Health insurance market summary
The publicly funded English National Health Service (NHS) is a tax-funded system that provides (largely) free healthcare at the point of use to all UK citizens. It is funded out of general taxation, rather than by a social insurance scheme or earmarked tax.
The NHS budget is currently devolved to regional bodies, called primary care trusts (PCTs), which must then fund care for their geographically defined populations, ranging from 90,000 people to over 1 million. PCTs sometimes provide primary care service by employing clinicians directly, but more likely are involved in 'commissioning' and paying for the care deemed necessary to meet the needs of its population. As there are few defined benefits under the NHS, a PCT must perform a juggling act and identify needs and clinical priorities to ensure that they have enough budget throughout the year to pay for care.
PCTs receive their budget allocation based on a risk-adjusted methodology, which takes into account the demographic structure of each population, as well as loadings for 'deprivation,' which is deemed to be a broad indication of clinical need. However, there is little sophisticated use of past clinical data to risk adjust the budgets according to a future perceived demand based on past health status. In addition, a PCT that underspends its allocation is likely to have the surplus redistributed to another PCT that has overspent. In this way, a very informal and uncodified ex-post experience adjustment is made to the budget allocation, but this adjustment could be as much to do with poor management as true risk differences.
Under proposed reforms, PCTs will be formally abolished and instead the funds will be held directly by clinicians (mainly GPs), grouped together in clinical commissioning groups (CCGs). Estimates vary on the size of these groups, but they are likely to cover populations similar in size or slightly larger than PCTs. It is unclear how budgets will be allocated to CCGs, or budgets allocated to practices within CCGs, but it is likely there will be a more sophisticated type of risk-adjustment methodology required than is currently used. In addition, there is an expectation that CCGs may be held fully accountable for their allocated budget with a series of penalties for overspending through poor performance, and financial incentives for underspending (combined with quality outcomes). But this is not clear as the new reforms are still evolving.
The private medical insurance (PMI) market in the UK provides complementary and top-up care to the NHS. Plans cover acute inpatient and outpatient treatment in private hospitals and private wings of NHS hospitals. They do not generally cover primary care or care of ongoing, chronic conditions, although many do provide full coverage of cancer.
PMI is purchased for a range of reasons, including faster access to operations (bypassing any NHS waiting lists), better and cleaner "hotel-style' hospital facilities, and more choice of consultants.
Approximately 12% of the UK population has PMI coverage, with the percentage being higher in the more affluent areas of London and South East England. Two-thirds of covered lives have the policy provided as part of their employee benefits package and paid for by the employer, and one-third has self-funded individual policies. Employers pay for PMI because it is seen as a necessary employee benefit to attract staff in certain industries, but also because it theoretically reduces the potential time an employee in need of an operation may be absent from work.
Because PMI is a 'luxury' purchase and the NHS is available for all health needs in theory, there are few restrictions on the rating or benefits of PMI policies, beyond the normal financial services regulations on 'treating customers fairly' and EU legislation on discrimination. Therefore, PMI insurers can charge whatever premiums they like, using whatever rating factors they wish, within reason, as long as they are not unfairly discriminating without statistical evidence of a differential in risk. There are no requirements to file rates or any restrictions on the levels of loss ratios that are acceptable. The PMI market is reasonably competitive--particularly for large corporate groups. Many of the very largest corporate groups self-insure all or most of the risk, using the "insurer' to provide stop-loss coverage and/or administration services or access to provider networks.
Aside from the domestic PMI market in the UK, a large number of international private medical insurers are also domiciled or operate from within the UK. These provide medical insurance to expatriates around the world (some also to indigenous populations). Benefits are usually far more comprehensive than those offered to the domestic UK market because of the necessity to cover all the potential medical needs of expatriates around the world.
Underwriting practice
Primary care trusts (PCTs) are not allowed to underwrite--they must accept to pay for care for any eligible citizen who lives within their geographically defined criteria. In theory, patients have a free choice of where they access secondary and tertiary care, but patients must currently access this care via their GP in their geographical area and at the surgery where they are registered. GPs are currently allowed to refuse to enroll a citizen if their list is deemed to be 'full' and the PCT is obliged to find the patient another GP within the area who will accept them. It is not clear how this will translate to a world where GPs hold the risk and budget and whether or not a GP practice will be allowed to turn down a patient thought to be an unacceptable risk, which is unlikely. However, it is hard to see how guaranteed acceptance would work, as some GPs will just not have the capacity to take on more patients.
Private medical insurance (PMI) insurers can underwrite with few restrictions. Most medium and large groups are not underwritten (generally anything over 50 lives), but are accepted on a medical-history-disregarded basis, i.e., full coverage for eligible benefits immediately. Individuals and small groups are generally offered a choice of moratorium underwriting--where there is no underwriting questionnaire to fill out, but preexisting conditions are declined at the point of claim--or full medical underwriting, where the applicant must complete a medical questionnaire and then may have specific exclusions applied. There is no rating for existing conditions. Instead, all applicants are accepted on standard rates, with exclusions, rather than loaded rates with coverage for preexisting conditions.
Contrary to this, insurers providing international medical insurance to expatriates around the world do often rate rather than exclude, though the final decision often rests with the subjective view of the underwriter based on the evidence presented at proposal.
Large groups are experience-rated and are often re-brokered each year, resulting in a highly competitive market. Small groups have some weight put on their experience, usually with some smoothing out of large claims, combined with book rates. Individuals are priced on book rates. Some companies have a no-claims-discount structure, where a policyholder who has not made a claim in the previous year will have a discount to their premium applied at renewal, partially offsetting the rise that is due to age.
Insurers are not allowed to medically re-underwrite policies at renewal, but must continue with the initial medical underwriting terms. Therefore, over time, an underwritten book of business has experience that gradually worsens, as the select effect of underwriting wears off. Insurers have to be careful to load rates appropriately for this in the first year, or face having to raise premiums sharply, leading to selective lapsing and a death spiral.
Some insurers offer continuation underwriting, where a member or group can switch insurers and continue on the same underwriting terms, with a small pricing load, compared with a fully newly underwritten group. Insurers also often offer continued underwriting terms to members of corporate schemes who leave service and take out their own individual policies without a break in coverage--some of this is on a guaranteed basis.
Risk adjustment and prediction is used in a number of ways in the NHS:
- To allocate budgets to PCTs
- For internal purposes within PCTs to identify patients at high risk for case management
Both of these are more prospective (e.g., predicting potential utilisation) than retrospective (e.g., equalisation).
Going forward, we expect to see it used much more once clinical commissioning groups (CCGs) are up and running, but until the financial arrangements have been clarified, it is difficult to speculate where it will be applicable.
Underwriting manuals are routinely used in the underwriting of PMI to assess and predict likely future risk, but risk adjustment is rarely used as they don't price for health status or conditions and they rarely receive enough information to use commercial risk adjuster products to predict spending or health status. Because PMI insurers only receive data on a small part of a patient's possible contact with the medical system, they only have limited information--for example, they would not have information on the patient's contacts with primary care, chronic conditions, excluded conditions, outpatient drugs, or even, sometimes, outpatient consults and diagnostic tests. When they do cover the condition/treatment, they only receive limited medical information--for example, they would know the primary procedure code and sometimes the primary diagnosis code, but not any secondary diagnosis codes. The results of diagnostic tests would not be part of the claim record. There is usually insufficient information on a PMI claim record to run an HRG grouper for example.
The lack of data means that most PMI insurers would not attempt to use risk adjustment and their business models do not imply a need to. Although many have case management departments covering cancer, back pain, and psychiatric claims, patients are referred to these at the point of claim and insurers do not attempt to predict in advance which patients are most likely to benefit from or have need of these. Arguably, these case managers might be more effective at preventing claims if they had some information available from a risk adjustment program that identified which patients are most at risk of having high-cost inpatient admissions, but the insurer would be unlikely to have sufficient historical data to give any sensible predictive value from a model.
Final thoughts
International (expatriate) private medical insurance (PMI) insurers may find sophisticated medical underwriting tools to be advantageous. These insurers often rate rather than simply exclude preexisting conditions, and underwriting is therefore critical in the risk management process.
Underwriting tools may also be able to provide potential benefits within the NHS, as a health risk assessment tool. Underwriting tools often have a clinical rather than statistical focus, and would align well with GPs for example. So, although it would be a non-traditional use, as GPs are not allowed to underwrite, these underwriting tools can provide assistance to individual GP practices that are looking to identify people who need more intensive case management when they first seek services from the practice. These tools also provide GPs with a way of tracking relative risk over time for their populations. Underwriting tools that make the link from drugs to diagnosed condition, and associated costs, would be particularly useful.
The use of risk adjustment models to allocate budgets and assess risk and performance is not new to the NHS. It is not yet clear what the financial and risk transfer arrangements will be between the different clinical commissioning groups and the GP practices within those groups. Though with an expectation of increased financial accountability falling on the shoulders of the clinical commissioning groups (CCGs), what is clear is that the use of risk adjustment models within the process of allocating financial budgets to commissioners will take on greater significance, particularly within the wider financial risk management framework.