Health Policy 116 (2014) 137–146

Contents lists available at ScienceDirect

Health Policy journal homepage: www.elsevier.com/locate/healthpol

Review

Systematic review of factors affecting pharmaceutical expenditures Mohamed Awad Mousnad a,∗ , Asrul Akmal Shafie b , Mohamed Izham Ibrahim c a

Senior Specialist of Pharmacoeconomics & Pharmacoepidemiology, National Health Insurance Fund, Khartoum, Sudan Assoc. Professor, Programme Chair, Discipline of Social & Administrative Pharmacy, School of Pharmaceutical Sciences, Universiti Sains Malaysia, 11800 Penang, Malaysia c Professor of Social & Administrative Pharmacy, College of Pharmacy, Qatar University, Doha, Qatar b

a r t i c l e

a b s t r a c t

i n f o

Article history: Received 12 June 2012 Received in revised form 11 February 2014 Accepted 18 March 2014

Keywords: Drug financing Drug quantities changes Drug therapies changes New drugs Pharmaceutical expenditures Spending determinants

Objective: To systematically identify the main factors contributing to the increase in pharmaceutical expenditures. Methods: A systematic search of published studies was conducted utilising major widely used electronic databases using the search terms ‘factors,’ ‘financing,’ ‘pharmaceutical,’ and ‘expenditures.’ To be included, the studies needed to: (1) measure at least one of the following outcomes: total growth in pharmaceutical expenditures, price growth or quantity growth; (2) mention a clear method for analysing the impact of factors affecting the increases in drug expenditures; (3) be written in English. Nonprimary articles that were published only as an abstract, a review, a commentary or a letter were excluded. Main results: From a total of 2039 studies, only 25 were included in the full review. The main determinant categories that were identified in the review were factors related to price, utilisation, therapeutic choice, demand and health care system. Conclusions: The major cost drivers were found to be changes in drug quantities and therapies as well as new drugs. It is important for policymakers to understand pharmaceutical spending trends and the factors that influence them in order to formulate effective cost containment strategies and design optimum drug policy. © 2014 Published by Elsevier Ireland Ltd.

1. Background The global demand for health services has increased dramatically in recent years due to changes in population characteristics, the introduction of health care technology and changes in the structural factors of the health care system [1]. In developing countries, this is caused by several factors, including the appearance of new diseases and the

∗ Corresponding author at: National Health Insurance Fund, P.O. Box 13267, Khartoum, Sudan. Tel.: +249 183 747620; fax: +249 183 747623. E-mail addresses: m [email protected], [email protected] (M.A. Mousnad). http://dx.doi.org/10.1016/j.healthpol.2014.03.010 0168-8510/© 2014 Published by Elsevier Ireland Ltd.

differences in the provision of health care to populations in different settings, including the provision of essential drugs and vaccines [2]. Pharmaceuticals, as the most important health component, contributed to more than half (55%) of all health-related trade in 2007 [3]. Faced with escalating health care costs in increasingly difficult economic conditions, many countries are looking for effective containment strategies for pharmaceutical costs. Unfortunately, there are multiple factors affecting pharmaceuticals. Pharmaceutical products involve a long chain of process from the discovery of new products, the promulgation of governmental regulations addressing licensing and safety issues, the marketing process and the behaviour of providers/consumers. A variety of fac-

138

M.A. Mousnad et al. / Health Policy 116 (2014) 137–146

tors emerge from this chain that may affect pharmaceutical expenditures. This multiplicity of factors is the origin of the difficulties facing policymakers as they seek to design effective pharmaceutical cost control interventions. In Europe, various policies were implemented not only for players in the health care market such as the pharmaceutical industry, wholesalers and retailers, consumers and prescribers [4], but also at the public and population level; demographic and disease incidence factors, which are outside of the pharmaceutical chain, are essential elements to consider in designing policies and allocating budgets [5]. A variety of strategies have been employed in different countries to rein in the increase in pharmaceutical expenditures, including educational, managerial, administrative and financial strategies [6]. The objective of this study is to identify the main factors that are contributing to the increase in pharmaceutical expenditures. Many studies have identified several factors that are responsible for the increase in pharmaceutical expenditures and have assessed several cost containment strategies [7]. These studies have explained the variation in pharmaceutical spending trends in different countries. Various factors have been explored, including the shift from older, cheaper medicines towards newer, more expensive medicines, increases in the utilisation of medicines, the introduction of new medicines, and increases in the prices of existing drugs [8]. However, due to differences in the healthcare systems in each setting, there are distinct variations in the classification of these factors. Furthermore, all of these studies were based in developed countries. This study attempted to systematically review the published literature regarding the factors that affect pharmaceutical expenditures. The findings may help health policymakers to gain a better understanding of the expenditures and trends in pharmaceutical services and may provide information that will assist them in setting priorities and in designing policies. This review presents findings, policy recommendations and suggestions for future research. 2. Objectives The objective of this review is to identify the main factors that are contributing to the increase in pharmaceutical expenditures. 3. Methods

technology. The studies also needed to have the following characteristics: (i) The study must measure at least one of the following outcomes: total growth in pharmaceutical expenditures, price growth or quantity growth. (ii) A clear method to analyse the impact of factors affecting the increases in drug expenditures must be mentioned. (iii) The study must be written in English. (iv) Non-primary research articles that were published only as an abstract, review, commentary or letter were excluded from the review. 3.2. Search methods for the identification of reviews Major widely used electronic databases were selected for the search, including Pub Med (from 1992 to 20 June 2010), Science Direct (from 1993 to 21 June 2010), ProQuest (from 1994 to 22 June 2010), SpringerLink (from 1994 to 20 June 2010), EBSCOhost (from 1995 to 13 December 2010 (last update)) and Google Scholar (from 1991 to 29 June 2010). The differences in the time frames were due to the availability of access to electronic databases from the Universiti Sains Malaysia library during the time in which the searches were performed. A search strategy was designed using the search terms ‘factors,’ ‘financing,’ ‘pharmaceutical’ and ‘expenditures.’ For PubMed, the medical subject headings (MeSH) ‘health expenditures,’ ‘economics,’ and ‘pharmaceutical’ were used (refer to the search strategy in Appendix 1). 3.3. Data extraction and management Two authors (MAM, AAS) independently assessed the titles and abstracts of each potential article for eligibility. Differences in decisions about inclusion and exclusion were resolved through consensus. A pre-developed data abstraction form was used to extract the following information from the included studies: the study title, author(s), country, year, objective(s), the statistical methods used to assess the factors affecting pharmaceutical expenditures, the results, and the authors’ conclusions. 3.4. Assessment of the methodological quality of the included reviews

This review was performed using an adaptation of the Cochrane guidelines for systematic reviews of health promotion and public health [9].

The quality assessment tool for quantitative studies [10] suggested by Cochrane Collaboration was adopted in this review for independent assessments by MAM and AAS of the quality of the included studies.

3.1. Criteria used to consider reviews for inclusion

3.5. Data synthesis

Studies were included if their main objective was to explore the factors that affect pharmaceutical expenditures. The factors affecting pharmaceutical expenditures are defined in this review as elements that influence the economic outcomes of a health care system driven by medical demand, the trend of care, or the effects of new

Narrative synthesis was used in this review. Metaanalysis was deemed to be unsuitable because of the heterogeneity of the studies in terms of methods, participants, settings and outcomes. The synthesis steps began with the organisation of the extracted data by the author [11]. A narrative description of the most common factors

M.A. Mousnad et al. / Health Policy 116 (2014) 137–146

139

Original search identified and screened N=2039 PubMed n=106 Science Direct n=605 Pro Quest n=104

SpringerLink n=126 EBSCOhost n=777 Google Scholar n=321 Excluded because: unrelated title and abstract N=1956 SpringerLink n=122 PubMed n=64 Science Direct n=593 EBSCOhost n=767 Pro Quest n=98 Google Scholar n=312

Search for eligibility N=83 PubMed n=42 Science Direct n=12 Pro Quest n=6

SpringerLink n=4 EBSCOhost n=10 Google Scholar n=9

Excluded from full review N=58 n=30 Only abstract PubMed n=30 SpringerLink n=0 EBSCOhost n=0 Science Direct n=0 Pro Quest n=0 Google Scholar n=0 Only abstract in English (foreign language) n=2 SpringerLink n=0 PubMed n=1 Science Direct n=0 EBSCOhost n=1 Pro Quest n=0 Google Scholar n=0

Publication meeting inclusion criteria N=25 PubMed n=1 Science Direct n=6 Pro Quest n=0

SpringerLink n=2 EBSCOhost n=8 Google Scholar n=8

Not designed to identify factors affecting pharmaceutical expenditure n=23 PubMed n=10 SpringerLink n=2 Science Direct n=6 EBSCOhost n=1 Pro Quest n=4 Google Scholar n=0 Duplicate publication n=3 PubMed n=0 SpringerLink n=0 EBSCOhost n=0 Science Direct n=0 Pro Quest n=2 Google Scholar n=1

Fig. 1. Process of determining studies’ eligibility.

identified from the review were reported and discussed in the results and discussion sections [11,12]. 4. Results 4.1. Description of the included reviews Fig. 1 shows the studies that were included in the systematic review and the reasons for the exclusion of the others. Out of 2039 studies, only 25 studies met the inclusion criteria and were included in the review. All of the studies included were from developed countries: USA (n = 10), Sweden (n = 2), Spain (n = 1), Canada (n = 7), France (n = 1), Denmark (n = 1), Greece (n = 1), Taiwan (n = 1) and Belgium (n = 1). 4.2. Characteristics and methodological aspects of the included reviews Table 1 shows the different study designs that were used in this review; none of the studies was a randomised

controlled trial (RCT) or clinical controlled trial. They included retrospective observational studies, repeated cross-sectional frameworks and time series data. Different methods and models were used. There was methodological variation in the studies’ exploration of the factors affecting pharmaceutical expenditures. The methods include: analysis of prescription claims [13] and the structural effect approach [14]; decomposition of pharmaceutical spending growth [15]; dynamics of expenditure through time [16]; decomposition of change in real drug spending using a multiplicative growth equation [17]; decomposition of real drug spending into price, quantity and residual [18,19]; prescription drug expenditures [20–22]; decomposing the growth of pharmaceutical expenditures into three components: treatment expansion; treatment substitution; and price effect [23]; a pharmacoepidemiological model based on incidence, treatment discontinuation and drug user mortality [24];decomposition of the increase in real drug expenditures into a price effect, a volume effect, and a product-mix effect [25]; multiplicative factor equations

140

M.A. Mousnad et al. / Health Policy 116 (2014) 137–146

Table 1 General characteristics and methods of the included studies. No.

Author(s)

Year

Country

Statistical methods to assess the factors affecting pharmaceutical expenditure

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Anderson and Kerluke [13] Cavalié [14] Chernew et al. [15] Darbà [16] Dubois et al. [17] Gerdtham and Lundin [18] Gerdtham et al. [19] Hanley and Morgan [20] Hoffman et al. [21] Hoffman et al. [22] Hsieh and Sloan [23] Kildemoes et al. [24] Lambrelli and Donnell [25] Morgan [26] Morgan [27] Morgan [28] Morgan et al. [29] Morgan [30] Mueller et al. [31] Mullins et al. [32] Sherman [33] Steinberg et al. [34] Suh et al. [35] Van Tielen et al. [36] Vandegrift and Datta [37]

1996 2003 2001 2003 2000 2004 1993 2009 2010 2008 2008 2010 2010 2004 2002 2005 2004 2006 1997 2001 1999 2000 1999 1998 2006

Canada France USA Spain USA Sweden Sweden Canada USA USA Taiwan Denmark Greece Canada Canada Canada Canada Canada USA USA USA USA USA Belgium USA

Analysis of prescription claims Structural effect approach Decomposition of pharmaceutical spending growth Dynamics of expenditure along time, decomposition of pharmaceutical expenditures A multiplicative growth equation, disaggregating spending trends Decomposition of the changes in real drug spending Decomposition of the changes in real drug spending Prescription drug expenditures Drug expenditure trends Drug expenditure trends Decomposition of the growth in pharmaceutical expenditures Model based on incidence, treatment discontinuation and drug user mortality Decomposition of the changes in real drug spending Disaggregating spending trends using Fisher indexes A multiplicative of expenditure inflation using Fisher’s ideal index Price and quantity indices using Fisher’s ideal formula Decomposition of expenditure using no stochastic index-theoretical methods Equation describing changes in age/sex-specific expenditures Drug spending trends Drug spending trends Evaluates price and utilisation trends among different types of drugs Analyses spending trends Analyses spending trends Estimates drug sales and the demographic impact of national reimbursements Fixed-effects panel data model to analyse prescription drug expenditures

using index numbers [26–29]; and equations describing changes in age/sex-specific expenditures [30]. Other studies included trend analysis [31–36] and fixed effects using regression analysis and the demographic impact method [37]. There were variations in the types of population, health care plans, health insurance, outpatient, hospital, health care systems and national income levels that were included in the studies. There were also variations in the type of outcomes measured in the studies, including the total growth of pharmaceutical expenditures, price change trends and the drug quantities utilised [38] (refer to Table 1). 4.3. Factors affecting pharmaceutical expenditures A summary of the statistically significant factors affecting pharmaceutical expenditures established in the review is presented in Table 2. The most common factors identified in the included reviews are price changes for existing drugs, changes in drug quantities and therapies and new drugs. Price changes for existing drugs have been identified by 18 out of the 25 studies included in this review. Some of these studies found that it has a positive effect on expenditures, i.e., increases in the prices of existing drugs increases the total spending [15–17,21,22,28–30,32,33,35]; while others found that it has a negative effect, i.e., a change in the price of existing drugs results in lower expenditures [14,18,19,23,25–27]. The results of this review showed that changes in drug quantities and therapies have had a major impact on drug expenditure increases. This impact has been identified by 18 out of the 25 studies included in this review. Twelve studies identified the effect of new drugs. The results showed that it is one of the major drivers of drug costs.

5. Discussion The pharmaceutical allocation has contributed substantially to the moderate rise in health care costs globally [39,40]. Therefore, it is very important that decision makers understand the elements behind the economic burden created by pharmaceuticals and its main driving factors. This review attempted to identify the factors affecting pharmaceutical expenditures that were investigated by the retrieved original studies in different population settings and countries. Several factors have been identified that significantly affect pharmaceutical expenditures. Although many of the factors are overlapping, this review identified three major factors: price changes for existing drugs, changes in drug quantities and therapies, and new drugs. Several study designs were used, including retrospective observational studies, repeated cross-sectional frameworks and time series data. Different methods and models were used. The most commonly used methods were decomposition analysis (10 studies) and expenditure trends analysis (8 studies). The different types of outcomes measured in the studies were assessed, including the total growth of pharmaceutical expenditures, price change trends and the drug quantities utilised. Changes in drug prices refer to changes in the prices charged for existing drugs. In the reviewed studies, changes in drug prices were measured in several ways. First, some of the studies analysed the overall growth in prescription drug expenditures by determining the change in drug unit cost (prescription) [34]. Secondly, changes in drug prices can be determined by inflation, i.e., so-called price inflation. In economics, inflation refers to the increase in the prices of goods and services over a period of time or it may refers to an increase in a particular set of prices or even in a

M.A. Mousnad et al. / Health Policy 116 (2014) 137–146

141

Table 2 Statistically significant factors affecting pharmaceutical expenditures in the review. No.

Author(s)

Country

Factors affecting the increase in pharmaceutical expenditures

1

Anderson and Kerluke [13]

Canada

2 3 4

Cavalié [14] Chernew et al. [15] Darbà [16]

France USA Spain

5

Dubois et al. [17]

USA

6

Gerdtham et al. [18]

Sweden

7

Gerdtham and Lundin [19]

Sweden

8 9

Hanley and Morgan [20] Hoffman et al. [21]

Canada USA

10 11

Hoffman et al. [22] Hsieh and Sloan [23]

USA Taiwan

12

Kildemoes et al. [24]

Denmark

13

Lambrelli and Donnell [25]

Greece

14

Morgan [26]

Canada

15 16

Morgan et al. [27] Morgan [28]

Canada Canada

17

Morgan [29]

Canada

18

Morgan [30]

Canada

19 20 21

Mueller et al. [31] Mullins et al. [32] Sherman [33]

USA USA USA

22 23

Steinberg et al. [34] Suh et al. [35]

USA USA

24 25

Van Tielen et al. [36] Vandegrift and Datta [37]

Belgium USA

Number of medicines received by patients (drug exposures); ageing; therapeutic drug categories Prices; quantities; new drugs Price changes; quantity changes (number of prescriptions and mix of prescriptions) Average price index; average population growth rate; average prescription per person growth rate, average residual. Average increase in volume existing drugs and new drugs; days per prescription; average increase in price factors inflation, new therapies; change in therapeutic mix The real price of drugs; the quantity (number of DDDs) and the residual increased (due to new drugs) The relative retail price index of drugs; the quantity of drugs (measured in DDDs) and residual (due to new drugs) Ageing; gender: socio-economic status (in terms of income); health status (morbidity profiles) Trend of increases in expenditures for specialised medications; biological biosimilars (and potential for biosimilar legislation), drugs in development, diffusion of new drugs, generic drugs, increased availability and use of generic equivalents of the blockbuster. Health care reform may exert greater influence on drug expenditures, drug safety concerns, and comparative effectiveness research Price; utilisation and mix; new drugs Quantity (due to the introduction of new drugs); the mix effect (due to treatment substitution) and pure price effects Increasing population-wide treatment prevalence; population ageing trends in incidence mortality and health care system (guidelines) The relative price of pharmaceuticals; the product-mix residual (this is mainly attributable to a switch to more innovative, but more expensive pharmaceuticals; the volume of drugs prescribed) Utilisation factor; drug exposure; therapeutic choices (narrow mix classes, broad mix), price-readied drivers (generics) and prices Price index; therapeutic categories; drug mix and generic drugs Population ageing; overall exposure; poly-therapy; drug class mix; drug mix; generic use, price changes Volume effects (prescription volume, prescription size), therapeutic choices (therapeutic mix drug mix price effects (price changes, generic use) Changes in the age-specific mix of drugs prescribed per prescription; the age-adjusted number of prescriptions filled per patient per therapeutic category increased population-wide per capita expenditures; age/sex-specific average numbers of categories used per ‘exposed’ patient caused per capita expenditures; drug price inflation (for existing products) Disease category factor; ageing; therapeutic classes Price gains in existing drugs, the use of existing drugs and product shift; use of newer drugs Direct-to-consumer advertising of new branded drugs; generic drugs; average prices per prescription; introduction and widespread use of new drugs Age; sex; disease category General inflation; prescription prices; population growth; growth in prescription drug use (the number of prescriptions) and growth in intensity per prescription Population ageing Obesity, population ageing; new drug approval; real incomes

single price. Put simply, inflation refers only to a drop in the purchasing power of money. Many types of inflation occur, e.g., oil price inflation, health care inflation and wage inflation [41]. Inflation is measured by calculating an index of changes in actual product prices, similar to the consumer price index (CPI) [17]. Price indexes measure the aggregate price change over a certain period by comparing the cost of purchasing a basket of goods at different times. Many different methods are used for the calculation of price indexes. Some of these methods include the Laspeyres price index, the Paasche price index and the Fisher ideal price index [42,43]. The widely used Laspeyres index is a constant quality type index; it is called a base-weighted index and measures price changes from a base year. The Paasche index (the output deflator) is called an end-year weighted price index and uses the end-year quantities as weights. The Fisher index (price deflator) is the geometric average of a Laspeyres and

a Paasche index for the same period, using an average of the observed quantities as weights. The geometric average is calculated by multiplying the Laspeyres index by the Paasche index and then taking the square root of the result [44–46]. Analysing pharmaceutical expenditures using price and quantity indexes exposes a widely known problem: they are not able to explain the increase in real drug expenditures [47]. This problem could be overcome by estimating a correct price index. This is performed by a relevant quantity index that is calculated using the number of prescriptions or Defined Daily Doses (DDDs) as the defined type of quantity used in the indexes, not the physical quantity [48]. Some studies in the review considered both inflation and price change in their analyses as price factors [17,19,34], while others used the change in a price index to measure the change in prices [16,19,23,26,27]. One study measured price changes by calculating the average of the changes

142

M.A. Mousnad et al. / Health Policy 116 (2014) 137–146

in the prices of all the items identified by ingredient(s), dosage, form and brand [28]. Price changes for existing drugs can be evaluated between different points of time. Traditional methods examine trend differences and calculate the differences in expenditures and the contribution of each factor that may affect the increase; these methods have limitations because the inflation rate is not taken into consideration, and thus, they produce inaccurate results. In recent studies, decomposition methods and econometric models have been used most frequently [17–19,23,28]; they calculate the real expenditures and real price changes (adjusted inflation). These methods can be applied in different settings and the results can be generalised. Since the real spending on pharmaceuticals is affected by economic inflation, it seems important to adjust the increases in expenditures to the general level of inflation in the economy. The most widely used measure is the pharmaceutical price index. Changes in drug quantities and therapies can be assessed by measuring the changes in the number of prescriptions or quantities of drugs used (changes in the growth of medication volume and size), measuring the real increases in the use of pharmaceutical therapies (intensity of treatment) and measuring the incidences in which multiple drug classes are used (poly-therapy). Intensity per prescription and exposure to treatment are terms that refer to the real increases in the use of pharmaceutical therapies and the real expenditure per prescription [34]. These factors are considered to be the most important factors for both the demand and the supply chain since prescribers are under the influence of patients or the promotions of pharmaceutical companies. Also, the introduction of new innovative medicines to the market may increase the utilisation of medicine. These facts are supported by several studies [49,50]. Dubois et al. [17] found that the growth in medication volume plays an important role in the increase in drug spending as opposed to the growth in the average price. Selection of the type of medicine, i.e., the active components and the specific medicine form, and the dose regimen are influenced by the introduction of newer, more expensive and effective drugs, the use of generic drugs, the therapeutic mix, the drug mix, biological therapies (biosimilars) and residuals. ‘Therapeutic mix’ refers to the type of drug choices within broad categories of treatment that result from changes in market shares [28]; another term for this is ‘broad mix’ [27].‘Drug mix’ refers to the more specific selection of drug products within therapeutic classes [28]; another term for this is ‘narrow mix’ [27]. The studies included in the review investigated the significant effect of the therapeutic mix and drug mix on pharmaceutical expenditures. Similar facts were found in a review study conducted in the USA, which confirms that the cost impact of more or less expensive strengths (strength mix) and more or less expensive drugs (therapeutic mix) makes a significant contribution to the average whole price (AWP) of pharmaceuticals [51]. Hoffman et al. [22] identifies additional important determinants of drug expenditures in the USA, including biological therapies (biosimilars and the potential for

biosimilars legislation) and specialised medications. These factors have emerged as the primary drivers of overall drug expenditure patterns, particularly in the outpatient setting. The introduction of new innovative drugs was investigated as one of the important causes of pharmaceutical expenditure growth in Europe [50]. Although the diffusion of these drugs has become widely accepted in the United States, it takes time for them to be accepted by prescribers due to the long approval process, the safety issues affecting new medications and the difficulty of accepting the ‘high prescribing costs’ of new drugs [22]. An increase in the number of new and expensive medications that are being approved for chronic conditions occurs every year [52]. Two studies conducted in the USA reported similar significant effects of newer and more effective drugs on the increase in pharmaceutical spending [51,53]. Many studies in the review have used decomposition methods to describe the factors responsible for the increases in drug expenditures. When these methods are utilised, the change in real spending is decomposed into three components: price, quantity and residual. Residual refers to the remaining part of the change in real drug expenditures other than price and quantity [18,19]. The residual reflects many factors, including the introduction of new drugs, the shift from cheaper to more expensive drugs, and the introduction of new and cheaper generic medicines [16,18,19,23,25]. The residual factor can be measured using index numbers [18,19]. The evidence in this review and in existing literature suggests that the residual plays a major role in the increase in pharmaceutical expenditures in many countries due to the shift from cheaper drugs to new, innovative and expensive drugs. Food and Drug Administration (FDA) in the United Sates defined generic drug as ‘identical or bioequivalent to a brand name drug in dosage form, safety, strength, route of administration, quality, performance characteristics and intended use’ [54]. The studies in this review show that the introduction of generic medicines has resulted in a significant reduction in pharmaceutical expenditures in many countries [27–29]. In the global market, price variations within drug classes or between drugs classes are a common phenomenon. Prescribers play an important role in therapeutic choices. Studies suggest that doctors have a poor understanding of pharmaceutical costs. This was confirmed in a systematic review by Allan et al. [55] which evaluated physicians’ level of awareness of the costs of prescription drug products. The review concluded that doctors were usually not accurate when asked to estimate the costs of drugs; doctors tended to underestimate the costs of expensive drugs and overestimate the costs of cheaper alternatives. This review also identified other important determinants of drug expenditures, including the ageing of the population and health care system factors. Studies in the current review have examined the effect of population changes on pharmaceuticals; several studies have determined that the ageing of the population has very little effect on pharmaceutical expenditures [14,24,35,37]. Although the ageing of the population does not have a major impact on expenditures, and is beyond policy control (it does not

M.A. Mousnad et al. / Health Policy 116 (2014) 137–146

affect the demand or supply side), it is of importance for planning, designing policies and allocating budgets. It is important to note that the age group 45–64 was found to be more likely to utilise a prescription than the group aged 65–84; furthermore, females are more likely to utilise medicine than males. Important determinants of drug expenditures have been identified in the USA, including drug safety concerns, changes in the drug supply chain, health care reform, drug shortages and comparative effectiveness research [22]. Among those determinants, health care reform may exert a greater influence in the coming years and may increase the unpredictability of drug expenditures. Among the health policy reforms, Medicare part D plays a central role in pharmaceutical expenditures, as evidenced by several authors [56–58]. Among the all factors compared, the major cost drivers found in this review were changes in drug quantities and therapies and new drugs. The least important factor affecting drug expenditures was price changes for existing drugs. 6. Policy implications Since health care needs are perpetually increasing, it is very important for policymakers to understand pharmaceutical spending trends, the factors influencing the rising demand for limited resources and the main cost containment strategies. To deal with the continuous growth in pharmaceutical expenditures, all of the above factors must be considered in formulating pharmaceutical policy. Several strategies have been implemented in different countries to rein in the increases in pharmaceutical expenditures, including educational, managerial, administrative and financial strategies [6,59,60]. In policy practices, educational strategies consist of certain tools, namely, practice guidelines, continuing education programmes, drug committees and patient information packages. Direct-to-consumer advertising (DTC) can be controlled by managerial and administrative strategies such as the regulation of marketing and commercial information. Managerial strategies consist of positive and negative lists, reference pricing systems (or incentive pricing), reimbursement schemes, disease management strategies, practice guidelines and the regulation of marketing and commercial information. Administrative and financial strategies include fixed or directive budgets for prescribing, price regulation, price/volume agreements, copayment schemes, financial incentives to pharmacists to improve pharmacy services and the promotion of generic drugs; these are the most widely tested tools to control increases in pharmaceutical expenditures [4,8,61–66]. Changes in drug quantities and therapies and new drugs are the key drivers of drug expenditures that have been identified in this review. These factors are endogenous to the pharmaceutical market and are among the factors over which policymakers, practitioners, and patients have the power to act. By focusing on these expenditure determinants and conducting related strategy interventions, an effective pharmaceutical policy can be reinforced. Policymakers should be oriented towards using information regarding essential medicines and financial

143

incentives for better spending trends and therapeutic selection. Policies should be targeted by the therapeutic category and should aim to increase educational interventions for prescribers and patients. Policymakers should remain informed regarding the trends in medication approvals and generic medication releases, as well as the patterns in the diffusion of newly released drugs. This information is important for the forecasting of drug spending and financial management [22]. Considerably more focus should be placed on the education of physicians regarding costs and the access to cost information [55]. Physicians should choose the least costly medication when there are no substantial differences in the safety and effectiveness between the least and most expensive medications. Physicians should choose therapeutically equivalent but less expensive drugs in order for substantial savings to occur. Generic substitution and reference price systems, which are strong pharmaceutical expenditure controlling strategies frequently used in Europe, Canada (British Columbia), New Zealand and Australia, should be implemented wherever possible [59,67,68]. Strategies such as co-payment schemes and co-insurance are also suggested, as they are reported to significantly reduce utilisation and costs. Practice guidelines should be reviewed and rationalised. Drug use reviews should be performed routinely to reduce pharmaceutical utilisation and costs [4,8,61–66,69,70]. Policies such as restricted listings encourage the selection of higher cost drugs only in cases in which the evidence suggests that such decisions are cost-effective. Health Technology Assessment (HTA) has been successfully implemented as a policy tool in many developed countries [71]. Policymakers and researchers should use HTA and pharmacoeconomic evaluations to promote better decision-making and to generate cost savings. Policymakers and researchers can use Evidence-based Medicine (EBM) methods to ensure the efficacy and effectiveness of pharmaceutical products. As this review focuses on the cost-effectiveness of pharmaceuticals, methods such as Comparative Effectiveness Research (CER) should be useful to ensure that, to the greatest extent possible, medical decisions are consistent with evidence of effectiveness and benefit [72]. CER has been the subject of much discussion in the United States in recent decades as a tool for the generation and synthesis of evidence. It has been found that CER has the potential to slow the growth of health care expenditures and has a clear positive impact on the quality and value of health care services. This fact is supported by several authors [73,74]. The Congressional Budget Office (CBO) estimated that CER could reduce direct spending by the federal government (Medicare and Medicaid) by $0.1 billion between 2008 and 2012 and by $1.3 billion between 2008 and 2017 [21]. CER studies or programmes will provide patients, clinicians, payers, and policymakers with valid and relevant evidence that could be used in HTA. 7. Review limitations There are some limitations to this review. None of the included studies were conducted in developing countries.

144

M.A. Mousnad et al. / Health Policy 116 (2014) 137–146

Although the review has determined important significant findings, they may vary from one country to another due to differences in health care systems, cultures and attitudes. A small number of studies were found to meet the review inclusion criteria. Finding studies on public health is much more complicated than retrieving medical studies because the literature is being more widely distributed [75].

Appendix 1. Search strategy Search databases

Keywords/search terms

Pub Med

#1 Factors [All Fields] AND #2 (‘health expenditures’ [MeSH Terms] AND ‘expenditures’ [All Fields] OR ‘health expenditures’ [All Fields] OR ‘expenditures’ [All Fields]) AND #3 (‘economics’ [Subheading] AND ‘financing’ [All Fields]) AND # 4 (‘economics, pharmaceutical’ [MeSH Terms] AND ‘pharmaceutical’ [All Fields] OR ‘economics, pharmaceutical’ [All Fields] OR ‘pharmaceutical’ [All Fields] #5 #2 OR #3 OR #4 Date: all articles published between 1992 and 20 June 2010. Factors AND financing AND pharmaceutical AND expenditures. All sources (economics, econometrics and finance, medicine and dentistry, pharmacology, toxicology and pharmaceutical science) Date: articles published between 1994 and 21 June 2010. Search results for ‘factors AND financing AND pharmaceutical AND expenditures’ with the filter: medicine. Date: articles published between 1994 and 22 June 2010. Factors AND financing AND pharmaceutical AND expenditures. Databases selected: multiple databases. Date range: articles published between 1994 and 20 June 2010. Searching: Academic Source Premier, Boolean/Phrase: TX Factors AND financing AND pharmaceutical AND expenditures. Scholarly (peer reviewed) journals, publication type: periodical, document type: article. Date: articles published between 1995 and 13 December 2010 (last update). With all of the words: factors AND financing AND pharmaceutical AND expenditures. With the exact phrase: pharmaceutical AND expenditures. Search only articles in the following subject areas: Biology, Life Sciences, Environmental Science, Business, Administration, Finance, Economics, Medicine, Pharmacology and Veterinary Science. Date: articles published between 1991 and 29 June 2010.

8. Conclusions The results of this review illustrate the value of understanding pharmaceutical cost determinants in policy making and planning. Moreover, the factors identified could be used in the cost-effective selection and choices of medicine, in allocating budgets to meet increasing needs using evidence-based information regarding their demographic causes, and in the monitoring and regulation of pharmaceutical marketing and commercial activities. Previous reviews have quantified the factors affecting pharmaceutical expenditures primarily in the context of the US and Canada; none of them provided a systematic review of the literature. A draft discussion paper by Morgan [5] reviewed the historical context of drug spending and provided a summary of the literature describing the underlying causes of drug spending trends in Canada. A review based on survey response data reviewed the studies, but only in the USA in the 1990s [7]. This review is the first systematic review of the factors affecting pharmaceutical expenditures. It provides a comprehensive analysis of the previously published studies found in widely used major electronic databases. The research design was a systematic review in addition to a wide scope of assessment utilised in different health care levels in different countries. Moreover, it captured other new factors that were not identified in the previous reviews, such as important determinants of drug expenditures in the context of the USA; these include biological therapies (biosimilars and potential for biosimilars legislation) and specialised medications; health care reform; drug safety concerns and changes in the drug supply chain. In conclusion, the major cost drivers identified in this review were changes in drug quantities and therapies and new drugs. More studies should be conducted regarding various types or spending levels using a simple economic analysis of drug spending trends and measuring the causes and consequences. Future studies should be conducted in developing countries to fill in the gap in the existing literature. Researchers should always take the real value of pharmaceutical expenditures into consideration when they conduct these studies in order to promote the future development of effective health care policies.

Financial resources The authors would like to acknowledge that this review was part of a PhD thesis conducted at the School of Pharmaceutical Sciences, Universiti Sains Malaysia, Penang, Malaysia.

Science Direct

SpringerLink

ProQuest

EBSCOhost

Google Scholar

References [1] Thomas R. Health services demand and utilization. Health Services Planning. Springer US; 2003. p. 55–73. [2] Annan KA. We the peoples: the role of the United Nations in the 21st century. United Nations; 2000. [3] Smith RD, Correa C, Oh C. Trade, TRIPS, and pharmaceuticals. The Lancet 2009;373(9664):684–91. [4] Ess SM, Schneeweiss S, Szucs TD. European healthcare policies for controlling drug expenditure. Pharmacoeconomics 2003;21(2):89–103. [5] Morgan SG. The determinants of prescription drug expenditure and what to do about them. UBC Centre for Health Services and Policy Research; April 2008. Available from: http://www.chspr.ubc.ca/ [cited 2.06.10]. [6] de Joncheere K. The role of the World Health Organization on pharmaceuticals in Europe. Value in Health 2001;4(3):217–24. [7] Merlis M. Explaining the growth in prescription drug spending: a review of recent studies; 2000. Available from: http://aspe.hhs. gov/health/reports/drug-papers/merlis/merlis-final.htm [8] Noyce PR, Huttin C, Atella V, Brenner G, Haaijer-Ruskamp FM, Hedvall M-B, et al. The cost of prescription medicines to patients. Health Policy 2000;52(2):129–45. [9] Armstrong R, Waters E, Jackson N, Oliver S, Popay J, Shepherd J, et al. Guidelines for Systematic reviews of health

M.A. Mousnad et al. / Health Policy 116 (2014) 137–146

[10]

[11] [12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28] [29] [30] [31]

[32]

[33]

[34]

[35]

promotion and public health interventions; 2007. Available from: www.vichealth.vic.gov.au/cochrane [cited 06.06.10]. EPHPP. The Effective Public Health Practice Project. The quality assessment tool for quantitative studies; 2010. Available from: http://www.city.hamilton.on.ca/phcs/EPHPP/ [cited 06.06.10]. Petticrew M, Roberts H. MyiLibrary. Systematic reviews in the social sciences: a practical guide. Oxford: Blackwell Pub.; 2006. Centre for Reviews and Dissemination. Systematic Reviews. CRD’s guidance for undertaking reviews in health care. York; 2009. Available from: www.york.ac.uk/inst/crd/systematic reviews book.htm Anderson G, Kerluke K. Distribution of prescription drug exposures in the elderly: description and implications. Journal of Clinical Epidemiology 1996;49(8):929–35. Cavalié P. Is therapeutic innovation responsible for the increase in drug expenditure? The European Journal of Health Economics 2003;4(3):184–94. Chernew ME, Smith DG, Kirking DM, Fendrick AM. Decomposing pharmaceutical cost growth in different types of health plans. The American Journal of Managed Care 2001;7(7):667. Darbà J. Pharmaceutical expenditure in Spain: evolution and cost containment measures during 1998–2001. The European Journal of Health Economics 2003;4(3):151–7. Dubois RW, Chawla AJ, Neslusan CA, Smith MW, Wade S. Explaining drug spending trends: does perception match reality? Health Affairs 2000;19(2):231–9. Gerdtham UG, Johannesson M, Gunnarsson B, Marcusson M, Henriksson F. The effect of changes in treatment patterns on drug expenditure. Pharmacoeconomics 1998;13(1 Pt. 2):127–34. Gerdtham U-G, Lundin D. Why did drug spending increase during the 1990?: a decomposition based on Swedish data. Pharmacoeconomics 2004;22(1):29–42. Hanley GE, Morgan S. Chronic catastrophes: exploring the concentration and sustained nature of ambulatory prescription drug expenditures in the population of British Columbia, Canada. Social Science & Medicine 2009;68(5):919–24. Hoffman JM, Doloresco F, Vermeulen LC, Shah ND, Matusiak L, Hunkler RJ, et al. Projecting future drug expenditures—2010. American Journal of Health-System Pharmacy 2010;67:919–28. Hoffman JM, Shah ND, Vermeulen LC, Doloresco F, Grim P, Hunkler RJ, et al. Projecting future drug expenditures—2008. American Journal of Health-System Pharmacy 2008;65(3):234–53. Hsieh C-R, Sloan FA. Adoption of pharmaceutical innovation and the growth of drug expenditure in Taiwan: is it cost effective? Value in Health 2008;11(2):334–44. Kildemoes HW, Andersen M, Støvring H. The impact of ageing and changing utilization patterns on future cardiovascular drug expenditure: a pharmacoepidemiological projection approach. Pharmacoepidemiology and Drug Safety 2010;19(12):1276–86. Lambrelli D, Donnell O. The impotence of price controls: failed attempts to constrain pharmaceutical expenditures in Greece. Health Policy 2010;101(2):162–71. Morgan SG. Quantifying components of drug expenditure inflation: the British Columbia seniors’ drug benefit plan. Health Services Research 2002;37(5):1243–66. Morgan SG, Agnew JD, Barer ML. Seniors’ prescription drug cost inflation and cost containment: evidence from British Columbia. Health Policy 2004;68(3):299–307. Morgan S. Drug spending in Canada: recent trends and causes. Materials Characterization 2004;42(7):635–42. Morgan SG. Booming prescription drug expenditure: a populationbased analysis of age dynamics. Medical Care 2005;99:6–1008. Morgan SG. Prescription drug expenditures and population demographics. Health Services Research 2006;41(2):411–28. Mueller C, Schur C, O’Connell J. Prescription drug spending: the impact of age and chronic disease status. Australian Journal of Public Health 1997;87(10):1626–9. Mullins CD, Wang J, Palumbo FB, Stuart B. The impact of pipeline drugs on drug spending growth. Health Affairs 2001;20(5): 210–5. Sherman D. Factors affecting the growth of prescription drug expenditures. Washington, DC: The National Institute for Health Care Management Research and Educational Foundation; 1999. Available from: www.nihcm.org/mainframe.html [cited 15.04.09]. Steinberg EP, Gutierrez B, Momani A, Boscarino JA, Neuman P, Deverka P. Beyond survey data: a claims-based analysis of drug use and spending by the elderly. Health Affairs 2000;19(2):198–211. Suh D-C, Lacy CR, Barone JA, Moylan D, Kostis JB. Factors contributing to trends in prescription drug expenditures. Clinical Therapeutics 1999;21(7):1241–53.

145

[36] Van Tielen R, Peys F, Genaert J. The demographic impact on ambulatory pharmaceutical expenditure in Belgium. Health Policy 1998;45(1):1–14. [37] Vandegrift D, Datta A. Prescription drug expenditures in the united states: the effects of obesity, demographics, and new pharmaceutical products. Southern Economic Journal 2006;73(2):515–29. [38] Higgins J. Green Se Cochrane handbook for systematic reviews of interventions. The Cochrane Collaboration; 2009. Available from: www.cochrane-handbook.org [cited 06.04.10]. [39] Andersson F. Methodological aspects of international drug price comparisons. Pharmacoeconomics 1993;4(4):247. [40] Mossialos E, Oliver A. An overview of pharmaceutical policy in four countries: France, Germany, the Netherlands and the United Kingdom. The International Journal of Health Planning and Management 2005;20(4):291–306. [41] Humpage OF. Rising relative prices or inflation: why knowing the difference matters. Economic Commentary 2008;(June). [42] Rovira J, Tremosa R, Gilabert A, Torralba M. The role of prices in drug expenditure analysis. The European Journal of Health Economics 2001;2(4):142–9. [43] Diewert WE. Index numbers. In: Eatwell JMM, Newman P, editors. The new Palgrave dictionary of economics. New York: Macmillan Reference USA; 1987. [44] Thomas J, Schondelmeyer S. Price indexes for pharmaceuticals used by the elderly. Health Care Financing Review 1992;14(1):91. [45] Danzon PM, Kim JD. International price comparisons for pharmaceuticals: measurement and policy issues. Pharmacoeconomics 1998;14:115–28. [46] Jönsson B. Cost sharing for pharmaceuticals—the Swedish reimbursement system. Pharmacoeconomics 1996;10(Suppl. 2): 68–74. [47] Gerdtham U-G, Johannesson M, Jönsson B. Drug expenditure and new drug introductions: the Swedish experience. Pharmacoeconomics 1993;4(3):215–25. [48] Gerdtham UG, Johannesson M, Gunnarsson B, Marcusson M, Henriksson F. Price indices of drugs and the switching to new drugs: two empirical examples. Pharmacoeconomics 1998;13(1 Part 2): 71–80. [49] Burstall ML. The management of the cost and utilisation of pharmaceuticals in the United Kingdom. Health Policy 1997;41(Suppl.):S27–43. [50] Ljungkvist MO, Andersson D, Gunnarsson B. Cost and utilisation of pharmaceuticals in Sweden. Health Policy 1997;41:S55–69. [51] Teitelbaum F, Roe CM, Fendrick AM. Pharmacy management trends: a guide for new physicians. Seminars in Medical Practice 2000;3(1):22–31. [52] Berndt ER. Health Affairs 2001;20(2):100–14. [53] Vogel RJ. The reasons behind increasing pharmaceutical expenditures. A technical paper; 2002. p. 1–12. Available from: http:// boulderblast.com/publications/studies research/pdfs/CoA Pharmaceutical Expenditures.pd [29.06.10]. [54] FDA. Generic drugs: questions and answers. USA: Food and Drug Administration; 2010. Available from: http://www.fda.gov/ Drugs/ResourcesForYou/Consumers/QuestionsAnswers/ ucm100100.htm [cited 12.11.11]. [55] Allan GM, Lexchin J, Wiebe N. Physician awareness of drug cost: a systematic review. PLoS Medicine 2007;4(9):e283, http://dx.doi.org/10.1371/journal.pmed.0040283. [56] Zhang Y, Donohue JM, Lave JR, O’Donnell G, Newhouse JP. The effect of Medicare Part D on drug and medical spending. New England Journal of Medicine 2009;361(1):52–61. [57] Fu AZ, Tang AS, Wang N, Du DT, Jiang JZ. Effect of Medicare Part D on potentially inappropriate medication use by older adults. Journal of the American Geriatrics Society 2010;58(5):944–9. [58] Millett C, Everett CJ, Matheson EM, Bindman AB, Mainous III AG. Impact of Medicare Part D on seniors’ out-of-pocket expenditures on medications. Archives of Internal Medicine 2010;170(15): 1325. [59] Lopez-Casasnovas G, Puig-Junoy J. Review of the literature on reference pricing. Health Policy 2000;54(2):87–123. [60] Dukes MNG. Drugs and money: prices, affordability and cost containment. Copenhagen: IOS Press; 2003. p. 15. Available from: http://www.euro.who.int/ data/assets/pdf file/0011/96446/ e79122.pdf [cited 10.05.10]. [61] Soumerai SB, Ross-Degnan D, Gortmaker S, Avorn J. Withdrawing payment for nonscientific drug therapy. The Journal of the American Medical Association 1990;263(6):831. [62] Dickson M, Redwood H. Pharmaceutical reference prices: how do they work in practice? Pharmacoeconomics 1998;14(5):471–9.

146

M.A. Mousnad et al. / Health Policy 116 (2014) 137–146

[63] Ioannides-Demos LL, Ibrahim JE, McNeil JJ. Reference-based pricing schemes: effect on pharmaceutical expenditure, resource utilisation and health outcomes. Pharmacoeconomics 2002;20(9): 577–91. [64] Rietveld AH, Haaijer-Ruskamp FM. Policy options for cost containment of pharmaceuticals. In: Dukes MNGH-R, de Joncheere FM, Rietveld CPAH, editors. Drugs and money. Copenhagen: Published on behalf of the World Health Organization Regional Office for Europe by IOS Press; 2003. p. 29–54. [65] Simoens S. International comparison of generic medicine prices. Current Medical Research and Opinion 2007;23(11): 2647–54. [66] Sebastian S. Reference drug programs: effectiveness and policy implications. Health Policy 2007;81(1):17–28. [67] Kanavos P, Reinhardt U. Reference pricing for drugs: is it compatible with U.S. Health Care? Health Affairs 2003;22(3):16–30. [68] Giuliani G, Selke G, Garattini L. The German experience in reference pricing. Health Policy 1998;44(1):73–85. [69] Aaserud M, Dahlgren A, Kösters J, Oxman A, Ramsay C, Sturm H. Pharmaceutical policies: effects of reference pricing, other pricing, and purchasing policies. Cochrane Database of Systematic Reviews

[70]

[71] [72]

[73]

[74]

[75]

[Internet] 2006;(2), http://dx.doi.org/10.1002/14651858.CD005979 [Art. No.: CD005979]. Available from: http://www.mrw.interscience. wiley.com/cochrane/clsysrev/articles/CD005979/frame.html [10.05.10]. Mrazek MF. Comparative approaches to pharmaceutical price regulation in the European Union. Croatian Medical Journal 2002;43(4):453–61. Yang B-M. The future of health technology assessment in healthcare decision making in Asia. Pharmacoeconomics 2009;27(11):891–901. Hochman M, McCormick D. Characteristics of published comparative effectiveness studies of medications. The Journal of the American Medical Association 2010;303(10):951–8. Perlroth DJ, Goldman DP, Garber AM. The potential impact of comparative effectiveness research on US health care expenditures. Demography 2010;47(1):S173–90. Basu A, Jena AB, Philipson TJ. The impact of comparative effectiveness research on health and health care spending. Journal of Health Economics 2011;30(4):695–706. Armstrong DJ. Reviews in health promotion and public health. In: Gse HJ, editor. Cochrane handbook for systematic reviews of interventions. Chichester, UK: John Wiley & Sons; 2008.