|
Ming-Yuan Tseng, MD, MPhil, MSc, PhD Department of Neurosurgery, Addenbrooke’s Hospital, University of Cambridge, U.K. Jen-Ho Tseng, MD Edwin Merchant, BA The authors reported no conflicts for disclosure. Received: March 4, 2007 Key Words: brain tumor, cancer registry, geography, glioma, socioeconomic Correspondence to: Ming-Yuan Tseng, [email protected] Abbreviations: CNS, central nervous system; CSR, crude survival
rate; HR, hazard ratio; ICD-O, International Classification of Diseases for
Oncology; MST, median survival time; NHS, National Health Services; SES, socioeconomic
status; |
Survival for gliomas of the CNS and particularly cerebral gliomas depends upon tumor characteristics and patient age (22-24). Even though diagnoses and management have been improved since the introduction of magnetic resonance imaging and adjuvant radiotherapy, comparable improvement in the survival seems to be unsatisfactory (8).
The NHS in the U.K. has set out aims and directives to provide the public with equal opportunities of accessing healthcare facilities. However, because of shortages in staff or equipment, patients from deprived SES or living in certain regions may have longer waiting time before being evaluated by specialists (10, 16), while the others can use private medical insurance or employment-related healthcare schemes to bypass the long waiting list and be treated in private hospitals (25). The effects from the inequity of SES and geographical variations on the survival have been demonstrated in some cancers (1, 7, 9), however very few studies have examined the independent effects of social vulnerability on survival for primary CNS gliomas (24).
This study investigates the effects of SES and geographical variations on survival for primary CNS gliomas at a national population level by using the dataset from the Cancer Registry in the U.K. which has the advantage of high accuracy (93 percent) and case coverage (> 94 percent) (9, 17).
Materials and Methods
Data
Anonymous data from the Cancer Registry, Office for National Statistics, London,
U.K. (6), were used and therefore approval from the Ethics Committee was
not required. Sources for the cancer registration included inpatient medical
records, reports from pathology laboratory, and records of outpatient or
other associated departments (radiotherapy, cytology, and general practitioners).
In order to solely investigate the effects of SES and geographical variation
on survival for primary CNS gliomas, tumors of metastasis, unknown origin,
non-glioma, unspecified behavior, death on certificates only, or those with
missing data on SES were excluded. The ICD-O was recategorized according
to the WHO classification as astrocytoma, oligodendroglioma, oligoastrocytoma,
ependymoma, and neuroembryonal tumors. The WHO grade was assigned to each
ICD-O code (14).
|
Abstract In order to compare effects of socioeconomic status, SES, and geographical variations on survival for patients with gliomas, data of 30,490 adults and 2,948 children with brain gliomas, and 336 adults and 63 children with primary spinal gliomas from the Cancer Registry in England and Wales were analyzed. The SES was categorized as higher (top 50 percent) and lower (bottom 50 percent) levels. The nine geographical regions were recategorized as Southern England, Midlands, and Northern England and Wales. One-, five- and 10-year rates for median survival time, MST, and crude survival rate, CSR, with respect to age, sex, morphology, World Health Organizaiton grade, periods of diagnosis, SES, and geographical regions were compared using the Kaplan-Meier method. Cox regressions were used for estimating the hazard ration, HR, to death from each variable. Results showed that in adults with brain gliomas, those from higher SES had lower risk of death (HR 0.94, p < 0.001), while those living outside Southern England had higher mortality (HR 1.08 to 1.10, p < 0.001). In children with brain gliomas, those residing in Midlands had increased risk of death (HR 1.16, p = 0.033), compared to other regions. However, effects from the SES on children were not significant. In patients with spinal gliomas (adults and children), the SES and geographical regions had no significant effect on survival. This study demonstrated that the SES had significant effects on the survival for adult brain gliomas, while the geographical variation influenced the survival for both adults and children with brain gliomas. |
Material deprivation was represented by scores of the Carstairs index derived from the census data in 1991, which included 109,578 census enumeration districts. Four variables (car ownership, household overcrowding, head of household in social class, and male unemployment) were used to calculate percentage values in each census enumeration district (5). The scores were ranked and the higher and the lower 50 percent of Carstairs scores were categorized into the higher and the lower SES, respectively. The nine NHS regions of England and Wales were recategorized as three major geographical regions: Southern England (South West, North Thames, South Thames, and Anglia and Oxford), Midlands (West Midlands, and Trent), and Northern England and Wales (North and Yorkshire, North West, and Wales).
Analysis
Analysis was performed using STATA for Windows (Intercool 8.0, College Station,
Texas). Survival time was computed as the number of days between the date
of diagnosis and the date of the last follow-up or death and divided by 365.25
to convert to elapsed years. Patients who were either alive or had emigrated
10 years after the diagnosis or on Dec. 31, 1995, were excluded, while the
others were followed until the time of their death. Therefore, those who
were diagnosed in the early 1970s had been monitored for up to 20 years.
Survival analyses were performed for brain gliomas in adults or children, and primary spinal gliomas in adults or children. Distributions among different variables were compared using a chi-square test. The estimated MST was defined as the point in time when the survival rate was 50 percent. The CSR was estimated by the Kaplan-Meier method, and the significance was determined by the log-rank test and defined as p < 0.05. The CSR at one, five, and 10 years was defined as the proportion of subjects surviving from diagnosis until those times, and the rates were compared with respect to SES and geographical regions. Potential confounders, including age, sex, ICD-O morphology, WHO grade, and period of diagnosis, were adjusted.
The Cox proportional hazard regression was used for multivariate logistic modelling. The probability of death for a specific category relative to a reference was presented as HR with 95 percent confidence interval. The likelihood ratio test was used for evaluating improvement in fit in each variable and the potential linear trend in certain categorical variables was tested for departure from a linear trend (one degree of freedom).
Data summary
There were 41,571 adults (≥ 15 years) and 4,485 children (< 15 years) registered
with CNS malignancy in the original dataset. After 2,998 patients were excluded
because of no data on SES, a total of 30,490 adults and 2,948 children with
brain gliomas, and 336 adults and 63 children with primary spinal gliomas were
eligible for final analyses. Information available included sex, dates of birth
and diagnosis, tumor morphology, calendar period of diagnosis (1971-1975, 1976-1980,
1981-1985, 1986-1990), vital status, region of residence, and Carstairs index.
There were more male than female patients (male-to-female ratio 1:4 in adults
and 1:2 in children). The most frequent morphology was astrocytoma (88.9 percent
brain gliomas, 62.4 percent spinal cord gliomas) and the majority of the brain
gliomas were of high grade (WHO grade III or IV, 86.9 percent). Only 7 percent
of the primary spinal gliomas were high grade.
In the whole study population, more patients came from higher SES or Southern England, but the distribution of SES among geographical regions varied; ratios of the higher to the lower SES in Southern England, Midlands, and Northern England and Wales were 1.92, 0.88, and 0.80, respectively (chi-square test, p < 0.001). The inequality of SES among geographical regions was most pronounced in adult patients with brain gliomas (Table 1).
|
| Table 1 *Chi-square test for each tumor population.
Abbreviations: SES, socioeconomic status; H/L, highlow, CNS, central
nervous system. † SES high and low represent data ranked into upper 50 persent (high) and Lower 50 percent (low) using the Carstairs index. |
Survival
The MST and CSR for patients with CNS glioma with respect to SES and geographical
regions are shown in Table 2. The MST and one-, five-, and 10-year CSR for
adult brain gliomas were 0.42 years, 29.1 percent, 12.0 percent, and 7.6
percent, respectively. The MST and one-, five-, and 10-year CSR for children
with brain gliomas were 9.5 years, 72.8 percent, 54.4 percent, and 49.6 percent,
respectively. The MST and the one-, five-, and 10-year CSR for adult spinal
gliomas were 8.7 years, 79.2 percent, 59.2 percent, and 48.5 percent, respectively.
The one-, five-, and 10-year CSR for children with spinal gliomas were 77.8
percent, 66.7 percent, and 63.5 percent, respectively, but the MST was too
lengthy to be available.
|
| Table 2 The log-rank test for the Kaplan-Meier method Abbreviations: CNS, central nervous system; CSR, crude survival rate; MSt, median survival time; N/A, not available, SES socioeconomic status. † SES high and low represent data ranked into upper 50 persent (high) and Lower 50 percent (low) using the Carstairs index. |
In adults with brain gliomas, those from higher SES or Southern England achieved longer survival than the others (log-rank test p < 0.001). In children with brain gliomas, those from higher SES seemed to fare better than those from lower SES (log-rank test p = 0.085), but no difference in CSR was seen among various geographical regions (log-rank test p = 0.22). In patients with primary spinal gliomas (adults and children), no significant difference in survival was seen between the two SES or among geographical regions.
The Cox regression showed that in adults with brain gliomas, those from the higher SES had reduced risk of deaths (HR 0.93, p < 0.001), while the others residing in regions outside Southern England had increased mortality (HR 1.09 to 1.10, p < 0.001) (Table 3A). In children with brain gliomas, those living in Midlands had higher risk of deaths (HR 1.17, p = 0.027), compared with the others (Table 3B), but no significant effects from the SES were seen. In patients with spinal gliomas (adults and children), no significant effects on the survival were seen with respect to either SES or geographical regions (Table 4).
|
| Table 3 The log-rank test for the Kaplan-Meier
method Abbreviations: CI, confidence interval; HR, hazard ratio; SES socioeconomic
status. † SES high and low represent data ranked into upper 50 persent (high) and Lower 50 percent (low) using the Carstairs index. |
|
| Table 4 The log-rank test for the Kaplan-Meier
method Abbreviations: CI, confidence interval; HR, hazard ratio; NA: not available
SES, socioeconomic status. † SES high and low represent data ranked into upper 50 persent (high) and Lower 50 percent (low) using the Carstairs index. |
Discussion
Results of this study confirmed that the SES had significant effects on the
survival for brain gliomas in adults, while the geographical regions influenced
the survival for brain gliomas both in adults and children. No similar effects
were seen in patients with spinal gliomas, probably because there were far
fewer patients, or it may be related to the indolent nature of the spinal
gliomas themselves (12, 18). The difference in survival curves among this
study population may be partially attributed to the biological characteristics
in both the tumor and the host (2).
Although the distribution of SES and geographical variations was similar in patients with brain gliomas, material deprivation obviously only affected the adults. This finding may be associated with the standardized treatment of, or more clinical trials for children, who are most frequently referred to certain specialized centers, irrespective of their family backgrounds (13, 20). Thus, in children, access to the NHS has been found to be equal across all SES levels (19).
For adults with brain gliomas, treatment protocols generally vary among centers. Furthermore, previous studies have suggested that patients from the more deprived SES are less likely to receive continuity of care because of their underlying social or economic problems (21). Even after being referred to the specialist they may be required to wait longer for diagnostic imaging because of severe shortages in staffing and infrastructure in certain regions (10), or because their residence is distant from secondary or tertiary referral centers, making arrangements for accessing these services more difficult (11, 15). Moreover, it has been found that these patients use the emergency department as their medical access more frequently than others (21). In contrast, patients from higher SES are more likely to use private medical insurance or employment-related health schemes to bypass NHS waiting lists or to be treated earlier in private hospitals (10, 16). The difference in waiting time may have caused “lead-time bias,” as patients with social disadvantages more frequently have advanced disease at diagnosis and subsequent shorter survival time (16). The significant effect of geography on survival for children with brain gliomas may be associated with similar shortages in staffing and infrastructures in certain regions as mentioned above.
For primary spinal gliomas, patients could be migrating among different geographical regions and neighborhoods because of their lengthy and subtle symptoms. The small number of patients in each region also may have made the effects of SES or geography difficult to demonstrate.
Although the 2,998 patients with missing data on SES only comprised 8.9 percent of the study population, the exclusion of them from analyses may have slightly underestimated the effect of SES as they were most likely to come from materially deprived neighborhoods (6). Furthermore, the geographical region where the diagnosis was made may not be necessarily the same place where the treatment was given, an important factor known to influence the outcome. Therefore, the effect of geographical region associated with the local health services also may have been underestimated.
The recent reports upon the financial management of NHS showed a deficit of 512 million pounds ($1,002,727,300) (3). The severe deficits have motivated the NHS executives to try to adopt several procedures for counterbalancing (4), including job cuts by modernization of the medical career for junior doctors, ward closure, stopping pay raises for the general practitioners, and delaying elective operations (BBC News June 7, 2006, and March 1 and 17, 2007). Therefore, we may expect that the differences in survival for patients with CNS gliomas among SES levels and geographical regions may become even more exaggerated in the future.
Conclusion
This study demonstrated that SES had significant effects on the survival for
adult brain gliomas, while the geographical variation influenced the survival
for both adults and children with brain gliomas. Effects of SES and geographical
variations may be even more pronounced in the cohort to be studied in the
future.
|
References 1. Baker D, Mead N, Campbell S: Inequalities in morbidity and consulting behaviour for socially vulnerable groups. Br J Gen Pract 52:124-130, 2002 2. Bloom HJ, Glees J, Bell J, Ashley SE, Gorman C: The treatment and long-term prognosis of children with intracranial tumors: a study of 610 cases, 1950-1981. Int J Radiat Oncol Biol Phys 18:723-745, 1990 3. Bourn J: Financial Management in the NHS: NHS (England) Summarised accounts 2004—05. National Audit Office, London, 2006 4. Carruthers I: Chief Executive’s report to the NHS: June 2006. Department of Health, Leeds, U.K., 2006 5. Carstairs V, Morris R: Deprivation and mortality: an alternative to social class? Community Med 11:210-219, 1989 6. Coleman MP, Babb P, Mayer D, Quinn MJ, Sloggett A: Cancer survival trends in England and Wales, 1971-1995: deprivation and NHS region (CDROM). Office for National Statistics, London, 1999 7. Coleman MP, Babb P, Sloggett A, Quinn M, De Stavola B: Socioeconomic inequalities in cancer survival in England and Wales. Cancer 91(1 Suppl):208-216, 2001 8. Davis FG, Freels S, Grutsch J, Barlas S, Brem S: Survival rates in patients with primary malignant brain tumors stratified by patient age and tumor histological type: an analysis based on Surveillance, Epidemiology, and End Results (SEER) data, 1973—1991. J Neurosurg 88:1-10, 1998 9. Dickinson HO, Salotti JA, Birch PJ, Reid MM, Malcolm A, Parker L: How complete and accurate are cancer registrations notified by the National Health Service Central Register for England and Wales? J Epidemiol Community Health 55:414-422, 2001 10. Duff SE, Wood C, McCredie V, Levine E, Saunders MP, OâDwyer ST: Waiting times for treatment of rectal cancer in North West England. J R Soc Med 97:117-118, 2004 11. Hippisley-Cox J, Pringle M. Inequalities in access to coronary angiography and revascularisation: the association of deprivation and location of primary care services. Br J Gen Pract 50:449—454, 2000. 12. Houten JK, Cooper PR: Spinal cord astrocytomas: presentation, management and outcome. J Neurooncol 47:219-224, 2000. 13. Johannesen TB, Langmark F, Lote K: Cause of death and long-term survival in patients with neuro-epithelial brain tumours: a population-based study. Eur J Cancer 39:2355-2363, 2003 14. Kleihues P, Cavenee WK: WHO classification of tumors. Pathology and genetics of tumors of nervous system. IARC, Lyon, France, 2000 15. MacKenzie R, Nimmo F, Bachoo P, Alozairi O, Brittenden J: The relationship between socio-economic status, geography, symptomatic carotid territory disease and carotid endarterectomy. Eur J Vas Endovasc Surg 26:145-149, 2003 16. Pell JP, Pell AC, Norrie J, Ford I, Cobbe SM: Effect of socioeconomic deprivation on waiting time for cardiac surgery: retrospective cohort study. BMJ 320:15-18, 2000 17. Quinn M, Babb P, Brock A, Kirby L, Jones J: Cancer trends in England and Wales 1950-1999. Office for National Statistics, London, 2001 18. Reimer R, Onofrio BM: Astrocytomas of the spinal cord in children and adolescents. J Neurosurg 63:669-675, 1985. 19. Saxena S, Eliahoo J, Majeed A: Socioeconomic and ethnic group differences in self reported health status and use of health services by children and young people in England: cross sectional study. BMJ 325:520-523, 2002 20. Stiller CA, Bunch KJ: Brain and spinal tumours in children aged under two years: incidence and survival in Britain, 1971-85. Br J Cancer (Suppl) 18:50-53, 1992 21. Sweeney KG, Gray DP: Patients who do not receive continuity of care from their general practitioner—are they a vulnerable group? Br J Gen Pract 45:133-135, 1995 22. Tseng JH, Tseng MY: Survival analysis of 459 adult patients with primary spinal cancer in England and Wales: a population-based study. Surg Neurol 67:53—58, 2007 23. Tseng JH, Tseng MY: Survival analysis of 81 children with primary spinal gliomas: A population-based study. Pediatr Neurosurg 42:347-353, 2006 24. Tseng MY, Tseng JH, Merchant E: Comparison of effects of socioeconomic and geographic variations on survival for adults and children with glioma. J Neurosurg (4 Suppl Pediatrics) 105:297-305, 2006 25. Williams B, Whatmough P, McGill J, Rushton L: Impact of private funding on access to elective hospital treatment in the regions of England and Wales. National records survey. Eur J Public Health 11:402-406, 2001 |