A moneymaking scan: Dual reimbursement systems and supplier-induced demand for diagnostic imaging.

In complex health systems with growing healthcare spending, combining reimbursement systems that incentivize cost-efficient healthcare provision within and across care sectors is key. This study investigates whether dual reimbursement systems lead hospitals to offset financial pressures in one care sector by inducing demand in another. We find that hospital imaging units induced demand for costly and unnecessary ambulatory imaging examinations reimbursed under fee-for-service, following a reform that introduced prospective payment and increased competition in the inpatient sector in Switzerland in 2012. Market structure, competitive pressures, and price regulations also influence demand inducement by varying the response to the reform. Reimbursement systems can influence supplier-induced demand in other care sectors within hospitals where revenue is tied to the intensity of care provision. In particular, the possibility to self-refer patients to high-margin diagnostic examinations bears negative consequences on healthcare expenditures and potentially patient health.

The impact of quality and accessibility of primary care on emergency admissions for a range of chronic ambulatory care sensitive conditions (ACSCs) in Scotland: longitudinal analysis.

BACKGROUND: Hospital admissions for Ambulatory Care Sensitive Conditions (ACSC) are those that could potentially be prevented by timely and effective disease management within primary care. ACSC admissions are increasingly used as performance indicators. However, key questions remain about the validity of these measures. The evidence to date has been inconclusive and limited to specific conditions. The aim of this study was to test the robustness of ACSC admissions as indicators of the quality of primary care. It is the first study to examine a wide range of ACSCs using longitudinal data which enables us to control for unmeasured characteristics which differ by practice but which are constant over time. METHODS: Using longitudinal data at the practice level, from 907 Scottish practices for the time period 1/4/2005 to 31/32012, we explored the relationships between the quality of primary care, and hospital admissions for multiple ACSCs controlling for a wide range of covariates including characteristics of GP practices, characteristics of the practice population, hospital effects and year effects. We examined the impact of two dimensions of quality of care: clinical quality of and access to daytime general practice. Generalised Estimating Equations taking the form of Negative Binomial regression models with the practice population included as the exposure term were estimated. RESULTS: We found that higher achievement on some clinical quality measures of primary care was associated with reduced ACSC emergency admissions. We also show that access to primary care was associated with ACSC emergency admissions. However, the effects were small and inconsistent and ACSC emergency admissions were associated with several confounding factors such as deprivation, rurality and distance to the hospital. CONCLUSIONS: The results suggest caution in the use of crude ACSC admission rates as a performance indicator of quality of primary care.

Choosing and booking-and attending? Impact of an electronic booking system on outpatient referrals and non-attendances.

Patient non-attendance can lead to worse health outcomes and longer waiting times. In the English National Health Service, around 7% of patients who are referred by their general practice for a hospital outpatient appointment fail to attend. An electronic booking system (Choose and Book-C&B) for general practices making hospital outpatient appointments was introduced in England in 2005 and by 2009 accounted for 50% of appointments. It was intended, inter alia, to reduce the rate of non-attendance. Using a 2004-2009 panel with 7,900 English general practices, allowing for the relaxation of constraints on patient of hospital, and for the potential endogeneity of use of C&B, we estimate that the introduction of C&B reduced non-attendance by referred patients in 2009 by 72,160 (8.7%).

Patient choice and mobility in the UK health system: internal and external markets.

The National Health Service (NHS) has been the body of the health care system in the United Kingdom (UK) for over 60 years and has sought to provide the population with a high quality service free of user charges for most services. The information age has seen the NHS rapidly transformed from a socialist, centrally planned and publicly provided system to a more market based system orientated towards patients as consumers. The forces of globalization have provided patients in the UK with greater choice in their health care provision, with NHS treatment now offered from any public or approved private provider and the possibility of treatment anywhere in the European Economic Area (EEA) or possibly further. The financial crisis, a large government deficit and austerity public spending policies have imposed a tight budget constraint on the NHS at a time of increasing demand for health care and population pressure. Hence, further rationing of care could imply that patients are incentivised to seek private treatment outside the constraints of the NHS, where the possibility of much greater choice exists in an increasingly globally competitive health care market. This chapter examines the evidence on the response of patients to the possibilities of increased choice and mobility within the internal NHS and external overseas health care markets. It also considers the relationships between patient mobility, health care provision and health policy. Patients are more mobile and willing to travel further to obtain better care outcomes and value for money, but are exposed to greater risk.

Healthcare in a pure gatekeeping system: utilization of primary, mental and emergency care in the prison population over time.

BACKGROUND: This study investigates the prisoner and prison-level factors associated with healthcare utilization (HCU) and the dynamic effects of previous HCU and health events. We analyze administrative data collected on annual adult prisoner-stay HCU (n = 10,136) including physical and mental chronic disease diagnoses, acute health events, penal circumstances and prison-level factors between 2013 and 2017 in 4 prisons of Canton of Vaud, Switzerland. Utilization of four types of health services: primary, nursing, mental and emergency care; are assessed using multivariate and multi-level negative binomial regressions with fixed/random effects and dynamic models conditional on prior HCU and lagged health events. RESULTS: In a prison setting with health screening on detention, removal of financial barriers to care and a nurse-led gatekeeping system, we find that health status, socio-demographic characteristics, penal history, and the prison environment are associated with HCU overtime. After controlling for chronic and past acute illnesses, female prisoners have higher HCU, younger adults more emergencies, and prisoners from Africa, Eastern Europe, and the Americas lower HCU. New prisoners, pretrial detainees or repeat offenders utilize more all types of care. Overcrowding increases primary care but reduces utilization of mental and emergency services. Higher expenditure on medical staff resources is associated with more primary care visits and less emergency visits. The dynamics of HCU across types of care shows persistence over time related to emergency use, previous somatic acute illnesses, and acting out events. There is also evidence of substitution between psychiatric and primary care. CONCLUSIONS: The prison healthcare system provides an opportunity to diagnose and treat unmet health needs for a marginalized population. Access to psychiatric and chronic disease management during incarceration and prevention of emergency or acute events can reduce future demand for care. Prioritization of high-risk patients and continuity of care inside and outside of prisons may reduce public health pressures in the criminal system. The prison environment and prisoners' penal circumstances impacts healthcare utilization, suggesting better coordination between the criminal justice and prison health systems is required.

Evaluating the Cost-Effectiveness of Early Compared with Late or No Biologic Treatment to Manage Crohn's Disease using Real-World Data.

BACKGROUND AND AIMS: We evaluated the cost-effectiveness of early [≤2 years after diagnosis] compared with late or no biologic initiation [starting biologics >2 years after diagnosis or no biologic use] for adults with Crohn's disease in Switzerland. METHODS: We developed a Markov cohort model over the patient's lifetime, from the health system and societal perspectives. Transition probabilities, quality of life, and costs were estimated using real-world data. Propensity score matching was used to ensure comparability between patients in the early [intervention] and late/no [comparator] biologic initiation strategies. The incremental cost-effectiveness ratio [ICER] per quality-adjusted life year [QALY] gained is reported in Swiss francs [CHF]. Sensitivity and scenario analyses were performed. RESULTS: Total costs and QALYs were higher for the intervention [CHF384 607; 16.84 QALYs] compared with the comparator [CHF340 800; 16.75 QALYs] strategy, resulting in high ICERs [health system: CHF887 450 per QALY; societal: CHF449 130 per QALY]. In probabilistic sensitivity analysis, assuming a threshold of CHF100 000 per QALY, the probability that the intervention strategy was cost-effective was 0.1 and 0.25 from the health system and societal perspectives, respectively. In addition, ICERs improved when we assumed a 30% reduction in biologic prices [health system: CHF134 502 per QALY; societal: intervention dominant]. CONCLUSIONS: Early biologic use was not cost-effective, considering a threshold of CHF100 000 per QALY compared with late/no biologic use. However, early identification of patients likely to need biologics and future drug price reductions through increased availability of biosimilars may improve the cost-effectiveness of an early treatment approach.

The Evolution of Health Care Utilisation and Costs for Inflammatory Bowel Disease Over Ten Years.

BACKGROUND AND AIMS: Inflammatory bowel disease [IBD] places an economic strain on health systems due to expensive pharmaceutical therapy, risk of hospitalisation and surgery, and long-term monitoring. The evolving treatment guidelines advocate rapid scale-up to biologic agents in order to improve health outcomes and quality of life. This study evaluated changes in health care utilisation and expenditures for IBD in Switzerland over time. METHODS: We extracted clinical, patient, and resource consumption data from the Swiss IBD Cohort Study between 2006 and 2016. Average unit costs for IBD-related events were derived from Swiss claims data and pharmaceutical price lists. We used multivariate regression, controlling for patient-level characteristics, to estimate trends and determinants of direct and indirect costs and resource utilisation. RESULTS: We included 2365 adults diagnosed with Crohn's disease [CD; N = 1353] and ulcerative colitis [UC; N = 1012]. From 2006-16, mean health care expenditures per patient per year were 9504 euros [70% drugs, 23% inpatient, 7% outpatient] for CD and 5704 euros [68% drugs, 22% inpatient, 10% outpatient] for UC. Health care costs increased by 7% [CD] and 10% [UC] per year, largely due to rising pharmaceutical expenditures driven by increased biologic agent use. Inpatient, outpatient, and indirect costs fluctuated and did not offset increased pharmaceutical costs. Disease characteristics were important predictors of costs. CONCLUSIONS: Increased expenditure for IBD was marked by a shift towards greater pharmaceutical management over the past decade. This study highlights the need to identify cost-effective treatment strategies in the face of increased uptake and expenditures associated with innovative treatments.

The impact of budgets for gatekeeping physicians on patient satisfaction: evidence from fundholding.

Between 1991 and 1998 English general practices had the option of holding budgets for prescribing and elective secondary care. Fundholding was reintroduced in 2005. We examine the effect of fundholding on patients' satisfaction with their practice, using a cross section of 4441 patients from 60 practices in the last year of fundholding (1998). We employ instrumental variables to allow for the endogeneity of fundholding. Patients of fundholders were less satisfied with the opening hours of their practice, their GP's knowledge of their medical history, with their GP's ability to arrange tests and willingness to refer to a specialist, and were more likely to agree that their doctor was more concerned about keeping costs down. Fundholder practices performed better on a number of process measures of care, and fundholding patients were more satisfied with additional non-medical services provided by the practice. The probability that patients were overall at least very satisfied with their GP practice was 0.073 (95% CI, 0.009-0.138) smaller in fundholding practices.

Socioeconomic inequality in small area use of elective total hip replacement in the English National Health Service in 1991 and 2001.

OBJECTIVES: To compare socioeconomic inequality in small area use of elective total hip replacement in the English National Health Service (NHS) in 1991 and 2001. METHODS: Hospital Episode Statistics and Census data were aggregated to a common geography of 'frozen' 1991 English electoral wards. The Townsend deprivation score was used as the primary indicator of socioeconomic status for each ward, and the sensitivity analysis used other Census indicators. Two main measures of inequality were examined: the indirectly age-sex standardized utilization rate ratio between most and least deprived quintile groups, and the concentration index of deprivation-related inequality in age-sex standardized utilization ratios between small areas. Each standardized utilization ratio is the observed use divided by the expected use, if each age and sex group in the study population had the same use rate as the national population. RESULTS: In both years, observed use was below expected use for the bottom third of areas by socioeconomic status. The standardized utilization rate ratio between top and bottom Townsend quintiles fell from 1.41 (95% confidence interval [CI] 1.36-1.47) in 1991 to 1.27 (95% CI 1.23-1.32) in 2001. The proportionate increase in use required to bring the bottom quintile to the level of top thus fell significantly from 41% to 27%. The Town-send-based concentration index also fell from 0.069 (95% CI 0.059-0.079) in 1991 to 0.060 (95% CI 0.050-0.071) in 2001, although this fall was not statistically significant (P = 0.085). Other socioeconomic indicators yielded a similar pattern. CONCLUSIONS: Socioeconomic small area inequality in use of total hip replacement appears to have fallen between 1991 and 2001. One possible explanation is that increased hip replacement rates in the 1990s may have lowered barriers to access, thus allowing this health technology to diffuse further among lower socioeconomic groups.

Addressing care-seeking as well as insurance-seeking selection biases in estimating the impact of health insurance on out-of-pocket expenditure.

Health Insurance (HI) programmes in low-income countries aim to reduce the burden of out-of-pocket (OOP) health care expenditure. However, if the decisions to purchase insurance and to seek care when ill are correlated with the expected health care expenditure, the use of naïve regression models may produce biased estimates of the impact of insurance membership on OOP expenditure. Whilst many studies in the literature have accounted for the endogeneity of the insurance decision, the potential selection bias due to the care-seeking decision has not been taken into account. We extend the Heckman selection model to account simultaneously for both care-seeking and insurance-seeking selection biases in the health care expenditure regression model. The proposed model is illustrated in the context of a Vietnamese HI programme using data from a household survey of 1,192 individuals conducted in 1999. Results were compared with those of alternative econometric models making no or partial allowance for selection bias. In this illustrative example, the impact of insurance membership on reducing OOP expenditures was underestimated by 21 percentage points when selection biases were not taken into account. We believe this is an important methodological contribution that will be relevant to future empirical work.

A systematic review of cost-effectiveness studies comparing conventional, biological and surgical interventions for inflammatory bowel disease.

BACKGROUND: Inflammatory bowel disease (IBD) is a chronic disease placing a large health and economic burden on health systems worldwide. The treatment landscape is complex with multiple strategies to induce and maintain remission while avoiding long-term complications. The extent to which rising treatment costs, due to expensive biologic agents, are offset by improved outcomes and fewer hospitalisations and surgeries needs to be evaluated. This systematic review aimed to assess the cost-effectiveness of treatment strategies for IBD. MATERIALS AND METHODS: A systematic literature search was performed in March 2017 to identify economic evaluations of pharmacological and surgical interventions, for adults diagnosed with Crohn's disease (CD) or ulcerative colitis (UC). Costs and incremental cost-effectiveness ratios (ICERs) were adjusted to reflect 2015 purchasing power parity (PPP). Risk of bias assessments and a narrative synthesis of individual study findings are presented. RESULTS: Forty-nine articles were included; 24 on CD and 25 on UC. Infliximab and adalimumab induction and maintenance treatments were cost-effective compared to standard care in patients with moderate or severe CD; however, in patients with conventional-drug refractory CD, fistulising CD and for maintenance of surgically-induced remission ICERs were above acceptable cost-effectiveness thresholds. In mild UC, induction of remission using high dose mesalazine was dominant compared to standard dose. In UC refractory to conventional treatments, infliximab and adalimumab induction and maintenance treatment were not cost-effective compared to standard care; however, ICERs for treatment with vedolizumab and surgery were favourable. CONCLUSIONS: We found that, in general, while biologic agents helped improve outcomes, they incurred high costs and therefore were not cost-effective, particularly for use as maintenance therapy. The cost-effectiveness of biologic agents may improve as market prices fall and with the introduction of biosimilars. Future research should identify optimal treatment strategies reflecting routine clinical practice, incorporate indirect costs and evaluate lifetime costs and benefits.

The effect of financial incentives on gatekeeping doctors: evidence from a natural experiment.

In many health care systems generalist physicians act as gatekeepers to secondary care. Under the English fundholding scheme from 1991/1992 to 1998/1999 general practices could elect to be given a budget to meet the costs of certain types of elective surgery (chargeable electives) for their patients and could retain any surplus. They did not pay for non-chargeable electives or for emergency admissions. Non-fundholding practices did not bear the cost of any type of hospital admissions. Fundholding is to be reintroduced from April 2005. We estimate the effect of fundholding using a differences in differences methodology on a large 4-year panel of English general practices before and after the abolition of fundholding. The abolition of fundholding increased ex-fundholders' admission rates for chargeable elective admissions by between 3.5 and 5.1%. The effect on the early wave fundholders was greater (around 8%) than on later wave fundholders. We also use differences in differences for two types of admissions (non-chargeable electives, emergencies) not covered by fundholding as additional controls for unobserved temporal factors. These differences in differences in differences estimates suggest that the abolition of fundholding increased ex-fundholders' chargeable elective admissions by 4.9% (using the non-chargeables DID) and by 3.5% (using the emergencies DID).

Quality of Disease Management and Risk of Mortality in English Primary Care Practices.

OBJECTIVE: To investigate whether better management of chronic conditions by family practices reduces mortality risk. DATA: Two random samples of 5 million patients registered with over 8,000 English family practices followed up for 4 years (2004/5-2007/8). Measures of the quality of disease management for 10 conditions were constructed for each family practice for each year. The outcome measure was an indicator taking the value 1 if the patient died during a specified year, 0 otherwise. STUDY DESIGN: Cross-section and multilevel panel data multiple logistic regressions were estimated. Covariates included age, gender, morbidity, hospitalizations, attributed socio-economic characteristics, and local health care supply measures. PRINCIPAL FINDINGS: Although a composite measure of the quality of disease management for all 10 conditions was significantly associated with lower mortality, only the quality of stroke care was significant when all 10 quality measures were entered in the regression. CONCLUSIONS: The panel data results suggest that a 1 percent improvement in the quality of stroke care could reduce the annual number of deaths in England by 782 [95 percent CI: 423, 1140]. A longer study period may be necessary to detect any mortality impact of better management of other conditions.

The demand for elective surgery in a public system: time and money prices in the UK National Health Service.

We construct a model of the admission process for patients from general practices for elective surgery in the UK National Health Service. Public patients face a positive waiting time, but a zero money price. Fundholding practices faced a positive money price for each patient admitted. The model is tested with data on general practice admission rates for cataract procedures in an English Health Authority. Admission rates are negatively related to waiting times and distance to hospital. Practices respond to financial incentives as predicted by the model: fundholding practices have lower admission rates than non-fundholders and respond differently to changes in waiting times and patient characteristics.

Variations in practice admission rates: the policy relevance of regression standardisation.

OBJECTIVES: To explore variations in general practice admission rates, comparing standardisation by regression with direct standardisation of the data to identify explained and unexplained variation. METHODS: Data from hospital episode statistics and the attribution dataset on 8048 cataract admissions from 109 practices in an English health district (North Yorkshire) between 1995 and 1998. Multiple regression was used to estimate the effect of practice characteristics, socio-economic factors, waiting times and distance on practice admission rates. Rankings of practices by the residuals from the regression were compared with rankings by directly standardised admission rates. RESULTS: The regression model yielded intuitively plausible results and explained 35% of the cross-practice variation in directly standardised admission rates. Standardisation by regression, compared with direct standardisation, made as least as much difference to the ranking of practices as direct standardisation compared with crude admission rates. Regression standardisation suggested that 10 practices not identified as 'unusual' by comparison of their rates to the district mean were in fact 'unusual', and that six practices identified as unusual by comparison with the district mean were not unusual once allowing for the explanatory factors used in the regression model. CONCLUSIONS: Given the increasing importance of systematic performance assessment to support quality improvement, care must be taken when interpreting variations in health care activity even after conventional standardisation of the data. If significant variations are detected, regression analysis can assist in explaining some of it, which is the starting point in informing discussions about whether variations are justified or unjustified.

Effect of a national primary care pay for performance scheme on emergency hospital admissions for ambulatory care sensitive conditions: controlled longitudinal study.

OBJECTIVE: To estimate the impact of a national primary care pay for performance scheme, the Quality and Outcomes Framework in England, on emergency hospital admissions for ambulatory care sensitive conditions (ACSCs). DESIGN: Controlled longitudinal study. SETTING: English National Health Service between 1998/99 and 2010/11. PARTICIPANTS: Populations registered with each of 6975 family practices in England. MAIN OUTCOME MEASURES: Year specific differences between trend adjusted emergency hospital admission rates for incentivised ACSCs before and after the introduction of the Quality and Outcomes Framework scheme and two comparators: non-incentivised ACSCs and non-ACSCs. RESULTS: Incentivised ACSC admissions showed a relative reduction of 2.7% (95% confidence interval 1.6% to 3.8%) in the first year of the Quality and Outcomes Framework compared with ACSCs that were not incentivised. This increased to a relative reduction of 8.0% (6.9% to 9.1%) in 2010/11. Compared with conditions that are not regarded as being influenced by the quality of ambulatory care (non-ACSCs), incentivised ACSCs also showed a relative reduction in rates of emergency admissions of 2.8% (2.0% to 3.6%) in the first year increasing to 10.9% (10.1% to 11.7%) by 2010/11. CONCLUSIONS: The introduction of a major national pay for performance scheme for primary care in England was associated with a decrease in emergency admissions for incentivised conditions compared with conditions that were not incentivised. Contemporaneous health service changes seem unlikely to have caused the sharp change in the trajectory of incentivised ACSC admissions immediately after the introduction of the Quality and Outcomes Framework. The decrease seems larger than would be expected from the changes in the process measures that were incentivised, suggesting that the pay for performance scheme may have had impacts on quality of care beyond the directly incentivised activities.

Effects of the Blair/Brown NHS reforms on socioeconomic equity in health care.

The central objectives of the 'Blair/Brown' reforms of the English NHS in the 2000s were to reduce hospital waiting times and improve the quality of care. However, critics raised concerns that the choice and competition elements of reform might undermine socioeconomic equity in health care. By contrast, the architects of reform predicted that accelerated growth in NHS spending combined with increased patient choice of hospital would enhance equity for poorer patients. This paper draws together and discusses the findings of three large-scale national studies designed to shed empirical light on this issue. Study one developed methods for monitoring change in neighbourhood level socioeconomic equity in the utilization of health care, and found no substantial change in equity between 2001-02 and 2008-09 for non-emergency hospital admissions, outpatient admissions (from 2004-05) and a basket of specific hospital procedures (hip replacement, senile cataract, gastroscopy and coronary revascularization). Study two found that increased competition between 2003-04 and 2008-09 had no substantial effect on socioeconomic equity in health care. Study three found that potential incentives for public hospitals to select against socioeconomically-disadvantaged hip replacement patients were small, compared with incentives to select against elderly and co-morbid patients. Taken together, these findings suggest that the Blair/Brown reforms had little effect on socioeconomic equity in health care. This may be because the 'dose' of competition was small and most hospital services continued to be provided by public hospitals which did not face strong incentives to select against socioeconomically-disadvantaged patients.

Does higher quality of diabetes management in family practice reduce unplanned hospital admissions?

OBJECTIVE: To investigate the association between indicators of quality of diabetic management in English family practices and emergency hospital admissions for short-term complications of diabetes. STUDY SETTING: A total of 8,223 English family practices from 2001/2002 to 2006/2007. STUDY DESIGN: Multiple regression analyses of associations between admissions and proportions of practice diabetic patients with good (glycated hemoglobin [HbA1c] ≤7.4 percent) and moderate (7.4 percent

Does better disease management in primary care reduce hospital costs? Evidence from English primary care.

We apply cross-sectional and panel data methods to a database of 5 million patients in 8000 English general practices to examine whether better primary care management of 10 chronic diseases is associated with reduced hospital costs. We find that only primary care performance in stroke care is associated with lower hospital costs. Our results suggest that the 10% improvement in the general practice quality of stroke care between 2004/5 and 2007/8 reduced 2007/8 hospital expenditure by about £130 million in England. The cost savings are due mainly to reductions in emergency admissions and outpatient visits, rather than to lower costs for patients treated in hospital or to reductions in elective admissions.

A person based formula for allocating commissioning funds to general practices in England: development of a statistical model.

OBJECTIVES: To develop a formula for allocating resources for commissioning hospital care to all general practices in England based on the health needs of the people registered in each practice DESIGN: Multivariate prospective statistical models were developed in which routinely collected electronic information from 2005-6 and 2006-7 on individuals and the areas in which they lived was used to predict their costs of hospital care in the next year, 2007-8. Data on individuals included all diagnoses recorded at any inpatient admission. Models were developed on a random sample of 5 million people and validated on a second random sample of 5 million people and a third sample of 5 million people drawn from a random sample of practices. SETTING: All general practices in England as of 1 April 2007. All NHS inpatient admissions and outpatient attendances for individuals registered with a general practice on that date. SUBJECTS: All individuals registered with a general practice in England at 1 April 2007. MAIN OUTCOME MEASURES: Power of the statistical models to predict the costs of the individual patient or each practice's registered population for 2007-8 tested with a range of metrics (R(2) reported here). Comparisons of predicted costs in 2007-8 with actual costs incurred in the same year were calculated by individual and by practice. RESULTS: Models including person level information (age, sex, and ICD-10 codes diagnostic recorded) and a range of area level information (such as socioeconomic deprivation and supply of health facilities) were most predictive of costs. After accounting for person level variables, area level variables added little explanatory power. The best models for resource allocation could predict upwards of 77% of the variation in costs at practice level, and about 12% at the person level. With these models, the predicted costs of about a third of practices would exceed or undershoot the actual costs by 10% or more. Smaller practices were more likely to be in these groups. CONCLUSIONS: A model was developed that performed well by international standards, and could be used for allocations to practices for commissioning. The best formulas, however, could predict only about 12% of the variation in next year's costs of most inpatient and outpatient NHS care for each individual. Person-based diagnostic data significantly added to the predictive power of the models.