Statewide and National Impact of California s Staffing Law on Pediatric Cardiac Surgery Outcomes

Transcription

1 JONA Volume 41, Number 5, pp Copyright B 2011 Wolters Kluwer Health Lippincott Williams & Wilkins THE JOURNAL OF NURSING ADMINISTRATION Statewide and National Impact of California s Staffing Law on Pediatric Cardiac Surgery Outcomes Patricia A. Hickey, PhD, MBA, RN, NEA-BC, FAAN Kimberlee Gauvreau, ScD Kathy Jenkins, MD, MPH Objective: The objective of the study was to examine the impact of staffing ratios on risk-adjusted outcomes for pediatric cardiac surgery programs in California and relative to other states combined. Background: California performs 20% of the nation s pediatric cardiac surgery and is the only state with a nurse ratio law. Understanding the imposition of mandated ratios on pediatric outcomes is necessary to inform the debate about nurse staffing. Data Sources: Patient variables were extracted from the Healthcare Cost and Utilization Project Kids Inpatient Database. The American Hospital Association database was used for institutional variables. Methods: Descriptive analyses were used to identify and describe patient, nursing, and hospital characteristics. Changes in nursing ratios and full-time equivalents (FTEs) between 2003 and 2006 were examined. Associations between nursing characteristics and each outcome variable were examined using general Author Affiliations: Vice President, Cardiovascular and Critical Care Services, Department of Nursing Patient Services (Dr Hickey), Research Associate in Cardiology, Department of Cardiology (Dr Gauvreau), Senior Vice President, Chief Patient Safety and Quality Officer, Program for Patient Safety and Quality (Dr Jenkins), Children s Hospital Boston; Department Chair (Dr Fawcett), Professor (Dr Hayman), College of Nursing and Health Sciences, University of Massachusetts, Boston. Correspondence: Dr Hickey, Cardiovascular and Critical Care Services, Bader 664, Children s Hospital Boston, 300 Longwood Ave, Boston, MA (Patricia.hickey@childrens.harvard.edu). Funding: This study was supported by funding from the Department of Nursing and the Cardiovascular Program, Children s Hospital Boston. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal s Web site ( DOI: /NNA.0b013e b2e Jacqueline Fawcett, PhD, RN, FAAN Laura Hayman, PhD, RN, FAAN estimating equation models. The RACHS-1 (Risk Adjustment for Congenital Heart Surgery) risk adjustment method was used for mortality. Results: Hospitals in California significantly increased RN FTEs (P =.025) and RN ratios (P =.036) after enactment of AB 394 in Neither RN FTEs nor RN ratios were associated with mortality, complications, or resource utilization after risk adjustment. After the law, California s standardized mortality ratio (SMR) decreased more (33%) than in all other states combined (29%). Standardized complication ratio (SCR) increased by 5% but decreased by 5% for all other states combined, and the increase in charge differential ($53,443) was more than twice the increase ($23,119) for other states combined. Conclusion: Hospitals in California made upward adjustments in nursing FTEs and ratios after enactment of AB 394. There was a substantial increase in California s charge differential, a decrease in SMR, and an increase in SCR after enactment of the legislation. As national healthcare reform evolves,substantial debate continues about the appropriate number of RNs necessary for provision of high-quality, cost-effective care in America s hospitals. Although there is no federal law that relates to nurse staffing, the state of California enacted the California Safe Staffing Law (AB 394) in 2004 and became a testing ground for mandated nurse staffing levels. To date, at least 17 states plus the District of Columbia have passed legislation and/or adopted regulations to address nurse staffing. 1 Although perspectives differ on how safe staffing should be achieved, the need for sufficient numbers of RNs to provide safe patient care is undisputed. 218 JONA Vol. 41, No. 5 May 2011

2 In the adult nursing literature, the associations between nurse staffing levels and patient outcomes of mortality and morbidity for hospitalized patients are well established. 2-6 However, the literature about nurse staffing and outcomes in California has been inconclusive. Since enactment of the law, several studies have demonstrated that although nursing full time equivalents (FTEs) may have increased in California, the anticipated positive impact on patient outcomes from the law was not actualized. 7,8 Interestingly, Aiken and colleagues 9 recently published results from a 3-state study comparing nurse workload and patient outcomes including mortality and failure to rescue. They concluded that California s mandated nurse staffing ratios were associated with lower mortality and nurse outcomes predictive of better retention. In pediatrics, the same relationship between nurse staffing and mortality and morbidity outcomes was not found. 10 Findings from a recent study by our group examined the relationship of nurse staffing to risk-adjusted mortality for congenital heart disease across 38 children s hospitals and did not reveal a significant association. 11 We concluded that mortality is probably insensitive to nursing characteristics as long as certain staffing thresholds have been achieved, which is often the case in pediatric hospitals. Nurse staffing in pediatrics is complex, and patient volume is a key driver. The state of California performs approximately 20% of the nation s pediatric cardiac surgery, and this group of patients consumes a disproportionately high share of hospital resources. 12 Building on our findings in children s hospitals, the purpose of this study was 2-fold: to identify and describe changes in RN FTEs and ratios that may have occurred because of enactment of AB 394 and to examine the impact of these changes on the clinical outcomes of risk-adjusted mortality, risk-adjusted complications, and risk-adjusted resource utilization for California versus other states combined. Conceptual Model The Conceptual Model of Nursing and Health Policy was used to guide this study. 13 (See Document, Supplemental Digital Content 1, Conceptual Model, Study Methods Data Sources The Healthcare Cost and Utilization Project Kids Inpatient Database (KID) is a set of pediatric hospital inpatient databases included in the Healthcare Cost and Utilization Project databases. (See page 2 of Document, Supplemental Digital Content 1, Data Sources section, American Hospital Association Annual Survey Database Institutional variables not available in the KID were analyzed by linking the KID 2003 and 2006 with American Hospital Association data for nurse staffing and nursing skill mix of specific institutions. The AHA data provide the most comprehensive data available on hospital organization structure, facilities, and personnel by occupational category, medical staff, finances, and administration ( Outcome Variables Three outcome variables measured at the patient level were obtained from the KID. Mortality was defined as in-hospital death, recorded as patient discharge status. A complication was defined as untoward harm associated with a therapeutic or diagnostic healthcare intervention, as measured by 20 International Classification of Diseases, Ninth Revision, Clinical Modification codes. Resource utilization was defined as total hospital charges. Nursing Variables Two nursing variables measured at the institution level were obtained from the AHA Survey Database. Fulltime equivalent was defined as the number of RN FTEs; each FTE is equal to 2,080 hours of worked time annually. Nursing ratio was defined as the numberofrnftesperpatient. Risk Adjustment Methods Three risk adjustment methods were used in this study. Risk Adjustment for Congenital Heart Surgery (RACHS-1) was used as a risk adjustment method for in-hospital mortality. 14 The Complication Screening Method for Congenital Heart Surgery 15 was used for the analysis of complications. Connor and colleagues 12 model for resource utilization was used to risk-adjust total hospital charges. (See page 2 of Document, Supplemental Digital Content 1, Risk Adjustment Methods section, Ethical Considerations and Institutional Review Board This study used deidentified secondary data. Therefore, the exempt category of approval was sought and approved by the institutional review boards of the University of Massachusetts, Boston, and Children s Hospital Boston. The databases were accessed by a single biostatistician for the duration of this study. JONA Vol. 41, No. 5 May

3 Statistical Analysis Patient demographic characteristics, clinical characteristics, and outcomes were summarized for children undergoing congenital heart surgery in California for calendar years 2003 and 2006 separately. Categorical variables were summarized using frequencies and percentages, and continuous characteristics were described using the mean, median, SD, range, and interquartile range. Nursing FTEs and nursing ratios were summarized at the hospital level using the median and interquartile range; the Wilcoxon signed rank test was used to assess differences between the 2 years. Associations between nursing ratios, nursing FTEs, and each of the 3 outcome variables were evaluated using generalized estimating equation models, which account for the intrainstitutional correlation among cases from the same hospital. The model for inhospital mortality included all factors that are part of the RACHS-1 risk adjustment method. Additional patient characteristics and institutional volume were also examined after adjusting for baseline risk; variables significant at the.10 level were retained in the model. Once a final model was determined, nursing FTEs and nursing ratio were added into the model separately; statistical significance was assessed using the likelihood ratio test. Similar analyses were performed for the outcomes, any complication, and total hospital charges. To investigate whether the California Safe Staffing Law impacted changes in risk-adjusted mortality, risk-adjusted complications, and risk-adjusted resource utilization in California relative to changes in other states combined, institutions in the KID data set performing congenital heart surgery in both 2003 and 2006 were identified. For the outcome inhospital mortality, a logistic regression analysis was used to fit the risk adjustment model for in-hospital mortality in the combined data set. For California and then for all other states combined, the observed mortality rate was defined as the number of cases in that region that had 1 or more deaths divided by the total number of cases in the state. The expected mortality rate was calculated by summing the probability of death (generated from the logistic regression model) for all cases within a given state to obtain the expected number of deaths for that state given its case mix. The expected number of deaths was then divided by the total number of cases performed in the state to get the expected mortality rate. The standardized mortality ratio (SMR) was calculated as the observed mortality rate divided by the expected mortality rate. Standardized mortality ratios were Table 1. Patient Clinical Characteristics and Outcomes in California KID 2003 (n = 2,120), n (%) KID 2006 (n = 2,294), n (%) RACHS-1 Risk Category (10.9) 189 (8.2) (33.4) 768 (33.5) (34.1) 836 (36.4) (9.0) 220 (9.6) 5 and 6 64 (3.0) 78 (3.4) Unable to be assigned 201 (9.5) 203 (8.8) Premature 56 (2.6) 85 (3.7) Major noncardiac structural anomaly 107 (5.0) 142 (6.2) Major chromosomal abnormality or syndrome 343 (16.2) 369 (16.1) 91 Surgery during admission 533 (25.1) 559 (24.4) Weekend admission 159 (7.5) 178 (7.8) Insurance Medicaid 839 (39.6) 1,020 (44.5) Private including HMO 964 (45.5) 984 (42.9) Self-pay and other 317 (15.0) 289 (12.6) Outcomes In-hospital death 92 (4.3) 80 (3.5) Any complication 751 (35.4) 855 (37.3) Total hospital charges, $ n = 2,072 n = 2,196 Median 106, ,516 Interquartile range 67, , , ,499 Range 3, ,376 9, ,938 Mean 163, ,905 SD 161, ,820 Abbreviation: RACHS-1, Risk Adjusted classification for Congenital Heart Surgery. 220 JONA Vol. 41, No. 5 May 2011

4 in Table 1. These characteristics were similar in both 2003 and The cases of congenital heart surgery were categorized using the RACHS-1 risk categories, with categories 2 and 3 accounting for most cases in both years. Outcomes in California In-hospital mortality rates decreased, and complications and hospital charges increased in 2006 compared with 2003 (Table 1). Figure 1. Nursing FTEs for California institutions, 2003 and calculated for California for year 2003, all other states combined for year 2003, California for year 2006, and all other states combined for year Analogous methodology was used to calculate SCRs (standardized complication ratios). Variation in total hospital charges was explored by examining the differences between the observed and expected mean charges for each institution (the Bcharge differential[). Linear regression analysis was used to fit the risk adjustment model for total hospital charges; because charges have a skewed distribution, the natural logarithm of total charges was used as the outcome variable. To calculate the charge differential, the natural logarithm of total charges was averaged for all cases within each institution to find the mean observed log charges for the center and then exponentiated to get the observed geometric mean charges. The risk adjustment model was then used to predict the log of total charges for each case in the data set, given the case s baseline characteristics. These predicted charges were averaged within each institution to find the mean expected log charges for the center and exponentiated to get the expected geometric mean charges. The difference between the observed mean charges and the expected mean chargesvcalled the mean charge differentialvrepresents the average dollar amount by which the institution Bovercharges[ or Bundercharges[ per case, relative to the data set as awhole. Results Patient Clinical Characteristics Patient demographic and clinical characteristics in the hospitals performing pediatric cardiac surgery in the KID data set in California are illustrated California Hospital Characteristics In 2003, 50% and, in 2006, 57% of the California hospitals performing pediatric cardiac surgery were large children s hospitals. In 2003, 100% and, in 2006, 92.9% of the hospitals were described as academic teaching hospitals. California Nursing Characteristics The median number of RN FTEs per hospital was 798 in 2003 and 1,122 in 2006 (Figure 1). Nursing ratios per staffed bed also increased from 1.85 in 2003 to 2.51 in 2006 (Figure 2). After enactment of the California Safe Staffing Law, RN FTEs significantly in creased (P =.025), and the increase in nurse ratios was also significant (P =.036) in 2006 compared with Relationships Between California Nursing Characteristics and Outcomes In California, RN FTEs and nurse ratios were not associated with risk-adjusted mortality in 2003 or Neither RN FTEs nor nursing ratios were significantly associated with risk-adjusted complications. RN FTEs and nursing ratios were not significantly associated with total hospital charges after risk adjustment (Table 2). Figure 2. Nursing ratios for California institutions, 2003 and JONA Vol. 41, No. 5 May

5 Table 2. Risk-Adjusted Relationships Between Nursing Characteristics and In-hospital Mortality, Complications, and Resource Utilization (Total Hospital Charges) Nursing Characteristics Odds Ratio 95% Confidence Interval P Risk-adjusted mortality RN FTEs (j100) 1.04 ( ).17 Nursing ratio (FTEs per staffed bed) (j1) 1.13 ( ).54 Risk-adjusted complications RN FTEs (j100) 0.94 ( ).24 Nursing ratio (FTEs per staffed bed) (j1) 0.52 ( ).17 Risk-Adjusted Resource Utilization (Total Hospital Charges) Nursing Characteristics " 95% Confidence Interval P RN FTEs (j100) j0.01 j0.03 to Nursing ratio (FTEs per staffed bed) (j1) j0.13 j0.33 to For RN FTEs, odds ratios were estimated for each increase of 100 FTEs. For nursing ratio, odds ratios were estimated for each increase of 1 FTE per staffed bed. Relationships Between California Outcomes and Other States Standardized mortality ratios decreased in California and in all other states in 2006 relative to There was a slightly greater decrease in the SMR for California than the decrease in SMR for all other states. The SCR increased in California in the year 2006 relative to However, SCR decreased for all other states in 2006 (Table 3). Total hospital charges increased in California and nationally in the year 2006 relative to Interestingly, the increase in the charge differential in California was more than twice the increase for all other states combined (Table 4). Discussion Major Findings There were several unique findings. This is the first known study to examine the impact of nurse staffing ratios on risk-adjusted mortality, risk-adjusted complications, and risk-adjusted resource utilization for pediatric cardiac surgery programs in California before and after enactment of AB 394 and relative to other states combined. The intent of the law was to ensure the presence of sufficient numbers of nurses to provide hospitalized patients with an appropriate level of nursing care. Hospitals with pediatric cardiac surgery programs in California significantly increased RN FTEs and RN ratios after enactment of AB 394 in 2006 compared with These findings are key because the law allows for licensed vocational nurses (LVNs) to comprise up to 50% of the nursing workforce within general acute-care hospitals. Prior to enactment of the law, nurse leaders were concerned that hospitals might hire more LVNs as a less expensive way to comply with the new legislation. 16,17 The data did not reveal less variation in RN FTEs and RN ratios in 2006 compared with No difference in variation may mean that an intention of the law was not achieved. Because an intent of the law was to Bstandardize[ the amount of nursing care received by hospitalized patients in the state, this intent may not have been realized. Or perhaps the RN FTEs and ratios were at the most desirable levels in these hospitals, in 2003, resulting in a ceiling effect. Nursing Characteristics on Risk-Adjusted Outcomes Higher nurse ratios and RN FTEs were not associated with lower risk-adjusted mortality, risk-adjusted Table 3. Standardized Mortality Ratios and Standardized Complication Ratios Observed MR Expected MR SMR Observed CR Expected CR SCR California, % 3.8% % 32.1% 1.10 All other, % 3.8% % 32.4% 1.00 California, % 4.4% % 32.6% 1.15 All other, % 4.1% % 32.6% 0.95 Abbreviations: MR, mortality rate; CR, complication rate. 222 JONA Vol. 41, No. 5 May 2011

6 Table 4. Total Hospital Charges: Charge Differential California, 2003 All other, 2003 California, 2006 All other, 2006 Observed Mean Charges Expected Mean Charges Charge Differential $116,146 $93,860 $22,287 $72,866 $91,240 j$18,373 $174,828 $99,098 $75,730 $99,492 $94,750 $4,746 complications, and risk-adjusted resource utilization, after controlling for additional patient and hospital characteristics in the year Contrary to the nurse staffing literature from general hospitals with adult patients, 3,18,19 RN FTEs and ratios were not independent explanatory variables for pediatric cardiac surgical mortality or complications. Consistent with our findings across children s hospitals, 11 the outcome variable of mortality may be insensitive to nursing characteristics in pediatric cardiac surgery programs as long as certain staffing thresholds have been achieved. Freestanding children s hospitals and adult hospitals with children s services may have staffing levels for pediatric cardiac surgery patients that are higher as a baseline than RN staffing for general adult services. Therefore, in the analysis for this outcome, nurse staffing may already be above the threshold for finding a difference in mortality. Also, in contrast to the literature about hospitalized adults 3,4 after risk adjustment, RN FTEs and ratios were not significantly associated with complications. It is possible that the broad definition of complications included in this study was less sensitive to changes in nurse staffing, as only some complicationsvsuch as pneumonia and urinary tract infectionsvwere nurse-sensitive. Further research is needed to understand the relation between nurse staffing and complications in pediatric patients, in general, and in pediatric cardiac surgery patients, in particular, before this finding can be understood more clearly. Neither RN FTEs nor ratios were associated with total hospital charges after risk adjustment. In the unadjusted univariate analysis, there was a relation between RN ratios and increased total hospital charges that weakened after risk adjustment. Interestingly, institutional volume of pediatric cardiac surgery was associated with risk-adjusted resource utilization; as volume increased total hospital charges actually decreased. As a result, volume was included in the risk adjustment model for resource utilization and may have accounted for mitigating the finding of a relation between RN ratios and increased total hospital charges in the unadjusted analyses. This finding may also support the well-established volume outcome relation in pediatric cardiac surgery programs. More specifically, pediatric institutions that perform larger numbers of cardiac surgery cases have lower mortality rates. 14,20 Standardized Mortality Ratio in California Versus Other States As expected, there was a slightly greater decrease in SMR in California (33%) than the decrease in SMR for all other states combined (29%) in This finding is consistent with literature that reveals a decrease in mortality over time for pediatric cardiac surgery programs. 14 It was important to verify this outcome in these data sets because the nature of congenital heart surgery is complex, and new procedures and treatments are being developed constantly to increase the survival of patients. The reason for the greater decrease in SMR in California compared with other states is not known but could be due to an increase in RN FTEs and ratios in these hospitals after enactment of the law. With higher RN ratios, there may have been more patients receiving successful nursing interventions to avert death. Standardized Complication Ratio in California Versus Other States Unexpectedly, a decrease in SCR was not found for California for 2006 relative to Although the SCR did decrease 5% for other states combined in 2006, the SCR in California increased by 5%. It is interesting that the change in California was opposite to what occurred nationally for SCR and opposite to the improved reduction in SMR in California over the same period. It is conceivable that cardiac surgical patients who would previously have died in California in 2003 survived in 2006 but experienced complications. Another possibility is that perhaps this occurred because hospitals in California added a significant number of RNs after enactment of AB 394 and the aggregate experience level of the nursing staff in each hospital was diluted because of the number of new graduate nurses. In such a scenario, the complication rate may have increased for this complex group of pediatric congenital heart surgery patients. Resource Utilization (Total Hospital Charges) in California Versus Other States The increase in charge differential in California from 2003 to 2006 was more than twice the increase for other states combined during the same period. This JONA Vol. 41, No. 5 May

7 finding may be because the increased benefit from higher RN staffing in reducing charges by reducing complications was not realized in California. These data suggest that there may be an opportunity for cost reduction strategies. Although increasing nurse-to-patient ratios has been recommended to improve patient safety in hospitals, the costeffectiveness of increasing RN staffing remains controversial because of data where causal links between nurse staffing and improved nurse-sensitive patient outcomes have not been demonstrated. 4,21 To economically survive, healthcare systems must tighten resources and maintain quality that is collaboratively defined by both users and providers in the system. Nurses are well positioned to drive this debate and articulate their value in creating safe passage for patients and in containing costs while achieving highquality outcomes for patients and families. Findings from this study will inform the national examination of the relation between nursing characteristics including education level and expertise and patient charges across states for pediatric congenital heart surgery services. Study Limitations As with any study relying on secondary data, there may be questions of data integrity and issues with inadequate reporting. 22 Hospital charges are not an ideal proxy for resource utilization; total hospital costs would be a better measure. In addition, in 2003 in California, there were 14 hospitals with pediatric cardiac surgery programs, so the study is limited by power and may not be generalizable to all US hospitals in which pediatric cardiac surgery is performed. Policy and Practice Implications for Nurse Executives These analyses provided information about a controversial state law and its impact on care in pediatric cardiac surgery programs. Before this study, it was uncertain whether the increased resources required by the law would be mitigated by reductions in mortality and morbidity. It is possible that FTEs and nursing ratios were already sufficiently high at hospitals that provide pediatric cardiac surgery services, in which case the law would have no impact and may actually be unnecessary. The California Safe Staffing Law has important economic, patient safety, and care quality implications. The study findings illuminated statewide outcomes of risk-adjusted mortality, complications, and resource utilization from primarily pediatric academic teaching institutions and can help to inform the debate about pediatric centers of excellence in tertiary and quaternary care. All of the findings from this study may guide and inform program development and healthcare policy at the institution, state, and federal levels. Conclusion The study findings indicate that hospitals in California made upward adjustments in RN FTEs and nurse ratios after enactment of the California Safe Staffing Law. There was a major increase in California s charge differential, which was 2.5 times more than other states, a decrease in SMR, and an increase in SCR for pediatric congenital heart surgery patients after enactment of the legislation. The findings of this study did not reveal a relation between changes and individual hospital RN FTEs or RN ratios. As healthcare reform is now the law of the land in these financially challenged times, states will be watching California and the results of its nurse staffing mandate. This study provides a foundation on which to build the next level of empirical studies to clarify relations between nurse-patient ratios, riskadjusted outcomes, financial performance, and both actual and opportunity costs, to better inform those who are evaluating the benefits and risks of this type of public policy. More empirical research is required to determine whether the mandated ratios have the intended effect on quality and safe care for congenital heart surgery patients. References 1. American Nurses Association. Safe Nurse Staffing Laws in State Legislatures isanadoing/statelegislation.aspx. Accessed March 10, Aiken LH, Clarke SP, Sloane DM, et al. Effects of hospital care environments on patient mortality and nurse outcomes. Journal of Nursing Administration. 2008;38(5): Aiken LH, Clarke SP, Sloane DM, et al. Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfaction. JAMA. 2002;288: Needleman J, Buerhaus P, Mattke S, Stewart M, Zelevinsky K. Nurse-staffing levels and the quality of care in hospitals. N Engl J Med. 2002;346(22): Tourangeau AE, Doran DM, Hall LM, et al. Impact of hospital nursing care on 30-day mortality for acute medical patients. J Adv Nurs. 2007;57(1): Van den Heede K, Lasaffre E, Diya L, et al. The relationship between inpatient cardiac surgery mortality and nurse numbers and educational level: analysis of administrative data. Int J Nurs Stud. 2009;46(6): Donaldson N, Burnes-Bolton L, Aydin C, Brown D, Elashoff J, Sandhu M. Impact of California s licensed nurse-patient ratios 224 JONA Vol. 41, No. 5 May 2011

8 on unit-level nurse staffing and patient outcomes. Policy Polit Nurs Pract. 2005;6: Burnes-Bolton L, Aydin CE, Donaldson N, et al. Mandated nurse staffing ratios in California: comparison of staffing and nursing-sensitive outcomes pre and post-regulation. Policy Polit Nurs Pract. 2007;8: Aiken LH, Sloane DM, Cimiotti JP, et al. Implications of the California nurse staffing mandate for other states. Health Serv Res. 2010;45(2): Mark B, Harless D, Berman W. Nurse staffing and adverse events in hospitalized children. Policy Polit Nurs Pract. 2007;8: Hickey P, Gauvreau K, Connor J, Sporing E, Jenkins K. The relationship of nurse staffing, skill mix, and magnet recognition to institutional volume and mortality for congenital heart surgery. JNursAdm. 2010;40(5): Connor JA, Gauvreau K, Jenkins KJ. Factors associated with increased resource utilization for congenital heart disease. Pediatrics. 2005;116: Fawcett J, Russell G. A conceptual model of nursing and health policy. Policy Polit Nurs Pract. 2001;2: Jenkins KJ, Gauvreau K. Center-specific differences in mortality: preliminary analyses using Risk Adjustment in Congenital Heart Surgery (RACHS-1) method. J Thorac Cardiovasc Surg. 2002;124: Benavidez OJ, Gauvreau K, Jenkins KJ. Medical injuries and congenital heart surgery. Pediatr Res. 2005;58: Buerhaus PI. What is the harm in imposing mandatory hospital nurse staffing regulations? Nurs Econ. 1997;15: Coffman JM, Seago JA, Spetz J. Minimum nurse-to-patient ratios in acute care hospitals in California: evidence is mixed on whether minimum staffing ratios will improve conditions for patients and for nurses. Health Aff. 2002;2: Aiken LH, Smith HL, Lake ET. Lower Medicare mortality among a set of hospitals known for good nursing care. Med Care. 1994;32: Pearson SD, Allison JJ, Kiefe CI, et al. Nurse staffing and mortality for Medicare patients with acute myocardial infarction. Med Care. 2004;42: Bazzani L, Marcin J. Case volume and mortality in pediatric cardiac surgery patients in California Circulation. 2007;115: Rothberg M, Abraham I, Lindenauer PK, et al. Improving nurse-to-patient staffing ratios as a cost-effective safety intervention. Med Care. 2005;43(8): LawthersAG,McCarthyEL,DavisR,PetersonB,PalmerR, Iezzoni L. Identification of in-hospital complications from claims data. Is it valid? Med Care. 2000;38: JONA Vol. 41, No. 5 May