3) Formulary decisions:

Provinces, hospitals, insurers, and other payers could use pharmacoeconomic evidence in determining new listings, whether to continue listings, or what portion of the cost of a given drug product they are willing to pay.

4) Clinical guidelines for prescribing decisions:

Those preparing clinical guidelines for providers need to consider not only the efficacy and effectiveness of clinical alternatives (including drugs), but also their cost-effectiveness. High quality pharmacoeconomic studies can provide important input into the development of such guidelines.

5) Post-marketing surveillance:

There is a need to monitor continuously the performance of drugs and to update periodically the pharmacoeconomic studies with accumulating evidence from actual utilization experience, including unanticipated effects, both beneficial and adverse. These updated studies, in turn, can be used to update decisions on pricing, formulary listings and clinical guidelines.

CUSTOMIZED PHARMACOECONOMIC MODELS¹¹

These modes provide the foundation of pharmacoeconomics, anyone individual model often does not meet the specific needs of the customers of pharmacoeconomic data. Although the traditional models are critical in the design of pharmacoeconomic projects, the following definitions are a more practical description of the types of research projects often undertaken.

Cost of care evaluations provide models for the current cost of providing care for a particular disease state. An important distinction between a cost of care versus a cost of illness (COI) evaluation is the fact that a COI study will often be conducted from a societal perspective and include associated indirect costs secondary to the disease. In a cost of care model, the perspective is more often that of the payer and, therefore, indirect costs are of less interest. These types of projects provide information of benefit to both the MCO and the pharmaceutical industry. The MCO gains a detailed perspective on how a certain disease is managed within the plan, and the pharmaceutical industry obtains information on the potential economic impact of its product.

Phase III economic trials are those that collect data on economic parameters of drug therapy from the patients participating in Phase III clinical trials. This provides valuable economic information about the drug at the time of launch. A disadvantage of this approach is that the economic parameters may be driven by the clinical trial protocol and therefore may not reflect "real world" use of the drug. Drummond¹² summarizes the specific methodologic issues that arise when integrating economic and clinical research such as design, collection of resources-use data, collection of outcomes data, and the interpretation and extrapolation of results.

Naturalistic prospective evaluations are designed to address the limitation of Phase III economic trials by prospectively collecting the economic parameters of a drug and its most relevant therapeutic alternative within a specific practice setting with minimal intervention. Although these studies can often provide valuable information towards therapeutic decisions, a disadvantage is that they can be time consuming and expensive.

Retrospective database analysis of prescription and medical claims within a specific practice setting can provide useful information in a short period of time at little expense. Information on the trends of prescribing patterns can be particularly helpful. A disadvantage is that claims may not be an accurate refection of actual care received if claims are submitted to maximize reimbursement.

Decision analysis modeling is an approach that uses information from epidemiologic studies, clinical trials, administrative claims and cost databases, and expert opinion to model current care and the impact of a specific new therapy. These models can be very useful in designing prospective economic trials as well as offering predictions on the impact of a pharmaceutical agent in the treatment of disease. Because it is not always possible to study all the effects of treatment with clinical trials, modeling techniques can be very useful in making therapeutic decisions. These principles were recently reviewed by Stergachis¹³

TIMING OF PHARMACOECONOMIC STUDIES:

Pharmacoeconomic studies can be undertaken at any point in a product's life cycle. The timing of studies depends upon the needs of the users of studies. Early studies, during the research and development (R&D) phase of the drug, may be undertaken by the company to guide future (i.e. Phase III and IV) R&D decisions and marketing planning. Phase III studies may have a particular role in pricing and formulary decisions early in the product’s life cycle. These studies may also play a role in initial clinical and prescribing guidelines. Phase IV pharmacoeconomic studies (post-marketing) would contribute by updating previous studies on the basis of the new effectiveness data; and provide better evidence regarding utilization and adverse events. These post-marketing review studies could be scheduled on the basis of time (3 to 5 years after the product is marketed), or on the basis of "trigger" events (changes in medical practice, costs, comparator[s], or the emergence of new adverse or beneficial events).

INDICATIONS BY TARGET POPULATION:

A pharmacoeconomic study clearly specify the target population for the drug. Target populations may be defined using baseline epidemiologic features describing the type of patient (e.g. age, gender, socio-economic status), with a specific disease, of a certain severity, with or without other co-morbidities or risk factors, their geographic distribution, usual compliance rates, typical patterns of treatment, and so on. Target population subgroups which are defined based on effectiveness (from previous research), cost and/or preferences may differ in terms of the cost effectiveness of an intervention used in those subgroups.

While subgroup differences may be important considerations for decision-makers ¹⁴, the precision of the cost-effectiveness estimate may be compromised by inadequate statistical power due to inadequate sample size. If these competing factors can be balanced in the study development phase, then subgroup analysis should be investigated. Because a drug may be cost effective for some subgroups of patients and not for others, it is important to identify clearly the groups under study a priori and, when appropriate, to undertake separate analyses for different groups.

The issue of subgroup analysis is a source of particular contention in economic evaluation. There are questions as to whether these analyses are statistically sound; and concerns that recommendations based on subgroup analysis may be misleading and result in harmful clinical or economic decisions ¹⁵. Therefore, caveats must be noted before subgroup analysis is contemplated.

TREATMENT COMPARATOR:

There may be a variety of relevant comparators for a drug, and they may differ across the various subgroups of patients. As previously noted, relevant comparators may include other drugs, other medical care (e.g. surgery or watchful waiting), and no treatment. In theory, all other possible treatments for the same patients are relevant comparators. In practice, studies will have to identify one, or a small number, of primary relevant comparators.The issue of relevant comparators is complicated because there are two possible questions. Is the new drug cost-effective relative to the existing drugs or treatments that it will in fact replace (local cost-effectiveness)? Or, is the new drug cost-effective relative to optimally cost-effective treatment (global cost-effectiveness)? Pharmacoeconomic researchers are encouraged to investigate both local and global cost-effectiveness of the new drug.

In the ideal situation, one would compare the current most cost-effective option (as reflected [theoretically] in current practice guidelines or criteria for use) to the new agent. Practically, one often cannot identify such a comparator and, therefore, will use the agent with the lowest treatment costs (i.e. the sum of drug costs, administration costs, and the costs of treating any side effects) for a given course of therapy. This is more appropriate than using the drug with the lowest unit price as the comparator. However, even choosing the lowest cost comparator can be difficult. The selection of an appropriate comparator requires input from the decision-makers, as the choice of comparator relates to the question(s) the target audience wants answered. Thus, analysts are encouraged to obtain input from decision-makers as they develop their research protocols.

In selecting comparators and interpreting incremental comparisons it is crucial to understand the concept of dominance ^16-18. The assessment of dominance is based on a comparison of the costs and effectiveness of each option. An option that has higher costs and lower effectiveness than another single option is said to be strongly dominated by that option. Weak dominance can occur when a new treatment has the same incremental cost but greater incremental effectiveness; or a lower incremental cost but the same incremental effectiveness. Finally, a more complicated type of dominance arises if an option is not dominated by any other single option, but rather is dominated by a weighted average of two other options. For example, option B might be dominated by a 50/50 weighted average of options A and C. This is known as extended dominance. The importance of the concept of dominance in cost-effectiveness and cost-utility analysis is that all dominated options, strong weak and extended, are “inefficient”. In the analysis, dominated options are all ruled out immediately. The non-dominated options form the efficient frontier, and the incremental cost-effectiveness or cost-utility ratios are formed along the efficient frontier. So, the practical implication is that a dominated option is never appropriate even as a comparator. The concepts of dominance are difficult to describe and understand in words, but easy to see graphically. For this reason, a graphical representation of cost-effectiveness and cost-utility results is strongly recommended within the final report of the evaluation¹⁷.

OUTCOMES OF PHARMACOECONOMIC STUDIES:

Efficacy versus Effectiveness:

Efficacy refers to the performance of a drug under highly controlled circumstances - that is, administered according to a strict written protocol by highly motivated, research-oriented clinicians to consenting, compliant patients who are a carefully selected subgroup of patients meeting restrictive inclusion and exclusion criteria. Effectiveness, on the other hand, refers to the performance of a drug. in the real world with a wide variety of providers administering the drug as they see fit to a broad heterogeneous group of patients who are less well-informed, less compliant, and liable to be influenced by a variety of concomitant diseases and/or medications not investigated in the original efficacy trials^19-22.

Pharmacoeconomic studies should use effectiveness data as their source of clinical evidence regarding the impact of an intervention. Unfortunately, the only data available prior to the launch of a new product are Phase III efficacy data. Thus, pre launch pharmacoeconomic studies must extrapolate from trial efficacy to utilization effectiveness using modeling techniques²³.

The process of obtaining efficacy or effectiveness data can present its challenges. Some would argue that prospective data reflecting the “real-life” experience of drug use (i.e. effectiveness) in a large number of patients are most desirable, while those derived prospectively from several large randomized controlled trials (RCTs; i.e. efficacy) would be next best. Retrospective data from either effectiveness or efficacy data sources represent viable but not ideal alternative information sources. In practical terms, the preferred source of data is dependent on the complexity of the question being investigated. Analysts must think carefully about the economic question at hand and the most appropriate sources of data for that question. No matter what the origin, analysts must make the presentation of the data transparent and explain the rationale for the source of data used in the study.

THE USE OF META-ANALYSIS:

It is often the case that studies reporting efficacy or effectiveness data are either insufficient or are conflicting, yet there is still the need for information to support valid retrospective model development. Meta-analysis a process of combining study results in such a way as to be able to draw conclusions about therapeutic effectiveness ²⁴. As such, it is a tool for increasing the precision of estimated differences between the proposed drug and appropriate comparators which can then be used in a pharmacoeconomic model. It can also highlight advantages and disadvantages of the proposed drug and its comparators which are too small to be detected accurately in individual trials.

While there are currently no standardized formats, the reporting of meta analyses is an area which is evolving. Similar to the efforts towards a standardized reporting structure for RCTs proposed in the CONSORT (Consolidated Standards of Reporting Trials) statement ^25-26, efforts are currently underway for a similar approach to reporting meta analyses.

1. Health-Related Quality of Life as an Outcome:

Quality of life is a broad concept that includes many aspects of living in addition to health, for example; wealth, freedom, political system, and cleanliness of the environment all contribute to the overall QOL²⁷. Health-related quality of life refers to those aspects of QOL that are related to health. The overall goal of the health system is to improve both survival (life expectancy) and HRQOL. Accordingly, many tools have been developed to measure HRQOL. Any drug product that demonstrates improved effectiveness over its comparator(s) and impacts on a patient’s HRQOL should probably be evaluated for this outcome using these tools. The methods are partitioned into three major sets: specific instruments, generic profiles, and utility (preference-based) measures^28-30.

No single measure of HRQOL has yet been accepted as the gold standard. If HRQOL is being measured in a prospective study, it is normally advisable to include one reasonably precise, reliable and valid scale from each of the following three types: generic, specific, and a preference measure. The choice of instrument(s) is based on many factors, including: the content of the tool(s) being considered, the basis of scoring, and the question and/or disease state being investigated.

Specific Measures:

Specific instruments include those that are targeted at specific diseases, such as the Functional Living Index - Cancer³¹ or the Western Ontario-McMaster Osteoarthritis Index ³²; specific populations, such as the Care and Resource Evaluation Tool for the Elderly ³³; and specific functions, such as visual function measured by the Activities of Daily Vision Scale^34-35. A cancer index may not detect a change brought about by an arthritis intervention. The advantage of specific instruments is that they would be expected to have higher responsiveness to change ³⁶.

Generic Measures:

Generic health profiles are applicable to a wide range of patients and diseases and, thus, are more generalizable but probably less responsive than specific instruments Three well known instruments in this category are the Short Form 36 ³⁷, the Sickness Impact

Profile ³⁸, and the Nottingham Health Profile ³⁹.

Preference-based Measures:

The advantage of preference-based measures is that they are the only approach that provides a score reflective of HRQOL that is suitable for use in CEA and CUA.

Modeling HRQOL:

The discussion has focused on HRQOL measurement in prospective evaluations. In retrospective modeling studies, the analyst does not have the luxury of specifying the HRQOL instruments that will be used to gather the data. Typically, the analyst must work with results from clinical trials that did not incorporate such instruments. In this case, the analyst can undertake a CEA using the primary clinical effectiveness measure from the trials. If the analyst wishes also to undertake a CUA, the effectiveness outcomes from the trial must be somehow mapped onto utility scores.

There are fundamentally three means by which effectiveness can be mapped onto utility scores. One method is to develop written scenarios that describe the relevant health states from the trial, and to measure the utility of these states on a sample of the general public using an technique like the standard gamble (SG). An alternative method is to map the health states from the trial onto a multi attribute system like the health utilities index (HUI). A third possibility would be to find patients currently in the health states relevant to the trial and to measure their utility for these states.

2. Outcomes for Cost-Utility Analysis:

In CUA the quantity of life improvement (survival) and the HRQOL improvement (morbidity) are combined into a single metric (e.g. quality-adjusted life years gained).

QALYs:

The current generally accepted method of combining quantity and quality is through the use of QALYs^7,16,40-42. A QALY is calculated by multiplying the number of life years added via a program by a standardized weight (between 0.0 and 1.0) that reflects the health-related quality of life during that time (where 0.0 is the weight given to immediate death and 1.0 is the weight given to perfect health for a defined period of time).The QALY approach is useful in policy analysis and program decision-making, in part, because it is completely general. It can be applied to any population, any disease, any intervention, and can be used to compare across quite diverse programs. However, for the comparisons to be valid, the QALY studies must use the same methodology; for example, the same QALY weights, the same perspective, the same discount rate, etc.

The QALY approach contains a number of assumptions and limitations:

1) It assumes that all QALYs are equal. For example, it assumes that it is equally desirable to provide a one QALY gain to a teenager or to a senior citizen, to a woman or a man, to a laborer or to a professional, and so on.

2) It also assumes that it is equally desirable to provide a small gain to many people or a large gain to a few, as long as the QALY totals are the same. For example, a gain of 0.1 QALY to each of 1000 people would be considered equal to a gain of 25 QALYs each to four individuals.

3) As usually practiced, the QALY approach assumes that the relative weights for health states are independent of the duration of the health states. However, it is possible to circumvent this assumption by measuring the weights specifically for the durations that are relevant ⁷

⁴⁾The QALY approach also assumes that the preferences that individuals have for paths of changing health states can be reasonably estimated by adding up the time-weighted preferences that the individual has for the components of that path ⁴³

Despite these assumptions and limitations, the QALY approach remains the most common approach for combining quantity and quality of life, and using cost-utility analysis. The frequency of use of this approach is probably the result of its clarity, simplicity, ease of application, face validity, and, when the weights are based on von Neumann-Morgenstern utilities, its theoretical foundation.

LIMITS OF PHARMACOECONOMIC EVALUATION:

Many problems limit our use of health economics in practice ⁴⁴

The whole process may be open to bias, in the choice of comparator drug, the assumptions made, or in the selective reporting of results. This suspicion arises because most studies are conducted or funded by pharmaceutical companies who obviously are interested in the results, and there is a publication bias towards those studies favorable to sponsoring companies ⁴⁵

Health economics is therefore sometimes misused as a marketing ploy. The same problems may however arise in studies funded by health care payers. To a specialist, this is not such a problem since the almost inevitable biases are usually clear. But since economic evaluation is less well understood by doctors and others, bias needs to be minimized.

Three problems are common:

(i) A short term outlook which limits the application of economic evaluations showing long term savings for the health service in return for increased spending now.

(ii) Many budgets operate in isolation, and it is not easy to move money between them: for instance, prescribing in primary care is often funded separately from hospital services, so any increased spending on drug therapy in primary care cannot be simply funded from a future reduction in hospital admissions.

(iii) A new intervention may simply not be affordable no matter how cost effective it might be.

Finally, health economics and pharmacoeconomics is a young science and is slowly developing and testing its methodologies.

REFERENCES:

1. Feeny D, Guyatt G, Tugwell P. Health care technology: effectiveness, efficiency and public policy. Montreal, QC: The Institute for Research on Public Policy; 1986.

2. Detsky AS, Naglie G. A clinician's guide to cost-effectiveness analysis. Annals of Internal Medicine 1990; 113(2): 147-154.

3. Freund DA, Dittus RS. Principles of pharmacoeconomic analysis of drug therapy. PharmacoEconomics 1992; 1(1): 20-32.

4. Laupacis A, Feeny D, Detsky A, Tugwell PX. How attractive does a new technology have to be to warrant adoption and utilization? Tentative guidelines for using clinical and economic evaluation. Canadian Medical Association Journal 1992; 146(4): 473-481.

5. Eisenberg JM, Schulman KA, Glick H, Koffer H. Pharmacoeconomics: economic evaluation of pharmaceuticals. In: Strom BL, editor. Pharmacoepidemiology. 2nd ed. Chichester, UK: John Wiley & Sons; 1994. p. 469-493.

6. Bootman JL, Townsend RJ, McGhan WF, editors. Principles of pharmacoeconomics. 2nd ed. Cincinnati, OH: Harvey Whitney Books Company; 1996.

7. Drummond MF, O’Brien BJ, Stoddart GL, Torrance GW. Methods for the economic evaluation of health care programmes. 2nd ed. Oxford: Oxford University Press; 1997.

8. Jacobs P, Bachynsky J, Baladi JF. A comparative review of pharmacoeconomic guidelines. PharmacoEconomics 1995; 8(3): 182-189.

9. Drummond MF, Jefferson TO. Guidelines for authors and peer reviewers of economic submissions to the BMJ. BMJ 1996; 313(7052): 275-283.

10. Ikeda S, Ikegami N, Oliver AJ, Ikeda M. A case for the adoption of pharmacoeconomic guidelines in Japan. PharmacoEconomics 1996;10(6):546-551.

11. Diana Brixner. Practitioner update: Outcomes Research, Pharmacoeconomics and the Pharmaceutical Industry JMCP Jan/Feb 1996 Val 2, No.1.

12. Drummond MF, Davies L. Economic analysis alongside clinical trials-revisiting the methodological issues. Int J Technol Assess Hea1th Care 1991; 7: 561-73.

13. Stergachis A. Overview of cost-consequence modeling in outcomes research. Pharmacotherapy 1995; 15: 405-425.

14. Baltussen R, Leidl R, Ament A. The impact of age on cost-effectiveness ratios and its control in decision-making. Health Economics 1996; 5(3): 227-239.

15. Oxman AD. Preparing and maintaining systematic reviews. In: Sackett D, Oxman A, editors. The Cochrane Collaboration handbook. Oxford, UK: The Cochrane Collaboration; 1996.

16. Weinstein MC. Principles of cost-effective resource allocation in health care organizations. International Journal of Technology Assessment in Health Care 1990; 6 (1): 93-103.

17. Seigel JE, Weinstein MC, Torrance GW. Reporting cost-effectiveness studies and results. In: Gold MR, Seigel JE, Russell LB and Weinstein MC, editors. Cost-effectiveness in health and medicine. New York: Oxford University Press; 1996. P. 276-303.

18. O’Brien BJ. A tale of two (or more) cities: geographic transferability of pharmacoeconomic data. American Journal of Managed Care. In press 1997.

19. Tugwell P, Bennett KJ, Sackett DL, Haynes RB. The measurement iterative loop: a framework for the critical appraisal of need, benefits and costs of health interventions. Journal of Chronic Diseases 1985; 38(4): 339-351.

20. Bailey KR. Generalizing the results of randomized clinical trials. Controlled Clinical Trials 1994; 15(1): 15-23.

21. Bloomfield Rubins H. From clinical trials to clinical practice: generalizing from participant to patient. Controlled Clinical Trials 1994;15(1):7-10.

22. Davis CE. Generalizing from clinical trials. Controlled Clinical Trials 1994; 15: 11-14.

23. Johannesson M, Jonsson B, Kjekshus J, Olsson AG, Pedersen TR, Wedel H. Cost effectiveness of simvastatin treatment to lower cholesterol levels in patients with coronary heart disease. New England Journal of Medicine 1997; 336(5): 332-336.

24. L’Abbé KA, Detsky AS, O’Rourke K. Meta-analysis in clinical research. Annals of Internal Medicine 1987; 107(2): 224-233.

25. Begg C, Cho M, Eastwood S, Horton R, Moher D, Olkin I, et al. Improving the quality of reporting of randomized controlled trials: the CONSORT statement. JAMA 1996; 276(8): 637-639.

26. Huston P, Hoey J. CMAJ endorses the CONSORT statement. Canadian Medical Association Journal 1996; 155(9): 1277-1279.

27. Patrick DL, Erickson P. Health status and health policy: quality of life in health care evaluation and resource allocation. New York, NY: Oxford University Press; 1993.

28. Guyatt GH, Veldhuyzen Van Zanten SJ, Feeny DH, Patrick DL. Measuring quality of life in clinical trials: a taxonomy and review. Canadian Medical Association Journal 1989; 140(12): 1441-1448.

29. Guyatt GH, Feeny DH, Patrick DL. Measuring health-related quality of life. Annals of Internal Medicine 1993; 118(8): 622-629.

30. Feeny DH, Torrance GW, Labelle R. Integrating economic evaluations and quality of life assessments. In: Spilker B, editor. Quality of life and pharmacoeconomics in clinical trials, 2^nd ed. Philadelphia: Lippincott-Raven; 1996. p. 85-95.

31. Clinch JJ. The functional living index - cancer: Ten years later. In: Spilker B, editor.Q uality of life and pharmacoeconomics in clinical trials. 2nd ed. Philadelphia, PA: Lippincott-Raven; 1996. p.215-225.

32. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. Journal of Rheumatology 1988; 15(12): 1833-1840.

33. Fretwell MD. Frail older patients: creating standards of care. In: Spilker B. editor. Quality of life and pharmacoeconomics in clinical trials. 2nd ed. Philadelphia, PA: Lippincott-Raven; 1996. p.809-817.

34. Mangione CM, Phillips RS, Seddon JM, Lawrence MG, Cook EF, Dailey R, et al. Development of the ‘Activities of Daily Vision Scale’. A measure of visual functional status. Medical Care 1992; 30(12): 1111-1126.

35. Mangione CM, Lee PP, Hayse RD. Measurement of visual functioning and health-related quality of life in eye disease and cataract surgery. In: Spilker B, editor. Quality of life and pharmacoeconomics in clinical trials. 2nd ed. Philadelphia, PA: Lippincott-Raven; 1996. p. 1045-1051.

36. Guyatt GH, Jaeschke R, Feeny DH, Patrick DL. Measurements in clinical trials: choosing the right approach. In: Spilker B, editor. Quality of life and pharmacoeconomics in clinical trials. 2nd ed. Philadelphia, PA: Lippincott-Raven; 1996. p. 41-48.

37. Ware JE. The SF-36 health survey. In: Spilker B, editor. Quality of life and pharmacoeconomics in clinical trials. 2nd ed. Philadelphia, PA: Lippincott-Raven; 1996. p. 337-345.

38. Damiano AM. The Sickness Impact Profile. In: Spilker B, editor. Quality of life and pharmacoeconomics in clinical trials. 2nd ed. Philadelphia: Lippincott-Raven; 1996. p. 347-354.

39. McEwen J, McKenna SP. Nottingham Health Profile. In: Spilker B, editor. Quality of life and pharmacoeconomics in clinical trials. 2nd ed. Philadelphia, PA: Lippincott-Raven; 1996. p. 281- 286.

40. Torrance GW, Feeny D. Utilities and quality-adjusted life years. International Journal of Technology Assessment in Health Care 1989; 5(4): 559-575.

41. Gold MR, Patrick DL, Torrance GW, Fryback DG, Hadorn DC, Kamlet MS, et al. Identifying and valuing outcomes. In: Gold MR, Russell LB, Seigel JE, Weinstein MC, editors. Cost-effectiveness in health and medicine. New York, NY: Oxford University Press; 1996. p. 82-134.

42. Torrance GW. Designing and conducting cost-utility analyses. In: Spilker B, editor.Q uality of life and pharmacoeconomics in clinical trials. 2nd ed. Philadelphia: Lippincott-Raven; 1996. p. 1105-1111.

43. Kuppermann M, Shiboski S, Feeny D, Elkin EP, Washington AE. Can preference scores for discrete states be used to derive preference scores for an entire path of events? An application to prenatal diagnosis. Medical Decision Making 1997; 17(1):42-55.

44. Grau S, Mateu-de Antonio J, Soto J, Marín-Casino M, Salas E. Pharm World Sci. 2005 Dec;27(6): 459-64.

45. Fairman KA, Motheral BR. Do decision-analytic models identify costeffective treatments? A retrospective look at Helicobacter pylori eradication. J Manag Care Pharm. 2003; 9(5): 430-40. Available at: www.amcp.org/data/jmcp/ Formulary%20Management-430-440.pdf.

Received on 24.07.2012

Modified on 06.08.2012

Accepted on 15.08.2012

Research J. Pharmacology and Pharmacodynamics. 4(5): September –October, 2012, 272-277