- Email: [email protected]
Insurance Status and Inequalities in Outcomes After Neurosurgery
PEER-REVIEW REPORTS
Insurance Status and Inequalities in Outcomes After Neurosurgery Abdulrahman M. El-Sayed1,2,3, John E. Ziewacz4,5, Matthew C. Davis6, Darryl Lau6, Hasan K. Siddiqi 6, Grettel J. Zamora-Berridi6, Stephen E. Sullivan4
Key words 䡲 Disparities 䡲 Inequalities 䡲 Insurance 䡲 Neurosurgery 䡲 Outcomes 䡲 Socioeconomic status Abbreviations and Acronyms ANCOVA: Analysis of covariance ANOVA: Analysis of variance ICU: Intensive care unit From the 1Department of Public Health, Oxford University, Oxford, United Kingdom; 2Department of Epidemiology, Columbia University, New York, New York, USA; 3College of Physicians and Surgeons, Columbia University, New York, New York, USA; 4Department of Neurosurgery, University of Michigan, Ann Arbor, Michigan, USA; 5Center for Surgery and Public Health, Harvard School of Public Health and Brigham and Women’s Hospital, Department of Surgery, Boston, Massachusetts, USA; and 6University of Michigan Medical School, Ann Arbor, Michigan, USA To whom correspondence should be addressed: Abdulrahman M. El-Sayed, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2011) 76, 5:459-466. DOI: 10.1016/j.wneu.2011.03.051
䡲 OBJECTIVE: Little is known about socioeconomic differences in postoperative outcomes after neurosurgery. We assessed the relation between insurance status and postoperative complication risk, neurosurgical intensive care unit stay, and hospital stay after neurosurgery. 䡲 METHODS: We collected data on 918 consecutive craniotomy or spine-related neurosurgical cases in patients at least 18 years of age at the University of Michigan Hospitals after April 2006. Bivariate 2 tests and analysis of variance were used to assess bivariate relations, and multivariable logistic regression models and analysis of covariance were used to adjust for potential confounders. 䡲 RESULTS: A total of 11.2% of privately insured patients, 23.6% of Medicare patients, 25.8% of Medicaid patients, and 27.3% of uninsured patients suffered complications within 30 days of surgery (P < 0.001). In adjusted models, odds of postoperative complications among Medicare (odds ratio [OR] ⴝ 2.1, 95% confidence interval [CI] 1.3–3.3), Medicaid (OR ⴝ 3.1, 95% CI 1.5– 6.1), and uninsured patients (OR ⴝ 3.6. 95% CI 1.3–10.3) were higher than among privately insured patients. By analysis of covariance, only Medicaid patients had significantly longer intensive care unit (P ⴝ 0.040) and hospital stays (P ⴝ 0.028) than privately insured patients. 䡲 CONCLUSIONS: Our findings suggest important socioeconomic disparities in outcomes after neurosurgical intervention. Access to postoperative outpatient care may mediate our findings.
Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter © 2011 Elsevier Inc. All rights reserved.
INTRODUCTION Disparities in both the quantity and quality of health services used among rich and poor are well documented in the health services literature (11, 12). Insurance status, a pivotal indicator of socioeconomic status with respect to health, has been demonstrated as an important predictor of several diverse and important health metrics in the United States, including cancer screening (25) and diagnosis (10, 16), treatment for heart disease (1), treatment and mortality after trauma (15), as well as overall mortality (13), with the uninsured and patients covered by Medicaid suffering worse outcomes. Of particular interest here is the consistent, well-documented relation between insurance status and outcomes after surgical
intervention (5, 7, 17, 21). For example, a recent study by LaPar et al. (21), using data on almost 900,000 patients undergoing diverse surgical procedures from the National Inpatient Sample, showed that mortality rates among patients on Medicaid and Medicare, as well as uninsured patients after surgery were significantly higher than among privately insured patients, and uninsured patients suffered the highest rates of mortality after surgery after adjusting for potential confounders. In addition, patients covered by Medicaid stayed longest in hospital and had the highest total costs associated with their care (21). The literature about socioeconomic disparities in outcomes after neurosurgery is limited. One study considered differences in metrics of postoperative outcomes among neurosurgery patients by insurance status. In a national sample of 99,665 pa-
WORLD NEUROSURGERY 76 [5]: 459-466, NOVEMBER 2011
tients undergoing craniotomy procedures between 1988 and 2004 from the National Inpatient Sample, Curry et al. (7) showed that patients paying for services by Medicaid had the highest rates of in-hospital mortality postoperatively, whereas privately insured patients had the lowest. Other studies have found similar disparities in outcomes by insurance status after treatment for intracranial aneurysms (17) and spinal fusion (5). Little is known about differences in risk for postoperative complications and/or hospital or neurosurgical intensive care unit (ICU) stays by insurance status among neurosurgical patients. Given the growth in the scale and scope of neurosurgical intervention, a more nuanced and systematic understanding of the socioeconomic determinants of postoperative complication and recovery is in order. Such an understanding
www.WORLDNEUROSURGERY.org
459
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
might empower policy-makers and clinicians to address these disparities, and prepare for and improve postoperative care. Using data about 918 consecutive spine and craniotomy neurosurgical cases at the University of Michigan Hospitals, we assessed the relation between insurance status and risk for 30-day postoperative complications, duration of ICU stay, and duration of hospital stay.
METHODS Data The University of Michigan Evidence-Based Neurosurgery Database collected retrospective data about 918 consecutive craniotomy or spine-related neurosurgical cases at the University of Michigan Hospitals from April 10, 2006 through May 4, 2009. Data were collected by trained medical students from full clinical registers in the electronic medical record system from the University of Michigan Health System. Patients younger than 18 years old at the time of operation were excluded from the database. Patients lacking 30-day follow-up, those with incomplete medical records, undergoing only ventriculostomy, head or neck cases that did not include craniotomy, or neuroendovascular procedures were excluded. From among 1331 patients originally enrolled, only 918 were suitable for inclusion in our analysis. Data were collected regarding patient demographics, comorbidities, postoperative complications, length of ICU stay, and length of hospital stay from neurosurgery clinical notes found in the electronic medical record system of the University of Michigan Hospitals. Data for explanatory covariates were collected regarding insurance status (analyzed as a categorical variable: private, Medicaid, Medicare, or none); gender; marital status (analyzed as a binary variable denoting the presence or absence of a spouse at time of operation); age (analyzed as a categorical variable: ⬍50, 50 –70, and 71 years or older); current tobacco and/or current alcohol use (both analyzed as binary variables: presence vs. absence); and case type (analyzed as a binary variable: spinal or craniotomy). Data about comorbidities were also collected, including obesity (body mass index ⬎30 kg/m2), diabetes mellitus, systemic hypertension, and/or
460
www.SCIENCEDIRECT.com
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
coronary artery disease (all analyzed as binary variables: presence vs. absence). Complications were defined as the occurrence of any of the following within 30 days of operation: death; coma ⬎24 hours; acute renal failure; postoperative bleeding requiring ICU stay, reoperation, or requiring more than four units of red blood cells within 72 hours; unplanned intubation; ventilation lasting more than 48 hours; pneumonia; cardiac arrest; myocardial infarction; pulmonary embolus; deep infection; sepsis; systemic inflammatory response syndrome; pseudomeningocele; deep vein thrombosis; seizure; and cerebrovascular accident (complications were analyzed as one binary variable denoting the presence or absence of any complication). Both length of ICU stay and total length of hospital stay were recorded in days for each patient, and were analyzed as continuous variables. This study was reviewed by the Medical Science Institutional Review Board of the University of Michigan. Analysis Because one of our outcomes of interest was categorical (i.e., complications) and the others were continuous (i.e., duration of hospital and ICU stay), we used multiple statistical analytical tools. We used bivariate 2 tests and multivariable logistic regression to assess relations between predictor covariates and categorical outcomes, and we used analysis of variance (ANOVA) and analysis of covariance (ANCOVA) to assess bivariate and multivariate relations, respectively, between predictor covariates and continuous outcomes. First, univariate statistics were calculated to describe our sample. Second, we used bivariate 2 tests to identify significant associations between explanatory covariates and insurance status, as well as between covariates of interest and categorical outcomes. Third, we used ANOVA to assess significant associations between explanatory covariates of interest and continuous outcomes. Fourth, a multivariable logistic regression model of 30-day complication risk by gender was fit, and was adjusted for potential confounders. We adjusted for covariates found to be significantly associated with postoperative complications in bivariate 2 tests. Fifth, ANCOVA models of ICU stay, as well as hospital stay by insurance
status were fit, and were adjusted for those covariates found to be significantly associated with the outcome of interest in ANOVA models. Associations with P ⬍ 0.05 were regarded as significant. SAS 9.2 (SAS Institute; Cary, North Carolina, USA) was used for all statistical analyses.
RESULTS Table 1 shows descriptive statistics and bivariate 2 tests between insurance status, demographic, and medical covariates. Insurance status was associated with marital status (P ⬍ 0.001), and those on private insurance and Medicare were significantly more likely to be married than their counterparts on Medicaid and without insurance. Insurance status was also associated with age (P ⬍ 0.001). Medicare patients and those with private insurance were older than their counterparts on Medicaid and without insurance. There was also a relationship between insurance status and tobacco use (P ⫽ 0.041), as patients on Medicaid were more likely to use tobacco than other groups. Insurance status also predicted case type (P ⬍ 0.001) (craniotomy vs. spinal cases), and those without insurance were more likely to undergo craniotomies than were other groups. Last, there were significant relationships between insurance status and hypertension (P ⬍ 0.001), coronary artery disease (P ⬍ 0.001), and diabetes mellitus (P ⬍ 0.001), and Medicare patients had higher risk for each of these comorbidities than other patient groups. Table 2 shows demographic characteristics and bivariate 2 tests between all covariates and 30-day complication risk. Of 918 patients in our study, 145 (15.8%) experienced a major complication within 30 postoperative days. Insurance status was associated with 30-day postoperative complication risk (P ⬍ 0.001). Among privately insured patients, 11.2% suffered a major complication, which was lower than among patients on Medicare (23.6%), Medicaid (25.8%), and uninsured patients (27.3%). Male gender (P ⬍ 0.001), older age (P ⫽ 0.006), craniotomy case type (P ⫽ 0.002), and emergent admissions status (P ⬍ 0.001), were all associated with postoperative complication risk relative to female gender, younger age, spinal case type, and elective admissions status, respectively. Among medical co-
WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2011.03.051
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
Table 1. Descriptive Statistics and Bivariate 2 Tests Between Insurance Status and Demographic and Medical Covariates Among 918 Patients Undergoing Neurosurgical Interventions at the University of Michigan Hospitals Insurance Status Patient Descriptives
Private
Medicare
Medicaid
None
Male
52.9
52.4
48.5
72.7
Female
47.1
47.6
51.5
27.3
No
32.4
36.8
74.2
68.2
Yes
67.6
63.2
25.8
31.8
⬍50 years
52.6
14.7
83.3
68.2
50–70 years
44.9
48.0
12.1
31.8
⬎70 years
2.5
37.3
4.6
0.0
No
59.6
57.2
40.6
59.1
Yes
40.4
42.8
59.4
40.9
No
63.3
66.1
57.8
54.6
Yes
36.7
33.9
42.2
45.4
Craniotomy
69.9
57.8
74.2
45.5
Spinal case
30.1
42.2
25.8
54.5
Emergent
10.5
12.4
16.7
13.6
Elective
89.5
87.6
83.3
86.4
No
66.7
64.9
69.7
68.2
Yes
33.3
35.1
30.3
31.8
P Value
Total Gender
0.089
⬍ 0.001
Married
⬍ 0.001
Age
Tobacco use
0.041
Alcohol use
0.516
Case type
0.001
Admission status
0.492
Obesity*
0.895
⬍0.001
Hypertension No
65.2
44.4
78.8
81.8
Yes
34.8
55.6
21.2
18.2
DISCUSSION ⬍0.001
Coronary artery disease No
95.7
80.4
95.4
100.0
Yes
4.3
19.6
4.6
0 ⬍0.001
Diabetes mellitus No
91.8
81.3
97.0
95.5
Yes
8.2
18.7
3.0
4.5
*Body mass index, ⬎30 kg/m . 2
morbidities, only coronary artery disease was associated with 30-day postoperative complication risk (P ⬍ 0.001). Table 3 shows mean postoperative hos-
pital and ICU stay, as well as ANOVA between explanatory covariates of interest and both continuous outcomes of interest among patients in our sample. Mean length
WORLD NEUROSURGERY 76 [5]: 459-466, NOVEMBER 2011
of stay in the ICU was 2.0 days (standard deviation 4.6 days), whereas the mean length of stay in the hospital was 6.6 days (standard deviation 9.4 days). Insurance status was not associated with length of ICU stay (P ⫽ 0.113). Craniotomy case type (relative to spinal case type) (P ⬍ 0.001), emergent admissions status (relative to elective) (P ⬎ 0.001), and obesity (P ⫽ 0.006) were the only predictors of length of postoperative ICU stay. Insurance status was associated with length of hospital stay (P ⫽ 0.020), along with male gender (relative to female gender) (P ⬍ 0.003), craniotomy case type (P ⬍ 0.001), emergent admissions status (P ⬍ 0.001), and coronary artery disease (P ⫽ 0.003). Table 4 shows a multivariable logistic regression model of 30-day postoperative complication risk by insurance status, adjusted for potential confounders. Compared with privately insured patients, odds of postoperative complications among patients on Medicare were 2.1 (95% confidence interval 1.3–3.3), odds among patients on Medicaid were 3.1 (95% confidence interval 1.5– 6.1), and odds among patients without insurance were 3.6 (95% confidence interval 1.3–10.3). Table 5 shows ANCOVA models of postoperative ICU stay and hospital stay by insurance status adjusted for potential confounders. Medicaid was a significant predictor of length of postoperative ICU stay (P ⫽ 0.031) and hospital stay (P ⫽ 0.028) relative to private insurance, but neither patients on Medicare nor those without insurance differed significantly from privately insured patients in either duration of ICU stay or duration of hospital stay.
In a study of 918 consecutive patients at a Midwestern teaching hospital, we found that relative to privately insured patients, patients on public insurance plans (Medicare or Medicaid) and uninsured patients had significantly increased risks for postoperative complications within 30 days of neurosurgery. Uninsured patients had the worst outcomes, followed by those on Medicaid, and then by those on Medicare. In addition, patients on Medicaid required significantly longer ICU and hospital stays than their privately insured counterparts. Our findings suggest important socioeco-
www.WORLDNEUROSURGERY.org
461
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
Table 2. Descriptive Statistics and Bivariate 2 Tests Between Demographic and Medical Covariates and Postoperative Complication Risk Among 918 Patients Undergoing Neurosurgical Interventions at the University of Michigan Hospitals 30-Day Complications Patient Descriptives
N
%
N
%
918
—
145
15.8
Private
439
58.4
49
11.2
Medicare
225
29.9
53
23.6
Medicaid
66
8.8
17
25.8
None
22
2.9
6
27.3
Total
⬍ 0.001
Insurance status
⬍ 0.001
Gender Male
459
50.0
93
20.3
Female
459
50.0
52
11.3
No
353
38.7
60
17.1
Yes
559
61.3
83
14.9
⬍50 years
409
44.6
48
11.7
50–70 years
388
42.3
70
18.0
⬎70 years
121
13.2
27
22.3
Married
0.550
Age
0.006
Tobacco use
0.651
No
525
57.8
79
15.1
Yes
384
42.2
62
16.2
No
574
63.2
90
15.7
Yes
334
36.8
51
15.3
Craniotomy
586
63.8
109
18.6
Spinal case
332
36.2
36
10.8
Emergent
103
11.2
45
43.7
Elective
814
88.8
100
12.3
No
610
66.5
91
14.9
Yes
308
33.6
54
17.5
No
546
59.5
78
14.3
Yes
371
40.5
67
18.1
No
835
91.1
114
13.7
Yes
82
8.9
31
37.8
816
89.0
124
15.2
101
11.0
21
20.8
Alcohol use
0.869
Case type
0.002
⬍ 0.001
Admission status
Obesity*
0.305
Hypertension
0.124
⬍ 0.001
Coronary artery disease
Diabetes mellitus
0.146
No Yes *Body mass index, ⬎30 kg/m . 2
462
P Value
www.SCIENCEDIRECT.com
nomic disparities in outcomes after neurosurgical intervention. Our findings are supported by the several other studies, which have considered differences in outcomes after neurosurgical intervention by insurance status (5, 7, 17). As mentioned previously, Curry et al. (7) found that patients paying for neurosurgery by Medicaid suffered the highest rates of inhospital mortality postoperatively, whereas privately insured patients suffered the lowest, and patients on Medicare as well as those without insurance suffered similar mortality rates, falling between those on Medicaid and the privately insured. Our findings affirm that privately insured patients had the lowest risk for postoperative complications. Although we found that uninsured patients had the highest risk for postoperative complications, rather than patients covered by Medicaid, as demonstrated in the study by Curry et al. (7), there were no significant differences in postoperative complication risk between patients on Medicaid, Medicare, or the uninsured in our analysis. Similar findings have also been demonstrated in outcomes after specific neurosurgical interventions (5, 17). Our findings may also be framed within the broader context of the literature about insurance status and postoperative outcomes among other surgical subspecialties. For example, our findings showed that patients covered by Medicaid had the longest postoperative ICU and surgical stays, which were significantly longer than those of privately insured patients. These findings are supported by work by LaPar et al. (21), reviewed previously, who showed that surgical patients on Medicaid had the longest postoperative hospital stays. In addition, the literature about differences in postoperative health metrics by insurance status among surgical patients has generally supported our findings that uninsured patients, as well as those covered by Medicare or Medicaid, suffer worse outcomes than their privately insured counterparts (5, 7, 17, 20-22). There are several plausible explanations for the disparities in postoperative outcomes by insurance status we observed here. First, by virtue of their insurance status, underinsured (Medicare and Medicaid) (9) and uninsured patients lack quality access to adequate primary health care (2). Hence, pathology among these groups is
WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2011.03.051
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
Table 3. Analysis of Variance Between Demographic and Medical Covariates and Postoperative Hospital and Intensive Care Unit Stay Among 918 Patients Undergoing Neurosurgical Interventions at the University of Michigan Hospitals Intensive Care Unit Stay Patient Descriptives Total
Days
SD
2.0
4.6
Insurance status
P Value
Hospital Stay Days
SD
6.6
9.4
0.113
P Value
Table 4. Logistic Regression Models of Complication Risk by Insurance Status Adjusted for Potential Confounders Among 918 Patients Undergoing Neurosurgical Interventions at the University of Michigan Hospitals Patient Descriptives
OR
95% CI
0.020 Insurance status
Private
1.8
4.0
5.7
8.1
Medicare
1.9
4.5
7.6
11.5
Private
ref
ref
Medicaid
3.1
4.9
8.6
9.4
Medicare
2.1
1.3–3.3
None
2.5
3.7
7.6
11.6
Medicaid
3.1
1.5–6.1
None
3.6
1.3–10.3
Male
1.6
1.1–2.4
Female
ref
ref
⬍50 years
ref
ref
50–70 years
1.8
1.1–2.8
⬎70 years
1.0
0.5–2.1
Craniotomy
1.8
1.2–2.8
ref
ref
Emergent
5.3
3.3–8.6
Elective
ref
ref
No
ref
ref
Yes
3.3
1.8–5.8
Gender
0.182
0.003
Male
2.2
5.0
7.5
10.8
Female
1.8
4.2
5.7
7.8
Married
0.592
0.086
No
2.0
4.3
7.2
10.2
Yes
1.9
4.6
6.1
8.8
Age ⬍50 years
0.292 1.7
3.8
50–70 years
2.1
5.1
⬎70 years
2.3
5.1
Tobacco use
Gender
Age 0.118
5.9
8.1
7.0
10.8
7.7
8.7
0.85
Case type 0.425
No
2.0
4.6
6.7
8.9
Spinal case
Yes
1.9
4.5
6.2
9.9
Admission status
No
2.0
4.9
6.4
8.7
Yes
1.9
4.1
6.7
10.4
Alcohol use
0.76
0.664
⬍ 0.001
Case type Craniotomy
2.7
5.3
7.7
10.6
Spinal case
0.7
2.6
4.8
6.6
⬍ 0.001
Admission status Emergent
6.3
8.2
Elective
1.4
3.6
Obesity*
⬍ 0.001 14.2
13.6
5.6
8.3
0.006
0.214
No
2.3
5.1
6.9
9.9
Yes
1.4
3.2
6.1
8.3
6.2
9.6
7.2
9.2
Hypertension
0.345
No
1.9
4.1
Yes
2.2
5.3
Coronary artery disease
0.109
0.078
0.003
No
1.9
4.4
6.3
9.3
Yes
2.8
5.8
9.6
10.3
No
2.0
4.6
6.5
9.5
Yes
2.0
4.8
7.4
8.8
Diabetes mellitus
Coronary artery disease ⬍ 0.001
0.889
0.359
SD, standard deviation. *Body mass index, ⬎30 kg/m2.
WORLD NEUROSURGERY 76 [5]: 459-466, NOVEMBER 2011
OR, odds ratio; 95% CI, 95% confidence interval; ref, referent.
likely to be more advanced at the time of presentation (3, 4, 21). Because more developed pathology is often more challenging to treat, and is associated with worse prognoses (3), differential access to primary health care may explain the differences in outcomes among neurosurgical patients by insurance status. Second, evidence suggests that distinct socioeconomic gradients in population health metrics persist independently of differences in access to health services (18, 28). For example, despite a universal health care system in the United Kingdom, there remain consistent socioeconomic gradients in several health metrics (8, 24). As insurance status is a reliable proxy for overall socioeconomic status, underinsured and
www.WORLDNEUROSURGERY.org
463
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
Table 5. Analysis of Covariance of Postoperative Hospital and Intensive Care Unit Stay by Insurance Status Adjusted for Potential Confounders Among 918 Patients Undergoing Neurosurgical Intervention at the University of Michigan Hospitals Intensive Care Unit Stay Patient Descriptives
Days
SD
Private
1.8
4.0
Medicare
1.9
4.5
Medicaid
3.1
4.9
None
2.5
3.7
P Value
Hospital Stay Days
SD
P Value
5.7
8.1
0.621
7.6
11.5
0.054
0.031
8.6
9.4
0.028
0.504
7.6
11.6
0.391
Insurance status
Gender
0.003
Male
7.5
10.8
Female
5.7
7.8
7.7
10.6
4.8
6.6
⬍0.001
Case type Craniotomy
2.7
5.3
Spinal case
0.7
2.6
⬍0.001
⬍0.001
Admission status
⬍0.001
Emergent
6.3
8.2
14.2
13.6
Elective
1.4
3.6
5.6
8.3
No
2.3
5.1
Yes
1.4
3.2
Obesity*
0.007
Coronary artery disease
0.016
No
6.3
9.3
Yes
9.6
10.3
SD, standard deviation. *Body mass index, ⬎30 kg/m2.
uninsured patients may be systematically more deprived than their privately insured counterparts. And as socioeconomic inequalities in health metrics persist independently of access to care, this population may have systematically worse baseline health than privately insured patients. Because worse baseline health is a demonstrated predictor of poor surgical outcomes (19, 29), it is plausible that differences in postoperative outcomes by insurance status may result from a priori differences in baseline health between privately insured and underand uninsured populations. Third, it is plausible that underinsured and uninsured patients, even after surgical intervention, may lack the coverage to support postoperative care after discharge. With compromised postsurgical care, it is plausible that these patients would suffer worse outcomes compared with privately
464
www.SCIENCEDIRECT.com
insured patients. In support of this explanation, several studies have demonstrated financial barriers to postoperative outpatient care among neurosurgical patients (14, 26). For example, one study by Gerszten et al. (14) showed that among 91 patients, uninsured patients had to wait significantly longer for placement into rehabilitation services than insured patients. Differences in access to outpatient postoperative care may also indirectly explain observed differences in the durations of ICU and hospital stays among Medicaid patients compared with others. It is possible that clinicians may purposely keep their Medicaid-covered patients in hospitalized care, where their expenses are more likely to be reimbursed by Medicaid, rather than discharging them where they may face more systematic financial barriers to outpatient care.
There are several limitations to our findings of which the reader should be aware. First, although a strong proxy for socioeconomic status at the population level, insurance status among some patients may not reliably approximate socioeconomic status (e.g., uninsured young people who choose not to purchase health insurance). In addition, this metric is partially age dependent (e.g., Medicare patients are, by definition, 65 years or older), and therefore, associations between insurance status and health may be confounded by age. Second, as with any observational analysis, although we adjusted for covariates that could potentially confound the association between insurance status and outcomes of interest, it is plausible that the relation remains residually confounded by unmeasured factors that are associated with both exposures and outcomes of interest. Third, our findings were derived from a sample that was bounded by time and space, and therefore, they may not generalize to other temporal and spatial contexts. In addition, our sample was limited to patients at a high-volume academic tertiary care center. Because outcomes have been shown to be better in these care contexts compared to low-volume centers (6, 27), and underinsured and uninsured patients may be more likely to seek care at low-volume centers (23), it is possible that our findings may underestimate the true disparities in neurosurgical outcomes nationally. Fourth, our sample included only patients undergoing a limited spectrum of neurosurgical operations, and therefore, our findings may not generalize to other patient populations. Fifth, we only considered outcomes within 30 days of operation, further limiting the generalizability of these findings. Despite these limitations, we believe that our observations have important implications for health policy, neurosurgical practice, and future research in this area. With regard to health policy, our findings suggest that the health effects of insurance extend beyond access to care—the population sampled in the present study had received neurosurgical care, yet there were distinct inequalities in outcomes by insurance status. Therefore, policy-makers intent on mitigating socioeconomic disparities in health metrics should consider means of health improvement, adjuvant to improvements in health access among the poor,
WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2011.03.051
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
which are independent of the formal health system. Interventions that focus on prevention alongside improved access to curative care may be in order. Second, alongside a growing literature that has demonstrated disparities in access to postoperative outpatient services among neurosurgical patients (14, 26), our findings suggest that barriers to postoperative outpatient care may compromise the neurosurgical outcomes of underinsured and uninsured patients. Therefore, policy-makers should seek to address these barriers so as to improve outcomes among the most deprived neurosurgical patients. Third, our findings demonstrated that Medicaid users required significantly longer hospital and neurosurgical ICU stays than their privately insured counterparts. We were unable to explain this troubling finding. However, postoperative neurosurgical care is costly, and a systematic requirement for greater care among publicly relative to privately insured patients, suggests inefficiencies in the public system. Policymakers may seek to address these differences by improving access to primary care, postoperative outpatient services, and/or improving baseline health metrics among Medicaid patients. With regard to clinical care, it is important that clinicians recognize and actively seek to remedy the outcome disparities we have noted. Although it is improbable that there are intraoperative differences in care that mediate the disparities that we have observed, differences in postoperative care may very well mediate this relation. In addition, our findings demonstrated higher risks for postoperative complications among underinsured and uninsured patients. Therefore, astute clinicians should work with these patients to negotiate barriers in access to quality postoperative care to assure optimal outcomes, as well as maintain a particular vigilance and attentiveness to signs and symptoms of postoperative complications among this population. Finally, our findings also have several implications for future research in this area. First, multicenter prospective studies about the relations between socioeconomic metrics and neurosurgical outcomes are needed to clarify the etiologies of differences in postoperative outcomes by insurance status. Such studies might consider differences in care trajectories after hospital discharge, as well as a more nuanced explo-
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
ration of differences in baseline preoperative health among privately insured and underinsured and uninsured patients. Second, investigators interested in inequalities in neurosurgical outcomes would be well served to consider relations between other metrics of socioeconomic status, such as education, income, and contextual deprivation, as well as race and ethnicity and neurosurgical outcomes.
CONCLUSIONS Our findings suggest that patients covered by Medicare and Medicaid, as well as uninsured patients, may have worse outcomes after neurosurgical intervention than privately insured patients. Differences in baseline health, access to primary care, or access to postoperative outpatient care may mediate the association between insurance status and postneurosurgical outcomes. Clinicians should work with underinsured and uninsured patients to negotiate barriers in access to outpatient care and pay particular attention to signs and symptoms of postoperative complication among this population.
REFERENCES
in the United States, 1988 –2004. Neurosurgery 66:427-438, 2010. 8. Davey Smith G, Dorling D, Mitchell R, Shaw M: Health inequalities in Britain: continuing increases up to the end of the 20th century. J Epidemiol Community Health 56:434-435, 2002. 9. Farley PJ: Who are the underinsured? Milbank Mem Fund Q Health Soc 63:476-503, 1985. 10. Farley TA, Flannery JT: Late-stage diagnosis of breast cancer in women of lower socioeconomic status: public health implications. Am J Public Health 79:1508-1512, 1989. 11. Fiscella K, Franks P, Doescher M, Saver BG: Disparities in health care by race, ethnicity, and language among the insured: findings from a national sample. Med Care 40:52-59, 2002. 12. Fiscella K, Franks P, Gold MR, Clancy CM: Inequality in quality: addressing socioeconomic, racial, and ethnic disparities in health care. JAMA 283:25792584, 2000. 13. Franks P, Clancy CM, Gold MR: Health insurance and mortality. Evidence from a national cohort. JAMA 270:737-741, 1993. 14. Gerszten PC, Witham TF, Clyde BL, Welch WC: Relationship between type of health insurance and time to inpatient rehabilitation placement for surgical subspecialty patients. Am J Med Qual 16:212-215, 2001. 15. Haas JSP, Goldman L: Acutely injured patients with trauma in Massachusetts: differences in care and mortality, by insurance status. Am J Public Health 84:1605-1608, 1994.
1. Ayanian JZ, Epstein AM: Differences in the use of procedures between women and men hospitalized for coronary heart disease. N Engl J Med 325:221225, 1991.
16. Halpern MT, Bian J, Ward EM, Schrag NM, Chen AY: Insurance status and stage of cancer at diagnosis among women with breast cancer. Cancer 110:403411, 2007.
2. Berk ML, Schur CL: Access to care: how much difference does Medicaid make? Health Aff 17:169-180, 1998.
17. Hoh BL, Rabinov JD, Pryor JC, Carter BS, Barker FG 2nd: In-hospital morbidity and mortality after endovascular treatment of unruptured intracranial aneurysms in the United States, 1996 –2000: effect of hospital and physician volume. AJNR Am J Neuroradiol 24:1409-1420, 2003.
3. Bradley CJ, Given CW, Dahman B, Luo Z, Virnig BA: Diagnosis of advanced cancer among elderly Medicare and Medicaid patients. Med Care 45:410-419, 2007. 4. Bradley C, Given C, Roberts C: Late stage cancers in a Medicaid-insured population. Med Care 41:722728, 2003. 5. Cahill KS, Chi JH, Day A, Claus EB: Prevalence, complications, and hospital charges associated with use of bone-morphogenetic proteins in spinal fusion procedures. JAMA 302:58-66, 2009. 6. Cowan JA Jr, Dimick JB, Leveque JC, Thompson BG, Upchurch GR Jr, Hoff JT: The impact of provider volume on mortality after intracranial tumor resection. Neurosurgery 52:48-54, 2003. 7. Curry WT Jr, Carter BS, Barker FG 2nd: Racial, ethnic, and socioeconomic disparities in patient outcomes after craniotomy for tumor in adult patients
WORLD NEUROSURGERY 76 [5]: 459-466, NOVEMBER 2011
18. Katz SJ, Hofer TP: Socioeconomic disparities in preventive care persist despite universal coverage: breast and cervical cancer screening in Ontario and the United States. JAMA 272:530-534, 1994. 19. Katz J, Stucki G, Lipson S, Fossel AH, Grobler LJ, Weinstein JN: Predictors of surgical outcome in degenerative lumbar spinal stenosis. Spine 24: 2229-2233, 1999. 20. Kelz RR, Gimotty PA, Polsky D, Norman S, Fraker D, DeMichele A: Morbidity and mortality of colorectal carcinoma surgery differs by insurance status. Cancer 101:2187-2194, 2004. 21. LaPar DJ, Bhamidipati CM, Mery CM, Stukenborg GJ, Jones DR, Schirmer BD, Kron IL, Ailawadi G: Primary payer status affects mortality for major surgical operations. Ann Surg 252:544-551, 2010.
www.WORLDNEUROSURGERY.org
465
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
22. Lidor AO, Gearhart SL, Wu AW, Chang DC: Effect of race and insurance status on presentation, treatment, and mortality in patients undergoing surgery for diverticulitis. Arch Surg 143:1160-1165, 2008. 23. Liu JH, Zingmond DS, McGory ML, SooHoo NF, Ettner SL, Brook RH, Ko CY: Disparities in the utilization of high-volume hospitals for complex surgery. JAMA 296:1973-1980, 2006. 24. Marmot MG, Smith GD, Stansfield S, Patel C, North F, Head J, White I, Brunner E, Feeney A: Health inequalities among British civil servants: the Whitehall II study. Lancet 337:1387-1393, 1991. 25. Rodriguez MA, Ward LM, Perez-Stable EJ: Breast and cervical cancer screening: impact of health insurance status, ethnicity, and nativity of Latinas. Ann Fam Med 3:235-241, 2005.
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
26. Taylor L, Simpson K, Bushardt R, Reeves C, Elkin TD, Fortson BL, Boll T, Patel S: Insurance barriers for childhood survivors of pediatric brain tumors: the case for neurocognitive evaluations. Pediatr Neurosurg 42:223-227, 2000. 27. Solomon RA, Mayer SA, Tarmey JJ: Relationship between the volume of craniotomies for cerebral aneurysm performed at New York state hospitals and in-hospital mortality. Stroke 27:13-17, 1996. 28. Veugelers PJ, Yip AM: Socioeconomic disparities in health care use: does universal coverage reduce inequalities in health? J Epidemiol Community Health 57:424-428, 2003.
propriateness, outcomes, and risk factors for complications. Stroke 28:891-898, 1997.
Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. received 2 December 2010; accepted 31 March 2011 Citation: World Neurosurg. (2011) 76, 5:459-466. DOI: 10.1016/j.wneu.2011.03.051 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter © 2011 Elsevier Inc. All rights reserved.
29. Wong JH, Findlay JM, Suarez-Almazor ME: Regional performance of carotid endarterectomy: ap-
Santiago Ramón y Cajal and Harvey Cushing: Two Forefathers of Neuroscience and Neurosurgery Grettel J. Zamora-Berridi2, Courtney Pendleton1, Gabriel Ruiz3, Aaron A. Cohen-Gadol4, Alfredo Quiñones-Hinojosa1
Key words 䡲 Cajal 䡲 Cushing 䡲 Neuroscience 䡲 Neurosurgery 䡲 Physician-scientist Abbreviations and Acronyms CNS: Central nervous system ICP: Intracranial pressure From the 1Department of Neurosurgery, Brain Tumor Stem Cell Laboratory, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA; 2University of Michigan Medical School, Ann Arbor, Michigan, USA; 3 Departamento de Psicología Experimental, Universidad de Sevilla, Sevilla, Spain; and 4Clarian Neuroscience Institute, Indianapolis Neurosurgical Group and Indiana University Department of Neurosurgery, Indianapolis, Indiana, USA
䡲 OBJECTIVE: To summarize the extraordinary accomplishments, and the commonalities, between Santiago Ramon y Cajal and Harvey Williams Cushing. 䡲 METHODS: Existing literature describing the lives and achievements of Ramón y Cajal and Cushing, as well as personal communication, and the surgical records of the Johns Hopkins Hospital, from 1896 to 1912, were reviewed. 䡲 RESULTS: Both Ramón y Cajal and Cushing were men of unusually broad interests and talents, and these shared characteristics undoubtedly influenced the career paths and scientific investigations they pursued. Although Santiago Ramón y Cajal and Harvey Williams Cushing never directly interacted, the links between them can be traced through some of their disciples, including Pío del Río Hortega, Wilder Penfield, and Percival Bailey. 䡲 CONCLUSIONS: Ramón y Cajal and Cushing are widely considered the forefathers of neuroscience and neurosurgery, respectively, and their discoveries have made lasting impressions on both the scientific and medical communities.
To whom correspondence should be addressed: Alfredo Quiñones-Hinojosa, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2011) 76, 5:466-476. DOI: 10.1016/j.wneu.2011.04.001 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter © 2011 Elsevier Inc. All rights reserved.
INTRODUCTION During the past century many people involved in science and medicine have been fascinated by the lives of both Santiago Ramón y Cajal (May 1, 1852–October 17, 1934) and Harvey Cushing (April 8, 1869 –
466
www.SCIENCEDIRECT.com
October 7, 1939). Much of our present-day knowledge, and the advancements made in both fields, are deeply rooted in the legacies left by Cajal and Cushing. Many historic events and scientific discoveries in the field of neuroscience preceded both Ramón y Cajal and Cushing. Such events include the articulation of the cell theory by Theodor Schwann and Matthias Schleiden in the late 1830s (90, 91), the depiction of the cell as an independent unit by Rudolf Virchow in 1855 (99), the first original descriptions of the central nervous
system by Rudolph Albert von Kölliker (22, 49, 93) and Jan Evangelista Purkinje (66), and the founding of the field of neurophysiology by Charles Sherrington (94). Likewise, the emphasis on patient-based learning, translational research, and meticulous technique by William Welch, Sir William Osler, and William Stuart Halsted, revolutionized clinical practice and medical training (11, 14, 29, 55). These connections between bench and bedside, coupled with the burgeoning field of neuroscience research, allowed both Cushing and Ramón y Cajal to
WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2011.04.001
Insurance Status and Inequalities in Outcomes After Neurosurgery Abdulrahman M. El-Sayed1,2,3, John E. Ziewacz4,5, Matthew C. Davis6, Darryl Lau6, Hasan K. Siddiqi 6, Grettel J. Zamora-Berridi6, Stephen E. Sullivan4
Key words 䡲 Disparities 䡲 Inequalities 䡲 Insurance 䡲 Neurosurgery 䡲 Outcomes 䡲 Socioeconomic status Abbreviations and Acronyms ANCOVA: Analysis of covariance ANOVA: Analysis of variance ICU: Intensive care unit From the 1Department of Public Health, Oxford University, Oxford, United Kingdom; 2Department of Epidemiology, Columbia University, New York, New York, USA; 3College of Physicians and Surgeons, Columbia University, New York, New York, USA; 4Department of Neurosurgery, University of Michigan, Ann Arbor, Michigan, USA; 5Center for Surgery and Public Health, Harvard School of Public Health and Brigham and Women’s Hospital, Department of Surgery, Boston, Massachusetts, USA; and 6University of Michigan Medical School, Ann Arbor, Michigan, USA To whom correspondence should be addressed: Abdulrahman M. El-Sayed, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2011) 76, 5:459-466. DOI: 10.1016/j.wneu.2011.03.051
䡲 OBJECTIVE: Little is known about socioeconomic differences in postoperative outcomes after neurosurgery. We assessed the relation between insurance status and postoperative complication risk, neurosurgical intensive care unit stay, and hospital stay after neurosurgery. 䡲 METHODS: We collected data on 918 consecutive craniotomy or spine-related neurosurgical cases in patients at least 18 years of age at the University of Michigan Hospitals after April 2006. Bivariate 2 tests and analysis of variance were used to assess bivariate relations, and multivariable logistic regression models and analysis of covariance were used to adjust for potential confounders. 䡲 RESULTS: A total of 11.2% of privately insured patients, 23.6% of Medicare patients, 25.8% of Medicaid patients, and 27.3% of uninsured patients suffered complications within 30 days of surgery (P < 0.001). In adjusted models, odds of postoperative complications among Medicare (odds ratio [OR] ⴝ 2.1, 95% confidence interval [CI] 1.3–3.3), Medicaid (OR ⴝ 3.1, 95% CI 1.5– 6.1), and uninsured patients (OR ⴝ 3.6. 95% CI 1.3–10.3) were higher than among privately insured patients. By analysis of covariance, only Medicaid patients had significantly longer intensive care unit (P ⴝ 0.040) and hospital stays (P ⴝ 0.028) than privately insured patients. 䡲 CONCLUSIONS: Our findings suggest important socioeconomic disparities in outcomes after neurosurgical intervention. Access to postoperative outpatient care may mediate our findings.
Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter © 2011 Elsevier Inc. All rights reserved.
INTRODUCTION Disparities in both the quantity and quality of health services used among rich and poor are well documented in the health services literature (11, 12). Insurance status, a pivotal indicator of socioeconomic status with respect to health, has been demonstrated as an important predictor of several diverse and important health metrics in the United States, including cancer screening (25) and diagnosis (10, 16), treatment for heart disease (1), treatment and mortality after trauma (15), as well as overall mortality (13), with the uninsured and patients covered by Medicaid suffering worse outcomes. Of particular interest here is the consistent, well-documented relation between insurance status and outcomes after surgical
intervention (5, 7, 17, 21). For example, a recent study by LaPar et al. (21), using data on almost 900,000 patients undergoing diverse surgical procedures from the National Inpatient Sample, showed that mortality rates among patients on Medicaid and Medicare, as well as uninsured patients after surgery were significantly higher than among privately insured patients, and uninsured patients suffered the highest rates of mortality after surgery after adjusting for potential confounders. In addition, patients covered by Medicaid stayed longest in hospital and had the highest total costs associated with their care (21). The literature about socioeconomic disparities in outcomes after neurosurgery is limited. One study considered differences in metrics of postoperative outcomes among neurosurgery patients by insurance status. In a national sample of 99,665 pa-
WORLD NEUROSURGERY 76 [5]: 459-466, NOVEMBER 2011
tients undergoing craniotomy procedures between 1988 and 2004 from the National Inpatient Sample, Curry et al. (7) showed that patients paying for services by Medicaid had the highest rates of in-hospital mortality postoperatively, whereas privately insured patients had the lowest. Other studies have found similar disparities in outcomes by insurance status after treatment for intracranial aneurysms (17) and spinal fusion (5). Little is known about differences in risk for postoperative complications and/or hospital or neurosurgical intensive care unit (ICU) stays by insurance status among neurosurgical patients. Given the growth in the scale and scope of neurosurgical intervention, a more nuanced and systematic understanding of the socioeconomic determinants of postoperative complication and recovery is in order. Such an understanding
www.WORLDNEUROSURGERY.org
459
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
might empower policy-makers and clinicians to address these disparities, and prepare for and improve postoperative care. Using data about 918 consecutive spine and craniotomy neurosurgical cases at the University of Michigan Hospitals, we assessed the relation between insurance status and risk for 30-day postoperative complications, duration of ICU stay, and duration of hospital stay.
METHODS Data The University of Michigan Evidence-Based Neurosurgery Database collected retrospective data about 918 consecutive craniotomy or spine-related neurosurgical cases at the University of Michigan Hospitals from April 10, 2006 through May 4, 2009. Data were collected by trained medical students from full clinical registers in the electronic medical record system from the University of Michigan Health System. Patients younger than 18 years old at the time of operation were excluded from the database. Patients lacking 30-day follow-up, those with incomplete medical records, undergoing only ventriculostomy, head or neck cases that did not include craniotomy, or neuroendovascular procedures were excluded. From among 1331 patients originally enrolled, only 918 were suitable for inclusion in our analysis. Data were collected regarding patient demographics, comorbidities, postoperative complications, length of ICU stay, and length of hospital stay from neurosurgery clinical notes found in the electronic medical record system of the University of Michigan Hospitals. Data for explanatory covariates were collected regarding insurance status (analyzed as a categorical variable: private, Medicaid, Medicare, or none); gender; marital status (analyzed as a binary variable denoting the presence or absence of a spouse at time of operation); age (analyzed as a categorical variable: ⬍50, 50 –70, and 71 years or older); current tobacco and/or current alcohol use (both analyzed as binary variables: presence vs. absence); and case type (analyzed as a binary variable: spinal or craniotomy). Data about comorbidities were also collected, including obesity (body mass index ⬎30 kg/m2), diabetes mellitus, systemic hypertension, and/or
460
www.SCIENCEDIRECT.com
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
coronary artery disease (all analyzed as binary variables: presence vs. absence). Complications were defined as the occurrence of any of the following within 30 days of operation: death; coma ⬎24 hours; acute renal failure; postoperative bleeding requiring ICU stay, reoperation, or requiring more than four units of red blood cells within 72 hours; unplanned intubation; ventilation lasting more than 48 hours; pneumonia; cardiac arrest; myocardial infarction; pulmonary embolus; deep infection; sepsis; systemic inflammatory response syndrome; pseudomeningocele; deep vein thrombosis; seizure; and cerebrovascular accident (complications were analyzed as one binary variable denoting the presence or absence of any complication). Both length of ICU stay and total length of hospital stay were recorded in days for each patient, and were analyzed as continuous variables. This study was reviewed by the Medical Science Institutional Review Board of the University of Michigan. Analysis Because one of our outcomes of interest was categorical (i.e., complications) and the others were continuous (i.e., duration of hospital and ICU stay), we used multiple statistical analytical tools. We used bivariate 2 tests and multivariable logistic regression to assess relations between predictor covariates and categorical outcomes, and we used analysis of variance (ANOVA) and analysis of covariance (ANCOVA) to assess bivariate and multivariate relations, respectively, between predictor covariates and continuous outcomes. First, univariate statistics were calculated to describe our sample. Second, we used bivariate 2 tests to identify significant associations between explanatory covariates and insurance status, as well as between covariates of interest and categorical outcomes. Third, we used ANOVA to assess significant associations between explanatory covariates of interest and continuous outcomes. Fourth, a multivariable logistic regression model of 30-day complication risk by gender was fit, and was adjusted for potential confounders. We adjusted for covariates found to be significantly associated with postoperative complications in bivariate 2 tests. Fifth, ANCOVA models of ICU stay, as well as hospital stay by insurance
status were fit, and were adjusted for those covariates found to be significantly associated with the outcome of interest in ANOVA models. Associations with P ⬍ 0.05 were regarded as significant. SAS 9.2 (SAS Institute; Cary, North Carolina, USA) was used for all statistical analyses.
RESULTS Table 1 shows descriptive statistics and bivariate 2 tests between insurance status, demographic, and medical covariates. Insurance status was associated with marital status (P ⬍ 0.001), and those on private insurance and Medicare were significantly more likely to be married than their counterparts on Medicaid and without insurance. Insurance status was also associated with age (P ⬍ 0.001). Medicare patients and those with private insurance were older than their counterparts on Medicaid and without insurance. There was also a relationship between insurance status and tobacco use (P ⫽ 0.041), as patients on Medicaid were more likely to use tobacco than other groups. Insurance status also predicted case type (P ⬍ 0.001) (craniotomy vs. spinal cases), and those without insurance were more likely to undergo craniotomies than were other groups. Last, there were significant relationships between insurance status and hypertension (P ⬍ 0.001), coronary artery disease (P ⬍ 0.001), and diabetes mellitus (P ⬍ 0.001), and Medicare patients had higher risk for each of these comorbidities than other patient groups. Table 2 shows demographic characteristics and bivariate 2 tests between all covariates and 30-day complication risk. Of 918 patients in our study, 145 (15.8%) experienced a major complication within 30 postoperative days. Insurance status was associated with 30-day postoperative complication risk (P ⬍ 0.001). Among privately insured patients, 11.2% suffered a major complication, which was lower than among patients on Medicare (23.6%), Medicaid (25.8%), and uninsured patients (27.3%). Male gender (P ⬍ 0.001), older age (P ⫽ 0.006), craniotomy case type (P ⫽ 0.002), and emergent admissions status (P ⬍ 0.001), were all associated with postoperative complication risk relative to female gender, younger age, spinal case type, and elective admissions status, respectively. Among medical co-
WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2011.03.051
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
Table 1. Descriptive Statistics and Bivariate 2 Tests Between Insurance Status and Demographic and Medical Covariates Among 918 Patients Undergoing Neurosurgical Interventions at the University of Michigan Hospitals Insurance Status Patient Descriptives
Private
Medicare
Medicaid
None
Male
52.9
52.4
48.5
72.7
Female
47.1
47.6
51.5
27.3
No
32.4
36.8
74.2
68.2
Yes
67.6
63.2
25.8
31.8
⬍50 years
52.6
14.7
83.3
68.2
50–70 years
44.9
48.0
12.1
31.8
⬎70 years
2.5
37.3
4.6
0.0
No
59.6
57.2
40.6
59.1
Yes
40.4
42.8
59.4
40.9
No
63.3
66.1
57.8
54.6
Yes
36.7
33.9
42.2
45.4
Craniotomy
69.9
57.8
74.2
45.5
Spinal case
30.1
42.2
25.8
54.5
Emergent
10.5
12.4
16.7
13.6
Elective
89.5
87.6
83.3
86.4
No
66.7
64.9
69.7
68.2
Yes
33.3
35.1
30.3
31.8
P Value
Total Gender
0.089
⬍ 0.001
Married
⬍ 0.001
Age
Tobacco use
0.041
Alcohol use
0.516
Case type
0.001
Admission status
0.492
Obesity*
0.895
⬍0.001
Hypertension No
65.2
44.4
78.8
81.8
Yes
34.8
55.6
21.2
18.2
DISCUSSION ⬍0.001
Coronary artery disease No
95.7
80.4
95.4
100.0
Yes
4.3
19.6
4.6
0 ⬍0.001
Diabetes mellitus No
91.8
81.3
97.0
95.5
Yes
8.2
18.7
3.0
4.5
*Body mass index, ⬎30 kg/m . 2
morbidities, only coronary artery disease was associated with 30-day postoperative complication risk (P ⬍ 0.001). Table 3 shows mean postoperative hos-
pital and ICU stay, as well as ANOVA between explanatory covariates of interest and both continuous outcomes of interest among patients in our sample. Mean length
WORLD NEUROSURGERY 76 [5]: 459-466, NOVEMBER 2011
of stay in the ICU was 2.0 days (standard deviation 4.6 days), whereas the mean length of stay in the hospital was 6.6 days (standard deviation 9.4 days). Insurance status was not associated with length of ICU stay (P ⫽ 0.113). Craniotomy case type (relative to spinal case type) (P ⬍ 0.001), emergent admissions status (relative to elective) (P ⬎ 0.001), and obesity (P ⫽ 0.006) were the only predictors of length of postoperative ICU stay. Insurance status was associated with length of hospital stay (P ⫽ 0.020), along with male gender (relative to female gender) (P ⬍ 0.003), craniotomy case type (P ⬍ 0.001), emergent admissions status (P ⬍ 0.001), and coronary artery disease (P ⫽ 0.003). Table 4 shows a multivariable logistic regression model of 30-day postoperative complication risk by insurance status, adjusted for potential confounders. Compared with privately insured patients, odds of postoperative complications among patients on Medicare were 2.1 (95% confidence interval 1.3–3.3), odds among patients on Medicaid were 3.1 (95% confidence interval 1.5– 6.1), and odds among patients without insurance were 3.6 (95% confidence interval 1.3–10.3). Table 5 shows ANCOVA models of postoperative ICU stay and hospital stay by insurance status adjusted for potential confounders. Medicaid was a significant predictor of length of postoperative ICU stay (P ⫽ 0.031) and hospital stay (P ⫽ 0.028) relative to private insurance, but neither patients on Medicare nor those without insurance differed significantly from privately insured patients in either duration of ICU stay or duration of hospital stay.
In a study of 918 consecutive patients at a Midwestern teaching hospital, we found that relative to privately insured patients, patients on public insurance plans (Medicare or Medicaid) and uninsured patients had significantly increased risks for postoperative complications within 30 days of neurosurgery. Uninsured patients had the worst outcomes, followed by those on Medicaid, and then by those on Medicare. In addition, patients on Medicaid required significantly longer ICU and hospital stays than their privately insured counterparts. Our findings suggest important socioeco-
www.WORLDNEUROSURGERY.org
461
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
Table 2. Descriptive Statistics and Bivariate 2 Tests Between Demographic and Medical Covariates and Postoperative Complication Risk Among 918 Patients Undergoing Neurosurgical Interventions at the University of Michigan Hospitals 30-Day Complications Patient Descriptives
N
%
N
%
918
—
145
15.8
Private
439
58.4
49
11.2
Medicare
225
29.9
53
23.6
Medicaid
66
8.8
17
25.8
None
22
2.9
6
27.3
Total
⬍ 0.001
Insurance status
⬍ 0.001
Gender Male
459
50.0
93
20.3
Female
459
50.0
52
11.3
No
353
38.7
60
17.1
Yes
559
61.3
83
14.9
⬍50 years
409
44.6
48
11.7
50–70 years
388
42.3
70
18.0
⬎70 years
121
13.2
27
22.3
Married
0.550
Age
0.006
Tobacco use
0.651
No
525
57.8
79
15.1
Yes
384
42.2
62
16.2
No
574
63.2
90
15.7
Yes
334
36.8
51
15.3
Craniotomy
586
63.8
109
18.6
Spinal case
332
36.2
36
10.8
Emergent
103
11.2
45
43.7
Elective
814
88.8
100
12.3
No
610
66.5
91
14.9
Yes
308
33.6
54
17.5
No
546
59.5
78
14.3
Yes
371
40.5
67
18.1
No
835
91.1
114
13.7
Yes
82
8.9
31
37.8
816
89.0
124
15.2
101
11.0
21
20.8
Alcohol use
0.869
Case type
0.002
⬍ 0.001
Admission status
Obesity*
0.305
Hypertension
0.124
⬍ 0.001
Coronary artery disease
Diabetes mellitus
0.146
No Yes *Body mass index, ⬎30 kg/m . 2
462
P Value
www.SCIENCEDIRECT.com
nomic disparities in outcomes after neurosurgical intervention. Our findings are supported by the several other studies, which have considered differences in outcomes after neurosurgical intervention by insurance status (5, 7, 17). As mentioned previously, Curry et al. (7) found that patients paying for neurosurgery by Medicaid suffered the highest rates of inhospital mortality postoperatively, whereas privately insured patients suffered the lowest, and patients on Medicare as well as those without insurance suffered similar mortality rates, falling between those on Medicaid and the privately insured. Our findings affirm that privately insured patients had the lowest risk for postoperative complications. Although we found that uninsured patients had the highest risk for postoperative complications, rather than patients covered by Medicaid, as demonstrated in the study by Curry et al. (7), there were no significant differences in postoperative complication risk between patients on Medicaid, Medicare, or the uninsured in our analysis. Similar findings have also been demonstrated in outcomes after specific neurosurgical interventions (5, 17). Our findings may also be framed within the broader context of the literature about insurance status and postoperative outcomes among other surgical subspecialties. For example, our findings showed that patients covered by Medicaid had the longest postoperative ICU and surgical stays, which were significantly longer than those of privately insured patients. These findings are supported by work by LaPar et al. (21), reviewed previously, who showed that surgical patients on Medicaid had the longest postoperative hospital stays. In addition, the literature about differences in postoperative health metrics by insurance status among surgical patients has generally supported our findings that uninsured patients, as well as those covered by Medicare or Medicaid, suffer worse outcomes than their privately insured counterparts (5, 7, 17, 20-22). There are several plausible explanations for the disparities in postoperative outcomes by insurance status we observed here. First, by virtue of their insurance status, underinsured (Medicare and Medicaid) (9) and uninsured patients lack quality access to adequate primary health care (2). Hence, pathology among these groups is
WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2011.03.051
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
Table 3. Analysis of Variance Between Demographic and Medical Covariates and Postoperative Hospital and Intensive Care Unit Stay Among 918 Patients Undergoing Neurosurgical Interventions at the University of Michigan Hospitals Intensive Care Unit Stay Patient Descriptives Total
Days
SD
2.0
4.6
Insurance status
P Value
Hospital Stay Days
SD
6.6
9.4
0.113
P Value
Table 4. Logistic Regression Models of Complication Risk by Insurance Status Adjusted for Potential Confounders Among 918 Patients Undergoing Neurosurgical Interventions at the University of Michigan Hospitals Patient Descriptives
OR
95% CI
0.020 Insurance status
Private
1.8
4.0
5.7
8.1
Medicare
1.9
4.5
7.6
11.5
Private
ref
ref
Medicaid
3.1
4.9
8.6
9.4
Medicare
2.1
1.3–3.3
None
2.5
3.7
7.6
11.6
Medicaid
3.1
1.5–6.1
None
3.6
1.3–10.3
Male
1.6
1.1–2.4
Female
ref
ref
⬍50 years
ref
ref
50–70 years
1.8
1.1–2.8
⬎70 years
1.0
0.5–2.1
Craniotomy
1.8
1.2–2.8
ref
ref
Emergent
5.3
3.3–8.6
Elective
ref
ref
No
ref
ref
Yes
3.3
1.8–5.8
Gender
0.182
0.003
Male
2.2
5.0
7.5
10.8
Female
1.8
4.2
5.7
7.8
Married
0.592
0.086
No
2.0
4.3
7.2
10.2
Yes
1.9
4.6
6.1
8.8
Age ⬍50 years
0.292 1.7
3.8
50–70 years
2.1
5.1
⬎70 years
2.3
5.1
Tobacco use
Gender
Age 0.118
5.9
8.1
7.0
10.8
7.7
8.7
0.85
Case type 0.425
No
2.0
4.6
6.7
8.9
Spinal case
Yes
1.9
4.5
6.2
9.9
Admission status
No
2.0
4.9
6.4
8.7
Yes
1.9
4.1
6.7
10.4
Alcohol use
0.76
0.664
⬍ 0.001
Case type Craniotomy
2.7
5.3
7.7
10.6
Spinal case
0.7
2.6
4.8
6.6
⬍ 0.001
Admission status Emergent
6.3
8.2
Elective
1.4
3.6
Obesity*
⬍ 0.001 14.2
13.6
5.6
8.3
0.006
0.214
No
2.3
5.1
6.9
9.9
Yes
1.4
3.2
6.1
8.3
6.2
9.6
7.2
9.2
Hypertension
0.345
No
1.9
4.1
Yes
2.2
5.3
Coronary artery disease
0.109
0.078
0.003
No
1.9
4.4
6.3
9.3
Yes
2.8
5.8
9.6
10.3
No
2.0
4.6
6.5
9.5
Yes
2.0
4.8
7.4
8.8
Diabetes mellitus
Coronary artery disease ⬍ 0.001
0.889
0.359
SD, standard deviation. *Body mass index, ⬎30 kg/m2.
WORLD NEUROSURGERY 76 [5]: 459-466, NOVEMBER 2011
OR, odds ratio; 95% CI, 95% confidence interval; ref, referent.
likely to be more advanced at the time of presentation (3, 4, 21). Because more developed pathology is often more challenging to treat, and is associated with worse prognoses (3), differential access to primary health care may explain the differences in outcomes among neurosurgical patients by insurance status. Second, evidence suggests that distinct socioeconomic gradients in population health metrics persist independently of differences in access to health services (18, 28). For example, despite a universal health care system in the United Kingdom, there remain consistent socioeconomic gradients in several health metrics (8, 24). As insurance status is a reliable proxy for overall socioeconomic status, underinsured and
www.WORLDNEUROSURGERY.org
463
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
Table 5. Analysis of Covariance of Postoperative Hospital and Intensive Care Unit Stay by Insurance Status Adjusted for Potential Confounders Among 918 Patients Undergoing Neurosurgical Intervention at the University of Michigan Hospitals Intensive Care Unit Stay Patient Descriptives
Days
SD
Private
1.8
4.0
Medicare
1.9
4.5
Medicaid
3.1
4.9
None
2.5
3.7
P Value
Hospital Stay Days
SD
P Value
5.7
8.1
0.621
7.6
11.5
0.054
0.031
8.6
9.4
0.028
0.504
7.6
11.6
0.391
Insurance status
Gender
0.003
Male
7.5
10.8
Female
5.7
7.8
7.7
10.6
4.8
6.6
⬍0.001
Case type Craniotomy
2.7
5.3
Spinal case
0.7
2.6
⬍0.001
⬍0.001
Admission status
⬍0.001
Emergent
6.3
8.2
14.2
13.6
Elective
1.4
3.6
5.6
8.3
No
2.3
5.1
Yes
1.4
3.2
Obesity*
0.007
Coronary artery disease
0.016
No
6.3
9.3
Yes
9.6
10.3
SD, standard deviation. *Body mass index, ⬎30 kg/m2.
uninsured patients may be systematically more deprived than their privately insured counterparts. And as socioeconomic inequalities in health metrics persist independently of access to care, this population may have systematically worse baseline health than privately insured patients. Because worse baseline health is a demonstrated predictor of poor surgical outcomes (19, 29), it is plausible that differences in postoperative outcomes by insurance status may result from a priori differences in baseline health between privately insured and underand uninsured populations. Third, it is plausible that underinsured and uninsured patients, even after surgical intervention, may lack the coverage to support postoperative care after discharge. With compromised postsurgical care, it is plausible that these patients would suffer worse outcomes compared with privately
464
www.SCIENCEDIRECT.com
insured patients. In support of this explanation, several studies have demonstrated financial barriers to postoperative outpatient care among neurosurgical patients (14, 26). For example, one study by Gerszten et al. (14) showed that among 91 patients, uninsured patients had to wait significantly longer for placement into rehabilitation services than insured patients. Differences in access to outpatient postoperative care may also indirectly explain observed differences in the durations of ICU and hospital stays among Medicaid patients compared with others. It is possible that clinicians may purposely keep their Medicaid-covered patients in hospitalized care, where their expenses are more likely to be reimbursed by Medicaid, rather than discharging them where they may face more systematic financial barriers to outpatient care.
There are several limitations to our findings of which the reader should be aware. First, although a strong proxy for socioeconomic status at the population level, insurance status among some patients may not reliably approximate socioeconomic status (e.g., uninsured young people who choose not to purchase health insurance). In addition, this metric is partially age dependent (e.g., Medicare patients are, by definition, 65 years or older), and therefore, associations between insurance status and health may be confounded by age. Second, as with any observational analysis, although we adjusted for covariates that could potentially confound the association between insurance status and outcomes of interest, it is plausible that the relation remains residually confounded by unmeasured factors that are associated with both exposures and outcomes of interest. Third, our findings were derived from a sample that was bounded by time and space, and therefore, they may not generalize to other temporal and spatial contexts. In addition, our sample was limited to patients at a high-volume academic tertiary care center. Because outcomes have been shown to be better in these care contexts compared to low-volume centers (6, 27), and underinsured and uninsured patients may be more likely to seek care at low-volume centers (23), it is possible that our findings may underestimate the true disparities in neurosurgical outcomes nationally. Fourth, our sample included only patients undergoing a limited spectrum of neurosurgical operations, and therefore, our findings may not generalize to other patient populations. Fifth, we only considered outcomes within 30 days of operation, further limiting the generalizability of these findings. Despite these limitations, we believe that our observations have important implications for health policy, neurosurgical practice, and future research in this area. With regard to health policy, our findings suggest that the health effects of insurance extend beyond access to care—the population sampled in the present study had received neurosurgical care, yet there were distinct inequalities in outcomes by insurance status. Therefore, policy-makers intent on mitigating socioeconomic disparities in health metrics should consider means of health improvement, adjuvant to improvements in health access among the poor,
WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2011.03.051
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
which are independent of the formal health system. Interventions that focus on prevention alongside improved access to curative care may be in order. Second, alongside a growing literature that has demonstrated disparities in access to postoperative outpatient services among neurosurgical patients (14, 26), our findings suggest that barriers to postoperative outpatient care may compromise the neurosurgical outcomes of underinsured and uninsured patients. Therefore, policy-makers should seek to address these barriers so as to improve outcomes among the most deprived neurosurgical patients. Third, our findings demonstrated that Medicaid users required significantly longer hospital and neurosurgical ICU stays than their privately insured counterparts. We were unable to explain this troubling finding. However, postoperative neurosurgical care is costly, and a systematic requirement for greater care among publicly relative to privately insured patients, suggests inefficiencies in the public system. Policymakers may seek to address these differences by improving access to primary care, postoperative outpatient services, and/or improving baseline health metrics among Medicaid patients. With regard to clinical care, it is important that clinicians recognize and actively seek to remedy the outcome disparities we have noted. Although it is improbable that there are intraoperative differences in care that mediate the disparities that we have observed, differences in postoperative care may very well mediate this relation. In addition, our findings demonstrated higher risks for postoperative complications among underinsured and uninsured patients. Therefore, astute clinicians should work with these patients to negotiate barriers in access to quality postoperative care to assure optimal outcomes, as well as maintain a particular vigilance and attentiveness to signs and symptoms of postoperative complications among this population. Finally, our findings also have several implications for future research in this area. First, multicenter prospective studies about the relations between socioeconomic metrics and neurosurgical outcomes are needed to clarify the etiologies of differences in postoperative outcomes by insurance status. Such studies might consider differences in care trajectories after hospital discharge, as well as a more nuanced explo-
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
ration of differences in baseline preoperative health among privately insured and underinsured and uninsured patients. Second, investigators interested in inequalities in neurosurgical outcomes would be well served to consider relations between other metrics of socioeconomic status, such as education, income, and contextual deprivation, as well as race and ethnicity and neurosurgical outcomes.
CONCLUSIONS Our findings suggest that patients covered by Medicare and Medicaid, as well as uninsured patients, may have worse outcomes after neurosurgical intervention than privately insured patients. Differences in baseline health, access to primary care, or access to postoperative outpatient care may mediate the association between insurance status and postneurosurgical outcomes. Clinicians should work with underinsured and uninsured patients to negotiate barriers in access to outpatient care and pay particular attention to signs and symptoms of postoperative complication among this population.
REFERENCES
in the United States, 1988 –2004. Neurosurgery 66:427-438, 2010. 8. Davey Smith G, Dorling D, Mitchell R, Shaw M: Health inequalities in Britain: continuing increases up to the end of the 20th century. J Epidemiol Community Health 56:434-435, 2002. 9. Farley PJ: Who are the underinsured? Milbank Mem Fund Q Health Soc 63:476-503, 1985. 10. Farley TA, Flannery JT: Late-stage diagnosis of breast cancer in women of lower socioeconomic status: public health implications. Am J Public Health 79:1508-1512, 1989. 11. Fiscella K, Franks P, Doescher M, Saver BG: Disparities in health care by race, ethnicity, and language among the insured: findings from a national sample. Med Care 40:52-59, 2002. 12. Fiscella K, Franks P, Gold MR, Clancy CM: Inequality in quality: addressing socioeconomic, racial, and ethnic disparities in health care. JAMA 283:25792584, 2000. 13. Franks P, Clancy CM, Gold MR: Health insurance and mortality. Evidence from a national cohort. JAMA 270:737-741, 1993. 14. Gerszten PC, Witham TF, Clyde BL, Welch WC: Relationship between type of health insurance and time to inpatient rehabilitation placement for surgical subspecialty patients. Am J Med Qual 16:212-215, 2001. 15. Haas JSP, Goldman L: Acutely injured patients with trauma in Massachusetts: differences in care and mortality, by insurance status. Am J Public Health 84:1605-1608, 1994.
1. Ayanian JZ, Epstein AM: Differences in the use of procedures between women and men hospitalized for coronary heart disease. N Engl J Med 325:221225, 1991.
16. Halpern MT, Bian J, Ward EM, Schrag NM, Chen AY: Insurance status and stage of cancer at diagnosis among women with breast cancer. Cancer 110:403411, 2007.
2. Berk ML, Schur CL: Access to care: how much difference does Medicaid make? Health Aff 17:169-180, 1998.
17. Hoh BL, Rabinov JD, Pryor JC, Carter BS, Barker FG 2nd: In-hospital morbidity and mortality after endovascular treatment of unruptured intracranial aneurysms in the United States, 1996 –2000: effect of hospital and physician volume. AJNR Am J Neuroradiol 24:1409-1420, 2003.
3. Bradley CJ, Given CW, Dahman B, Luo Z, Virnig BA: Diagnosis of advanced cancer among elderly Medicare and Medicaid patients. Med Care 45:410-419, 2007. 4. Bradley C, Given C, Roberts C: Late stage cancers in a Medicaid-insured population. Med Care 41:722728, 2003. 5. Cahill KS, Chi JH, Day A, Claus EB: Prevalence, complications, and hospital charges associated with use of bone-morphogenetic proteins in spinal fusion procedures. JAMA 302:58-66, 2009. 6. Cowan JA Jr, Dimick JB, Leveque JC, Thompson BG, Upchurch GR Jr, Hoff JT: The impact of provider volume on mortality after intracranial tumor resection. Neurosurgery 52:48-54, 2003. 7. Curry WT Jr, Carter BS, Barker FG 2nd: Racial, ethnic, and socioeconomic disparities in patient outcomes after craniotomy for tumor in adult patients
WORLD NEUROSURGERY 76 [5]: 459-466, NOVEMBER 2011
18. Katz SJ, Hofer TP: Socioeconomic disparities in preventive care persist despite universal coverage: breast and cervical cancer screening in Ontario and the United States. JAMA 272:530-534, 1994. 19. Katz J, Stucki G, Lipson S, Fossel AH, Grobler LJ, Weinstein JN: Predictors of surgical outcome in degenerative lumbar spinal stenosis. Spine 24: 2229-2233, 1999. 20. Kelz RR, Gimotty PA, Polsky D, Norman S, Fraker D, DeMichele A: Morbidity and mortality of colorectal carcinoma surgery differs by insurance status. Cancer 101:2187-2194, 2004. 21. LaPar DJ, Bhamidipati CM, Mery CM, Stukenborg GJ, Jones DR, Schirmer BD, Kron IL, Ailawadi G: Primary payer status affects mortality for major surgical operations. Ann Surg 252:544-551, 2010.
www.WORLDNEUROSURGERY.org
465
PEER-REVIEW REPORTS ABDULRAHMAN M. EL-SAYED ET AL.
22. Lidor AO, Gearhart SL, Wu AW, Chang DC: Effect of race and insurance status on presentation, treatment, and mortality in patients undergoing surgery for diverticulitis. Arch Surg 143:1160-1165, 2008. 23. Liu JH, Zingmond DS, McGory ML, SooHoo NF, Ettner SL, Brook RH, Ko CY: Disparities in the utilization of high-volume hospitals for complex surgery. JAMA 296:1973-1980, 2006. 24. Marmot MG, Smith GD, Stansfield S, Patel C, North F, Head J, White I, Brunner E, Feeney A: Health inequalities among British civil servants: the Whitehall II study. Lancet 337:1387-1393, 1991. 25. Rodriguez MA, Ward LM, Perez-Stable EJ: Breast and cervical cancer screening: impact of health insurance status, ethnicity, and nativity of Latinas. Ann Fam Med 3:235-241, 2005.
INSURANCE STATUS AND INEQUALITIES AFTER NEUROSURGERY
26. Taylor L, Simpson K, Bushardt R, Reeves C, Elkin TD, Fortson BL, Boll T, Patel S: Insurance barriers for childhood survivors of pediatric brain tumors: the case for neurocognitive evaluations. Pediatr Neurosurg 42:223-227, 2000. 27. Solomon RA, Mayer SA, Tarmey JJ: Relationship between the volume of craniotomies for cerebral aneurysm performed at New York state hospitals and in-hospital mortality. Stroke 27:13-17, 1996. 28. Veugelers PJ, Yip AM: Socioeconomic disparities in health care use: does universal coverage reduce inequalities in health? J Epidemiol Community Health 57:424-428, 2003.
propriateness, outcomes, and risk factors for complications. Stroke 28:891-898, 1997.
Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. received 2 December 2010; accepted 31 March 2011 Citation: World Neurosurg. (2011) 76, 5:459-466. DOI: 10.1016/j.wneu.2011.03.051 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter © 2011 Elsevier Inc. All rights reserved.
29. Wong JH, Findlay JM, Suarez-Almazor ME: Regional performance of carotid endarterectomy: ap-
Santiago Ramón y Cajal and Harvey Cushing: Two Forefathers of Neuroscience and Neurosurgery Grettel J. Zamora-Berridi2, Courtney Pendleton1, Gabriel Ruiz3, Aaron A. Cohen-Gadol4, Alfredo Quiñones-Hinojosa1
Key words 䡲 Cajal 䡲 Cushing 䡲 Neuroscience 䡲 Neurosurgery 䡲 Physician-scientist Abbreviations and Acronyms CNS: Central nervous system ICP: Intracranial pressure From the 1Department of Neurosurgery, Brain Tumor Stem Cell Laboratory, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA; 2University of Michigan Medical School, Ann Arbor, Michigan, USA; 3 Departamento de Psicología Experimental, Universidad de Sevilla, Sevilla, Spain; and 4Clarian Neuroscience Institute, Indianapolis Neurosurgical Group and Indiana University Department of Neurosurgery, Indianapolis, Indiana, USA
䡲 OBJECTIVE: To summarize the extraordinary accomplishments, and the commonalities, between Santiago Ramon y Cajal and Harvey Williams Cushing. 䡲 METHODS: Existing literature describing the lives and achievements of Ramón y Cajal and Cushing, as well as personal communication, and the surgical records of the Johns Hopkins Hospital, from 1896 to 1912, were reviewed. 䡲 RESULTS: Both Ramón y Cajal and Cushing were men of unusually broad interests and talents, and these shared characteristics undoubtedly influenced the career paths and scientific investigations they pursued. Although Santiago Ramón y Cajal and Harvey Williams Cushing never directly interacted, the links between them can be traced through some of their disciples, including Pío del Río Hortega, Wilder Penfield, and Percival Bailey. 䡲 CONCLUSIONS: Ramón y Cajal and Cushing are widely considered the forefathers of neuroscience and neurosurgery, respectively, and their discoveries have made lasting impressions on both the scientific and medical communities.
To whom correspondence should be addressed: Alfredo Quiñones-Hinojosa, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2011) 76, 5:466-476. DOI: 10.1016/j.wneu.2011.04.001 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter © 2011 Elsevier Inc. All rights reserved.
INTRODUCTION During the past century many people involved in science and medicine have been fascinated by the lives of both Santiago Ramón y Cajal (May 1, 1852–October 17, 1934) and Harvey Cushing (April 8, 1869 –
466
www.SCIENCEDIRECT.com
October 7, 1939). Much of our present-day knowledge, and the advancements made in both fields, are deeply rooted in the legacies left by Cajal and Cushing. Many historic events and scientific discoveries in the field of neuroscience preceded both Ramón y Cajal and Cushing. Such events include the articulation of the cell theory by Theodor Schwann and Matthias Schleiden in the late 1830s (90, 91), the depiction of the cell as an independent unit by Rudolf Virchow in 1855 (99), the first original descriptions of the central nervous
system by Rudolph Albert von Kölliker (22, 49, 93) and Jan Evangelista Purkinje (66), and the founding of the field of neurophysiology by Charles Sherrington (94). Likewise, the emphasis on patient-based learning, translational research, and meticulous technique by William Welch, Sir William Osler, and William Stuart Halsted, revolutionized clinical practice and medical training (11, 14, 29, 55). These connections between bench and bedside, coupled with the burgeoning field of neuroscience research, allowed both Cushing and Ramón y Cajal to
WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2011.04.001