Abstract
This study assessed and described the episode rate, duration of illness, and health care utilization and costs associated with acute gastroenteritis (AGE), norovirus gastroenteritis (NVGE), and rotavirus gastroenteritis (RVGE) in physician office, emergency department (ED), and inpatient care settings in the United States (US). The retrospective analysis was conducted using an administrative insurance claims database (2006–2011). AGE episode rates were assessed using medical (ICD-9-CM) codes for AGE; whereas a previously published “indirect” method was used in assessing estimated episode rates of NVGE and RVGE. We calculated per-patient, per-episode and total costs incurred in three care settings for the three diseases over five seasons. For each season, we extrapolated the total economic burden associated with the diseases to the US population. The overall AGE episode rate in the physician office care setting declined by 15% during the study period; whereas the AGE episode rate remained stable in the inpatient care setting. AGE-related total costs (inflation-adjusted) per 100 000 plan members increased by 28% during the 2010–2011 season, compared with the 2006–2007 season ($832,849 vs. $1 068 116) primarily due to increase in AGE-related inpatient costs. On average, the duration of illness for NVGE and RVGE was 1 day longer than the duration of illness for AGE (mean: 2 days). Nationally, the average AGE-related estimated total cost was $3.88 billion; NVGE and RVGE each accounted for 7% of this total. The episodes of RVGE among pediatric populations have declined; however, NVGE, RVGE and AGE continue to pose a substantial burden among managed care enrollees. In conclusion, the study further reaffirms that RVGE has continued to decline in pediatric population post-launch of the rotavirus vaccination program and provides RVGE- and NVGE-related costs and utilization estimates which can serve as a resource for researchers and policy makers to conduct cost-effectiveness studies for prevention programs.
Keywords: acute gastroenteritis, norovirus, rotavirus, health care cost, duration of illness
Introduction
Acute gastroenteritis (AGE) is defined as inflammation of the stomach and small and large intestines, resulting in diarrhea, nausea, vomiting, fever, or abdominal pain.1 Approximately 179 million incident cases of AGE occur annually in the United States (US); of these cases, approximately 600 000 patients require hospitalization and nearly 5000 individuals die because of AGE.1,2 In the US, 12% of all AGE cases across all age groups are attributable to norovirus gastroenteritis (NVGE).3 Annually, NVGE accounts for over 70 000 hospitalizations and 800 deaths in young children and older adults.3 Rotavirus gastroenteritis (RVGE) is another cause of AGE in the US, with disproportionally higher infections observed among young children than adults.
Currently, there are two vaccines available to prevent rotavirus infections: RotaTeq (a live, oral, human-bovine reassortant vaccine) and Rotarix (a live, oral, human-attenuated vaccine), which have high efficacy in preventing rotavirus gastroenteritis as well as hospitalizations due to rotavirus gastroenteritis.4,5 RVGE has declined considerably since the introduction in 2006 of the rotavirus vaccination program. For example, a study assessing rotavirus-related hospitalization rates among pediatric population (age <5 y) reported 75% decline in 2007–2008 and 60% in 2008–2009 in hospitalization rates compared with pre-vaccination period of 2001–2006.6 Several groups are working to develop a norovirus vaccine, with the lead candidate already tested in humans, and at least 2 more candidates planned to enter clinical trials in the next few years.7
In general, AGE, NVGE, and RVGE lead to a significant burden to the health care system. A recent study suggested that hospitalization due to NVGE cost nearly $500 million annually.8 Similarly, before the introduction of the rotavirus vaccination program, the total annual direct and indirect costs for RVGE were estimated at approximately $1 billion, of which $264 million were attributable to direct medical costs.9 However, a recent study conducted in a managed care setting reported a reduction of over $157 (33%) million and $119 (25%) million in hospitalization costs during the postvaccination program periods 2007–2008 and 2008–2009, respectively, compared with the prevaccination program period (2001–2006) costs (473 million).6
Overall, recent studies have compared the prevaccination program and postvaccination program RVGE rates in pediatric populations;9-12 however, limited data exist that report the current cost of RVGE across all age groups in the US. Furthermore, there are few studies that assess AGE, NVGE, and RVGE postvaccination program rates among individuals enrolled in employer-sponsored managed care health plans. Thus, the objective of this study was to assess and describe the episode rates, durations of illness, and health care utilization and costs associated with AGE, NVGE, and RVGE in physician’s office, emergency department (ED), and inpatient care settings. Prior research suggests that the use of NVGE and RVGE-specific diagnosis codes is infrequent and a majority of AGE claims include non-specific organism codes.6,8 Thus, to address this issue we used an indirect method (based on Poisson regression) as described by Lopman and colleagues5 in which the estimation of norovirus-related AGE claims was conducted using the cause-unspecified AGE claims. Details on the indirect approach used in the estimation of norovirus and rotavirus-related AGE based on cause-unspecified AGE claims are described in the methods section.
Results
Patient Characteristics
Demographic characteristics for the three study groups, by care setting and seasons, are presented in Table 1.
Table 1. Demographic Characteristics for Patients With Acute Gastroenteritis, Norovirus, and Rotavirus in the Physician’s Office, Emergency Department, and Inpatient Care Settings, by Seasons.
| Acute Gastroenteritis | Norovirus | Rotavirus | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2006 -2007 | 2007 -2008 | 2008 -2009 | 2009 -2010 | 2010 -2011 | 2006 -2007 | 2007 -2008 | 2008 -2009 | 2009 -2010 | 2010 -2011 | 2006 -2007 | 2007 -2008 | 2008 -2009 | 2009 -2010 | 2010 -2011 | |
| Care setting: physician’s office | |||||||||||||||
| Total (N) | 592,214 | 767,690 | 964,699 | 908,674 | 781,366 | 47 | 32 | 59 | 84 | 84 | 2,199 | 1,624 | 2,154 | 938 | 1,060 |
| Sex | |||||||||||||||
| Male (%) | 45% | 45% | 45% | 44% | 44% | 45% | 59% | 47% | 38% | 42% | 54% | 53% | 56% | 50% | 51% |
| Female (%) | 55% | 55% | 55% | 56% | 56% | 55% | 41% | 53% | 62% | 58% | 46% | 47% | 44% | 50% | 49% |
| Age | |||||||||||||||
| Mean (SD) | 30 (24) | 30 (23) | 30 (23) | 32 (23) | 33 (24) | 42 (24) | 41 (28) | 38 (22) | 41 (25) | 40 (21) | 5 (12) | 6 (14) | 6 (13) | 10 (18) | 10 (17) |
| Median | 30 | 30 | 29 | 31 | 32 | 44 | 41 | 38 | 42 | 39 | 1 | 1 | 1 | 1 | 2 |
| Region | |||||||||||||||
| Northeast (%) | 10% | 13% | 15% | 15% | 15% | 6% | 6% | 34% | 24% | 24% | 7% | 8% | 10% | 14% | 14% |
| North Central (%) | 21% | 23% | 24% | 22% | 21% | 21% | 19% | 22% | 12% | 38% | 26% | 27% | 30% | 19% | 25% |
| South (%) | 51% | 45% | 43% | 44% | 42% | 28% | 38% | 27% | 46% | 20% | 55% | 48% | 46% | 52% | 46% |
| West (%) | 18% | 17% | 16% | 18% | 20% | 43% | 34% | 15% | 17% | 15% | 12% | 15% | 12% | 13% | 13% |
| Unknown (%) | 1% | 2% | 2% | 2% | 2% | 2% | 3% | 2% | 1% | 2% | 1% | 2% | 2% | 2% | 2% |
| Care setting: emergency department | |||||||||||||||
| Total (N) | 129,030 | 159,793 | 213,233 | 202,689 | 174,376 | 9 | 12 | 16 | 23 | 21 | 951 | 444 | 899 | 158 | 386 |
| Sex | |||||||||||||||
| Male (%) | 42% | 42% | 42% | 40% | 40% | 67% | 67% | 44% | 48% | 29% | 56% | 55% | 61% | 58% | 57% |
| Female (%) | 58% | 58% | 58% | 60% | 60% | 33% | 33% | 56% | 52% | 71% | 44% | 45% | 39% | 42% | 43% |
| Age | |||||||||||||||
| Mean (SD) | 32 (23) | 32 (22) | 31 (22) | 33 (21) | 34 (23) | 57 (29) | 43 (29) | 38 (30) | 48 (29) | 49 (27) | 4 (12) | 5 (13) | 5 (11) | 10 (18) | 8 (15) |
| Median | 30 | 30 | 30 | 31 | 32 | 62 | 45 | 46 | 45 | 48 | 1 | 1 | 2 | 3 | 3 |
| Region | |||||||||||||||
| Northeast (%) | 10% | 13% | 17% | 17% | 16% | 22% | 8% | 25% | 22% | 19% | 5% | 5% | 9% | 9% | 10% |
| North Central (%) | 23% | 24% | 25% | 24% | 25% | 22% | 17% | 31% | 17% | 48% | 35% | 36% | 42% | 22% | 35% |
| South (%) | 51% | 45% | 42% | 43% | 41% | 44% | 33% | 25% | 30% | 14% | 51% | 45% | 40% | 50% | 39% |
| West (%) | 15% | 15% | 14% | 15% | 16% | 11% | 42% | 13% | 30% | 14% | 8% | 11% | 6% | 17% | 13% |
| Unknown (%) | 1% | 2% | 3% | 2% | 2% | 0% | 0% | 6% | 0% | 5% | 1% | 3% | 3% | 1% | 3% |
| Care setting: inpatient | |||||||||||||||
| Total (N) | 46,468 | 61,882 | 82,123 | 78,593 | 75,100 | 39 | 32 | 66 | 70 | 81 | 1,683 | 753 | 1,459 | 262 | 701 |
| Sex | |||||||||||||||
| Male (%) | 40% | 40% | 40% | 39% | 39% | 49% | 34% | 52% | 41% | 48% | 53% | 54% | 59% | 52% | 53% |
| Female (%) | 60% | 60% | 60% | 61% | 61% | 51% | 66% | 48% | 59% | 52% | 47% | 46% | 41% | 48% | 47% |
| Age | |||||||||||||||
| Mean (SD) | 45 (26) | 47 (25) | 47 (25) | 49 (24) | 52 (24) | 54 (28) | 49 (34) | 48 (29) | 51 (27) | 53 (27) | 4 (11) | 5 (13) | 5 (11) | 7 (15) | 8 (16) |
| Median | 49 | 51 | 51 | 52 | 56 | 56 | 57 | 51 | 54 | 56 | 1 | 1 | 2 | 2 | 3 |
| Region | |||||||||||||||
| Northeast (%) | 11% | 15% | 18% | 18% | 17% | 3% | 25% | 20% | 17% | 21% | 9% | 9% | 11% | 15% | 11% |
| North Central (%) | 25% | 27% | 28% | 28% | 30% | 26% | 25% | 36% | 13% | 33% | 28% | 26% | 33% | 14% | 28% |
| South (%) | 49% | 41% | 37% | 38% | 35% | 38% | 31% | 29% | 37% | 21% | 51% | 42% | 47% | 53% | 43% |
| West (%) | 14% | 15% | 14% | 15% | 17% | 33% | 19% | 14% | 33% | 21% | 10% | 20% | 8% | 18% | 14% |
| Unknown (%) | 1% | 2% | 3% | 1% | 2% | 0% | 0% | 2% | 0% | 4% | 1% | 3% | 2% | 1% | 3% |
SD = standard deviation. Patient demographic characteristics based on medical claims for medically coded for acute gastroenteritis, norovirus infection, and rotavirus infection.
The mean age for patients with a medically coded diagnosis for RVGE in a physician’s office or ED setting was considerably lower (range: 5–10 y) than for patients with AGE (range: 30–34 y) in these setting; for patients with an NVGE diagnosis, mean age was higher (range: 38–42 y) in these settings. Patients diagnosed with AGE and NVGE in an inpatient setting were on average 10 or more years older than patients diagnosed with AGE and NVGE in a physician’s office or ED setting. Across all three care settings, a greater proportion of patients with AGE were females, whereas a higher proportion of patients with medically coded diagnoses of RVGE were males. Sex distribution for patients with medically coded diagnoses of NVGE varied by care settings and seasons.
AGE-, NVGE-, and RVGE-Related Durations of Illness
Across all seasons, the average duration of illness (across all age groups) for AGE was 2 d (Table 2). The average duration of illness for NVGE (3 d) and for RVGE (3 d) was 1 d longer. For the three cohorts, the duration of illness varied by age groups. The NVGE duration of illness varied from 2 d to 5 d among patients aged 0 to 4 y. By comparison, RVGE duration of illness (3 d) and AGE duration of illness (2 d) were stable in this age group. For a majority of seasons, the NVGE duration of illness was greater among the elderly than among the pediatric and adult age groups. Finally, across all age groups, NVGE and RVGE-related durations of illness were greater than AGE-related duration of illness.
Table 2. Duration of Illness Associated With Acute Gastroenteritis, Norovirus, and Rotavirus in Physician’s Office, Emergency Department and Inpatient Care Settings, by Seasons.
| 2006–2007 | 2007–2008 | 2008–2009 | 2009–2010 | 2010–2011 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean | SD | Median | Mean | SD | Median | Mean | SD | Median | Mean | SD | Median | Mean | SD | Median | |
| Overall duration of illness (in days) | |||||||||||||||
| AGE related | 2 | 3 | 1 | 2 | 3 | 1 | 2 | 3 | 1 | 2 | 3 | 1 | 2 | 3 | 1 |
| Norovirus related | 3 | 4 | 1 | 3 | 5 | 1 | 3 | 4 | 1 | 3 | 4 | 1 | 3 | 4 | 1 |
| Rotavirus related | 3 | 7 | 1 | 3 | 4 | 1 | 3 | 4 | 1 | 2 | 3 | 1 | 3 | 3 | 1 |
AGE = acute gastroenteritis; SD = standard deviation.
AGE-, NVGE-, and RVGE-Related Episode Rates
The AGE episode rate gradually declined (except for 2008–2009) over time in the physician office setting. The AGE episode rate was 1,824 per 100 000 plan members in 2006–2007, declining to 1,557 per 100 000 plan members in 2010–2011, an overall reduction of 15% (Table 3). Although the AGE episode rate declined over time among patients aged 0 to 4 y, this group had the highest AGE episode rate in the physician’s office setting than the other age groups (range: 5928–8068 per 100 000 plan members). The NVGE episode rate in the physician’s office setting was 59 per 100 000 plan members during the 2006–2007 season. The rate peaked during the 2007–2008 season (95 per 100 000 plan members) and then remained relatively stable for the 2009–2010 (60 per 100 000 plan members) and 2010–2011 (53 per 100 000 plan members) seasons. Similarly, the RVGE episode rate in the physician’s office setting was 119 per 100 000 plan members during the 2007–2008 season and remained in the range 38 to 62 per 100 000 plan members for the other seasons. Among patients aged 0 to 4 y, the RVGE episode rate in the physician’s office setting declined more than 62% during the study period, from 1033 per 100 000 plan members in the 2006–2007 season to 391 per 100 000 plan members in the 2010–2011 season.
Table 3. Season Episode Rate of Acute Gastroenteritis, Norovirus, and Rotavirus in Physician’s Office, Emergency Department, and Inpatient Care Settings, by Seasons.
| Acute Gastroenteritis | Norovirus | Rotavirus | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2006 -2007 | 2007 -2008 | 2008 -2009 | 2009 -2010 | 2010 -2011 | 2006 -2007 | 2007 -2008 | 2008 -2009 | 2009 -2010 | 2010 -2011 | 2006 -2007 | 2007 -2008 | 2008 -2009 | 2009 -2010 | 2010 -2011 | |
| Physician’s office: episode rate per 100,000 plan members | |||||||||||||||
| Overall | 1,824 | 1,647 | 1,796 | 1,638 | 1,557 | 59 | 95 | 34 | 60 | 53 | 54 | 119 | 46 | 38 | 62 |
| By age groups | |||||||||||||||
| 0–4 y | 8,068 | 6,982 | 7,741 | 5,928 | 6,001 | 874 | 838 | 447 | 482 | 232 | 1,033 | 703 | 587 | 483 | 391 |
| 5–17 y | 1,473 | 1,398 | 1,580 | 1,434 | 1,297 | 94 | 95 | 106 | 112 | 130 | 145 | 92 | 133 | 139 | 145 |
| 18–64 y | 1,476 | 1,360 | 1,478 | 1,408 | 1,303 | 66 | 54 | 36 | 59 | 69 | 36 | 60 | 31 | 53 | 70 |
| 65–74 y | 1,544 | 1,215 | 1,275 | 1,312 | 1,482 | 110 | 66 | 88 | 34 | 94 | 79 | 63 | 87 | 28 | 96 |
| 75–84 y | 2,296 | 1,787 | 1,893 | 1,979 | 2,239 | 145 | 227 | 112 | 89 | 126 | 132 | 169 | 151 | 137 | 120 |
| 85 y and over | 1,698 | 1,312 | 1,425 | 1,439 | 1,675 | 150 | 81 | 132 | 149 | 99 | 297 | 149 | 138 | 166 | 121 |
| Emergency department: episode rate per 100,000 plan members | |||||||||||||||
| Overall | 358 | 309 | 357 | 327 | 309 | 51 | 36 | 55 | 51 | 29 | 33 | 49 | 35 | 24 | 15 |
| By age groups | |||||||||||||||
| 0–4 y | 1,225 | 985 | 1,182 | 789 | 824 | 217 | 237 | 284 | 94 | 156 | 283 | 159 | 293 | 102 | 121 |
| 5–17 y | 236 | 219 | 262 | 256 | 224 | 59 | 36 | 59 | 41 | 30 | 63 | 37 | 61 | 38 | 39 |
| 18–64 y | 326 | 289 | 332 | 320 | 295 | 46 | 31 | 43 | 81 | 31 | 35 | 30 | 52 | 51 | 24 |
| 65–74 y | 283 | 196 | 205 | 191 | 225 | 70 | 24 | 34 | 33 | 25 | 62 | 30 | 36 | 47 | 26 |
| 75–84 y | 488 | 355 | 385 | 372 | 431 | 105 | 49 | 51 | 54 | 46 | 97 | 43 | 68 | 61 | 42 |
| 85 y and over | 550 | 381 | 409 | 386 | 488 | 174 | 80 | 96 | 87 | 81 | 101 | 81 | 82 | 27 | 83 |
| Inpatient: episode rate per 100,000 plan members | |||||||||||||||
| Overall | 132 | 124 | 143 | 132 | 139 | 3 | 5 | 8 | 9 | 4 | 8 | 6 | 10 | 5 | 6 |
| By age groups | |||||||||||||||
| 0–4 y | 348 | 237 | 282 | 164 | 171 | 41 | 47 | 57 | 31 | 10 | 153 | 64 | 88 | 39 | 33 |
| 5–17 y | 47 | 42 | 50 | 44 | 42 | 1 | 4 | 6 | 6 | 5 | 5 | 6 | 8 | 6 | 6 |
| 18–64 y | 112 | 109 | 126 | 122 | 118 | 5 | 4 | 6 | 6 | 4 | 5 | 4 | 7 | 4 | 5 |
| 65–74 y | 236 | 222 | 237 | 233 | 281 | 30 | 17 | 24 | 15 | 30 | 28 | 18 | 17 | 18 | 17 |
| 75–84 y | 533 | 501 | 549 | 528 | 629 | 98 | 33 | 44 | 46 | 24 | 81 | 29 | 48 | 25 | 16 |
| 85 y and over | 649 | 606 | 681 | 602 | 769 | 101 | 65 | 100 | 55 | 55 | 138 | 42 | 92 | 44 | 48 |
During the five seasons, AGE episode rate in the ED setting varied from 309 per 100 000 plan members in the 2006–2007 season to 358 per 100 000 plan members in the 2010–2011 season. Similarly, among pediatric patients aged 0 to 4 y, AGE rate in the ED setting was highest (1225 per 100 000 plan members) during the 2006–2007 season and lowest (789 per 100 000 plan members) during the 2009–2010 season. In comparison, children aged 5 to 17 y had a much lower episode rate in the ED setting, ranging from 219 to 262 per 100 000 plan members in 2006–2007 and 2010–2011, respectively. Finally, during the five seasons, no specific trends of increasing or decreasing AGE episode rates in the ED setting were observed over time among the adult (18–64 y) and elderly age groups.
Among patients diagnosed with AGE in the inpatient setting, the AGE episode rate remained relatively stable during the five seasons, ranging from 124 per 100 000 plan members in 2007–2008 to 143 per 100 000 plan members in 2008–2009. The AGE episode rate in the inpatient setting among patients aged 0 to 4 y was 348 per 100 000 plan members in the 2006–2007 season, which declined to 171 per 100 000 plan members during the 2010–2011 season. The AGE episode rate in the inpatient setting was higher among the elderly than in the adult age group (range: 109–126 per 100 000 plan members) and increased with increasing age (65–74 y: range, 222–281; 75–84 y: range, 501–628; ≥ 85 y: range, 602–769 per 100 000 plan members). The NVGE episode rate in the inpatient setting was low and ranged from 3 to 9 per 100,000 plan members during all the study seasons. A greater NVGE episode rate in the inpatient setting was observed among the elderly (65–74 y: range, 15–30; 75–84 y: range, 24–98; ≥ 85 y: range, 55–101 per 100,000 plan members) than among patients in the adult age group. The RVGE episode rate in the inpatient setting ranged from 5 to 10 per 100 000 plan members during the all seasons considered for this study.
AGE-, NVGE-, and RVGE-Related Health Care Costs
During the five seasons, we observed a gradual increase in AGE-, NVGE-, and RVGE-related median physician’s office costs per episode. The AGE-related median physician’s office costs per episode increased by 8% during the 2010–2011 season when compared with costs for the 2006–2007 season (US $56) (Table 4). Similarly, the NVGE- and RVGE-related median physician’s office costs per episode increased by 14% and 13%, respectively, during the 2010–2011 season as compared with the 2006–2007 season. The median physician’s office costs per episode were highest in the adult age group (18–64 y) (range: $62-$68) across all seasons than in all other age groups. In contrast, for NVGE (range: $67-$115) and RVGE (range: $67-$102), the median physician’s office costs per episode across all seasons were highest in patients aged 0- to 4 y. The AGE-related median physician’s office costs per episode among the elderly ranged from $12 to $19.
Table 4. Per-Patient and per-Episode Costs Associated With Acute Gastroenteritis, Norovirus, and Rotavirus in Physician’s Office, Emergency Department and Inpatient Care Settings, by Seasons.
| 2006–2007 | 2007–2008 | 2008–2009 | 2009–2010 | 2010–2011 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean | SD | Median | Mean | SD | Median | Mean | SD | Median | Mean | SD | Median | Mean | SD | Median | |
| Disease-related per patient and per episode, by care setting | |||||||||||||||
| Physician’s office | |||||||||||||||
| Per-patient AGE-related costs | $92 | $342 | $59 | $99 | $421 | $62 | $103 | $404 | $64 | $103 | $336 | $64 | $103 | $396 | $63 |
| Per-patient norovirus-related costs | $63 | $64 | $52 | $63 | $102 | $52 | $66 | $60 | $58 | $70 | $83 | $58 | $79 | $75 | $59 |
| Per-patient rotavirus-related costs | $99 | $108 | $65 | $129 | $152 | $81 | $128 | $163 | $73 | $191 | $260 | $87 | $145 | $210 | $74 |
| AGE-related per episode costs | $77 | $143 | $56 | $82 | $159 | $59 | $85 | $161 | $61 | $85 | $160 | $61 | $84 | $154 | $60 |
| Norovirus-related per episode costs | $63 | $64 | $52 | $63 | $102 | $52 | $66 | $60 | $58 | $70 | $83 | $58 | $77 | $74 | $59 |
| Rotavirus-related per episode costs | $88 | $79 | $65 | $100 | $89 | $78 | $101 | $93 | $72 | $120 | $111 | $83 | $110 | $133 | $74 |
| Emergency department | |||||||||||||||
| Per-patient AGE-related costs | $528 | $838 | $287 | $523 | $835 | $295 | $521 | $854 | $300 | $528 | $835 | $302 | $555 | $879 | $313 |
| Per-patient norovirus-related costs | $1502 | $2294 | $518 | $598 | $890 | $304 | $340 | $397 | $262 | $450 | $472 | $303 | $350 | $367 | $176 |
| Per-patient rotavirus-related costs | $631 | $810 | $376 | $628 | $2141 | $321 | $615 | $913 | $346 | $780 | $1299 | $368 | $672 | $782 | $406 |
| AGE-related per episode costs | $516 | $815 | $283 | $511 | $805 | $290 | $509 | $827 | $296 | $516 | $808 | $297 | $542 | $854 | $309 |
| Norovirus-related per episode costs | $1502 | $2294 | $518 | $598 | $890 | $304 | $340 | $397 | $262 | $450 | $472 | $303 | $350 | $367 | $176 |
| Rotavirus-related per episode costs | $625 | $796 | $374 | $628 | $2141 | $321 | $615 | $913 | $346 | $780 | $1299 | $368 | $672 | $782 | $406 |
| Inpatient | |||||||||||||||
| Per-patient AGE-related costs | $12 368 | $38 824 | $4830 | $14 370 | $42 710 | $5810 | $15 014 | $43 096 | $6133 | $15 406 | $43 186 | $6456 | $14 948 | $42 521 | $6446 |
| Per-patient norovirus-related costs | $6209 | $7490 | $4160 | $6,666 | $7,968 | $3050 | $8123 | $11 551 | $4199 | $25 816 | $107 929 | $6,488 | $17,984 | $33 225 | $6984 |
| Per-patient rotavirus-related costs | $4477 | $7019 | $3300 | $6414 | $16,219 | $3600 | $6504 | $24 153 | $3786 | $10 306 | $32 755 | $4,166 | $6513 | $8680 | $4703 |
| AGE related per episode costs | $11 240 | $34 991 | $4766 | $12 764 | $34 932 | $5715 | $13 322 | $36 495 | $6042 | $13 788 | $37 314 | $6348 | $13 346 | $35 890 | $6339 |
| Norovirus-related per episode costs | $6209 | $7,490 | $4160 | $6666 | $7968 | $3050 | $7574 | $10,085 | $4199 | $18 913 | $56 483 | $6488 | $17 373 | $32 526 | $6984 |
| Rotavirus-related per episode costs | $4457 | $6,957 | $3300 | $6303 | $15657 | $3600 | $6477 | $24 099 | $3786 | $10 306 | $32 755 | $4166 | $6505 | $8678 | $4703 |
AGE = acute gastroenteritis; SD = standard deviation.
In the inpatient care setting, the median costs per AGE, NVGE, or RVGE episode increased by 33%, 68%, and 42%, respectively, during the study period. In comparison with the 2006–2007 season ($832,849 per 100 000 plan members), the total AGE-related costs across all care settings increased by 28% in the 2010–2011 season ($1 068 116 per 100 000 plan members) (Fig. 1). Across all seasons, costs incurred for AGE in the inpatient setting accounted for over 75% of the total AGE-related costs. Finally, given the considerably higher episode rate for the three conditions and across all care settings among patients aged 0 to 4 y, the costs burden per 100,000 plan members across all seasons was highest of all age groups in the study.
Figure 1. Acute Gastroenteritis, Norovirus and Rotavirus-Related Total Costs per 100,000 Plan Members in Physician’s Office, Emergency Department, and Inpatient Care Setting, by Seasons. AGE = acute gastroenteritis; ED = emergency department; US = United States.
National Estimates for AGE-, NVGE-, and RVGE-Related Episodes and Health Care Costs
The national estimates for AGE, NVGE, and RVGE are presented in Table 5. Across the five seasons, the estimated total number of AGE-related episodes in the physician’s office setting ranged from 5.2 to 5.9 million. During the same seasons, the estimated total number of NVGE and RVGE episodes in the physician’s office ranged from 0.4 to 0.5 million. The estimated AGE-related episodes in the ED and inpatient settings ranged from 1.0 million to 1.1 million and from 0.4 million to 0.5 million, respectively. The total AGE-related seasonal costs burden across the five seasons ranged from $3.30 billion (2006–2007) to $4.28 billion (2010–2001), with inpatient utilization accounting for a majority (77–83%) of the total costs. During the five seasons, NVGE-related total seasonal costs ranged from $180 million to $355 million. The RVGE-related total seasonal costs ranged from $217 million to $367 million.
Table 5. Estimated National Episodes and Costs Associated With Acute Gastroenteritis, Norovirus, and Rotavirus in Physician’s Office, Emergency Department, and Inpatient Care Setting, by Seasons.
| 2006–2007 | 2007–2008 | 2008–2009 | 2009–2010 | 2010–2011 | |
|---|---|---|---|---|---|
| Physician’s office | |||||
| Estimated annual episodes | |||||
| Acute gastroenteritis | 5 904 052 | 5 319 289 | 5 865 348 | 5 337 473 | 5 189 408 |
| Norovirus | 399 498 | 372 044 | 257 276 | 299 910 | 294 605 |
| Rotavirus | 402 677 | 350 267 | 295 687 | 308 404 | 338 865 |
| Estimated annual costs | |||||
| Acute gastroenteritis | $387 306 574 | $368 643 591 | $418 034 406 | $381 349 985 | $366 331 748 |
| Norovirus | $24 236 175 | $22 836 597 | $17 567 740 | $20 320 794 | $20 513 698 |
| Rotavirus | $30 667 305 | $32 007 708 | $24 778 282 | $29 895 240 | $29 226 426 |
| Emergency department | |||||
| Estimated annual episodes | |||||
| Acute gastroenteritis | 1 137 870 | 975 492 | 1 136 838 | 1 032 064 | 1 001 326 |
| Norovirus | 198 585 | 142 246 | 191 658 | 217 514 | 124 026 |
| Rotavirus | 187 179 | 124 930 | 214 085 | 161 072 | 108 838 |
| Estimated annual costs | |||||
| Acute gastroenteritis | $376 875 628 | $332 056 483 | $394 438 668 | $359 368 172 | $362 370 531 |
| Norovirus | $120 444 526 | $50 617 450 | $58 731 394 | $77 212 475 | $25 512 913 |
| Rotavirus | $82 131 868 | $47 005 785 | $86 782 695 | $69 464 761 | $51 789 779 |
| Inpatient | |||||
| Estimated annual episodes | |||||
| Acute gastroenteritis | 455 770 | 421 112 | 486 034 | 447 580 | 478 210 |
| Norovirus | 41 495 | 29 824 | 43 333 | 33 869 | 24 947 |
| Rotavirus | 65 909 | 32 706 | 50 886 | 28 206 | 27 652 |
| Estimated annual costs | |||||
| Acute gastroenteritis | $2 545 707 077 | $2 820,356,324 | $3 441 446 597 | $3 329 521 191 | $3 552 262 036 |
| Norovirus | $202 269 807 | $106,599 601 | $213 248 266 | $257 489 428 | $204 178 967 |
| Rotavirus | $254 909 581 | $137 985 242 | $225 734 224 | $137 694 016 | $152 383 709 |
| Overall | |||||
| Estimated annual costs | |||||
| Acute gastroenteritis | $3 309 889 279 | $3 521 056 398 | $4 253 919 672 | $4 070 239 348 | $4 280 964 315 |
| Norovirus | $346 950 507 | $180 053 649 | $289 547 400 | $355 022 697 | $250 205 578 |
| Rotavirus | $367 708 754 | $216 998 735 | $337 295 200 | $237 054 017 | $233 399 913 |
Discussion
The primary objective of this study was to estimate the episode rate of AGE, NVGE, and RVGE in physician’s office, ED, and inpatient care settings for all patients and as stratified by age groups. Additionally, this study assessed disease duration and disease-related health care utilization and costs associated with AGE, NVGE, and RVGE across the three care settings. We observed an overall decrease in the rate of RVGE episodes across all three care settings, especially in patients aged 0 to 4 y. Overall, AGE, NVGE, and RVGE rates varied considerably across the five seasons considered for this study. The overall rate of AGE episodes appeared to peak in the 2006–2007 and 2008–2009 seasons. The cost trends per 100 000 plan members followed the episode rate trends for AGE, NVGE, and RVGE.
Across all age groups, we did not observe any consistent (e.g., increase or decline) trends in NVGE across the five seasons considered for this study. Overall, NVGE accounted for 4% of the total AGE episodes in the physician’s office setting and 13% of the total AGE episodes in the ED setting. However, these proportions varied considerably with age groups. For example, among patients aged 0 to 4 y and 5 to 17 y, NVGE accounted for 15% and 10% of the total AGE episodes in the inpatient setting, respectively. Similarly, NVGE accounted for greater proportions of the total AGE episodes in the inpatient setting among the three elderly age groups (range: 7% to 11%), than among the adult age group (4%). The variability in episode rate across seasons and age groups observed in our study was consistent with a recent study assessing NVGE-associated hospital discharges by age groups and season, using a nationally representative data set of hospital inpatient stays.6 Additionally, the study conducted by Lopman and colleagues reported increasing incidence of cause-unspecified and NVGE-related hospital discharges among the elderly, which peaked during the 2006–2007 season.6 Similarly, we observed the highest episode rate of AGE and NVGE among the elderly during the 2006–2007 season across the three care settings; this rate dropped by over 40% in the following season (2007–2008) and remained relatively stable during the remaining three seasons. However, similar trends were not observed for the pediatric and adult age groups. The increase in the NVGE episode rate among the elderly during the 2006–2007 season coincides with the 2006–2007 AGE and NVGE epidemic, during which two new strains of norovirus were identified.13,14 During the 2006–2007 norovirus outbreak, a large proportion of AGE cases were identified among residents of long-term care facilities, which may help explain the increased episode rate among the elderly observed in this study.13
The NVGE episode rate across seasons in the ED and inpatient care settings among patients aged 0 to 4 y followed the RVGE episode rate, which peaked during the 2008–2009 season but declined during all other seasons. Caution should be exercised in interpreting the declining NVGE episode rate among patients aged 0 to 4 y. Using the indirect method, we made use of cause-unspecified cases to predict NVGE and RVGE episodes. The introduction in 2006 of the rotavirus vaccination program may have also resulted in a lower number of cause-unspecified AGE infections, which would impact our cause-unspecified prediction model for NVGE. The decline in the rate in the pediatric age group may be reflective of the rotavirus vaccination program and not a decline in the actual episode rate of NVGE. A recent study conducted using laboratory-confirmed NVGE cases among children younger than aged 5 y reported a higher rate of NVGE-related outpatient, ED, and inpatient utilization than the rates observed in our study.15 Thus, the indirect method employed likely underestimates the burden of NVGE. Additional research is required to assess trends in the rate of laboratory-confirmed NVGE cases during the 2006–2011 period across all age groups.
In this study, on average across all age groups, RVGE accounted for 5% and 9% of the total AGE episodes in the ED and inpatient care settings, respectively. Among pediatric patients aged 0 to 4 y, RVGE accounted for 18% and 29% of the total AGE episodes in the ED and inpatient care settings, respectively. However, during the five seasons, RVGE episodes declined considerably among patients aged 0 to 4 y in both the ED (36% decline) and the inpatient (57% decline) settings. These findings are consistent with prior studies that have reported a decline in RVGE-related cases following the introduction in 2006 of the rotavirus vaccine.6,10-12 For example, a study conducted by Payne and colleagues reported that compared with the 2006 season, a 87%, 96% and 92% reduction in rotavirus-related hospitalization rates was observed among patients in the age groups 6–11 mo, 12–23 mo and 24–35 mo, respectively.11 Similarly, findings from a study conducted another by Desai and colleagues indicate that compared with the 2000–2006 period (15 per 10 000), rotavirus-related hospitalized declined by 66% in 2008 (6 per 10 000) and 60% in 2009 (6 per 10 000) among children <5 y of age.12 Similar reductions in rotavirus rates have been reported in studies from other countries following the introduction of a rotavirus vaccine.16-18 A study conducted by Zlamy and colleagues reported that there was a 74% reduction of RVGE-related hospitalizations per year after introduction of the universal mass vaccination program against rotavirus in 2008 in Austria.16
In comparison with previous publications, this study documents the previously unknown RVGE episode rate among older children (5 to 17 y), adults (18 to 64 y) and elderly (≥65 y) enrolled in employer-sponsored managed care health plans. Among older children (5 to 17 y), RVGE in the ED and inpatient care settings accounted for 19% and 13% of the total AGE episodes, respectively. Furthermore, RVGE accounted for a greater proportion of the total AGE episodes in the inpatient setting among elderly patients than in adult patients. Previous studies that suggest a decline in the RVGE cases among older children and adults after introduction of the pediatric RVGE vaccination program.19-21 However, we did not observe a decline across the five seasons in the RVGE episode rate among older children, adults, and elderly following the introduction of the pediatric RVGE vaccination. Factors such as differences in study design, assessment of RVGE cases, study population, and period of incidence assessment may account for the observed variation. For example, the study conducted by Lopman and colleagues reported a markedly lower number of RVGE- and cause-unspecified AGE discharges in 2008 than in 2000–2006 among patients aged 0 to 4 y and 5 to 14 y.21 In contrast, our study focused only on the post-vaccination period and did not compare episode rates with the pre-vaccination rates. Similarly, a study conducted by Anderson and colleagues was limited to a hospital in Chicago and compared the rates of laboratory-confirmed pre-vaccination program RVGE cases with post-vaccination program cases.19 Additional research comparing the rate of RVGE among these populations is required to assess whether the pediatric vaccination program offers herd immunity to older children, adults, and the elderly enrolled in employer-sponsored managed care plans.
Another important finding of this study is that among patients with NVGE and RVGE, the duration of illness was 1 d longer than among patients with AGE. This also likely explains the greater per-episode costs observed for most years in physician’s office and ED settings among patients with NVGE and RVGE than among patients with AGE. Finally, our findings indicated that the estimated national AGE-related seasonal costs burden was considerable (range: $3.30 billion–$4.28 billion) and increased by over 29% during the study period. On average, NVGE and RVGE each accounted for 7% of the total estimated national AGE-related seasonal costs.
Several limitations should be considered when interpreting findings from this study. The selection of patients with clinically coded AGE, NVGE, and RVGE was based on International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes, and coding inaccuracies may have led to misidentification of patients as having AGE. Due to infrequent laboratory testing to determine the cause of gastroenteritis, a majority of cases did not have an organism-specific diagnosis code in the database. For example, in the office setting on average less that 5% of the total estimated RVGE-related episodes had a confirmed rotavirus diagnosis code. Similarly, less than 1% of the total estimated NVGE-related episodes in physician office setting had a confirmed norovirus diagnosis code. Depending on the nature of the miscoding, this could have led to either an underestimation or overestimation of case numbers in each study group. In addition, we used an indirect method in estimating the total number of NVGE and RVGE episodes as a function of cause-unspecified AGE episode claims included in the database for each season. Thus, an estimation of the number of NVGE and RVGE episodes from the cause-unspecified pool of AGE was heavily influenced by the accurate coding of other cause-specific AGE and by the overall burden of AGE observed during the season. For example, if an exceptionally large number of cause-unspecified AGE episodes were observed during a season, it could have led to larger residuals resulting in inflated estimates for both NVGE and RVGE episodes.
This study only assessed the direct medical costs associated with these diseases. To assess the overall disease burden, further research estimating the indirect costs related to these diseases is required. Additionally, the direct per-patient and per-episode costs for NVGE and RVGE were based solely on the costs observed from the clinically coded episodes for these infections. In some instances, due to several levels of stratification (by season, by setting, and by age group), the sample of clinically coded episodes was small. In such cases, an outlier with very high or very low costs would likely significantly impact the mean per-episode and per-patient costs.
We included the employed elderly receiving employer-sponsored supplemental health coverage in the study population. For elderly patients with both employer-sponsored supplemental health coverage and Medicare coverage, the claims covered by Medicare are not included in this database of claims covered by the supplemental health plan. Thus, this likely led to underestimation of AGE-, NVGE-, RVGE-related cost and utilization among the elderly included in this study. This study was limited to managed care enrollees; thus, the results may not be generalized to the entire US population or to individuals covered by other federal insurance programs (e.g., Medicaid, Veterans Affairs). Finally, in estimating national costs burden for the three conditions, we used the 2008 Healthcare Cost and Utilization Project Nationwide Inpatient Sample (HCUP NIS) data to assess payer distribution and payer ratios based on AGE-related inpatient visits. However, the payer distribution and payer ratios associated with the AGE-related inpatient visits may not be representative of the payer distribution and payer ratios for NVGE- and RVGE-related visits in inpatient or other care settings or across the five seasons.
Even with these limitations, this is the first study to describe resource utilization and costs among managed care enrollees with AGE, NVGE, or RVGE. NVGE varied considerably across age groups and by seasons; the NVGE episode rate peaked during the 2008–2009 season for the pediatric groups; the rate peaked during the 2006–2007 season for the elderly groups. As anticipated, a steady decline in RVGE and associated costs was observed among patients aged 0 to 4 y. Nevertheless, with an estimated seasonal cost of $278 million, the burden of RVGE in the US remains substantial. Even with a significant reduction in the incidence of RVGE among pediatric populations, AGE due to various causative organisms continues to affect a substantial proportion of managed care enrollees across all age groups.
Overall the RVGE-related costs and utilization estimated presented in this study can be utilized as model inputs for studies assessing cost-effectiveness of rotavirus vaccination programs for individuals enrolled in managed care health plans. Furthermore, the substantial national NVGE-related utilization and costs estimates presented in this study may emphasize the need for norovirus prevention programs. Finally, this methodology can be used for future evaluations of vaccination program effectiveness and cost outcomes.
Materials and Methods
Data Source
For this retrospective analysis, we used the MarketScan Commercial Claims and Encounters (CCAE) database. This database primarily consists of employer-sourced and health plan-sourced data containing medical and outpatient pharmacy utilization claims for over 40 million unique individuals annually. The database also includes pharmacy and medical (i.e., inpatient, outpatient, physician office, and ancillary services) claims for Medicare-eligible retirees with employer-sponsored Medicare supplemental plans. Over 100 large employers and 12 unique health plans in the US contribute to the CCAE database. The data also include enrollment information for each plan enrollee, such as periods of health plan enrollment, demographic details (e.g., age, sex, region), and plan information (e.g., health plan type).
The age distribution for individuals in the CCAE database (i.e., individuals with at least 1 d of health plan enrollment) is well matched with the age distribution of the US census population, except for individuals 18 to 64 y of age (Appendix Table A1). Because the CCAE includes data from employer-sponsored health plans, a greater proportion of working adults of age 18 to 64 y are included in the database.
Table A1. Age Distribution Comparison, MarketScan Commercial Claims and Encounters Database Population Vs. US Census Population.
| MarketScan CCAE Population | US Census Population | ||||
|---|---|---|---|---|---|
| Age Group (Years) | N | % | Age Group (Years) | N | % |
| 2006–2007 | |||||
| 0–4 y | 2 066 317 | 5 | 0–4 y | 20 125 962 | 7 |
| 5–17 y | 7 153 438 | 18 | 5–17 y | 53 893 443 | 18 |
| 18–64 y | 27 645 619 | 70 | 18–64 y | 189 386 091 | 63 |
| 65–74 y | 1 588 857 | 4 | 65–74 y | 19 698 727 | 7 |
| 75–84 y | 920 695 | 2 | 75–84 y | 13 087 439 | 4 |
| 85 y and over | 357 761 | 1 | 85 y and over | 5 039 545 | 2 |
| Total | 39 732 687 | 100 | Total | 301 231 207 | 100 |
| 2007–2008 | |||||
| 0–4 y | 2 912 414 | 5 | 0–4 y | 20 271 127 | 7 |
| 5–17 y | 9 896 988 | 18 | 5–17 y | 53 833 475 | 18 |
| 18–64 y | 38 503 710 | 69 | 18–64 y | 191 211 743 | 63 |
| 65–74 y | 2 516 884 | 5 | 65–74 y | 20 505 679 | 7 |
| 75–84 y | 1 417 012 | 3 | 75–84 y | 13 076 102 | 4 |
| 85 y and over | 590 279 | 1 | 85 y and over | 5 195 840 | 2 |
| Total | 55 837 287 | 100 | Total | 304 093 966 | 100 |
| 2008–2009 | |||||
| 0–4 y | 3 239 265 | 5 | 0–4 y | 20 244 518 | 7 |
| 5–17 y | 11 197 834 | 17 | 5–17 y | 53 889 649 | 18 |
| 18–64 y | 44 271 381 | 69 | 18–64 y | 193 014 187 | 63 |
| 65–74 y | 2 988 926 | 5 | 65–74 y | 21 233 099 | 7 |
| 75–84 y | 1 600 100 | 2 | 75–84 y | 13 022 775 | 4 |
| 85 y and over | 708 559 | 1 | 85 y and over | 5 367 301 | 2 |
| Total | 64 006 065 | 100 | Total | 306 771 529 | 100 |
| 2009–2010 | |||||
| 0–4 y | 3 341 234 | 5 | 0–4 y | 20 200 529 | 7 |
| 5–17 y | 11 600 164 | 18 | 5–17 y | 53 995 231 | 17 |
| 18–64 y | 45 690 947 | 69 | 18–64 y | 194 716 348 | 63 |
| 65–74 y | 3 161 647 | 5 | 65–74 y | 21 841 372 | 7 |
| 75–84 y | 1 620 344 | 2 | 75–84 y | 13 063 453 | 4 |
| 85 y and over | 776 912 | 1 | 85 y and over | 5 532 756 | 2 |
| Total | 66 191 248 | 100 | Total | 309 349 689 | 100 |
| 2010–2011 | |||||
| 0–4 y | 2 911 230 | 5 | 0–4 y | 20 162 058 | 6 |
| 5–17 y | 10 378 656 | 17 | 5–17 y | 53 772 214 | 17 |
| 18–64 y | 41 233 673 | 69 | 18–64 y | 196 263 504 | 63 |
| 65–74 y | 3 238 468 | 5 | 65–74 y | 22 481 738 | 7 |
| 75–84 y | 1 598 152 | 3 | 75–84 y | 13 175 230 | 4 |
| 85 y and over | 771 413 | 1 | 85 y and over | 5 737 173 | 2 |
| Total | 60 131 592 | 100 | Total | 311 591 917 | 100 |
CCAE, Commercial Claims and Encounters; US, United States.
Patient Selection
We initially identified patients in the CCAE database who had at least one medical claim with a diagnosis of AGE (ICD-9-CM codes 001.0–001.9, 002.0–002.9, 003.0–003.1, 003.2x, 003.8–003.9, 004.0–004.9, 005.0–005.4, 005.8x, 005.9, 006.0–006.2, 006.9, 007.0–007.9, 008.0x, 008.1–008.3, 008.4x, 008.5, 008.6x, 008.8, 009.0–009.3, 558.9, and 787.91) during the five seasons considered for this study (July 2006-June 2007, July 2007-June 2008, July 2008-June 2009, July 2009-June 2010, and July 2010-June 2011). Among the selected patients with AGE, we created AGE episodes, defined by the dates of the first and last observed AGE-related medical claims (primary or nonprimary diagnosis), followed by a 14-d clean period (i.e., no AGE-related medical claim).22 An AGE episode with at least one medical claim for NVGE (ICD-9-CM code: 008.63) or RVGE (ICD-9-CM code: 008.61) infection was categorized as a NVGE or RVGE episode, respectively. Using this episode definition, we created three nonexclusive cohorts (i.e., AGE, NVGE, and RVGE) for each season. The selected patients with at least one medically-coded diagnosis code observed during an episode were used in documenting patient characteristics for the three cohorts, according to the setting in which the episode occurred.
Prior studies suggest that diagnosis codes for NVGE and RVGE are infrequently used.6,8 For example, a study conducted by Lopman and colleagues reported that nationally approximately 200 inpatient admissions per year had evidence of norovirus specific code (ICD-9-CM: 008.63), thereby using only specific diagnosis codes to identify claims for NVGE underestimates the burden of this disease.8 Thus, for patients with AGE episodes with cause-unspecified diagnosis codes (ICD-9-CM codes: 009.0–009.3, 558.9, 787.91, 008.8), we took additional steps to attribute the AGE episodes to specific pathogens, using a previously published “indirect” method.6 With this method, cause-unspecified AGE episodes observed in each month were modeled as a function of pathogen-specific (i.e., RVGE, other viral, bacterial, Clostridium difficile, or parasitic) AGE episodes in the same month. Specifically, in order to estimate the number of NVGE episodes during a season, we first developed separate Poisson regression models with an identity link for each month of a season to predict the number of cause-unspecified AGE episodes as a function of observed counts of cause-specified AGE types, excluding NVGE:
| Cause-unspecifiedEcnt = ∝ + β1RVGE0-4Ecnt + β2OtherViralEcnt + β3BacterialEcnt + β4CdifficileEcnt + β5ParasiticEcnt + Timey, | (1) |
with Cause-unspecifiedEcnt, RGVE0–4Ecnt, OtherViralEcnt, BacterialEcnt, CdifficileEcnt, and ParasiticEcnt being counts for cause-unspecified, RVGE (for patients aged 0–4 y because RVGE primarily affects this age group), other viral, bacterial, C. difficile, and parasitic episodes, respectively. Timey was an indicator for the month and year in which the episode started. Using each model’s residuals, calculated as predicted cause-unspecified counts minus actual counts, we estimated the number of NVGE-related episodes for each month as the residual of that month minus the minimum monthly residual of the season. Finally, the monthly estimates of NVGE episodes were summed over each season and added to the medically coded NVGE episode counts for a total seasonal count of NVGE episodes. A similar approach was used for the estimation of seasonal NVGE episode counts; however, in the Poisson model shown (Eqn. 1), β1RVGE was replaced with β1NVGE. These steps were repeated for each care setting (i.e., physician’s office, ED, and inpatient) and for each patient age group (0–4 y, 5–17 y, 18–64 y, 65–74 y, 75 84 y, and ≥ 85 y). Additional details on the methods, regression equations and statistical programs involved these estimations will be provided to researchers upon request.
Patient Characteristics
For the three patient groups (AGE, NVGE, and RVGE), we assessed demographic characteristics, including age (at first diagnosis during a season), sex, and region (Northeast, North Central, South, West, unknown). Assessment of patient characteristics was based only on data for medically coded episodes of AGE, NVGE, and RVGE.
AGE-, NVGE-, and RVGE-Related Episode Rates
For each study cohort for each disease, we separately assessed the episode rates for physician’s office, ED, and inpatient care settings. The total number of disease-related episodes during a season, by care setting, served as the numerator in calculating episode rates. In addition to patients with AGE, the number of individuals enrolled in the health plan for at least 3 mo during a season served as the denominator in the episode rate calculation. The disease-related episode rate was calculated as follows.
Duration of Illness
Using medical claims in the CCAE database, we assessed the overall duration of illness (across all care settings) per AGE, NVGE, and RVGE episode. In calculating the duration of an AGE episode, the first observed AGE diagnosis claim (irrespective of causative pathogen) defined the episode start date, and a subsequent AGE claim followed by a 14-d clean period (i.e., no AGE-related medical claim) defined the episode end date.22 Episode state and end dates for NVGE and RVGE were calculated in a similar manner, where an AGE episode with at least one medically coded diagnosis claim for NVGE or RVGE was categorized as an NVGE or RVGE episode, respectively.
AGE-, NVGE-, and RVGE-Related Costs
For each study group and for each season, we reported the following cost measures for the three care settings of interest (physician’s office, ED, and inpatient). All cost data were adjusted to 2011 US dollars, using appropriate inflators from the medical care component of the US Consumer Price Index.
Care setting-specific per-episode and per-patient costs
For each patient, costs recorded on claims during an episode with medically coded AGE, NVGE, and/or RVGE were aggregated to provide disease-related episode costs for each setting. For each patient, the setting-specific costs incurred during a season for a disease (e.g., NVGE) were summed to obtain the total disease-specific per-patient seasonal costs. These costs were further divided by the total number of pathogen-coded (e.g., NVGE) episodes incurred during a season to obtain total setting- and disease-specific per-episode seasonal costs.
Care setting-specific total disease-related costs per 100 000 plan members
For each season, the estimated total number of disease-related episodes per 100 000 plan members was multiplied by the median disease-related per-episode costs of medically coded episodes. Because only a small number of patients had a medically coded diagnosis claim for an NVGE or RVGE, the mean disease-related per episode costs were skewed. Thus, we used median disease-related per-episode costs in calculating the care setting-specific total disease-related episode costs per 100 000 plan members.
National AGE-, NVGE-, and RVGE-Related Estimated Costs
For each season, the national episode rate estimates for the three diseases were calculated by multiplying the overall seasonal episode rate with the US population estimate obtained from the US Census Bureau for that season. Further, we obtained payer distributions for AGE-related visits in an inpatient setting using the HCUP NIS. The payer distributions from the HCUP NIS database were applied to the estimated national episode rates to obtain payer-specific national episode rate estimates for the three conditions for each season. We also assessed median costs by payer type (i.e., private, Medicare, Medicaid, other, self-pay, unknown) for AGE-related visits using the HCUP NIS data. The median AGE-related visit cost by private payer type from the HCUP NIS data was considered the base and payer ratios for all other payer types were generated from the base value. For example, the median AGE-related visit costs for private and Medicaid payer types were US $5858 and $5223, respectively. Thus, the median AGE-related visit costs for Medicaid payer type were 90% (payer ratio = $5223 ÷ $5858) of the private payer type. For each season, we applied these ratios to the care setting-specific median episode costs obtained from the CCAE database to estimate the payer and care setting-specific per-episode costs for the three diseases. For each season, these payer-specific per-episode costs then were multiplied by the payer-specific national episode rate estimates to obtain the national AGE-, NVGE-, and RVGE-related costs estimates. Appendix Table A2 includes details on the payer distribution for AGE-related visits and median AGE-related visit costs and payer ratios derived from the HCUP NIS database.
Table A2. Payer Distribution for Acute Gastroenteritis-Related Visits and Median Acute Gastroenteritis-Related Visit Costs and Payer Ratios, Based on the Healthcare Cost and Utilization Project Nationwide Inpatient Sample Database.
| Payer Type | Acute Gastroenteritis-Related Visit | Payer Distribution | Acute Gastroenteritis-Related Visit Costs, Median | Payer Ratios |
|---|---|---|---|---|
| Private | 405 853 | 28.0% | $5858 | 1.00 |
| Medicare | 727 281 | 50.2% | $8048 | 1.37 |
| Medicaid | 205 928 | 14.2% | $5223 | 0.89 |
| Other | 39 891 | 2.8% | $6184 | 1.06 |
| Self-pay | 60 139 | 4.2% | $5525 | 0.94 |
| Unknown | 9066 | 0.6% | $6205 | 1.06 |
Statistical Analyses
All outlined study measures were analyzed descriptively. Descriptive analyses entailed tabular display of frequency distributions for categorical outcome measures and mean, standard deviation, and median continuous outcome measures of interest. Using Pearson correlation coefficients and corresponding P values, we assessed whether or not seasonal trends in AGE episode rates per 100 000 plan members were correlated with NVGE and RVGE episode rates. All analyses were conducted using SAS® version 9.3 (Cary, North Carolina).
Acknowledgments
GlaxoSmithKline Biologicals SA funded this study and was involved in all stages of study conduct, including analysis of the data. GlaxoSmithKline Biologicals SA also covered all costs associated with the development and publication of this manuscript.
The authors thank Heather Santiago (publication manager, GlaxoSmithKline Vaccines) for editorial assistance and manuscript coordination.
Sudeep Karve, Jennifer Korsnes, and Sean D. Candrilli are employees of RTI Health Solutions, an independent contract research organization that has received research funding from GlaxoSmithKline SA for this study. Girishanthy Krishnarajah and Adrian Cassidy are employees of GlaxoSmithKline group of companies, and Girishanthy Krishnarajah has restricted shares in GlaxoSmithKline group of companies. The publication of this study’s results is not contingent upon the sponsor’s approval or censorship of the manuscript. Sudeep Karve, Girishanthy Krishnarajah, Adrian Cassidy, and Sean Candrilli were the primary developers of the study design. As principal investigator, Sudeep Karve had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Jennifer Korsnes, with support from Sudeep Karve and Sean Candrilli, led all statistical analyses. Sudeep Karve served as the primary writer in drafting the manuscript text and in interpreting the findings. Girishanthy Krishnarajah, Adrian Cassidy, Jennifer Korsnes, and Sean Candrilli assisted in interpreting the study findings and drafting the manuscript text. They also served as the primary reviewers of the manuscript text. All authors were responsible for approving the manuscript and its contents. All authors are accountable for all aspects of the work and are responsible for ensuring that questions related to the accuracy or integrity of any part of this work will be appropriately investigated and resolved.
Glossary
Abbreviations:
- AGE
acute gastroenteritis
- CCAE
Commercial Claims and Encounters
- ED
emergency department
- HCUP NIS
Healthcare Cost and Utilization Project Nationwide Inpatient Sample
- ICD-9-CM
International Classification of Diseases, Ninth Revision, Clinical Modification
- NVGE
norovirus gastroenteritis
- RVGE
rotavirus gastroenteritis
- US
United States
References
- 1.Hall AJ, Rosenthal M, Gregoricus N, Greene SA, Ferguson J, Henao OL, Vinjé J, Lopman BA, Parashar UD, Widdowson M-A. Incidence of acute gastroenteritis and role of norovirus, Georgia, USA, 2004-2005. Emerg Infect Dis. 2011;17:1381–8. doi: 10.3201/eid1708.101533. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Jones TF, McMillian MB, Scallan E, Frenzen PD, Cronquist AB, Thomas S, Angulo FJ. A population-based estimate of the substantial burden of diarrhoeal disease in the United States; FoodNet, 1996-2003. Epidemiol Infect. 2007;135:293–301. doi: 10.1017/S0950268806006765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Centers for Disease Control and Prevention. burden of norovirus illness and outbreaks [Internet]. Atlanta (GA): Centers for Disease Control and Prevention; c2012 [updated 2013 September 27; cited 2013 Jun 18]; [about 2 screens]. Available from: http://www.cdc.gov/norovirus/php/illness-outbreaks.html
- 4.Dennehy PH, Vesikari T, Matson DO, Itzler RF, Dallas MJ, Goveia MG, DiNubile MJ, Heaton PM, Ciarlet M. Efficacy of the pentavalent rotavirus vaccine, RotaTeq® (RV5), between doses of a 3-dose series and with less than 3 doses (incomplete regimen) Hum Vaccin. 2011;7:563–8. doi: 10.4161/hv.7.5.15406. [DOI] [PubMed] [Google Scholar]
- 5.O’Ryan M, Linhares AC. Update on Rotarix: an oral human rotavirus vaccine. Expert Rev Vaccines. 2009;8:1627–41. doi: 10.1586/erv.09.136. [DOI] [PubMed] [Google Scholar]
- 6.Cortes JE, Curns AT, Tate JE, Cortese MM, Patel MM, Zhou F, Parashar UD. Rotavirus vaccine and health care utilization for diarrhea in U.S. children. N Engl J Med. 2011;365:1108–17. doi: 10.1056/NEJMoa1000446. [DOI] [PubMed] [Google Scholar]
- 7.Willyard C. First vaccines targeting ‘cruise ship virus’ sail into clinical trials. Nat Med. 2013;19:1076–7. doi: 10.1038/nm0913-1076. [DOI] [PubMed] [Google Scholar]
- 8.Lopman BA, Hall AJ, Curns AT, Parashar UD. Increasing rates of gastroenteritis hospital discharges in US adults and the contribution of norovirus, 1996-2007. Clin Infect Dis. 2011;52:466–74. doi: 10.1093/cid/ciq163. [DOI] [PubMed] [Google Scholar]
- 9.Cortese MM, Parashar UD, Centers for Disease Control and Prevention (CDC) Prevention of rotavirus gastroenteritis among infants and children: recommendations of the Advisory Committee on Immunization Practices (ACIP) MMWR Recomm Rep. 2009;58(RR-2):1–25. [PubMed] [Google Scholar]
- 10.Centers for Disease Control and Prevention (CDC) Reduction in rotavirus after vaccine introduction--United States, 2000-2009. MMWR Morb Mortal Wkly Rep. 2009;58:1146–9. [PubMed] [Google Scholar]
- 11.Payne DC, Staat MA, Edwards KM, Szilagyi PG, Weinberg GA, Hall CB, Chappell J, Curns AT, Wikswo M, Tate JE, et al. New Vaccine Surveillance Network (NVSN) Direct and indirect effects of rotavirus vaccination upon childhood hospitalizations in 3 US Counties, 2006-2009. Clin Infect Dis. 2011;53:245–53. doi: 10.1093/cid/cir307. [DOI] [PubMed] [Google Scholar]
- 12.Desai R, Curns AT, Steiner CA, Tate JE, Patel MM, Parashar UD. All-cause gastroenteritis and rotavirus-coded hospitalizations among US children, 2000-2009. Clin Infect Dis. 2012;55:e28–34. doi: 10.1093/cid/cis443. [DOI] [PubMed] [Google Scholar]
- 13.Centers for Disease Control and Prevention (CDC) Norovirus activity--United States, 2006-2007. MMWR Morb Mortal Wkly Rep. 2007;56:842–6. [PubMed] [Google Scholar]
- 14.Yen C, Wikswo ME, Lopman BA, Vinje J, Parashar UD, Hall AJ. Impact of an emergent norovirus variant in 2009 on norovirus outbreak activity in the United States. Clin Infect Dis. 2011;53:568–71. doi: 10.1093/cid/cir478. [DOI] [PubMed] [Google Scholar]
- 15.Payne DC, Vinjé J, Szilagyi PG, Edwards KM, Staat MA, Weinberg GA, Hall CB, Chappell J, Bernstein DI, Curns AT, et al. Norovirus and medically attended gastroenteritis in U.S. children. N Engl J Med. 2013;368:1121–30. doi: 10.1056/NEJMsa1206589. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zlamy M, Kofler S, Orth D, Würzner R, Heinz-Erian P, Streng A, Prelog M. The impact of Rotavirus mass vaccination on hospitalization rates, nosocomial Rotavirus gastroenteritis and secondary blood stream infections. BMC Infect Dis. 2013;13:1–10. doi: 10.1186/1471-2334-13-112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Braeckman T, Van Herck K, Meyer N, Pirçon JY, Soriano-Gabarró M, Heylen E, Zeller M, Azou M, Capiau H, De Koster J, et al. RotaBel Study Group Effectiveness of rotavirus vaccination in prevention of hospital admissions for rotavirus gastroenteritis among young children in Belgium: case-control study. BMJ. 2012;345:e4752. doi: 10.1136/bmj.e4752. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Dudareva-Vizule S, Koch J, An der Heiden M, Oberle D, Keller-Stanislawski B, Wichmann O. Impact of rotavirus vaccination in regions with low and moderate vaccine uptake in Germany. Hum Vaccin Immunother. 2012;8:1407–15. doi: 10.4161/hv.21593. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Anderson EJ, Shippee DB, Weinrobe MH, Davila MD, Katz BZ, Reddy S, Cuyugan MG, Lee SY, Simons YM, Yogev R, et al. Indirect protection of adults from rotavirus by pediatric rotavirus vaccination. Clin Infect Dis. 2013;56:755–60. doi: 10.1093/cid/cis1010. [DOI] [PubMed] [Google Scholar]
- 20.Dey A, Wang H, Menzies R, Macartney K. Changes in hospitalisations for acute gastroenteritis in Australia after the national rotavirus vaccination program. Med J Aust. 2012;197:453–7. doi: 10.5694/mja12.10062. [DOI] [PubMed] [Google Scholar]
- 21.Lopman BA, Curns AT, Yen C, Parashar UD. Infant rotavirus vaccination may provide indirect protection to older children and adults in the United States. J Infect Dis. 2011;204:980–6. doi: 10.1093/infdis/jir492. [DOI] [PubMed] [Google Scholar]
- 22.Wang FT, Mast TC, Glass RJ, Loughlin J, Seeger JD. Effectiveness of the pentavalent rotavirus vaccine in preventing gastroenteritis in the United States. Pediatrics. 2010;125:e208–13. doi: 10.1542/peds.2009-1246. [DOI] [PubMed] [Google Scholar]