Data Collection and Analysis
The clinicians who agreed to participate in the study were sent protocol instructions and study materials. They were asked to complete data collection for each patient visit in which care was provided to a secondary patient during a 1-week period. They also reported the total number of all patient visits during the study period.
The data forms were sent to the ASPN central office where they were manually checked for completeness and key entered. The data reported by the clinicians were merged with information on the characteristics of the clinician (age, sex, years in practice) and practice characteristics (rural, urban, suburban) obtained from the ASPN member database.
Descriptive statistics are reported for primary patients, secondary patients, visits, and clinicians. We used chi-square tests for comparisons involving categorical variables and Student t tests to compare means for continuous variables. Significance was reported at P <.05.
Results
The 170 clinicians in 50 ASPN practices who participated in the study reported a total of 6957 patient visits during the 1-week reporting period. Ninety-five of the clinicians (56%) reported 1 or more instances of providing secondary care, yielding a total of 422 (6.1%) visits involving secondary care. Seventy-five clinicians reported no secondary care. The secondary encounter was most often initiated by the primary patient (55%) and least often by the clinician (15%). Secondary patents were present in the office 39% of the time and initiated the secondary care during 30% of the visits.
Clinicians estimated that the secondary care required an average of 4.9 minutes to deliver (range=1-60 minutes). They also reported that 64% of the secondary encounters were likely to have substituted for a separate office visit, while additional billing for the care was reported in only 5.2% of secondary encounters.
Categories of Service to the Secondary Patient
Advice was the discernable service provided in more than half the visits Table 1. Approximately 30% were accounted for equally by prescription, assessment or explanation of symptoms, and general discussion of condition. In addition, advice was also the most frequent service when the secondary patient encounter was judged by the clinician to have substituted for a separate office visit. In fact, advice was the most common in almost every secondary patient category, except secondary patients who were aged 65 years or older, where follow-up of a previous problem was the service category most likely to occur (data not shown).
Finally, certain services were more likely to be initiated by clinicians than patients Table 1. Follow-up and general discussion of a condition were associated with clinician-initiated secondary care, while advice, assessment or explanation of symptoms, and prescriptions were associated with patient-initiated secondary care.
Characteristics Associated with Secondary Care
There were few differences between clinicians who reported secondary care and those who did not. Physicians reporting secondary care were older (P <.05) and more likely to practice in a rural area (P <.05). Clinician sex and years in practice were not remarkable.
Table 2 shows the characteristics of primary and secondary patients. There were a greater percentage of women than men in the primary patient group (64%). The secondary patient was most often a spouse, parent, or child of the primary patient. Eighty-seven percent of the secondary patients were enrolled as patients in the practice.
Discussion
ASPN clinicians reported providing secondary care during approximately 6% of primary care visits and rarely billed for the service. Secondary care was provided primarily in the form of advice to another family member. An episode with a secondary patient was reported to take an average of 5 minutes and to substitute for a visit more than 60% of the time.
This is the first study to examine the content of care given to a secondary patient in community primary care practices. Although arranging a referral, dispensing a prescription (perhaps a renewal), or providing follow-up care might not be unexpected, ASPN clinicians reported more instances of the provision of more time-intensive and complex services, such as advice, assessment or explanation of symptoms, and general discussion of condition. The fact that clinicians reported this secondary care could substitute for an actual office visit 60% of the time further suggests some complexity of the service provided.
The observation that certain services were more likely associated with clinician—rather than patient-initiated—secondary care might relate to how comfortable a clinician was with a particular service. However, the strength of the association for follow-up of a previous episode of care supports the Institute of Medicine definition of primary care as continuous and accountable.16
A limitation of our study is the reliance on physician self-report, which might vary from the report of the patient or an objective observer. The lower frequency of secondary care than reported in the direct observational study by Flocke and colleagues11 is likely due to the lower sensitivity of physician self-report versus direct observation of service delivery.17-19