Economic Impact of a Primary Care Career: A Harsh Reality for Medical Students and the Nation : Academic Medicine

Secondary Logo

Journal Logo

Primary Care Careers

Economic Impact of a Primary Care Career: A Harsh Reality for Medical Students and the Nation

Palmeri, Martin MD, MBA; Pipas, Catherine MD; Wadsworth, Eric PhD, MBA; Zubkoff, Michael PhD

Author Information
doi: 10.1097/ACM.0b013e3181f5b754


Primary care (PC) plays an integral role in the overall health tapestry of the United States. Recent health care reforms focus on preventive care and quality measures that rely on a strong base of PC physicians,1 but fewer U.S. medical graduates are entering PC practices.2 Numerous scholars have explored the role of medical school training experiences, educational debt, and specialty reimbursement differences in hopes of understanding the complex interplay of factors that go into choosing a career in PC; however, no publications to date have detailed a young physician's expenses or explored the relationship between fiscal budgetary constraints and PC reimbursement even though financial planners routinely conduct such budgetary analyses on an individual basis.

Becoming a physician involves many expenses, which include predominantly educational loan costs, lost opportunity costs, and delayed savings and investing penalties. Using financial planning tools, we attempted to quantify the expenses that a young PC physician in the United States will face when entering her or his career. As the debate regarding the value of PC services and physician reimbursement continues, we hope this analysis will serve as a quantitative foundation for policy makers to rationally assess PC physician income and expenses.


Model structure

We explored a PC physician's income and developed a budget consisting of major expense drivers. Economic modeling is based on a variety of dynamic factors that can fluctuate from year to year and/or from state to state. Our model uses the data available in mid-2009. Many economic analyses typically do not describe the assumptions on which they are based to the degree of detail we have (List 1). Detailing our assumptions allows for greater transparency and for the reproducibility of our results for secondary analysis. As many of these assumptions are open to debate, we conducted a sensitivity analysis to evaluate potential differences in low, average, and high expense models.

List 1 Assumptions Used in Analysis

Regarding retirement planning, we used publicly available models. The T. Rowe Price (a global investment management firm) retirement planner uses an Ibbotson simulation supported by Morningstar (a financial resource specializing in fund investing). Additional information regarding Ibbotson simulations and Monte Carlo modeling are available in the paper by Ibbotson,3 “National savings rate guidelines for individuals.”

Data sources

We compiled physician income from published reports using data from the Physician Compensation Survey and the Medical Group Management Association.2,4 We based our educational loan costs on data from the Association of American Medical Colleges (AAMC) Graduation Questionnaire All Schools Reports that were published by Steinbrook.5 We obtained the data regarding home pricing from the National Association of Realtors, and we based interest rates on Bank of America's most recently published rates (2009). We obtained the data regarding college costs and inflation from the Bureau of Labor Statistics, U.S. News and World Report, and the College Board. Because the literature reports no specific data for medical student budgetary expenditures, we based the “other costs” used in the analysis on professional cohort data from the Bureau of Labor Statistics. Some evidence suggests that physicians have a higher relative consumption (i.e., more expensive life insurance, homes, cars) compared to similar income cohorts,6 but we could find no quantifiable data to use in this analysis.


We used the following basic financial formulas or iterations thereof:

Future value = Present value × (1 + i)t


Present value =


  • i = interest rate or rate of return
  • t = number of periods involved in the analysis
  • PMT = payment amount or cash flows (loan payment, mortgage payment, college savings, etc.)

Both Microsoft Excel (version 2003) and a standard financial calculator have present value and future value functions. For the model, we used both tools (see Supplemental Digital Personal Financial Calculator,

Calculating net income

In calculating net physician income after expenses, we used averaged low cost and high cost parameters to develop a model that would be most applicable to a medical student pool; however, in an effort to reflect the possible range of results, we explored several extremes including various debt loads, low expense models, and high expense models.


Educational loan costs

Over the past few years, sizeable increases in the amount of medical student debt have occurred. In 2007, 87% of medical students had debt at an average of $145,000 for students attending public medical schools and an average of $180,000 for students attending private schools.5 Data from the AAMC Graduation Questionnaire All Schools Reports shows that approximately 23% of medical students (a number that in 2008 was more than double the percentage in 2004) have more than $200,000 in debt5 (Figure 1). Also disconcerting, the interest rate on federal Stafford loans has been increasing for an unsubsidized loan—at the time of this analysis in 2009, the rate was 6.5% (the rate goes from 6.8% to 6.5% after the initial payment). For the purpose of this model, we used an average of $162,500 for the individual educational loan debt. If a graduate defers the payment of the loan for three years for residency training, the total debt is $199,159. Amortized over 10 years, this debt results in a monthly payment of $2,261.

Figure 1:
Distribution of total educational debt for medical students in the United States at graduation, 2004–2008. Debt figures include premedical or college education debt. In 2008, 38% of graduates had debt from premedical education (median amount, $20,000). The distribution of the 23% of students with total debt of $200,000 or more was 15% with $200,000 to $249,999 in debt, 6% with $250,000 to $299,999 in debt, and 3% with >$300,000 in debt (numbers do not sum to 23% because of rounding). Data from the AAMC Graduation Questionnaire All Schools Reports. Source: Steinbrook R. N Engl J Med. 2008;359:2629–2632. © 2008 Massachusetts Medical Society. All rights reserved. Used with permission.

Retirement savings

Several mechanisms are available for calculating retirement savings needs, and these needs—in and of themselves—can be quite variable depending on economic conditions, health care costs, life expectancy, and Social Security. For the purpose of this calculation, we compared several online retirement savings planners. List 1 describes some of the parameters we used to determine monthly retirement savings. We entered these parameters into the T. Rowe Price and Bank of America online retirement planning applications,7,8 which yielded a figure of $895 a month for retirement savings.

Home expenses

Reliable data for the parameters of the home expenses search were more difficult to secure because home prices vary regionally. According to the National Association of Realtors, the mean existing home price in 2008 was $242,700.9 As of 2009, the reported 30-year conventional fixed mortgage annual percentage rate was 4.908%.10 At 4.908% interest across 30 years, a $242,700 mortgage would require a monthly payment of $1,302.

The Bureau of Labor Statistics publishes annual average expenses for various populations. The bureau notes that master's, professional, or doctorate degree households spend, on average, $5,181 annually for property taxes, home maintenance, and insurance.11 Budgeted monthly, these expenses come to $432.

Children's college savings

Some could consider children's college savings an extraneous expense, but it is one that most families agree is important to their goals and sense of responsibility. For calculating children's college savings, we used several educational and bank resources that provide information on future educational costs and savings requirements. In addition, recent reports show net tuition rising 8.8% a year at public colleges and 6.7% a year at private ones.12 Using the average of in-state and private college tuition13 with a tuition inflation rate of 7%, we calculated an approximate future college cost of $686,884.14 Assuming that one starts saving at age 30 and that return on investment is 5%, the monthly contribution would be $1,967.

Other costs

Referring again to the 2007 Bureau of Labor Statistics data on master's, professional, or doctorate degree households, we found estimates for several other average annual expenses, including food ($8,734), utilities ($4,254), housekeeping supplies ($1,169), vehicle maintenance/gasoline ($6,030), health care ($4,105), and life/personal insurance ($538). Yearly, these expenses total an additional $24,83011—or $2,069 monthly. Of note, “other costs” do not include discretionary spending; therefore, categories such as clothing, entertainment, hobbies, and travel are not accounted for.

Net physician income

As noted previously, the average starting salary for a PC physician (including family medicine, pediatrics, and internal medicine) is $130,000.2 Assuming an average tax rate of 25%, the after-tax monthly income is $8,125. In Table 1, the calculated expenses for education loan repayment, retirement savings, children's college savings, home mortgage, and other household expenses are deducted from a PC physician's average monthly income. This results in a net negative balance of $801. Using average starting salary data from 2007,2Table 1 also compares the results for PC with those for psychiatry and radiology.

Table 1:
Monthly Budgetary Shortfalls of Primary Care Compared With Other Specialties

Figure 2 plots various expense models (high costs, average costs, and low costs) against after-tax PC earnings. The low cost analysis includes the following parameters:

  • medical students had no educational debt (reflective of approximately 13% of the medical student pool5),
  • medical students tailored college savings for in-state public tuition for two children, and
  • no changes to Social Security benefits occurred.
Figure 2:
Low, average, and high expense budget models compared with after-tax primary care (PC) physician income. A budget reflecting higher projected expenses and a budget reflecting minimal expenses are plotted against a PC physician's after-tax earnings. The higher expense model will result in budgetary shortfalls for approximately seven years. The low expense model will have budgetary surpluses immediately on entering into practice. An average of the high and low models is also plotted, resulting in a budgetary shortfall of approximately two years.

We based the high cost analysis on the following parameters:

  • medical students' educational debt equaled the average for the upper cohort ($180,000),
  • medical students deferred debt payments for the three years of residency,
  • medical students tailored college savings for sending two children to private schools, and
  • medical students did not rely on any Social Security benefits.

Figure 3 explores the effect of altering only medical student debt within the average expense model. As illustrated, high debt burden can create a scenario in which it could take more than three years for income to exceed projected expenses.

Figure 3:
Comparison of various educational debt levels with after-tax primary care (PC) physician income. Medical student educational debt levels of $220,000, $140,000, and $60,000 are entered into an average expense budgetary model and are plotted against a PC physician's after-tax earnings. Within an average expense model, higher debt will result in budgetary shortfalls for approximately 3.5 years. An educational debt of $60,000 will result in a budgetary surplus immediately on entering into practice.

An important point to consider for the above analysis is that during the period when expenses exceed earnings, students will need to make decisions about foregoing contributions to certain accounts such as retirement and college savings. Depending on the saving and investment strategy, significantly higher rates of savings or payments will be necessary if delayed until midcareer.

Finally, another number we can add to this equation is the opportunity cost of becoming a physician. We define opportunity costs as earnings students forego while they are pursuing a medical education; for example, we compare the physician with a person who enters the workforce immediately after high school. Assuming the latter earns an average of $20,000, the physician has an opportunity cost of $190,982—amortized over 20 years, this cost is $1,260 monthly. If we include this opportunity cost in the calculations, the net monthly deficit for the first year of PC practice increases to $2,061. We included opportunity costs only in the high expense budgetary model.


Our results both support and challenge the conventional wisdom that a career in PC is financially unsound. As Figures 2 and 3 illustrate, PC physicians in the first three to five years following residency will continue to have expenses that exceed their earnings. This finding aligns with those in a recent publication noting that PC physicians do not begin to have a positive cumulative net worth until approximately age 33.15 Therefore, the first few years after residency will be financially lean but feasible (except in the high expense model) if the newly practicing physician reigns in expenses by, for example, extending educational debt repayment from 10 to 20 years. The two caveats to this are (1) that starting wages need to be approximately $130,000 and (2) that income quickly rises to the mean of $188,272.2

Some may argue that physicians could defer saving for college and retirement until their earnings increase sufficiently, but deferring saving would result in having to save more aggressively five years down the road. For instance, a 5-year delay in retirement savings (from 30 to 35) requires a 50% increase in annual funding. A 10-year delay (pushing the start date to age 40) necessitates doubling the annual savings.3 This approach is not recommended by financial planners.

Every individual's circumstances are unique, and certain preferences, such as living in a region of the country with high home prices, high taxes, and/or lower average reimbursement, will alter one's budgetary restrictions. Such uncertainty could serve as a significant deterrent for medical students considering PC. Surprisingly, despite these risks, once PC physicians attain mean income levels, they experience a significant increase in discretionary income.

Physicians cannot make the decision to follow a PC career path lightly. Although declining interest in the gate-keeping disciplines of internal medicine, family medicine, and pediatrics suggests that students intuitively grasp this problem, these data make the career choice disincentives explicit. Clearly, subspecialization allows a budget surplus for discretionary spending, ranging from modest to substantial. The financial situation of those who pursue subspecialization contrasts sharply with the dilemma of PC physicians who face difficult economic choices, as this paper demonstrates. The stark differences perhaps distort the choices newly minted physicians might make otherwise (e.g., choosing a subspecialty because of monetary concerns).

This analysis should be viewed as a budgetary scaffold and not in absolute terms. The essence of the analysis is to provide a financial benchmark to guide PC policy makers and to serve as a warning that any erosion in PC physician income may make a career as an internist, family physician, or pediatrician untenable. Rising interest rates on educational loans, increasing student debt, declining Medicare/Medicaid reimbursement, and inflation can significantly reduce PC physician income.


Several of our assumptions might be subject to criticism. We designed the assumptions to be from the prospective perspective of a medical student using average nationwide data. However, the transparency of our assumptions could allow for greater interpretation and criticism of the results when applied to a broader audience. For example, average home prices are higher in Massachusetts than in South Dakota, and taxes are higher in New York compared with Tennessee. Using regional or personal variations would alter the final net income result. For the purpose of this analysis, we selected very conservative assumptions. The majority of potential revisions to our assumptions (higher home costs, higher college savings, economic downturns, greater budgeting for other expenses) would result in a larger budgetary deficit for the PC physician.

The assumptions most likely to lead to criticism are our choice of a 25% average tax rate. The average tax rate is the ratio of taxes paid to taxable income. This tax rate accounts for applicable tax deductions to which an individual is subject. The marginal tax rate is the “tax bracket” that is based on a taxpayer's last dollar of earnings, which is generally higher than the average tax rate. The average tax rate would be a conservative estimate of taxable income for a PC cohort.

Another concern is the assumption that the physician is the sole wage earner. Many physicians are married to spouses or otherwise live with partners who work and contribute to household income; however, when single medical students are exploring their career choices, they cannot presume that they will have a future spouse or partner whose earnings will offset any financial shortfalls from the cost of their specialty.

Life expectancy and duration of retirement savings might also be subject to criticism. A variety of industry recommendations are available to help consumers with retirement planning, but an industry standard is to err on the side of living longer so that a person does not prematurely exhaust his or her resources. Recommended ages for death in these calculations ranged from 85 to 95. For the purpose of this analysis, we used 90 years.

Despite the multitude of variables that can be adjusted, only a few seemed to have significant effects on net income, particularly medical student debt.

Future directions

The passage of the Patient Protection and Affordable Care Act (PPACA) relies heavily on a strong PC physician foundation.16 A growing amount of evidence shows that increased reliance on subspecialist care leads to higher expenditures, increased mortality, and lower quality.17 However, states like Massachusetts have discovered that guaranteeing coverage does not guarantee access18,19; therefore, the theoretical financial and health advantages of improved preventive services and decreased emergency room care do not necessarily come to fruition.

To achieve high-quality health care for the entire United States, Americans must first have access to PC, and thus, the PC workforce must grow in an urgent and timely fashion. The U.S. Department of Health and Human Services estimates that 16,643 PC providers are currently needed to address care shortages.20 The AAMC's Center for Workforce Studies projects that when the PPACA is fully implemented by 2025, this shortage will be 46,000 PC physicians.21

The medical community must rejuvenate student interest in PC. As the percentage of U.S. medical students choosing PC continues to decline,22 policy makers and medical leaders must address the real and perceived financial risks we have outlined in this report. Innovative support and financial security can overcome the barriers in both recruitment and retention of PC physicians.

The success of the PPACA depends in part on providing every American access to care and ensuring high-quality, affordable health care for all Americans. We must begin to close the salary gap, ensure that quality care models and reimbursements align, and provide loan repayment incentives directed to PC debt load. The National Health Service of England has successfully increased PC income 58% to improve recruitment.23 The United States can achieve such success if the current models in innovative payments continue and grow. Some of these innovative payment schemes include increased fee-for-service for PC, expanded per-member, per-month fees for non-face-to-face care, and shared savings efforts to support quality and cost reduction. The PPACA addresses educational debt with limited loan repayment options (up to $35,000 a year) for PC physicians who agree to work in high-need areas.16 Whether these incentives affect PC physicians' career decisions significantly enough to meet the current PC needs is an area in need of additional study. A five-year, 10% increase in Medicare reimbursement starting in 2011 for PC physicians has also been proposed but has yet to be enacted.18 Successful efforts to develop and align high-value care models and outcomes with reimbursement are ongoing. Such efforts include national endorsement and support for a new, patient-centered medical home model that encompasses comprehensive, high-quality, and low-cost efforts.24,25 We now have a voluntary assessment tool, the Physician Practice Connections–Patient- Centered Medical Home, developed by the National Committee for Quality Assurance, which physicians can use to measure practice performance and to qualify their practice as a patient- centered medical home, thus becoming eligible for additional reimbursement benefits.26

Transformation of the reimbursement models to match the reengineered practice models is the key change to align, on one hand, compensation and incentives with, on the other, measures attributed to high quality and low cost. It is necessary to have these reimbursements in direct alignment with the activities of the PC practice. Appropriate compensation for the new activities is necessary to promote their success. Two such activities, coordination of care and team-based care, take time and would, in the current visit- or volume-based, episodic, fee-for-service system, be penalized. We know that better coordination and continuity of care can result in greater satisfaction, higher quality, and lower costs.27,28

Global changes to the PC reimbursement models have the potential to improve the quality of care provided as well as the physician practice environment and job satisfaction. The ramifications of current reforms will take years to fully have an effect that can translate into a significant motivator for medical student career choices. The need for more PC physicians cannot wait. Based on the models we provide above, a competitive PC starting salary covering full loan repayment would significantly or completely reduce the financial deficits and risks PC physicians face early in their careers. Several loan repayment programs are available; however, many such incentives have numerous restrictions and obligations attached. Liberalizing such programs or having novel tax incentives for practices to forgive PC debt could be a powerful motivator for young physicians.

Developing incentives—financial and otherwise—for recruiting PC physicians is not a simple task, but the consequences to the health and economy of the United States are grave if we do not provide a comprehensive and coordinated approach to building a skilled and solid PC foundation.



Other disclosures:


Ethical approval:

Not applicable.


1 The White House. The Agenda: Health Care. Available at: [no longer available]. Accessed March 2009.
2 Ebell M. Future salary and US residency fill rate revisited. JAMA. 2008;300:1131–1132.
3 Ibbotson R, Xiong J, Kreitler R, Kreitler CF, Chen P. National savings rate guidelines for individuals. J Financ Plan. 2007;20:50–61. Available at: Accessed August 3, 2010.
4 Bodenheimer T, Berenson R, Rudolf P. The primary care–specialty income gap: Why it matters. Ann Intern Med. 2007;146:301–306.
5 Steinbrook R. Medical student debt—Is there a limit? N Engl J Med. 2008;359:2629–2632.
6 Stanley TJ, Danko WD. The Millionaire Next Door: The Surprising Secrets of America's Wealthy. New York, NY: Pocket Books; 1996.
7 T. Rowe Price. Retirement. Available at: Accessed July 23, 2010.
8 Bank of America. Individual retirement accounts. Available at: Accessed July 23, 2010.
9 National Association of Realtors. Existing-home sales price and sales statistics. Available at: Accessed July 23, 2010.
10 Bank of America. Mortgage purchase: Rates. Available at: Accessed June 20, 2010.
11 Bureau of Labor Statistics. Table 10: Education of reference person: Average annual expenditures and characteristics. In: 2007 Consumer Expenditure Survey. Available at: Accessed July 23, 2010.
12 Glater JD. College costs outpace inflation rate. New York Times. October 23, 2007. Available at: Accessed July 23, 2010.
13 The College Board. College cost calculator. Available at: Accessed on July 23, 2010.
14 Paying of College. U.S. News and World Reports: Loan calculator and amortization. Available at: Accessed on June 28, 2010.
15 Vaughn B, DeVrieze S, Reed SD, Schulman KA. Can we close the income and wealth gap between specialists and primary care physicians? Health Aff (Millwood). 2010;29:933–940.
16 Government Printing Office. Patient Protection and Affordable Care Act. Available at: Accessed July 23, 2010.
17 Wennberg JE, Fisher ES. Tracking the Care of Patients With Severe Chronic Illness: The Dartmouth Atlas of Health Care 2008. Lebanon, NH: The Dartmouth Institute for Health Policy and Clinical Practice, Center for Health Policy Research; 2008.
18 Iglehart JK. Health reform, primary care, and graduate medical education. N Engl J Med. 2010;363:584–590. Available at: Accessed July 30, 2010.
19 Massachusetts Medical Society. Physician Workforce Study. Available at: Accessed July 30, 2010.
20 U.S. Department of Health and Human Services. Shortage designation: HPSAs, MUAs & MUPs. Available at: Accessed July 23, 2010.
21 O'Reilly K. Health reform's next challenge: Who will care for the newly insured? Am Med News. April 12, 2010. Available at: Accessed July 23, 2010.
22 Barr M; American College of Physicians. The Advanced Medical Home: A Patient-Centered, Physician-Guided Model of Health Care. Available at: Accessed July 23, 2010.
23 Ham C. Chronic care in the English National Health Service: Progress and challenges. Health Aff (Millwood). 2009;28:190–201.
24 Nutting PA, Miller WL, Crabtree BF, Jaen CR, Stewart EF, Stange KC. Initial lessons from the first national demonstration project on practice transformation to a patient-centered medical home. Ann Fam Med. 2009;7:254–260.
25 Rittenhouse D, Shortell S. The patient-centered medical home: Will it stand the test of health reform? JAMA. 2009;201:2038–2040.
26 Takach M, Kaye N, Beesla R. Strategies states can use to support the infrastructure of a medical home. State Health Policy Briefing. May 2008;2. Available at: Accessed July 23, 2010.
27 Atlas SJ, Grant RW, Ferris TG, Chang Y, Barry MJ. Patient–physician connectedness and quality of primary care. Ann Intern Med. 2009;150:325–335.
28 Zerehi MR. How Is a Shortage of Primary Care Physicians Affecting the Quality and Cost of Medical Care? A Comprehensive Evidence Review. Available at: Accessed July 23, 2010.

Supplemental Digital Content

© 2010 Association of American Medical Colleges