6. Health service provision Economic incentives and organization of the hospital sector II

1. 6. Health service provision Economic incentives and organization of the hospital sector II Tor Iversen

2. Problem to be addressed: Does activity based financing promote hospital efficiency? • Main reference: Biørn, Hagen, Iversen & Magnussen… • 1980-1997: Hospitals in Norway mainly financed by fixed budgets from the county councils • Activity based financing (ABF) was implemented from July 1st 1997. • A fraction of the block grant from the state to the county councils was replaced by a matching grant depending upon the number and composition of hospital treatments. • The number of treated patients adjusted for case-mix is measured by Diagnostic Related Groups (DRG) • All inpatient treatments are classified into 495 groups with a certain weight. • For each DRG point achieved the state pays a certain amount of NOK

3. 1997: the state paid 30 per cent of the DRG  based cost of a treatment 1998: 40 per cent 1999: 50 per cent 2002: 55 per cent 2003: 60 per cent 2004: 40 per cent 2005: 60 per cent The government’s reasons for introducing ABF: • Encourage the counties to increase the amount of resources allocated to hospitals • Increase the number of treated patients by encouraging increased hospital efficiency • Reduce the patients’ waiting time before hospital treatment For inpatient stays measured in DRG-equivalents there was an average yearly increase in hospital activity of 3.2 per cent in the period from 1997 to 2000, compared with 2.0 per cent per year in the period from 1992 to 1996

4. We predict that hospital efficiency will increase because the benefit from cost-reducing efforts in terms of number of treated patients is increased under ABF as compared with global budgets. Why? • The argument is elaborated by means of a model rather similar to Chalkley and Malcomson • A change from a fixed budget to a combination of fixed budget and revenue per treatment (ABF) is predicted to result in an increase in the level of effort and hence, an increase in technical efficiency. The reason is: Increased effort gives a higher reward in terms of number of treated patients under ABF compared with fixed budgets. Hence: ABF has incentives in the direction of increasing the level of effort • A change from a fixed budget to a combination of fixed budget and revenue per treatment is expected to be have an equal or smaller effect on cost efficiency than on technical efficiency.

5. U = u(n) + h(B+wn-c(n,e)-g) - γ(e) where u(), h() and γ() are functions. We assume where the superscript ‘ (‘’) denotes first (second) order derivative. n is the number of treated patients B is a fixed budget w is the revenue per treatment e is cost-reducing efforts g is an exogenous cost component Assumptions about the cost function

6. Maximizing U wrt n and e gives the first order condition: and the second order condition similar to the paper We model the change from a global budget to a mixed system as an increase in w and a reduction in B of a magnitude allowing the hospital to choose the same n and e after the change as chosen before the change. The reduction in the fixed budget is then assumed to be -n0Δw, where n0 is the optimal number of patients treated under a global budget and Δw is the increase in revenue per treatment.

7. By means of differentiating f.o.c. we find: and Summing up predictions: • An increase in the budget is in general predicted to have an indeterminate effect on cost-reducing effort and on hospital efficiency. • A change from a fixed budget to a combination of fixed budget and revenue per treatment is predicted to result in an increase in the level of effort and hence, an increase in technical efficiency. • A change from a fixed budget to a combination of fixed budget and revenue per treatment is expected to be have an equal or smaller effect on cost efficiency than on technical efficiency.

8. Empirical analysis Data • Hospitals are multi-product firms, treating a variety of patients with a variety of inputs. Outputs: • Inpatient care: Number of patients adjusted for case-mix by weighting patients by diagnosis related groups (DRGs). • Outpatient care: Number of outpatient visits weighted with the price paid by the state for each visit. Thus a hospital´s income from outpatient care is an approximation of a case-mix adjusted measure of outpatient care.

9. Inputs: • Physician FTEs (full time equivalents): The physician input is measured as number of FTEs per year. • Other labor FTEs: other types of labor are merged in one category. • Medical expenses: Medical expenses are measured in NOK, and deflated to 1999 prices. • Total running expenses: Total running expenses are used as input in one model, where the purpose is to provide a measure of cost-efficiency. • Data are collected for the period 1992 to 2000. We do not possess cost data for medical equipment The efficiency frontier is estimated from Data Envelopment Analysis (DEA) based on a pooled set of observations; i.e. we calculate an intertemporal efficiency frontier. This is done in order to be able to compare efficiency between years. Two models • technical efficiency: inputs in physical units: physicians, other labor and medical expenses. Adjusted number of treatments per unit of labor. • cost-efficiency: inputs in NOK. Adjusted number of treatments per NOK

10. Development in technical efficiency and cost efficiency in Norwegian hospitals 1992-2003 (1992 = 100). Source: Samdata Somatikk 2003. SINTEF Helse Rapport 1/04.

12. TABLE Static models. Estimates (t-values)

13. Results from the statistical analysis: • The introduction of ABF has improved efficiency when measured as technical efficiency • The result is less uniform with respect to the effect on cost-efficiency • What about quality and activity not measured, as teaching and research? • What about the composition of patients? A shortage of many types of health personnel when ABF was introduced – worked in the direction of increased cost of labor related to the increased production The importance of factor markets for the result of reforms in the hospital sector. In a situation with excess supply of health personnel, employment and production could have been increased with a roughly constant wage level and hence with a more beneficial effect on cost-efficiency.

14. The theoretical predictions were based on a model that assumed full information of costs and hard budget constraints. The institutional structure of the Norwegian hospital sector has probably more in common with institutional environment described as soft budget constraints described by In that case the effect of the formal financing system is expected to be small, because of the cost-compensating properties of soft budget constraints. Hence, the impact of economic incentives, as described in our model, depends on the existence of reasonably hard budget constraints. This suggests that measures that work in the direction of hardening budget constraints are essential to achieve efficiency gains in the hospital sector.

15. The effect of activity based financing on the composition of treatments and admitted patients Two types of principles behind setting priorities: Medical priority-setting (cf. rule A, B,): The cost of one treatment relative to another treatment should not influence the priority between two groups of treatments Economic priority-setting (cf. Rule C): The cost of one treatment relative to Another treatment should influence the priority between two groups of treatments ABF creates a difference between local and social costs: Hospital: Cost of resources per treatment (C) – income per treatment (- A) State: Hospitals income per treatment is the state’s cost (A) Social: Cost of resources per treatment (income per treatment cancels out) (C) From a social point of view the treatment specific income from the state is a transfer and not a real cost

16. The problem with a treatment specific income (A) is that it may vary between treatments as a proportion of a hospital’s total cost of doing that treatment (C) That causes no problem if hospital decision-makers apply medical priority- setting. But it may cause problems if hospital decision-makers apply economic priority- setting based on local costs (A-C). If the income per treatment as a proportion of the hospital’s cost of that treatment (A/C) varies between treatments, the actual priority-setting may diverge from the optimal one from a social point of view Examples