Hospitalizations and outcomes for diabetic gastroparesis in North Carolina.

ABSTRACT

Background. Population-based data to assess the extent of gastroparesis in the diabetic population are scarce. We examined the demographic and clinical characteristics associated with hospital admissions for diabetic gastroparesis in North Carolina.

Methods. Data from the 1998 North Carolina Hospital Discharge database were abstracted from records in which gastroparesis and diabetes mellitus were listed as simultaneous diagnoses.

Results. There were 1476 discharges meeting our criteria, with total charges of $11,378,446 over 7850 total hospital days. Most patients were female (65.8%), [greater than or equal to] 45 years of age (54.5%), and had Medicare as the primary payer (52.1 %). While most of these patients were admitted under emergency or urgent circumstances, the vast majority had routine discharges.

Conclusion. Despite some limitations, these data indicate that diabetic gastroparesis is not uncommon, but can be treated effectively.

**********

NEUROPATHY is a common sequela of diabetes mellitus. Neuropathy occurring in the gastrointestinal system in people with diabetes is known as diabetic gastroparesis; it is believed to develop in about 20% of persons with type 1 diabetes, and less often in persons with type 2 diabetes. (1) About 50% of patients with type 1 or type 2 diabetes experience some delayed gastric emptying due to gastric neuropathy. (2-4) Population data to describe the impact of diabetic gastroparesis are scarce, however. (5) Diabetic gastroparesis can greatly impact glycemic control by delaying gastric emptying, which disrupts the timing of insulin or oral hypoglycemic agents with nutrient intake. (6,7) The purpose of this report is to indicate the degree to which persons with diabetes in North Carolina are hospitalized with diabetic gastroparesis, and the demographic and clinical characteristics associated with these admissions.

METHODS

Data from the North Carolina Hospital Discharge Database of the State Center for Health Statistics were analyzed to ascertain the frequency, charges, and length of stay for diabetic gastroparesis. Data for all discharges from nonfederal North Carolina hospitals for calendar year 1998 were abstracted. Data from these discharge records are based on the UB-92 medical billing form, which has a maximum of 9 diagnostic fields (1 primary and 8 contributing diagnoses). All hospital discharges with a diagnosis of gastroparesis (International Classification of Diseases, Ninth Revision [ICD-9]; code 536.3) as the principal or secondary diagnosis were abstracted, including those with a simultaneous diagnosis of diabetes mellitus (ICD-9 code 250). Data for those patients with both discharge diagnoses were stratified according to sex, age, payer source, admission source and type, and discharge status.

RESULTS

A total of 155 discharges of patients with gastroparesis as the principal diagnosis, and 3,639 discharges of patients with gastroparesis as a secondary diagnosis were identified (Table) . Forty-five (29%) of those patients discharged with a principal diagnosis of gastroparesis had diabetes as a secondary diagnosis, while 1,431 (39%) of those discharged with gastroparesis as a secondary diagnosis had diabetes listed as the principal diagnosis. Patients with discharge diagnoses of both diabetes and gastroparesis (n = 1,476) were hospitalized a sum total of 7,850 days, with an average stay of 5.3 days per admission; they accumulated a total of $11,378,386 in hospital charges, at an average cost of $7,709 per hospital visit.

About two thirds of all patients with discharge diagnoses of diabetic gastroparesis were female (65.8%), and slightly over half were 45 years of age (54.5%). Fifty-two percent of these hospital charges were paid by Medicare and 16.9% were paid by Medicaid. While the majority of patients were admitted as emergency (56.2%) or urgent cases (38.6%), and over half were admitted directly from the emergency room (56.9%), the vast majority (83.4%) had routine discharges. Nine patients died in the hospital (0.6%), and 8 left against medical advice (0.5%). About 15% were transferred to another facility.

DISCUSSION

Diabetes mellitus is a growing problem in many societies. Approximately 16 million Americans have diabetes, or about 5.9% of the total population of the United States. (8) According to Behavioral Risk Factor Surveillance System data, about 360,000 adults in North Carolina had a diagnosis of diabetes in 1998 (personal communication, State .Center for Health Statistics, Raleigh, NC) . The growing prevalence of diabetes reflects an increase in the number of persons being diagnosed with diabetes, as well as an increase in survival time of those already diagnosed with diabetes. The increase in the prevalence of diabetes is asscciated with an increase in the number of cases of diabetic complications, including diabetic neuropathy.

Diabetic gastroparesis, resulting from neuropathy of the vagus and other nerves that control the musculature of the gastrointestinal system, is a recognized problem in type 1 diabetes, and a growing problem in type 2 diabetes. While population-based data are scarce for diabetic gastroparesis, clinical studies have indicated that about 20% of patients with type 1 diabetes will develop diabetic gastroparesis, and about half of patients with type 1 or type 2 diabetes will experience some delay in gastric emptying. (1-4) Data from the 1987-1991 National Hospital Discharge Surveys indicated that adult patients aged [greater than or equal to]45 years with a discharge diagnosis of a stomach function disorder (ICD-9 code 536) were more than twice as likely to have a simultaneous diagnosis of diabetes mellitus than to not have the diabetes diagnosis. (1)

Data from the North Carolina Discharge Database for 1998 indicated a total of 1,476 discharges in which the diagnoses of diabetes mellitus and gastroparesis were listed simultaneously. These represent about 39% of all discharges for gastroparesis, and about 1% of all discharges for diabetes mellitus (n = 134,540). Current data on the occurrence of diabetic gastroparesis relative to data from prior studies may be inflated due to the recently developed advanced techniques for diagnosing gastroparesis, such as electrogastropathy, gastric scintigraphy, breath testing for carbon 13, and manometry. (9) These data may also reflect an increasing life expectancy of patients with diabetes as a result of earlier diagnosis of diabetes, and improved treatment modalities for diabetes and cardiovascular disease and its risk factors. Since diabetic gastroparesis is much more common in long-standing diabetes, the increased frequency of diabetic gastroparesis may be indicative of better diagnostic procedures, as well as increa sed susceptibility to conditions related to increased duration of diabetes.

These data also indicate that, while many of these patients were admitted to the hospital under dire circumstances, most were discharged in stable condition; over 83% of patients had routine discharges. This high success rate may be reflective of improved detection modalities, as well as improved nutritional, pharmacologic, and surgical interventions for diabetic gastroparesis.

These data are hampered by at least 2 limitations. First, they are reflective only of hospital discharges, and only reflect charges over 1 calendar year. These data do not separate multiple discharges for the same individual, and do not follow individuals longitudinally. Second, these data may be biased due to 1 of 2 sources: 1) misclassification of diabetes on the hospital discharge, and 2) possible discriminant capabilities of hospitals across the state to diagnose gastroparesis. The first of these issues may lead to an underreporting of the severity of the problem, since it is likely that misclassification of diabetes would lean heavily in favor of diabetes not being recorded on the discharge summary. The second of these issues is difficult to ascertain, and may be remedied by patients traveling to hospitals with capabilities for diagnosing and treating this condition to receive care for gastroparesis. Nonetheless, these data indicate that diabetic gastroparesis is not an uncommon complication of diabetes, and that most hospitalizations for diabetic gastroparesis have satisfactory outcomes.

TABLE

Hospital Discharge Data for patients With Diabetic Gastroparesis in
North Carolina in 1998

                             Principal   Secondary       Total
                             Diagnosis   Diagnosis     Discharges
Characteristic               (n = 45)    (n = 1,43)    (N= 1,476)

Length of Stay (Total days)     229        7,621         7,850
Costs (Total $)               364,851    11,013,595    11,378,446

                              No.(%)       No.(%)        No.(%)
Sex
  Man                        16 (35.6)    489 (34.2)    505 (34.2)
  Women                      29 (64.4)    942 (65.8)    971 (65.8)
Age Group
  < 45 Years                 12 (26.7)    659 (46.1)    671 (45.5)
  [greater than or           33 (73.3)    772 (53.9)    805 (54.5)
    equal to] 45 Yea
Payer Source
  Medicare                   23 (51.1)    746 (52.2)    769 (52:1)
  HMO/Private                12 (26.6)    389 (27.2)    401 (27.2)
  Medicaid                    9 (20.0)    241 (16.9)    250 (16.9)
  Self-Pay, Indigent          1  (2.2)     49  (3.4)     50  (3.4)
  Other                       0             6  (0.4)      6  (0.4)
Admission Source
  Emergency room             24 (53.3)    816 (57.0)    840 (56.9)
  Referral                   20 (44.4)    582 (40.7)    602 (40.8)
  Transfer                    1  (2.2)     33  (2.3)     34  (2.3)
Admission Type
  Emergency                  22 (48.9)    807 (56.4)    829 (56.2)
  Urgent                     23 (51.1)    547 (38.2)    570 (38.6)
  Elective                    0            77  (5.4)     77  (5.2)
Discharge Status
  Routine                    39 (86.7)  1,192 (83.3)  1,231 (83.4)
  Transfer                    5 (11.1)    223 (15.6)    228 (15.4)
  Left against advice         1  (2.2)      7  (0.5)      8  (0.5)
  Death                       0             9  (0.6)      9  (0.6)

References

(1.) National Institute of Diabetes and Digestive and Kidney Diseases: Gastroparesis and Diabetes. National Institutes of Health, US Department of Health and Human Services, NIH Publication 99-4348, 1999

(2.) Lyrenas E, Olsson E, Arvidsson U, et al: Prevalence and determinants of solid and liquid gastric emptying in unstable type 1 diabetes: relationship to postprandial blood glucose concentrations. Diabetes Care 1997;20:413-418

(3.) Horowitz M, Harding PE, Maddox AF, et al: Gastric and oesophageal emptying in patients with type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia 1989;32:151-159

(4.) Mearin F, Malagelada JR: Gastroparesis and dyspepsia in patients with diabetes mellitus. Eur J Gastroenterol Hepatol 1995;7:717-723

(5.) Kong M-F, Horowitz M, Jones KL, et al: Natural history of diabetic gastroparesis. Diabetes Care 1999;22:503-507

(6.) Enck P, Frieling T: Pathophysiology of diabetic gastroparesis. Diabetes 1997;46 (suppl 2):S77-S81

(7.) Horowitz M, Wishart JM, Jones KL, et al: Gastric emptying in diabetes: an overview. Diabet Med 1996; 13(suppl 5):S16-S22

(8.) Centers for Disease Control and Prevention: National Diabetes Fact Sheet: National Estimates and General Information on Diabetes in the United States. Atlanta, US Department of Health and Human Services, Centers for Disease Control and Prevention, 1997

(9.) O'Leary C, Quigley EM: Evaluation of upper-gastrointestinal symptoms in the diabetic patient. Pract Diabetol 2000;19:7-14

RELATED ARTICLE: KEY POINTS

* Approximately 1,500 hospital discharges each year in North Carolina fit the criteria for diabetic gastroparesis, representing about $11 million in charges and 7,800 hospital days.

* Patients hospitalized for diabetic gastroparesis in North Carolina are mostly female, 45 years of age, and have Medicare as the primary payer.

* In general, outcomes for hospitalizations for diabetic gastroparesis are favorable and the discharges are routine.

From the Department of Public Health Sciences, Section on Epidemiology, and the Department of Family and Community Medicine, wake Forest University School of Medicine; and the State Center for Health Statistics, Division of Public Health, North Carolina Department of Health and Human Services, Winston-Salem, NC.

Presented in poster format at the Centers for Disease Control and Prevention Diabetes Translation Conference 2000, New Orleans, La, April 17-20, 2000.

Reprint requests to Ronny A. Bell, PhD, MS. Department of Public Health Sciences, wake Forest University School of Medicine, Medical Center Blvd, Winston-Salem, NC 27157-1063.