Interpersonal Continuity of Care and Patient Satisfaction: A Critical Review

INTRODUCTION

Continuity of care is considered to be a core principle of family medicine^1,2 and primary care,³ but conclusive proof regarding its value has eluded primary care and health policy researchers. The first step in establishing this proof requires a specific definition of the most important elements of continuity. Previous authors have addressed different aspects of continuity and have proposed definitions with multiple dimensions.^1,4–,¹⁰ Recently, 3 of these dimensions were organized into a hierarchy ranging from the availability of accurate information from one health care encounter to another (informational continuity), through a pattern of health care utilization at a particular site of care (longitudinal continuity), to a personal doctor-patient relationship characterized by loyalty and trust (interpersonal continuity).¹¹ For family physicians the concept of interpersonal continuity holds particular importance.¹² Interpersonal continuity refers to a special type of longitudinal continuity in which an ongoing personal relationship between the patient and clinician is characterized by personal trust and responsibility.¹¹ Changes in the American health care system during the past decade have made it increasingly difficult to establish such long-term trusting relationships between physicians and patients.^13–,¹⁹ Some authors have questioned whether a personal model of care is feasible, as health plans increasingly have required provider changes for economic reasons.²⁰ As these changes occur, health policy planners are looking for evidence supporting the value of interpersonal continuity.

Proving the value of interpersonal continuity would require us to show that outcomes or efficiency of care is somehow better when this attribute is present, but which clinical outcomes are most important to analyze and which are most amenable to methodologically sound analysis? If interpersonal continuity of care improves diabetes outcomes, does it necessarily follow that hypertension outcomes would also be improved? A logical starting point is to examine the degree to which interpersonal continuity improves patient satisfaction. Patient satisfaction is generally viewed as a desirable quality characteristic,³ and there are standardized techniques to measure satisfaction.²¹

Interpersonal continuity of care has been a common subject for review articles and philosophy articles in the medical literature.^{4–11,13,15,20,22–},³⁶ Several of these articles have concluded that interpersonal continuity improves patient satisfaction.^5,7,24,30,33 Others have suggested that a personal model of primary care may be an outdated concept.^20,32 Nearly all of these reviews have commented on the inherent difficulty of conducting research in this area. This article describes our efforts to examine systematically the evidence regarding the association between interpersonal continuity and one aspect of care quality, the level of patient satisfaction.

METHODS

A search of the MEDLINE database from January 1, 1966, to April 30, 2002, was conducted using the medical subject heading “continuity of patient care,” including all possible subheadings. Titles and reference citations of each of the resulting 2,424 citations in the English language articles were reviewed, and references were eliminated if they were letters to the editor, if they addressed health professions other than medicine, or if they clearly addressed only aspects of continuity other than interpersonal continuity. Excluding these articles left 379 citations that appeared to address interpersonal continuity as an attribute of the relationship between clinicians and patients in general medical care. Full-text copies of each of these articles were obtained and read in detail by the primary author. In addition, the bibliographies of each of these articles were scanned to identify potential references that were missed by the MEDLINE search. The process of article review and exclusion has been previously described.¹¹ This review process yielded 142 articles that directly related to the concept of interpersonal continuity. Forty-one were review or theory articles dealing with continuity of care in general. The remaining 101 were original research reports. All article citations were entered into a bibliographic database, and each was classified by study methods, primary research question(s), and how continuity was defined and measured. This classification process has been described.¹¹

Seventy-eight of the 142 articles addressed the relationship between interpersonal continuity of care and the outcome, cost, or patient satisfaction with care. An analysis of the relationship between interpersonal continuity and care outcomes, including cost efficiency, will be the subject of a future paper. Thirty of the articles addressed the relationship between interpersonal continuity and patient satisfaction and are the subject of this report.^{5,7,24,30,32,33,37–},⁶⁰ Twenty-two of these articles described original research on the subject, and the remaining 8 were reviews or related articles. The authors reviewed all 30 articles and systematically analyzed the 22 research articles to assess the significance and strength of this relationship. The goals of our analysis were to determine the quality of this evidence and to assess the methods used by the authors to examine the association.

We originally considered a meta-analysis of these data, but found it impossible because of substantial differences in how patient satisfaction was defined and measured among these articles. Thus, our analysis consisted of a systematic review of each article to determine the study method, the method by which interpersonal continuity was measured, the results of each study, and an assessment of the quality of evidence in each study. Quality was evaluated by determining the degree to which each study addressed 5 quality characteristics. We chose these 5 quality characteristics after reading the articles and discussing the project with colleagues. We then reviewed each study independently and assigned a score of 2, 1, or 0 for each quality characteristic that was met, partially met, or not met, respectively.

The 2 scores were averaged, and in those cases where the scores differed by more than 1 point (one score was 2 and the other 0), the article in question was reread by both authors and discussed to reach consensus. The 5 scores were totaled as a measure of the overall study quality. Rather than assigning a quality score based on study method, we sorted by study method to allow comparison of studies using different methods.

The 22 original research reports describe the results of 20 studies addressing the association between interpersonal continuity and patient satisfaction. These studies, as well as our qualitative assessment of the strength of evidence in each study, can be found in Appendix 1, which is available online as supplemental data at http:// www.annfammed.org/cgi/content/full/2/5/445/DC1. Because each study was evaluated by the 5 criteria listed as questions heading the columns in the appendix, there were 100 quality scores assigned for the 20 studies. In 12 of the 100 cells, our quality score differed by more than 1 point. In 8 instances the areas of disagreement related to the criteria dealing with the measurement of satisfaction, primarily the degree to which subjective vs objective measures were used and whether the instruments had been previously validated. In all other cells the score is the average of our 2 scores. The sum of the scores for each criterion is listed in the last column. Table 1⇓ lists further details about the setting and findings of each study, as well as the summative quality score from Appendix 1.

RESULTS

The relationship between continuity and patient satisfaction has been studied in 4 randomized trials that were identified in this review. All 4 found aspects of improved patient satisfaction in study groups with higher continuity of care. In the 1960s, Alpert and colleagues^47,57 randomly assigned 931 low-income children to receive care in a comprehensive pediatric clinic or in a traditional pediatric walk-in clinic in Boston. The care of these 2 groups was compared at 2 and 4 years. Overall patient satisfaction was significantly higher in the intervention group, and mothers in this group also rated the ease of talking with the physician and clinic staff more highly. In a similar study, Becker and colleagues^48,49 randomly assigned 125 low-income pediatric patients to either a clinic with continuity or a walk-in clinic and compared patient satisfaction 1 year later. Overall satisfaction was significantly higher in the continuity group, as was satisfaction with staff-patient interactions, clinic procedures, time spent with the physician, perceived quality of care, and confidence in the physicians’ recommendations. A study by Wasson et al reported in 1984 ⁵⁰ randomly assigned 776 men older than 55 years at a Veterans Administration (VA) clinic to either a continuity clinic or an outpatient clinic without continuity. Overall satisfaction and satisfaction with continuity of care, clinician thoroughness, and patient education were significantly higher in the continuity group after 2 years. In 1995 Rowley and colleagues⁵⁸ reported a trial that randomly assigned 405 pregnant Australian women to receive care either from a continuity clinic staffed by 6 midwives or from a university teaching clinic without continuity. The intervention group had higher scores for 3 aspects of patient satisfaction: information giving, participation in decision making, and relationship with caregivers. None of these trials identified any aspects of satisfaction that were worse with continuity.

All of these trials were flawed by study methods that failed to isolate continuity as the only uncontrolled difference between the study groups. In 2 of the 4 trials, continuity of care was not measured in either the intervention or control group.^48,49,58 Although the intervention groups were assigned to receive care in clinics that were designed to deliver care in an interpersonal continuity model, there is no evidence in these reports that the continuity was in fact better in these clinics. In the remaining 2 trials by Alpert et al^47,57 and Wasson et al,⁵⁰ interpersonal continuity was measured by the percentage of patients who could name their physicians. Wasson et al also reported the usual provider continuity index (UPC) and the sequential continuity index (SECON) to show higher continuity scores in the intervention group. In spite of these flaws, all 4 clinical trials showed a significant improvement in patient satisfaction when intervention and control groups were compared.

Cohort studies have addressed the relationship between continuity and patient satisfaction among patients with hypertension in Seattle,⁴² obstetric patients in the United States,⁵² pediatric patients in an academic practice in Cleveland,⁵³ and patients in a VA internal medicine clinic.⁴⁶ All of these studies rated or measured the continuity of care and attempted to correlate the scores with patient satisfaction. Two of these cohort studies^52,53 did not find a significant association between continuity and patient satisfaction. Shear et al⁵² found higher satisfaction in patients receiving care with higher continuity in a family practice clinic compared with patients in an obstetrics-gynecology clinic, but the difference did not achieve statistical significance, and this study was confounded by differing providers and care models in the two settings. Breslau and Haug⁵³ examined the change in continuity and satisfaction that occurred when 2 pediatricians moved their practices from the community into a teaching center. Although the patients were less likely to see their own physicians after the move, overall satisfaction did not change significantly. The remaining 2 cohort studies by Shortell et al⁴² and Smith⁴⁶ both found a significant association between improved continuity and patient satisfaction. Shortell et al measured continuity in an unusual way by simply counting the number of clinicians seen. This measurement technique has been used only twice in studies addressing interpersonal continuity of care¹¹ and may, in fact, be more of a measure of longitudinal continuity. Smith⁴⁶ measured continuity of care in a small subset of the patients enrolled in their study, raising the question of whether interpersonal continuity truly differed in the 2 cohorts.

Most of the work relating continuity and satisfaction has been done in cross-sectional surveys or other correlational studies. Five of these 12 articles reported studies done in Europe^{39,40,45,54,60} and a sixth was done in both the United States and United Kingdom.⁵⁵ Three of these studies compared British general practices using combined patient lists with those using personal lists.^39,45,60 Combined list practices assign patients to the practice site, whereas personal list practices assign them to an individual physician. All 3 showed patient satisfaction advantages with personal lists, but 2 of the 3 studies did not measure patient satisfaction directly.^39,45 A common weakness of these correlational studies is inconsistency in how patient satisfaction is measured. In 6 of the 12 studies, patient satisfaction was measured using a previously validated and standardized instrument.^{37,38,41,44,59,60} In 2 others, satisfaction is explicitly measured by a simple Likert scale or patient interview that had not been previously validated.^40,43 The remaining 4 studies do not directly attempt to measure satisfaction, but instead address issues that might be considered proxies for satisfaction, such as the desire to change clinicians or the willingness to wait for an appointment.^39,45,54,55 All these correlational studies except one show an association between at least one aspect of improved satisfaction and interpersonal continuity. The exception found no relationship between clinician continuity and satisfaction among prenatal patients at the University of Missouri.³⁸

DISCUSSION

The available data suggest a consistent significant association between interpersonal continuity of care and patient satisfaction. All these studies, however, are limited by a set of problems that impair our ability to isolate and study interpersonal continuity of care as a variable. Most Western health care systems allow patients to choose their own physicians, making it difficult to assign patients randomly to different care models. In the absence of random assignment, it is difficult to tell whether continuity leads to satisfaction or satisfaction leads to continuity. In fact, it seems likely that causality is bidirectional. Perhaps a better understanding of interpersonal continuity may come from studies using continuity as the dependent rather than the independent variable in a prospective design. Such a study might examine newly enrolled patients and follow them as a cohort to see which ones become long-term patients in the practice. The differences between those who stay and those who leave could reveal important elements of interpersonal continuity of care.

Selection bias is also a challenge to continuity researchers. Will interpersonal continuity improve satisfaction for all patients or for only those who care enough about it to choose such a model of care? Many of the studies in this review have shown increased patient satisfaction in practices that promote a personal physician for each patient. This finding does not mean that patients who are not in such practices would be more satisfied if their care model were changed. A few studies have examined ways to improve interpersonal continuity in a practice.^{45,46,53,60–},⁶⁴ It would be useful to add preintervention and postintervention satisfaction surveys to such studies in the future. Although not always feasible, a randomized trial design remains an ideal method to examine the relationship between continuity of care and satisfaction, but the intervention in such a trial does not need to be patient assignment. Instead, the intervention could be measures taken by a practice to improve interpersonal continuity.

There are some important limitations in the method we used for this review. Each report in this review used a slightly different approach to how continuity was defined and measured. Although these approaches were similar and all attempted to address interpersonal continuity, these differing techniques make it difficult to generalize from one study to the next. Our evaluation of the quality of each study was limited to only 5 criteria and was subjective. Finally, there was considerable diversity from one study to the next in the characteristics of the patient population being studied, and articles were included in the review from several western countries. On the one hand, this might increase the validity of our conclusions, but such diversity may also mask differences that limit our ability to generalize from one study to the next.

The availability of electronic medical records is enhancing our ability to communicate information efficiently from one clinician to another. We are no longer as dependent as we once were on the knowledge of a personal physician for many of the tasks of informational continuity. Some authors have predicted the end of personal care as we move to team-based care in the office and hospitalist systems for inpatient care.³² It is therefore more important than ever to establish those aspects of personal care that may be of enduring value to our patients. This review suggests that interpersonal continuity in the doctor-patient relationship has importance to patients in the United States, Great Britain, Australia, and several other countries.

In 1991, Gayle Stephens wrote, “As we struggle to direct our own evolution, let us not be intimidated or enticed to give up our most indispensable ideal—personal medicine. The need for it will never become passé, although we must continue to learn what it means and how to do it.”¹² Research into continuity remains limited by differing definitions and measurement techniques, but evidence that interpersonal continuity matters to patients and their physicians is sufficient to warrant caution as our health care system continues to change. Many aspects of the traditional interpersonal model of care may be essential elements of the future rather than outdated characteristics of the past.