Sascha G. Wolf, Forecasting Health Care Expenditure – Review of the Literature in:

Sascha G. Wolf

Pharmaceutical Expenditure in Germany, page 18 - 21

Future Development, Political Influence and Economic Impact

1. Edition 2009, ISBN print: 978-3-8329-4164-2, ISBN online: 978-3-8452-2005-5

Series: Neue Studien zur Politischen Ökonomie, vol. 6

Bibliographic information
18 sive dataset enables separate examination of the single indication areas of the “Red List”, the pharmaceutical catalogue of the SHI. By means of grouping drugs whose members are characterised by common attributes, we will show that support for both the medicalisation and the compression theses can be found depending on what kind of disturbances of health are affected. Considering different evolutions of diverse disease patterns facilitates more sophisticated forecasting than can currently be found in literature. The chapter is organised as follows: since the research done on the pharmaceutical sector has been minimal, Section 2.2 gives an overview of the literature which is devoted to the prediction of total SHI health care expenditure. We will arrive at the conclusion that the results of the forecasts differ depending on the assumptions about the dynamic development of age-speci? c health care demand. Therefore, section 2.3 gives insights into the long-lasting discussion about the impact of age on expenditure and introduces the medicalisation and the compression theses. Afterwards, in section 2.4, we will start with the empirical analysis. By means of principal components and cluster analysis, groups of drugs with similar characteristics will be built and analysed due to their age-dependency. Based on these results, in section 2.5 an outlook on SHI’s future drug disbursements will be accomplished. Section 2.6 evaluates the predictions in respect to their political implications. Finally, section 2.7 draws the conclusions. 2.2. Forecasting Health Care Expenditure – Review of the Literature Forecasting SHI’s future expenditure development has been subject to numerous studies since the 1980s. A way of simplifying these surveys can be to divide them into those that consider only purely demographic impacts and those that additionally include the in? uence of technological progress. In general, “purely demographic impact” means that the prognoses rest upon the assumption that per capita age-related expenditure remains ? xed over time when the state of medical technology is controlled. In such a “status quo approach”, (Breyer and Felder 2004, p. 3) future expenditure changes arise solely from alterations in the demographic structure. Everything else, i.e. standards of diagnosis and therapies, quantity and quality of treatments, morbidity and mortality etc., are assumed to be constant and increasing disbursements are only due to an absolute increase in the number of the elderly. Recent surveys in particular state that this purely demographic impact will have only minor cost-driving in? uence and will lead to moderate increases in contribution rates of between 15 and 19 % by 2040 (? gure 2.1).2 However, as soon as technological progress is included, a far more dynamic expenditure development is being forecasted. 2 Studies which consider only demographic impact (selection): Schmähl (1983), Erbsland and Wille (1995), Knappe (1995), Oberdieck (1998), PROGNOS (1998), Buttler et al. (1999), Erbsland et al. (1999), Knappe and Optendrenk (1999), Hof (2001). 19 The interaction of both determinants could result in contribution rates increasing of up to more than 30 % by 2040 and almost 40 % by 2050 (Postler 2003).3 Figure 2.1: Review of the Literature: Range of Predicted SHI Contribution Rates (Projection Year: 2040). 15,3 19,419,3 32,0 0 5 10 15 20 25 30 35 Purely Demographic Effect Incl. Technological Progress Co n tri bu tion Rates in Pe rc e n t Lowest Highest While the studies mentioned which analyse only the purely demographic effect come to quite similar predicted contribution rates of between 15.3 and 19.3 %, the results of the estimates which additionally consider technological progress differ signi? cantly from 19.4 to 32 %. This is not only due to unequally applied forecasting methods, but also to the strong data restrictions they use. In contrast to the static view of purely demographic effects, a sophisticated analysis which considers dynamic cost development needs much more information and many additional assumptions. Because of the lack of information about age-related expenditure pro? les, the ? rst attempts to forecast SHI disbursements applied theoretical approaches. Schmähl (1983) used a simpli? ed budget constraint. Under the tenet that revenues equal expenditure in the same year, he came up with an equation which showed that contribution rates depend on four relationships: (i) the proportion of average per-head expenditure of employees to the average gross income (which determines the premium of the compulsory health insurance), (ii) the per-head expenditure relation between retirees and SHI-members, (iii) the ratio of retirees to non-retirees, and (iv) the relative amount of pensions. Holding all relationships constant, with the exception of the quota of retirees to non-retirees (see iii), Schmähl derived the purely demographic effect which will lead to a contribution rate of around 16 % by 2040. Knappe (1995) chose the same approach and reached comparable results for the demographic effect. Additionally, he added a second step which varied the expenditure relation between retired 3 Studies which consider both demographic development and technological progress (selection): Dudey (1994), Knappe (1995), Oberdieck (1998), PROGNOS (1998), Breyer and Ulrich (2000), Buttler et al. (1999), Hof (2001). There also exists the opinion that neither demographic ageing nor technological progress coercively determines increasing expenditure. See Pfaff (1994). 20 and non-retired persons (see ii) according to the developments between 1960 and 1992. He did this to try to simulate technological progress. Together, these effects result in a contribution rate of 25 % by 2030. An analogical approach was chosen by Oberdieck (1998), who distinguished between retirees and employees (see iii) in a more sophisticated way. Under the assumption that the average retirement age will raise to 62 years, he determined a contribution rate of 31.2 % in 2040. Finally, the study by Postler (2003) can also be assigned to this group of SHI budget restraint referring surveys. Assuming an incremental decrease of pensions to 48 % of gross income level and by extrapolating the expenditure which correspond to the development between 1990 and 2000, in the worst case he estimated a contribution rate of 39.5 % in 2050. So far the studies quoted used the proportion of SHI members beyond a certain age to younger members to picture the insurants’ age structure. This approach delivers only inexact results about the impact of gradual population ageing because health care expenditure differs within the two groups and the intense aggregation camou? ages age-related cost developments. Therefore Erbsland and Wille (1995) divided the SHI members not only into employees and retirees, but into 15 separate age groups. They resorted to an estimated age-related expenditure pro? le for the year 1995. Additionally, they analysed six different cost sectors separately: ambulant treatments, dental treatments, in-patient costs, pharmaceuticals, curative facilities, and dental prostheses. The authors came to the conclusion that pharmaceutical expenditure is the most agerelated cost-unit in the SHI. Considering only demographic effects, they estimated an increase of contribution rates by 4 percentage points by 2030. In the same way Buttler et al. (1999) used age-related expenditure pro? les and computed a purely demographic driven contribution rate of 19.1 % by 2040. The econometric approach is traced back to Breyer and Ulrich (2000). By means of ordinary least squares (OLS) regression the authors tried to explain per-head SHI expenditure. The included exogenous variables: the ratio of people which are older than 65 years to the total number of members in the SHI, the gross income, which determines the amount of SHI contributions, and a time trend. Based on the 8th Coordinated Population Projection of the Federal Statistical Of? ce, they estimated a contribution rate of 23.1 % in 2040. Similarly, by applying the population projection of the German Institute for Economic Research (DIW), Breyer et al. (2001) predicted a contribution rate of 34 % in 2040. Apart from methodological problems concerning non-stationarity (Breyer et al. 2001, p. 40) and the arguable assumption of consistent robust interrelations between the variables (Breyer and Ulrich 2000), the econometric approach is much more ? exible and allows additional in? uencing determinants to be considered. For example, Breyer and Felder (2004) applied the econometric approach to more comprehensive Swiss data and included a separate inspection of persons in their last four years of life. They came to the conclusion that the impact of technological progress on health expenditure is much larger than the impact of ageing. These results were corroborated by Stearns and Norton (2004). Further regressions were accomplished by Hof (2001) and Sauerland (2002). Hof calculated a contribution rate of 22.6 % by 2050 and Sauerland between 25 and 31 % by 2040. 21 2.3. Age and Health Care Expenditure The great differences of the predicted contribution rates primarily arise from the assumptions about the in? uence of medical technological progress and its impact on the long-term development of the age-speci? c health care expenditure. Age-speci? c health care disbursements can be visualised by means of “per capita age-related health care expenditure pro? les” (in the following referred to as “expenditure pro? les”), which show health care costs as a function of age. The current (status quo) expenditure pro? le of the SHI shows the typical shape (? gure 2.2): except for comparatively high costs for infants, disbursements stay very low in younger ages and reveal only a slight upward trend. Starting at the age of 40, however, the gradient of the expenditure pro? le suddenly becomes much steeper. This observation has led to the popular belief that there is a strong relationship between the ageing population and increasing health spending, thus “as the elderly’s share of the population increases, so too will the demand for health care” (Longman 1987, p. 88). Figure 2.2: Per Capita Age-Related Health Care Expenditure Pro? le of the SHI in 2004 (Status Quo) and Hypothetical Future Pro? les According to Medicalisation and Compression Theses. 0 100 200 300 400 500 600 700 800 900 1000 0 10 20 30 40 50 60 70 80 90 Ag e 1 0 = 1 00 P er ce nt Age Medicalisation Thesis Status Quo Compression Thesis Source: Based on data of the German Federal (Social) Insurance Authority. Presentation following Fetzer (2005). Of course, it cannot be denied that a direct cost-driving effect results from the ageing population. Since older people generate higher expenditure than younger people, an increasing number of the elderly may cause rising health care disbursements. But, in fact, health care expenditure depends on a wider set of in? uences (e.g. Gray 2005). In particular, technological progress leads to changes in medical therapies and treatments and affects morbidity as well as mortality (Ulrich 2000). Consequently, the expenditure pro? le is not ? xed, but varies over time. The way the expenditure pro? le’s shape

Chapter Preview



Der Arzneimittelsektor der Gesetzlichen Krankenversicherung stand wiederholt im Fokus zahlreicher Gesundheitsreformen. Dennoch ist es bislang nicht gelungen, den Trend steigender Ausgaben nachhaltig zu bremsen. Die vorliegende Untersuchung leistet einen Beitrag dazu, die Ursachen dieser Entwicklung zu erklären und Lösungsansätze aufzuzeigen. Mittels Hauptkomponenten- und Cluster-Analyse wurden Gruppen von Arzneimitteln mit vergleichbaren Konsumeigenschaften gebildet. Jede Gruppe wurde auf den Einfluss der Altersabhängigkeit und des technologischen Fortschritts hin analysiert. Aufbauend auf diesen Ergebnissen wurde eine Prognose der zukünftigen Ausgabenentwicklung bis zum Jahr 2050 erstellt. Obwohl die Hauptkostenfaktoren exogen sind, steht der Gesetzgeber dem vorhergesagten ansteigenden Kostenpfad nicht hilflos gegenüber. Im Gegenteil: Anhand ökonometrischer Tests wird gezeigt, dass die Gesundheitspolitik in der Vergangenheit durch wahl- und klientelorientierte Interessendurchsetzung geprägt war. Mehr Effizienz in der Arzneimittelversorgung könnte durch die Einführung individueller Gesundheitssparkonten erzielt werden. Dies bestätigen die Resultate eines vertikal differenzierten Wettbewerbsmodells.