Nonphysician Clinicians

The Cost Effectiveness of Nurse Practitioners

Many studies have claimed that Nurse Practitioners (NPs)–as well as Physician Assistants (PAs)–are adequate substitutes for primary care physicians.  Researchers claim that NPs can perform a great majority of the tasks currently carried out by primary care physicians, and should be used more frequently since NP and PA salaries are usually half of primary care physicians.  The Bureau of Labor Statistics reports that the median PA income in 2004 was $69,410, as compared to a family practice doctor with less than two years of specialization who would earn a median salary of $137,199.  

Two of the more reliable studies which analyze the cost-effectiveness of Nurse Practitioners are those of Mundinger, et al. (2000) and Ettner, et al. (2005).  In the Mundinger study, individuals were randomly assigned to one of two clinics.  The first clinic was staffed with nurse practitioners and the other clinic was staffed by primary care physicians.  Both clinics had access to the same pool of specialists, inpatient units, and emergency departments.  A follow-up survey was conducted after six months.  The study concludes that:

  • There is no significant difference in patient health status, with the exception that in patients with hypertension, the diastolic value was statistically significantly lower for NP patients.
  • There were no significant differences in utilization rates between the two clinics.
  • There were no significant differences in patients satisfaction, with the exception that after 6 months, the patients rated physicians higher (4.2 vs. 4.1 on a scale of 1-5; P=.05) in the category of provider attributes.

There were some problems with this study however.  First, the patient base was almost exclusively Dominican immigrants on Medicaid.  Thus, it would be difficult to generalize these findings (especially those on patient satisfaction) to the wider U.S. population.  Secondly, it is possible that the physicians were of superior quality, but because measuring the true value of medical inputs on patient health is difficult, this might have caused the lack of any statistically significant difference.  Otherwise, this study is fairly robust.

In Ettner, et al. (2006), the researchers divided the 5th floor of an academic medical center into two wings.  In the West wing medical personnel used a traditional style of care but in the East wing there was an intervention which “…consisted of adding a nurse practitioner (NP) to each of the 2 general medicine teams on 5E” as well as some other changes in care practices.  The authors find the following:

  •  Intervention costs were $1187 per patients but savings were $3331 per patient resulting in a net benefit per patient of $1484. 
  • Of the total savings amount ($3331), $1947 was due to reduced cost during the impatient stay and the remainder was due to decreased hospital utilization after the discharge.
  • The authors re-run the analysis to take into account the possibility of attrition bias and find that the net benefits are still $947/patient.

The biggest problem with this study is that it analyzes a Multi-Disciplinary Doctor-NP Model (MDNP) model which involves a variety of changes in how provider teams treat patients.  It is difficult to analyze how much of the net savings is from the use of a nurse practitioner and how much of the savings is through more effective management of medical personnel.  Also, one must worry that the Hawthorne effect and not MDNP may have been the cause of additional worker productivity in the East wing.  The study does use robust statistical methods and puts forth a convincing argument for the use of MDNP in more hospitals. 

Mundinger, Kane, Lenz, Totten, Tsai, Cleary, Friedewald, Siu, Shelanski (2000); “Primary care outcomes in patients treated by nurse practitioners of physicians: A randomized trial,” JAMA, Vol 283(1), pp. 59-68.

Ettner, Kotlerman, Afifi, Vazirani, Hays, Shapiro, Cowan (2006); “Reducing the costs of patient care? A controlled trial of the Multi-Disciplinary Doctor-Nurse Practitioner (MDNP) model,” Medical Decision Making, Jan-Feb; pp. 9-17.