I work in investment banking and cover hospital facilities (i.e. the "hospital chains"). You are right that the Kaiser model, or what the industry calls fully integrated systems, can often produce more efficient healthcare. For people that aren't as familiar with the healthcare system, I thought it might be helpful to provide a brief overview of why that is the case.
Since the ACA, there has been an emphasis on shifting from a fee-for-service model (FFS) to a value-based care model (VBC). Under a FFS model, doctors and hospitals are paid for each service they provide and make their profit from the margins built into the prices they charge for their services. In an ideal VBC world, doctors and hospitals are paid a certain amount for each patient they cover, and don't make additional revenue when they provide services to patients.
You can immediately see the incentives in each model. In a FFS world, doctors and hospitals are incentivized to give you the most care possible at the highest possible prices. For example, you could see how a doctor could be motivated to give a patient an unnecessary MRI. In a VBC world, the best-case scenario for the doctor is that he never sees you, and that you never enter the hospital. Each time you receive care, he spends time and money providing that care, but doesn't get paid any additional money for it. In other words, his margins decrease when he provides care (or, as those in the industry think about it, his medical loss ratio increases). When you become sick, the doctor would prefer that you seek care in a setting that is as low-cost as possible - via telemedicine, an urgent care center, or his office. That way, he keeps more of the monthly premium.
Practically, though, in the VBC world, the doctor probably wants to see you once or twice a year, to make sure you are healthy and that you are taking preventative measures to avoid becoming sick and utilizing healthcare. That investment of resources can help reduce your need for healthcare in the future, protecting his profit margin going forward.
So in a perfect VBC world, everyone's incentives line up. You spend as little time in the hospital or at the doctor as possible, and check in every once in a while to make sure you are healthy. Doctors and hospitals make more money when they don't have to treat you, and want to keep you out of the healthcare system.
The problem is, the current system is a mix of FFS and VBC. Most hospitals and doctors aren't compensated on a fully capitated basis (capitation is a concept that, in layman's terms, measures how close the system is to VBC vs FFS - more highly capitated = closer to VBC). Under FFS, doctors don't make much money providing preventative care, so there is no incentive to keep people away from the hospital. Unless you are a member of a system like Kaiser, you are most likely covered by a plan that is partially capitated - your insurer may share profits over a certain % with your doctor as a reward for keeping you healthy, but still pays for services on a FFS basis. Often, this profit sharing does not compare with the potential revenue from providing additional acute care services, so the old FFS incentives are still at play. (Hence the $600 1-minute consultation.)
Now, wouldn't it be great if every health system operated like Kaiser, where all the incentives are aligned and the objective is to keep people out of the healthcare system? Yes, but the answer is not as simple as requiring healthcare systems to provide insurance. In fact, Kaiser is one of the only success stories involving provider-sponsored health plans (Presbyterian in New Mexico is another).
The reason why many of these provider-sponsored health plans fail is twofold:
First, most healthcare in the US (70% I believe - but that is from memory) is provided by regional or community healthcare systems. These systems only serve certain communities (their primary service area, or PSA) and therefore certain populations. These populations are often not of significant enough size to provide adequate risk diversification for healthcare systems that provide insurance plans. One of the key reasons insurance works is risk diversification - but these hospitals can't diversify their coverage beyond their community's population. In fact, one of the most common criticisms of Kaiser from the investor community is that its membership is too concentrated in California - and Kaiser has over 10 million members on the East and West coasts.
Second, these systems often don't have the capability to price their insurance and healthcare services correctly to account for the levels of risk embedded in their insurance plans. Sometimes that is due to lack of actuarial experience, sometimes that is due to lack of risk diversification causing risk to exceed estimates, and sometimes that is due to existing healthcare prices driving up costs to a level beyond where they can reasonably charge premiums. Also, it is hard for these systems to predict who will enroll in their health plan, and what their overall risk level will be after enrollment season. There are companies dedicated to helping systems operate provider-sponsored health plans (e.g. Evolent), but this has proved to be a difficult problem. For examples of premier systems getting this wrong, look up Partners (they own Mass Gen), Catholic Health Initiatives, Northwell, Banner Health.
While this is a hard problem to solve, and I don't have the answer, you are right that the Kaiser model has in many ways proven to be more cost effective than FFS or other capitated models. It might be a good intermediate step for some of the larger systems. But there is still a lot of work to do to fix the system, and even solutions that sound good on paper have unintended consequences (for example, if you move to a single payer system to reduce prices, would lower drug prices disincentivize pharma R&D, hurting development in the US and the rest of the world? Would medical device companies making prosthetics go out of business if prices decreased below their cost levels? etc.).
Hopefully this was helpful to people who aren't as familiar with the way the system works. A less-than-perfect analogy I often use is:
The old FFS model is a "supermarket" model: the supermarket makes money by selling you as many gallons of milk as possible, and pricing the milk at a premium to their cost. The higher they can price the milk, or the more milk they can convince you to buy, the better off they are.
The current model is a "Costco" model: you pay a recurring membership fee to Costco, and can buy their products at a lower price, but (let's assume) you still pay a slight margin on those products to Costco. So, Costco would love to have as many members as possible paying membership fees - and, in fact, could offer its products close to cost if there were enough members who didn't use the store. But, they would still prefer that members use the store as much as possible, and buy as much product as possible at the highest allowable margin for Costco.
The "ideal" VBC model is a "Netflix" model: you pay a recurring membership fee to Netflix, and can stream any of their videos for free. Netflix starts out the month with its $10 of revenue from your membership, and each video you stream causes them to incur streaming costs and royalty payments, reducing their margin on your $10 throughout the month. From a pure profit perspective, Netflix would love to have millions of members who never used the service, allowing Netflix to keep 100% of their membership fees. However, in order to grow and be successful, Netflix needs members to use and love the service, so it "invests" some of its membership fee by streaming videos to users. If users are going to stream, Netflix would prefer that users stream its proprietary content, which is lower-cost for them to provide. (That last part of the analogy is stretching it a bit, but meant to demonstrate that for the health of the system, some utilization is required, and the provider would prefer that utilization to be as low-cost as possible.)