NEWSLETTER - October 2008
From Anmarie Widener
DOES PAID FAMILY AND SICK LEAVE IMPROVE OUR HEALTH?
By Anmarie Widener and Adam Korobow
There are no U.S. micro level studies (at the individual level) that statistically show the health impact of paid leave. This study is the first of its kind, sampling over 50 thousand children using the Medical Expenditure Panel Survey, a survey sponsored and conducted by AHRQ, the Agency for Health Research and Quality. This study is the first of a two part study examining the health impact of paid leave. Given the current debate in Congress on the importance of mandated paid leave, studies like these are timely and necessary if Congress is going to make policy decisions that are efficient and effective.
The ability for working parents to meet the health needs of their children and families may be contingent upon the important component of paid leave. This paper analyzes the impact of paid leave on child health outcomes.
Having no paid leave from employment may be an obstacle for families needing to care for the health and illnesses of family members or their own health care needs, as an infinite number of anecdotal evidence shows. Yet many U.S. working parents continue to lack paid leave[1]: 70 percent of private sector employees in the U.S. are not entitled to paid sick days to care for a sick child or other family member; 77 percent of low-income families and 48 percent of all private sector employees have no paid sick days at all (Lovell, 2004; Levin-Epstein, 2006).
Even general paid leave such as paid time off (PTO) (leave that includes time for illness and holiday and personal time all wrapped up in a set number of days per year) is not universal for U.S. employees. As far as vacation time goes, 23% of private sector employees have no paid vacation; 31% of female workers have no paid vacation; and 37% of low income female workers have no paid vacation time (Ray & Schmitt, 2007; De Graaf, 2008).
A recent report from the Center for Economic and Policy Research reviewed the national policies of 21 wealthy countries that support parents’ work/life balance and ranked the U.S. last when it comes to the generosity and length of paid leave (Ray, Gornick, & Schmitt, 2008).
We hypothesize that children whose parents have access to paid leave will have more child wellness visits and healthcare visits which result in better child health status because parents will be more likely to take time off work to care for health needs. In this paper, we test the hypothesis that having access to paid-leave will increase the likelihood that a child will receive either a wellness visit or what we will call a care visit. We consider the cost of a physician visit as being comprised of two components: 1) the out-of-pocket price paid to the physician and 2) the opportunity cost of lost wages if the parent has to take time from work that is not reimbursed.
Having access to paid leave should reduce the overall price of a physician visit since the opportunity cost of foregone wages is removed from the decision. In other words, if leave from work to visit the doctor is not paid for, then this should result in a higher price of a physician visit, and might lead people to undertake fewer visits. Thus, offering paid-leave should reduce the price of a physician visit and might result parents being more likely to take their child to a physician when care is needed, assuming parents can use paid-leave for their children (Clemans-Cope, et al, 2008).
We know that parents without paid leave are less likely to take their child or themselves to a doctor when sick. When they do, many lose their jobs; in fact, one of the leading causes of job loss is taking care of children’s health. In 2007, an estimated $28.4 billion in lost productivity was due to job turnover, going to work sick and spreading illness, and present but not productive employees due to the employees’ own illness and stress (presenteeism). And children in low income families are more likely to have poorer health compared to children in higher income families.
Our sample consisted of children in high, middle, and low income families between the ages of 0 and 17. Children had to be in families with at least one working parent employed all year, thus single and dual parent families are included. We use data from multiple years of the Medical Expenditure Panel Survey (MEPS). MEPS provides nationally representative estimates of health care use and expenditures for the U.S. civilian non-institutionalized population and is sponsored by the Agency for Healthcare Research and Quality (AHRQ) and the National Center for Health Statistics (NCHS).
About 94 percent of the children in our sample have health coverage; 74 percent of these children have private health coverage while the rest are covered by a public plan. The average age of a child in our sample is approximately nine years. Average total parent income is about 60 thousand dollars. Most children have at least one parent in the household (88 percent) who completed high school (or has a GED equivalent). Approximately half the children in the sample (46 percent) have a parent in the household who has a college degree.
About 68 percent of children in our sample have at least one parent who has access to paid sick leave. For children who have two full-time working parents, access to paid sick leave rises to about 92%. Single parent households have the lowest access to paid sick leave, 43%. Access to paid sick leave also rises sharply with household income. About 20% of children in households which earn less than $20,000 per year have at least one parent that has access to paid sick leave. For children in households with income of greater than $60,000, access to paid sick leave for at least one parent rises to just over 85%.
Children from households where at least one parent is working and the household has access to sick leave receive a higher average number of wellness visits, 1.09, as compared to children whose parents do not have access to leave, 0.96.[2] This difference becomes more pronounced when looking at the number of wellness visits by households where both parents are working and have access to paid leave (1.06 visits) compared to households where both parents work and have no access to paid leave (0.85 visits). When we examine care visits, the same result is found, namely that children with parents who have access to paid-sick leave exhibit a higher number of average annual care visits (2.82) than children whose parents do not have access to paid-sick leave (2.25).
The largest magnitude determinant in our models of both wellness and care visits is whether a child has some form of health coverage. Having at least one parent in the household with a college degree increases the odds that a child receives a wellness visit and a care visit. Having at least one parent with high-school degree or GED also increases the odds of a visit but the magnitude of this effect is smaller than having a college degree. Paid sick leave had the next largest impact on likelihood of receiving a wellness or doctor visit. Specifically, having access to paid sick leave increased the odds of a child receiving a wellness visit by 1.14 while paid sick leave increased the odds of a care-based visit by 1.18. We also list the corresponding average marginal effects from the logistic regressions. In particular, the average marginal impact of having access to paid-leave across our sample is to increase the likelihood of a wellness visit by about 3% and increase the likelihood of a care-based visit by about 4%. Only having health insurance (approximately 20%) and education level (approximately 6%) had a higher impact. Paid leave is statistically significant at the 95 percent level in both the wellness and care-based models. Moreover, the magnitude of the impact is close to the same impact as estimated by the logistic regressions, namely, having access to paid leave is associated with about a 3% higher likelihood of receiving a wellness visit in a year and a 3% higher likelihood of receiving a care-based visit.
Further analysis on the impact of paid leave on subgroups within the dataset are needed. For example, what is the impact of paid sick leave on asthmatic children, on children with special health care needs, on school absenteeism. These types of breakdowns as well as distinguishing between types of paid leave (e.g., sick, family, and vacation) and their differential impact on child health outcomes are necessary to tease out the actual true impact of policies that support paid leave.
Paid sick leave has been found to significantly increase the likelihood of receiving child wellness and child doctor visits. Working families may depend on paid leave policies to make time to give quality care without fear of losing one’s job and to decrease the cost of a doctor visit. Paid leave may be one mechanism by which parents are better able to participate in preventative care for their children, to manage their child’s acute illnesses well, and to ensure their children are growing in safe and healthy environments.
A more integrated health system would consider not only decreasing costs and increasing health insurance coverage, but also increasing an individual’s ability to receive care when ill while maintaining gainful employment. True cost-cutting mechanisms emphasizes prevention which means making time to care for one’s own health and the health of their family members, before acute illness occurs. Access to paid sick leave may be one mechanism employees depend on to provide healthcare for themselves and their children.
Anmarie J. Widener, MSW, Ph.D., is a health policy analyst at LMI, a not-for-profit government organization based in McLean, VA. Adam Korobow, Ph.D., is a research fellow at the LMI Research Institute.
[1] We are interested in paid leave as opposed to unpaid leave. The Family and Medical Leave Act (1993) provides 12 weeks of unpaid leave to workers; about half of all private sector employees are eligible. A survey by the Department of Labor found more than three quarters of eligible employees reported needing but not using FMLA because they could not afford to take leave without pay (Cantor, et al, 2001).
[2] We tested this difference using a Wald Test and found statistical significance at the 1 percent level.