009;25:207–12. 10. Boscarino JA, Chang J. Survival after coronary artery bypass graft surgery and community socioeconomic status: clinical and research implications. Med Care 1999;37:210–6. 11. Ancona C, Agabiti N, Forastiere F, et al. Coronary artery bypass graft surgery: socioeconomic inequalities in access
15. 16.
17. 18. 19. 20. 21. 22. 23.
24.
25.
and in 30 day mortality. A population-based study in Rome, italy. J Epidemiol Community Health 2000;54:930–5. Blais C, Hamel D, Rinfret S. Impact of socioeconomic deprivation and area of residence on access to coronary revascularization and mortality after a first acute myocardial infarction in Qu ebec. Can J Cardiol 2012;28:169–77. Roy A, Na Y. Batch production of driving distances and times using SASÒ and Web Map APIs. Paper 091-2012, SAS Global Forum 2012. Postal codes conversion file (PCCF). Reference guide, catalogue no. 92-153-g, no 02. Ottawa: Minister of Industry, July 2011. Pampalon R, Hamel D, Gamache P, Raymond G. A deprivation index for health planning in Canada. Chronic Dis Can 2009;29:178–91. e publique du Qu ebec (INSPQ). Institut national de sant Deprivation index. Equivalence table between the dissemination/enumeration areas and the different geographic territories, version: March 3, 2010. http://www2.Inspq.Qc.Ca/ santescope/indicedefavoeng.Asp?Noindd¼9&lg¼en. Ferraris VA, Ferraris SP, Harmon RC, Evans BD. Risk factors for early hospital readmission after cardiac operations. J Thorac Cardiovasc Surg 2001;122:278–86. Pitblado JR, Pong RW. Geographic distribution of physicians in Canada. Sudbury, ON: Centre for Rural and Northern Health Research, Laurentian University; 1999. Pong RW, Pitblado JR. Geographic distribution of physicians in Canada: beyond how many and where. Ottawa: Canadian Institute for Health Information; 2005. Tu K, Gong Y, Austin PC, Jaakimanian L, Tu JV. An overview of the types of physicians treating acute cardiac conditions in Canada. Can J Cardiol 2004;20:282–91. Roos NP. Hospitalization style of physicians in Manitoba: the disturbing lack of logic in medical practice. Health Serv Res 1992;27:361. Roos NP, Flowerdew G, Wajda A, Tate RB. Variations in physicians’ hospitalization practices: a population-based study in Manitoba, Canada. Am J Public Health 1986;76:45–51. Taylor RS, Brown A, Ebrahim S, et al. Exercise-based rehabilitation for patients with coronary heart disease: systematic review and meta-analysis of randomized controlled trials. Am J Med 2004;116:682–92. Vanasse A, Courteau J, Cohen A, Orzanco M, Drouin C. Rural-urban disparities in the management and health issues of chronic diseases in Quebec (Canada) in the early 2000s. Rural Remote Health 2010;10:1–15. Harju BL, Wuensch KL, Kuhl EA, Cross NJ. Comparison of rural and urban residents’ implicit and explicit attitudes related to seeking medical care. J Rural Health 2006;22: 359–63.