When is caring sharing? Primary care provider interdependence and continuity of care : JAAPA

Secondary Logo

Journal Logo

AAPA Members can view Full text articles for FREE. Not a Member? Join today!
Original Research

When is caring sharing? Primary care provider interdependence and continuity of care

Everett, Christine M. PhD, MPH, PA-C; Christy, Jacob; Morgan, Perri A. PhD, PA-C; Docherty, Sharron L. PhD, PNP-BC, FAAN; Smith, Valerie A. DrPH; Anderson, John B. Jr. MD, MPH; Viera, Anthony MD; Jackson, George L. PhD, MHA

Author Information
JAAPA 36(1):p 32-40, January 2023. | DOI: 10.1097/01.JAA.0000902896.51294.47
  • Free


Efforts to improve access to high-quality, efficient primary care have highlighted the need for team-based care. The United States' aging population and the increasing prevalence of chronic illnesses will not only increase the demand for primary care visits, but also will result in an increase in the average number, duration, and complexity of issues addressed at each visit. The move from volume- to value-based care also has placed increased focus on primary care and created accountability for meeting quality, cost, and patient-experience targets.1-5

Team-based primary care, a founding principle of the physician associate/assistant (PA) profession, occurs when two or more primary care providers (PCPs) work collaboratively to deliver care to a panel of patients.6,7 A key structural feature of a team approach is sharing the workload, resulting in task interdependence between team members.8,9 Task interdependence is the reliance of team members on each other to complete their work.9-12 In primary care, task interdependence can occur in a visit or between visits.8,10,11 In-visit interdependence can be extensive and involve many professionals, including medical assistants, nurses, PAs, NPs, physicians, pharmacists, and social workers.10,13 Examples include performing screening questionnaires, taking vital signs, history taking, physical examinations, and patient education.10 Because primary care is comprehensive and longitudinal, between-visit tasks result in between-visit interdependencies.14 A common example is nurse management of PCP inbox messages.15 Additionally, between-visit interdependence can occur when PCPs share patients, such as when PAs or physicians see each other's patients.16 Research indicates that this sharing of common patients, or PCP interdependence, has the potential to improve disease-specific quality of care.11,17,18 However, most primary care teams are designed to minimize patient sharing, in an attempt to preserve interpersonal continuity of care between patient and PCP.8,19-21

Continuity of care is a defining characteristic of primary care and is associated with improved patient outcomes—particularly for outcomes relevant to medically complex patients, for whom disease-specific quality measures may not be ideal.22-27 Evidence suggests that higher continuity of care leads to better care experiences, and is associated with fewer ED visits and hospitalizations and lower costs of care.23,28,29 However, certain patient and organizational characteristics, such as PCP turnover, have been demonstrated to affect continuity of care.30 Patient factors that can reduce continuity of care include sex, age, and medical complexity.30-32 However, when simultaneously considered, organizational factors, such as PCP turnover, resulted in bigger disruptions in continuity of care than patient factors.30 What remains unknown is the effect of PCP interdependence (that is, patient sharing) on continuity of care.

The first objective of this study was to describe a novel measure of PCP interdependence. This measure reflects the structural interdependence of PCPs who practice in the same clinic.9,12,33 Structural interdependence is considered a critical team design feature because of its effect on team performance.34,35 Historically, patients in primary care are divided into panels under a single PCP (called the usual provider) who has responsibility for the quality of care delivered to the patients on their panel. This reduces task and goal interdependence and improves patient continuity of care.14,33 Similarly, rewards such as production bonuses also frequently are calculated at the PCP level.36,37 However, because primary care is comprehensive and longitudinal, a panel's usual provider usually is unable to provide all the care to all the patients on the panel.16,38,39 As a result, PCPs in the same clinic often see patients from other PCPs' panels (that is, they are supplemental providers on other panels), resulting in PCP interdependence (Figure 1).8,16 Therefore, we propose that measuring PCP interdependence in primary care should be considered at the panel level and reflect the number of shared patients seen by each supplemental provider. The degree to which PAs and NPs need to be interdependent with physicians and still deliver quality care remains a point of significant debate. As a structural feature, it can be manipulated by team leaders and members to provide the most beneficial outcomes for patients.33

PCP interdependence on panel levelX = PCP, square box = patient

Because evidence suggests that patient and organizational factors affect continuity of care, the second objective of this study was to evaluate the association between patient characteristics, malleable structural team characteristics including PCP interdependence, and continuity of care.23,24 To enhance the potential for using this information, we also estimated the effect of PCP interdependence and other team structural characteristics such as patient characteristics, panel size, and PCP type on continuity of care.23,24 Models of team effectiveness postulated complex relationships between team structure, care delivery, and patient outcomes.40 Key factors of team structure such as patient characteristics, care delivery setting, and interdependence can directly and indirectly affect how care is delivered, including continuity of care.40,41 Understanding these relationships will provide us with a better understanding of how to match features of team design, such as interdependence, to patient needs.


Study setting

This cohort study was conducted with data from patients who received care at 26 healthcare-system affiliated primary care practices in central North Carolina. Although these practices are affiliated with an academic healthcare system, the practices themselves generally do not operate as academic practices. For example, only one of the 26 practices included a teaching clinic with residents providing care, and the vast majority of PCPs did not have regular rank academic appointments. The results are based on the in-person visits delivered by 134 physicians, 20 NPs, and 10 PAs at family medicine and internal medicine clinics. These clinics provide more than 500,000 visits per year to 193,000 patients. This study was approved by the Duke Health institutional review board, which waived the requirement for individual patient consent.

Sample and data sources

The sample consisted of adults with diabetes having at least two in-person primary care visits in 2016 (N = 18,808). Diabetes was chosen because these patients routinely use primary care and it is a frequently used condition to measure healthcare quality.42 All data were obtained from the electronic health record (EHR). Patients were identified as having diabetes using billing codes from outpatient and inpatient data.43

The usual PCP was determined by a previously published approach.16-18 Briefly, the usual clinic of each patient was identified as the primary care clinic that provided the majority of their in-person primary care visits. Next, the PCP (physician, NP, or PA) was the provider who delivered the majority of the in-person visits in the usual clinic. In the event of ties, patients were assigned to the clinic and PCP with the most recent visit. A PCP panel was defined by patients for whom that PCP delivered the majority of in-person primary care visits.


Patient variables included the dependent variable as well as independent variables. The dependent variable is a density measure of continuity ranging from 0 to 1, with 1 being perfect continuity of care.44 This measure was calculated as the percentage of primary care visits with usual PCP in the year. Independent patient variables included sociodemographic and clinical variables. Sociodemographic variables included age, sex, and race. Two measures of overall morbidity burden were used. A count of chronic conditions other than diabetes was calculated and categorized. The hierarchical clinical category score was used as an overall measure of healthcare service use due to morbidity.45

Panel-level (that is, usual provider) measures were constructed by aggregating (averaging) patient-level measures. Panel variables include PCP variables and panel characteristics. Usual provider type is a binary measure indicating either a physician or a PA or NP. The number of supplemental providers on the panel represents the number of PCPs who shared patients with the usual PCP by providing at least one visit to one patient on the panel in the year.17 Interdependence is a continuous measure of patient sharing between the usual PCP and other PCPs in the same clinic (number of shared patients divided by the number of supplemental providers). Panel characteristics are represented by panel size (that is, the number of patients with diabetes with the same usual PCP), average patient age, proportion female, average hierarchical clinical category score, and primary care visits (total visits, visits with the usual PCP and visits with the supplemental providers).


Descriptive analyses, including means, standard deviations, and percentages were calculated for patient and panel characteristics including PCP interdependence. To evaluate the association of patient and panel characteristics on patient continuity of care, binomial regression was performed with clustering by practice. Postestimation margins of continuity of care were estimated using the delta method for standard error for two patients. The first represented the highest complexity of patient seen in the dataset. This patient profile is of an 80-year-old White woman with nine or more chronic conditions who had 43 in-person primary care visits in the year. The second patient profile represented a heathier-than-average patient with diabetes with average primary care use (a 55-year-old Black man with two chronic conditions and four primary care visits). Continuity of care was estimated for each patient using nine panel scenarios each for physicians and PAs or NPs (total of 18 scenarios). The first two scenarios represented the values that would provide the highest continuity of care (largest panel of patients with no supplemental providers and therefore no PCP interdependence) (Figure 2). The second two scenarios represented the values with the worst possible continuity of care (smallest panel size, highest number of supplemental providers, and PCP interdependence). The third two scenarios represented the mean for panel size, number of supplemental providers, and interdependence. To show the potential effect of varying panel size, four more scenarios set the number of supplemental providers and interdependence at the mean, and then set the panel size to the minimum and maximum (Figure 3). Finally, we chose four more scenarios (Figure 4): the mean panel size and provider team with the minimum PCP interdependence; and the mean panel size and provider team with the maximum provider interdependence.

Best, mean, and worst continuity of care
Continuity of care variation by panel sizes
Continuity of care variation by panel interdependence


Patient characteristics

Patients in the sample (N = 18,808) were demographically mixed, reasonably healthy, and on average, had most of their visits with their usual PCP (Table 1). About 55% of the sample was female, with 54% White, 39% Black, and 7% other or not reported. The average age was 63 years with an average of two chronic conditions and four primary care visits per year. On average, patients had 86% of their primary care visits with their usual PCP. Patients with physicians as usual PCPs were, on average, older than those with PAs or NPs as usual PCPs (mean age, 63.61 versus 58.03). Similarly, patients of physicians were better insured compared with patients of PAs and NPs (7.51% uninsured or not reported versus 12.86%). Patient clinical characteristics were similar for those with physicians and PAs or NPs as usual PCPs.

TABLE 1. - Patient descriptive statistics
Patients (N = 18,808) Total mean (SD) or % Physician (n = 16,389) NP or PA (n = 2,419) Standardized mean difference
Age 62.9 (12.97) 63.61 (12.77) 58.03 (13.25) 0.428
Female 54.27 52.93 63.46 0.214
Race 0.083
   White 54.55 55.06 51.18 -
   Black 38.84 38.33 42.33 -
   Other/not reported 6.6 6.61 6.49 -
Type of insurance 0.286
     Private 32.95 31.9 40.06 -
     Medicare 58.85 60.59 47.09 -
     Medicaid/other government insurance/uninsured/missing 8.2 7.51 12.86 -
Hierarchical condition category score 0.84 (0.21) 0.84 (0.21) 0.83 (0.19) 0.061
Count of chronic conditions 1.80 (2.13) 1.82 (2.14) 1.69 (2.04) 0.062
Chronic condition count category 0.091
   1 27.64 27.59 27.95 -
   2 30.78 30.56 32.29 -
   3 to 5 32.12 32.11 32.24 -
   6 to 8 6.51 6.77 4.75 -
   9 or more 2.95 2.98 2.77 -
Primary care use
   Total visits 4.43 (3.14) 4.45 (3.19) 4.29 (2.8) 0.054
   Visits with usual PCP 3.71 (2.48) 3.76 (2.51) 3.37 (2.22) 0.166
   Visits with supplemental provider 0.72 (1.31) 0.69 (1.3) 0.92 (1.34) -0.175
Interpersonal continuity 0.87 (0.19) 0.88 (0.18) 0.81 (0.22) 0.336

Panel characteristics including PCP interdependence

Panels of physicians and NPs or PAs showed differences in demographics and use (Table 2). The average panel consisted of 111 adults with diabetes, with physicians on average having larger panels (121 patients) than NPs or PAs (72 patients). Average age of PA or NP panels was lower (57.24 years versus 62.98 years) and contained a higher proportion of women (59% versus 51.51%). Hierarchical clinical category score was higher for physicians (0.84 versus 0.83). Supplemental providers on average delivered about 21% of the in-person visits to each panel of patients with diabetes. On average, each panel of patients with diabetes had eight supplemental providers, but PA and NP panels had fewer supplemental providers (7.47 versus 8.20). However, panels with NPs or PAs had higher proportions of visits with supplemental providers (21%) than physicians (16%). PCP interdependence, on average, was 5.7 (that is, 5.7 patients with diabetes per supplemental provider) and ranged from 0 to 20.8. Panels of PA or NP usual PCPs had lower interdependence (5 versus 5.84).

TABLE 2. - Patient descriptive statistics
Panels (N = 172) Total mean (SD) Physician (n = 138) NP or PA (n = 34) Standardized mean difference
Size 111.35 (68.73) 121.02 (68.83) 72.12 (47.56) 0.819
   Age 61.85 (4.72) 62.98 (4.07) 57.24 (4.41) 1.354
   Female 52.99 (19.27) 51.51 (19.47) 59 (17.45) -0.405
   Hierarchical condition category score 0.84 (0.04) 0.84 (0.03) 0.83 (0.04) 0.359
Primary care use
   Total visits 484.10 (328.64) 528.23 (336.46) 305 (220.25) 0.785
   Visits with usual PCP 405.51 (279.8) 446.42 (284.52) 239.47 (184.96) 0.862
   Visits with supplemental provider 78.59 (64.95) 81.81 (67.88) 65.53 (50.10) 0.273
   Number of supplemental providers 8.05 (4.83) 8.20 (5.08) 7.47 (3.69) 0.163
Interdependence 5.67 (3.64) 5.84 (3.74) 5.00 (3.2) 0.241

Characteristics that predict continuity of care

In the regression analysis, patient and panel characteristics predicted continuity of care (Table 3). Controlling for all patient and panel characteristics, female sex was the patient characteristic that predicted the greatest reduction in continuity of care (OR = 0.81, P < .001, 95% CI [0.756, 0.866]). Increases in an individual patient's in-person visits was also associated with reduced patient continuity of care (OR = 0.91, P < .001, 95% CI [0.900, 0.921]). Compared with White patients, patients of a race other than Black had improved continuity of care (OR = 1.16, P = .031, 95% CI [1.014, 1.332]). Increases in continuity of care were seen for patients with two chronic conditions compared with only one (OR = 1.13, P = .037, 95% CI [1.007, 1.258]). However, no differences were seen with three or more chronic conditions. Continuity of care also increased slightly with age (OR = 1.01 for each year of age, P < .001, 95% CI [1.004, 1.010]).

TABLE 3. - Characteristics predicting continuity (N = 18,808)
OR Robust SE 95% CI
Patient demographics
   Age 1.01 0.001 1.004 1.01
   Female 0.81 0.028 0.756 0.866
   White ref
   Black 0.96 0.043 0.88 1.047
   Other/not reported∗∗ 1.16 0.081 1.014 1.332
Type of insurance
   Private ref
   Medicare 1.04 0.05 0.946 1.14
   Medicaid/other government insurance/uninsured/missing 1.03 0.052 0.929 1.133
   Hierarchical condition category score 0.91 0.054 0.812 1.026
Chronic condition count category
   1 reference
   2∗∗ 1.13 0.064 1.007 1.258
   3 to 5 1.07 0.064 0.954 1.203
   6 to 8 1.07 0.094 0.902 1.271
   9 or more 1.01 0.078 0.871 1.178
Total primary care visit count 0.91 0.005 0.9 0.921
Panel size∗∗ 1.01 0.002 1.001 1.008
Usual PCP: NP or PA∗∗ 0.69 0.122 0.486 0.973
Interdependence 0.86 0.016 0.827 0.891
Number of supplemental providers∗∗ 0.94 0.019 0.908 0.982
Average age 0.98 0.023 0.931 1.022
Percentage female 1 0.001 0.999 1.004
Average hierarchical condition category score 0.48 0.431 0.08 2.813
Average chronic condition count 0.91 0.076 0.778 1.076
Total panel visits∗∗ 1 <0.001 1 1.002
P < .001
∗∗P < .05

Panel characteristics both reduced and increased continuity of care. Controlling for all patient and panel characteristics, the largest reductions in continuity of care were seen with interdependent PCPs. As interdependence increased (that is, as more patients were shared with supplemental providers), continuity of care showed significant reductions (OR = 0.86, P < .001, 95% CI [0.827, 0.891]). Additional reductions in continuity of care occurred with each new supplemental provider serving a panel of patients (OR = 0.94, P = .004, 95% CI [0.908, 0.982]). Increases in continuity of care were seen with increases in workload. Increases in the number of patients with diabetes on the panel increased continuity of care (OR = 1.01, P = .002, 95% CI [1.001, 1.008]) as did the number of in-person visits provided to the panel (OR = 1.001, P < .001, 95% CI [1.000, 1.002]).

Postestimations of the regression suggest that complex patients have a greater challenge receiving high continuity of care, but patients with high and average complexity experienced the best continuity of care when on large panels with no supplemental providers and no interdependence. The lowest continuity of care was seen with both types of PCPs when panels were small (11 patients), interdependence was high (34 patients/supplemental provider), and there was a large number of supplemental providers (20.8 providers) (Figure 2). The complex patient (an 80-year-old White woman with more than nine chronic conditions, Medicare, and 43 primary care visits in the year) experienced continuity of care ranging from 0.002 to 0.73. Estimates of continuity of care for the less-complex patient (a 55-year-old Black man with two chronic conditions, private insurance, and four primary care visits in the year) ranged between 0.078 and 0.99 (Figures 3 through 5). Continuity of care with the smallest PCP team (usual PCP plus one supplemental provider) ranged from 0.43 to 0.97 when PCP interdependence was at the mean of 5.7 patients per PCP. When PCP interdependence was increased to 20.8 patients, continuity of care ranged from 0.07 to 0.79 (Figure 6).

Continuity of care variation by supplemental providers
Continuity of care with small teams


This study describes between-visit PCP interdependence serving adults with diabetes. We found that the extent of structural interdependence, or patient sharing, between PCPs in the same clinic varied widely. We also found that a range of patient and panel characteristics affects continuity of care, including PCP interdependence. As would be expected, the best continuity of care occurred on a large panel with no supplemental providers and no patient sharing. When PCP interdependence occurred, patients appeared to have better continuity of care when on large panels with fewer supplemental providers and lower interdependence. However, for the most-complex patients, minimizing the number of supplemental providers and interdependence may not be enough to achieve adequate continuity of care with an individual PCP.22

Interdependence between PCPs in the same clinic generally occurs as patient sharing in the form of provision of in-person visits to patients assigned to another PCP's panel.8,16 The measure of PCP interdependence varied widely within this study sample, ranging from no shared patients on a single panel, to 20 patients per supplemental provider. The mean interdependence was about 5.7 for all PCPs. No large differences were found in mean interdependence between PCP types. However, physicians had the highest mean (5.84) and PAs or NPs had the lowest mean (5). The number of supplemental providers also was similar by PCP type. PAs and NPs had lower mean number of supplemental providers on their panels (7.47); physicians had the highest (8.2). This finding may provide at least a partial mechanism for recent findings that similar quality of care is delivered by all usual PCP types, PCP teams deliver higher-quality chronic disease care, and PCP teams are more likely to provide the full range of primary care services.11,17,18,46

Patient characteristics and panel characteristics are associated with continuity of care. This study found that female sex and clinical complexity reduced continuity of care. In contrast, age and unreported race or race other than White or Black increased continuity of care. These findings are consistent with current literature looking at continuity of care over multiple years.24,30,47 Unique to this study are the findings related to panel-level team structural characteristics and their association with continuity of care. Reductions in continuity of care were seen with increases in PCP interdependence and increases in the number of supplemental providers serving the panel. This finding makes intuitive sense, as it is expected that sharing a higher density of patient care among more PCPs would reduce continuity of care. Given that PCPs are working in larger clinics and systems, this finding would suggest that creating formal PCP teams in each clinic may help mitigate a loss of continuity of care for patients.48,49 Although evidence is insufficient to make recommendations on panel size, perhaps more surprising is the finding that larger panel sizes with more visits increased continuity of care.50 We believe this finding may be due to clinicians who do not work in clinic full-time. Because PCPs who work more clinic hours will likely have more patients using our method, the larger panel size may actually be a marker for full-time practice—potentially explaining the improved continuity of care.

The best potential for improving practice from this study is determining which patients do best on which team structures. The lowest estimates for both less-complex and extremely complex patients with diabetes occurred on the largest panels, with a large number of supplemental providers seeing a large number of patients. Therefore, it is safe to say that such team structures may best be avoided if continuity of care with the usual PCP is a goal. Of note if the number of supplemental providers is reduced to one and PCP interdependence is set at the mean or maximum, both the less-complex (0.72 to 0.96) and more-complex patients (0.07 to 0.51) saw increases in continuity of care. These increases moved the less-complex patient, but not the most-complex patient, into a continuity of care range (that is, continuity of care greater than 0.6) associated with lower total patient charges and hospitalizations.22 Such changes in team structure have the potential to improve continuity of care while maintaining the benefits of multiple PCP teams.12,18,47,51 However, the most-complex patients did not achieve an estimated continuity of care over 0.6 until they were placed on a larger panel, with few supplemental providers. In the context of existing literature, it makes hypothesizing on cost and use difficult, because current literature has conflicting results for complex patients.22,52


This study has several key limitations. First, the sample is limited to a single academic physician group; therefore, our findings may not be generalizable to other primary care practices. However, the sample is from 26 primary care clinics that vary in size and staffing patterns as well as in the full range of geographic locations (rural, urban, suburban). Second, the study uses data from the EHR only. Therefore, important variables, such as professional and patient perceptions of the team, are not in the analysis. Although this may affect findings, evidence suggests that the objective measures of structure are the most salient for continuity of care.14,40 Similarly, the interdependence measure is a ratio, so a finding of 5 could mean that five patients on the panel received care from a single supplemental provider or 10 patients received care from two supplemental providers. To limit the effect of this issue and ensure a true density measure, we included the number of supplemental providers in the regression and postestimations. Finally, the study is limited to adults with diabetes who use primary care. Although this may not reflect all primary care patients, diabetes is prevalent, costly, and is an index condition that frequently is used to understand overall care delivery.51


In this study, a wide range of PCP interdependence is seen in 26 adult primary care clinics associated with an academic medical center. A range of patient and panel factors work together to affect continuity of care, making it critical to place patients on teams with structures that can achieve their goals. Higher PCP interdependence can improve continuity of care for less-complex patients with diabetes. However, lower PCP interdependence is needed to achieve adequate continuity of care for the most-complex patients. More research on understanding the effect of PCP interdependence on the full range of patient outcomes is needed to allow for evidence-based optimization of primary care teams.


1. Putera I. Redefining health: implication for value-based healthcare reform. Cureus. 2017;9(3):e1067.
2. Gray M. Value based healthcare. BMJ. 2017;356:j437.
3. Burke T. Accountable care organizations. Public Health Rep. 2011;126(6):875–878.
4. Rittenhouse DR, Shortell SM, Fisher ES. Primary care and accountable care—two essential elements of delivery-system reform. N Engl J Med. 2009;361(24):2301–2303.
5. Fisher ES. Building a medical neighborhood for the medical home. N Engl J Med. 2008;359(12):1202–1205.
6. Mitchell P, Wynia MK, Golden R, et al. Core principles and values of effective team-based health care. Institute of Medicine discussion paper. Washington, DC, 2012. https://nam.edu/wp-content/uploads/2015/06/VSRT-Team-Based-Care-Principles-Values.pdf. Accessed September 15, 2022.
7. Everett CM, Cawley J, Hooker R. Physician Assistants in Primary Care. Physician Assistants: Policy and Practice. Philadelphia PA: FA Davis; 2017.
8. Everett CM, Docherty SL, Matheson E, et al. Teaming up in primary care: membership boundaries, interdependence, and coordination. JAAPA. 2022;35(2):1–10.
9. Kiggundu MN. Task interdependence and the theory of job design. Acad Manage Rev. 1981;6(3):499–508.
10. Green LV, Savin S, Lu Y. Primary care physician shortages could be eliminated through use of teams, nonphysicians, and electronic communication. Health Aff (Millwood). 2013;32(1):11–19.
11. Pany MJ, Chen L, Sheridan B, Huckman RS. Provider teams outperform solo providers in managing chronic diseases and could improve the value of care. Health Aff (Millwood). 2021;40(3):435–444.
12. Kiggundu MN. Task interdependence and job design: test of a theory. Organ Behav Hum Perform. 1983;31(2):145–172.
13. Holman GT, Beasley JW, Karsh B-T, et al. The myth of standardized workflow in primary care. J Am Med Inform Assoc. 2016;23(1):29–37.
14. Starfield B. Primary Care: Balancing Healthy Needs, Services, and Technology. New York, NY: Oxford University Press; 1998.
15. Eaton P. Managing messages. Fam Pract Manag. 2012;19(5):25–29.
16. Everett CM, Thorpe CT, Palta M, et al. Division of primary care services between physicians, physician assistants, and nurse practitioners for older patients with diabetes. Med Care Res Rev. 2013;70(5):531–541.
17. Everett C, Thorpe C, Palta M, et al. Physician assistants and nurse practitioners perform effective roles on teams caring for medicare patients with diabetes. Health Aff (Millwood). 2013;32(11):1942–1948.
18. Jackson GL, Smith VA, Edelman D, et al. Intermediate diabetes outcomes by primary care provider type: are there differences among patients of physicians, nurse practitioners, and physician assistants. Ann Intern Med. 2018;169(12):825–835.
19. Klein S. The Veterans Health Administration: implementing patient-centered medical homes in the nation's largest integrated delivery system. Case study high-performing health care organization. www.commonwealthfund.org/publications/case-study/2011/sep/veterans-health-administration-implementing-patient-centered? Accessed October 20, 2022.
20. Leach B, Morgan P, Strand de Oliveira J, et al. Primary care multidisciplinary teams in practice: a qualitative study. BMC Fam Pract. 2017;18(1):115.
21. Mainous AG 3rd, Koopman RJ, Gill JM, et al. Relationship between continuity of care and diabetes control: evidence from the Third National Health and Nutrition Examination Survey. Am J Public Health. 2004;94(1):66–70.
22. Bazemore A, Petterson S, Peterson LE, et al. Higher primary care physician continuity is associated with lower costs and hospitalizations. Ann Fam Med. 2018;16(6):492–497.
23. Saultz JW. Defining and measuring interpersonal continuity of care. Ann Fam Med. 2003;1(3):134–143.
24. Haggerty JL, Reid RJ, Freeman GK, et al. Continuity of care: a multidisciplinary review. BMJ. 2003;327(7425):1219–1221.
25. Cabana MD, Jee SH. Does continuity of care improve patient outcomes. J Fam Pract. 2004;53(12):974–980.
26. Boyd C, Smith CD, Masoudi FA, et al. Decision making for older adults with multiple chronic conditions: executive summary for the American Geriatrics Society guiding principles on the care of older adults with multimorbidity. J Am Geriatr Soc. 2019;67(4):665–673.
27. Tinetti ME, McAvay G, Trentalange M, et al. Association between guideline recommended drugs and death in older adults with multiple chronic conditions: population based cohort study. BMJ. 2015;351:h4984.
28. Maciejewski ML, Powers BJ, Sanders LL, et al. The intersection of patient complexity, prescriber continuity and acute care utilization. J Gen Intern Med. 2014;29(4):594–601.
29. Rodriguez HP, Rogers WH, Marshall RE, Safran DG. Multidisciplinary primary care teams: effects on the quality of clinician-patient interactions and organizational features of care. Med Care. 2007;45(1):19–27.
30. Everett CM, Morgan P, Smith VA, et al. Interpersonal continuity of primary care of veterans with diabetes: a cohort study using electronic health record data. BMC Family Practice. 2018;19(1):132.
31. Nutting PA, Goodwin MA, Flocke SA, et al. Continuity of primary care: to whom does it matter and when. Ann Fam Med. 2003;1(3):149–155.
    32. Overland J, Yue DK, Mira M. Continuity of care in diabetes: to whom does it matter. Diabetes Res Clin Pract. 2001;52(1):55–61.
    33. Courtright SH, Thurgood GR, Stewart GL, Pierotti AJ. Structural interdependence in teams: an integrative framework and meta-analysis. J Appl Psychol. 2015;100(6):1825–1846.
    34. Wageman R. Interdependence and group effectiveness. Adm Sci Q. 1995;40(1):145–180.
    35. Van Der Vegt GS, Emans BJM, Van De Vliert E. Patterns of interdependence in work teams: a two-level investigation of the relations with job and team satisfaction. Personnel Psychol. 2001;54(1):51–69.
    36. Boyd CM, Darer J, Boult C, et al. Clinical practice guidelines and quality of care for older patients with multiple comorbid diseases: implications for pay for performance. JAMA. 2005;294(6):716–724.
    37. Pham HH, Schrag D, O'Malley AS, et al. Care patterns in Medicare and their implications for pay for performance. N Engl J Med. 2007;356(11):1130–1139.
    38. Østbye T, Yarnall KS, Krause KM, et al. Is there time for management of patients with chronic diseases in primary care? Ann Fam Med. 2005;3(3):209–214.
    39. Yarnall KS, Pollak KI, Østbye T, et al. Primary care: is there enough time for prevention? Am J Public Health. 2003;93(4):635–641.
    40. Lemieux-Charles L, McGuire WL. What do we know about health care team effectiveness? A review of the literature. Med Care Res Rev. 2006;63(3):263–300.
    41. Cohen SG, Bailey DE. What makes teams work: group effectiveness research from the shop floor to the executive suite. J Manage. 1997;23(3):239–290.
    42. Agency for Healthcare Research and Quality. 2019 national healthcare quality and disparities report. Rockville, MD, 2021.
    43. Hebert PL, Geiss LS, Tierney EF, et al. Identifying persons with diabetes using Medicare claims data. Am J Med Qual. 1999;14(6):270–277.
    44. Jee SH, Cabana MD. Indices for continuity of care: a systematic review of the literature. Med Care Res Rev. 2006;63(2):158–188.
    45. Pope GC, Kautter J, Ellis RP, et al. Risk adjustment of Medicare capitation payments using the CMS-HCC model. Health Care Financ Rev. 2004;25(4):119–141.
    46. Auerbach DI, Levy DE, Maramaldi P, et al. Optimal staffing models to care for frail older adults in primary care and geriatrics practices in the US. Health Aff (Millwood). 2021;40(9):1368–1376.
    47. Kristjansson E, Hogg W, Dahrouge S, et al. Predictors of relational continuity in primary care: patient, provider and practice factors. BMC Fam Pract. 2013;14(1):72.
    48. Barnes H, Martsolf GR, McHugh MD, Richards MR. Vertical integration and physician practice labor composition. Med Care Res Rev. 2022;79(1):46–57.
    49. Jones D, Furukawa M, Machta R, et al. Employment of advanced practice clinicians in vertically integrated health systems reflects larger trends towards consolidation. In: Mathmatica. Washington, DC: Agency for Healthcare Research and Quality; 2021.
    50. Paige NM, Apaydin EA, Goldhaber-Fiebert JD, et al. What is the optimal primary care panel size. Ann Intern Med. 2020;172(3):195–201.
    51. Nolte E, Bain C, McKee M. Diabetes as a tracer condition in international benchmarking of health systems. Diabetes Care. 2006;29(5):1007–1011.
    52. Morgan PA, Smith VA, Berkowitz TSZ, et al. Impact of physicians, nurse practitioners, and physician assistants on utilization and costs for complex patients. Health Aff (Millwood). 2019;38(6):1028–1036.

    primary care; interdependence; continuity of care; complex patients; PAs; NPs

    Copyright © 2023 American Academy of Physician Associates