Healthcare 2 (2014) 232–237
Contents lists available at ScienceDirect
Healthcare journal homepage: www.elsevier.com/locate/hjdsi
Time spent with patients by physicians, nurse practitioners, and physician assistants in community health centers, 2006–2010 Perri Morgan a,n, Christine M. Everett a, Esther Hing a,b a b
Department of Community and Family Medicine, Duke University Medical Center, Durham, NC 27710, USA National Center for Health Statistics, Centers for Disease Control, USA
art ic l e i nf o
a b s t r a c t
Article history: Received 25 April 2014 Received in revised form 14 August 2014 Accepted 25 September 2014 Available online 31 October 2014
Background: As health systems struggle to meet access, cost and quality goals in the setting of increased demand, nurse practitioners (NPs) and physician assistants (PAs) are expected to help meet the need for care. The amount of time spent with each patient can affect the clinical productivity, quality of care, and satisfaction of patients and clinicians. This paper compares time spent per patient in community health centers by whether the provider is a physician, NP, or PA. Methods: This paper uses National Ambulatory Medical Care Survey (NAMCS) Community Health Center (CHC) data from 2006–2010. The NAMCS CHC strata is a national sample of CHCs, providers within CHCs, and patient visits to CHCs. Provider characteristics and variables related to time spent with patients across provider types were compared using t tests and chi square tests of association. Multivariate linear regression analysis was used to compare time spent with patients, controlling for patient and visit characteristics. Results: There were no differences in the number of visits by provider type, but PAs saw patients for a slightly larger portion of the week (3.8 days) than did physicians (3.5 days, po0.05) or NPs (3.4 days, po 0.05). There were no statistical differences in the mean time spent per patient in the crude and adjusted analyses. Conclusions: Time spent per patient in CHCs is similar for physicians, NPs and PAs. This information may be useful to planners concerned with health system capacity and cost efficiency, and has implications for patient and provider satisfaction. & 2014 Elsevier Inc. All rights reserved.
Keywords: Physician assistants Nurse practitioners Community health centers Clinical productivity Efficiency
1. Background The “triple aim” for improvement of the US healthcare system calls for improving patient outcomes and the health of populations while simultaneously reducing per capita healthcare costs.1 New models of care, such as the patient-centered medical home model, strive to achieve multiple concurrent goals.2 In these new models, improvements in both clinical outcomes and the patient experience are integral. Improving the health of populations requires assuring timely access to care. At the same time that healthcare organizations are expected to meet these new goals, many experts predict that system capacity may not be adequate.3,4 Concern about whether the supply of physicians will be adequate to meet the need for services is especially acute in primary care, a sector that suffers from lower salaries, high burnout, and low prestige.4 As health systems struggle to meet diverse goals1 – broad access to care, improved clinical quality, high patient satisfaction, cost
n
Corresponding author. E-mail address: [email protected] (P. Morgan).
http://dx.doi.org/10.1016/j.hjdsi.2014.09.009 2213-0764/& 2014 Elsevier Inc. All rights reserved.
control, and an attractive work climate – and the nation expects a surge in demand for care,5 the supply of physicians might be inadequate unless their labor is bolstered by other professionals.6– 8 Care by nurse practitioners (NPs) and physician assistants (PAs) can achieve high quality and patient satisfaction, and could help expand access and control costs.9–11 Questions remain, however, about the use of NPs and PAs, including how their clinical productivity compares to that of physicians, and how this clinical productivity impacts organizational efficiency and system capacity. How many primary care patients can an NP or PA care for, compared to a physician? The Negotiated Rule Committee established through the Affordable Care Act (ACA) to update guidelines for designating health professional shortage areas (HPSAs) recommended to the Administrator of the Health Resources and Services Administration in 2011 that each NP, PA, or clinical nurse midwife (CNM) be counted as 75% of a physician for purposes of determining HPSA status. The committee acknowledged, however, that the evidence for this recommendation was inadequate.12 The amount of time that providers spend caring for each patient is an important component of a clinical productivity
P. Morgan et al. / Healthcare 2 (2014) 232–237
assessment. With clinical productivity defined as the amount of output per unit input, time spent with the patient is central to characterizing the denominator of this equation. The amount of time that providers spend with patients also impacts their ability to fill the need for health services. Spending less time per patient might improve organizational efficiency, since more patients would be seen per unit of labor cost. If the goal is expanding access to care, providers who can see patients more quickly may be able to extend services to more patients. On the other hand, spending more time with patients has been associated with small to modest improvements in patient satisfaction.13–16 Adequate time with patients could allay provider frustration with the often hectic pace of primary care and improve provider satisfaction.17,18 The amount of time that providers spend with patients may affect the type and quality of care that they provide.19 Some types of care, such as preventive services and patient education, may require more time.19 Chronic disease care outcomes have been shown to be improved by positive provider–patient communication,20 and this communication may require more time.21 Finally, the attractiveness of a health profession or of primary care specialty practice to potential entrants may be affected by perceptions of whether time with patients will be rushed.22 For example, PA school applicants frequently cite their perception that they will be able to spend more time with patients as a PA than as a physician as a reason for choosing a career as a PA.22 To our knowledge, concrete support for this perception is lacking. Spending more time with patients, then, could both help and hinder attainment of health system goals. Healthcare organizations should consciously construct teams to meet their specific goals and to balance the advantages and disadvantages of variations in team composition.23 As organizations include more NPs and PAs in their staffing mix, it would be helpful to know how the amount of time that they spend with patients compares to that of physicians. Community health centers (CHCs) provide care for vulnerable populations and are an important component of the healthcare safety net. CHCs have expanded over the past decade, and are expected to continue to expand to meet the need for care among the newly insured as the ACA is implemented.24 Because these centers rely on staffing patterns that employ a large proportion of NPs and PAs, compared to physicians25, they provide an instructive setting for study of the practice characteristics of NPs and PAs. This paper uses national data from community health centers (CHCs) to compare time spent with patients by whether the visit provider was a physician, NP, or PA.
2. Methods This study uses 2006–2010 data from the National Ambulatory Medical Care Survey (NAMCS) Community Health Center Stratum.26 The NAMCS is designed to reflect physician practice in the U.S. and uses a three stage probability sample based on geographic primary sampling units (PSUs), physician practices within the PSUs, and patient visits within physician practices. The NAMCS CHC sample draws CHCs from a health center roster and takes representative samples of physicians, NPs and PAs within those centers. This CHC stratum was added to the NAMCS family of surveys in 2006 and consists of approximately 104 CHCs per year. Sampled CHCs include Federally Qualified Health Center (FQHC) clinics that receive Section 330 grants under the Public Health Service Act, “look-alike” health centers that meet FQHC requirements, and federally-qualified Indian Health Service clinics. Separate response rates taking CHC participation and visit response into account were computed by provider type. Among CHC nonphysician providers (NPs, PAs, and CNMs), the combined response
233
rate was 86.6%; among CHC physicians, it was 85.5% (unpublished calculations by authors). Our study includes only visits to Section 330 grantees and “look-alike” CHCs.27 Our study included 670 physicians, 245 NPs, and 103 PAs, as well as the 24,528 visits that patients made to these providers in CHCs from 2006–2010. The 2010 data are the most recent data available. All sampled providers were asked to complete a provider induction survey and 30 patient visit forms for a randomly selected sample of patients over a randomly selected one-week period. Our estimates are based on the provider type who actually saw the patient. For some patients, time spent with the patient was recorded for two provider types, suggesting that two providers saw the patient. These visits were rare (0.4% of visits) and were excluded from our analysis. The nonresponse rate for most questions was less than 5%. Exceptions were race (20.9%), ethnicity (16.4%), and number of past visits during the previous 12 months (10.5%). All of these variables, however, were imputed by National Center for Health Statistics analysts and used in our analysis. Imputation was accomplished by randomly assigning a value from another record with similar characteristics. Imputations, in general, were based on physician specialty, geographic region, and diagnosis codes.28 Sampling weights were used to obtain national estimates for all analyses. To adjust for the complex sample design, standard errors were obtained using Taylor-series approximation with SUDAAN software. Because CNMs constituted a very small portion of our sample (1.7%), we included them in the NP category. When possible, we analyzed PAs separately from NPs. For the provider demographics and trend analyses, NP or PA sample size limitations frequently produced unreliable estimates, so we combined NPs and PAs into a single group for analysis. However, we only combined NPs and PAs into a single analytic category after first confirming that both were more similar to each other than either as a separate category was similar to physicians with regard to the attribute being analyzed. We used a weighted least-squares regression analysis to determine the significance of trends in numbers and percent of each provider type by year.29 All other analyses combined all five years of data. We compared provider characteristics and time with patient variables across provider types using t tests and chi square tests of association. We performed linear regression on the time spent with provider, adjusting for the patient and visit factors that might affect the amount of time required for a visit. We adjusted for patient demographic and socioeconomic factors, reason for visit, visit complexity (measured by number of chronic conditions of the patient and number of services provided at the visit), factors indicating the patient's relationship to the clinic and to the provider (whether the provider is the patient's primary care provider, whether the patient is new to the clinic, number of times the patient has been seen in the clinic during the previous twelve months), and timing (study year and time of year). This regression analysis was performed at the visit level and included 21,125 patient encounters. Because our previous work42 with this dataset showed that patient visit attributes for PAs were often more similar to those of physicians than to those of NPs, we did not combine PA and NP visits for the visit level regression analysis. We used a significance level of p o0.05 for all analyses. The Duke University Medical Center Institutional Review Board declared this research exempt from full review.
3. Results When taking all five years together, physicians constituted 69% of CHC providers, NPs 21%, and PAs 10%. Due to sample size limitations, separate annual estimates for NPs and PAs are not
234
P. Morgan et al. / Healthcare 2 (2014) 232–237
shown, but NPs were present at approximately twice the rate of PAs throughout the time period studied. The estimated number of CHC providers (physicians, NPs, and PAs) increased from 2006– 2009 and then decreased in 2010 (Fig. 1). Trend analyses showed a statistically significant increase in total providers (p o0.01) and in NPs/PAs (combined, p o0.01) over the five years studied, but physician numbers did not change significantly (p¼ 0.10). Fig. 2 shows the estimates for the proportion of providers who were physicians compared to the proportion who were NPs/PAs (combined). By 2010, NPs and PAs together comprised 35% of CHC providers. Trend tests did not show significant differences in the proportions by provider type (p ¼0.21). Provider-level variables are shown in Table 1. Physicians tended to be younger (p o0.05), with more physicians in the 35–44 year old age group, and more NPs and PAs in the 45–54 year old age group. Sex differences by provider type were both substantial and statistically significant, with 83% of NPs and PAs compared to 45% of physicians in CHCs being female. Physicians were less common in rural CHCs (11% of CHC physicians), compared to NPs and PAs (30% of CHC NPs/PAs)(p o0.01). NPs and PAs were more often nonHispanic whites than were physicians (p o0.01). Regional distributions did not vary by provider type. During the reporting week, PAs saw patients for a slightly – but statistically significant – larger portion of the week (3.8 days) than physicians (3.5 days) or NPs (3.4 days). There was not a statistical difference in the number of patients seen by provider type when the analysis unit was the entire reporting week, but when we adjusted for the number of days spent seeing patients, PAs saw fewer patients per clinical care day than did physicians (14.8 patients per day for physicians and 12.9 for PAs, p o0.05). The number of patients seen per day spent in patient care for NPs was intermediate at 14.2, but this number was not statistically different
Number of providers
20000
Estimated number of providers in CHCs, 2006-2010
15000 Total
10000
Physician NP+PA
5000 0
2006
2007
2008
2009
2010
Fig. 1. Trend test: p o0.01 for all visits, p¼ 0.10 for physicians, p o 0.01 for NPsþ Pas. Source: National Ambulatory Medical Care Survey.
Estimated Percent of Community Health Center Providers, 2006-2010 28
26
30
33
Table 1 Provider characteristics of physicians, nurse practitioners, or physician assistants in Community Health Centers, 2006–2010. Source: National Ambulatory Medical Care Survey. Provider type
p values
Physicians
NPs þPAs
o 35 35–44 45–54 55–64 65 or over Sex (%)
15 31 24 22 8
16 20 35 22 7
Male Female MSA (%)
55 45
17 83
Urban Rural Region (%)
89 11
70 30
Northeast Midwest South West Race/ethnicity (%)
28 17 25 31
26 18 27 28
62 11 9 18
79 8 8 a, 5
Physician vs. NP/PA
Age group (%) o 0.05
o 0.01
o 0.01
0.93
o 0.01 Non-Hispanic white Non-Hispanic black Hispanic Non-Hispanic other
a The estimates for NPs and PAs in this category had large standard error with relative standard error40.3.
than that for physicians or PAs. In this unadjusted analysis (Table 2), there was not a difference in the mean amount of time (varied between 18.6 and 20 minutes) that the provider types spent on each patient. In the adjusted linear regression analysis (Table 3), the mean time spent with patients was 19.82 minutes for physicians, 19.96 minutes for NPs, and 20.15 minutes for PAs. These differences are very small (all within 20 seconds per visit) and not statistically significant. Patient factors that reduced the mean time spent were being under 18 years of age, being seen in 2009 (compared to 2010), having Medicare or Medicaid/SCHIP insurance (compared to private insurance) and having been seen in the clinic before. Patient factors that increased the mean time spent were being seen for a flare-up of a chronic condition (vs. for preventive care), receiving more health services, receiving more health education services, being seen by a provider who considers themselves the patient's primary care provider (vs. not knowing whether they are the primary care provider), and being seen for mental healthcare, prevention or pregnancy, or ill-defined or vision and hearing symptoms. Full regression analysis results are shown in Appendix 1.
35 4. Discussion
% 72
74
2006
2007
70
2008 Physician
67
65
2009
2010
NP+PA
Fig. 2. Trend test: p¼ 0.21. Source: National Ambulatory Medical Care Survey.
NPs and PAs comprised a substantial portion (31%) of the providers in CHCs during the time period studied. The increased tendency that we found for NPs/PAs to work in rural CHCs supports other research that found NPs and PAs particularly contribute to access to care in high need rural areas.30–32 Overall, measures of time spent with patients were remarkably similar across provider types in both crude and adjusted analyses. PAs spent two to three more hours during the reporting week
P. Morgan et al. / Healthcare 2 (2014) 232–237
235
Table 2 Comparison of time spent in patient care and time spent per patient by provider type, Community Health centers, 2006–2010. Source: National Ambulatory Medical Care Survey. Provider type
Mean number of visits in reporting week Mean number of days patients seen in reporting week Number of visits per day (calculated) Mean time spent per patient (minutes)
P values
Physician
NP
PA
Physician vs. NP
Physician vs. PA
NP vs. PA
53.3 3.5 14.8 18.6
48.7 3.4 14.2 19
50 3.8 12.9 20
0.22 0.47 0.48 0.53
0.46 o 0.05 o 0.05 0.06
0.7752 o0.05 0.13 0.26
Number of visits per day was calculated by dividing the mean number of visits by the mean number of days patients were seen.
Table 3 Time spent with patients by provider type, adjusted. Source: National Ambulatory Medical Care Survey. Provider type
Mean time spent per patient (minutes)
P value for difference from physician time
Physician NP PA
19.82 19.96 20.15
– 0.62 0.75
Linear regression analysis was used to adjust for patient demographic (age group, sex, race/ethnicity, rural/urban status), socioeconomic (expected payment source, county median household income) and medical complexity (number of chronic conditions) factors, visit (major reason for visit, number of services provided, number of medications prescribed, number of health education/prevention services provided, and type of care provided according to Fenton categories based on visit ICD-9 codes47), patient relationship to the clinic (was patient new to the clinic, number of times patient was seen in the past year, whether the provider considers him/herself the patients' primary care provider), and survey administration (survey year, season of the year) variables.
providing patient care. Despite this finding, PAs did not see more patients or spend more time within a visit. This finding could suggest that PAs spend more time on their patient care days in activities other than seeing patients who would be reported on the NAMCS patient encounter form, such as communicating with patients who are not in the clinic. The time allowed for patient visits is often established by clinic scheduling procedures that establish a standard time allocation for each visit. If physicians, NPs and PAs are all given the same fixed amount of time to see each patient, this could influence the amount of time that providers spend with a patient.33 Visits for returning patients are often scheduled for 15 minute or 20 minute time slots. Other factors can affect the time actually spent, including longer appointment blocks that some clinics use for new patients and those scheduled for yearly check-ups or complex visits. Still, the results of our study and others19,33 find that visit length for all provider types averages very near twenty minutes per patient. When patient needs require additional time, providers typically schedule return appointments to address those needs. Assessment of the relative amount of time required to care for patients by provider type would ideally include the rate of return visits.34 Few studies have examined total office visit use in relation to provider type, but those that have done so have not found increased office visit resource use with nonphysician care35–37. The relative increase that we found in use of NPs and PAs in CHCs is similar to that seen in other reports on CHCs38 and in other settings39. The large increase in total providers in 2009 can be attributed to funding for CHC expansion as part of the American Recovery and Reinvestment Act40, and the subsequent decrease in 2010 was likely caused by uncertainty about whether increased levels of funding would be sustained41. The relatively large decrease in mean time spent per patient (about 1.7 minutes) in 2009 might reflect the fact that CHC provider growth between 2008 and 2009 (39%) did not keep pace with the increase in
number of visits to CHCs (57%).42 For CHCs providers to handle the large increase in demand for patient visits to safety net providers, caused presumably by widespread job losses during the economic recession,43 they likely had to decrease the amount of time they spent with each patient. The CHC population is younger and more socially and medically complex than the general US population,44,45 which limits generalizability of our results. A recent analysis of primary care visits by Bruen and colleagues using NAMCS data found that care of CHC patients was similar to that of patients in other settings with regard to content of care and time spent per visit.33 Like the Bruen group, we found that the major determinants of time spent with patients were their health needs and the type and quantity of care provided. The organization of care and characteristics of providers are different in CHCs than in the private sector, and these differences might also limit generalizability of our findings. Study limitations include that data were obtained by provider or clinic staff report rather than through direct observation. Other studies have found that physicians overestimate the time spent with patients,46 but we have no reason to suspect that such overestimation would vary by whether the provider is a physician, NP, or PA. Since a large number of services can be reported in NAMCS, we used summary measures of services ordered or provided (total health services, total health education services, total other services). However, the services tallied for these summary measures varied widely in complexity and in the amount of time that would be required. For example, items in the total health services summary measure ranged from simple activities such as taking a temperature to more complex activities such as obtaining and interpreting an electrocardiogram. Similarly, we could not distinguish between brief or extended preventive counseling. Although we attempted to adjust our analysis for patient complexity, residual confounding by complexity may remain.
5. Conclusions Physicians, NPs, and PAs seem to spend similar amounts of time —about 20 minutes‐‐ with each patient in CHCs. This allocation of time is similar to that seen in other settings and may be driven by clinic scheduling practices. Information about the relative amount of time that physicians, NPs, and PAs spend with patients has implications for workforce planning to meet goals of organizational efficiency and increased health system capacity, but may also affect patient health outcomes and patient and provider satisfaction.
Disclaimer The findings and conclusions in this paper are those of the authors and do not necessarily represent the views of CDC.
236
P. Morgan et al. / Healthcare 2 (2014) 232–237
Conflict of interest disclosure statement This statement accompanies the article “Time Spent with Patients by Physicians, Nurse Practitioners, and Physician Assistants in Community Health Centers, 2006–2010,” authored by Perri Morgan, Christine Everett, and Esther Hing and submitted to Healthcare as an original article. Below all authors have disclosed relevant commercial associations that might pose a conflict of interest: Consultant arrangements: None. Stock/other equity ownership: None. Patent licensing arrangements: None. Grants/ research support: None. Employment: None. Speakers' bureau: None Expert witness: None. Other: None.
Acknowledgement This work was supported in part by an unrestricted grant from the Physician Assistant Education Association.
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.hjdsi.2014.09.009. References 1. Berwick D, Nolan T, Whittington J. The triple aim: care, health, and cost. Health Aff (Millwood). 2008;27(3):11. 2. Jackson G, Powers B, Chatergee R, et al. The patient-centered medical home: a systematic review. Ann. Intern Med. 2013;158(3):169–178 (February). 3. Association of American Medical Colleges. The Impact of Health Care Reform on the Future Supply and Demand for Physicians: Updated Projections Through 2025. 2011; 〈https://www.aamc.org/download/158076/data/updated_projec tions_through_2025.pdf〉. 4. Council on Graduate Medical Education. Twentieth Report: Advancing Primary Care. 2010; 〈http://www.hrsa.gov/advisorycommittees/bhpradvisory/cogme/ Reports/〉. Accessed May 10, 2013. 5. Keehan SP, Sisko AM, Truffer CJ, et al. National health spending projections through 2020: economic recovery and reform drive faster spending growth. Health Aff (Millwood). 2011;30(8):1594–1605 (Aug). 6. Gottlieb S, No Emanuel EJ. There won't be a doctor shortage. New York Times; 2013 NY, NY (December 4). 7. Auerbach DI, Chen PG, Friedberg MW, et al. Nurse-managed health centers and patient-centered medical homes could mitigate expected primary care physician shortage. Health Aff (Millwood). 2013;32(11):1933–1941 (November 1). 8. Shipman S, Sinsky C. Expanding primary care capacity by reducing waste and improving the efficiency of care. Health Aff (Millwood). 2013;32(11):1990–1997 (November 1). 9. Laurant M, Harmsen M, Wollersheim H, Grol R, Faber M, Sibbald B. The impact of nonphysician clinicians: do they improve the quality and cost-effectiveness of health care services? Med Care Res Rev. 2009;66(6 Suppl):S36–S89 (Dec). 10. Newhouse RP, Stanik-Hutt J, White KM, et al. Advanced Practice Nurse Outcomes 1990-2008: a systematic review. Nurs Econ. 2011;29:5. 11. Cronenwett L, Dzau V. Who Will Provide Primary Care and How Will They Be Trained?, Proceedings of a Conference Sponsored by the Josiah Macy, Jr. Foundation. Josiah Macy, Jr. Foundation; 2010: 〈http://www.josiahmacyfounda tion.org/documents/JMF_PrimaryCare_Monograph.pdf〉. Accessed May 26, 2010. 12. Negotiated Rulemaking Committee on the Designation of Medically Underserved Populations and Health Professional Shortage Areas. Negotiated Rulemaking Committee on the Designation of Medically Underserved Populations and Health Professional Shortage Areas Report: Appendices and Addenda. October 31, 2011; 〈http://www.hrsa.gov/advisorycommittees/shortage/〉. Accessed October 23, 2013. 13. Lin CT, Albertson GA, Schilling LM, et al. Is patients' perception of time spent with the physician a determinant of ambulatory patient satisfaction? Arch Intern Med. 2001;161(11):1437–1442. 14. Landau D-A, Bachner YG, Elishkewitz K, Goldstein L, Barneboim E. Patients' views on optimal visit length in primary care. J Med Pract Manage. 2007;23 (1):12–15 (Jul–Aug). 15. Otani K, Kurz RS, Harris LE. Managing primary care using patient satisfaction measures. (Sep–Oct). J Healthc Manag. 2005;50(5):311–324 (discussion 324315). 16. Wood GC, Spahr R, Gerdes J, Daar ZS, Hutchison R, Stewart WF. Patient satisfaction and physician productivity: complementary or mutually exclusive? Am J Med Qual. 2009;24(6):498–504 (Nov–Dec).
17. Chen PG, Curry LA, Nunez-Smith M, Bradley EH, Desai MM. Career satisfaction in primary care: a comparison of international and US medical graduates. J Gen Intern Med. 2012;27(2):147–152 (Feb). 18. Linzer M, Konrad TR, Douglas J, et al. Managed care, time pressure, and physician job satisfaction: results from the physician worklife study. J Gen Intern Med. 2000;15(7):441–450 (Jul). 19. Chen LM, Farwell WR, Jha AK. Primary care visit duration and quality: does good care take longer? Arch Intern Med. 2009;169(20):1866–1872 (Nov 9). 20. Stewart MA. Effective physician–patient communication and health outcomes: a review. CMAJ Can Med Assoc J. 1995;152(9):1423–1433 (May 1). 21. Howie JG, Porter AM, JF. F.. Quality and the use of time in general practice: widening the discussion. BMJ.. 1989;15(298):1008–1010 (April 15). 22. Strand J. The myth of more time. Assoc Physician Assist Progr (APAP) Update. 2004 (March). 23. Everett C, Thorpe C, Palta M, Carayon P, Gilchrist V, Teams Smith M. Roles, and goals: primary care physician assistant and nurse practitioner roles and outcomes for older patients with diabetes. Health Aff (Millwood). 2013;32 (11):1942–1948. 24. Adashi EY, Geiger HJ, Fine MD. Health care reform and primary care—the growing importance of the community health center. N Engl J Med. 2010;362 (22):2047–2050 (Jun 3). 25. Hing E, Hooker RS, Ashman JJ. Primary health care in community health centers and comparison with office-based practice. J. Community Health. 2011;36 (3):406–413 (Jun). 26. Centers for Disease Prevention and Control. Public Use Data File Documentation: 2006–2010 National Ambulatory Medical Care Survey.. 〈http://www.cdc. gov/nchs/ahcd.htm〉. Accessed September 16, 2013. 27. Indian Health Service (IHS) clinics were not included. Since IHS clinics have markedly different staffing patterns, inclusion of these clinics would have skewed our results. The IHS clinic sample was quite small compared to the Section 330 grantees and the lookalike clinics. 28. National Center for Health Statistics. Public Use microdata file documentation: National Ambulatory Medical Care Survey, 2006–2010. 2012; 〈ftp://ftp.cdc.gov/ pub/Health_Statistics/NCHS/Dataset_Documentation/NAMCS/〉. Accessed November 11, 2013. 29. Sirkem MG, Shimizu I, French DK, DB B. Manual on Standards and Procedures for Reviewing Statistical Reports. Hyattsville, MD: National Center for Health Statistics; 1990. 30. Skillman SM, Fordyce MA, Yen W, Mounts T. Washington State Primary Care Provider Survey, 2011-2012: Summary of Findings. Seattle, WA: WWAMI Center for Health Workforce Studies; 2012. 31. Grumbach K, Hart LG, Mertz E, Coffman J, Palazzo L. Who is caring for the underserved? A comparison of primary care physicians and nonphysician clinicians in California and Washington Ann of Family Med. 2003;1(2):97–104. 32. Dehn RW. The distribution of physicians, advance practice nurses, and physician assistants in Iowa. J Physician Assist Educ. 2006;17(1):36–39. 33. Bruen BK, Ku L, Lu X, Shin P. No evidence that primary care physicians offer less care to medicaid, community health center, or uninsured patients. Health Aff (Millwood). 2013;32(9):1624–1630. 34. Fishman PA, Hornbrook MC. Assigning resources to health care use for health services research: options and consequences. Med. Care. 2009;47(7 Suppl 1): S70–S75 (Jul). 35. Hooker RS. A cost analysis of physician assistants in primary care. J Am Acad Physician Assist. 2002;15(11):39–42 (45, 48 passim, 2002 Nov). 36. Morgan PA, Shah ND, Kaufman JS, Albanese MA. Impact of physician assistant care on office visit resource use in the United States. Health Serv Res. 2008;43(5 Pt 2):1906–1922 (Oct). 37. Lenz ER, MOM RL, Kane SC, Hopkins, Lin SX. Primary care outcomes in patients treated by nurse practitioners or physicians: two-year follow-up. Med Care Res Rev. 2004;61(3):332–351 (September 1, 2004). 38. National Association of Community Health Centers. Expanding Access to Primary Care: The Role of Nurse Practitioners, Physician Assistants, and Certified Nurse Midwives in the Health Center Workforce. Fact Sheet # 0913, 2013; 〈http://www.nachc.com/client/documents/Workforce_FS_0913.pdf〉. 39. Morgan P, Abbott D, McNeill R, Fisher D. Characteristics of Primary Care Office Visits to Nurse Practitioners, Physician Assistants, and Physicians in United States Veterans Health Administration Facilities, 2006-2010: a retrospective cross-sectional analysis. Hum Resour Health. 2012;10:42. 40. National Association of Community Health Centers. More Patients Gain Access To Health Center Care Thanks to Stimulus Funds. 2010; 〈http://www.nachc.org/ client/ARRA%20Fact%20Sheet%20Final%20revised.pdf〉. Accessed December 16, 2010. 41. Morris A. FY 2011 Appropriations Reaches Conclusion; Health Centers Slated for $600 Million Cut. Nat Assoc Community Health Cent blog. 2011 (Accessed October 25, 2013). 42. Morgan PA, Everett CM, Hing E. Nurse Practitioners, Physician Assistants, and Physicians in Community Health Centers, 2006–2010. Health Care, in press, http://dx.doi.org/10.1016/j.hjdsi.2014.06.002. 43. National Association of Community Health Centers. Recession Brings More Patients to Community Health Centers. Fact Sheet #0209 2009; 〈https://www. nachc.com/client/documents/rising_patient_%20demand_093.pdf〉. Accessed January 17, 2014. 44. Harrison B, Rittenhouse D, Phillips Jr. RL, Grumbach K, Bazemore AW, Dodoo MS. Title VII. is criticial to the community health center and national health service corps workforce. Am Fam Physician.. 2010;81(2): (132-132).
P. Morgan et al. / Healthcare 2 (2014) 232–237
45. National Association of Community Health Centers. Community Health Centers: The Local Prescription for Better Quality and Lower Costs. 2011; 〈http:// www.nachc.com/client/A%20Local%20Prescription%20Final%20brief%203% 2022%2011.pdf〉. Accessed July 15, 2013.
237
46. Gilchrist VJ, Stange KC, Flocke SA, McCord G, Bourguet CC. A comparison of the national ambulatory medical care survey (NAMCS) measurement approach with direct observation of outpatient visits. Med Care. 2004;42(3):276–280.
Contents lists available at ScienceDirect
Healthcare journal homepage: www.elsevier.com/locate/hjdsi
Time spent with patients by physicians, nurse practitioners, and physician assistants in community health centers, 2006–2010 Perri Morgan a,n, Christine M. Everett a, Esther Hing a,b a b
Department of Community and Family Medicine, Duke University Medical Center, Durham, NC 27710, USA National Center for Health Statistics, Centers for Disease Control, USA
art ic l e i nf o
a b s t r a c t
Article history: Received 25 April 2014 Received in revised form 14 August 2014 Accepted 25 September 2014 Available online 31 October 2014
Background: As health systems struggle to meet access, cost and quality goals in the setting of increased demand, nurse practitioners (NPs) and physician assistants (PAs) are expected to help meet the need for care. The amount of time spent with each patient can affect the clinical productivity, quality of care, and satisfaction of patients and clinicians. This paper compares time spent per patient in community health centers by whether the provider is a physician, NP, or PA. Methods: This paper uses National Ambulatory Medical Care Survey (NAMCS) Community Health Center (CHC) data from 2006–2010. The NAMCS CHC strata is a national sample of CHCs, providers within CHCs, and patient visits to CHCs. Provider characteristics and variables related to time spent with patients across provider types were compared using t tests and chi square tests of association. Multivariate linear regression analysis was used to compare time spent with patients, controlling for patient and visit characteristics. Results: There were no differences in the number of visits by provider type, but PAs saw patients for a slightly larger portion of the week (3.8 days) than did physicians (3.5 days, po0.05) or NPs (3.4 days, po 0.05). There were no statistical differences in the mean time spent per patient in the crude and adjusted analyses. Conclusions: Time spent per patient in CHCs is similar for physicians, NPs and PAs. This information may be useful to planners concerned with health system capacity and cost efficiency, and has implications for patient and provider satisfaction. & 2014 Elsevier Inc. All rights reserved.
Keywords: Physician assistants Nurse practitioners Community health centers Clinical productivity Efficiency
1. Background The “triple aim” for improvement of the US healthcare system calls for improving patient outcomes and the health of populations while simultaneously reducing per capita healthcare costs.1 New models of care, such as the patient-centered medical home model, strive to achieve multiple concurrent goals.2 In these new models, improvements in both clinical outcomes and the patient experience are integral. Improving the health of populations requires assuring timely access to care. At the same time that healthcare organizations are expected to meet these new goals, many experts predict that system capacity may not be adequate.3,4 Concern about whether the supply of physicians will be adequate to meet the need for services is especially acute in primary care, a sector that suffers from lower salaries, high burnout, and low prestige.4 As health systems struggle to meet diverse goals1 – broad access to care, improved clinical quality, high patient satisfaction, cost
n
Corresponding author. E-mail address: [email protected] (P. Morgan).
http://dx.doi.org/10.1016/j.hjdsi.2014.09.009 2213-0764/& 2014 Elsevier Inc. All rights reserved.
control, and an attractive work climate – and the nation expects a surge in demand for care,5 the supply of physicians might be inadequate unless their labor is bolstered by other professionals.6– 8 Care by nurse practitioners (NPs) and physician assistants (PAs) can achieve high quality and patient satisfaction, and could help expand access and control costs.9–11 Questions remain, however, about the use of NPs and PAs, including how their clinical productivity compares to that of physicians, and how this clinical productivity impacts organizational efficiency and system capacity. How many primary care patients can an NP or PA care for, compared to a physician? The Negotiated Rule Committee established through the Affordable Care Act (ACA) to update guidelines for designating health professional shortage areas (HPSAs) recommended to the Administrator of the Health Resources and Services Administration in 2011 that each NP, PA, or clinical nurse midwife (CNM) be counted as 75% of a physician for purposes of determining HPSA status. The committee acknowledged, however, that the evidence for this recommendation was inadequate.12 The amount of time that providers spend caring for each patient is an important component of a clinical productivity
P. Morgan et al. / Healthcare 2 (2014) 232–237
assessment. With clinical productivity defined as the amount of output per unit input, time spent with the patient is central to characterizing the denominator of this equation. The amount of time that providers spend with patients also impacts their ability to fill the need for health services. Spending less time per patient might improve organizational efficiency, since more patients would be seen per unit of labor cost. If the goal is expanding access to care, providers who can see patients more quickly may be able to extend services to more patients. On the other hand, spending more time with patients has been associated with small to modest improvements in patient satisfaction.13–16 Adequate time with patients could allay provider frustration with the often hectic pace of primary care and improve provider satisfaction.17,18 The amount of time that providers spend with patients may affect the type and quality of care that they provide.19 Some types of care, such as preventive services and patient education, may require more time.19 Chronic disease care outcomes have been shown to be improved by positive provider–patient communication,20 and this communication may require more time.21 Finally, the attractiveness of a health profession or of primary care specialty practice to potential entrants may be affected by perceptions of whether time with patients will be rushed.22 For example, PA school applicants frequently cite their perception that they will be able to spend more time with patients as a PA than as a physician as a reason for choosing a career as a PA.22 To our knowledge, concrete support for this perception is lacking. Spending more time with patients, then, could both help and hinder attainment of health system goals. Healthcare organizations should consciously construct teams to meet their specific goals and to balance the advantages and disadvantages of variations in team composition.23 As organizations include more NPs and PAs in their staffing mix, it would be helpful to know how the amount of time that they spend with patients compares to that of physicians. Community health centers (CHCs) provide care for vulnerable populations and are an important component of the healthcare safety net. CHCs have expanded over the past decade, and are expected to continue to expand to meet the need for care among the newly insured as the ACA is implemented.24 Because these centers rely on staffing patterns that employ a large proportion of NPs and PAs, compared to physicians25, they provide an instructive setting for study of the practice characteristics of NPs and PAs. This paper uses national data from community health centers (CHCs) to compare time spent with patients by whether the visit provider was a physician, NP, or PA.
2. Methods This study uses 2006–2010 data from the National Ambulatory Medical Care Survey (NAMCS) Community Health Center Stratum.26 The NAMCS is designed to reflect physician practice in the U.S. and uses a three stage probability sample based on geographic primary sampling units (PSUs), physician practices within the PSUs, and patient visits within physician practices. The NAMCS CHC sample draws CHCs from a health center roster and takes representative samples of physicians, NPs and PAs within those centers. This CHC stratum was added to the NAMCS family of surveys in 2006 and consists of approximately 104 CHCs per year. Sampled CHCs include Federally Qualified Health Center (FQHC) clinics that receive Section 330 grants under the Public Health Service Act, “look-alike” health centers that meet FQHC requirements, and federally-qualified Indian Health Service clinics. Separate response rates taking CHC participation and visit response into account were computed by provider type. Among CHC nonphysician providers (NPs, PAs, and CNMs), the combined response
233
rate was 86.6%; among CHC physicians, it was 85.5% (unpublished calculations by authors). Our study includes only visits to Section 330 grantees and “look-alike” CHCs.27 Our study included 670 physicians, 245 NPs, and 103 PAs, as well as the 24,528 visits that patients made to these providers in CHCs from 2006–2010. The 2010 data are the most recent data available. All sampled providers were asked to complete a provider induction survey and 30 patient visit forms for a randomly selected sample of patients over a randomly selected one-week period. Our estimates are based on the provider type who actually saw the patient. For some patients, time spent with the patient was recorded for two provider types, suggesting that two providers saw the patient. These visits were rare (0.4% of visits) and were excluded from our analysis. The nonresponse rate for most questions was less than 5%. Exceptions were race (20.9%), ethnicity (16.4%), and number of past visits during the previous 12 months (10.5%). All of these variables, however, were imputed by National Center for Health Statistics analysts and used in our analysis. Imputation was accomplished by randomly assigning a value from another record with similar characteristics. Imputations, in general, were based on physician specialty, geographic region, and diagnosis codes.28 Sampling weights were used to obtain national estimates for all analyses. To adjust for the complex sample design, standard errors were obtained using Taylor-series approximation with SUDAAN software. Because CNMs constituted a very small portion of our sample (1.7%), we included them in the NP category. When possible, we analyzed PAs separately from NPs. For the provider demographics and trend analyses, NP or PA sample size limitations frequently produced unreliable estimates, so we combined NPs and PAs into a single group for analysis. However, we only combined NPs and PAs into a single analytic category after first confirming that both were more similar to each other than either as a separate category was similar to physicians with regard to the attribute being analyzed. We used a weighted least-squares regression analysis to determine the significance of trends in numbers and percent of each provider type by year.29 All other analyses combined all five years of data. We compared provider characteristics and time with patient variables across provider types using t tests and chi square tests of association. We performed linear regression on the time spent with provider, adjusting for the patient and visit factors that might affect the amount of time required for a visit. We adjusted for patient demographic and socioeconomic factors, reason for visit, visit complexity (measured by number of chronic conditions of the patient and number of services provided at the visit), factors indicating the patient's relationship to the clinic and to the provider (whether the provider is the patient's primary care provider, whether the patient is new to the clinic, number of times the patient has been seen in the clinic during the previous twelve months), and timing (study year and time of year). This regression analysis was performed at the visit level and included 21,125 patient encounters. Because our previous work42 with this dataset showed that patient visit attributes for PAs were often more similar to those of physicians than to those of NPs, we did not combine PA and NP visits for the visit level regression analysis. We used a significance level of p o0.05 for all analyses. The Duke University Medical Center Institutional Review Board declared this research exempt from full review.
3. Results When taking all five years together, physicians constituted 69% of CHC providers, NPs 21%, and PAs 10%. Due to sample size limitations, separate annual estimates for NPs and PAs are not
234
P. Morgan et al. / Healthcare 2 (2014) 232–237
shown, but NPs were present at approximately twice the rate of PAs throughout the time period studied. The estimated number of CHC providers (physicians, NPs, and PAs) increased from 2006– 2009 and then decreased in 2010 (Fig. 1). Trend analyses showed a statistically significant increase in total providers (p o0.01) and in NPs/PAs (combined, p o0.01) over the five years studied, but physician numbers did not change significantly (p¼ 0.10). Fig. 2 shows the estimates for the proportion of providers who were physicians compared to the proportion who were NPs/PAs (combined). By 2010, NPs and PAs together comprised 35% of CHC providers. Trend tests did not show significant differences in the proportions by provider type (p ¼0.21). Provider-level variables are shown in Table 1. Physicians tended to be younger (p o0.05), with more physicians in the 35–44 year old age group, and more NPs and PAs in the 45–54 year old age group. Sex differences by provider type were both substantial and statistically significant, with 83% of NPs and PAs compared to 45% of physicians in CHCs being female. Physicians were less common in rural CHCs (11% of CHC physicians), compared to NPs and PAs (30% of CHC NPs/PAs)(p o0.01). NPs and PAs were more often nonHispanic whites than were physicians (p o0.01). Regional distributions did not vary by provider type. During the reporting week, PAs saw patients for a slightly – but statistically significant – larger portion of the week (3.8 days) than physicians (3.5 days) or NPs (3.4 days). There was not a statistical difference in the number of patients seen by provider type when the analysis unit was the entire reporting week, but when we adjusted for the number of days spent seeing patients, PAs saw fewer patients per clinical care day than did physicians (14.8 patients per day for physicians and 12.9 for PAs, p o0.05). The number of patients seen per day spent in patient care for NPs was intermediate at 14.2, but this number was not statistically different
Number of providers
20000
Estimated number of providers in CHCs, 2006-2010
15000 Total
10000
Physician NP+PA
5000 0
2006
2007
2008
2009
2010
Fig. 1. Trend test: p o0.01 for all visits, p¼ 0.10 for physicians, p o 0.01 for NPsþ Pas. Source: National Ambulatory Medical Care Survey.
Estimated Percent of Community Health Center Providers, 2006-2010 28
26
30
33
Table 1 Provider characteristics of physicians, nurse practitioners, or physician assistants in Community Health Centers, 2006–2010. Source: National Ambulatory Medical Care Survey. Provider type
p values
Physicians
NPs þPAs
o 35 35–44 45–54 55–64 65 or over Sex (%)
15 31 24 22 8
16 20 35 22 7
Male Female MSA (%)
55 45
17 83
Urban Rural Region (%)
89 11
70 30
Northeast Midwest South West Race/ethnicity (%)
28 17 25 31
26 18 27 28
62 11 9 18
79 8 8 a, 5
Physician vs. NP/PA
Age group (%) o 0.05
o 0.01
o 0.01
0.93
o 0.01 Non-Hispanic white Non-Hispanic black Hispanic Non-Hispanic other
a The estimates for NPs and PAs in this category had large standard error with relative standard error40.3.
than that for physicians or PAs. In this unadjusted analysis (Table 2), there was not a difference in the mean amount of time (varied between 18.6 and 20 minutes) that the provider types spent on each patient. In the adjusted linear regression analysis (Table 3), the mean time spent with patients was 19.82 minutes for physicians, 19.96 minutes for NPs, and 20.15 minutes for PAs. These differences are very small (all within 20 seconds per visit) and not statistically significant. Patient factors that reduced the mean time spent were being under 18 years of age, being seen in 2009 (compared to 2010), having Medicare or Medicaid/SCHIP insurance (compared to private insurance) and having been seen in the clinic before. Patient factors that increased the mean time spent were being seen for a flare-up of a chronic condition (vs. for preventive care), receiving more health services, receiving more health education services, being seen by a provider who considers themselves the patient's primary care provider (vs. not knowing whether they are the primary care provider), and being seen for mental healthcare, prevention or pregnancy, or ill-defined or vision and hearing symptoms. Full regression analysis results are shown in Appendix 1.
35 4. Discussion
% 72
74
2006
2007
70
2008 Physician
67
65
2009
2010
NP+PA
Fig. 2. Trend test: p¼ 0.21. Source: National Ambulatory Medical Care Survey.
NPs and PAs comprised a substantial portion (31%) of the providers in CHCs during the time period studied. The increased tendency that we found for NPs/PAs to work in rural CHCs supports other research that found NPs and PAs particularly contribute to access to care in high need rural areas.30–32 Overall, measures of time spent with patients were remarkably similar across provider types in both crude and adjusted analyses. PAs spent two to three more hours during the reporting week
P. Morgan et al. / Healthcare 2 (2014) 232–237
235
Table 2 Comparison of time spent in patient care and time spent per patient by provider type, Community Health centers, 2006–2010. Source: National Ambulatory Medical Care Survey. Provider type
Mean number of visits in reporting week Mean number of days patients seen in reporting week Number of visits per day (calculated) Mean time spent per patient (minutes)
P values
Physician
NP
PA
Physician vs. NP
Physician vs. PA
NP vs. PA
53.3 3.5 14.8 18.6
48.7 3.4 14.2 19
50 3.8 12.9 20
0.22 0.47 0.48 0.53
0.46 o 0.05 o 0.05 0.06
0.7752 o0.05 0.13 0.26
Number of visits per day was calculated by dividing the mean number of visits by the mean number of days patients were seen.
Table 3 Time spent with patients by provider type, adjusted. Source: National Ambulatory Medical Care Survey. Provider type
Mean time spent per patient (minutes)
P value for difference from physician time
Physician NP PA
19.82 19.96 20.15
– 0.62 0.75
Linear regression analysis was used to adjust for patient demographic (age group, sex, race/ethnicity, rural/urban status), socioeconomic (expected payment source, county median household income) and medical complexity (number of chronic conditions) factors, visit (major reason for visit, number of services provided, number of medications prescribed, number of health education/prevention services provided, and type of care provided according to Fenton categories based on visit ICD-9 codes47), patient relationship to the clinic (was patient new to the clinic, number of times patient was seen in the past year, whether the provider considers him/herself the patients' primary care provider), and survey administration (survey year, season of the year) variables.
providing patient care. Despite this finding, PAs did not see more patients or spend more time within a visit. This finding could suggest that PAs spend more time on their patient care days in activities other than seeing patients who would be reported on the NAMCS patient encounter form, such as communicating with patients who are not in the clinic. The time allowed for patient visits is often established by clinic scheduling procedures that establish a standard time allocation for each visit. If physicians, NPs and PAs are all given the same fixed amount of time to see each patient, this could influence the amount of time that providers spend with a patient.33 Visits for returning patients are often scheduled for 15 minute or 20 minute time slots. Other factors can affect the time actually spent, including longer appointment blocks that some clinics use for new patients and those scheduled for yearly check-ups or complex visits. Still, the results of our study and others19,33 find that visit length for all provider types averages very near twenty minutes per patient. When patient needs require additional time, providers typically schedule return appointments to address those needs. Assessment of the relative amount of time required to care for patients by provider type would ideally include the rate of return visits.34 Few studies have examined total office visit use in relation to provider type, but those that have done so have not found increased office visit resource use with nonphysician care35–37. The relative increase that we found in use of NPs and PAs in CHCs is similar to that seen in other reports on CHCs38 and in other settings39. The large increase in total providers in 2009 can be attributed to funding for CHC expansion as part of the American Recovery and Reinvestment Act40, and the subsequent decrease in 2010 was likely caused by uncertainty about whether increased levels of funding would be sustained41. The relatively large decrease in mean time spent per patient (about 1.7 minutes) in 2009 might reflect the fact that CHC provider growth between 2008 and 2009 (39%) did not keep pace with the increase in
number of visits to CHCs (57%).42 For CHCs providers to handle the large increase in demand for patient visits to safety net providers, caused presumably by widespread job losses during the economic recession,43 they likely had to decrease the amount of time they spent with each patient. The CHC population is younger and more socially and medically complex than the general US population,44,45 which limits generalizability of our results. A recent analysis of primary care visits by Bruen and colleagues using NAMCS data found that care of CHC patients was similar to that of patients in other settings with regard to content of care and time spent per visit.33 Like the Bruen group, we found that the major determinants of time spent with patients were their health needs and the type and quantity of care provided. The organization of care and characteristics of providers are different in CHCs than in the private sector, and these differences might also limit generalizability of our findings. Study limitations include that data were obtained by provider or clinic staff report rather than through direct observation. Other studies have found that physicians overestimate the time spent with patients,46 but we have no reason to suspect that such overestimation would vary by whether the provider is a physician, NP, or PA. Since a large number of services can be reported in NAMCS, we used summary measures of services ordered or provided (total health services, total health education services, total other services). However, the services tallied for these summary measures varied widely in complexity and in the amount of time that would be required. For example, items in the total health services summary measure ranged from simple activities such as taking a temperature to more complex activities such as obtaining and interpreting an electrocardiogram. Similarly, we could not distinguish between brief or extended preventive counseling. Although we attempted to adjust our analysis for patient complexity, residual confounding by complexity may remain.
5. Conclusions Physicians, NPs, and PAs seem to spend similar amounts of time —about 20 minutes‐‐ with each patient in CHCs. This allocation of time is similar to that seen in other settings and may be driven by clinic scheduling practices. Information about the relative amount of time that physicians, NPs, and PAs spend with patients has implications for workforce planning to meet goals of organizational efficiency and increased health system capacity, but may also affect patient health outcomes and patient and provider satisfaction.
Disclaimer The findings and conclusions in this paper are those of the authors and do not necessarily represent the views of CDC.
236
P. Morgan et al. / Healthcare 2 (2014) 232–237
Conflict of interest disclosure statement This statement accompanies the article “Time Spent with Patients by Physicians, Nurse Practitioners, and Physician Assistants in Community Health Centers, 2006–2010,” authored by Perri Morgan, Christine Everett, and Esther Hing and submitted to Healthcare as an original article. Below all authors have disclosed relevant commercial associations that might pose a conflict of interest: Consultant arrangements: None. Stock/other equity ownership: None. Patent licensing arrangements: None. Grants/ research support: None. Employment: None. Speakers' bureau: None Expert witness: None. Other: None.
Acknowledgement This work was supported in part by an unrestricted grant from the Physician Assistant Education Association.
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.hjdsi.2014.09.009. References 1. Berwick D, Nolan T, Whittington J. The triple aim: care, health, and cost. Health Aff (Millwood). 2008;27(3):11. 2. Jackson G, Powers B, Chatergee R, et al. The patient-centered medical home: a systematic review. Ann. Intern Med. 2013;158(3):169–178 (February). 3. Association of American Medical Colleges. The Impact of Health Care Reform on the Future Supply and Demand for Physicians: Updated Projections Through 2025. 2011; 〈https://www.aamc.org/download/158076/data/updated_projec tions_through_2025.pdf〉. 4. Council on Graduate Medical Education. Twentieth Report: Advancing Primary Care. 2010; 〈http://www.hrsa.gov/advisorycommittees/bhpradvisory/cogme/ Reports/〉. Accessed May 10, 2013. 5. Keehan SP, Sisko AM, Truffer CJ, et al. National health spending projections through 2020: economic recovery and reform drive faster spending growth. Health Aff (Millwood). 2011;30(8):1594–1605 (Aug). 6. Gottlieb S, No Emanuel EJ. There won't be a doctor shortage. New York Times; 2013 NY, NY (December 4). 7. Auerbach DI, Chen PG, Friedberg MW, et al. Nurse-managed health centers and patient-centered medical homes could mitigate expected primary care physician shortage. Health Aff (Millwood). 2013;32(11):1933–1941 (November 1). 8. Shipman S, Sinsky C. Expanding primary care capacity by reducing waste and improving the efficiency of care. Health Aff (Millwood). 2013;32(11):1990–1997 (November 1). 9. Laurant M, Harmsen M, Wollersheim H, Grol R, Faber M, Sibbald B. The impact of nonphysician clinicians: do they improve the quality and cost-effectiveness of health care services? Med Care Res Rev. 2009;66(6 Suppl):S36–S89 (Dec). 10. Newhouse RP, Stanik-Hutt J, White KM, et al. Advanced Practice Nurse Outcomes 1990-2008: a systematic review. Nurs Econ. 2011;29:5. 11. Cronenwett L, Dzau V. Who Will Provide Primary Care and How Will They Be Trained?, Proceedings of a Conference Sponsored by the Josiah Macy, Jr. Foundation. Josiah Macy, Jr. Foundation; 2010: 〈http://www.josiahmacyfounda tion.org/documents/JMF_PrimaryCare_Monograph.pdf〉. Accessed May 26, 2010. 12. Negotiated Rulemaking Committee on the Designation of Medically Underserved Populations and Health Professional Shortage Areas. Negotiated Rulemaking Committee on the Designation of Medically Underserved Populations and Health Professional Shortage Areas Report: Appendices and Addenda. October 31, 2011; 〈http://www.hrsa.gov/advisorycommittees/shortage/〉. Accessed October 23, 2013. 13. Lin CT, Albertson GA, Schilling LM, et al. Is patients' perception of time spent with the physician a determinant of ambulatory patient satisfaction? Arch Intern Med. 2001;161(11):1437–1442. 14. Landau D-A, Bachner YG, Elishkewitz K, Goldstein L, Barneboim E. Patients' views on optimal visit length in primary care. J Med Pract Manage. 2007;23 (1):12–15 (Jul–Aug). 15. Otani K, Kurz RS, Harris LE. Managing primary care using patient satisfaction measures. (Sep–Oct). J Healthc Manag. 2005;50(5):311–324 (discussion 324315). 16. Wood GC, Spahr R, Gerdes J, Daar ZS, Hutchison R, Stewart WF. Patient satisfaction and physician productivity: complementary or mutually exclusive? Am J Med Qual. 2009;24(6):498–504 (Nov–Dec).
17. Chen PG, Curry LA, Nunez-Smith M, Bradley EH, Desai MM. Career satisfaction in primary care: a comparison of international and US medical graduates. J Gen Intern Med. 2012;27(2):147–152 (Feb). 18. Linzer M, Konrad TR, Douglas J, et al. Managed care, time pressure, and physician job satisfaction: results from the physician worklife study. J Gen Intern Med. 2000;15(7):441–450 (Jul). 19. Chen LM, Farwell WR, Jha AK. Primary care visit duration and quality: does good care take longer? Arch Intern Med. 2009;169(20):1866–1872 (Nov 9). 20. Stewart MA. Effective physician–patient communication and health outcomes: a review. CMAJ Can Med Assoc J. 1995;152(9):1423–1433 (May 1). 21. Howie JG, Porter AM, JF. F.. Quality and the use of time in general practice: widening the discussion. BMJ.. 1989;15(298):1008–1010 (April 15). 22. Strand J. The myth of more time. Assoc Physician Assist Progr (APAP) Update. 2004 (March). 23. Everett C, Thorpe C, Palta M, Carayon P, Gilchrist V, Teams Smith M. Roles, and goals: primary care physician assistant and nurse practitioner roles and outcomes for older patients with diabetes. Health Aff (Millwood). 2013;32 (11):1942–1948. 24. Adashi EY, Geiger HJ, Fine MD. Health care reform and primary care—the growing importance of the community health center. N Engl J Med. 2010;362 (22):2047–2050 (Jun 3). 25. Hing E, Hooker RS, Ashman JJ. Primary health care in community health centers and comparison with office-based practice. J. Community Health. 2011;36 (3):406–413 (Jun). 26. Centers for Disease Prevention and Control. Public Use Data File Documentation: 2006–2010 National Ambulatory Medical Care Survey.. 〈http://www.cdc. gov/nchs/ahcd.htm〉. Accessed September 16, 2013. 27. Indian Health Service (IHS) clinics were not included. Since IHS clinics have markedly different staffing patterns, inclusion of these clinics would have skewed our results. The IHS clinic sample was quite small compared to the Section 330 grantees and the lookalike clinics. 28. National Center for Health Statistics. Public Use microdata file documentation: National Ambulatory Medical Care Survey, 2006–2010. 2012; 〈ftp://ftp.cdc.gov/ pub/Health_Statistics/NCHS/Dataset_Documentation/NAMCS/〉. Accessed November 11, 2013. 29. Sirkem MG, Shimizu I, French DK, DB B. Manual on Standards and Procedures for Reviewing Statistical Reports. Hyattsville, MD: National Center for Health Statistics; 1990. 30. Skillman SM, Fordyce MA, Yen W, Mounts T. Washington State Primary Care Provider Survey, 2011-2012: Summary of Findings. Seattle, WA: WWAMI Center for Health Workforce Studies; 2012. 31. Grumbach K, Hart LG, Mertz E, Coffman J, Palazzo L. Who is caring for the underserved? A comparison of primary care physicians and nonphysician clinicians in California and Washington Ann of Family Med. 2003;1(2):97–104. 32. Dehn RW. The distribution of physicians, advance practice nurses, and physician assistants in Iowa. J Physician Assist Educ. 2006;17(1):36–39. 33. Bruen BK, Ku L, Lu X, Shin P. No evidence that primary care physicians offer less care to medicaid, community health center, or uninsured patients. Health Aff (Millwood). 2013;32(9):1624–1630. 34. Fishman PA, Hornbrook MC. Assigning resources to health care use for health services research: options and consequences. Med. Care. 2009;47(7 Suppl 1): S70–S75 (Jul). 35. Hooker RS. A cost analysis of physician assistants in primary care. J Am Acad Physician Assist. 2002;15(11):39–42 (45, 48 passim, 2002 Nov). 36. Morgan PA, Shah ND, Kaufman JS, Albanese MA. Impact of physician assistant care on office visit resource use in the United States. Health Serv Res. 2008;43(5 Pt 2):1906–1922 (Oct). 37. Lenz ER, MOM RL, Kane SC, Hopkins, Lin SX. Primary care outcomes in patients treated by nurse practitioners or physicians: two-year follow-up. Med Care Res Rev. 2004;61(3):332–351 (September 1, 2004). 38. National Association of Community Health Centers. Expanding Access to Primary Care: The Role of Nurse Practitioners, Physician Assistants, and Certified Nurse Midwives in the Health Center Workforce. Fact Sheet # 0913, 2013; 〈http://www.nachc.com/client/documents/Workforce_FS_0913.pdf〉. 39. Morgan P, Abbott D, McNeill R, Fisher D. Characteristics of Primary Care Office Visits to Nurse Practitioners, Physician Assistants, and Physicians in United States Veterans Health Administration Facilities, 2006-2010: a retrospective cross-sectional analysis. Hum Resour Health. 2012;10:42. 40. National Association of Community Health Centers. More Patients Gain Access To Health Center Care Thanks to Stimulus Funds. 2010; 〈http://www.nachc.org/ client/ARRA%20Fact%20Sheet%20Final%20revised.pdf〉. Accessed December 16, 2010. 41. Morris A. FY 2011 Appropriations Reaches Conclusion; Health Centers Slated for $600 Million Cut. Nat Assoc Community Health Cent blog. 2011 (Accessed October 25, 2013). 42. Morgan PA, Everett CM, Hing E. Nurse Practitioners, Physician Assistants, and Physicians in Community Health Centers, 2006–2010. Health Care, in press, http://dx.doi.org/10.1016/j.hjdsi.2014.06.002. 43. National Association of Community Health Centers. Recession Brings More Patients to Community Health Centers. Fact Sheet #0209 2009; 〈https://www. nachc.com/client/documents/rising_patient_%20demand_093.pdf〉. Accessed January 17, 2014. 44. Harrison B, Rittenhouse D, Phillips Jr. RL, Grumbach K, Bazemore AW, Dodoo MS. Title VII. is criticial to the community health center and national health service corps workforce. Am Fam Physician.. 2010;81(2): (132-132).
P. Morgan et al. / Healthcare 2 (2014) 232–237
45. National Association of Community Health Centers. Community Health Centers: The Local Prescription for Better Quality and Lower Costs. 2011; 〈http:// www.nachc.com/client/A%20Local%20Prescription%20Final%20brief%203% 2022%2011.pdf〉. Accessed July 15, 2013.
237
46. Gilchrist VJ, Stange KC, Flocke SA, McCord G, Bourguet CC. A comparison of the national ambulatory medical care survey (NAMCS) measurement approach with direct observation of outpatient visits. Med Care. 2004;42(3):276–280.
