Abstract
Objective To promote managed competition in Dutch health insurance, the insured are now able to change health insurers. They can choose a health insurer with a low flat‐rate premium, the best supplementary insurance and/or the best service. As we do not know why people prefer one health insurer to another, we investigated their reasons for selecting their health insurer and assessed the importance of the supplementary benefit package and the flat‐rate premium.
Methods A self‐administered questionnaire was completed by 468 of a total of 884 (52.9%). Data were compared among three groups. The first group comprised those who left one health insurer for another (exit). The second group had joined the health insurer (entry) and the third group comprised those who did not switch (stayers).
Results Those in the entry group were statistically significantly less satisfied with their former insurance organization than those in the other groups (exit and stayers) with the insurance organization under investigation. They were also less satisfied than the other groups in respect of the flat‐rate premium. Those in the exit group were younger and seemed to be in better health. In general, the insured were only aware of small differences between health insurance funds and the three groups did not differ from each other in this respect. About a quarter of the entry group reported the flat‐rate premium as a reason for selecting a particular health insurance fund. However, the most frequently reported reason, for both exit and entry, was the benefit package of the supplementary insurance.
Conclusions In the absence of clear differences between insurance organizations, the advantages of managed competition maybe too difficult to achieve.
Keywords: managed competition, consumer preferences, health insurance
Introduction
During the past decade, an increasing number of industrialized countries have introduced managed competition in their social health insurance scheme. 1 , 2 , 3 Managed competition should allow social health insurance organizations to become more consumer‐oriented and more active in managing the provision of health care. Key features are freedom of choice for subscribers and effective price competition among organizations.
Historically, social health insurance programmes were designed to guarantee financial access to health care, as large numbers of people in Europe could not afford to consult doctors or other health providers. In many countries, such as Germany, Switzerland, the Netherlands and Belgium, social health insurance organizations were founded. During the 1970s, the Dutch social health insurance scheme evolved into a system with regional monopolies, where individuals living in a certain region automatically subscribed to a regional health insurance fund. All funds charged the same income‐dependent premiums and provided uniform benefit packages. So, in order to guarantee universal access, the health care system was heavily regulated at the expense of incentives for efficiency and innovation. 4
A solution to the problem of combining equal access to health care and incentives for efficiency and quality can, in theory, be found in the model of managed competition. 5 , 6 Preconditions for managed competition are: an annual opportunity for subscribers to enrol in one of several competing health plans; an obligation for insurers to accept anyone (irrespective of health status) on the same terms; a standardized benefit package and an adequate system of risk‐adjusted payments to insurance organizations for part of their expenses. Insurance organizations could then engage in a value‐for‐money competition. 7
Recent reforms in the Dutch social health insurance seem to be close to introducing managed competition in health care, because the above‐mentioned conditions for competition have been secured. 8 , 9 , 10 In 1992, freedom to choose a health insurance fund was introduced. At the same time the premium structure was changed from wholly income‐dependent to partly income‐dependent and partly flat rate. Health insurance funds became responsible for an increasing share (currently 38%) of their expenses, leading to different flat‐rate premiums between funds. However, the number of people switching from one health insurance fund to another is still relatively small. In the past 5 years, only a minority have considered changing and only very few (7%) did so. The main instrument of competition, the flat‐rate premium, was not the main reason for changing health insurance. 11
In order to gain more detailed knowledge of the reasons why people switch between insurance funds, we conducted a study among the members of one of the largest Dutch health insurance funds. Information about the reasons for changing insurer may be useful in understanding why only a few people change from one health insurance fund to another. Low levels of switching may not provide adequate demand signals in a system of managed competition.
Research on switching health insurance organizations in European health care systems is scarce. In Germany, contribution rates between organizations differ significantly and switching occurs on a large scale. 12 In Belgium, no national information is available, but people do not seem to switch very often and the main instruments of competition are the benefits of supplementary insurance. In Switzerland, although premium differences are quite considerable, competition seems to relate to risk‐selection rather than cost. 13
Much more research has been carried out in the United States in the context of choosing a health plan. Nearly all empirical US studies have found insurance premiums to have a statistically significant negative effect on the probability of enrolling in a health plan. 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 Consumers are most cost‐sensitive when their new health plan is similar to their old health plan. 23 About two‐thirds of the people who change their health plan stay in the same plan type. 24 Feldman et al. 25 also report that changing prices leads to health plan switching within similar health plans.
Various studies in the United States have indicated that people do not only look at price, but also at benefits and the availability and quality of doctors. 20 , 26 , 27 , 28 , 29 Only a few studies have taken all these variables into account. 30 , 31 Chakraborty et al. 31 report the coverage of hospital care to be the main determinant in health plan choice, followed by choice of doctors, cost, dental cover and choice of hospitals.
When Dutch consumers change from one health insurance fund to another, conditions are similar and benefits uniform (see Box 1). So they change to a very similar type of health plan. Because of these similarities, we would expect the premium to have an influence on switching behaviour. As the flat‐rate premium of the insurance fund under investigation is among the lowest in the country, it is expected that it will be important to people entering the fund and less important to people leaving it.
Table Box 1 .
Health insurance system in the Netherlands (situation in 2002)
| Health care system | Exceptional Medical Expenses Act (AWBZ) provides cover for expensive and long‐term health care for all residents (40% of total expenditure). More than 64% of the population are members of a social health insurance fund (37% of total expenditure). Some 31% have taken out private insurance voluntarily and the remaining 5% have medical insurance under a public law scheme (together 19% of total expenditure) |
| Membership of health insurance funds | Obligatory for employees under the income ceiling of 31 750€ (2003) and their families, some groups of social security dependents, old age pensioners (income ceiling 20 200€), self‐employed (income ceiling 20 250€). |
| Benefits of health insurance funds | Medical care, pharmaceutical prescriptions, hospital care, dental care <18 years. Supplementary, voluntary insurance possible (e.g. for dental care >18 years and parts of physiotherapy). Supplementary benefits differ between funds |
| Premium | Income‐dependent part is uniform, paid by employees and employers to the central fund, distributed with risk‐adjustment to individual funds. Flatrate part of premium determined by the individual funds. Difference between cheapest and most expensive is 124€ per year (2002) Supplementary insurance premium determined by individual funds |
| Number | Twenty‐four different health insurance organizations |
| Health insurance fund under investigation | About 2 million members, mainly in the south of the Netherlands; flat‐rate premium 15€ per month Five different supplementary insurance schemes (3–20€ per month) |
Although the basic benefit package in Dutch social health insurance is uniform, a small part of the benefits (2.2% in terms of costs) can vary (see Box 1). About 95% of the compulsory insured have bought supplementary insurance. 32 Because of the differences among health insurers as regards the supplementary benefit packages, we would expect the supplementary benefits and the supplementary premium to play a role in the considerations of switchers. Freedom of provider choice is a minor issue in the Netherlands. Although general practitioners are gatekeepers to hospital care, patients have freedom of choice among hospitals and doctors.
We have defined two steps in the process of moving to a new insurer. First, it can be assumed that the insured have been dissatisfied with their former insurer. Otherwise it would not make much sense to switch, except in cases where the switch was not initiated by the people themselves (general practitioners, for instance, sometimes encourage patients to switch to insurance funds inside their region because they do not have contracts with all insurance organizations). Secondly, there have to be differences between health insurance organizations, for instance, in the premium charged or in other areas, at least as perceived by those switching. Without such differences there may be reasons to change but nowhere to go to. So people who are dissatisfied with their insurer and also perceive differences between organizations are most likely to switch.
Usually, satisfaction scores can only be interpreted in a relative sense. A satisfaction score for those who switch between organizations can best be compared with satisfaction scores of people who do not switch. So to construct a comparative base for the satisfaction scores, a group of non‐switchers is included in the study. The same procedure is applied to construct a comparative base for the perceived differences between insurance organizations.
The following research questions have been formulated:
-
1
Are people who switch from one insurer to another less satisfied than people who do not switch?
-
2
Do people who switch from one insurer to another perceive more differences between insurance funds than people who do not switch?
-
3
For what reasons do people leave or enter specific health insurance funds and what is the role of the flat‐rate premium?
-
4
Do people who are both dissatisfied with their insurer and perceive differences between organizations report the same reasons as other people?
Methods
Survey sample
In the Netherlands, membership of a health insurance fund is obligatory for about 60% of the total population (see Box 1 for details). For the purpose of our research, we collaborated with one of the largest health insurance funds in the Netherlands. The study population was divided into three stratified groups. The first group comprised those who had left the health insurance fund on 1 January 2001, changing to another health insurance fund. (People leaving the fund because of a collective insurance contract with their employer were not considered.) This group was labelled exit. The second group comprised those who entered the health insurance fund on 1 January 2001 from another health insurance fund. (People entering the fund because of a collective insurance contract with their employer were not considered.) This group was labelled entry. The third group comprised those who were members of the insurance fund under investigation for at least 1 year and did not switch on 1 January 2001. This group was called stayers.
In the spring of 2001, 884 people (aged 18 years or above) were randomly selected – about 300 in each stratification – because we wanted to have a power of 0.80 to detect small/medium differences (effect size 0.30) among the groups, assuming that half the sample would respond. After 3 weeks, non‐respondents received a reminder letter and after a further 3 weeks they were sent the questionnaire again. The total response rate was 468 (52.9%). As the response in the stayers group was rather low, these people were also contacted by telephone. Within the three groups, the response rate was as follows: entry 47.8%; exit 63.4%; and stayers 47.2%. Seventy‐nine respondents were excluded because they had to switch as a result of their employer's health insurance contract. Non‐respondents did not differ much from respondents in terms of sex and age. Only two of six age/sex comparisons showed statistically significant differences. Respondents in the entry group were slightly older than non‐respondents (mean age is 44.0 as against 40.6 years) and in the stayers group women responded more often than men.
Postal questionnaire
A postal questionnaire was designed to measure the concepts of the research questions. Satisfaction with health insurance was measured by a set of 10 Likert‐type items. One example is: to what extent do you agree with the following statement: ‘I was satisfied with the flat‐rate premium of my health insurance fund’. The answer categories are described in Table 2. The reliability of the scale in terms of Cronbach's α is 0.86. Both the items and the scale score were analysed. An additional single question was asked to determine satisfaction with the insurance fund in general.
Table 2.
Mean satisfaction scores according to group (for exit and stayers the figures relate to the health insurance fund under investigation, for entry they relate to the other health insurance funds in the Netherlands)
| Exit (n = 113) | Entry (n = 148) | Stayers (n = 128) | P (F‐test) | Post hoc contrasts (Scheffé) | |
|---|---|---|---|---|---|
| Flat‐rate premium | 3.82 | 3.28 | 3.75 | 0.000 | Exit, stayers vs. entry |
| Supplementary insurance premium | 3.65 | 3.22 | 3.62 | 0.000 | Exit, stayers vs. entry |
| Supplementary insurance benefits | 3.24 | 3.07 | 3.52 | 0.002 | Entry vs. stayers |
| Speed with which bills are compensated | 3.70 | 3.61 | 3.65 | 0.783 | – |
| Waiting time for medical specialists | 3.49 | 3.46 | 3.11 | 0.009 | Exit, entry vs. stayers |
| Accessibility by telephone | 3.60 | 3.97 | 3.65 | 0.000 | Exit, stayers vs. entry |
| Opening hours insurance fund | 3.83 | 3.58 | 3.57 | 0.029 | None |
| Client centredness | 3.63 | 3.33 | 3.74 | 0.005 | Entry vs. stayers |
| Information from insurance fund | 3.52 | 3.30 | 3.58 | 0.081 | – |
| Possibility of filing complaints | 3.39 | 3.23 | 3.54 | 0.067 | – |
| Mean satisfaction scale score of 10 items | 3.57 | 3.26 | 3.60 | 0.000 | Exit, stayers vs. entry |
| Satisfaction in general | 3.44 | 3.24 | 3.65 | 0.002 | Entry vs. stayers |
Scale of items: 1, completely disagree; 2, disagree; 3, agree/disagree; 4, agree; 5, completely agree.
The respondents’ perception of differences between health insurance funds was also measured with a set of 11 Likert‐type items. We would like to stress that these differences are the perception of the respondents and do not necessarily reflect real differences. An example of one item is: ‘What do you think of the differences between health funds as regards the benefits of the supplementary insurance?’. Answer categories are described in Table 3. The reliability of the scale in terms of Cronbach's α is 0.92. Both the items and the scale score were analysed.
Table 3.
Mean perceived difference scores according to group
| Exit (n = 113) | Entry (n = 148) | Stayers (n = 128) | P (F‐test) | Post hoccontrasts (Scheffé) | |
|---|---|---|---|---|---|
| Basic health insurance benefits | 2.77 | 2.92 | 2.91 | 0.447 | – |
| Flat‐rate premium | 3.00 | 3.17 | 2.97 | 0.218 | – |
| Supplementary insurance benefits | 3.38 | 3.57 | 3.19 | 0.006 | Entry vs. stayers |
| Supplementary insurance premium | 3.25 | 3.50 | 3.21 | 0.030 | None |
| Efforts to reduce waiting lists | 2.95 | 3.00 | 3.02 | 0.890 | – |
| Speed with which bills are compensated | 2.96 | 3.21 | 3.14 | 0.179 | – |
| Medication benefits | 2.78 | 2.65 | 2.89 | 0.242 | – |
| Medical specialist benefits | 2.90 | 2.75 | 2.83 | 0.608 | – |
| Accessibility by telephone | 2.87 | 3.37 | 2.92 | 0.002 | Exit, stayers vs. entry |
| Client centredness | 2.95 | 3.21 | 2.90 | 0.105 | – |
| Handling of complaints | 3.03 | 3.35 | 3.23 | 0.142 | – |
| Mean score of 11 items | 2.97 | 2.78 | 2.87 | 0.076 | – |
Scale of items: 1, no differences; 2, very small differences; 3, small differences; 4, large differences; 5, very large differences, between health insurance organizations.
On the basis of the median split of the satisfaction and the perceived difference variables, the total sample was divided into two groups. Subsequently the respondents who were both dissatisfied and perceived the biggest differences between health insurance funds were compared with those who scored in the upper half of the satisfaction or in the lower half of the perceived difference variable. This division was made in order to study the process of choice. In addition to the satisfaction and the perceived difference variables we investigated the reason for switching insurance funds. Box 2 indicates the domain the variables refer to in terms of the health insurance fund under investigation or the other health insurance funds in the Netherlands.
Table Box 2 .
Domain to which variables relate by group
| Variables | Exit (n = 113) | Entry (n = 168) | Stayers (n = 128) |
|---|---|---|---|
| Satisfaction | Health insurance fund under investigation | Other health insurance funds | Health insurance fund under investigation |
| Perceived differences | Health insurance funds in general | Health insurance funds in general | Health insurance funds in general |
| Reasons for leaving | Health insurance fund under investigation | Other health insurance funds | None |
| Reasons for entering | Other health insurance funds | Health insurance fund under investigation | None |
For the exit group, satisfaction with their former health insurance fund and reasons why they have left the former fund relates to the health insurance fund under investigation. For the entry group these variables relate to the health insurance fund they came from, which could be any other health insurance fund in the Netherlands. For the stayers, group satisfaction relates to their current fund, which is, of course, the health insurance fund under investigation. Perceived differences is a variable that relates to health insurance funds in general.
Statistical analyses
Comparisons among the three groups were made using F‐tests for interval variables and chi‐squared tests for nominal variables. When the F‐test was statistically significant, post hoc contrasts (Scheffé) were calculated to compare group means.
Results
The result section will address the satisfaction of people who switched from one insurer to another and the extent of perceived differences between insurance organizations consecutively, followed by the reported reasons for switching and the role of the flat‐rate premium. First, some characteristics are given to compare the three samples.
Table 1 shows five respondents’ characteristics according to groups. People who entered the insurance fund were, on average, older than those who left and those who stayed. There are also relatively more women in the entry group. The three groups do not significantly differ statistically as regards health and education. Finally, the percentage of the employed in the group who left the insurance fund is statistically, significantly higher than the percentages in the other groups.
Table 1.
Respondents’ characteristics according to group
| Exit (n = 113) | Entry (n = 148) | Stayers (n = 128) | P* | |
|---|---|---|---|---|
| Age in years (mean) | 34.8 | 44.0 | 34.5 | 0.000 |
| Men (%) | 48.4 | 32.0 | 49.2 | 0.015 |
| Excellent or good subjective health (%) | 47.3 | 44.9 | 37.1 | 0.243 |
| Highest education on scale 1–10 (mean) | 5.12 | 5.85 | 5.30 | 0.053 |
| Employed (%) | 85.0 | 78.8 | 66.4 | 0.003 |
*F‐test for age and highest education, chi‐square test otherwise.
Table 2 shows the mean scores of the patient satisfaction items. The entry group was less satisfied than the other groups in respect of the flat‐rate premium. The same applies with the supplementary insurance premium. Three items do not show statistically significant differences among the three groups (speed with which bills are paid, office opening hours of insurance fund and information from insurance fund). The total sum of the item scores also showed that those in the entry group were statistically significantly less satisfied than the others, the same was true for satisfaction in general (3.24 vs. 3.44 and 3.65 for the entry, exit and stayers, respectively).
Table 3 shows the perceived differences between health insurance organizations in general. Most people perceive only small differences. All the mean scores are somewhere in the region of scale point 3. With the exception of three items, there is no distinction among the groups. The entry group perceived larger differences as regards telephone accessibility and the supplementary insurance benefits and premium. The total perceived differences score does not differ statistically significantly across the three groups.
In order to study the reasons for leaving one insurance fund and entering another, the group of stayers who did not switch were excluded.
Table 4 shows the reported reasons for leaving a health insurance fund. The flat‐rate premium level is reported by 7.3% of the exit group and by 19.6% of the entry group as their reason for leaving. The difference is statistically significant. The most frequently reported reason by both the exit and the entry groups is the restriction of the supplementary benefits (32.7 and 33.8%, respectively). Both groups also gave various other reasons for leaving their insurance fund.
Table 4.
Reasons to leave (for exit the figures relate to the health insurance fund under investigation, for entry they relate to the other health insurance funds in the Netherlands)
| Exit, % (n = 113) | Entry, % (n = 168) | P (chi‐squared test) | |
|---|---|---|---|
| Supplementary insurance benefits too restricted | 32.7 | 33.8 | 0.859 |
| Via partner | 15.9 | 6.8 | 0.018 |
| Via employer | 13.6 | 16.2 | 0.567 |
| Dissatisfied with service | 13.6 | 19.6 | 0.208 |
| Change of employment | 11.8 | 14.9 | 0.480 |
| Unclear information | 10.9 | 12.8 | 0.637 |
| Flat‐rate premium too high | 7.3 | 19.6 | 0.005 |
| Supplementary insurance premium too high | 6.4 | 12.8 | 0.088 |
| Moved to another region | 6.4 | 0.7 | 0.009 |
| Compensation of bills too slow | 5.5 | 6.8 | 0.668 |
| Family doctors’ advice | 3.6 | 0.0 | 0.019 |
| Bad reputation | 1.8 | 1.4 | 0.764 |
| Do not recall | 1.8 | 0.7 | 0.397 |
| Otherwise | 20.9 | 23.6 | 0.067 |
Table 5 shows the reported reasons for entering a health insurance fund. To some extent the reasons for entering a particular fund mirrors the reasons for leaving. The flat‐rate premium level is reported by 13.2% of the exit group and by 24.7% of the entry group as the reason for entering. As the supplementary insurance premium does not differentiate the exit group from the entry group in respect both of reasons for leaving and reasons for entering, we have concluded that these groups consider the supplementary insurance premium to be less important.
Table 5.
Reasons to enter (for exit the figures relate to the other health insurance funds in the Netherlands, for entry they relate to the health insurance fund under investigation)
| Exit, % (n = 113) | Entry, % (n = 168) | P (chi‐squared test) | |
|---|---|---|---|
| Supplementary insurance benefits | 44.3 | 39.5 | 0.437 |
| Family and friends | 24.5 | 13.0 | 0.018 |
| Good service | 19.8 | 12.2 | 0.100 |
| Clear information | 17.9 | 8.2 | 0.019 |
| Most well‐known | 17.0 | 4.8 | 0.001 |
| Supplementary insurance premium | 16.0 | 22.4 | 0.206 |
| Flat‐rate premium | 13.2 | 24.7 | 0.024 |
| Via employer | 12.3 | 27.2 | 0.004 |
| Good reputation | 2.8 | 12.9 | 0.095 |
| Otherwise | 16.8 | 19.9 | 0.539 |
The most frequently reported reason for choosing to enter a health insurance fund are the supplementary insurance benefits. The entry group does not significantly differ statistically from the exit group. This indicates that for some people the benefits of other insurance organizations are attractive, while for others the benefits of the insurance fund under investigation are attractive. Table 5 shows five discriminating reasons. In the exit group, the clarity of information, the influence of family and friends and familiarity with the insurance fund are reported more frequently than in the entry group. The influence of the employer and the flat‐rate premium are shown to be more important in the entry group.
The last research question relates to the group of people who are dissatisfied and also perceive differences between insurance funds. This group comprises 25.9% of the total sample. Compared with the other respondents, this group reported more reasons for dissatisfaction. Furthermore, they appear to be more influenced by the flat‐rate premium, the supplementary insurance premium and the supplementary insurance benefits (data not in table).
Discussion
In this article, we examined the reasons why people switch between insurance organizations. Three groups were compared: people who left the health insurance fund for another one (exit), people who entered the health insurance fund coming from another one (entry) and people who were members of the insurance fund under investigation for at least 1 year and did not switch (stayers). Two steps were defined in the process of choosing a new insurance fund. First, there had to be some dissatisfaction with the insurance fund they were currently registered with, and secondly, there had to be differences between health insurance organizations, for instance in the premium charged or in other areas, at least as perceived by the people who were considering changing their health insurer. If they were not dissatisfied, there was no reason to change their insurance and without differences, there may be reasons to change but nowhere to change to.
The results show that people who entered the insurance fund under investigation were less satisfied with their former insurance fund than the other groups. They were also less satisfied about the flat‐rate premium. It was shown that people are only aware of small differences among the health insurance funds and the three groups only deviated a little from each other. About a quarter of the entry group reported the flat‐rate premium as a reason for choosing a health insurance fund. However, the most frequently reported reason, for both exit and entry, was shown to be supplementary insurance benefits.
Process of choice
Those in the entry group were less satisfied than the other groups with their former health insurance fund. So dissatisfaction seemed to motivate their decision to switch. However, contrary to our expectations, those in the exit group were no less satisfied with the insurance fund under investigation when compared with the people who did not change. Most people perceived only small differences between insurance organizations. In fact, the group of switchers perceived equally little difference. The reason for this is not clear. In conclusion, the process of choice is only partly validated. Some further support for the process of choice is found in the reasons for entry reported by the group of persons who are both dissatisfied and perceive relatively more differences between health insurance funds. Compared with the others this group is more sensitive to the more rational reasons: the flat‐rate premium, the supplementary insurance premium and the supplementary insurance benefits. When categorizing the set of reasons, we can distinguish three broad groups: cognitive reasons, affective reasons and reasons of convenience. Reflections about premiums and benefits are in the cognitive domain, while considerations about reputation, acquaintance and the influence of family and friends are in the affective domain. Moving to another region or change of employment is a matter of convenience. Kerssens et al. 11 report, on the basis of a random sample of Dutch households, that the most frequently mentioned reasons for joining a particular health insurance fund are: starting or changing employment, joining in early adulthood and familiarity. The majority of consumers still choose traditional regional funds. 33 So reasons in domains of affection and convenience seem to be more important than reasons in the cognitive domain.
Strong/weak points of the health insurance fund under investigation
Five reasons distinguish the exit, from the entry, group (Table 5). In the exit group, the influence of family and friends, the clarity of information and familiarity with the newly chosen insurance fund are reported more frequently than in the entry group. Improving the clarity of information seems to be an important marketing objective for the insurance fund under investigation. The fourth discriminating reason relates to change of employment, although people with collective contract health insurance were excluded from the analyses. According to the national health insurance information centre, only 5% of the insured have a collective contract. 34 Whenever people start a new job, they have to complete a number of forms. Usually, an enrolment form for health insurance is one of them, even when completion is not obligatory. That is why change of employment was reported as a reason for switching health insurance.
The low flat‐rate premium of the insurance fund under investigation compared with others seems to be another important marketing instrument because dissatisfaction with the flat‐rate premium is reported as a reason for being in the entry group (fifth discriminating reason). The supplementary insurance premium does not differentiate the exit from the entry group in terms of reasons for leaving or entering, so the supplementary insurance premium seems to be less important.
In both the exit and the entry groups, the most frequently reported reasons for entering a health insurance fund are the supplementary insurance benefits. The exit group does not significantly differ statistically from the entry group. This means that for some people the benefits of other insurance organizations are attractive, while for others the benefits of the insurance fund under investigation are attractive.
When looking at the respondents’ characteristics there were differences in the three groups. Those in the exit group were younger and in better health. A larger proportion were employed compared with the entry group. This was partly due to the selection of the entry sample, because it is difficult to implement a random selection in the insurance fund computers. The mean age of the total entry population was 42 (instead of 46). But, the health insurance fund under investigation seemed to lose the ‘better risks’. The response rate at 52.9% is an obvious methodological weakness of the study. How well can the results be generalized? Kerssens et al. did a similar study in a sample of the Dutch population with a much higher response rate (76.9%), 12 in which their main results were comparable with the current study. For example, the reasons given for leaving in the switching group were: supplementary benefits too restricted (30.6%), flat‐rate premium too high (18.7%) and dissatisfaction with service (25.4%). 12 These figures are the same as those of the entry group shown in Table 4, which seems to indicate that the respondents are a representative sample of the switching population. (As the reasons for leaving in the switching group in the population sample relate to all Dutch insurance funds, while those of the entry group refer to all Dutch insurance funds, except the insurance fund under investigation, the entry group is the most appropriate group for comparison.)
Managed competition
Introducing freedom of choice does not automatically lead to switching. The reasons for considering a change are at least some dissatisfaction with the quality of one's current insurance fund and sufficient obvious differences between health insurance funds. Dutch men do not often perceive differences and consequently, the number of switchers is actually quite small. However, in the absence of clear differences between insurance funds, the benefits of freedom of choice for consumers – having the opportunity to choose a better health insurance fund – and for the insurance organizations – switching as a signal of low or decreasing quality – are difficult to achieve. Accordingly, although one of the basic vehicles of consumer influence – the exit option – has been introduced, 35 , 36 other conditions of consumer influence are still underdeveloped.
Increasing differences, especially in the main instrument for competition, the flat‐rate premium, will probably lead to more switching. In Germany many people have changed their insurance within only a few years and many health insurance funds have won or lost a significant number of insured because premium differences are very large. 12 The cheapest and the most expensive Dutch fund average about 44€ per year, while German differences are almost 10‐fold. 12 An effective way of increasing differences is to enlarge the financial risk for Dutch social health insurance funds (today, Dutch funds are responsible for 38% of their expenses, while in Germany the figure is 100%). But how large do the differences need to be for effective price competition? What is the price‐elasticity of the flat‐rate premium? Schut 37 calculated a zero price‐elasticity for the flat‐rate premium. A zero price‐elasticity obviously means that increasing differences will not lead to more effective price competition. And increasing financial risks too much could lead to adverse selection instead of price competition. 37
Perhaps a better way of encouraging the tendency to switch is to reduce search costs. Although the Dutch social insurance system seems to be quite clear – with standardized benefits and price difference only – the varying supplementary insurance benefits obstruct our vision. Furthermore, in the Netherlands, managed competition has been introduced in an individual insurance market, in contrast to the United States where large employers facilitate plan choice by offering a limited number of health plans to choose from. 9 Dutch consumers have an unlimited choice of numerous supplementary insurance options, as there are more than 20 funds, each offering several. The most frequently reported reason in our study, for both exit and entry, concerns the supplementary insurance benefits. Furthermore, Schut 37 calculated a statistically significant price‐elasticity for the supplementary insurance premium. So the main instrument for competition seems to be the supplementary insurance, rather than the flat‐rate premium. An active government policy aimed at structuring supplementary insurance would certainly facilitate consumer choice of social health insurance in managed competition.
Acknowledgement
This study was supported by VGZ‐zorgverzekeraar.
References
- 1. Altenstetter C, Bjorkman JW. Health Policy Reform, National Variations and Globalization. London: MacMillan, 1997. [B4257] [Google Scholar]
- 2. Light DW. Comparative institutional response to economic policy managed competition and governmentality. Science and Medicine, 2001; 52: 1151–1166. [DOI] [PubMed] [Google Scholar]
- 3. Gross R, Rosen B, Shirom A. Reforming the Israeli health system: findings of a 3‐year evaluation. Health Policy, 2001; 56: 1–20. [DOI] [PubMed] [Google Scholar]
- 4. Schut FT, van Doorslaer EKA. Towards a reinforced agency role of health insurers in Belgium and the Netherlands. Health Policy, 1999; 48: 47–67. [DOI] [PubMed] [Google Scholar]
- 5. Enthoven AC. Theory and Practice of Managed Competition in Health Care Finance. Amsterdam: North‐Holland, 1988. [B1955] [Google Scholar]
- 6. Enthoven AC, Kronick R. A consumer choice health plan for the 1990s: universal health insurance in a system designed to promote quality and economy. New England Journal of Medicine, 1989; 320: 29–37 (first part), 94–101 (second part). [DOI] [PubMed] [Google Scholar]
- 7. Enthoven AC. The history and principles of managed competition. Health Affairs, 1993; 12(Suppl.): 24–48. [DOI] [PubMed] [Google Scholar]
- 8. Schut FT. Health care systems in transition: The Netherlands. Part I: Health care reforms in The Netherlands: miracle or mirage? Journal of Public Health Medicine, 1996; 18: 278–284. [DOI] [PubMed] [Google Scholar]
- 9. Schut FT, Gress S, Wasem J. Consumer price sensitivity and social health insurer choice in Germany and The Netherlands. International Journal of Health Care Finance and Economics, 2003; 3: 117–138. [DOI] [PubMed] [Google Scholar]
- 10. Lieverdink H. The marginal success of regulated competition policy in The Netherlands. Social Science and Medicine, 2001; 52: 1183–1194. [DOI] [PubMed] [Google Scholar]
- 11. Kerssens JJ, Delnoij DMJ, Verweij JA, Van Der Schee E. De keuze van ziekenfondsverzekerden voor een zorgverzekeraar (Consumer preferences for particular social health insurers). Tijdschrift voor Sociale Gezondheidszorg, 2002; 80: 35–42. [Google Scholar]
- 12. Gress S, Groenewegen P, Kerssens J, Braun B, Wasem J. Free choice of sickness funds in regulated competition: evidence from Germany and The Netherlands. Health Policy, 2002; 60: 235–254. [DOI] [PubMed] [Google Scholar]
- 13. Colombo F. Labour Market and Social Policy-Occasional Papers No. 53. Paris: OECD, 2001. [Google Scholar]
- 14. McGuire TG. Price and membership in a prepaid group medical practice. Medical Care, 1981; 19: 172–183. [DOI] [PubMed] [Google Scholar]
- 15. Welch WP. The elasticity of demand for health maintenance organizations. Journal of Human Resources, 1986; 21: 252–266. [Google Scholar]
- 16. Long SH, Settle RF, Wrightson CW. Employee premiums, availability of alternative plans, and HMO disenrollment. Medical Care, 1988: 26: 927–938. [DOI] [PubMed] [Google Scholar]
- 17. Short PF, Taylor AK. Premiums, benefits, and employee choice of health insurance options. Journal of Health Economics, 1989; 8: 293–311. [DOI] [PubMed] [Google Scholar]
- 18. Dowd B, Feldman R. Premium elasticities of health plan choice. Inquiry, 1994; 31: 438–444. [PubMed] [Google Scholar]
- 19. Barringer MW, Mitchell OS. Workers’ preferences among company‐provided health insurance plans. Industrial and Labor Relations Review, 1994; 48: 141–152. [Google Scholar]
- 20. Scanlon DP, Chernew M, Lave JR. Consumer health plan choice: current knowledge and future directions. Annual Review of Public Health, 1997; 18: 507–528. [DOI] [PubMed] [Google Scholar]
- 21. Cutler DM, Reber SJ. Paying for health insurance: the trade‐off between competition and adverse selection. Quarterly Journal of Economics, 1998; 113: 433–466. [Google Scholar]
- 22. Royalty AB, Solomon N. Health plan choice. Price elasticities in a managed competition setting. Journal of Human Resources, 1999; 34: 1–41. [Google Scholar]
- 23. Buchmueller TC, Feldstein PJ. Consumer's sensitivity to health plan premiums: evidence from a natural experiment in California. Health Affairs, 1996; 15: 143–151. [DOI] [PubMed] [Google Scholar]
- 24. Cunningham PJ, Kohn L. Health plan switching: choice or circumstance? Health Affairs, 2000; 19: 158–164. [DOI] [PubMed] [Google Scholar]
- 25. Feldman R, Finch M, Dowd B, Cassou S. The demand for employment‐based health insurance plans. Journal of Human Resources, 1989; 24: 115–142. [Google Scholar]
- 26. Smith HL, Rogers RD. Factors influencing consumers’ selection of health insurance carriers. Journal of Health Care Marketing, 1986; 6: 6–14. [PubMed] [Google Scholar]
- 27. Gibbs DA, Sangl JA, Burrus B. Consumer perspectives on information needs for health plan choice. Health Care Financial Review, 1996; 18: 55–73. [PMC free article] [PubMed] [Google Scholar]
- 28. Robinson S, Brodie M. Understanding the quality challenge for health consumers: the Kaiser/AHCPR Survey. Journal of Quality Improvement, 1997; 23: 239–244. [DOI] [PubMed] [Google Scholar]
- 29. Hibbard JH, Jewett JJ. Will quality report cards help consumers? Health Affairs, 1997; 16: 218–228. [DOI] [PubMed] [Google Scholar]
- 30. Hershey JC, Kunreuther H, Schwartz JS, Williams SV. Health insurance under competition: would people choose what is expected? Inquiry, 1984; 21: 349–360. [PubMed] [Google Scholar]
- 31. Chakraborty G, Ettenson R, Gaeth G. How consumers choose health insurance. Journal of Health Care Marketing, 1994; 14: 21–33. [PubMed] [Google Scholar]
- 32. ZFR/CTU . Onderzoek effecten aanvullende verzekering (Investigation of the Effects of Supplementary Insurance). Amstelveen: ZFR/CTU, 1999. [Google Scholar]
- 33. van den Brink RJ. Een efficiente ziekenfondsmarkt? (An efficient social health insurance market?). Economische Statistische Berichten, 2001; 86: 220–223. [Google Scholar]
- 34. Vektis . Financiering van de zorg in 1998 (Financing care in 1998). Zeist: Vektis; bv (The national information center of health insurers), 1999. [NE 156 (99)] [Google Scholar]
- 35. Hirschman AO. Exit, Voice, and Loyalty: Response to Decline in Firms, Organizations and States. Cambridge, MA: Harvard University Press, 1970. [PeterG] [Google Scholar]
- 36. Chinitz D. Regulated competition and citizen participation: lessons from Israel. Health Expectations, 2000; 3: 90–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Schut FT. Prijsconcurrentie ziekenfondsen nog niet effectief (Price competition between health insurers is not yet effective). Economisch Statistische Berichten, 2001; 86: 172–175. [Google Scholar]