Medical procedures are moving into outpatient facilities, mainly due to technological advances such as minimally invasive surgical procedures. But value-based care incentives are also playing a role in this trend.
CLINICAL innovation, patient preferences and financial incentives are tilting the balance in favour of outpatient settings for hospital services. Aggregate hospital income from outpatient services grew from 30 per cent in 1995 to 47 per cent in 2016.1 Some of this change is driven by patient preference and clinical and technological advances such as minimally invasive surgical procedures and new anaesthesia techniques that reduce complications and allow patients to return home sooner.
Financial incentives have likely played a role as well. Health plans and government programme payment policies support providing services in lower-cost care settings, including outpatient facilities.2 Health systems have also been acquiring or partnering with physicians and physician practices, further driving up the volume of services3 performed in outpatient settings.4
Moreover, these payers also are often using shared-savings, bundles and other arrangements that tie payment amounts to cost and quality performance. One reason for the growth in outpatient care might be health systems’ strategies to perform well under these arrangements by reducing inpatient care by shifting patients to outpatient settings. To gain greater insight into the factors driving growth in outpatient services and decline in inpatient care, the Deloitte Center for Health Solutions conducted descriptive and regression analyses using Medicare claims data between 2012 and 2015. Three key findings emerged:
Given the shift from inpatient to outpatient care, health systems will want to consider building effective strategies to grow capacity and infrastructure for outpatient services. These strategies generally have three components:
Hospital inpatient stays have declined 6.6 per cent over the past decade despite population growth and demographic shifts (such as an increasingly older, sicker Medicare population).7 In contrast, between 2005 and 2015, visits to outpatient facilities (see sidebar “Type of outpatient care settings”) increased by 14 per cent—from 197 visits per 100 people in 2005 to 225 visits per 100 people in 2015.8 Hospitals’ gross outpatient income per visit increased at an even faster pace. Between 2010 and 2015, gross outpatient income per visit grew 45 per cent, from $1,352 per visit in 2010 to $1,962 per visit in 2015.9 Health systems and hospitals have also increased their capital investments in outpatient facilities.10 As a result, as figure 1 illustrates, the aggregate share of outpatient services in total hospital income has grown over time—from 28 per cent in 1994 to almost half (47 per cent) in 2016.
Health care services can be categorised into inpatient and outpatient depending on where the procedure is performed and the length of stay. Outpatient care refers to medical services and procedures, typically low-acuity ones that do not require an overnight hospital stay. Figure 2 below describes the primary types of hospital-based outpatient facilities.
Between 2012 and 2015, outpatient income grew faster than inpatient income in all but two states, according to our analyses. But, as figure 3 shows, the mix of hospital inpatient and outpatient services in 2015 varied significantly by state. In Nevada, for example, outpatient services accounted for 35 per cent of total hospital income, while they made up 69 per cent of Vermont hospitals’ income. This variation largely reflects a combination of regional differences in physician practice patterns and other market factors.
The increase in hospital outpatient services was pronounced in Medicare fee for service (FFS) between 2005 and 2015. During this period, outpatient services per beneficiary—which include outpatient visits and imaging services—grew 47 per cent, according to the Medicare Payment Advisory Commission (MedPAC). Between 2006 and 2015, Medicare outpatient spending per beneficiary grew 8 per cent annually from $885 in 2006 to $1,753 in 2015, according to MedPAC.11
Innovation and improvements in clinical procedures likely played an important role in enabling this change.12 Many surgeries and medical and diagnostic procedures that once required an inpatient stay can now be performed safely in an outpatient setting. Patients have embraced these changes as outpatient services tend to cost less—and be more convenient—than inpatient services. Inpatient facilities tend to maintain more staff and have a wider range of capabilities, services and equipment, including resource-intensive technologies that drive up costs. Furthermore, minimally invasive surgical procedures—such as laparoscopy and robotic surgery—and new anaesthesia techniques that help prevent complications, have helped reduce recovery time for outpatient services and improved patient convenience.
Under Medicare payment policy, on-campus hospital-owned physician practices are paid more than independent physicians for the same services, which provides health systems with the incentive to buy physician practices.13 A MedPAC report found that physician-hospital consolidation increased between 2012 and 201414 and that in 2014, 39 per cent of physicians who billed for Medicare in a large national database were affiliated with a health system or hospital. This consolidation could lead to more services being performed in hospital outpatient settings.
In addition to these trends, the increase in value-based payments might spur greater shifts from inpatient to outpatient care, to reduce total cost of care and improve patient experience. Health plans and Medicare and Medicaid programmes are experimenting with payment models that reward better value (see sidebar “Main types of quality and value contracts”). These provide participating health systems with the incentive to shift services to lower-cost care settings, including outpatient ones.15 Indeed, some health care systems are building clinically integrated networks to help them perform more effectively in quality and value-payment models, partly by acquiring or partnering with physicians and physician practices.16
We wanted to explore whether hospitals that receive a higher share of income from quality and value contracts are seeing more services shift to outpatient settings. This question has not been studied well so far. We analysed inpatient and outpatient claims data from a nationally representative 5 per cent sample of Medicare beneficiaries from 2012 to 2015. We combined this data with information about hospital and market characteristics—such as hospital size, location (urban or rural), ownership type, teaching status, and case and payer mix. We also categorised hospitals by the degree to which they receive income from quality and value contracts (see the Appendix for further details).
Some health plans are tying payment to provider cost and quality performance through new payment arrangements such as:
We used hospital income data from quality and value contracts (“incentives”) to classify hospitals into three groups (see the Appendix for details):
Between 2012 and 2015, hospitals with any income from quality and value contracts accounted for about 10 per cent of the approximately 3,500 hospitals in our database. We divided that group into two: those with large incentives had an average of 23 per cent of their income from quality and value contracts and those with small incentives received 3 per cent of their income from such arrangements.
Hospitals with any incentives (large or small) generally differed from the rest. Hospitals with large incentives were more likely to be medium-sized (48 per cent vs. 34 per cent) and not for profit (73 per cent vs. 49 per cent), as well as to have a disproportionate share status (68 per cent vs. 44 per cent) and higher patient case mix (1.14 vs. 0.7), compared to hospitals with no incentives. To control for the possible influence of hospital characteristics on the association between outpatient services mix and quality and value incentives, we used a seemingly unrelated regressions estimation framework (see the Appendix).
Regression results reveal that, on average and controlling for their other characteristics, hospitals with any incentives had more outpatient visits and income than other hospitals. Moreover, we saw an even stronger relationship between outpatient services and quality and value contracts for hospitals with large incentives (figure 4). Compared with hospitals that did not report any income from quality and value contracts:
However, we did not see larger drops in inpatient visits and income for hospitals with any incentives, compared with other hospitals during the period we examined (figure 4).
Was the relationship between growth in outpatient services and presence of incentives more pronounced in certain therapeutic areas? We found the relationship was strongest for major diagnostic categories (MDCs) with higher rates of physician-hospital affiliation and technological change. Outpatient income was 18 per cent higher for diseases of the circulatory system17 and 13 per cent higher for diseases of the musculoskeletal system18 among hospitals with large incentives.
We found that compared to hospitals reporting no income from quality and value contracts:
What might explain the relationship between incentives and outpatient volume in the different therapeutic areas? We see a stronger relationship between incentives and outpatient visits and income for therapeutic areas that have seen high physician-hospital affiliation and technological change throughout the period of our study. Among physicians who bill Medicare, for instance, 53 per cent of cardiologists and 35 per cent of orthopedists reported hospital or health system affiliation in 2014.19 Outpatient income from diseases of the circulatory system20 was 18 per cent higher among hospitals with large incentives (MDC 5 in figure 5). For diseases of the musculoskeletal system,21 outpatient income was 13 per cent higher (MDC 8).
Diseases and disorders of the musculoskeletal system. Laser spine surgery is a minimally invasive procedure that no longer requires an inpatient stay. Endoscopy and live imaging are used to visualise the damaged disc and the damaged tissue is removed using a precision laser. Since the surgical scar is small, little or no postsurgery care is typically needed.22
Diseases and disorders of the circulatory system. Certain cardiology interventions—such as catheterization, percutaneous coronary intervention (PCI) and stent and percutaneous transluminal coronary angioplasties—are increasingly performed in outpatient settings.23 For instance, over 45 per cent of all PCI procedures shifted from the inpatient to the outpatient setting between 2004 and 2014.24 The change was largely driven by safety improvements stemming from clinical and technological innovations such as the use of radial access, less contrast material, bleeding risk assessments, better anticoagulation options and improved disposable products.
Diseases and disorders of the digestive system. A growing number of bariatric surgeries are performed on an outpatient basis. For instance, gastric balloons ingested by patients to achieve weight loss can now be removed endoscopically, without the need for anaesthesia or incision.25
Diseases and disorders of the ear, nose, throat and mouth. Improvements in safety, combined with technological advancements such as “dropless” surgery, mean that most cataract surgeries can now be performed in outpatient settings.26
Diseases and disorders of the respiratory system. More than 70 per cent of patients who undergo thoracoscopic surgery can be discharged on the day of surgery itself due to the use of new techniques and technologies such as short endoscopes with small incisions and advanced robotic technological aids.27
Our original hypothesis was that we would find a more pronounced shift from inpatient to outpatient care among health systems with greater value and quality incentives. While we found higher use of outpatient care, we did not find lower use of inpatient care than for other hospitals. One reason may be the very small proportion of hospitals with any type of incentive contracts at all, the relatively recent experiences with these contracts, or the limited amount of risk these hospitals may be facing.
Nevertheless, it is interesting to find that hospitals with incentives have greater outpatient services. Many hospitals are trying to increase their outpatient services both as a defensive mechanism to react to new and more aggressive competitors and to diversify their revenues. Greater outpatient business may also position hospitals to do well under contracts that consider the whole spectrum of care in the future and that reward closer physician-health system collaboration.
Going forward, hospitals and health systems, especially those that get a large portion of their income from value contracts, will likely have to address the need to move treatment from inpatient to outpatient settings. Is there a road map for this transition?
Health systems may want to consider their investments in both human and physical capital. Expanding outpatient services may call for building partnerships with organisations that now have the capacity (for example, ambulatory surgery centres, outpatient clinics and retail centres) and human capital (physicians and other clinical staff) to support care in these settings, as well as considerations around referral patterns, workflow and operational improvements. Building physician relationships and networks through partnerships or affiliations can help increase capacity and attract patients. Capacity and capabilities can help health systems succeed in both fee-for-service payment systems and value-payment arrangements.
Virtual care/technology can be a part of the outpatient strategy, allowing health systems to add capacity and generate referrals as well as provide a lower-cost setting for treatment.
Finally, technology can help health systems manage operations and patient care more efficiently. For example, case management, supported by analytics, can help health systems work with patients to decide on which care setting is the most effective, safe and efficient.
The Deloitte Center for Health Solutions performed regression analyses to study the association between quality and value incentives and hospital inpatient and outpatient visits and income. We used controls for factors that could influence this association, including hospital organisational characteristics (such as hospital size, urban/rural location, ownership type, service mix, teaching status and being part of a system), case and payer mix, as well as local market conditions.
Our main regression specification was a system of four linear equations (one for each of the four hospital service metrics) of the following form:
Hospital services metrics = f (quality and value incentive indicators, hospital organisational characteristics, case and payer mix, local market characteristics and year indicators)
The variables are as follows:
In these models, the unit of observation is the hospital-year cell. In the MDC analyses, the unit of observation is the hospital-MDC-year cell. Since we include state and year indicators, the association between quality and value incentives and hospital service mix is estimated from changes in incentives in a given hospital over time, as compared to other hospitals with similar characteristics in the same state. We use a seemingly unrelated regression estimation framework to account for the correlations between our hospital service metrics and we correct the standard errors for clustering on hospital referral regions (HRRs). The adjusted R-squared in our estimations varied between 70 and 79 per cent.
We mapped the ICD-9 and ICD-10 codes from the Medicare LDS claims data to their respective diagnosis-related groups (MS-DRGs), which were in turn mapped to their respective MDCs. MDCs were devised by physician panels to ensure DRGs are clinically coherent, since MDCs are mutually exclusive categorizations of all possible diagnosis codes. Each MDC corresponds to a single organ, system, or medical speciality. Public health departments28 use MDC coding in their inpatient discharge and emergency department modules.
In our data, information was not available for MDC 15 (newborns and neonates with conditions). The 24 other MDCs we analysed are listed below in table 1.