Factors Influencing RNs’ Decisions to Work Carol S. Brewer, Ph.D.* Chris T. Kovner, Ph.D.** William Greene, Ph.D.** Yow Wu-Yu, Ph.D.* Liu Yu, Ph.D. (cand.)* This work was supported by a grant from AHRQ R01 HS011320 Presented at AcademyHealth, June 6, 2004 *University at Buffalo ** New York University Participation PT FT Why important: – If only 10% of the PT RN population worked FT, it would add 31,000 RNs to supply Part of larger analysis also looking at work/notwork – If the RN works, how much (PT or FT) FT defined as >35 hrs per week for all jobs Research questions What factors are associated with the work decision (WK/NW) and amount of work (FT/PT)? Are the WK/NW and PT/FT decisions made together or separately? Data sources The National Sample Survey of Registered Nurses March 2000 (Spratley et al., 2001) – County level data (some restrictions) – Female RNs in 300 MSAs represented MSA/County level variables – InterStudy Competitive Edge Part III Regional Market Analysis (2002) – Area Resource File (2002) Sample 35,358 registered nurses Exclusions: – Did not live or work in the USA – Missing MSA codes for job and living location – Did not work (or live) in an MSA Analytic sample was 21,123 females Married 14, 898 Single 6,225. Economic Environment Variables Induced demand HYP. Means – Medical/surgical specialists per 1000 pop + 1.74 – Primary care practitioners per 1000 pop + 0 .24 – % of HMO services paid FFS + 17.4% Managed care/demand – Index of competition - .68 – Penetration rate of managed care - 29.6% Poverty/demand – % non-HMO Medicaid as % of total MSA pop + 7.4% – % uninsured pop ? 13.6% – % families living in poverty ? 8.1% Unemployment rate + 1.8% Demographics Characteristics Working Non working Modal Age 40-44 50-54 Modal Tot Inc $50-75,000 $50-75,000 Non-white 16.1% 12.8% Marital status 69.8% 74.4% Any kids < 6 18.1% 15.7% Student 7.4% 2.9% Working RNs Characteristics Dominant function direct care 51.6% Staff/general duty nurses 50.9% Work in hospitals 60.5% Satisfaction (mean) 1= extremely satisfied married 2.31 single 2.42 Analysis Analytic method: bivariate probit regression – Tested hypothesis that WKNW / FTPT decisions are related Single RNs Rho= -0.45, p= 0.02 Married RNs, Rho=-0.51, p= 0.00 Results Interpretation of marginal effects Probability of working or working FT changes (+ or -) by amount of marginal effect at mean of variable Ex: The probability of a 25-30 yr old RN working FT decreases by 0.12 compared to a RN < 25 Significant marginal effects PT/FT regression: Economic variables Probability of FT Married Single decreases Primary care physician -0.18 -0.23 ratio Other sig var (very small effects): Unemployment rate, penetration rate for both % non HMO M’caid, Specialist ratio for single only Significant marginal effects PT/FT regression: Economic variables Probability of FT Married Single increases Index of competition 0.12 0.11 Other sig var (very small effects): sig for both % families in poverty Size of MSA (small, medium, compare to large) Significant Demographic variables in Part-time / Full-time regression Probability of FT decreases – All age categories: Stronger effect for married, >60 – if any children < 6 Stronger for married (-0.30 vs -0.17) – Baccalaureate RN vs. AD Probability of FT increases – Minorities married, ME=0.16 – Total family income, (non linear) NS for married 0.30 to 0.19 for single – Student status NS for married PT student or not a student Significant organizational variables in PT/FT regression Probability of FT decreases – Satisfaction: small ME= - 0.01 married, ONLY – Settings: Educators, student health, ambulatory care SIG vs. hospital RNs Probability of FT increases – Function: Supervisors, teachers, administrators vs. direct care RNs: ME=0.09- 0.21 married, ONLY – Positions: ALL other (NP, CNS, administrator, etc) vs. staff RNs, Stronger for married Conclusions MSA level economic variables – Influential on PT/FT decision, but not decision WK/NW Influence of demographic variables – Age, children, minority, income and student status more effect on FT work decision than WK – Education (BSN-married, Master’s single) weak but negative = concern Organization variables – satisfaction significant, neg, if married – Hospital, direct care and staff RNs most likely to be PT – Functions and positions indicating career path more likely to be sig Implications Need to target single vs married RNs What organizations can change: – Career orientation may influence PT/FT chicken or egg ? Develop career paths early – Age related work conditions, esp after age 55 – Improve satisfaction – Recruit minorities Work decision different from how much to work Implications Government policy – Clarify education: rewards need to be clear – Economic variables-need to understand What can Govt manipulate? May help in predicting regional variability in shortages. Job market or health of population? – For ex: IOC- perhaps hospitals are competing for nurses and end up with more full-time workers due to higher wages
"Factors Influencing Registered Nurses Decisions to Work"