Provider Practice Characteristics That Promote Interpersonal Continuity

Tyler S. Mittelstaedt; Motomi Mori; William E. Lambert; John W. Saultz

doi:10.3122/jabfm.2013.04.120306

Article Figures & Data

Figures

Tables

Download figure
Open in new tab
Figure 1.
Simple linear regression of Usual Provider Continuity Index (UPC) (percentage of total clinic visits with assigned provider) on clinic frequency (monthly half-day clinic sessions held).
Download figure
Open in new tab
Figure 2.
Simple linear regression of Usual Provider Continuity Index (UPC) (percentage of total clinic visits with assigned provider) on patient load (panel size-to-clinic frequency ratio).
Download figure
Open in new tab
Figure 3.
Multivariate prediction of Usual Provider Continuity Index (UPC) (percentage of total clinic visits with assigned provider) based on clinic frequency (monthly half-day clinic sessions held), stratified by provider type (physicians vs. mid-level providers). Predicted UPC calculated at fixed values for patient load (panel-to-half-day ratio, 38.2) and duration in practice (6.35 years). Dashed lines denote data extrapolation outside of the observed range of clinic frequency for either provider type.

Tables

Figures

View popup

Table 1. Descriptive Summary of Outcome and Predictor Variables, Stratified by Provider Type

Variable	Mean	SD	Min	Max	P^*
UPC (%)^{^†}
All providers	62.92	11.55	37.30	87.40
Physician	60.97	9.84	37.29	78.64	.033
Mid-level	67.78	14.16	41.93	87.41
Clinic frequency (monthly half-day clinic sessions held)
All providers	16.83	7.27	4.92	31.30
Physician	13.63	5.04	4.92	24.42	.000
Mid-level	24.85	5.63	15.06	31.30
Panel size (count of patients assigned to provider)
All providers	577.4	315.8	65	1377
Physician	540.5	274.3	92	1288	.144
Mid-level	669.7	395.4	65	1377
Patient load (ratio of panel size to half-day clinic frequency)
All providers	38.21	19.96	2.40	93.67
Physician	41.32	18.80	5.90	93.70	.049
Mid-level	30.41	21.22	2.41	72.55
Attendance ratio^{^‡}
All providers	1.01	0.26	0.56	2.15
Physician	1.04	0.30	0.56	2.14	.140
Mid-level	0.93	0.13	0.68	1.29
Duration in practice (years)
All providers	6.35	6.38	0.42	24.92
Physician	7.19	6.85	0.67	24.92	.099
Mid-level	4.26	4.50	0.42	16.75

↵* Two-sample t test for difference in mean by provider type.
↵† Outcome variable: percentage of total clinic visits with assigned provider.
↵‡ Ratio of actual clinic frequency to expected clinic frequency.
UPC, Usual Provider Continuity Index; SD, standard deviation.

View popup

Table 2. Simple Linear Regression of Usual Provider Continuity Index^* on Predictor Variables

Predictor Variable	Β	95% CI	r	P
Clinic Frequency^{^†}	0.944	0.618–1.271	0.595	<.0001
Panel Size^{^‡}	−0.009	−0.018 to 0.000	0.209	.044
Patient Load^{^§}	−0.370	−0.483 to −0.256	0.639	<.0001
Attendance Ratio^{^‖}	0.050	−0.06 to 0.162	0.115	.370
Duration in Practice^{^¶}	−0.265	−0.724 to 0.193	0.146	.252

↵* Percentage of total clinic visits with assigned provider.
↵† Monthly half-day clinic sessions held.
↵‡ Count of patients assigned to provider.
↵§ Ratio of panel size to clinic frequency.
↵‖ Ratio of actual clinic frequency to expected clinic frequency.
↵¶ Duration (years) in current practice.

View popup

Table 3. Optimal Multivariate Linear Model for Prediction of Usual Provider Continuity Index^*

Variables	β	95% CI	P^{^†}
Clinic frequency	1.52	0.84–2.19	<.001
Patient load	−0.32	−0.45 to −0.20	<.001
Duration in practice	0.61	0.27–0.95	.001
Provider type^{^‡}	23.01	5.62–40.41	.010
Type and patient load interaction	−0.81	−1.58 to −0.04	.039
Intercept^{^§}	62.91	54.74–71.07	<.001

↵* Percentage of total clinic visits with assigned provider.
↵† P values reflect partial F-tests for significance of individual terms within the model.
↵‡ Reference group is mid-level providers (mid-level = 0, physician = 1).
↵§ Data centered around mean clinic frequency of 16.8 monthly clinic half-days (mean clinic frequency for all providers).

View popup

Table 4. Qualitative Data Summary from Focus Group Interviews

Themes	Subthemes	Description
1. Clinic diversity	1.1 PCP assignment accuracy	Clinics have different systems for ensuring that PCP fields are frequently updated and accurate.
	1.2 Scheduling	Clinics place variable importance on scheduling patients with their PCP rather than the first available provider.
	1.3 Location/type	University, FQHC, and rural clinics may have intrinsic differences in patient population and structure.
2. Provider diversity	2.1 Provider type	Physician (MD/DO) and mid-level (PA/FNP) providers may have differences in both scope of practice and approach to care.
	2.2 Provider sex	Female and male providers may have differences in both scope of practice and approach to care.
	2.3 Scope of practice	Providers who provide maternity care, inpatient care, sports medicine, suboxone, resident precepting, etc., may have different UPC patterns than providers who solely conduct outpatient continuity clinics.
	2.4 Nonclinical duties	Providers with more academic/administrative responsibilities may have different UPC patterns than those with strictly clinical practices.
	2.5 Clinic scheduling	Providers with more open schedules or night/weekend clinics may have different UPC patterns than those with primarily prescheduled daytime clinics.
	2.6 Location of residency	Providers who trained at OHSU may have more developed panels for their year in practice than providers who trained elsewhere.
3. Patient diversity	3.1 Panel demographics	SES, sex, race, ethnicity, age, medical complexity, visit frequency, and other panel demographics are likely to influence UPC for a given provider.
3. Patient diversity	3.2 Importance of continuity to patient	Some patients frequently change providers for a variety of reasons, making their PCP field relatively arbitrary.
4. Visit type	4.1 Acute care	Acute care visits may not be as important as chronic or ongoing care in terms of actual continuity.
4. Visit type	4.2 Nonoffice visits	Current calculation of UPC does not take into account phone, E-mail, or MyChart encounters, where a meaningful interaction occurs without face-to-face contact.
5. Non-PCP continuity	5.1 Team continuity	Patients may value continuity with a team of providers more than with a specific provider.
	5.2 Clinic continuity	Patients may value continuity with a particular clinic more than with an individual or team of providers.
	5.3 Family continuity	Continuity across families may be more important than continuity with individual patients.
6. Absences	6.1 Planned vs. unplanned	Are all absences from clinic a diversion from continuity (eg, deliveries, inpatient), or is it just unplanned absences (eg, illness)?
6. Absences	6.2 New parent leave	Specific extended absence that may behave differently than other absences in terms of continuity.

DO, osteopathic doctor; FNP, family nurse practitioner; FQHC, federally qualified health center; MD, medical doctor; OHSU, Oregon Health & Sciences University; PA, physician assistant; PCP, primary care provider; UPC, Usual Provider Continuity Index.