Family Practice Vol. 17, No. 3, 222-229
© Oxford University Press 2000
Who needs a gatekeeper? Patients' views of the role of the primary care physician
a Health Policy Research Unit, JDC-Brookdale Institute and
b Faculty of Life Sciences, Ben-Gurion University; Department of Family Medicine, Haemek Hospital; Kupat Holim Clalit, Israel.
Revital Gross, JDC-Brookdale Institute, POB 13087, Jerusalem, Israel 91130.
Gross R, Tabenkin H and Brammli-Greenberg S. Who needs a gatekeeper? Patients' views of the role of the primary care physician. Family Practice 2000; 17: 222–229.
Received 26 July 1999; Accepted 21 December 1999.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. The primary care physician serving as a ‘gatekeeper’ can make judicious decisions about the appropriate use of medical services, and thereby contribute to containing costs while improving the quality of care. However, in Israel, sick funds competing for members have not adopted this model for fear of endangering their competitive stance. The purpose of this study was to examine, for the first time, the stated preferences and actual behaviour of a national sample of members of the four Israeli sick funds regarding self-referral to specialists, and to identify the characteristics of patients who prefer the gatekeeper model.
Methods. Data were derived from a national telephone survey carried out in 1997. A random representative sample of 1084 of all adult sick fund members were interviewed, with a response rate of 81%. Bivariate analysis was conducted using over all chi-square tests, and multivariate analysis was performed using logistic regression models.
Results. A third of all respondents prefer self-referral to a specialist, 40% prefer their family physician to act as gatekeeper and 19% prefer the physician to co-ordinate care but to refer themselves to a specialist. Independent variables predicting preference for the gatekeeper model are: living in the periphery, sick fund membership, low level of education, being male, fair or poor health status, having a permanent family physician and being satisfied with the professional level of the family physician. A significant correlation was found between practising self-referral and preference for self-referral.
Conclusions. The findings indicate the importance of surveying patients' attitudes as an input in policy formulation. The study identified specific population groups which prefer the gatekeeper model, and explored the advantages of a flexible model of gatekeeping.
Keywords. Gatekeeping, patient surveys, primary care.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The task of a primary care physician who serves as a ‘gatekeeper’ is to manage and co-ordinate a patient's care, as well as to be the sole referring agent to specialists. On the one hand, the gatekeeper is the patient's ‘health adviser’. On the other hand, he can make judicious decisions about the best and most appropriate use of medical services, and thereby contribute to containing costs while improving the quality of care.1–3 In countries where primary care physicians co-ordinate care and control access to specialists and the utilization of associated services, the cost of health services and their share of the national economy are lower compared with developed countries who offer direct access to medical services.4–6
In Israel, as in other countries, health care organizations are considering implementation of the gatekeeper model in light of these advantages. However, despite its potential benefits, there are obstacles to the actual implementation of the gatekeeper model. In a survey of members of one Israeli sick fund, resistance of service recipients who prefer direct access to special-ists was identified. Furthermore, 33% said that they would leave the sick fund if self-referral was prohibited.7 Therefore, sick funds competing for members have not yet adopted service models which limit access to specialists for fear of endangering their competitive stance.
The purpose of this study was to examine, for the first time, the stated preferences and actual behaviour of a national sample of adult members of all four Israeli sick funds regarding self-referral to specialists, and to identify the characteristics of patients who prefer the gatekeeper model.
Knowing the preferences of patients will assist health care organizations to develop a gatekeeper model that will fit the preferences of different patient populations. It will enable the organizations to offer service models that contribute to cost containment, while meeting the expectations of the public. By identifying population groups that oppose this model, it will be possible to design educational and guidance programmes to gain their co-operation.
Background—the Israeli health care system
Primary health care in Israel is provided through four non-profit-making sick funds that deliver services to their members based on a model similar to that of health maintenance organizations in the USA. Services are delivered at the sick funds' own facilities or through contracted providers. Kupat Holim Clalit (KHC) insures ~60% of the population, Maccabi insures ~20%, and Meuchedet and Leumit each insure 10%.8 The Ministry of Health (MOH) is responsible for: planning, regulation and co-ordination of the health system; for the general assessment and supervision of sick fund operation; and for implementation of legislation concerning health care and its provision. The MOH is also the major provider of in-patient services, public health services and community-based psychiatric care. KHC owns and operates eight general hospitals, or ~17% of the country's general hospital beds.
The National Health Insurance Law which came into effect in January 1995 mandates compulsory health insurance for all residents in a sick fund of their choice. In all sick funds, members are entitled to choose a primary care physician and change physicians if not satisfied. The sick funds differ regarding their policy for access to specialists. Since 1993, KHC has changed its gatekeeping policy and allows direct access to dermatologists, ophthalmologists, otolaryngologists and orthopaedists. Other specialists still require referrals by the family physician. The other three sick funds have always had a policy of self-referral to almost all specialists. None of the sick funds require an extra payment if patients go directly to a specialist, and the appointment procedure is the same as for patients who were referred by their family physician. Family physicians who take responsibility for referring patients to specialists do not receive extra reimbursement for performing this role.
In 1997, the total expenditure on health was ~8.4% of the Gross National Product. Life expectancy in 1995 was 75.5 years for men and 79.5 years for women. On average, in 1996, there were 2.4 visits per person to a family practitioner in 6 months, and 0.9 to a specialist.9
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Between August and October 1997, the JDC-Brookdale Institute conducted a telephone survey which included questions concerning respondents' attitudes toward the role of the family physician and their behaviour when needing specialist care (i.e. self-referral or through the family physician).
The study population comprised all permanent residents of Israel over the age of 22 years. A random representative sample of 1600 telephone numbers was drawn using the national computerized telephone listing of Bezek, the Israel Telecommunications Corporation. For each listing sampled that met the criteria for the study population, one randomly chosen adult household member was interviewed. Repeated efforts were made over a 2 month period to contact the individual sampled. Furthermore, the questionnaire was translated into Arabic and Russian for respondents who did not speak Hebrew. The sample was weighted according to the probability of each respondent, i.e. family size.
Of the 1600 telephone numbers sampled, 16.4% did not fit the criteria for the study population (i.e. they belonged to businesses, institutions, fax-modems, foreign workers or people under the age of 22 years). In all, 1084 questionnaires were completed, for a response rate of 81% and a refusal rate of 10.6%. (Seventy-six per cent of the questionnaires were completed in Hebrew, 10% in Arabic and 14% in Russian.) In 8.3% of the cases, no contact was made with the individual sampled (either because there was no answer at any hour of the day or night, or because the individual sampled was not at home), or else contact was made but the interview was not conducted (due to language problems, medical or emotional problems, etc.).
The socio-demographic characteristics of the study population are presented in Appendix 1.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Patients' attitudes toward the role of the family physician as gatekeeper
In order to learn what patients thought about the role of the family physician as gatekeeper, they were asked: "Would you like your family physician to become your personal physician, to co-ordinate all your care, and to be exclusively responsible for referring you to a specialist?" The respondents were then asked to choose one of the following categories: (i) "I would like to refer myself to a specialist"; (ii) "I would like the family physician to co-ordinate care and be exclusively responsible for referring me to a specialist"; (iii) "I would like the family physician to co-ordinate care but to refer myself to a specialist".
Table 1
shows the distribution of responses by various background variables. As can be seen, a slightly higher percentage of respondents (39.5%) prefer that their family physician co-ordinate their care and refer them to a specialist, compared with those who preferred self-referral (34%). Furthermore, an additional 19% prefer self-referral but would still like the physician to co-ordinate their care.
|
The proportion of those preferring that their family physician co-ordinate their care was higher among members of KHC (46%) than among members of Maccabi (29%) and Meuchedet (27.3%) sick funds. More women preferred self-referral to a specialist than men, and more respondents aged 44 years and under preferred self-referral than did those aged 65 years and over. More respondents with 12 or more years of schooling preferred self-referral than did those with up to 8 years of schooling. Higher proportions of preference for self-referral were found among respondents with high income or who lived in Jerusalem, Tel Aviv or central Israel, than among residents of the periphery and of Arab towns and villages. The proportion of those preferring self-referral was higher among those who did not report severe medical problems and those who were employed during the previous 3 months.
High percentages of preference for self-referral were also found among those who did not have a permanent family physician, and among those who were dissatisfied with the professional level or attitude of their family physician. No differences in preference regarding referral were found in terms of general satisfaction with the sick fund.
A multivariate logistic regression analysis was conducted to determine which variables have an independent influence on the preference for referral to a specialist through a family physician, versus self-referral. Table 2 shows the outcome of this analysis, which revealed that the predictor variables that have an independent effect on the preference for referral to a specialist through the family physician are: living outside of Tel Aviv and central Israel, being a member of KHC, having <12 years of schooling, being male, having self-reported fair or poor health, having a permanent family physician, and being satisfied with the professional level of the family physician. (Income does not have a significant influence on the preference for referral to a specialist through a family physician. The variables ‘income’ and ‘education’ were not used in the same model because of their interdependence.)
|
Patients' actual behaviour
Ninety per cent of the respondents reported having a permanent family physician, and ~40% reported that they had visited a specialist in the last 3 months. All respondents were asked about their actual behaviour regarding referral to a specialist: "When you want to see a specialist, do you turn to him directly or do you turn first to your family physician"? (These two questions did not appear on the questionnaire consecutively in order to prevent any response bias.) About half (46.6%) reported that they go directly to a specialist when necessary, 37.1% reported that they first turn to their family physician, and 11.6% reported that it depends on the type of problem (Table 3
).
|
The rate of self-referral was high among the members of Meuchedet and Maccabi (67% each) and the members of Leumit (62.5%), compared with the members of KHC (32.8%). The proportion of women who turn directly to a specialist was higher than that of men. In addition, more of those aged 25–44 years turned directly to a specialist than those aged 65 years and over, as did more of those with 12 or more years of schooling relative to those with up to 8 years of schooling, those with high income relative to those with lower income, those who did not report severe medical problems, those who had seen a specialist during the last 3 months and those who were employed during the last 3 months. The proportion of those turning directly to a specialist was also higher among residents of Jerusalem, Tel Aviv, central Israel and Arab towns and villages than among residents of Haifa, and the north and south of Israel.
Those who did not have a permanent family physician were more apt to go directly to a specialist than were those with a permanent family physician. No differences were found in the likelihood of going to a family physician in terms of satisfaction with the professional level or attitude of the family physician, or overall satisfaction with the sick fund.
A significant correlation was found between practising self-referral and preference for self-referral, such that 65.9% of the respondents who reported preferring self-referral to a specialist did indeed go directly to a specialist when necessary. In addition, ~61% of those who reported preferring that their family physician co-ordinate their care indeed turned to their family physician for a referral to a specialist. Of those who reported preferring that their family physician co-ordinate their care and refer them to specialists, 26.7% actually went directly to a specialist when necessary, and 18.3% of those who reported preferring self-referral actually turned first to their family physician. Finally, 30.4% of those who were unsure whether they preferred self-referral or physician referral to a specialist actually went to their family physician whenever they needed a referral to a specialist.
We conducted a multivariate regression to determine which variables had an independent effect on going to a family physician for referral to a specialist. As seen in Table 4, the independent variables predicting which sick fund members would request referral to a specialist from their family physician are: residence outside of Tel Aviv and central Israel, being a member of KHC, having fewer than 12 years of schooling, being male, not working and having a permanent family physician. (Income has significant influence on actual referral to a specialist through a family physician, such that having a low income positively influences referral through a family physician. The variables ‘income’ and ‘education’ were not used in the same model because of their interdependence.)
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The findings of this study indicate that there are differences among the members of KHC compared with the other sick funds regarding preferences and actual behaviour when seeking specialist care. In the three sick funds which traditionally have had a policy of self-referral to almost all specialists, a high percentage of members prefer that option, but an even higher percentage employ self-referral when needing specialist care. On the other hand, in KHC, which in the past employed a policy of gatekeeping and even today restricts self-referral to a certain extent, a larger proportion of members prefer visiting their primary care physician first. These findings can be a result of self-selection in which members choose sick funds which fit their views on self-referral. On the other hand, it is known that offering a clinical service to the public creates a demand for this service,10,11 and such may be the case regarding self-referral.
There are also differences in preferences among population groups. In each of the sick funds there are a considerable proportion of patients who prefer the family physician to co-ordinate care and be the sole referring agent. Characteristics correlated with this preference are: having a lower socio-economic status, being elderly, living in the periphery, being male and having fair or poor health status. These findings indicate that primary care providers can design differential models of care for different populations, in which members will participate voluntarily. This means that the sick funds should explicitly offer all options and encourage their members to choose the option which fits their preferences: self-referral, gatekeeping or co-ordinated care with self-referral.
A gatekeeping model which is flexible may have the advantage of responsiveness to a patient's needs as well as improving co-ordination of care and cost containment. This approach is in line with core values in primary care stressing the key role of the physician in the healing process, and the importance of responsiveness to individual patient needs and of trust in the relationship between physician and patient.12 This approach is also in line with recent social trends which see medicine as a service and stress increased personal freedom, rights of individuals, free choice and consumerism.13 A flexible model in which gatekeeping is a voluntary option for patients who prefer it has a better chance of being adopted. Furthermore it would not arouse objections as does the practice of managed care organizations limiting access by administrative means.14
In time, the gatekeeper model can be expanded to include additional population groups if primary care providers undertake an active campaign to explain the benefits to members. At the same time, it might be worthwhile providing members with incentives for turning to specialists through their family physician. Possible incentives include giving priority for appointments with a specialist to those who were referred by their physician, or waiving extra payments, where these are required, for members who are referred to a specialist through their family physician.
The study also revealed a correlation between sick fund members' preferences and their satisfaction with their family physician (e.g. the physician's professional level, attitude toward patients). Therefore, sick funds that are considering asking their primary care physicians to be gatekeepers must ensure that each of their members has a permanent family physician to whom they may turn under all circumstances, and that primary care physicians receive appropriate training. This involves not only the physicians' formal knowledge, but also their approach and attitude toward patients, which have been found to have a significant effect on patient preferences. In addition, the sick funds will have to examine how to increase their primary care physicians' motivation to serve as gatekeepers. It can be assumed that implementing gatekeeping voluntarily, only for patients who prefer this model, will be more acceptable to physicians, since a primary cause of physician opposition, i.e. that it might have a detrimental effect on patient relations, will not exist in a voluntary model.15 Furthermore, being a gatekeeper under these circumstances, i.e. without confrontation with the patient, may enrich the family physician's job16 by enhancing their responsibility and authority and providing an opportunity to practise medicine in a comprehensive manner, including having control over the course of treatment.
The findings of this study have implications for primary health care providers in other countries considering implementing a gatekeeper model. The findings indicate the importance of surveying patients' attitudes as an input in policy formulation. The findings also suggest the benefits of developing a flexible gatekeeping model adjusted to the preferences and needs of different population groups. In the Israeli case, we have identified characteristics of people who prefer their family physician to co-ordinate their care and refer them to specialists when needed, while others prefer self-referral. Therefore, implementing a gatekeeper model among members voluntarily may be a recommended policy which is financially beneficial for health care orgnizations, as well as responsive to patients' needs. Future studies are needed in order to assess the generalizability of the findings to other health care systems.
|
|
Acknowledgments |
|---|
This study was funded by the National Institute for Health Policy and the JDC-Brookdale Institute.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Franks P, Clancy CM. Gatekeeping revisited—protecting patients from overtreatment. N Engl J Med 1992; 327: 424–429.[Web of Science][Medline]
2
Starfield B, Powe NR, Weiner JR. Costs vs quality in different types of primary care settings. J Am Med Assoc 1994; 272: 1903–1908.
3 Martin DP, Diehr P, Price KF et al. Effect of gatekeeper plan on health services use and charges: a randomized trial. Am J PH 1989; 79: 1628–1632.
4
Grumbach K, Fry J. Managing primary care in the US and in the UK. N Engl J Med 1993; 328: 940–945.
5 Starfield B. Primary Concept Evaluation and Policy. Oxford: Oxford University Press, 1992.
6 Wachter RH. Rationing health care: preparing for a new era. South Med J 1995; 88: 25–31.[Medline]
7
Tabenkin H, Gross R, Brammli S, Shvartzman P. Patients' views of direct access to specialists. J Am Med Assoc 1998; 279: 1943–1948.
8 Bendlack J. Sick Fund Membership. Survey Series 150. Jerusalem: The National Insurance Institute, 1998.
9 Ministry of Health. Health in Israel. State of Israel, Jerusalem, 1998.
10 Ginsberg E. The restructuring of US health care. Inquiry 1985; 22: 272–281.[Medline]
11 Bishop C, Wallack SS. National health expenditure limits: the case for a global budget process. Milbank Q 1996; 74: 361–376.[Web of Science][Medline]
12 McWhinney IR. Core values in a changing world. Br Med J 1998; 346: 1807–1809.
13 Boerma GW, Fleming DH. The Role of General Practice in Primary Health Care. WHO Publication: Stationery Office Ltd, 1998.
14 Davies K. Managed Care: The Patients Perspective. New York: Commonwealth Fund, 1995.
15 Taylor T. Pity the poor gatekeeper: a transatlantic perspective on cost containment in clinical practice. Br Med J 1989; 299: 1323–1325.
16 Herzberg F. The Managerial Choice. Homewood IL: Dow Jones Irwin, 1976.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||