In the world of IDS, timing is everything.
The purpose of this article is to provide an overview of IDS formation nationally, based upon an ongoing inventory of local hospital systems and networks. And based on identifiable patterns of IDS development, derive some implications for investment in information systems or integrated delivery information systems (IDISs).
The vast majority of IDSs are formed at the initiative of hospital-based systems. There are notable exceptions such as the physician-based Mulligan system in the Los Angeles area, but overall the hospital-based systems are playing a central role in bringing providers together in most local markets.
Although the great variety of IDSs may be overwhelming, some general patterns can be discerned. To help make sense of the diversity, see Figure 1 on page 34 for a typology.
As indicated, we see three primary dimensions of IDS formation:
* Geographic spread--local, regional or state.
* Tightness of coupling--loose (based on affiliations, partnerships, contracts, etc.) or tight (based on ownership or control via contract management, lease or sponsorship).
* Service integration--limited (predominantly acutecare focused) or comprehensive (combines most major levels of care).
Most IDSs are local, such as the Columbia/HCA system in Orlando called the Colombia Park Healthcare System, while others are spread regionally or statewide like the California Health Network, which combines noted systems from Sacramento to San Diego. Most have achieved only a limited level of service integration. However, virtually all IDSs are in the process of building more comprehensive systems.
For instance, hospital-based systems rapidly are acquiring physician practices, corporately restructuring and moving into managed care. A few already have combined with major insurance players (HealthSpan and Medica forming Allina in Minneapolis) and others are creating their own insurance products or strategically aligning with insurance or HMO companies.
The most common model is the local, tightly coupled system that is still limited in its service integration (cell A in Figure 1). Currently, more than 460 of this model have been identified. The most rapidly growing model in terms of numbers of networks are the local, loosely coupled networks that are limited in their degree of service integration (cell B in Figure 1).
Many hospitals are members of more than one network, which is an indication of the formative stage of networks. In the state of Florida many local systems also are aligned regionally to compete within the boundaries of Florida's Community Health Purchasing Alliances (CHPAs) and have joined statewide networks to compete with Columbia/HCA, which itself controls approximately a quarter of the hospital capacity in the state. Including all such varieties, to date there are nearly 200 strategic networks and this represents an estimated two-thirds of the more loosely coupled forms that have or will be created (not all networks were included in this count).
This study requires that the networks be "strategic" combinations, that is, the major players within them have invested in the networks, formed shared boards that represent their collective interests and/or designated the network to represent them collectively in managed-care negotiations and contracting. Many networks are not strategic, but involve only limited contractual arrangements (many PPOs are of this type). Nonstrategic forms have been excluded from the data base.
Interestingly, some of the more tightly coupled forms have themselves joined strategic alliances or networks, thus mixing tight and loose structures within the same IDS. In fact, there is much overlapping across the cells of the table as the tight systems and loose networks continue to evolve into the more fully integrated forms. Note the arrows in the figure represent the primary directions of change among existing networks and systems.
Patterns in IDS formation
The combination of hospitals into local clusters began as early as the turn of the century. The most rapid movement, however, occurred in the 1980s as hospitals and other providers reacted to a stream of financing and regulatory initiatives. Table 1 summarizes the percentages of urban, nonfederal acute-care hospitals and beds that were in urban clusters (two or more hospitals in the same system in the same metropolitan statistical area (MSA)) at three points in time, 1982, 1989 and 1995 (as of January). As can be seen, the percentages have doubled in 13 years, increasing from 20 to nearly 40 percent of urban hospitals and beds that are in clusters. And, the number is still climbing.
Going beyond mere urban clusters to include other varieties of local systems and networks, the percentages of urban hospitals and beds that are in some form of IDS ranges up to 65 percent. This is shown in Table 2 on page 36.
To correct for the current underestimate of the network form itself, this number would approach 70 percent to 75 percent. This percentage is significant as it approximates the number of hospitals expected to ultimately end up in IDSs, once the consolidation phase is completed. It is speculated that from 20 percent to 30 percent of urban hospitals will either not fit into IDSs, be closed or converted for other purposes.
Table 3 (on page 36) displays the percentages by MSA size. Clearly, involvement of hospitals in IDSs is greater in the larger MSAs (where at least 70 percent of urban hospital beds are in IDSs). There are a number of reasons for this, including the greater penetration of managed care in the larger markets and the probability of finding compatible local partner increases with market size.
Although not shown, their percentages are fairly equal across the four major regions of the country (Northeast, South, Midwest and West). Historically, the Northeast has lagged other regions in terms of hospital participation in multihospital systems and managed care. But the Northeast is now among the most active in terms of both IDS and managed-care development.
IDSs and information systems
In a recent survey, hospital executives said that managed care was their top priority for IDIS development, representing a shift from earlier surveys where system and administrative purposes were given higher priority. A similar finding was reported in a study conducted by Argus Business' Research Division for this journal. This is given the significant ongoing movement toward managed care and system integration. However, the reality of what IDISs are capable of contributing lags considerably behind current expectations of system leaders.
The IDISs vary significantly in the degree they are able to integrate their participating members into a singular system of delivery. Many are at too early a stage of system formation to be capable of investing rationally in infrastructure, including in IDISs. There are notable exceptions of course (e.g., see the cover story on Allina in HMT Dec. '94).
Network forms are both fragile and evolving. Within a year or two some may either not exist or have evolved into more tightly coupled forms. And many of the more tightly coupled IDSs are still at very early stages of system evolution.
Recommendations for planning investments in IDIS capacity
* Avoid over generalizing based on the capabilities of leading systems.
It is common to assume that "follower" systems or networks are capable of moving in the same directions and at the same pace as the relatively small number of more advanced IDSs (e.g. Intermountain Health Care, Henry Ford, Sharp, Sentara, Group Health of Puget Sound). Some leading systems might be sufficiently developed to invest with purpose in more advanced IDIS capacities, such as in sophisticated switching capacities to integrate other wise incompatible provider units.
Most systems and networks, however, are in such formative stages or have opted for such loosely coupled structures that they may be incapable of going much beyond less complicated options (e.g. standardizing reports or agreeing upon data collection protocols required in managed-care contracts).
* Avoid overvaluing the power of common ownership
Being in the same system does not necessarily mean that integration among local system members is achievable or even desired by local members. Some hospitals that are in the same system based on ownership and in the same market have had to merge formally in order to achieve legally and practically "same-systemness."
* Avoid over estimating the virtues of integration
The payoffs of horizontal combinations (e.g., eliminating redundancies and excess capacity) are clear, especially if done within more tightly coupled structures. But, the rewards of vertical integration may be elusive. The advantages of integration may be significantly diminished when the costs of coordination among loosely coupled partners are subtracted out. This may be especially true for the more geographically expanded network forms.
* Avoid failing to prioritize adequately among the major functions IDIS can perform within IDSs. Information systems have three functions:
1. Internal management--cost management and generally increasing management effectiveness
2. Integrated delivery--interfacing among the various providers, introducing integrated medical records
3. Managed care--reporting and providing other responses to care and other external constituencies
Here again an IDS's priorities will depend on timing and environmental conditions. An IDS located in a small market where there is little managed-care penetration obviously will need only limited capability to respond to the demands of external constituencies. On the other hand, were the IDS is geographically dispersed, it may need to emphasize medical record transfers along established referral channels.
The more tightly coupled forms may give priority to the function of internal management. Howver, since these systems have more formal organizational structures in place, the need for information systems to support operation controls may be somewhat less than is true for the more loosely structured IDSs.
* Avoid impatience with I/S vendor product quality and capacity
I/S vendors are as new to the IDIS game as are many of the providers. They are caught in a transitional trap where they are forced to service two worlds simultaneously--one of individual institutions and providers, and the other of integrated systems. Given the organizational immaturity of many IDSs, it is little wonder the vendors may under invest, at least for now in IDIS technology and software.
The major issues facing those who invest in IDIS have to do with timing and degree. Failure to appreciate the importance of the IDS stage of development and the need to match the pace of IDIS investment with the pace of IDS evolution could lead to costly mistakes. Clearly, more time is needed for IDSs to mature and the managed-care environment to evolve before information system managers move too aggressively into the complex and expensive world of IDIS investment.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||integrated delivery systems for the healthcare industry|
|Author:||Luke, Roice D.; Olden, Peter C.|
|Publication:||Health Management Technology|
|Date:||Jul 1, 1995|
|Previous Article:||Blending the old with the new.|
|Next Article:||Information systems and integrated healthcare: an essential partnership.|