Poplaski, Joseph J. PhD; Poe, James A. BA; Menn, Eric R. BA
National health reform has been debated in the United States for more than 100 years.1 In March, 2010, Congress passed the Patient Protection and Affordable Care Act (PPACA, Public Law 111–148) and companion legislation, the Health Care and Education Reconciliation Act (H.R. 4872), that will take effect during the next 4 years.
The Commonwealth of Massachusetts enacted health care reform (An Act Providing Access to Affordable, Quality, Accountable Heath Care—Chapter 58 of the Acts of 2006) 4 years before the rest of the United States.2,3
Debate preceding passage of both Massachusetts Chapter 58 and the PPACA centered on myriad issues including coverage, implementation, expense, and roles.4 For the purposes of this discussion, cost is the principal issue, because private employers provide health insurance coverage for 55.8% of, or approximately 170 million, Americans.5 In Massachusetts, approximately 68% of its nearly six-and-a-half million residents are covered by employer-sponsored insurance.6,7
Following passage of the Patient Protection and Affordable Care Act we initiated a study to test the hypothesis that universal access to health providers will lead to significant increases in diagnosis-specific short-term disability (STD) durations associated with specialist physician populations in short supply. These increased durations will directly affect employer costs because short-term disability payments are generally borne by employers.
The study began as an examination of short-term disability (STD) claim durations for Massachusetts-resident claimants whose conditions generally required consultation or treatment by specialist physicians.
Since 2001 the Massachusetts Medical Society has published a Physician Workforce Study to track physician supply in the Commonwealth (PWS 2009).8 Surveys are conducted across 18 specialties [See Table 1]. The PWS 2009 identified seven of these specialties to be in “severe” or “critical” shortage; whereas the 2008 physician workforce data showed 12 specialties facing severe or critical shortages.
Liberty Mutual assigns ICD-9CM codes into Major Diagnostic Categories (MDC). Matching our MDCs to the specialties surveyed by PWS 2009, we compared lengthening of disability durations by diagnosis with specialties that the PWS 2009 identified as in severe or critical shortage in 2008 and 2009.
We used sequential analysis9,10 to evaluate our study population. This method incrementally tests if a hypothesis can be accepted, or rejected, to a specified confidence level, given a presumed probability distribution, as data is received over time. A typical use of sequential analysis is in clinical trials where the entrance and exit of study participants is staggered. The study continues until the method provides a confident assessment of efficacy or inefficacy.
We modified this method to return what confidence could be placed in the affirmative hypothesis that the duration of STD claims was longer after Massachusetts Chapter 58 than in the year prior to its implementation. The distributions of control and test population claim durations were modeled as gamma distributions with identical coefficients of variation, and means equal to the individual populations observed.
We examined a fixed block of Liberty Mutual, Massachusetts residents, closed STD claims up to July 19, 2010. Dates of disability in 2006 (n = 1603) were isolated as a control population because they preceded implementation of Massachusetts Chapter 58. Claims with dates of disability in 2008 (n = 2098), and 2009 (n = 2136) were compared as distinct test populations. Claims incurred in 2007 were not included to avoid “burn-in” effects that might bias the analysis during the first year of Massachusetts Chapter 58 implementation.
Since the study population consisted of employed Massachusetts's residents with medical coverage we presumed an existing relationship with a primary care provider.
The bulk of claims were from four employees in industries, based on Standard Industrial Code (SIC) groupings [See Table 2]. This weighting is not reflective of the actual workforce of Massachusetts [See Table 3]11. It represents the distribution of businesses that are buyers of short-term disability insurance coverage and services from Liberty Mutual.
Geographic Distribution of Claimants Studied
Table 4 shows the geographic distribution of short-term disability claimants included in our study matched against the distribution, by region, of all employed persons in Massachusetts as of April 2009.12
Five specialties were identified by the PWS 2009 to be in short supply in 2008 or 2009. Some specialties studied by the Massachusetts Medical Society were linked to more than one Liberty Mutual MDC. They are associated with Liberty Mutual MDCs as given in Table 5.
Eleven specialties (Anesthesiology, Dermatology, Emergency Medicine, Family Medicine, General Surgery, Internal Medicine, Obstetrics/Gynecology, Pediatrics, Radiology, Urology, and Vascular Surgery) were excluded from our study for the following reasons:
* They were not significant in short-term disability claim management (Dermatology, Family Medicine, Internal Medicine, Obstetrics/Gynecology, Pediatrics);
* A patient would already have an on-going (nondisability related) relationship with the specialist (Dermatology, Family Medicine, Internal Medicine, Obstetrics/Gynecology);
* They were not associable with one of our Major Diagnostic Category (MDC) groupings of ICD-9CM primary diagnosis codes (Anesthesiology, Emergency Medicine, Urology, General Surgery, Radiology, and Vascular Surgery).
* As a control, data associated with two specialties identified in PWS 2009 as not in short supply in either 2008 or 2009 that aligned with Liberty Mutual MDCs were tracked for the study periods:
* Cardiology—Liberty Mutual MDC “heart & circulatory”
* Gastroenterology—Liberty Mutual MDC “digestive system”
Two specialties in short supply according to PWS 2009, neurology and oncology, showed a significant increase in duration during 2008 (See Table 6) and 2009 (See Table 7).
The specialties identified as neurology and neurosurgery in the PWS 2009 correlated to Liberty Mutual's MDC “nervous system.” Average durations for nervous system claims were 46.7 days in 2006. In 2008 durations rose to 54.0 days and in 2009 to 52.3 days. A significance of more than 80% confidence was associated with this change. Relative duration change in 2008 from 2006 was 15.6% with an absolute duration change of 7.3 days; whereas 2009 showed a relative duration change of 11.9% and an absolute duration change of 5.6 days.
The PWS 2009 specialty oncology aligned with Liberty Mutual's MDC “cancer.” Average durations for cancer claims were 57.0 days in 2006. In 2008 durations rose to 63.6 days and in 2009 to 64.4 days. A significance of more than 90% confidence was associated with this change. Relative duration change in 2008 from 2006 was 11.5% with an absolute duration change of 6.5 days; whereas 2009 showed a relative duration change of 12.9% and an absolute duration change of 7.3 days.
Another specialty in short supply, orthopedics (Liberty Mutual MDC “musculoskeletal”), had a significant duration change only in 2009 (>90% confidence). Average musculoskeletal claims were 55.7 days in 2006 and increased to 58.7 days in 2009 with a relative duration change of 5.4% and an absolute duration increase of 3.0 days.
The remaining specialty in short supply, psychiatry (Liberty Mutual MDCs “mental & behavioral” and “drugs & alcohol” [M&B-D&A]), did not have a significant change in disability duration for the test periods.
Cardiology and gastroenterology, the two specialties identified in the PWS 2009 as not in short supply that we compared as a control showed no significant changes in disability durations.
Tables 8 and 9 illustrate increased costs associated with physician specialists in short supply. The four Liberty Mutual MDCs associated with PWS 2009 specialties in severe or critical shortage showed a significant increase in costs of 1.2% in 2008 and 6.5% in 2009. Conversely, Liberty Mutual MDCs tied to PWS 2009 specialties not in short supply showed no significant changes in cost during 2008 and 2009 when compared with 2006.
Actual 2006 costs cannot be directly compared to 2008 costs as differences in benefits amounts skew results. Employer costs were presumed to vary directly with expected durations—the percentage of increase in employer costs would be equal to the percentage of increase in durations. Therefore, 2006 costs were derived by adjusting 2008 costs to reflect the percentage of change in duration between 2006 and 2008.
Disability is a medical event that requires physician treatment. Most employer-sponsored disability plans require that claimants submit medical information to substantiate his or her disability and be under the care and treatment of a physician to continue to receive benefits. Generally, claimants need releases for return to work from their physicians to be allowed to resume their jobs.
Disability plans we studied pay benefits after seven consecutive days of illness or injury. These types of medical episodes usually require specialist referrals. Although most employed individuals have existing relationships with primary care MDs, they usually would not have existing relationships with specialists.
Short-term disability durations for certain diagnoses have increased since the introduction of health care reform in Massachusetts. These disabilities can be associated with specialties that the Massachusetts Medical Society identified as in severe or critical shortage in 2008 and 2009. These results cannot be extrapolated however because employee populations, benefit plan designs, and individual case mixes differ from the Liberty Mutual book of business.
Since the expansion in STD durations associated with physician specialists in short supply is an emerging trend, it is not surprising that a literature search did not reveal any investigation of this topic. We studied it with the hope that our results would inspire others to collect more data and probe further into these phenomena for both Massachusetts and the nation.
To determine if this trend in duration change is Massachusetts-specific, rather than regional in nature, we conducted similar analyses of claims from Northeastern states with sufficient Liberty Mutual claims data. Following the same methodology as our Massachusetts-based study, we reviewed group short-term disability claims of Liberty Mutual customers in Connecticut, New Jersey, and New York (See Table 10). The duration results in these states do not show similar patterns as experienced in Massachusetts. This indicates that there is no regional or global trend driving the results we observed in Massachusetts, which supports our hypothesis that Massachusetts-based duration changes can be attributable to health care reform.
We believe the data demonstrates that longer specialist encounter delays may be attributable to increased demand as a consequence of coverage expansion under Massachusetts's health care reform. The delays seen among employees in Massachusetts may be indicative of trends that might appear as health care reform is established nationwide. Additional data from a broader geographic sample, accumulated over time, may offer a clearer view of STD costs and their implications for employers who generally are the primary payers of short-term disability benefits.
Ms. Erin M. Finnemore provided vital editorial review and consultation in preparation of this manuscript. Glenn S. Pransky, MD, Director, Liberty Mutual Center for Disability Research provided invaluable comments in reviewing our study and findings.
1. Birn AE, Brown TM, Fee E, Lear WJ Struggles for national health reform in the United States. Am J Public Health 93: 86–91.
3. Steinbrook R Health care reform in Massachusetts—a work in progress. N Engl J Med 2006;354:2095–2098.
5. U.S. Department of Commerce. U.S. Census Bureau. Health, poverty, and health insurance coverage in the United States: 2009. Current population reports. September 2010. pp. 30–31.
6. Long SK, Cook A, Stockley K Health insurance coverage in Massachusetts: Estimates from the 2008 Massachusetts health insurance survey. Boston; Massachusetts Division of Health Care Finance and Policy. 2008. p. 17. http://www.shadac.org/files/MA_2008_HH_Findings.pdf
. Accessed April 8, 2011.
9. Hogg RV, Craig A Introduction to mathematical statistics, 4th Edition, New York, NY: MacMillan; 1978; pp. 374–385.
10. Lindgren BW Statistical theory, 2nd Edition, New York, NY: Collier-MacMillan Intern. Editions; 1968; pp. 320–321.