Pharmacoeconomic evaluation of diabetic nephropathic patients attending nephrology department in a tertiary care hospitalJV Jose1, M Jose2, P Devi1, R Satish3
1 Department of Pharmacology, St. John's National Academy of Health Sciences, Bengaluru, Karnataka, India
2 Department of Pharmacology, Jubilee Mission Medical College and Research Institute, Thrissur, Kerala, India
3 Department of Nephrology, St. John's National Academy of Health Sciences, Bengaluru, Karnataka, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.194199
Source of Support: None, Conflict of Interest: None
Aims: To evaluate the cost of pharmacotherapy and its determinants in diabetic nephropathy (DN) in the nephrology department of a tertiary care hospital. Materials and Methods: A prospective observational study was conducted among adult patients visiting nephrology outpatient department (February-July 2015). Data on demography, investigations, and medications prescribed, direct cost and indirect costs were analyzed. We used Chi-squared test for categorical variables and multivariate linear regression analysis to identify determinants of cost of pharmacotherapy and total cost. Results: Of 100 patients, 50 were above 60 years and 75 were male. Ninety-seven patients had hypertension, which was the most common comorbidity. The majority (60 patients) belonged to Stage 5 DN and 59 patients were on dialysis. The mean number of drugs per patient was 7.60 ± 2.44. The total monthly cost per patient amounted to INR 24,203.27 with total direct cost of INR 21,013.90 (87%) and indirect cost of INR 3189.30 (13%). The monthly cost of dialysis and pharmacotherapy per patient were INR 9060.00 (37%) and INR 2535.98 (11%), respectively. Stage of DN (unstandardized coefficient, B = 7553.96, 95% confidence interval [CI] [6175.09-8932.82], P < 0.001) was a significant determinant of total cost. Number of drugs (B = 636.694, 95% CI [335.670-937.718], P < 0.001) and stage of DN (B = 852.986, 95% CI [297.043-1408.928], P = 0.003) were predictors of cost of pharmacotherapy. Conclusion: Stage of DN and number of drugs prescribed were major determinants of cost of pharmacotherapy.
Keywords: Diabetic nephropathy, direct cost, indirect cost, pharmacotherapy
Diabetic nephropathy (DN) is a common microvascular complication of diabetes. Approximately 40% of persons with diabetes develop DN.  It accounts for 44% of new cases of end-stage renal disease (ESRD) in the US.  DN is the most common cause of chronic renal failure in India. It accounts for 30% of patients with chronic renal failure seen in 10 years.  The Chennai Urban Population Study showed that mortality due to renal causes in diabetic patients were three times more than nondiabetic patients (23.5% vs. 6.1%, P = 0.072).  In the United Kingdom Prospective Diabetes Study conducted in a cohort of 5097 cases, the annual rate of progression from normo- to microalbuminuria was 2%, from micro- to macroalbuminuria was 2.8% and from macroalbuminuria to elevated serum creatinine was 2.3%. The annual death rate was 0.7% in normo-, 2.0% in micro-, 3.5% in macroalbuminuric patients and 12.1% in those with elevated serum creatinine.  The management of DN includes good glycemic control, tight control of blood pressure, and reduction of proteinuria with cessation of smoking, lipid control, salt and protein restriction. Therapeutic intervention is intended to prevent or retard the progression of diabetic renal disease as well as to reduce cardiovascular complications. The drugs used in the management of DN are antidiabetic drugs, antihypertensives, erythropoietin stimulating agents and iron therapy, lipid-lowering drugs, phosphate binders, Vitamin D analogs, and vitamin supplements such as folic acid and Vitamin B12.  Thus, pharmacotherapy in DN is complex and polypharmacy is commonly seen. Outcomes in DN are cardiovascular events such as angina, myocardial infarction, cardiac failure, neurological events such as stroke, coma, seizure, and death. 
The 2010 USRDS report shows that Medicare spent US dollars (USD) 29 billion in 2009, or almost 6% of the annual Medicare budget, for people with ESRD.  Across all stages of chronic kidney disease (CKD), the 2011 USRDS report shows that annual Medicare spending among adults aged 65 years or older was USD 20,432 per person.  A study on 1364 subjects with type 2 diabetes mellitus (T2DM) among Michigan health maintenance organization showed that ESRD treated with dialysis had 11-fold increase in cost compared with T2DM patients without complications.  The annual cost of treatment of DN is USD 16.8 billion and 1.2 billion in the US and the UK, respectively.  Among those undergoing dialysis, diabetic patients have to pay approximately 27% higher than nondiabetic patients.  In the post-CKD period, costs directly related to treatment of CKD accounted for 9%-19% of all-cause medical service costs - 9.2% for patients with diabetes, 11.6% for patients with hypertension, and 18.8% for patients with both diabetes and hypertension.
There are limited data on the economic burden of DN from India (PubMed search using Medical Subject Headings terms-cost analysis, economic evaluation, DN, India from 2000 to 2015). Hence, we aimed to determine the cost of pharmacotherapy, total cost of illness, and estimate the determinants of cost among patients with DN attending Nephrology Department of a tertiary care hospital.
We conducted a prospective observational study at St. John's Medical College, a tertiary care hospital in Bengaluru. The study involved 100 adult patients diagnosed with DN from the nephrology outpatient department (OPD). Patients undergoing peritoneal dialysis and those who underwent renal transplants were excluded from the study. Patient recruitment was done over 6 months from February to July 2015. The study was approved by the Institutional Ethics Committee with reference number - 115/2015. Informed consent was obtained from all the participants before recruitment.
A structured case report form was used to collect data including demographic and baseline details (age, gender, tobacco, and alcohol use), socioeconomic status (SES) (assessed using Kuppuswamy's SES scale), medical history (comorbidities, dialysis history, and family history), clinical laboratory investigations, and prescription data. Data on indirect cost (transportation and lost wages) are collected by direct patient interview. Direct cost is calculated based on hospital price list 2015.
Direct, indirect, and total monthly costs were calculated.
Direct costs included cost of dialysis, drugs, investigation, medical consultation, and arteriovenous (AV) fistula charges. The cost of management of adverse drug reactions (ADRs) was excluded as most of the ADRs were of mild category and none of them had any significant impact on the outcomes such as prolongation of hospital stay or disability. The medication cost arising from other comorbidities and overall investigation costs are included. However, OPD visits for other comorbidities are excluded as we conducted this study in the Nephrology Department only. The unit cost of dialysis was fixed as INR 1500 which included cost of dialyzer and tubings. The monthly cost of dialysis is calculated as frequency of dialysis multiplied by unit cost of dialysis. The details of drug treatment are obtained from prescription order in the patient chart. The unit cost of each drug was obtained from Current Index of Medical Service 2012. The dose prescribed daily and the duration of treatment of each drug is used to calculate monthly treatment cost. The cost of consultation was INR 120. The costs related to investigations (radiology, hematology, and biochemical) were taken from the "Hospital Price List" for the year 2015.
Indirect costs included cost of transportation and lost wages. Productivity loss (loss of wages) for attending the hospital and days off due to dialysis were determined based on the income of patients per month. Each outpatient visit and inpatient day was assumed to account for a full day of work loss. We used shadow pricing (taking only monetary value of the lost wages on per day basis as a true measure of loss of productivity without placing value on time and quality of productive work) to calculate the indirect costs, and the patients/caretakers loss of wages was used to value the loss of productivity. On average, one attendant (relative) accompanied the patient during dialysis. To calculate the productivity losses arising from outpatient visits or hospitalizations, the numbers of absent working days were multiplied by the loss of productivity value (income per day). The cost of travel was also included. The costs were collected from patient perspective. We used a bottom-up approach for cost analysis. We have calculated the cost for a month (short period), so discounting was not done in this study.
We used descriptive statistics to summarize demographic data, clinical and cost data. Categorical variables were compared using Chi-squared test. All continuous variables were checked for normality and compared using independent t-test and Mann-Whitney U-test. To assess the predictors and determinants of cost of pharmacotherapy and total monthly cost, we used multivariable linear regression analysis. For analysis of determinants, the dependent variable was total monthly cost and independent variables were patients' characteristics (age, gender, and SES), stage of DN, number of comorbidities. For the analysis of predictors, the dependent variable was cost of pharmacotherapy and independent variables were age, gender, SES, stage of DN, number of drugs, and cost of investigations. P < 0.05 was considered statistically significant for all tests. Statistical analyses were performed using commercially available software Statistical Package for the Social Sciences version 19.0 software (SPSS Inc, Chicago, IL, USA).
The mean age was 59 (standard deviation ± 10) years and majority (75 patients) were males. The baseline characteristics of patients are summarized in [Table 1]. Majority (64) were from middle class and lower socioeconomic class (25).
Majority of patients belonged to Stage 5 (60) followed by Stage 3 (19) [Figure 1]. Among these patients, 59 were on dialysis. Mean number of dialysis per month was 7.27 ± 5 and mean number of drugs per patient per day was 7.60 ± 2.44. Almost all patients had hypertension (97) as comorbidity followed by diabetic retinopathy (36) and coronary artery disease (27) [Figure 2].
The total monthly cost per patient amounted to INR 24,203 with total direct costs of INR 21.013 and the indirect cost per patient of INR 3189, respectively . The direct and indirect costs represented 87% and 13%, respectively, of the total cost of DN treatment. The monthly direct medical expenses included cost of dialysis (INR 9060), cost of medications (INR 5513), cost of investigations (INR 3364), and AV fistula charges (INR 3076). The monthly indirect costs include cost of travel (INR 1586) and lost wages due to hospital visits (INR 1603). The details of cost of therapy in DN are presented in [Table 2].
Stage of DN (unstandardized coefficient, B = 852.986, 95% confidence interval [CI] (297.043-1408.928), P = 0.003), number of drugs taken by the patient (B = 636.694, 95% CI (335.670-937.718), P < 0.001), and the cost of investigation (B = 1.710, 95% CI (0.822-2.599), P < 0.001) are found to significantly influence the cost of pharmacotherapy [Table 3]. The significant determinant of total cost was stage of DN (B = 7553.96, 95% CI (6175.09-8932.82), (P < 0.001) [Table 4].
The economic impact of diabetes is huge particularly as the major proportion of the costs related to diabetes are accounted for by the late stage complications arising from diabetes.  One of the recent population-based studies highlighted the median annual direct and indirect cost associated with diabetes care in India. The annual cost for diabetes was 1541.4 billion INR in 2010, by extrapolating the direct and indirect cost estimates to the Indian population.  Therefore, the cost of diabetes care is high and it further escalates in the presence of complications.
The present study was undertaken as there is a lack of sufficient information on pharmacoeconomic aspects of DN from India. It was found that monthly cost of dialysis contributed most (37%) to the total cost in our study. An earlier study from India reported that cost per dialysis in India ranges from INR 150 in government hospitals to INR 2000 in private hospitals.  In most of the private hospitals, the average cost of dialysis per patient per month is INR 12,000, and the yearly cost is INR 140,000 equivalent of USD 3000, and this is in sharp contrast to the annual cost of USD 60,000 in the United States and the United Kingdom.  In our study, monthly average cost of dialysis per patient was INR 9060. The patients on an average underwent eight dialysis sessions per month in our study which was comparable with study by Satyavani et al. done in a South Indian state. As per the Kidney Disease Outcomes Quality Initiative clinical practice guidelines,  patients should undergo 3-4 dialysis sessions per week. This was not affordable to many patients as around 89% belonged to lower to middle class socioeconomic strata in our study. A study by Satyavani et al.  showed that among patients on dialysis, 80% had monthly income below INR 20,000 and 47% below INR 10,000. The cost of diabetes treatment is an out-of-pocket expenditure for many patients in developing countries such as India. In the absence of insurance policies for diseases such as diabetes and meager financial support from the public health-care sector, patients spend from their personal savings and face a huge financial crisis.
The stage of DN and total number of drugs prescribed were major determinants of cost of pharmacotherapy in our study and majority belonged to Stage 5. A study by Kumpatla et al. showed that total expenditure on treating diabetes complications correlated significantly with duration of diabetes, number of hospitalized days and presence of complications.  The huge amount which is being spent on diabetes can be brought down by preventing patients from progressing to Stage 3 CKD and end-stage renal failure. In a study by Satyavani et al.,  the cost for a person on hemodialysis was four times higher than for persons with CKD only (INR 61,170 vs. 12,664). A study by O'Brien et al.  in Canada has shown that early stage complications (e.g., microalbuminuria: USD 62 event cost; USD 10 state cost) has relatively low financial burden compared to more costly advanced stages (e.g., end-stage renal disease, USD 63,045 state cost). Kidney transplantation remains the most cost-effective treatment for ESRD, offering considerable savings and improvement in quality of life in these patients  as shown in a study by Satyavani et al.  The total median cost of treatment of diabetics with ESRD was very high (on average INR 500,000) compared with transplant patient (INR 345,000) over a period of 2 years whereas the average cost of treatment of diabetes with and without CKD were INR 100,000 and INR 30,000, respectively. Improvement in diabetes control has the potential to reduce direct costs involved in the treatment of complications.  Strict glycemic and blood pressure control can reduce the incidence and slow the progression of DN.  Improvement in diabetes control, early diagnosis of DN, and treatment has the potential to reduce the direct cost involved in treatment of DN.  Patients should be made aware about the potential benefits of effective control of diabetes and its complications through various educational programs.
This study is first of its kind in India to evaluate the cost of pharmacotherapy in DN and its determinants and provides the baseline data for future studies. This study has got some limitations. First, this was a single center study and patterns of drug usage may be different in different health-care settings. The costs reported in this study are the actual charges incurred by the patient's to avail treatments. We did not discount the future costs due to the short duration of this study. In addition, cost of treatment of ADRs and intangible costs were not included in our study due to logistic reasons. We were not able to include the cost arising from visiting other departmental OPDs for comorbidities. The findings of this study can provide a frame work for large longitudinal studies to find out a cost-effective strategy for treating complications of DN such as CKD.
The cost of treatment of DN was substantial. Stage of DN and total number of drugs prescribed were major determinants of cost of pharmacotherapy.
The authors would like to thank Mrs. Sumithra, biostatistician, at St. John's Research Institute for conducting the statistical analysis.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]