The influence of socioeconomic inequalities in pancreatic cancer patients and especially its effect in patients who had a resection is not known. Hospital type in which resection is performed might also influence outcome. Patients diagnosed with pancreatic cancer from 1989 to 2011 (n = 34,757) were selected from the population-based Netherlands Cancer Registry. Postal code was used to determine SES. Multivariable survival analyses using Cox regression were conducted to discriminate independent risk factors for death. Patients living in a high SES neighborhood more often underwent resection and more often were operated in a university hospital. After adjustment for clinicopathological factors, risk of dying was increased independently for patients with intermediate and low SES compared to patients with high SES. After resection, no survival difference was found among patients in the three SES groups. However, survival was better for patients treated in university hospitals compared to patients treated in non-university hospitals. Low SES was an independent risk factor for poor survival in patients with pancreatic cancer. SES was not an adverse risk factor after resection. Resection in non-university hospitals was associated with a worse prognosis.
Citation: van Roest MHG, van der Aa MA, van der Geest LGM, de Jong KP (2016) The Impact of Socioeconomic Status, Surgical Resection and Type of Hospital on Survival in Patients with Pancreatic Cancer. A Population-Based Study in The Netherlands. PLoS ONE 11(11): e0166449. https://doi.org/10.1371/journal.pone.0166449
Editor: Aamir Ahmad, University of South Alabama Mitchell Cancer Institute, UNITED STATES
Received: April 11, 2016; Accepted: October 28, 2016; Published: November 10, 2016
Copyright: © 2016 van Roest et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors are unable to publicly share the data due to a sharing agreement with the Netherlands Cancer Registry which reports cancer facts in Cijfers over Kanker. However, all interested parties will be able to obtain the dataset in the same manner that the authors did. The data set is available on request from the Netherlands Cancer Registry after approval of the Netherlands Cancer Registry Supervisory Committee via https://www.iknl.nl/cijfers/gegevens-aanvragen (English website available).
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Despite advances in knowledge concerning risk factor reduction and improvements in early detection and treatment of several cancers, socioeconomic inequalities persist in incidence and survival[1–5]. Low socioeconomic status (SES) has proven to be an important risk factor for developing upper aerodigestive tract cancer. It is also a risk factor for poor survival in patients with cancer of lung, stomach, and breast, as well as hepatocellular carcinoma. Furthermore it is associated with late presentation and recurrence in colorectal cancer.
Like in other countries, mortality rates from pancreatic cancer in the Netherlands remain high. The 1-year survival is 18% and only four percent of the patients is alive after five years (period 1989–2010). The incidence of pancreatic carcinoma is slightly increasing in the Netherlands; 8.9/100.000 persons in 1989 and 9.4/100.000 persons in 2015(ESR: European standard Ratio). Pancreatic resection is the only potentially curative treatment, but only a minority of the patients with pancreatic cancer are eligible for surgery.
Some studies showed that pancreatic cancer patients of low SES neighborhoods were less likely to receive surgical resection, chemotherapy, or radiotherapy, and had significantly higher perioperative and long-term mortality rates than patients with higher SES[2,14]. This may be the result of poorer access to health care for people with low SES, which results in delayed diagnosis and inferior treatment. Although we would not expect this to be the case in the Netherlands, as the obligatory insurance coverage should prevent inequalities in health care access, a higher resection rate was found in a study about stomach cancer in the Netherlands. We therefore aimed to evaluate the patients who underwent pancreatic resection. Since there is a known relationship between hospital volume and mortality after resection of pancreatic cancer[15,16], we investigated whether patients from higher SES groups were more often referred to high-volume university hospitals for treatment. To test these hypotheses we performed a nationwide study in the Netherlands using the population-based database from the Netherlands Cancer Registry.
All patients diagnosed with pancreatic carcinoma between January 1st, 1989 and December 31, 2011 (n = 34,757) were selected from the population-based Netherlands Cancer Registry, which contains data on all patients newly diagnosed with cancer in the Netherlands. The cancer registry obtains notifications from PALGA (Pathologisch Anatomisch Landelijk Geautomatiseerd Archief), the nationwide network and registry of histopathology and cytopathology in the Netherlands. Additional sources are the national registry of hospital discharge, which in general accounts for another eight percent of new cases, and–in a minority of cases- radiotherapy institutions. Information on patient characteristics and tumor characteristics such as subsite (International Classification of Diseases for Oncology (ICD-O-3), histology, stage (Tumor lymph Node Metastasis (TNM) classification), and grade, are obtained routinely from the medical records. In patients who underwent resection (14% of the total study group) stage was based on pathological TNM (I,II,III,IV). In the 86% unresectable patients, the clinical stage was based on clinical TNM or the one-dimensional Extent of Disease was used. These were combined into one variable with 4 categories; 1) ‘local’ = tumor irrespective of size but confined to the pancreas, classification T1 or T2 (TNM 6th and 7th edition); 2) ‘beyond pancreas’ = tumor extension beyond pancreas, classification T2 (according to TNM classification 4th and 5th edition), or T3 or T4, or positive lymph nodes; 3) metastasized disease, M1; and 4) unknown stage. The site of the tumor in the pancreas is reported as head (ICD-O C25.0), not head (ICD-O C25.1–78) and “not otherwise specified/overlapping” (C25.8–9).
The type of hospital in which the pancreatic resection was performed was categorized into: (1) university hospitals (n = 8), (2) non-university teaching hospitals (n = 46) and (3) non-university non-teaching hospitals (n = 46; classification in 2010). A teaching hospital is defined as any hospital which provides medical training to surgical residents to become board-certified surgeons. If the hospital of surgery was not registered, the hospital of diagnosis was assumed to be the hospital of surgery.
The quality of the data in the Netherlands Cancer Registry is high, due to thorough training of the administrators and computerized consistency checks at regional and national levels and completeness is estimated to be at least 95%. Follow-up of vital status of all patients was calculated as the time from the date of diagnosis to the date of death or to January 1st, 2013. The information on vital status was initially obtained from municipal registries and from 1995 onwards from the nationwide municipal population registries network.
Socioeconomic status (SES) scores
Postal code at time of diagnosis was used to determine SES. SES scores are available for each of the 4,002 four-digit postal code neighborhood in the Netherlands. SES scores were calculated in 2006, and these formed the basis for the current analysis. The mean number of inhabitants was 4080 per postal code in 2006. SES scores are provided by the Netherlands Institute for Social Research (Sociaal Cultureel Planbureau) and based on the following items which were collected per six-digit postal code: 1) mean annual income per household, 2) the percentage of households with a low income and 3) the percentage of households with a low education.
SES was divided into three groups based on the delivered rank numbers: low (1st-3rd deciles, n = 10,294), intermediate (4th-7th, n = 13,775) and high (8th-10th, n = 10,274) SES.
Associations between SES of neighborhood, stage of disease, localization of the tumor, histological grade of the tumor and treatment were analyzed by Chi-square analysis and calculation of 95% confidence intervals (95%CI). Association between SES and age at diagnosis were analyzed by one-way Anova analysis.
Overall 5-year survival rates were calculated. Cox’ regression models were used to compute multivariable rates (Hazard Ratio = HR, forward procedure) and 95% confidence intervals (95%CI). Statistical analysis was performed with Stata version 12.
In the period 1989–2011 34,757 patients were diagnosed with pancreatic cancer in the Netherlands. The mean age at diagnosis was 70 years (range 15 to 101 years). Most patients were diagnosed with metastatic disease (45%).
Table 1 shows differences in patient and tumor characteristics between different SES groups. The low SES group is characterised by more females, slightly older age, and less frequent resection of the tumor. Of note, 10% of the patients living in high SES neighborhoods underwent resection, whereas in the low SES neighborhoods this was 9% (p = 0.006, Table 1). Patients with higher SES were treated with adjuvant chemotherapy more often than patients with lower SES (22% versus 18%, p = 0.02).
The highest volume of operated patients was observed in university hospitals; eight hospitals performed 1,198 resections (= 35% (34–37% 95%CI)), as compared to 1,350 resections (= 40% (38–41% 95%CI)) in 46 non-university teaching hospitals and 833 resections (= 25% (23–26% 95% CI)) in 41 non-university non-teaching hospitals.
Table 2 summarizes the prognostic factors in all patients presenting with pancreatic cancer. Favorable prognostic factors in multivariable analysis were a medium and high SES and resectable tumors, while adverse prognostic factors were older age, non-head tumor localization, tumor differentiation other than well differentiated, and TNM stage I.
Table 3 shows the prognostic factors in 3,381 patients who underwent a resection. On multivariable analysis adverse prognostic factors were: operation in other than university hospitals and poorly differentiated tumors. Patients with TNM stage I disease had a more favorable prognosis compared to patients with other stages of disease. Of note, whereas SES was a risk factor in the whole group of pancreatic cancer patients, this was not the case anymore in the group of patients who underwent a resection (univariable analysis SES Low (reference) HR: 1; SES Medium HR1.04 (95% CI 0.9–1.1; SES High HR 0.98 (95% CI 0.89–1.07).
The main findings of this nationwide study in patients with pancreatic cancer are that low SES is an independent risk factor for poor survival, whereas in pancreatic cancer patients who underwent resection, SES is not a risk factor. These results confirm findings in other studies in which pancreatic cancer patients with lower SES tend to have worse survival and are less likely to receive adequate treatment (Table 4). In the two series reporting on patients who underwent resection the results are contradictory: the Lim paper  (n = 396 patients) describes lower survival in the low SES group, whereas in the Kuhn paper  (n = 117 patients) no effect was found.
Among the variables which are significantly different in the three SES groups–gender, age, tumor localization, differentiation, stage and resection—we consider only resection a clinical relevant variable, because this has a major impact on outcome. In this study, patients with low SES less often underwent resection and had therefore more often an unknown stage than patients with high SES. After adjustment for confounding factors, SES remained an independent prognostic variable. However, in patients with resectable tumors, SES was no longer a relevant variable for survival. This was also found in a case series of 117 patients in Germany in which the SES was obtained from individual patients.
We found that patients with higher SES were more often operated in a university hospital. Furthermore, resection for pancreatic cancer in a non-university (teaching or non-teaching) hospital was associated with an increased risk of dying as compared to resection in a university hospital. University hospitals have higher volumes of pancreatic resections, because of referral patterns for this type of surgery. The lower volumes in most of the non-university hospitals might be a part of the explanation for the worse outcome, because low volume is a well-known risk factor for poor outcome in various surgical procedures including pancreatic resections [15,16,24]. For example, a study in a population of patients with periampullary cancer in the Southern part of the Netherlands also revealed that patients who underwent resection in a university hospital had a better three-month survival as compared to patients treated in a low volume hospital (performing less than five resections a year) in the region. However, apart from differences in patient selection of patient referral, other explanations might be differences in staff-to-patient ratios, financial resources, more easy access to specialized diagnostic and treatment possibilities, and higher qualified intensive care units.
Health disparities originate in the complex interplay of patient, health-care provider and institutional factors. Major determinants of survival are tumor characteristics, and the presence of disseminated disease. Other factors like psycho-social and bio-behavioral influences, environmental influences, and income-related lifestyle factors may play a role in cancer pathogenesis[5,38,39]. Patients of lower SES probably have more associated comorbidity and additionally may lack the knowledge to comprehend the complex details of their diagnosis, may not be well informed about the possible improved outcome at university hospitals and may lack social support and structure. Data from a study in the United States showed that patients with pancreatic carcinoma with lower SES were treated more frequently in low volume hospitals, compared with patients with higher SES. However, when patients from lower SES were treated in a high volume and or teaching hospital, they still had increased perioperative mortality and shorter median survival. In a recently published paper on patients after pancreaticoduodenectomy, we could demonstrate that in the Netherlands 30-day mortality was 4.6% and 90-day mortality 7.8%. In that paper, describing only patients who underwent resection, SES was not predicting peri-operative death, neither 30-day mortality nor 90-day mortality. Thus, we extrapolate that postoperative mortality and SES are not predictors of cancer survival.
In our study, patients from higher SES neighborhoods were more likely to receive adjuvant chemotherapy. We cannot exclude the likelihood that patients who received adjuvant therapy were less ill and thus judged to be better candidates for such treatment. Treatment with adjuvant chemotherapy has been introduced in the Netherlands in 2007 after publication of the CONKO-001 study. Therefore only limited data are available precluding a multivariable analysis with adjuvant chemotherapy. Possibly this chemotherapy contributes to better survival in the latest period.
In 40% of patients the presumed diagnosis of pancreatic cancer could not be proven by histology or cytology, because most of the patients with pancreatic cancer are unresectable, which also explains that around 75% of the patients in each SES group has an unknown histological grade.
Limitations of our study are the way of measuring SES and the lack of information on specific referral patterns in relation to SES. We used an ecological measure of SES, which was assigned to each individual patient according to the postal code of residence at the time of diagnosis. Theoretically it is possible that individual people are misclassified and that inferences at the area level do not directly transfer to individuals. In practice, validation studies, however have found that an area-based measure of SES is a good indicator of SES for individuals. Another limitation is the absence of information on comorbidity. Comorbidity may be a factor underlying lower survival in cancer patients with low SES.
In conclusion, this study demonstrated that SES was an independent prognostic factor in a nationwide-study of a large cohort of pancreatic cancer patients in the Netherlands. Performing a pancreatic resection is the most important prognostic factor and is more frequently performed in patients with high SES versus patients with low SES. Remarkably, in patients who underwent resection, SES is no longer a risk factor. Resection in university hospitals is associated with the highest chance of survival. With the knowledge of the differences in treatment between patients with different SES, specialists treating patients with pancreatic cancer should make effort to explain the different treatment options to all patients groups clearly and afford referral to a center for pancreatic surgery.
The authors thank the registration team of the Netherlands Comprehensive Cancer Organisation (IKNL) for the collection of data for the Netherlands Cancer Registry (NCR) as well as IKNL staff for scientific advice.
- Conceptualization: KPDJ.
- Data curation: MAVDA LGMVDG.
- Formal analysis: MAVDA LGMVDG.
- Methodology: MAVDA LGMVDG.
- Project administration: MHGVR.
- Supervision: KPDJ.
- Validation: MAVDA LGMVDG.
- Visualization: MHGVR.
- Writing – original draft: MHGVR.
- Writing – review & editing: MHGVR MAVDA LGMVDG KPDJ.
- 1. Niu X, Pawlish KS, Roche LM. Cancer survival disparities by race/ethnicity and socioeconomic status in New Jersey. J Health Care Poor Underserved. 2010;21: 144–160. pmid:20173261
- 2. Cheung MC, Yang R, Byrne MM, Solorzano CC, Nakeeb A, Koniaris LG. Are patients of low socioeconomic status receiving suboptimal management for pancreatic adenocarcinoma? Cancer. 2010;116: 723–733. pmid:19998350
- 3. Clegg LX, Reichman ME, Miller BA, Hankey BF, Singh GK, Lin YD, et al. Impact of socioeconomic status on cancer incidence and stage at diagnosis: selected findings from the surveillance, epidemiology, and end results: National Longitudinal Mortality Study. Cancer Causes Control. 2009;20: 417–435. pmid:19002764
- 4. Murphy MM, Tseng JF, Shah SA. Disparities in cancer care: an operative perspective. Surgery. 2010;147: 733–737. pmid:19962161
- 5. Shapiro M, Chen Q, Huang Q, Boosalis VA, Yoon CH, Saund MS, et al. Associations of Socioeconomic Variables With Resection, Stage, and Survival in Patients With Early-Stage Pancreatic Cancer. JAMA Surg. 2015.
- 6. Conway DI, McKinney PA, McMahon AD, Ahrens W, Schmeisser N, Benhamou S, et al. Socioeconomic factors associated with risk of upper aerodigestive tract cancer in Europe. Eur J Cancer. 2010;46: 588–598. pmid:19857956
- 7. Yang R, Cheung MC, Byrne MM, Huang Y, Nguyen D, Lally BE, et al. Do racial or socioeconomic disparities exist in lung cancer treatment? Cancer. 2010;116: 2437–2447. pmid:20209616
- 8. Siemerink EJ, Hospers GA, Mulder NH, Siesling S, van der Aa MA. Disparities in survival of stomach cancer among different socioeconomic groups in North-East Netherlands. Cancer Epidemiol. 2011;35: 413–416. pmid:21470931
- 9. Aziz Z, Iqbal J, Akram M, Anderson BO. Worsened oncologic outcomes for women of lower socio-economic status (SES) treated for locally advanced breast cancer (LABC) in Pakistan. Breast. 2010;19: 38–43. pmid:19892552
- 10. Artinyan A, Mailey B, Sanchez-Luege N, Khalili J, Sun CL, Bhatia S, et al. Race, ethnicity, and socioeconomic status influence the survival of patients with hepatocellular carcinoma in the United States. Cancer. 2010;116: 1367–1377. pmid:20101732
- 11. Bowman KC, Tabrizian P, Telem DA, Boudourakis L, Divino CM. Health disparity in complicated colorectal cancer. Am Surg. 2010;76: 164–167. pmid:20336893
http://www.cijfersoverkanker.nl?language=en, Accessed September 2016
- 13. Bernards N, Haj Mohammad N, Creemers GJ, de Hingh IH, van Laarhoven HW, Lemmens VE. Ten weeks to live: a population-based study on treatment and survival of patients with metastatic pancreatic cancer in the south of the Netherlands. Acta Oncol. 2015;54: 403–410. pmid:25263080
- 14. Cress RD, Yin D, Clarke L, Bold R, Holly EA. Survival among patients with adenocarcinoma of the pancreas: a population-based study (United States). Cancer Causes Control. 2006;17: 403–409. pmid:16596292
- 15. Finks JF, Osborne NH, Birkmeyer JD. Trends in hospital volume and operative mortality for high-risk surgery. N Engl J Med. 2011;364: 2128–2137. pmid:21631325
- 16. de Wilde RF, Besselink MG, van der Tweel I, de Hingh IH, van Eijck CH, Dejong CH, et al. Impact of nationwide centralization of pancreaticoduodenectomy on hospital mortality. Br J Surg. 2012;99: 404–410. pmid:22237731
Wittekind C, Greene F, Hutter R, editors. International Classification of Diseases for Oncology. Berlin: UICC; 2005.
Sobin L, Hoppener P, editors. TNM Atlas. New York: UICC; Wiley-Liss; 2002.
- 19. Schouten LJ, Hoppener P, van den Brandt PA, Knottnerus JA, Jager JJ. Completeness of cancer registration in Limburg, The Netherlands. Int J Epidemiol. 1993;22: 369–376. pmid:8359950
Knol FA. From high to low; from low to high: the development of social status of neighbourhoods between 1971–1995 (In Dutch). 1998.
- 21. Lim JE, Chien MW, Earle CC. Prognostic factors following curative resection for pancreatic adenocarcinoma: a population-based, linked database analysis of 396 patients. Ann Surg. 2003;237: 74–85. pmid:12496533
- 22. Kuhn Y, Koscielny A, Glowka T, Hirner A, Kalff JC, Standop J. Postresection survival outcomes of pancreatic cancer according to demographic factors and socio-economic status. Eur J Surg Oncol. 2010;36: 496–500. pmid:19748206
- 23. Blot WJ, Fraumeni JF Jr, Stone BJ. Geographic correlates of pancreas cancer in the United States. Cancer. 1978;42: 373–380. pmid:667808
- 24. Janes RH Jr, Niederhuber JE, Chmiel JS, Winchester DP, Ocwieja KC, Karnell JH, et al. National patterns of care for pancreatic cancer. Results of a survey by the Commission on Cancer. Ann Surg. 1996;223: 261–272. pmid:8604906
- 25. Krzyzanowska MK, Weeks JC, Earle CC. Treatment of locally advanced pancreatic cancer in the real world: population-based practices and effectiveness. J Clin Oncol. 2003;21: 3409–3414. pmid:12972517
- 26. van Oost FJ, Luiten EJ, van de Poll-Franse LV, Coebergh JW, van den Eijnden-van Raaij AJ. Outcome of surgical treatment of pancreatic, peri-ampullary and ampullary cancer diagnosed in the south of The Netherlands: a cancer registry based study. Eur J Surg Oncol. 2006;32: 548–552. pmid:16569495
- 27. Zell JA, Rhee JM, Ziogas A, Lipkin SM, Anton-Culver H. Race, socioeconomic status, treatment, and survival time among pancreatic cancer cases in California. Cancer Epidemiol Biomarkers Prev. 2007;16: 546–552. pmid:17372250
- 28. Le H, Ziogas A, Rhee JM, Lee JG, Lipkin SM, Zell JA. A population-based, descriptive analysis of malignant intraductal papillary mucinous neoplasms of the pancreas. Cancer Epidemiol Biomarkers Prev. 2008;17: 2737–2741. pmid:18843017
- 29. Seyedin S, Luu C, Stabile BE, Lee B. Effect of socioeconomic status on surgery for pancreatic adenocarcinoma. Am Surg. 2012;78: 1128–1131. pmid:23025956
- 30. Cheung R. Racial and social economic factors impact on the cause specific survival of pancreatic cancer: a SEER survey. Asian Pac J Cancer Prev. 2013;14: 159–163. pmid:23534717
- 31. Enewold L, Harlan LC, Tucker T, McKenzie S. Pancreatic cancer in the USA: persistence of undertreatment and poor outcome. J Gastrointest Cancer. 2015;46: 9–20. pmid:25403828
- 32. Wolfson JA, Sun CL, Wyatt LP, Hurria A, Bhatia S. Impact of care at comprehensive cancer centers on outcome: Results from a population-based study. Cancer. 2015;121: 3885–3893. pmid:26218755
- 33. Markossian TW, O'Neal CM, Senkowski C. Geographic disparities in pancreatic cancer survival in a southeastern safety-net academic medical center. Aust J Rural Health. 2015.
- 34. Shahian DM, Nordberg P, Meyer GS, Blanchfield BB, Mort EA, Torchiana DF, et al. Contemporary performance of U.S. teaching and nonteaching hospitals. Acad Med. 2012;87: 701–708. pmid:22534588
- 35. van Roest MH, Gouw AS, Peeters PM, Porte RJ, Slooff MJ, Fidler V, et al. Results of pancreaticoduodenectomy in patients with periampullary adenocarcinoma: perineural growth more important prognostic factor than tumor localization. Ann Surg. 2008;248: 97–103. pmid:18580212
- 36. Effenberger KE, Schroeder C, Eulenburg C, Reeh M, Tachezy M, Riethdorf S, et al. Disseminated tumor cells in pancreatic cancer-an independent prognosticator of disease progression and survival. Int J Cancer. 2012;131: E475–83. pmid:21932421
- 37. Antoni MH, Lutgendorf SK, Cole SW, Dhabhar FS, Sephton SE, McDonald PG, et al. The influence of bio-behavioural factors on tumour biology: pathways and mechanisms. Nat Rev Cancer. 2006;6: 240–248. pmid:16498446
- 38. McCormack VA, Boffetta P. Today's lifestyles, tomorrow's cancers: trends in lifestyle risk factors for cancer in low- and middle-income countries. Ann Oncol. 2011;22: 2349–2357. pmid:21378201
- 39. Nilsen TI, Vatten LJ. A prospective study of lifestyle factors and the risk of pancreatic cancer in Nord-Trondelag, Norway. Cancer Causes Control. 2000;11: 645–652. pmid:10977109
- 40. Louwman WJ, Aarts MJ, Houterman S, van Lenthe FJ, Coebergh JW, Janssen-Heijnen ML. A 50% higher prevalence of life-shortening chronic conditions among cancer patients with low socioeconomic status. Br J Cancer. 2010;103: 1742–1748. pmid:20978508
- 41. van der Geest LG, Besselink MG, van Gestel YR, Busch OR, de Hingh IH, de Jong KP, et al. Pancreatic cancer surgery in elderly patients: Balancing between short-term harm and long-term benefit. A population-based study in the Netherlands. Acta Oncol. 2015: 1–8.
- 42. Oettle H, Neuhaus P, Hochhaus A, Hartmann JT, Gellert K, Ridwelski K, et al. Adjuvant chemotherapy with gemcitabine and long-term outcomes among patients with resected pancreatic cancer: the CONKO-001 randomized trial. JAMA. 2013;310: 1473–1481. pmid:24104372