Is simultaneous pancreas kidney transplant the most cost-effective strategy for type 1 diabetes patients with renal failure?

Abstract

Introduction:

Simultaneous pancreas kidney transplant (SPK) has shown beneficial outcomes in type 1 diabetes patients with renal failure (IDDM-RF). The objective of this study was to assess its cost-effectiveness compared with other treatment strategies for IDDM-RF.

Methods:

A decision analytic model was developed for IDDM-RF treatment with four possible strategies: deceased donor kidney transplant (DDKT), living donor kidney transplant (LDKT), SPK and dialysis. A cost-utility analysis from the healthcare provider perspective was conducted based on a five-year model. Local data were used whenever possible except for SPK survival variables, wherein data from United Network for Organ Sharing and Scientific Registry of Transplant Recipients were used. Sensitivity analyses were performed to evaluate the impact of uncertainties around key variables.

Results:

In the baseline analysis, LDKT was the most cost-effective strategy with the lowest cost per quality-adjusted life year gained, i.e. SGD77,068, followed by SPK (SGD82,991), DDKT (SGD92,432) and dialysis (SGD181,192). The DDKT was extended dominated by dialysis and LDKT strategies. Incremental cost-effectiveness ratios with dialysis as a reference for LDKT and SPK strategies were SGD43,094 and SGD56,201, respectively. Both strategies are considered highly cost-effective under World Health Organization (WHO) guidelines. In the sensitivity analysis, a 6% increase in both SPK kidney graft survival and patient survival would make SPK the most cost-effective strategy.

Conclusions:

Both LDKT and SPK are highly cost-effective strategies in the treatment of IDDM-RF. SPK is potentially the most cost-effective strategy if a 6% increase in both SPK kidney graft survival and patient survival is achieved.

Keywords

Cost-effectiveness simultaneous pancreas kidney transplant cost-utility analysis type 1 diabetes renal failure

Introduction

Pancreas transplant has been performed for over four decades and is a well-established procedure in diabetic patients.^1,2 The potential benefits of pancreas transplant are improved quality of life,^3–5 cost-effective treatment,⁶ and prolonged survival,⁷ It is also the only treatment that can render a normoglycemic state for type I diabetes mellitus.⁸ It has been most widely applied in type 1 diabetes with renal failure (IDDM-RF), particularly when done as a simultaneous pancreas kidney transplant (SPK) with increased survival rate compared with solitary kidney transplants or dialysis.^8–11 The rationale for a SPK is that it is an appropriate therapy for patients with IDDM-RF. The trade-off for such patients is the additional operative risk associated with pancreas transplant, as they are obligated to lifelong chronic immunosuppression due to their need for renal transplant. This additional operative risk can be justified as pancreas transplantation represents the gold standard of intensified insulin therapy, as it normalises glucose levels far better than any other strategy available for type 1 diabetes.^12,13

Pancreas transplant is widely available in many major US and European transplant centres, but at the time of the current study it is not yet available in Singapore or in the south-east Asian region. This may be partly attributed to a significant variation by at least 100-fold in the worldwide incidence of type 1 diabetes, being highest in Finland and Sardinia (Italy) and lowest in Venezuela and China.¹⁴ The limited adoption of pancreas transplant in Asia could be due to the very low incidence of type 1 diabetes in Asia.¹⁵ However, there is considerable evidence that the incidence of childhood type 1 diabetes is increasing globally, with the greatest increase in the youngest age group (0–4 years).¹⁶ This development is coupled with countries such as China and India which are starting to face increased childhood obesity, where autoimmunity, obesity and genetic factors may all contribute to early onset of type 2 diabetes; concern exists that increased onset of type 1 diabetes will follow suit.¹⁷ The age-standardised incidence rate for type 1 diabetes in Singapore was 2.46 per 100,000 children aged 0–12 years old, for the period 1992–1994. The data seemed to indicate a rising incidence in this population, being 1.4/100,000 in 1992, 2.4/100,000 in 1993 and 3.8/100,000 in 1994.¹⁸ This worrying trend in the region prompted the establishment of a national pancreas transplant programme in Singapore, leveraging on the existing national liver and kidney transplant programmes.

Even though studies from overseas have proven that SPK is a cost-effective treatment strategy for IDDM-RF,^19,20 there has been no analysis done in the region. The objective of the current study was to assess its cost-effectiveness compared with other treatment strategies for IDDM-RF in anticipation of a national pancreas transplant programme in Singapore.

Research design and methods

Model structure and assumptions

A decision analytic model was built for possible treatment strategies for IDDM-RF, consisting of deceased donor kidney transplant (DDKT), living donor kidney transplant (LDKT), SPK and dialysis. The model was based on the assumptions that all options are available to patients and that transplantations are performed and managed according to standard techniques and immunosuppressive regimens. The decision tree was adapted from Douzdjian and colleagues’ model.¹⁹ The time horizon of five years was adopted in this study and the model was performed from the perspective of the healthcare provider. The decision analytic model was analysed using TreeAge Pro software (TreeAge Inc., Williamstown, Massachusetts, USA) and the model structure is presented in Figure 1.

Figure 1.

Decision analytic model.

Probabilities

All patients and graft survival probabilities were based on five-year survival analyses, with the exception of “dies from operation or complication”, where complement of survival probability at one year was used. All survival values used in the baseline analysis were obtained from the Singapore Renal Registry data, except for all SPK survival variables where data from the United Network for Organ Sharing and Scientific Registry of Transplant Recipients (OPTN/SRTR) were used as no local data were available (Table 1).

Table 1.

Baseline survival probability and cost data with their variation used in sensitivity analysis.

Variable	Baseline values	Variation	Source
Survival probability, %
Kidney graft survival – LDKT	95.30	91.5–99.0	(24)
Kidney graft survival – DDKT	81.30	76.5–86.0	(24)
Kidney graft survival – SPK	78.9	±15	(31)
Pancreas graft survival – SPK	72.8	±15	(31)
Patient survival – LDKT	96.6	93.2–99.9	(24)
Patient survival – DDKT	89.1	85.1–93.0	(24)
Patient survival – SPK	86.6	±15	(31)
Patient survival – dialysis (diabetes nephropathy)	34.1	31.9–36.3	(24)
Patient survival – dialysis (non-diabetes)	65.9	63.5–68.3	(24)
One-year survival – LDKT	99.4	98.2–100	(24)
One-year survival – DDKT	96.0	93.8–98.1	(24)
One-year survival – SPK	95.1	±15	(31)
Cost data, SGD*
Cost of diabetes management per year	2795	±20%	(22)
Cost of dialysis (haemodialysis) per year	30,300	±20%	(23)
1st year DDKT cost	91,763	±20%	^
1st year failed DDKT cost	149,656	±20%	^
1st year LDKT costs	92,036	±20%	^
1st year failed LDKT cost	101,627	±20%	^
1st year SPK cost	130,303	±20%	Expert opinion
1st year failed SPK cost	175,908	±20%	Expert opinion
Annual LDKT follow-up cost	27,475	±20%	^
Annual DDKT follow-up cost	24,123	±20%	^
Annual SPK follow-up cost	27,741	±20%	Expert opinion
Total quality-adjusted life year at year 5, state
Dialysis and insulin free	4.72^#	Discount rate: 0–5%	(19)
Dialysis free, insulin dependent	2.83^#	Discount rate: 0–5%	(19)
Dialysis dependent, insulin free	2.36^#	Discount rate: 0–5%	(19)
Dialysis and insulin dependent	1.89^#	Discount rate: 0–5%	(19)

Note: All patients and graft survival probabilities were based on five-year survival, unless otherwise specified.

All costs were adjusted to 2010 value;

unpublished data abstracted from Medical Board Paper;

3% of discount rate had been used.

Outcomes and cost analysis

Outcomes for treatment strategies were measured in terms of total cost and quality-adjusted life year (QALY) in five years. The QALY is a measure of disease burden, including both the quality and the quantity of life lived. Health utilities for each treatment option were obtained from the previous study using the Standard Gamble method¹⁹ (Table 1).

Only direct medical costs related to the utilisation of healthcare resources were considered in this study. All cost components in the model were adjusted to 2010 values using the healthcare component of the Singapore Consumer Price Index.²¹ We adopted a 3% annual discount rate for all costs and effects beyond one year in the baseline case analysis, which converted values that would occur in the future to their present values. In the event of graft failure, costs were calculated based on three years of graft survival, with the assumption of graft failure happening in year 3 as suggested in the earlier study.¹⁹ Diabetes cost derived was based on the results of a local study on cost of diabetes treatment and related complications over a five-year period in patients with uncontrolled type 2 diabetes mellitus.²² The patients evaluated in the study were all on insulin, so the cost was assumed to be similar to the treatment cost of type 1 diabetes mellitus patient. The complication events considered in the study included myocardial infarction, stroke, microvascular events (such as retinopathy and nephropathy), heart failure, amputation and severe hypoglycaemia. Cost of dialysis was obtained from a Ministry of Health Singapore Information Paper based on a public hospital, i.e. National University Hospital Dialysis Centre’s costs data.²³ Haemodialysis cost was used as majority of the patients, i.e. 81.2% were on haemodialysis in Singapore.²⁴ Kidney transplant costs included the cost of transplant, re-hospitalisation, outpatient visits and drugs used (immunosuppressants and non-immunosuppressants) and were based on average cost of actual patients’ data (unpublished data abstracted from Medical Board Paper) from Singapore General Hospital, a public hospital with the biggest share of kidney transplantation in Singapore.²⁴ Cost of failed kidney was estimated from the cost of transplant with complications, based on actual patients’ data (unpublished data abstracted from Medical Board Paper). Based on expert opinion of the local surgical team, first-year SPK transplant cost and annual follow-up cost were estimated to be around 40% and 15% higher than the first-year DDKT and annual follow-up cost, respectively. First-year failed pancreas is estimated to be 35% higher cost than the functioning pancreas and kidney (average of 10% and 60% higher cost for failed LDKT and DDKT, respectively). Tables 1 and 2 summarise all the cost components used in the analysis. All indirect cost to society and individuals (e.g. loss of productivity, travelling costs, loss of work days) and intangible cost (pain and suffering due to illness) were excluded from the analysis.

Table 2.

Calculated annual follow-up cost for different graft outcome.

Strategy	Graft outcome	Annual follow-up cost, SGD	Remarks
DDKT	Functioning kidney	26,917	DDKT annual follow-up cost plus diabetes management cost
	Failed kidney	36,030	Dialysis cost plus diabetes management cost
LDKT	Functioning kidney	30,270	LDKT annual follow-up cost plus diabetes management cost
	Failed kidney	36,030	Dialysis cost plus diabetes management cost
SPK	Functioning pancreas and kidney	27,741	Expert opinion: 15% higher cost than the DDKT annual follow-up cost
	Failed kidney	60,977	SPK transplant follow-up cost plus dialysis cost
	Failed pancreas	30,536	SPK transplant follow-up cost plus diabetes management cost
	Failed pancreas and kidney	36,030	Dialysis cost plus diabetes management cost

Cost-utility analysis

The cost-effectiveness of each strategy was evaluated in terms of cost-utility ratio (CUR), i.e. cost per QALY gained. Incremental cost-utility ratio (ICUR) was also calculated for each treatment strategy versus the least costly strategy. The ICUR is defined as the additional cost incurred to achieve an extra unit of QALY gained and was calculated as follows:

{ICUR}_{A vs . B} = (Cost A - Cost B) / (QALY gained for A - QALY gained for B)

Under the World Health Organization (WHO) guidelines, ICUR below a value of 1 times gross domestic product (GDP) per capita is considered as “highly cost-effective” and below 3 times GDP per capita as “cost-effective”²⁵ (GDP per capita for Singapore in 2010 = SGD59,813).²⁶

Sensitivity analysis

Sensitivity analyses were performed to evaluate the impact of uncertainties around key variables, such as survival probability, cost data and QALY. For survival variables, the variations used in the sensitivity analysis were based on 95% confidence interval provided by the Singapore Renal Registry. For all SPK survival variables, variations were based on ±15% of the baseline values as the data were subject to a higher level of uncertainty, as no local data were available. For cost variables, the variations were calculated based on ±20% of the baseline values. The variations used for the QALY ranged from 0–5% of the discount rate. Threshold analyses were undertaken to determine the critical value(s) for the most cost-effective strategy in terms of the lowest CUR. Besides one-way sensitivity analysis, the two most critical variables which impact on the CUR were chosen for further evaluation using two-way sensitivity analyses.

Results

Baseline analysis

Expected values at five years for each treatment strategy in the baseline analysis are summarised in Table 3. Not surprisingly, dialysis strategy costs the least at five years (i.e. SGD116,777), while the SPK costs the most among all strategies (i.e. SGD251,099). However, dialysis also yielded the least QALY at five years (0.64) and the SPK the most at five years (3.03). From the modelling, the LDKT was the most cost-effective strategy with the lowest CUR (i.e. cost per QALY gained) followed by the SPK. Cost per QALY for LDKT was SGD77,068; SPK was SGD82,991; DDKT was SGD92,432; and dialysis was SGD181,192. The DDKT strategy was extended dominated by dialysis and LDKT strategies as it falls above the cost-effectiveness efficiency frontier (Figure 2). ICURs with dialysis as a reference for LDKT and SPK strategies were SGD43,094 and SGD56,201, respectively. Both strategies were therefore considered highly cost-effective under WHO guidelines.

Table 3.

Expected values in the baseline analysis.

Treatment option	Cost, SGD	Quality-adjusted life year, QALY	Cost-utility ratio, SGD	Incremental cost-utility ratio (versus dialysis), SGD
Dialysis	116,777	0.64	181,192	NA
DDKT	192,602	2.08	92,432	Dominated strategy
LDKT	201,900	2.62	77,068	43,094
SPK	251,099	3.03	82,991	56,201

Figure 2.

Cost-utility analysis for IDDM-RF treatment strategies.

Sensitivity analysis

Table 4 summarises the results of one-way sensitivity analyses performed across wide ranges of parameters. Dialysis, being the least sensitive strategy for the CUR, remained unchanged in most of the ranges tested in the sensitivity analyses. In contrast, SPK is the most sensitive strategy for most variables tested for impact on the CUR, and the impact of each variable is illustrated in a tornado diagram (Figure 3). As illustrated in the tornado diagram, kidney graft survival probability and patients’ survival probabilities for SPK strategy were the most influential variables on the CUR. In the threshold analyses, with a 10% increase in kidney graft survival (86.8% versus 78.9% used in the baseline) or 12% increase in patient survival (97.2% versus 86.6% used in the baseline), SPK would become the most cost-effective strategy in terms of the lowest CUR. Based on cost components, SPK would become the most cost-effective strategy, if the LDKT annual follow-up cost is 16% higher (SGD31,871 versus SGD27,475 used in the baseline) or first-year LDKT cost was 18% higher (SGD108,480 versus SGD92,036 used in the baseline) than the baseline values. In the two-way sensitivity analysis, when the two most critical variables, namely SPK kidney graft survival and SPK patient survival, varied concurrently, a 6% increase in both SPK kidney graft survival and patient survival was the most optimistic combination, which would make SPK the most cost-effective strategy (Table 5).

Table 4.

One-way sensitivity analysis on cost-utility ratio for different treatment strategies.

	Sensitivity range	Cost-utility ratio, SGD
	Sensitivity range	Dialysis	DDKT	LDKT	SPK
1st year failed DDKT cost, SGD	119,725–179,587	181,192	89,854–95,011	77,068	82,991
1st year failed LDKT cost, SGD	81,302–121,952	181,192	92,432	76,705–77,430	82,991
1st year failed SPK cost, SGD	140,726–211,090	181,192	92,432	77,068	77,748–88,234
1st year DDKT costs, SGD	73,410–110,116	181,192	85,206–99,659	77,068	82,991
1st year LDKT costs, SGD	73,629–110,443	181,192	92,432	70,370–83,766	82,991
1st year SPK cost, SGD	104,242–156,362	181,192	92,432	77,068	78,262–87,721
Annual DDKT follow-up cost, SGD	19,298–28,948	181,192	85,274–99,590	77,068	82,991
Annual LDKT follow-up cost, SGD	21,980–32,970	181,192	92,432	69,617–84,519	82,991
Cost of dialysis	24,240–36,360	148,014–214,370	91,796–93,068	76,936–77,199	82,416–83,566
Probability of death from operation/ complications, %:
• DDKT	1.9–6.2	181,192	91,490–93,465	77,068	82,991
• LDKT	0–1.8	181,192	92,432	76,857–77,497	82,991
• SPK	3.74–5.06	181,192	92,432	77,068	82,593–83,395
Kidney graft survival (DDKT), %	76.5–86.0	181,192	87,679–97,703	77,068	82,991
Kidney graft survival (LDKT), %	91.5–99.0	181,192	92,432	74,910–79,419	82,991
Kidney graft survival (SPK), %	67.07–90.74	181,192	92,432	77,068	74,147–93,534
Pancreas graft survival (SPK), %	61.88–83.72	181,192	92,432	77,068	78,008–88,491
Patient survival, %
• Dialysis (diabetes)	31.9–36.3	172,125–191,510	92,158–92,709	77,022–77,113	82,940–83,042
• Dialysis (non-diabetes)	63.5–68.3	181,192	92,432	77,068	82,824–83,160
• DDKT	85.1–93.0	181,192	89,444–95,766	77,068	82,446–83,559
• LDKT	93.2–99.9	181,192	92,432	75,181–79,150	82,991
• SPK	73.61–99.59	181,192	92,432	77,068	75,742–92,055
QALY, state
• Dialysis and insulin free	4.55–5.0	181,192	92,432	77,068	79,493–85,270
• Dialysis free, insulin dependent	2.7–3.0	181,192	87,470–95,624	72,748–79,857	82,147–83,496
• Dialysis dependent, insulin free	2.27–2.5	181,192	92,432	77,068	82,231–83,231
• Dialysis and insulin dependent	1.82–2.0	171,227–188,161	92,135–92,623	77,016–77,101	82,935–83,027

Figure 3.

Tornado diagram for sensitivity analysis on SPK treatment strategy.

Table 5.

Two-way sensitivity analysis for two most critical variables on cost-utility ratio.

Variation of SPK kidney graft survival probability	0.671 (-15%)	0.694 (-12%)	0.718 (-9%)	0.742 (-6%)	0.765 (-3%)	0.789 (0%)	0.813 (+3%)	0.836 (+6%)	0.860 (+9%)	0.884 (+12%)	0.907 (+15%)
Variation of SPK patient survival probability
0.736 (–15%)	$103,131	$100,763	$98,475	$96,262	$94,122	$92,051	$90,045	$88,102	$86,218	$84,391	$82,618
0.762 (–12%)	$101,038	$98,688	$96,420	$94,228	$92,110	$90,061	$88,078	$86,158	$84,298	$82,496	$80,748
0.788 (–9%)	$99,037	$96,707	$94,459	$92,289	$90,192	$88,166	$86,206	$84,310	$82,474	$80,696	$78,972
0.814 (–6%)	$97,123	$94,813	$92,586	$90,437	$88,363	$86,359	$84,423	$82,550	$80,738	$78,984	$77,285
0.840 (–3%)	$95,290	$93,001	$90,795	$88,668	$86,616	$84,635	$82,722	$80,872	$79,084	$77,353	$75,678
0.866 (0%)	$93,534	$91,264	$89,080	$86,975	$84,945	$82,987	$81,097	$79,271	$77,506	$75,799	$74,147
0.892 (+3%)	$91,848	$89,600	$87,437	$85,354	$83,347	$81,411	$79,544	$77,741	$75,999	$74,315	$72,687
0.918 (+6%)	$90,230	$88,003	$85,861	$83,800	$81,815	$79,902	$78,058	$76,277	$74,558	$72,898	$71,292
0.944 (+9%)	$88,675	$86,469	$84,349	$82,310	$80,347	$78,457	$76,634	$74,876	$73,180	$71,542	$69,958
0.970 (+12%)	$87,179	$84,995	$82,896	$80,879	$78,938	$77,070	$75,270	$73,534	$71,860	$70,243	$68,682
0.996 (+15%)	$85,740	$83,577	$81,500	$79,504	$77,585	$75,739	$73,960	$72,247	$70,594	$68,999	$67,460

Note: Scenarios when SPK would become the most cost-effective strategy are in bold.

Discussion

To the best of our knowledge, the current decision analytic model is the only economic evaluation done in the region for IDDM-RF treatment strategies. Douzdjian and colleagues’ model was adapted in the current decision analytic model for comparison with our healthcare system. In the current model, LDKT is the most cost-effective strategy in the baseline analysis, with SPK being the most cost-effective strategy in some of the variations used in the sensitivity analyses. This deviates from analysis based on the US healthcare system which had shown that SPK remained the most cost-effective strategy in the baseline analysis as well as across all variations used in the sensitivity analyses. However, geographic transferability of cost-effectiveness analyses from one country to other countries is subject to validity issues. Hence, blind application across geographies with vastly different healthcare financing and delivery systems is probably ill-advised.²⁷

Our analysis has a few strengths over the earlier study. First, the current model has taken into account the discount rate for both costs and effects to reflect the fact that dollars and health effects occurring in the future should not weigh as heavily as dollars and health effects occurring today, due to the existence of time preference as recommended in the literature.²⁸ Secondly, actual patient-level data were used in the current model, and an effort was made to distinguish costs for LDKT and DDKT where the earlier study weighted the same cost for these two types of transplants. Likewise, an effort has been made to differentiate the costs for a functioning and failed kidney in the current study.

There are several limitations in the current analyses. It is likely that our effectiveness for SPK strategy was underestimated as conservative approaches have been adopted in the current study. As no SPK data are available in Singapore at the time of this evaluation, OPTN data had been used in the baseline analysis. As five-year survival rates for kidney graft were higher in Singapore (95.3% and 81.3% for living and cadaveric kidney transplant, respectively) than those reported by the OPTN (80.8% and 67.9%, respectively), likely in part due to the stringent selection criteria for potential recipients,²⁹ one of our expert sources suggested that kidney graft survival for SPK in Singapore could be similarly higher than the data provided in the OPTN. Besides, patients selected to undergo SPK would generally be younger in age than the typical kidney transplant patients; it is thus plausible that the kidney survival outcomes for SPK could be better than the DDKT locally. In addition, kidney graft survival rate for SPK in the OPTN data was also shown to be higher than the kidney graft survival for DDKT alone (78.9% versus 67.9%). Therefore, using SPK kidney graft survival rate adopted from OPTN data, i.e. 78.9% (which is lower than the kidney graft survival rate for DDKT in Singapore, i.e. 81.3%), the current analysis could be an underestimate for possible SPK kidney graft survival in Singapore. On the other hand, one-way sensitivity analyses have shown that with a slight increase of 10% from the baseline for SPK kidney graft survival, i.e. 86.8% instead of 78.9% used in the baseline, or 12% increase in SPK patient survival rate, i.e. 97.2% versus 86.6% used in the baseline, SPK would become the most cost-effective strategy. Similarly, SPK would become the most cost-effective strategy with a 6% increase in both SPK kidney graft survival and patient survival in the two-way sensitivity analyses. As the variation of Singapore data and OPTN data is probably between 10–12% based on both countries’ registry data, this is within the comfortable zone to interpret that SPK could be the most cost-effective strategy in Singapore. Secondly, even though studies have been conducted locally to evaluate quality of life of kidney transplant patients, a profile-based quality of life instrument was used which does not place wellness on a continuum and has multiple-outcome dimensions, i.e. no utilities values could be determined from the study for the QALY calculation.³⁰ Furthermore, SPK transplant has not been performed in Singapore, making data collection for this group of patients unfeasible. Consequently, utilities values used in the current analysis were based on the earlier study conducted overseas. To overcome this shortcoming, a sensitivity analysis was conducted to test the robustness of the study findings, for which the results remained unchanged within the wide variations tested.

Conclusions

In conclusion, both LDKT and SPK are considered highly cost-effective strategies in the treatment of IDDM-RF under WHO guidelines. Compared with dialysis, LDKT and SPK cost an extra SGD43,094 and SGD56,201 per QALY gained, respectively. The study model has taken into consideration the risk and complications as well as cost involved for the procedures. This cost-effectiveness analysis could contribute to the decision-making process, in this case, in the evaluation of Singapore’s new national pancreas transplant programme. Once the centre is established, the service will not only benefit patients in Singapore, but around the regions.

Footnotes

Declaration of Conflicting Interest

None declared.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Sutherland

Gruessner

Dunn

. Lessons learned from more than 1,000 pancreas transplants at a single institution. Ann Surg 2001; 233: 463–501.

Gruessner

Sutherland

Gruessner

. Pancreas transplantation in the United States: A review. Curr Opin Organ Transplant 2010; 15: 93–101.

Milde

Hart

Zehr

. Pancreatic transplantation. Impact on the quality of life of diabetic renal transplant recipients. Diabetes Care 1995; 18: 93–95.

Kiebert

van Oosterhout

van Bronswijk

. Quality of life after combined kidney-pancreas or kidney transplantation in diabetic patients with end-stage renal disease. Clin Transplant 1994; 8: 239–245.

Ziaja

Bozek-Pajak

Kowalik

. Impact of pancreas transplantation on the quality of life of diabetic renal transplant recipients. Transplant Proc 2009; 41: 3156–3158.

Kiberd

Larson

. Estimating the benefits of solitary pancreas transplantation in nonuremic patients with type 1 diabetes mellitus: A theoretical analysis. Transplantation 2000; 70: 1121–1127.

Smets

Westendorp

van der Pijl

. Effect of simultaneous pancreas-kidney transplantation on mortality of patients with type-1 diabetes mellitus and end-stage renal failure. Lancet 1999; 353: 1915–1919.

Sutherland

Gruessner

. Pancreas transplantation for treatment of diabetes mellitus. World J Surg 2001; 25: 487–496.

Martins

Henriques

Dias

. One hundred eleven simultaneous pancreas-kidney transplantations: 10-year experience from a single center in Portugal. Transplant Proc 2011; 43: 205–208.

10.

Arjona

Martinez-Cecilia

Ruiz-Rabelo

. One hundred one simultaneous pancreas-kidney transplantations: Long-term outcomes at a single center. Transplant Proc 2009; 41: 2463–2465.

11.

Navarro

Perez

del Castillo

. Simultaneous pancreas-kidney transplant compared with kidney transplant in type I diabetic patients with end-stage renal disease. Transplant Proc 2002; 34: 204–205.

12.

Aguera

Navarro

Perez-Calderon

. Simultaneous pancreas-kidney transplant: A single-center long-term outcome. J Nephrol 2007; 20: 173–176.

13.

Demartines

Schiesser

Clavien

. An evidence-based analysis of simultaneous pancreas-kidney and pancreas transplantation alone. Am J Transplant 2005; 5: 2688–2697.

14.

Borchers

Uibo

Gershwin

. The geoepidemiology of type 1 diabetes. Autoimmun Rev 2010; 9: A355–A365.

15.

Park

. Why is type 1 diabetes uncommon in Asia? Ann N Y Acad Sci 2006; 1079: 31–40.

16.

Chan

. Diabetes: Incidence of childhood type 1 diabetes: A worrying trend. Nat Rev Endocrinol 2009; 5: 529–530.

17.

Chan

Malik

Jia

. Diabetes in Asia: Epidemiology, risk factors, and pathophysiology. JAMA 2009; 301: 2129–2140.

18.

Lee

Ooi

Thai

. The incidence of IDDM in Singapore children. Singapore Med J 1998; 39: 359–362.

19.

Douzdjian

Ferrara

Silvestri

. Treatment strategies for insulin-dependent diabetics with ESRD: A cost-effectiveness decision analysis model. Am J Kidney Dis 1998; 31: 794–802.

20.

Douzdjian

Escobar

Kupin

. Cost-utility analysis of living-donor kidney transplantation followed by pancreas transplantation versus simultaneous pancreas-kidney transplantation. Clin Transplant 1999; 13: 51–58.

21.

Department of Statistics Singapore. Statistics, Prices (Themes), http://www.singstat.gov.sg/stats/themes/economy/ess/essa31.pdf (accessed 30 May 2011).

22.

Chow

Goh

Ying

. Treatment of poorly controlled type 2 diabetes (T2DM) patients with insulin glargine – An economic analysis, http://www.singhealth.com.sg/Research/HealthServicesResearch/Publications/Documents/Glargine%20Cost%20Analysis.pdf (accessed 26 April 2011).

23.

Chowdhury

. MOH Information Paper: 2006/ 013. Kidney Dialysis, http://www.moh.gov.sg/mohcorp/uploadedFiles/Publications/Information_Papers/2006/Kidney_Dialysis_0.pdf (accessed 26 April 2011).

24.

National Registry of Disease Office, Health Promotion Board. Six report of the Singapore renal registry 2005/2006, http://www.nrdo.gov.sg/uploadedFiles/NRDO/Publications/SRR%202005%202006%20Report%20v1.2.4%20d20080524.pdf (accessed 26 April 2011).

25.

WHO. Making choices in health: WHO guide to cost-effectiveness analysis. Geneva: World Health Organization, 2003.

26.

Department of Statistics. Key Annaul Indicators, http://www.singstat.gov.sg/stats/keyind.html#keyind (accessed 30 May 2011).

27.

O’Brien

. A tale of two (or more) cities: Geographic transferability of pharmacoeconomic data. Am J Manag Care 1997; 3(Suppl): S33–S39.

28.

Drummond

Sculpher

Torrance

. Methods for the economic evaluation of health care programmes. 3rd ed. New York: Oxford University Press, 2005.

29.

Vathsala

Chow

. Renal transplantation in Singapore. Ann Acad Med Singapore 2009; 38: 291–299.

30.

Ong

van der Erf

Chow

. Will health-related quality of life better after kidney transplantation among patients in Singapore? Proceeding of Annual Scientific Meeting, Singapore, 8–9 April 2011.

31.

U.S Department of Health and Human Services. OPTN/SRTR Annual Report. Survival Rate Data Tables. Available at: http://optn.transplant.hrsa.gov/ar2008/survival_rates.htm. Accessed 20 April 2011 .