Elsevier

Social Science & Medicine

Volume 48, Issue 4, February 1999, Pages 547-557
Social Science & Medicine

HMO physicians' use of referrals

https://doi.org/10.1016/S0277-9536(98)00380-3Get rights and content

Abstract

Clinical uncertainty is a source of variation in medical decision-making as well as a source of work-related stress. Increasing enrollment in organized health care systems has intensified interest in understanding referral utilization as well as issues such as physician dissatisfaction and burnout. We examined whether primary care physicians' affective reactions to uncertainty and their job characteristics were associated with use of referrals and burnout. Data came from mail surveys of primary care physicians practicing in two large group model health maintenance organizations (HMOs) in the USA. Consistent with past research, we found that younger physicians had higher referral rates than older physicians, and that general internists had higher rates than either family practitioners or pediatricians. Greater stress from uncertainty increased referrals and referrals were negatively correlated with heavier work demands (patient visits per hour). Greater stress from uncertainty, perceived workload (too high) and a sense of loss of control over the practice environment were associated with higher levels of burnout.

Introduction

The growth of organized health care systems and increasing use of primary care physicians as gatekeepers have intensified interest in referrals from primary care physicians to specialists. Patient dissatisfaction with policies to restrict direct access to specialists is growing and there is concern that factors other than patients' medical needs may influence referral patterns (Mechanic, 1996; Dolen et al., 1997). Increasing evidence pointing to medical practice variations and the role of nonmedical factors in explaining these variations have heightened these concerns (Eisenberg, 1986; Greenfield et al., 1992; Langley et al., 1992).

Uncertainty is ubiquitous in clinical medicine and is assumed to be one of the important causes of variation in clinical decision-making (Wennberg et al., 1982; Eddy, 1984) as well as a determinant of physician work-related stress (Logan and Scott, 1996). A relevant dimension of uncertainty is the physician's affective reaction to it. Gerrity et al. (1990)have developed scales to measure two components of this reaction: Stress from Uncertainty Scale (SUS) and Reluctance to Disclose Uncertainty Scale (RDUS). Whether physicians' affective reactions to uncertainty actually influence their clinical decisions and use of resources has not been established.

Physicians' utilization behavior is also shaped by the practice environment and associated job characteristics of physicians. Pineault observes: ``The social organization of medical practice consists of occupational members specially trained for the performance of therapeutic and diagnostic procedures. However, in the performance of these acts physicians are located within organizations that have requirements and demands of their own on the performance of their members'' (Pineault, 1974, p. 16).

Increasingly, physicians are practicing in organized health care systems that integrate the financing and delivery of health care. These are often referred to as `managed care' and can take many forms, ranging from modified fee-for-service indemnity plans with minimal controls over utilization and costs to highly organized systems that alter reimbursement mechanisms and control both the delivery system and provider behavior (Weiner and de Lissovoy, 1993; Freeborn and Pope, 1994). Although many view managed care as primarily a way of controlling costs, others see it as a means to higher-quality, more coordinated services for patients and enrollees (Luft and Greenlick, 1996). The most well-known type of managed care in the US is the health maintenance organization or HMO. Five major characteristics distinguish the HMO from other types of health care delivery systems (Luft, 1981):

  • 1.

    an enrolled population,

  • 2.

    responsibility for delivering necessary medical care within a fixed budget,

  • 3.

    low or no financial barriers to enrollee use,

  • 4.

    risk sharing by providers, not just the insurer,

  • 5.

    voluntary choice of plan

In HMOs, a variety of controls and incentives change the dynamic of traditional decision-making. These include displacement of financial risk to the practitioner and increased scrutiny of medical decision-making. These mechanisms are designed to increase efficiency and to improve the quality of care, but they also increase the tension in clinical decisions. HMO primary care physicians (and physicians in capitated and prepaid systems generally) frequently deal with undifferentiated illness, heavy patient demands and resource and time restrictions (Mechanic, 1975; Luft, 1981; Barr, 1983; Lichtenstein, 1984; Freeborn and Pope, 1994). A number of analysts have suggested that this situation leads to a sense of diminished control over both administrative and clinical aspects of practice and results in negative attitudes about the practice environment (e.g. professional dissatisfaction and/or burnout) (Mechanic, 1975; Luft, 1981; Lichtenstein, 1984; Schmoldt et al., 1994; Freeborn and Pope, 1994). However, few recent empirical studies have addressed these issues.

The use of primary care physicians as gatekeepers is a prominent feature of HMOs and other forms of managed care. The aim of the gatekeeper role is to improve the coordination of care and to encourage primary care physicians to use resources more efficiently. However, some suggest that it may motivate primary care physicians to overuse resources (e.g. diagnostic tests, prescriptions and referrals) as a way of dealing with time constraints and overdemanding patients (Freidson, 1973; Luft, 1981; Barr, 1983; Mechanic, 1986). Research on this topic is also very sparse.

The primary objective of this study was to determine whether HMO primary care physicians' affective reactions to uncertainty and their job characteristics are associated with use of referrals, after adjusting for physicians' demographic characteristics and characteristics of physicians' patient panels. A secondary objective was to examine the relationship between uncertainty, job characteristics and physician burnout. We hypothesized that greater stress from uncertainty, higher job demands and a decreased sense of control would be associated with higher referral rates and higher levels of physician burnout.

In a rapidly changing health care system, rationalizing the use of generalist and specialist services is an increasingly important issue. Although primary care gatekeepers are a central feature of most forms of managed care, the best use of general and specialist care remains unclear. Ambulatory care by specialists has been linked to increased costs and more intensive resource utilization (Glenn et al., 1987; Greenfield et al., 1992), while demonstration of improved outcomes remains controversial (Franks et al., 1992; Kassirer, 1994; Greenfield et al., 1995). Because of the high costs associated with specialist care, primary care physicians' referral behavior has been under scrutiny in the UK (Cummins et al., 1981; Armstrong et al., 1988; Roland et al., 1990; Wilkin, 1992) and increasingly so in the US (Salem-Schaatz et al., 1994). Many US managed care organizations require preauthorization for specialty care or use profiling to monitor primary care referrals (Kerr et al., 1995).

Wide interpractitioner variation in referral utilization by generalist physicians has been observed consistently. Explanations of this variation have been made on multiple levels, including patient factors such as diagnoses (Wilkin, 1992; Blancquert et al., 1992; Dovey et al., 1993), age (Penchansky and Fox, 1970; Armstrong et al., 1988; Coulter, 1992; Wilkin, 1992; Salem-Schaatz et al., 1994) and patients' request for specialty care (Armstrong et al., 1991; Langley et al., 1992); and physician factors such as expertise and training (Rothert et al., 1984; Reynolds et al., 1991; Calman et al., 1992) and case mix (Salem-Schaatz et al., 1994); and practice environment factors such as visit length (Franks and Clancy, 1997), rural versus urban setting (Penchansky and Fox, 1970) and availability and perceived quality of specialists (Ludke, 1982; Roland and Morris, 1988). The financing of care can also affect referral utilization. HMO patients are referred more frequently than fee-for-service patients (Mayer, 1982; Franks and Clancy, 1997), although the use of specialty care in the management of chronic diseases (such as hypertension and diabetes) is markedly decreased (Greenfield et al., 1992). Much of the interpractitioner variation remains unexplained and is hypothesized to involve practice style and unmeasured psychologic factors such as intolerance of uncertainty (Roland, 1988; Kassirer, 1989; Coulter, 1992; Nutting et al., 1992).

Although theoretically appealing, an association between intolerance of uncertainty and increased utilization of diagnostic testing or subspecialty care has little empiric support. Epstein et al. (1984)found no relationship between physicians' scores on Budner's `intolerance of ambiguity' scale and their use of diagnostic testing in hypertensive patients. In one recent study (Pearson et al., 1995), there was a trend towards increased admissions of emergency room chest-pain patients among physicians with higher scores on the Stress from Uncertainty Scale from Gerrity et al. (1990). Similarly, among European general practitioners, willingness to tolerate more uncertainty and to take risks in patient care was correlated with fewer specialty referrals (Grol et al., 1990).

Gerrity et al. (1990)recognize that physicians' decisions are influenced by factors other than uncertainty. Their conceptual model takes into account patient, physician and organizational characteristics and assumes that these interact with physician uncertainty to explain differences among physicians in their use of resources. Gerrity et al. (1990), however, did not test this model or analyze the relationship between stress from uncertainty and physicians' use of resources.

A number of researchers have examined variation in resource use among primary care physicians, but only a few of these studies have focused on referrals (Penchansky and Fox, 1970; Mayer, 1982; Eisenberg, 1985). Physician age, years of clinical experience and specialty have consistently been found to affect physicians' use of resources (Freeborn et al., 1972; Pineault, 1974, Pineault, 1976; Eisenberg, 1985, Eisenberg, 1986), and some research suggests that intensity of work load (measured by patient visits per hour) also influences physicians' utilization behavior (Freeborn et al., 1972; Luft, 1981; Eisenberg, 1985; Freeborn et al., 1983, Freeborn et al., 1987, Freeborn et al., 1989). The general finding is that primary care physicians' use of lab tests, imaging procedures and other outpatient resources decreases as patient visits per hour increase.

Burnout is a syndrome of physical and emotional exhaustion (Pines et al., 1981). Hypothesized effects of physician burnout include negative attitudes towards work and coworkers, decreased job satisfaction, increased likelihood of job attrition and more negative attitudes toward patients (Schmoldt et al., 1994). Few, if any, empirical studies have examined the relationship between burnout and physician referrals, but it is often assumed that burned out physicians tend to refer more.

Prevalence of burnout among HMO physicians has been reported to be in the range of 13–19%, depending on the measure used (Schmoldt et al., 1994). Previous research on physician burnout has emphasized the importance of both personal and organizational factors. In one recent study of HMO physicians (Deckard et al., 1994), emotional exhaustion was higher in younger physicians and female physicians, groups previously noted by Gerrity et al. (1990)to be more affected by uncertainty. This suggests that consistently high levels of stress from uncertainty may contribute to physician burnout. Other predictors of burnout are long hours and dissatisfaction with workload and scheduling (Schmoldt, 1991). Burnout tends to be associated with the kinds of stresses that HMO primary care clinicians face routinely — the stress of inadequate time per patient visit, long hours, resource constraints and lack of control over their schedules and other administrative aspects of practice. Low social supports and inability to divulge stress to others also appear to increase burnout levels (Pines et al., 1981; Schmoldt, 1991).

Section snippets

Method

The data for this study, conducted in 1991–1992, were obtained by mail surveys of physicians practicing in two regions of Kaiser Permanente (KP), the Northwest region and the Ohio region, USA. KP is a nonprofit prepaid group practice HMO that provides integrated inpatient and outpatient care for an enrolled population. The two KP regions serve over 600,000 members. Members resemble the overall area population in age distribution as well as health status and sociodemographic characteristics (

Stress from uncertainty

Consistent with Gerrity et al., 1990, we found that Stress from Uncertainty (SUS) was higher in younger physicians than in older physicians (ANOVA; p=0.03) and female physicians had higher levels of SUS than male physicians (ANOVA; p=0.01). In terms of specialty, general internists had higher levels than family practitioners and pediatricians (p=0.005). SUS was not associated with practice location or years with the HMO.

Referrals

The average percent of patients referred for consultations was 7.1, with a

Discussion

In this study, the average referral rate was 7.1%, with high interphysician variance. These results are similar to other studies of referral utilization (Cummins et al., 1981; Mayer, 1982; Armstrong et al., 1988; Wilkin, 1992; Calman et al., 1992), including the recent study by Franks and Clancy (1997). Our finding of higher referral rates among general internists (compared with family physicians) is also consistent with other studies (Penchansky and Fox, 1970; Barr, 1983; Rothert et al., 1984;

Acknowledgements

The authors thank Vicky Burnham for research assistance throughout this project and for her editing expertise during manuscript preparation. We also acknowledge the contributions of Ralph A. Schmoldt, Ph.D. and Harvey D. Klevit, MD. Both assisted in the design of the physician questionnaire and the carrying out of the physician surveys. We also thank those Northwest and Ohio Permanente physicians who made this study possible by their participation in the surveys.

References (72)

  • R.L Logan et al.

    Uncertainty in clinical practice: implications for quality and costs of health care

    Lancet

    (1996)
  • J.E Wennberg et al.

    Professional uncertainty and the problem of supplier induced demand

    Soc. Sci. Med.

    (1982)
  • D Armstrong et al.

    Measuring general practitioner referrals: patient, workload and list size effects

    J. R. Coll. Gen. Pract.

    (1988)
  • D Armstrong et al.

    Doctors' perceptions of pressure from patients for referral

    Br. Med. J.

    (1991)
  • J.K Barr

    Physicians' views of patients in prepaid group practice: reasons for visits to HMOs

    J. Health Soc. Behav.

    (1983)
  • K.D Bertakis et al.

    Gatekeeping in primary care: a comparison of internal medicine and family practice

    J. Fam. Pract.

    (1987)
  • I.R Blancquert et al.

    Referral patterns for children with chronic diseases

    Pediatrics

    (1992)
  • N.S Calman et al.

    Variability in consultation rates and practitioner level of diagnostic certainty

    J. Fam. Pract.

    (1992)
  • Coulter, A., 1992. The interface between primary and secondary care. In: Roland, M., Coulter, A. (Eds.), Hospital...
  • R.O Cummins et al.

    Do general practitioners have different `referral thresholds?'

    Br. Med. J.

    (1981)
  • G Deckard et al.

    Physician burnout: an examination of personal, professional and organizational relationships

    Med. Care

    (1994)
  • K Dolen et al.

    The new medical market place: physicians' views

    Health Aff.

    (1997)
  • S.M Dovey et al.

    New Zealand general practitioner referral patterns

    NZ Med. J.

    (1993)
  • D.M Eddy

    Variations in physician practice: the role of uncertainty

    Health Aff.

    (1984)
  • J.M Eisenberg

    Physician utilization: the state of the research about physicians' practice patterns

    Med. Care

    (1985)
  • Eisenberg, J.M., 1986. Doctors Decisions and the Cost of Medical Care. Health Administration Press, Ann...
  • G.J Elwyn et al.

    Avoidable referrals: analysis of 170 consecutive referrals to secondary care

    Br. Med. J.

    (1994)
  • A.M Epstein et al.

    The effects of physicians' training and personality on test ordering for ambulatory patients

    Am. J. Public Health

    (1984)
  • A Fertig et al.

    Understanding variation in rates of referral among general practitioners: are inappropriate referrals important and would guidelines help reduce rates?

    Br. Med. J.

    (1993)
  • P Franks et al.

    Referrals of adult patients from primary care: demographic disparities and their relationship to HMO insurance

    J. Fam. Prac.

    (1997)
  • P Franks et al.

    Gatekeeping revisited-protecting patients from overtreatment

    N. Eng. J. Med.

    (1992)
  • D.K Freeborn et al.

    The determinants of medical care utilization: physicians' use of laboratory services

    Am. J. Public Health

    (1972)
  • Freeborn, D.K., Johnson, R.E., Mullooly, J.P., 1983. Physicians' Use of Ambulatory Resources in a Prepaid Group...
  • D.K Freeborn et al.

    Length of service, leadership and outpatient resource use among HMO physicians

    GHAA J.

    (1987)
  • Freeborn, D.K., Mullooly, J.P., Johnson, R.E., 1989. Practice Variations among HMO Physicians: the Effects of...
  • Freeborn, D.K., Pope, C.R., 1994. Promise and Performance in Managed Care. John Hopkins University Press,...
  • E Freidson

    Prepaid group practice and the new `demanding patient'

    Milbank Memorial Fund Q./Health Soc.

    (1973)
  • M.S Gerrity et al.

    Physician's reactions to uncertainty in patient care

    Med. Care

    (1990)
  • J.K Glenn et al.

    Physician referral in a competitive environment: an estimate of the economic impact of a referral

    JAMA

    (1987)
  • S Greenfield et al.

    Variation in resource utilization among medical specialties and systems of care: results from the Medical Outcomes Study

    JAMA

    (1992)
  • S Greenfield et al.

    Outcomes of patients with hypertension and non-insulin dependent diabetes mellitus treated by different systems and specialties: results from the Medical Outcomes Study

    JAMA

    (1995)
  • Greenlick, M.R., Freeborn, D.K., Pope, C.R., 1988. Health Care Research in an HMO. Johns Hopkins University Press,...
  • R Grol et al.

    Work satisfaction of general practitioners and the quality of patient care

    Fam. Pract.

    (1985)
  • R Grol et al.

    Attitudes to risk taking in medical decision making among British, Dutch and Belgian general practitioners

    Brit. J. Gen. Pract.

    (1990)
  • Karasek, R., Theorell, T., 1990. Healthy Work. Basic Books, New...
  • J.P Kassirer

    Our stubborn quest for diagnostic certainty: a cause of excessive testing

    N. Eng. J. Med.

    (1989)
  • Cited by (67)

    • Defensive practices among non-medical health professionals: An overview of the scientific literature

      2019, Journal of Healthcare Quality Research
      Citation Excerpt :

      Furthermore, unnecessary invasive procedure and the increased intervention rate create risks of patient harm,8,31 and costs rise.11,40 In addition, job dissatisfaction and poor working relationship can cause health professionals burnout, that, subsequently, can damage also patients and the whole health care system.11,41–48 Some critical issues concerning the internal validity of the present study should be stated.

    View all citing articles on Scopus
    View full text