The n e w e ng l a n d j o u r na l of m e dic i n e

esis promoted by chronic inflammation is B lymphocyte depen- pabilities of B cells and CD19-positive plasma
dent. Cancer Cell 2005;7:411-23.
3. Ammirante M, Luo J-L, Grivennikov S, Nedospasov S, Karin
M. B-cell-derived lymphotoxin promotes castration-resistant
prostate cancer. Nature 2010;464:302-5.
4. Balkwill F, Montfort A, Capasso M. B regulatory cells in
cancer. Trends Immunol 2013;34:169-73.
5. Hansmann L, Blum L, Ju C-H, Liedtke M, Robinson WH,
Davis MM. Mass cytometry analysis shows that a novel memory
phenotype B cell is expanded in multiple myeloma. Cancer Im-
munol Res 2015;3:650-60.
DOI: 10.1056/NEJMc1512760
The authors reply: In reply to Shimabukuro-
Vornhagen and colleagues: the mechanism of ac-
tion underlying the efficacy of CTL019 in the
case of multiple myeloma we reported remains
uncertain. In our discussion, we acknowledged
the possibility that elimination of non-neoplastic
B cells by CTL019 might be at least partially re-
sponsible for its therapeutic activity, and we cited
prior studies showing the tumor-promoting ca-
cells.1,2 Through this mechanism, CTL019 may
have clinical usefulness in other cancers that do
not express CD19, particularly in combination
with other immunotherapies.
AlfredL. Garfall, M.D.
EdwardA. Stadtmauer, M.D.
CarlH. June, M.D.
Perelman School of Medicine at the University of Pennsylvania
Philadelphia, PA
cjune@exchange.upenn.edu
Since publication of his article, Dr. Garfall reports receiving
consulting fees from Novartis. No further potential conflict of
interest relevant to this letter was reported.
1. Affara NI, Ruffell B, Medler TR, et al. B cells regulate mac-
rophage phenotype and response to chemotherapy in squamous
carcinomas. Cancer Cell 2014;25:809-21.
2. Shalapour S, Font-Burgada J, Di Caro G, et al. Immunosup-
pressive plasma cells impede T-cell-dependent immunogenic
chemotherapy. Nature 2015;521:94-8.
DOI: 10.1056/NEJMc1512760

Dependent Coverage under the ACA and Medicaid Coverage

for Childbirth
To the Editor: In the United States, rates of ment sources for childbirth among young adults
uninsurance have been historically high among (details of the methods are provided in the
young adults (19 to 26 years of age). One of the Supplementary Appendix, available with the full
first implemented provisions of the Affordable
Care Act (ACA) was a mandate that all private- Figure 1 (facing page). Time Trends in Payment Sources
insurance family policies cover dependents until for Childbirth, According to Age Group.
26 years of age. Estimates suggest that this pro- Seasonally adjusted time trends in payment sources
vision has reduced the rate of uninsurance for childbirth according to the Centers for Disease
among young adults by approximately 10%.1 Control and Prevention natality data obtained from
Childbirth, the major reason for hospitalization states that provided health insurance information are
shown.3 The payment sources were private insurance
of young adults, has received little attention in (Panels A and B), Medicaid (Panels C and D), self-pay
prior literature on the mandate. ment (Panels E and F), and other payment (Panels G
Childbirth is financed differently from other and H). Each circle in the scatter plot represents the
U.S. health care services. Before the ACA, Med- percentage of births paid for by the particular payment
type in that month for that age group. The category of
icaid covered all pregnant women earning less
other payment included the Indian Health Service,
than 133% of the federal poverty level. During the TRICARE military health system (formerly known
this time, the government covered approximate- as the Civilian Health and Medical Program of the Uni
ly half of U.S. births.2 Though 32.2% of young formed Services), other government insurance at the
adults were uninsured in 2009, for instance, only federal, state, and local levels, and all other insurance.
The time period was January 2009 through December
approximately 5% of births to these adults were
2012, except for the period from March 2010 through
uninsured; more than 60% were covered by December 2010 (the staggered implementation period).
Medicaid. 1
We used regression modeling of the individual-level
In our study, we used data from the natality data with the payment type as the outcome in order to
files of the Centers for Disease Control and Pre- calculate the monthly seasonal adjustments as coeffi
cients on monthly dummy variables. We then subtract
vention and the Nationwide Inpatient Sample
3
ed the seasonal adjustments from averages calculated
from the Agency for Healthcare Research and at the age and month levels.
Quality to consider the effect of the ACA on pay-

194 n engl j med 374;2nejm.org January 14, 2016

The New England Journal of Medicine

Downloaded from nejm.org on February 10, 2017. For personal use only. No other uses without permission.
Copyright 2016 Massachusetts Medical Society. All rights reserved.
 Correspondence

text of this letter at NEJM.org). Using a differ- (mothers who were 27 to 29 years of age) before
ence-in-differences analysis, we compared a and after the policy was implemented. As shown
treatment population (mothers who were 19 to in Figure 1 and the regression analysis in Table
25 years of age) with a control population S1 in the Supplementary Appendix, the young-

A Private Insurance for Age 1925 Yr B Private Insurance for Age 2729 Yr
53.5
26.5
53.0
26.0
Percentage

Percentage
25.5 52.5

25.0 52.0

24.5 Staggered 51.5 Staggered

implemen- implemen-
24.0 tation 51.0 tation
period period
0.0 0.0
2009 2010 2011 2012 2013 2009 2010 2011 2012 2013
Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan.
Month of Birth Month of Birth

C Medicaid Payment for Age 1925 Yr D Medicaid Payment for Age 2729 Yr
66 39.5

65 39.0
Percentage

Percentage

38.5
64
38.0
63
Staggered 37.5 Staggered
implemen- implemen-
62 tation 37.0 tation
period period
0 0.0
2009 2010 2011 2012 2013 2009 2010 2011 2012 2013
Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan.
Month of Birth Month of Birth

E Self-payment for Age 1925 Yr F Self-payment for Age 2729 Yr

5.0 5.0

4.8
4.5
Percentage

Percentage

4.6

4.4
4.0
Staggered 4.2 Staggered
implemen- implemen-
3.5 tation 4.0 tation
period period
0.0 0.0
2009 2010 2011 2012 2013 2009 2010 2011 2012 2013
Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan.
Month of Birth Month of Birth

G Other Payment for Age 1925 Yr H Other Payment for Age 2729 Yr
7.0 5.8

6.8 5.6
Percentage

Percentage

6.6 5.4

6.4 5.2

6.2 Staggered 5.0 Staggered

implemen- implemen-
6.0 tation 4.8 tation
period period
0.0 0.0
2009 2010 2011 2012 2013 2009 2010 2011 2012 2013
Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan. Jan.
Month of Birth Month of Birth

n engl j med 374;2nejm.org January 14, 2016 195

The New England Journal of Medicine
Downloaded from nejm.org on February 10, 2017. For personal use only. No other uses without permission.
Copyright 2016 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

adult provision led to an increase of 2.5 percent-
age points (95% confidence interval [CI], 1.6 to
3.4) in private-insurance reimbursement a
9.9% relative increase from baseline. Medicaid
payments decreased by 2.1 percentage points
(95% CI, 3.0 to 1.3), and self-payments decreased
by 0.3 percentage points (95% CI, 0.5 to 0.2).
Most of the changes involved payments for
ever, childbirth tends to be the most costly part
of pregnancy and payment sources for childbirth
probably are similar to payment sources for con-
current care.
Yaa AkosaAntwi, Ph.D.
Indiana UniversityPurdue University Indianapolis
Indianapolis, IN
yakosaan@iupui.edu

births to unmarried mothers (Table S1 in the Jie Ma, M.A.

Supplementary Appendix). Kosali Simon, Ph.D.
In our confirmatory analysis using the Na- Indiana University
tionwide Inpatient Sample, we found similar re- Bloomington, IN

sults. Coverage by private insurance increased, Aaron Carroll, M.D.

and Medicaid-funded and uninsured deliveries
decreased (Table S2 in the Supplementary Ap-
pendix).
Our study shows that the young-adult provi-
sion was associated with a significant increase
in private coverage and a significant decrease in
Medicaid coverage of childbirth among women
19 to 26 years of age. As such, it suggests a shift
in financing of childbirth from Medicaid to pri-
vate insurance in this population. This research
may be limited insofar as it only looks at two
specific points of pregnancy and delivery: points
immediately before and after childbirth. How-
Indiana University School of Medicine
Indianapolis, IN
Disclosure forms provided by the authors are available with
the full text of this letter at NEJM.org.
1. Akosa Antwi Y, Moriya AS, Simon K. Effects of federal policy
to insure young adults: evidence from the 2010 Affordable Care
Acts dependent-coverage mandate. American Economic Jour-
nal: Economic Policy 2013;5:1-28.
2. Markus AR, Andres E, West KD, Garro N, Pellegrini C. Med-
icaid covered births, 2008 through 2010, in the context of the
implementation of health reform. Womens Health Issues 2013;
23:e273-80.
3. Centers for Disease Control and Prevention, National Center
for Health Statistics. Birth data 20092012 (http://www.cdc.gov/
nchs/births.htm).
DOI: 10.1056/NEJMc1507847

State Medicaid Expansion and Changes in Hospital Volume

According to Payer
To the Editor: The Affordable Care Act (ACA)
has many potential implications for the hospital
industry. One of the most closely followed issues
is the expansion of Medicaid, which became a
state option as a result of the Supreme Court
decision of 2012.1 As of this writing, 31 states
and Washington, D.C., have elected to expand
Medicaid, and enrollment grew by 21% to more
than 71 million persons between January 2014
and March 2015.2 State decisions about Medicaid
expansion potentially have important implica-
tions for hospital payment sources and revenue.
A number of reports have shown a reduced
volume of uninsured patients in hospitals in
expansion states. However, most of these data
have come from for-profit hospitals or from a
single state.3,4
We performed a study using hospital-dis-
charge data from the Hospital ACA Monitoring
Project (HAMP), a voluntary surveillance effort
funded by the Robert Wood Johnson Founda-
tion.5 All state hospital associations were invited
to participate by submitting quarterly data on
inpatient admissions and emergency department
visits according to payer. Of the 21 states cur-
rently participating, 11 have expanded Medicaid.
Data submitted through HAMP are highly repre-
sentative of overall hospital volume in their re-
spective states, including, on average, 98% of
acute care hospitals. States that participate in
the study, as compared with nonparticipating
states, have smaller Hispanic populations and
lower rates of uninsurance and poverty, as
shown in Table S1 in the Supplementary Appen-
dix, available with the full text of this letter at
NEJM.org.
Table 1 shows changes in hospital volume per
capita between 2013 and 2014 according to

196 n engl j med 374;2nejm.org January 14, 2016

The New England Journal of Medicine

Downloaded from nejm.org on February 10, 2017. For personal use only. No other uses without permission.
Copyright 2016 Massachusetts Medical Society. All rights reserved.