Primary Care Office Visits For Acute Care Dropped Sharply In 2002–15, While ED Visits Increased Modestly

Acute care, defined as the evaluation and treatment of newly arising health concerns (including acute exacerbations of chronic illnesses), accounts for more than one-third of all US ambulatory care visits, of which approximately 28 percent are to the emergency department (ED).¹ While inpatient capacity and utilization have declined over the past two decades,² ED visits in the US have increased substantially, from 90 million in 1996 to more than 136 million in 2015, outpacing population growth.^3,4 This has garnered considerable attention from policy makers and has motivated efforts to reduce ED visits across the nation—especially those that are potentially suitable for office-based care, which are often labeled nonurgent visits.^5,6 Underlying this narrative is the assumption that acute care is available and accessible in lower-cost office settings. But the proliferation of urgent care centers and retail clinics suggests that there is a growing deficit in the supply of timely office-based acute care visits.⁷ However, recent trends in office-based acute care, particularly in primary care settings, have not been thoroughly examined.

Over the past two decades the redesign of primary care delivery—with novel models such as the patient-centered medical home, accountable care organizations, Comprehensive Primary Care (CPC) Initiative, and CPC Plus (CPC+)—has become a policy priority for professional societies, policy makers, and federal agencies.^8–14 These efforts, in which public payers have played a significant role, have aimed to improve chronic disease care, prevention, and care coordination within a fragmented health care system. The redesigned delivery models also emphasized the availability of timely visits, including access to same-day and after-hours appointments.^14,15 While recent primary care redesign efforts have improved access to preventive services and chronic disease care, and have improved patient and provider satisfaction,^16–18 their effect on the use of acute care visits to the office setting is largely unknown.

We examined trends in acute care visits to EDs and office-based primary care providers in the period 2002–15, combining data from nationally representative surveys. We hypothesized that rates of both sets of visits per 1,000 population were increasing, particularly among the growing populations of Medicaid beneficiaries and adults ages sixty-five and older, either in Medicare or privately insured.

Study Data And Methods

Data Source

We analyzed publicly available data from the National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey ED subsample (NHAMCS-ED). Both are annual surveys conducted by the National Center for Health Statistics (NCHS) using stratified, multistage samples of visits to outpatient offices and EDs in the United States. Each visit is assigned a weight and survey design variables to generate nationally representative estimates. Because the NCHS began to use single-stage survey design variables in 2002 and the most recent data available were from 2015, we chose 2002–15 for our study period. Details about the survey methodology are available on the NCHS website.¹⁹

We obtained the denominator for annual population-based visit rates for 2002–15 from the Annual Social and Economic Supplement of the Census Bureau’s Current Population Survey.²⁰ The NAMCS and NHAMCS-ED protocols have been approved by the NCHS research ethics review board.

Variable Definitions

We defined acute care visits in NAMCS as office visits with responses to the item “major reason for visit” as “new problem ($<3$ mo. onset)” or “chronic problem, flare up.” Other office visits were considered nonacute visits. We considered all ED visits in NHAMCS-ED as acute care visits.

In our analysis we used the NAMCS definition of primary care visits, which includes visits to physicians in general practice, family medicine, general internal medicine, general pediatrics, and obstetrics-gynecology. We considered all other NAMCS visits as specialty visits. As recommended by the NCHS, we excluded community health center visits (which accounted for approximately 2.5 percent of weighted visits), to maintain comparability across included survey years.

We identified Medicaid beneficiaries from the multinomial variable “expected payment type.” To create mutually exclusive categories, we defined patients dually eligible for Medicare and Medicaid as having Medicare; therefore, they were not included as Medicaid beneficiaries. Detailed variable definitions are available in online appendix A.²¹

Analysis

We first estimated the volume of acute care visits in three settings: primary care office, specialist office, and ED. We then calculated yearly population-based visit rates by dividing the weighted visit counts by the US reference population. We plotted figures by calculating visit rates over two-year periods.

We examined the significance of linear trends in annual visit rates using weighted least squares regression, with weighted visit rates for each data year as data points and the inverse of their variances as analytic weights (see appendix A for details on our statistical methods).²¹ We defined the level of significance at an α of 0.01 for two-sided tests, as recommended by the NCHS.

We replicated the above analysis in two growing subpopulations: Medicaid beneficiaries and adults ages sixty-five and older (Medicare or privately insured). To draw a more complete picture of the overall burden of care, we calculated annual acute and nonacute visit rates in specialty and primary care offices. Our analysis used STATA/MP, version 15.

Sensitivity Analysis Of Primary Care Acute Visits

After 2004 NAMCS discontinued the use of the “episode of care” variable, which served to identify visits as an “initial” visit versus a “follow-up” visit within an episode of illness. Because of this, we performed several sensitivity analyses that examined the trends in acute care visits to primary care practices. First, we examined annual population-based visit rates of acute care visits to primary care and specialty offices but excluded visits for “chronic problem, flare up.” We also performed this analysis in the Medicaid and older adult subpopulations.

Second, we examined annual population-based visit rates of acute primary care visits limited to those with a symptom or disease common in primary care visits (see appendix exhibit A1 for the list of common primary care symptoms or diseases),²¹ adopted from a prior study.²²

Third, we plotted biannual rates of acute care visits to primary care offices, stratified by primary care specialty.

Fourth, we plotted biannual rates of acute care visits to primary care offices, stratified by practice type—a variable available in the period 2002–11. Because clinicians at urgent care centers may be primary care or specialty providers (such as emergency medicine physicians), we also examined the linear trends in overall per population acute care visit rates to these centers.

Lastly, we grouped reasons for visits into major categories (see appendix exhibit A2)²¹ and identified the ten most common categories in ED visits as well as in acute and nonacute visits to primary care and specialty offices. We further plotted the biannual visit rates for the five most common major categories of reasons for acute and nonacute visits to primary care offices.

Limitations

This study had a number of limitations. First, it was limited by the cross-sectional design of the surveys. While the data allowed us to estimate the changes in visit volumes, they did not allow us to establish the underlying cause of the changes.

Second, our definition of acute care visits in office practice included follow-up visits for acute illnesses. The episode-of-care indicator in NAMCS that specified the visits as an initial versus a follow-up visit for an acute illness was available only for the first few years of our study period (2002–04). During that time, 34.0 percent of acute care office visits were follow-up visits (see appendix exhibit A3 for the proportion of office-based acute care visits in 2002–04 by major reason for visit and episode of care),²¹ and thus not directly comparable to unplanned ED visits. However, follow-up visits—whether to check illness progression or continue therapy—also play an important role in office-based care for an acute illness. Therefore, although we could not differentiate initial visits from follow-up visits, changes in acute visit rates over time remain informative of trends in office-based care for episodic acute illnesses. Moreover, the proportions of acute care visits as initial or follow-up visits did not change significantly over the years when the episode-of-care indicator was available in NAMCS (2001–04) (see appendix exhibit A4 for proportions of acute care visits by episode of care in this period).²¹

Third, we did not include the hospital outpatient department sample of NHAMCS (NHAMCS-OPD) because the NCHS had released data from it only up to 2011. In a prior analysis, NHAMCS-OPD accounted for approximately 7.1 percent of all acute care visits, including those in both NAMCS and NHAMCS-ED.¹ Acute care visit rates in NHAMCS-OPD also did not change significantly from 2002 to 2011 (annual trend: 0.8 visits per 1,000 population; $p=0.45$).

Fourth, our results measured the volume of acute care visits to different outpatient care settings. However, the NAMCS and NHAMCS-ED data sets do not include comorbid conditions or billing levels, so we could not account for the changes in patient complexity and work intensity for each visit.

Fifth, because our data extended only up to 2015, we could not evaluate the effects of the Affordable Care Act (ACA), whose first major coverage expansions (state expansions of eligibility for Medicaid and the creation of the health insurance Marketplaces) were implemented in 2014. But efforts to redesign primary care began in the early 2000s, preceding the ACA.¹¹ The law continued to support novel primary care delivery models such as accountable care organizations.^12,13 Moreover, state-level studies of the effects of coverage expansion on ED visit trends are mixed, with no clear national effect of the ACA.^23–25 Therefore, we believe that the trends in acute primary care and ED visits we observed in this analysis reflect the structural changes in health care delivery that not only predated the ACA but have continued after its implementation.

Study Results

Trends In Acute Care Visits

In the period 2002–15 NAMCS sampled 85,694 acute care visits to primary care practices and 92,515 to specialist practices. NHAMCS-ED sampled 354,219 ED visits. During the study period the weighted estimate of acute care visits to primary care practices declined from about 268 million visits in 2002 to about 203 million in 2015 (exhibit 1). However, in the same period ED visits rose from about 110 million to about 137 million, and acute care visits to specialty practices increased from about 131 million to about 172 million.

When we accounted for total population growth, acute care visits to primary care practices decreased from 937.5 to 637.0 visits per 1,000 population during the study period, while ED visit rates increased from 385.2 to 429.5 visits per 1,000 population. Exhibit 2 shows biannual rates of acute care visits to primary care and specialty offices and to EDs per 1,000 population in the study period.

SOURCE Authors’ analysis of data for 2002–15 from the National Ambulatory Medical Care Survey and the National Hospital Ambulatory Care Survey emergency department (ED) subsample. NOTES The total US population was abstracted from the Annual Social and Economic Supplement of the Current Population Survey. “Primary care” and “specialty” are explained in the notes to exhibit 1.Community health center visits are excluded. All standard errors are less than 15 percent of the estimated number of visits.

Acute Care Visits Among Medicaid Beneficiaries And Older Adults

For Medicaid beneficiaries, the weighted number of acute care visits increased substantially across all settings (exhibit 1). However, when we accounted for the increase in the number of Medicaid beneficiaries, the rate of acute care visits to primary care offices decreased from 908.4 to 770.4 visits per 1,000 beneficiaries. Notably, ED visit rates showed no significant change over time. Exhibit 3 shows biannual rates of acute care visits to primary care and specialty offices and to EDs per 1,000 Medicaid beneficiaries in the study period.

SOURCE Authors’ analysis of data for 2002–15 from the National Ambulatory Medical Care Survey and the National Hospital Ambulatory Care Survey emergency department (ED) subsample. NOTES The total US population of Medicaid beneficiaries was abstracted from the Annual Social and Economic Supplement of the Current Population Survey. “Primary care” and “specialty” are explained in the notes to exhibit 1. Community health center visits and the population dually eligible for Medicare and Medicaid are excluded. All standard errors are less than 15 percent of the estimated number of visits.

Similarly, among adults ages sixty-five and older (with Medicare or private insurance), rates of acute care visit to primary care practices decreased from 1,252.8 to 861.7 visits per 1,000 older adults, while ED visit rates had no significant changes over time (exhibit 1). Exhibit 4 shows biannual rates of acute care visits to primary care and specialty practices and to EDs per 1,000 adults ages sixty-five and older in the study period.

SOURCE Authors’ analysis of data for 2002–15 from the National Ambulatory Medical Care Survey and the National Hospital Ambulatory Care Survey emergency department (ED) subsample. NOTES The total US population of adults ages sixty-five and older (with Medicare or private insurance) was abstracted from the Annual Social and Economic Supplement of the Current Population Survey. “Primary care” and “specialty” are explained in the notes to exhibit 1. Community health center visits are excluded. All standard errors are less than 15 percent of the estimated number of visits.

While we found a substantial decline in acute care visits to primary care practices, rates of nonacute visits to those practices decreased slightly in the study period, from 1,017.3 to 948.4 visits per 1,000 population ($p=0.005$) (appendix exhibits A6 and A7).²¹ In contrast, nonacute visits to specialty practices increased from 692.3 to 981.9 visits per 1,000 population ($p=0.006$).

Sensitivity Analysis Of Acute Primary Care Visits

When we excluded visits for the exacerbation of chronic problems, substantial declines in the rate of acute care visits to primary care practices persisted in the overall population as well as for Medicaid beneficiaries and adults ages sixty-five and older (appendix exhibit A8).²¹ We also found a downward trend in primary care acute visit rates after we limited those visits to the ones with common primary care symptoms and diseases as the reasons for the visit (appendix exhibit A6).²¹

When we stratified acute care visits to primary care practices by primary care specialty type, we found that rates of acute care visits to general/family practice physicians experienced the largest decline, followed by rates of visits to internal medicine and pediatric primary care physicians (appendix exhibit A9).²¹ When we stratified the visits by practice setting, in the period 2002–11, 87.2 percent of primary care acute visits occurred at private solo or group practices, and the rates of those visits decreased over time (appendix exhibit A10).²¹ Concurrently, the overall rates of acute care visits to primary care and specialty providers at urgent care centers did not change significantly over time (annual trend: −3.08 visits per 1,000 population; $p=0.15$).

Appendix exhibit A11 shows the most common categories of reasons for visits by care setting.²¹ Overall, respiratory, digestive, and musculoskeletal symptoms and diseases were among the most common reasons for acute care visits across all settings. General symptoms (such as fever and fatigue) were the most common reason for ED visits and the third most common reason for acute care visits to primary care practices. When we examined the biannual rates of acute care visits to those practices, we found that the rates for the five most common reasons for a visit decreased during the study period—particularly for respiratory symptoms and diseases (appendix exhibit A12a).²¹ The biannual rates for the five most common reasons for a nonacute visit did not have consistent changes (appendix exhibit A12b).²¹

Discussion

In the period 2002–15 overall ED visit rates increased, but there was a decline of nearly one-third in the rate of acute care visits to primary care practices. Although we could not differentiate initial visits from follow-up visits, our results suggest that over time, patients were less often attending to their acute care needs through primary care offices. This is consistent with the recently published Health Care Cost Institute report that noted an 18 percent decrease in commercial claims for nonpreventive primary care office visits from 2012 to 2016.²⁶ We also found declines in rates of acute care visits to primary care offices among both Medicaid beneficiaries and adults ages sixty-five and older, whereas ED visit rates did not change significantly among either of these subgroups. These concurrent trends suggest that acute care has increasingly shifted away from primary care offices and toward other settings.

This decrease is unlikely to be due to a decline in acute care needs. Although comprehensive primary care through models such as the patient-centered medical home may reduce acute care needs through improved preventive care and chronic disease management, the evidence of the effect of comprehensive primary care on acute care use remains mixed.^16–18 Rather, we suspect that the demand for acute care has remained unchanged or even increased, given growing patient complexity and needs across all settings.^27–29

The substantial decline in rates of acute care visits to primary care offices, despite a 10 percent increase in the number of primary care physicians from 2008 to 2014,^30,31 may be because of increasing difficulties that primary care practices face in providing episodic care. As primary care practices move toward novel delivery models, there is a growing emphasis on providing comprehensive services—including care coordination, patient and caregiver engagement, and integration with behavioral health care.^11,14 The nationwide implementation of electronic health record (EHR) systems facilitated the wider availability of electronic access to clinicians. However, studies have not found enhanced electronic clinician access able to replace in-person visits.³² Instead, evidence has consistently shown that EHRs added substantial administrative and documentation burdens.^33,34 With increasing patient complexity, acute illnesses may have become less suitable for office-based care.^27–29 Low reimbursement for primary care visits further disincentivizes providers from reserving excess capacity for acute care visits, since practices may lose revenue opportunities if appointment slots remain unfilled.^35,36 Lastly, the rising trend of part-time practice among primary care physicians may create additional difficulties for patients in accessing their primary care physician when an unanticipated acute need arises.³⁷ While our study could not evaluate the underlying reasons for the decrease in rates of primary care acute visits, these factors may contribute to an environment in which it is increasingly difficult to provide episodic acute care in primary care practices. Furthermore, the concurrent decrease in rates of nonacute primary care visits suggests that there may also be a broader change in the overall capacity of primary care practices in the context of these primary care delivery redesigns.

Given the changes in rates of primary care acute visits, the demand for acute care previously satisfied by primary care offices may be going unmet or, more likely, may have shifted to other acute care settings, including the ED. The growing need for complex diagnosis and treatment may be an underlying driver of increased ED visits, as evidenced by the increasing practice intensity in EDs, particularly for older adults.^28,38 Although we found that ED visit rates increased only slightly, the care received at each ED visit might have required multiple visits to the office setting.

The rapid rise of alternative acute care sites such as retail clinics, urgent care centers, and telemedicine likely also plays a part in the reduction in rates of acute primary care visits. Unfortunately, no reliable current national estimate of care provided at alternative acute care sites is available. While we did not observe a significant trend in national estimates of urgent care visit rates from 2002 to 2011, more recent data show a substantial rise in urgent care, retail clinic, and telemedicine visits among the commercially insured population.³⁹ Alternative acute care sites were shown to have replaced a significant proportion of office-based visits among commercially insured patients,^40,41 but they were associated with only a small decrease in low-acuity ED visits.^39,42 These findings may suggest that these alternative acute care venues likely mainly replaced office-based visits, although their effects on ED and office visits have not been directly compared. Notably, the availability of alternative acute care sites also induced an overall increase in visits and costs.^39–41 The considerable variability in the capabilities of alternative acute care sites also introduces substantial uncertainty about their role in acute care delivery.⁴³ Future research should examine the interaction between the alternative acute care sites, EDs, and primary care offices to further understand the underlying etiology of the decreasing rates of acute care visits to primary care offices.

The absence of significant changes in ED visit rates among Medicaid beneficiaries runs contrary to the common narrative of increasing ED use in this population.^3,23 The concurrent decrease in rates of acute care visits to primary care providers further suggests that most Medicaid beneficiaries do not substitute ED visits for acute primary care visits. Instead, higher ED visit rates among Medicaid beneficiaries likely reflect the difficulties they have in obtaining timely appointments, compared to people with private insurance.^44,45 Qualitative studies have also found that Medicaid beneficiaries tend to perceive hospital-based care to be of higher quality and value than office-based ambulatory care.⁴⁶ Therefore, strategies to economically disincentivize “unnecessary” ED visits, such as increased copayments or coverage denial, have had a limited impact while raising substantial concerns about unintended patient harm.^47,48

Future redesigns of health care delivery must acknowledge this national shift in outpatient acute care.

Future discussions and redesigns of health care delivery must acknowledge this national shift in outpatient acute care. Since the primary care workforce is projected to remain undersupplied in the foreseeable future,³⁰ constructing an interconnected network to provide outpatient acute care—including physician offices, EDs, and alternative acute care sites—should become a policy priority.

Two broad strategies should be emphasized. First, with the increasing ubiquity of EHRs, interoperability will be essential to allow the free flow of clinical information across care settings. As the traditional model of primary care offices as the main care provider of episodic illnesses diminishes, acute care will occur in increasingly diverse settings. Seamless information exchange will be important for coordinating care transitions across various acute care sites, as highlighted by the recent National Quality Forum report on ED transitions of care.⁴⁹

Second, future payment model reforms should align payer and provider incentives with patients’ need for timely, comprehensive care for the entire episode of an acute illness. Such reforms need to include key acute care settings such as the ED.⁵⁰ For example, global budgets and capitation have been used to ensure that health care delivery systems are accountable for the total cost and outcome of the acute care episode as a whole, incentivizing more efficient management of care needs throughout an episode of acute illness and across acute care settings within the delivery system.^51,52

Conclusion

In the period 2002–15 there was a large decline in the population-based rate of acute care visits to primary care offices without a correspondingly large increase in ED visits. There was a similar decrease in acute primary care visits without substantial changes in ED use among economically vulnerable subgroups. As acute care delivery in the US occurs in increasingly diverse settings, it is time to redesign acute care and integrate it within the broader redesign of health care delivery.

ACKNOWLEDGMENTS

Arjun Venkatesh reports career development support from the National Center for Advancing Translational Sciences (Grant No. KL2TR001862) and the Yale Center for Clinical Investigation. Seth Trueger received stipends as social media editor for Annals of Emergency Medicine and Emergency Physicians Monthly. He also became digital media editor for JAMA Network Open on July 1, 2018.

NOTES

• 1 Pitts SR, Carrier ER, Rich EC, Kellermann AL. Where Americans get acute care: increasingly, it’s not at their doctor’s office. Health Aff (Millwood). 2010;29(9):1620–9. Go to the article, Google Scholar
• 2 Henry J. Kaiser Family Foundation. Hospital inpatient days per 1,000 population by ownership type [Internet]. San Francisco (CA): KFF; 2016 [cited 2018 Dec 3]. Available from: https://www.kff.org/other/state-indicator/inpatient-days-by-ownership/ Google Scholar
• 3 Tang N, Stein J, Hsia RY, Maselli JH, Gonzales R. Trends and characteristics of US emergency department visits, 1997–2007. JAMA. 2010;304(6):664–70. Crossref, Medline, Google Scholar
• 4 National Center for Health Statistics. National Hospital Ambulatory Medical Care Survey: 2015 emergency department summary tables [Internet]. Hyattsville (MD): NCHS; [cited 2018 Dec 3]. Available from: https://www.cdc.gov/nchs/data/nhamcs/web_tables/2015_ed_web_tables.pdf Google Scholar
• 5 Agency for Healthcare Research and Quality. Innovations to reduce non-urgent use of emergency services [Internet]. Rockville MD: AHRQ; 2016 Aug 17 [cited 2018 Dec 3]. Available from: https://innovations.ahrq.gov/issues/2016/08/17/innovations-reduce-non-urgent-use-emergency-services Google Scholar
• 6 Uscher-Pines L. Applying what works to reduce non-urgent emergency department use. RAND Blog [blog on the Internet]. 2013 May 22 [cited 2018 Dec 3]. Available from: https://www.rand.org/blog/2013/05/applying-what-works-to-reduce-non-urgent-emergency.html Google Scholar
• 7 Mehrotra A. The convenience revolution for treatment of low-acuity conditions. JAMA. 2013;310(1):35–6. Crossref, Medline, Google Scholar
• 8 Grumbach K, Bodenheimer T. A primary care home for Americans: putting the house in order. JAMA. 2002;288(7):889–93. Crossref, Medline, Google Scholar
• 9 Kilo CM, Wasson JH. Practice redesign and the patient-centered medical home: history, promises, and challenges. Health Aff (Millwood). 2010;29(5):773–8. Go to the article, Google Scholar
• 10 Koh HK, Sebelius KG. Promoting prevention through the Affordable Care Act. N Engl J Med. 2010;363(14):1296–9. Crossref, Medline, Google Scholar
• 11 American Academy of Family Physicians, American Academy of Pediatrics, American College of Physicians, American Osteopathic Association. Joint principles of the patient-centered medical home [Internet]. Leawood (KS): AAFP; 2007 Mar 7 [cited 2018 Dec 3]. Available from: http://www.aafp.org/dam/AAFP/documents/practice_management/pcmh/initiatives/PCMHJoint.pdf Google Scholar
• 12 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):2301–3. Crossref, Medline, Google Scholar
• 13 Davis K, Abrams M, Stremikis K. How the Affordable Care Act will strengthen the nation’s primary care foundation. J Gen Intern Med. 2011;26(10):1201–3. Crossref, Medline, Google Scholar
• 14 CMS. Comprehensive Primary Care Initiative [Internet]. Baltimore (MD): Centers for Medicare and Medicaid Services; [last updated 2018 Nov 29; cited 2018 Dec 3]. Available from: https://innovation.cms.gov/initiatives/comprehensive-primary-care-initiative/ Google Scholar
• 15 National Committee for Quality Assurance. NCQA PCMH standards and guidelines. Washington (DC): NCQA; 2017. Google Scholar
• 16 Jackson GL, Powers BJ, Chatterjee R, Bettger JP, Kemper AR, Hasselblad Vet al. The patient centered medical home. A systematic review. Ann Intern Med. 2013;158(3):169–78. Crossref, Medline, Google Scholar
• 17 Jabbarpour Y, De Marchis E, Bazemore A, Grundy P. The impact of primary care practice transformation on cost, quality, and utilization: a systematic review of research published in 2016 [Internet]. Washington (DC): Patient-Centered Primary Care Collaborative; 2017 Jul [cited 2018 Dec 3]. Available from: https://www.pcpcc.org/sites/default/files/resources/pcmh_evidence_report_08-1-17%20FINAL.pdf Google Scholar
• 18 Peikes D, Dale S, Ghosh A, Taylor EF, Swankoski K, O’Malley ASet al. The Comprehensive Primary Care Initiative: effects on spending, quality, patients, and physicians. Health Aff (Millwood). 2018;37(6):890–9. Go to the article, Google Scholar
• 19 National Center for Health Statistics. Ambulatory health care data [Internet]. Hyattsville (MD): NCHS; [last updated 2018 Nov 28; cited 2018 Dec 3]. Available from: https://www.cdc.gov/nchs/ahcd/index.htm Google Scholar
• 20 Census Bureau. Current Population Survey (CPS): data [Internet]. Washington (DC): Census Bureau; [cited 2018 Dec 20]. Available from: https://www.census.gov/programs-surveys/cps/data-detail.html Google Scholar
• 21 To access the appendix, click on the Details tab of the article online.
• 22 Kale MS, Federman AD, Ross JS. Visits for primary care services to primary care and specialty care physicians, 1999 and 2007. Arch Intern Med. 2012;172(18):1421–3. Crossref, Medline, Google Scholar
• 23 Sommers BD, Simon K. Health insurance and emergency department use—a complex relationship. N Engl J Med. 2017;376(18):1708–11. Crossref, Medline, Google Scholar
• 24 Nikpay S, Freedman S, Levy H, Buchmueller T. Effect of the Affordable Care Act Medicaid expansion on emergency department visits: evidence from state-level emergency department databases. Ann Emerg Med. 2017;70(2):215–225.e6. Crossref, Medline, Google Scholar
• 25 Miller S, Wherry LR. Health and access to care during the first 2 years of the ACA Medicaid expansions. N Engl J Med. 2017;376(10):947–56. Crossref, Medline, Google Scholar
• 26 Health Care Cost Institute. 2016 health care cost and utilization report [Internet]. Washington (DC): HCCI; 2018 Jun 19 [cited 2018 Dec 3]. Available from: http://www.healthcostinstitute.org/report/2016-health-care-cost-utilization-report/ Google Scholar
• 27 Rich E, Lipson D, Libersky J, Parchman M. Coordinating care for adults with complex care needs in the patient-centered medical home: challenges and solutions [Internet]. Rockville (MD): Agency for Healthcare Research and Quality; 2012 Jan [cited 2018 Dec 3]. (AHRQ Publication No. 12-0010). Available from: https://pcmh.ahrq.gov/page/coordinating-care-adults-complex-care-needs-patient-centered-medical-home-challenges-and Google Scholar
• 28 Pines JM, Mullins PM, Cooper JK, Feng LB, Roth KE. National trends in emergency department use, care patterns, and quality of care of older adults in the United States. J Am Geriatr Soc. 2013;61(1):12–7. Crossref, Medline, Google Scholar
• 29 Tinetti ME, Fried TR, Boyd CM. Designing health care for the most common chronic condition—multimorbidity. JAMA. 2012;307(23):2493–4. Crossref, Medline, Google Scholar
• 30 Petterson SM, Liaw WR, Phillips RL, Rabin DL, Meyers DS, Bazemore AW. Projecting US primary care physician workforce needs: 2010–2025. Ann Fam Med. 2012;10(6):503–9. Crossref, Medline, Google Scholar
• 31 Petterson SM, Liaw WR, Tran C, Bazemore AW. Estimating the residency expansion required to avoid projected primary care physician shortages by 2035. Ann Fam Med. 2015;13(2):107–14. Crossref, Medline, Google Scholar
• 32 Irizarry T, DeVito Dabbs A, Curran CR. Patient portals and patient engagement: a state of the science review. J Med Internet Res. 2015;17(6):e148. Crossref, Medline, Google Scholar
• 33 Shipman SA, Sinsky CA. Expanding primary care capacity by reducing waste and improving the efficiency of care. Health Aff (Millwood). 2013;32(11):1990–7. Go to the article, Google Scholar
• 34 Tai-Seale M, Olson CW, Li J, Chan AS, Morikawa C, Durbin Met al. Electronic health record logs indicate that physicians split time evenly between seeing patients and desktop medicine. Health Aff (Millwood). 2017;36(4):655–62. Go to the article, Google Scholar
• 35 Holtrop JS, Luo Z, Alexanders L. Inadequate reimbursement for care management to primary care offices. J Am Board Fam Med. 2015;28(2):271–9. Crossref, Medline, Google Scholar
• 36 Candon M, Zuckerman S, Wissoker D, Saloner B, Kenney GM, Rhodes Ket al. Declining Medicaid fees and primary care appointment availability for new Medicaid patients. JAMA Intern Med. 2018;178(1):145–6. Crossref, Medline, Google Scholar
• 37 Bodenheimer T, Haq C, Lehmann W. Continuity and access in the era of part-time practice. Ann Fam Med. 2018;16(4):359–60. Crossref, Medline, Google Scholar
• 38 Pitts SR, Pines JM, Handrigan MT, Kellermann AL. National trends in emergency department occupancy, 2001 to 2008: effect of inpatient admissions versus emergency department practice intensity. Ann Emerg Med. 2012;60(6):679–686.e3. Crossref, Medline, Google Scholar
• 39 Poon SJ, Schuur JD, Mehrotra A. Trends in visits to acute care venues for treatment of low-acuity conditions in the United States from 2008 to 2015. JAMA Intern Med. 2018;178(10):1342–9. Crossref, Medline, Google Scholar
• 40 Ashwood JS, Gaynor M, Setodji CM, Reid RO, Weber E, Mehrotra A. Retail clinic visits for low-acuity conditions increase utilization and spending. Health Aff (Millwood). 2016;35(3):449–55. Go to the article, Google Scholar
• 41 Ashwood JS, Mehrotra A, Cowling D, Uscher-Pines L. Direct-to-consumer telehealth may increase access to care but does not decrease spending. Health Aff (Millwood). 2017;36(3):485–91. Go to the article, Google Scholar
• 42 Martsolf G, Fingar KR, Coffey R, Kandrack R, Charland T, Eibner Cet al. Association between the opening of retail clinics and low-acuity emergency department visits. Ann Emerg Med. 2017;69(4):397–403.e5. Crossref, Medline, Google Scholar
• 43 Trueger NS, Chua KP, Hussain A, Liferidge AT, Pitts SR, Pines JM. Incorporating alternative care site characteristics into estimates of substitutable ED visits. Med Care. 2017;55(7):693–7. Crossref, Medline, Google Scholar
• 44 Asplin BR, Rhodes KV, Levy H, Lurie N, Crain AL, Carlin BPet al. Insurance status and access to urgent ambulatory care follow-up appointments. JAMA. 2005;294(10):1248–54. Crossref, Medline, Google Scholar
• 45 Chou SC, Deng Y, Smart J, Parwani V, Bernstein SL, Venkatesh AK. Insurance status and access to urgent primary care follow-up after an emergency department visit in 2016. Ann Emerg Med. 2018;71(4):487–496.e1. Crossref, Medline, Google Scholar
• 46 Kangovi S, Barg FK, Carter T, Long JA, Shannon R, Grande D. Understanding why patients of low socioeconomic status prefer hospitals over ambulatory care. Health Aff (Millwood). 2013;32(7):1196–203. Go to the article, Google Scholar
• 47 Mortensen K. Copayments did not reduce Medicaid enrollees’ nonemergency use of emergency departments. Health Aff (Millwood). 2010;29(9):1643–50. Go to the article, Google Scholar
• 48 Kellermann AL, Weinick RM. Emergency departments, Medicaid costs, and access to primary care—understanding the link. N Engl J Med. 2012;366(23):2141–3. Crossref, Medline, Google Scholar
• 49 Oh A, Martsolf GR, Friedberg MW. Association between sponsorship and findings of medical home evaluations. JAMA Intern Med. 2017;177(9):1375–6. Crossref, Medline, Google Scholar
• 50 Schuur JD, Lin M, Muhlestein DB. 4^EMF engagement of accountable care organizations in acute care redesign: results of a national survey. Ann Emerg Med. 2017;70(4 Suppl):S171. Crossref, Google Scholar
• 51 Markovich P. A global budget pilot project among provider partners and Blue Shield of California led to savings in first two years. Health Aff (Millwood). 2012;31(9):1969–76. Go to the article, Google Scholar
• 52 Selevan J, Kindermann D, Pines JM, Fields WW. What accountable care organizations can learn from Kaiser Permanente California’s acute care strategy. Popul Health Manag. 2015;18(4):233–6. Crossref, Medline, Google Scholar