News

Repopulating New Orleans: Big Problem for the Big Easy

One week after Hurricane Katrina and the subsequent flooding from damaged levees in August and September 2005, the population of New Orleans had dwindled from nearly half a million to no more than a few thousand. RAND researchers estimate that the city’s population will recover gradually to only about a quarter million, or roughly half the pre-Katrina population, by September 2008, fully three years after the hurricane.

“Because so many New Orleanians lived in areas that experienced flood damage that was so serious or severe, the challenges of rebuilding those areas strongly dampen the potential for the population to recover in those areas,” according to Narayan Sastry, coauthor of the RAND study, The Repopulation of New Orleans After Hurricane Katrina.

The key factor determining how quickly New Orleans can be repopulated is housing availability.

At the request of the Bring New Orleans Back Commission, RAND researchers conducted a rapid assessment — from late November 2005 to early January 2006 — to develop estimates of the future population of New Orleans. The researchers found that the key factor determining how quickly New Orleans can be repopulated is housing availability: 55 percent of the city’s pre-Katrina population of 485,000 suffered severe damage to their housing after parts of New Orleans were inundated by floodwaters more than four feet deep.

By September 2008, the Population of New Orleans Is Estimated to Grow to 56 Percent of Pre-Katrina Levels

SOURCE: The Repopulation of New Orleans After Hurricane Katrina, 2006.

The researchers provided four estimates: an immediate one (over the next three to six months) and three near-term ones (over the next one to three years). As shown in the figure, the 2008 population estimate is 272,000 — about 56 percent of the pre-Katrina population. The population was projected to rise rapidly to about 155,000 in March 2006 as basic repairs and stabilization of housing were completed, public services and infrastructure were restored, and schools and universities were reopened. The repopulation was projected to taper off thereafter.

As to where the repopulation will occur, the findings show that the population returning to the city will be initially concentrated in the higher-elevation areas on either bank of the Mississippi River, with repopulation in other parts of the city proceeding very slowly.

The study argues that policymakers could speed the repopulation by enacting policies and procedures to streamline the process of obtaining permits to repair and reconstruct housing, something that has already begun. Repopulation could also be accelerated if government officials provide clear and comprehensive information about progress toward and ultimate goals for restoring essential city services and systems, such as public transportation, levees, public safety, public education, and hospitals.

“The more government at every level can do to reduce these uncertainties, the more rapid the conclusion of the recovery process is likely to be,” said coauthor Kevin McCarthy.

For more information:
The Repopulation of New Orleans After Hurricane Katrina, ISBN 0-8330-3940-7, 2006.

Full Document

---

Only Half in Flood-Prone Areas Purchase Flood Insurance

While Hurricanes Katrina and Rita have focused public attention on flooding, it is not a problem just for those living on the coasts.

Millions of homes across the United States lie in areas (so-called Special Flood Hazard Areas, or SFHAs) that have at least a 1 percent chance of flooding per year. Still, “only about half of homeowners living in SFHAs buy federal flood insurance, leaving millions of families at risk for severe financial losses when floods strike,” said Lloyd Dixon, lead author of a new RAND report.

“Many homeowners who suffered substantial damage from Hurricane Katrina were in this voluntary market.”

Coming up with reliable estimates of who buys flood insurance — the “market penetration rate” — is challenging because of a lack of data. Using an innovative approach based on property parcel data, researchers have found that about 50 percent of the 3.6 million single-family homes that are in SFHAs have flood insurance. Only 1 percent of the 76 million homes that are not in SFHAs do. There is also substantial geographic variation: Rates of coverage for homes within SFHAs in the South and West are about 60 percent, while rates range from 20 to 30 percent in the Northeast and Midwest.

Few homeowners buy flood insurance unless they are required to do so. Federal law requires most homes in SFHAs with mortgages to purchase flood insurance. As shown in the figure, 50 to 60 percent of homes in SFHAs are subject to this mandatory purchase requirement, and a substantial proportion of these homes do have flood insurance.

In contrast, roughly only one in five homeowners in SFHAs who are not subject to the mandatory purchase requirement buys flood insurance. “Many homeowners who suffered substantial damage from Hurricane Katrina were in this voluntary market,” Dixon noted.

Only One in Five Homes Not Subject to Mandatory Purchase Requirement Carries Flood Insurance

SOURCE: The National Flood Insurance Program’s Market Penetration Rate: Estimates and Policy Implications, 2006.

When it comes to identifying additional factors that may explain these market penetration rates, the study finds that the number of homes in an SFHA community is key. The penetration rate is 66 per-cent in communities with more than 5,000 homes in the SFHA but declines precipitously to only 16 percent when there are 500 or fewer homes in the SFHA.

While such communities provide a growth opportunity for flood insurance, Dixon pointed out that “the sheer number of such communities — 19,000 of the 20,000 communities where federal flood insurance is available — makes it difficult for policymakers to develop strategies.” Market penetration is also much lower in inland communities (35 percent) than in communities subject to coastal flooding (63 percent).

Although the study was completed before Hurricanes Katrina and Rita, the results are useful for evaluating reforms that are now being debated for the National Flood Insurance Program, such as recent proposals to extend the mandatory purchase requirements to all homes in SFHAs.

For more information:
The National Flood Insurance Program’s Market Penetration Rate: Estimates and Policy Implications, online only, 2006.

Full Document

---

Quality of Health Care Is an Equal Opportunity Gamble

Virtually every adult in the United States who uses the health care system faces a significant risk of failing to receive needed care, regardless of race, gender, income, age, or insurance status, according to a RAND study published in the March 16 issue of the New England Journal of Medicine.

The study assessed the quality of care received by a national sample of people living in major metropolitan areas, measuring quality across the continuum of care for 30 acute and chronic conditions that constitute the leading causes of death and disability in this country and for key preventive services.

“No one can afford to be complacent. The quality-of-care problem in this country is profound and systemic.”

As previously reported by RAND, adults nationwide receive 55 percent of recommended care overall. When looking at socioeconomic subgroups, we can see (as the figure shows) that although differences do exist across gender, income, age, and race categories, “the differences pale in comparison to the chasm between where we are today [in overall quality of care] and where we should be,” said lead author Steven Asch. The study also found that health insurance status was largely unrelated to the quality of care among those with at least some access to care.

“These findings tell us that no one can afford to be complacent, and they underscore that the quality- of-care problem in this country is profound and systemic,” Asch said.

Small Differences Exist, but the Gap Between Observed and Desired Quality of Care Is Large for All Groups Receiving Care

SOURCE: “Who Is at Greatest Risk for Receiving Poor-Quality Health Care?” The New England Journal of Medicine, Vol. 354, No. 11, March 16, 2006, Steven M. Asch, Eve A. Kerr, Joan Keesey, John L. Adams, Claude M. Setodji, Shaista Malik, Elizabeth A. McGlynn.

To be included in the study, participants had to have at least one encounter with the health care system over a two-year period, which indicates that they all had some minimal access to care. Other studies have shown disparities in access to care by race, income, and insurance groups.

The RAND study found that access alone was not enough to guarantee the appropriate delivery of care. Other barriers to receiving necessary care exist once patients get in the door.

“Policies to improve access, while critically important, will not by themselves fix the quality problem,” Asch noted.

Given the large quality-of-care chasm between observed and optimal care for all patients — and the relatively small gaps across demographic subgroups —the authors concluded that in addition to programs that are designed to reduce disparities in access and health outcomes among demographic subgroups, broader efforts to improve care for all are needed.

According to coauthor Elizabeth McGlynn, “We need to fundamentally redesign the health system to ensure that no matter who you are or where you go for care you will get what you need.”

---

Arthritis Care for Older Patients Deemed Poor; Medication Safety, Poorer

Arthritis is a debilitating disease that affects anywhere from 50 to 80 percent of the elderly population. But when it comes to treating the elderly who have arthritis, the quality of care they receive is often poor, and even more so when it comes to addressing the potential hazards of arthritis medications, according to a RAND study published in the April 15 issue of Arthritis Care & Research.

In previous studies, researchers relied only on medical record reviews, which could underestimate the quality of care because medical records do not document such things as education and counseling provided or the use of nonprescription medicines. The new study used clinical data received directly from 339 patients 75 or older receiving care for arthritis in two large medical groups in the western United States, thus enabling researchers to capture a more complete image of the quality of care provided.

Overall, doctors took the indicated actions 57 percent of the time across all eight indicators.

The figure shows the key results of the study, which looked at four indicators of treatment quality and four indicators of medication safety. The bars measure whether doctors took the indicated action in a particular situation. For example, the first bar measures whether a doctor performed an annual assessment of the functional status and the degree of pain for each patient with arthritis.

Overall, doctors took the indicated actions 57 percent of the time across all eight indicators. But researchers found that failing to address medication safety was the most severe quality problem. On average, patients received the recommended care with respect to medication safety just 44 percent of the time. In contrast, patients received some recommended treatment 64 percent of the time.

“The quality of arthritis care is similar to other medical conditions — which is not very good,” said David Ganz, the study’s lead author. “Arthritis dramatically affects the quality of life of older people. Better care would make a real difference in these patients’ lives.”

Efforts to improve arthritis care should focus on instructing patients about safe medication use, said Ganz, particularly since older patients tend to take many prescription drugs.

The study was conducted by a collaboration of researchers from the RAND Corporation; the David Geffen School of Medicine at the University of California, Los Ange- les; and the Greater Los Angeles Veteran Affairs Healthcare System. Partial funding for the project came from the Arthritis Foundation and the Robert Wood Johnson Foundation.

---

Cutting Copayments for Sickest Patients Can Yield Big Payoffs

A new RAND study on prescription drug benefits has identified a way to both improve patient care and hold down rising medical costs in the United States: a targeted policy of reducing drug copayments for the sickest patients taking certain drugs.

Researchers tested this “variable copayment” approach by simulating its effect on the 6.3 million U.S. adults with either private insurance or Medicare coverage who take statin drugs to reduce their cholesterol, dividing the group into high-, medium-, and low-risk categories. The study found that reducing drug copayments for the sickest patients increased compliance with their medications.

“As a result, many people had fewer health problems and were hospitalized less often,” said coauthor Dana Goldman.

Improvements added up to more than $1 billion in estimated savings.

Compared with a base case where patients in all risk categories had a $10 copayment for a one-month supply, the variable copayment approach charged high- and medium-risk patients no copayment and low-risk patients a $22 copayment. Outcomes included a health measure (whether patients fully complied with the treatment), two cost measures (costs to health insurance plans and to patients), and two usage measures (effects on hospitalizations in general and emergency room visits in particular). The results appeared in the January 2006 issue of The American Journal of Managed Care.

The table shows that the variable copayment approach improved most outcomes. Full compliance went up for higher-risk patients. Full compliance went down for lower-risk ones, but this group does not benefit nearly as much from therapy. Overall drug spending increased modestly for health insurance plans, but patient out-of-pocket payments decreased. More significantly, the number of hospitalizations and emergency room visits declined by 79,000 and 31,000, respectively. Taken together, these improvements added up to more than $1 billion in estimated savings.

Although the approach appears promising, researchers warn that there are some potential problems. For example, health insurance plans using this approach might attract higher numbers of sicker patients while deterring healthier patients. Thus, insurance plans pursuing the approach might want to use some of the cost savings to lower premiums for healthy patients or to create some incentives for healthy patients who closely follow their drug regimens.

“We don’t want to penalize people for being healthy,” said coauthor Geoffrey Joyce. “The goal is to encourage people to be healthier — both by taking their medicine and by following healthier lifestyles overall.”

While the research looked only at statins, it seems likely that lowering patient copayments for other drugs that are used to treat chronic illnesses could also prompt more patients to take their prescribed medications, thus lowering medical costs in the United States even further.

---

Most Reservists Earn More Money When Called to Active Duty

“Reservists Pay Steep Price for Service,” trumpeted a headline in a 2003 USA Today story about U.S. military reservists losing income when deployed. But according to Jacob Klerman, lead author of a 2005 RAND study, “most U.S. military reservists called to active duty in 2003 actually saw their earnings rise — a finding that contradicts the conventional wisdom.”

“The average income boost was… about $10,000 per year.”

The conventional wisdom was based on a May 2004 defense department survey of reservists who were called to active duty. In that survey, 51 percent of the reservists reported an earnings loss, including 44 percent of the total who reported a drop of 10 percent or more and an overlapping 21 percent of the total who reported an income loss of 20 percent or more.

But the accuracy of the 2004 survey is questionable. It relied on self-reported earnings, which are subject to error, and only 30 percent of those surveyed responded, which means it may not be representative of all reservists. Moreover, the survey explicitly instructed reservists to report before-tax income, ignoring the fact that military allowances and pay received while serving in a combat zone are tax free.

Reservists Activated for More Days Were Less Likely to Experience Losses

SOURCE: Early Results on Activations and the Earnings of Reservists, 2005. NOTE: In 2001, the reservists who were studied had been activated fewer than 90 active-duty days.

RAND researchers took a different tack, using civilian and military payroll data to estimate earnings losses. Using payroll data addresses many of the concerns with the survey data: Everyone “responds” to payroll data, the data come from computer systems (not individual recall), and it is possible to impute the value of the military tax benefit.

The researchers found that 72 percent of the reservists had experienced pay increases. “The average income boost was 25 percent over what they had earned when not serving on active duty — about $10,000 per year,” said Klerman.

Researchers did find that 28 percent of the reservists studied lost pay after being called to active duty and that about 20 percent of all reservists lost 10 percent or more of their normal income (see the figure). Still, this is a far smaller problem than was noted in the survey data.

The percentage of reservists who experienced gains in income increased with the number of active-duty days served. As noted, across all activations of any duration, an average 72 percent of reservists experienced an income gain and 28 percent experienced a loss. But for reservists who were activated, for example, for more than 271 days during 2002 and 2003, 83 percent experienced a gain and 17 percent experienced a loss.

Despite the overall gains, Klerman pointed out that “the increase in earnings might not be sufficient to compensate reservists for the hardships of active-duty service, like being away from family and in harm’s way.” For that reason, the U.S. Department of Defense could still face challenges when it comes to recruiting and retaining reservists.

For more information:
Early Results on Activations and the Earnings of Reservists, ISBN 0-8330-3819-2, 2005.

Full Document