Adjusting for bias introduced by instrumental variable estimation in the Cox proportional hazards model.

Instrumental variable (IV) methods are widely used for estimating average treatment effects in the presence of unmeasured confounders. However, the capability of existing IV procedures, and most notably the two-stage residual inclusion (2SRI) algorithm recommended for use in non-linear contexts, to account for unmeasured confounders in the Cox proportional hazard model is unclear. We show that instrumenting an endogenous treatment induces an unmeasured covariate, referred to as an individual frailty in survival analysis parlance, which if not accounted for leads to bias. We propose a new procedure that augments 2SRI with an individual frailty and prove that it is consistent under certain conditions. The finite sample-size behavior is studied across a broad set of conditions via Monte Carlo simulations. Finally, the proposed methodology is used to estimate the average effect of carotid endarterectomy versus carotid stenting on the mortality of patients suffering from carotid artery disease. Results suggest that the 2SRI-frailty estimator generally reduces the bias of both point and interval estimators compared to traditional 2SRI.

How fast will the registered nurse workforce grow through 2030? Projections in nine regions of the country.

BACKGROUND: After an unprecedented increase in nursing school enrollment and graduates in the past 10 years, projected shortages of nurses have been erased at a national level. However, nursing markets are local, and an uneven distribution of health care providers of all types is a longstanding feature of health care in the United States. PURPOSE: The purpose of this study was to understand how the outlook for future registered nurse (RN) supply varies regionally across the United States. METHODS: We apply our nursing supply model to the nine U.S. Census Divisions to produce separate supply forecasts for each region. DISCUSSION: We find dramatic differences in expected future growth of the nursing workforce across U.S. regions. These range from zero expected growth in the number of RNs per capita in New England and in the Pacific regions between 2015 and 2030 to 40% growth in the East South Central region (Mississippi, Alabama, Tennessee, Kentucky) and in the West South Central region (Texas, Oklahoma, Arkansas, Louisiana). CONCLUSION: Assuming growth in the demand for RNs per population, some regions of the United States are expected to face shortfalls in their nursing workforce if recent trends do not change.

Recent Changes in the Number of Nurses Graduating from Undergraduate and Graduate Programs.

Since the 1970s, a number of initiatives have attempted to increase the proportion of nursing graduates with a baccalaureate degree, but with little national effect. Now market forces, health reforms, and an Institute of Medicine report (2011) have combined to transform the educational composition of the nursing workforce. Today, there are considerably more graduates of baccalaureate nursing programs than associate degree programs. The educational transformation of the nursing workforce is not limited to baccalaureate education but includes the rapidly increasing numbers of registered nurses who have earned graduate degrees. These changes in nursing education are increasing the readiness of nursing professionals to capitalize on new opportunities, overcome challenges, and take on new roles and responsibilities as the nation's health care delivery and payments systems evolve in coming years.

Will the RN Workforce Weather the Retirement of the Baby Boomers?

IMPORTANCE: After forecasts made more than a decade ago suggested dire nursing shortages, enrollment in nursing schools doubled. The implications of this unprecedented change for the nursing workforce have not yet been fully explored. OBJECTIVE: To forecast the size and age distribution of the nursing workforce to the year 2030 and to compare to demand recently projected by the Health Resources and Services Agency. DESIGN: A retrospective cohort analysis of employment trends by birth year and age were used to project age and employment of registered nurses (RNs) through 2030. SETTING: Data on employed RNs from the United States Bureau of the Census Current Population Survey (1979-2000, N=72,222) and American Community Survey (2001-2013, N=342,712). PARTICIPANTS: RNs between the ages of 23 and 69 years. MAIN OUTCOME MEASURE: Annual full-time equivalent (FTE) employment of RNs in total and by single year of age. RESULTS: Annual retirements from the nursing workforce will accelerate from 20,000 a decade ago to near 80,000 in the next decade as baby boomer RNs continue to age. We project that this outflow will be more than offset by continued strong entry of new RNs into the workforce. Overall, we project that the registered nursing workforce will increase from roughly 2.7 million FTE RNs in 2013 to 3.3 million in 2030. We also find that the workforce will reach its peak average age in 2015 at 44.4. This increase in workforce size, which was not expected in forecasts made a decade ago, is contingent on new entry into nursing continuing at its current rate. Even then, supply would still fall short of demand as recently projected by the Health Resources and Services Agency in the year 2025 by 128,000 RNs (4%). CONCLUSIONS: The unexpected surge of entry of new RNs into the workforce will lead to continued net growth of the nursing workforce, both in absolute FTE and FTE per capita. While this growth may not be sufficient to meet demand, such projections are uncertain in the face of a rapidly evolving health care delivery system.

Nurse Staffing in Neonatal Intensive Care Units in the United States.

The neonatal intensive care unit (NICU) is a setting with high nurse-to-patient ratios. Little is known about the factors that determine nurse workload and assignment. The goals of this study were to (1) develop a measure of NICU infant acuity; (2) describe the acuity distribution of NICU infants; (3) describe the nurse/infant ratio at each acuity level, and examine the factors other than acuity, including nurse qualifications and the availability of physicians and other providers, that determined staffing ratios; and (4) explore whether nurse qualifications were related to the acuity of assigned infants. In a two-stage cohort study, data were collected in 104 NICUs in 2008 by nurse survey (6,038 nurses and 15,191 infants assigned to them) and administrators reported on unit-level staffing of non-nurse providers; in a subset of 70 NICUs in 2009-2010, census data were collected on four selected shifts (3,871 nurses and 9,276 infants assigned to them). Most NICU infants (62%) were low-acuity (Levels 1 and 2); 12% of infants were high-acuity (Levels 4 and 5). The nurse-to-infant ratio ranged from 0.33 for the lowest-acuity infants to 0.95 for the highest-acuity infants. The staffing ratio was significantly related to the acuity of assigned infants but not to nurse education, experience, certification, or availability of other providers. There was a significant but small difference in the percentage of high-acuity (Levels 4 and 5) infants assigned to nurses with specialty certification (15% vs. 12% for nurses without certification). These staffing patterns may not optimize patient outcomes in this highly intensive pediatric care setting.

Do Associate Degree Registered Nurses Fare Differently in the Nurse Labor Market Compared to Baccalaureate-Prepared RNs?

Roughly 40% of the nearly 3 million registered nurses (RNs) in the United States have an associate's degree (ADN) as their highest level of nursing education. Yet even before the recent Institute of Medicine report on The Future of Nursing, employers of RNs have increasingly preferred baccalaureate-prepared RNs (BSNs), at least anecdotally. Data from the American Community Survey (2003-2013) were analyzed with respect to employment setting, earnings, and employment outcomes of ADN and BSN-prepared RNs. The data reveal a divergence in employment setting: the percentage of ADN-prepared RNs employed in hospitals dropped from 65% to 60% while the percentage of BSN-prepared RNs employed in hospitals grew from 67% to 72% over this period. Many ADNs who would have otherwise been employed in hospitals seem to have shifted to long-term care settings.

Health care reform and the health care workforce--the Massachusetts experience.

Massachusetts' experience with health care reform may be predictive of the effects of national health care reform. Data on employment in the health care industry were examined to determine the impact of the Massachusetts reform on the state's health care workforce.

Reliability of surgical outcomes for predicting future hospital performance.

BACKGROUND: Because of small sample sizes and low event rates, risk-adjusted surgical outcomes often do not meet reliability benchmarks for distinguishing hospital performance. Nonetheless, it is unclear whether these measures may still be useful for predicting future hospital surgical performance. METHODS: We used national Medicare data to analyze patients undergoing colectomy from 2007 to 2010 (n=462,959 patients). We first quantified 2007-2008 outcome reliability (ability to differentiate quality differences) and ranked hospitals based on their 2007-2008 risk-adjusted outcome rates. To assess the ability of adjusted outcomes to predict true performance, we evaluated future (2009-2010) outcomes across quintiles of past performance. We then systematically sampled 2007-2008 cases to evaluate performance prediction when hospitals' past performance was measured with progressively lower reliability levels. RESULTS: Outcomes in 2007-2008 were good predictors of outcomes in the next 2 years (2009-2010), but predictive strength depended upon reliability. With progressive sampling of 2007-2008 caseloads, outcome reliability and predictive strength decreased. With 100% sampling of 2007-2008 caseloads, the worst versus best hospital quintile based on past performance had 1.52 [95% confidence interval (CI), 1.44-1.60] times the odds of mortality and 1.50 (95% CI, 1.44-1.56) times the odds of complications in 2009-2010. With 10% sampling, outcome reliability was well below commonly accepted benchmarks, but the worst quintile of hospitals in 2007-2008 still had 1.12 (95% CI, 1.06-1.19) times the odds of mortality and 1.16 (95% CI, 1.11-1.21) times the odds of complications in 2009-2010 compared with the best quintile of hospitals. CONCLUSIONS: Even at very low reliability levels, risk-adjusted outcome measures may distinguish best and worst hospitals' surgical performance. This study suggests that commonly accepted reliability thresholds may be too high, especially in the context of selective referral.

School Choice, School Quality and Postsecondary Attainment.

We study the impact of a public school choice lottery in Charlotte-Mecklenburg schools on college enrollment and degree completion. We find a significant overall increase in college attainment among lottery winners who attend their first choice school. Using rich administrative data on peers, teachers, course offerings and other inputs, we show that the impacts of choice are strongly predicted by gains on several measures of school quality. Gains in attainment are concentrated among girls. Girls respond to attending a better school with higher grades and increases in college-preparatory course-taking, while boys do not.

The Rapid Growth of Graduates From Associate, Baccalaureate, And Graduate Programs in Nursing.

Growth in the number of RN graduates from 2002-2012 has been dramatic and broad based, occurring between both associate and baccalaureate programs, and has included people from all racial and eth- nic backgrounds. This growth has occurred in all types of public, private not-for- profit, and proprietary institu- tions. The growth of RNs with gradu- ate degrees has also increased, particularly since 2004. Given the rapid production of nursing graduates, leaders in academic nursing education are urged to focus on the quality of nursing graduates, take steps to assure that graduates are well prepared for growth in nonhospital settings, ensure graduates are aware of the many challenges they will confront, and are well prepared to seize opportunities that will unfold during an era of health reform.

Projecting the Future Registered Nurse Workforce After the COVID-19 Pandemic.

Importance: Health care delivery systems rely on a well-prepared and adequately sized registered nurse (RN) workforce. The US RN workforce decreased by more than 100 000 in 2021 during the COVID-19 pandemic-a far greater single-year drop than observed over the past 4 decades. The implication for the longer-term growth of the RN workforce is unknown. Objective: To describe recent trends in RN employment through 2023 and forecast the growth of the RN workforce through 2035. Design, Setting, and Participants: Descriptive analysis of recent trends since the start of the COVID-19 pandemic in RN employment using data from the US Bureau of the Census Current Population Survey and including employed RNs aged 23 to 69 years from 1982 through 2023, and retrospective cohort analysis of employment trends by birth year and age to project the age distribution and employment of RNs through 2035. Main Outcome and Measures: Annual full-time equivalent (FTE) employment of RNs by age, demographics, and sector of employment; forecast of RN workforce by age through 2035. Results: The final sample included 455 085 RN respondents aged 23 to 69 years. After a sharp decline in 2021, RN employment recovered, and the total number of FTE RNs in 2022 and 2023 was 6% higher than in 2019 (3.35 million vs 3.16 million, respectively). Using data on employment, education, and population through 2022, the size of the RN workforce was projected to increase by roughly 1.2 million FTEs to 4.56 million by 2035, close to prepandemic forecasts. Growth will be driven primarily by RNs aged 35 to 49 years, who are projected to compose nearly half (47%) of the RN workforce in 2035, up from 38% in 2022. Conclusions and Relevance: In this study, the rebound in the total size of the US RN workforce during 2022 and 2023 indicates that the earlier drop in RN employment during the first 2 years of the COVID-19 pandemic was likely transitory. Updated forecasts of the future RN workforce are very close to those made before the pandemic.

Composite measures for profiling hospitals on surgical morbidity.

CONTEXT: Although risk-adjusted morbidity is widely used as a surgical quality indicator, it may not always be a reliable indicator of hospital quality. In this study, we assess the value of a novel composite measure for improving the reliability of hospital morbidity rankings. DESIGN, SETTING, AND PATIENTS: Using data from the American College of Surgeons' National Surgical Quality Improvement Program (ACS-NSQIP), we studied all patients undergoing 4 surgical procedures (2008-2009): colectomy, ventral hernia repair, abdominal aortic aneurysm repair, and lower extremity bypass surgery. For these procedures, we created a composite measure by combining quality indicators from several distinct domains of quality: morbidity, reoperation, length of stay, and morbidity with other potentially related procedures. We empirically weighted each measure and adjusted for reliability using empirical Bayes techniques. To validate this approach, we assessed how well composite measures from 1 year (2008) predict morbidity in the next year (2009) compared with the standard ACS-NSQIP approach for assessing hospital rates of risk-adjusted morbidity. RESULTS: For all 4 operations, the composite measures explained a higher proportion of hospital-level variation in morbidity than the standard approach: ventral hernia repair (58% for the composite vs 8% for the standard approach), colon resection (33% vs 14%), abdominal aortic aneurysm repair (51% vs 38%), and lower extremity bypass surgery (32% vs 3%). When evaluating the ability to discriminate future performance, the composite approach performed best for ventral hernia repair. For this procedure, the bottom 20% of hospitals based on the composite approach had nearly threefold higher (odds ratio: 2.65; 95% confidence interval: 1.83-3.85) morbidity rates than the top 20% of hospitals. However, when using the standard approach, there was only a 1.3-fold difference (odds ratio: 1.30; 95% confidence interval: 0.87-1.96). Although the differences were smaller in magnitude, the composite measure also outperformed the standard approach for the other 3 procedures. CONCLUSIONS: Composite measures better reflect hospital quality than simple rates of risk-adjusted morbidity. In the context of ACS-NSQIP, composite measures would give hospitals a better sense of where they stand and help identify truly exemplary hospitals for benchmarking.

Composite quality measures for common inpatient medical conditions.

BACKGROUND: Public reporting on quality aims to help patients select better hospitals. However, individual quality measures are suboptimal in identifying superior and inferior hospitals based on outcome performance. OBJECTIVE: To combine structure, process, and outcome measures into an empirically derived composite quality measure for heart failure (HF), acute myocardial infarction (AMI), and pneumonia (PNA). To assess how well the composite measure predicts future high and low performers, and explains variance in future hospital mortality. RESEARCH DESIGN: Using national Medicare data, we created a cohort of older patients treated at an acute care hospital for HF (n=1,203,595), AMI (n=625,595), or PNA (n=1,234,299). We ranked hospitals on the basis of their July 2005 to June 2008 performance on the composite. We then estimated the odds of future (July to December 2009) 30-day, risk-adjusted mortality at the worst versus best quintile of hospitals. We repeated this analysis using 2005-2008 performance on existing quality indicators, including mortality. RESULTS: The composite (vs. Hospital Compare) explained 68% (vs. 39%) of variation in future AMI mortality rates. In 2009, if an AMI patient had chosen a hospital in the worst versus best quintile of performance using 2005-2008 composite (vs. Hospital Compare) rankings, he or she would have had 1.61 (vs. 1.39) times the odds of dying in 30 days (P-value for difference <0.001). Results were similar for HF and PNA. CONCLUSIONS: Composite measures of quality for HF, AMI, and PNA performed better than existing measures at explaining variation in future mortality and predicting future high and low performers.

Projections of the long-term growth of the registered nurse workforce: a regional analysis.

Providing regional projections of the RN workforce will allow underlying differences in the age structure of the RN workforce to become more visible. By providing regional-level projections, it will also be possible to identify those regions whose RN workforce is expected to grow at a slower rate relative to other regions. States in the South and Midwest have a greater supply of younger-aged RNs available to replace fewer numbers of older-age RNs compared to other regions. In contrast, the Northeast and West have fewer younger RNs currently in their workforce yet a relatively larger number of older age RNs to replace. These differences in age structure may be partly due to differences in nursing school enrollment and expansion in nursing education capacity across regions. This information can help guide national and state health workforce planners, employers, educators, and others in developing policies and initiatives that may impact nursing supply in their states.

Characteristics of rural registered nurses and the implications for workforce policy.

PURPOSE: Rural registered nurses (RNs) play an integral role in providing care for an underserved population with worse health outcomes than urban counterparts. However, little information is available on the profile of this workforce, which is necessary to understand the capacity of these nurses to provide quality and demanded care presently and in the future. METHODS: We utilize data from the American Community Survey to provide a contemporary analysis on the supply of rural RNs in the United States. FINDINGS: While the number of physicians serving rural populations has decreased in recent years, and rural nurse practitioners (NPs) remain in short supply, rural RNs have steadily grown in numbers at a rate comparable to urban RNs. Rural RNs are markedly less diverse than the populations they serve and only half of rural RNs had a bachelor's degree or higher compared to over 70% for urban RNs. In their supply, young rural nurses appear on pace with urban nurses to adequately replace older nurses and continue to grow the workforce, based on data through 2019. CONCLUSIONS: The rural RN workforce is projected to steadily grow amidst declining rural physicians and limited rural NPs. The burgeoning investments in the rural health workforce present opportunities to help diversify, increase educational access, and further rural readiness for rural RNs moving forward.

Marriage, Children, and Sex-Based Differences in Physician Hours and Income.

Importance: A better understanding of the association between family structure and sex gaps in physician earnings and hours worked over the life cycle is needed to advance policies addressing persistent sex disparities. Objective: To investigate differences in earnings and hours worked for male and female physicians at various ages and family status. Design, Setting, and Participants: This retrospective, cross-sectional study used data on physicians aged 25 to 64 years responding to the American Community Survey between 2005 and 2019. Exposures: Earned income and work hours. Main Outcomes and Measures: Outcomes included annual earned income, usual hours worked per week, and earnings per hour worked. Gaps in earnings and hours by sex were calculated by family status and physician age and, in some analyses, adjusted for demographic characteristics and year of survey. Data analyses were conducted between 2019 and 2022. Results: The sample included 95 435 physicians (35.8% female, 64.2% male, 19.8% Asian, 4.8% Black, 5.9% Hispanic, 67.3% White, and 2.2% other race or ethnicity) with a mean (SD) age of 44.4 (10.4) years. Relative to male physicians, female physicians were more likely to be single (18.8% vs 11.2%) and less likely to have children (53.3% vs 58.2%). Male-female earnings gaps grew with age and, when accumulated from age 25 to 64 years, were approximately $1.6 million for single physicians, $2.5 million for married physicians without children, and $3.1 million for physicians with children. Gaps in earnings per hour did not vary by family structure, with male physicians earning between 21.4% and 23.9% more per hour than female physicians. The male-female gap in hours worked was 0.6% for single physicians, 7.0% for married physicians without children, and 17.5% for physicians with children. Conclusions and Relevance: In this cross-sectional study of US physicians, marriage and children were associated with a greater earnings penalty for female physicians, primarily due to fewer hours worked relative to men. Addressing the barriers that lead to women working fewer hours could contribute to a reduction in the male-female earnings gap while helping to expand the effective physician workforce.

The nursing workforce: a comparison of three national surveys.

The termination of the National Sample Survey of Registered Nurses (NSSRN) represents a loss of a key source of information on the nursing workforce that has been available for more than 30 years. At the same time, this loss presents new opportunities to address some of the biases associated with using licensing data to construct lists of RNs to construct a sampling frame. Given the loss of this key national data source on the nursing workforce, the NSSRN is compared to existing alternatives for nursing workforce data in two U.S. Census Bureaus surveys: the Current Population Survey and the American Communities Survey. The endeavor of workforce planning and a deeper understanding of whether we are equipped to meet the nation's changing health care landscape will benefit from a thoughtful consideration of how best to obtain accurate and comparable data from future surveys.