Background
As in other managed care organizations, the expenditures of Israel’s health plans on their members are influenced by a variety of factors, including sociodemographic characteristics, health behaviors, and morbidity levels, including preventable diseases. In contrast, in Israel, the government’s compensation mechanism is currently based almost entirely on a risk-adjusted capitation formula with only three variables: age, sex, and residence in the geographic periphery. As in other countries, at times a gap may arise between actual expenditures on members and the compensation received by the health plans. This gap can create an incentive for “cream skimming” (i.e., selecting relatively healthy members) and for avoiding accepting members whose care entails high expenditure or avoiding providing them optimal care. At the same time, in Israel as in other countries, the opposite situation, where compensation fully reflects expected expenditures, may reduce incentives for disease prevention: if improved health lowers expenditures but does not generate financial gain, health plans may have less motivation to invest in prevention. This phenomenon is known in the literature as the “prevention paradox,” as it weakens incentives to promote health.
This study examined differences between actual healthcare utilization and normative utilization (an estimate calculated according to the capitation formula), by population groups. The difference between actual and normative utilization makes it possible to assess the effectiveness of the capitation mechanism in terms of fairness in the allocation of funds and reducing incentives for selection. The study focuses on actual and normative utilization according to engagement in primary prevention behaviors (physical activity and avoiding smoking) and two preventable diseases (heart attack and/or stroke and diabetes).
The study was commissioned by the Health Economics Unit of the Ministry of Health’s Strategic and Economic Planning Administration headed by Dr. Ayelet Greenbaum, and was accompanied by Dr. Hani Schroeder, Rachel Brenner-Shalom, and Alexei Belinsky.
Objectives
- Identify variables that may help explain differences between actual healthcare utilization and normative utilization and assist in identifying points of intervention to improve the current compensation mechanism.
- Examine whether the prevention paradox is present in the Israeli compensation mechanism, and, specifically, a comparison between (a) the actual and normative use of healthcare services by population groups based on engagement in primary prevention behaviors, and (b) the actual and normative use of healthcare services by population groups according to the number of people with each of the preventable diseases examined.
- Use the findings to propose improvements to the compensation mechanism, including identifying intervention points that could strengthen incentives for health plans to promote healthy behaviors and prevent disease.
Method
The analysis drew on data from the National Health Survey conducted by the Israel Center for Disease Control in the Ministry of Health in 2013–2015 (4,511 respondents, 3,604 of whom lived in central Israel) and on SHARE data (the Survey of Health, Ageing and Retirement in Europe and Israel) from Waves 8 and 9 (Wave 8: 2019–2020, pre-COVID-19; Wave 9: 2021–2022) (1,252 respondents aged 55+, with no duplicates).
For each respondent, two measures were calculated: actual utilization of healthcare services, and normative utilization according to the existing Israeli risk-adjusted capitation formula. The gap between the two measures was used to assess whether health plans experienced profit or loss for different groups, defined by health behaviors (smoking, physical activity), preventable diseases (heart attack and/or stroke, diabetes), income, and education.
Key Findings
- A consistent pattern was observed: individuals with higher education and/or income are profitable for health plans, while those with lower education and/or income generate losses (hereafter: are unprofitable). These differences existed even independent of smoking, physical activity, or disease.
- Smokers, individuals who do not engage in physical activity, and individuals with preventable diseases are unprofitable for health plans compared to others.
- The analysis also showed that members of higher socioeconomic status are profitable for health plans when they exhibit healthy behaviors and before disease onset, but are unprofitable when they develop disease, similar to ‒ or even greater than ‒ other groups.
Recommendations
- Consider incorporating socioeconomic risk factors, such as income support benefits or disability benefits, into the capitation formula to reduce incentives for selection.
- Avoid including specific diseases in the formula, due to concern that doing so could weaken incentives to treat them. However, weighted morbidity measures may be considered.
- Add retrospective components that encourage disease prevention, especially among older adults, where short-term benefits are more observable.
- Strengthen oversight of mechanisms designed to promote prevention to ensure they are not used for “cream skimming.”
Citing suggestion: Brammli-Greenberg, S., & Gorelik, Y. (2025). Preventive Medicine and the Health Plans’ Compensation Mechanism: Is There a Paradox? S-242-25. Myers-JDC-Brookdale Institute. (Hebrew)