Glycemic control is fundamental in the management of diabetes mellitus .If lifestyle intervention and full tolerated doses of one or two oral glucose lowering drugs (OGLDs) fail to achieve or sustain glycemic goals, insulin should be initiated. New insulin analogs are generated to improve glycemic control .New insulin analogs are generated to improve glycemic control,However, the cost of these analogs is a major problem .The aim of this piggy back evaluation was to assess the effect of BIAsp 30 versus NPH plus regular human insulin on metabolic control as well as its cost-effectiveness in people with type 2 diabetes in the Iranian setting.

Study design and participants This was a randomized controlled clinical trial of 48 weeks on subjects with T2DM between July 2011 and October 2012. Two hundred and four diabetic subjects aged 28-65 years were included in the study. Subjects had A1C >8.0% and had indication for starting insulin. They were randomly allocated to two groups using a simple randomization method. Any current and prior medications were acceptable for inclusion of participant except any type of insulin being evaluated. Demographic and anthropometric variables were recorded. Hypoglycemia events and laboratory data including FPG, HbA1c and lipid profiles were measured every three months. In addition, quality of life was assessed with self-administered standard EQ-5D questionnaire (Brooks, Rabin et al. 2003).

Subjects were excluded for any of the following criteria: Alteration in insulin sensitivity such as major surgery, infection, renal failure (Glomerular Filtration Rate < 50), glucocorticoid treatment, recent (within 2 weeks) serious hypoglycemic episode (requires assistance of another), simultaneous participating in another clinical study, using any type of insulin, sight or hearing impaired, active proliferative retinopathy or maculopathy require treatment within 6 months prior to screening, breast feeding, pregnancy or nursing of the intention of becoming pregnant or not using adequate contraceptive measures.

The study was approved by ethics committee of Tehran University of Medical Sciences and was recorded in IRCT (IRCT code: 201112038282N1). All eligible subjects agreed and signed written informed consent after full explanation of the purpose and nature of all procedures used. All patients were free to withdraw at will at any time.

Assessments and outcome measures The insulin therapies were prescribed by a single physician in the clinic. The starting dose of BIAsp 30 (NovoMix® 30-pen, NovoNordisk) was 0.2-0.6 unit/kg per day in 2 divided doses according to level of glycemia in the intervention group. The control group received NPH/Reg insulin (from Exir pharmaceuticals, Lorestan, Iran) in a Ratio of 2:1 with initiation dose of 0.2-0.6 unit/kg in 2 divided doses. Two-thirds of the dose was given before breakfast and the remainder before dinner in NPH/Reg insulin group. Changes to oral glucose lowering drugs (OGLDs) at the time of starting the insulin analog, or thereafter, were entirely at the discretion of the participant and physician. Para clinical data were measured in a referral laboratory every three months. After an overnight fast of at least 12 hours, blood samples were obtained for measurement of fasting blood sugar (FBS) using a the glucose oxidase method, lipid profileswere determined using enzymatic methods. These measurements were conducted using commercial Parsazmun kits (Tehran, Iran) and a Hitachi 704 automatic analyzer (Tokyo, Japan)(Lentjes, Harff et al. 1987). HbA1c was measured using the high performance liquid chromatography method(Esteghamati, Jamali et al. 2010). Post prandial blood glucose (PPG) was measured two hours after breakfast using a glucose analyzer (YSI 2700 Select, YSI, Inc., Yellow Springs, OH).

Trial visits were defined as 0, 12, 24, 36 and 48 weeks from baseline. All participants were asked to record their 7-point blood glucose values in three consecutive days before each visit. Seven-point self-monitoring blood glucose includes three pre-meals, three post-meals, and bedtime blood glucose values during each day. Insulin doses were adjusted by a titration regimen according to self-monitored blood glucose. For both groups, treatment goals were as follows: fasting blood glucose of 80-120 mg/dl, postprandial glucose <160 mg/dl, A1C<7% and the before dinner blood glucose target for the NPH/Reg insulin group was 100 mg/dl, with a stepwise increase of both type of insulin depending on the pre-meal blood glucose values to achieve targets for plasma glucose (PG)as follows: +2 IU/day where 126 mg/dl < PG < 140 mg/dl, +4 IU/day where 140 mg/dl < PG < 160 mg/dl, +6 IU/day where 160 mg/dl < PG < 180 mg/dl, +8 IU/day where 180 mg/dl < PG < 200 mg/dl and +10 IU/day where PG > 200 mg/dl(Hermansen, Davies et al. 2006; Liebl, Prager et al. 2009), unless symptoms of hypoglycemia occurred. Hypoglycemia was defined as blood glucose <70 mg/dl. Severe hypoglycemia was defined as an event with symptoms consistent with hypoglycemia where the individual required the assistance of another person and was not treated with oral carbohydrate due to confusion or unconsciousness and was associated with a blood glucose level <40 mg/dl with recovery with intravenous glucose, or glucagon administration. Nocturnal hypoglycemia was defined as hypoglycemia occurred at night and which commonly known as hypoglycemia while asleep.

Data were collected from the physician's clinical notes and participants' recall and self-monitoring blood glucose values at each visit.

Costs

Direct medical costs:

We collected medical costs of each patient by a checklist. All patients had been asked to refer to the clinic every one month during the study. Clinical events or hospital episodes and also all related costs were determined at each visit. Any pharmaceutical, laboratory/diagnostic and rehabilitative care, as well as any contact with specialists, general practitioners, nurses, opticians, and dieticians were recorded for patients with/without complication. Finally total costs were calculated.

Direct nonmedical costs Any services such as transportation for patients and their family to clinic and taking care of dependents were assessed for non-medical expenditures by a patient self-estimate questionnaire.

Indirect costs The lost productivity costs due to health problems of DM were determined by days absent from work, poor work performance, low earnings capacity from disabilities, and mortality. We calculated number of days in each visit that the patient could not be present in their job because of DM related health care. The average net hourly wage was asked from each patient. For unemployed patients, we considered average wage of population who were economically active and in employment (Javanbakht, Baradaran et al. 2011). Lost earnings owing to premature mortality were defined as the mortality costs. Costs from the health provider perspective, were converted from Iranian Rials (IRR) into USA dollar (USD) at an official exchange rate of 12,260 IRR/1USD 2012 to have an international comparison (Central Bank of Iran).

Utility calculation Utility scores were calculated with two recent well known measurements;EQ-5D and EQ-VAS between 0-1.These are two standardized measures of health status and quality adjusted life year (QALY), developed by the EuroQol group in order to provide a simple, generic measure of health for clinical and economic appraisal (1990).

Cost Effectiveness Analysis Hypoglycemic events and QALYs were considered as outcomes and the incremental cost effectiveness ratio (ICER) per patient was calculated according to the below formula: ICER= ∆Cost/ ∆Outcome (Rasccati Karen L. 2009).The comparison of ICERs was held afterwards per each outcome.

Analysis All data are presented as mean ± standard deviation (SD). Significant differences in general characteristics were determined by Chi-square and Student's t-test. Change from baseline HbA1c, FBS, PPBG, and blood lipids was analyzed using an analysis of variance (ANOVA) model with baseline characteristics as variates. SPSS for Windows (Version 14; SPSS Inc., Chicago, IL) was used for data analyses and P values < 0.05 were considered statistically significant.

Inclusion Criteria:

• diabetes type 2
• HbA1c 8% or higher
• age 25 to 65

Exclusion Criteria:

• alteration in insulin sensitivity such as major surgery, infection, renal failure (Glomerular Filtration Rate < 50),
• glucocorticoid treatment,
• recent (within 2 weeks) serious hypoglycemic episode (requires assistance of another),
• simultaneous participating in another clinical study,
• using any type of insulin,
• sight or hearing impaired,
• active proliferative retinopathy or maculopathy require treatment within 6 months prior to screening,
• breast feeding,
• pregnancy or nursing of the intention of becoming pregnant or
• not using adequate contraceptive measures.