Role of exercise for type 2 diabetic patient management.

ABSTRACT

Background. Exercise is integral to the management of type 2 diabetes. Unfortunately, the majority of adults with type 2 diabetes do not engage in regular exercise.

Methods. Three hundred patients with type 2 diabetes were randomly selected from a patient pool of diabetic patients encountered in 1996 at the Department of Veterans Affairs Medical Center in New Orleans, Louisiana. Medical records from October 1996 to June 1999 were reviewed. Information about exercise, alcohol intake, smoking, medications, laboratory results, and other variables was extracted from medical records. Patients with mean glycosylated hemoglobin ([HbA.sub.1c]) <8.0 (good diabetic control) were compared with those who had poor diabetic control (mean ([HbA.sub.1c]) [greater than or equal to]8.0). The effect of exercise in the management of type 2 diabetes was assessed.

Results. After adjustment for other variables, patients without exercise had an odds ratio of 2.71 (95% CI, 1.38-5.32) for poor diabetic control compared with patients with exercise.

Conclusions. These findings suggest that exercise by itself is important for type 2 diabetes management.

TYPE 2 DIABETES is a major health problem in the United States, affecting an estimated 16 million people, costing more than $90 billion per year and accounting for about one sixth of all expenditures for health care. (1) Ninety percent of patients with diabetes have type 2 diabetes (formerly known as non-insulin-dependent diabetes). The prevalence of type 2 diabetes has increased over the past two decades in the United States. (1) The mortality rate among patients with diabetes is 11 times higher than that among people without type 2 diabetes.

Life-style interventions that combine counseling for dietary management, weight reduction, and increased physical activity are pivotal in type 2 diabetes management with respect to glycemic control. (2) Adequate physical activity relative to food (calories) intake prevents obesity, which is a major modifiable risk factor for type 2 diabetes. In addition, physical activity may independently enhance insulin sensitivity and glucose tolerance. (3-7) The possible benefits of exercise for patients with type 2 diabetes are substantial; data from recent studies support the importance of long-term exercise programs for the treatment and prevention of type 2 diabetes and its complications.

Evidence has accumulated suggesting that the progressive decrease with aging in fitness, muscle mass, and strength is in part preventable by maintaining regular exercise. The decrease in insulin sensitivity with aging is also partly due to a lack of physical activity. Lower levels of physical activity are especially likely in the population with type 2 diabetes. The benefit of exercise in improving the metabolic abnormalities of type 2 diabetes is probably greatest when it is used early in its progression from insulin resistance to impaired glucose tolerance to overt hyperglycemia requiring treatment with oral glucose-lowering agents and finally to insulin dependence. (8) Unfortunately, it is estimated that 60% to 80% of the adult population in the United States do not meet the recommended levels of physical activity, (9) and data from a nationally representative survey study have shown that adults with diabetes are even less likely to engage in regular physical activity than adults in general. (10)

With the publication of new clinical reviews, it is becoming increasingly clear that exercise may be a therapeutic tool in a variety of patients with or at risk for diabetes; however, like any therapy, its effects are not clearly understood. (10-12) This study was conducted in patients with type 2 diabetes at the Department of Veterans Affairs Medical Center in New Orleans (VAMCNO). The aim of the study was to evaluate the effects of exercise in management of patients with type 2 diabetes.

METHODS

Subjects

This study was conducted using data from a quality evaluation of diabetic care at VAMCNO. In this study, 300 type 2 diabetic patients (ICD-9 codes 250.00-250.93) were randomly selected from a patient pool of 1,632 diabetic patients encountered at VAMCNO in 1996. Data were extracted from medical records, including hard copies, Veterans Health Information System and Technology Architecture, and Computerized Patient Records System. Medical records from October 1996 to June 1999 were reviewed.

Study Design

Veterans with type 2 diabetes who received care at the VAMCNO were eligible for the study. The study was designed as a case-control study; eligible subjects were classified according to mean glycosylated hemoglobin ([HbA.sub.1c]) level to assess diabetic control. Ninety-two patients identified as having poor diabetic control ([HbA.sub.1c] [greater than or equal to]8) were classified as cases, and 176 patients identified as having good diabetic control ([HbA.sub.1c] <8) were classified as controls. The groups were compared with respect to exercise, medicine, diet, and other important variables. Thirty-two patients were excluded from the analysis because of the lack of [HbA.sub.1c] value. Results of [X.sup.2] tests indicate no statistical difference in patient characteristics between patients with missing [HbA.sub.1c] values and patients with recorded [HbA.sub.1c] values.

Assessment of Physical Activity

A routine preventive medicine questionnaire was given to each patient visiting VAMCNO. Patients were asked if they performed regular exercise, though regular exercise was not defined on the questionnaire. Most patients reported walking 30 to 60 minutes daily.

Data Extraction 

Age, race, body mass index (BMI), sex, medication, clinical laboratory results, smoking history, diet, and other information were extracted from patients' medical records. Crosscheck of different medical records was also done to ensure accurate data collection. One physician reviewed the medical records of all 300 patients. Since the study was designed to assess diabetic quality care at VAMCNO, the reviewer was blinded to the participants' case-control status.

Data Analysis

All analyses were completed using SAS for Windows. (13) Descriptive statistics on sociodemographics, health status, reported exercise, medications, and preventive diabetic care is presented for all study subjects. The relationship between mean [HbA.sub.1c] level and age, sex, VA eligibility, and race was analyzed using [X.sup.2] analyses. Logistic regression analysis was done to assess the relationship between mean [HbA.sub.1c] level and exercise after controlling for other possible confounding variables. Odds ratios for experiencing poor diabetic control were derived with logistic regression models, which included age, race, BMI, medication, alcohol consumption, exercise, diet, smoking, diabetic education, and other variables. Exercise was classified as either regular exercise or no regular exercise, with regular exercise as the baseline reference.

RESULTS

Selected Characteristics of Study Subjects

Of the 268 patients chosen for the analysis, 229 (89.2%) were aged 50 or older, and 179 (67%) were 60 or older. One hundred fifty-eight patients (59%) were black, 104 (38.8%) were white, and 6 (2.2%) were another race. Eighty-eight patients (35.2%) had normal weight (BMI <28), and 162 (64.8%) were overweight (BMI [greater than or equity to]28). Sixty-five patients (24.7%) currently drank alcohol, and 161 (61.0%) did not drink. One hundred sixteen patients (43.3%) exercised regularly, and 114 (42.5%) did not exercise regularly. Sixty-six patients (24.7%) were current smokers, and 163 (61.0%) were not current smokers. One hundred seventy-six patients (65.7%) had good diabetic control (mean [HbA.sub.1c] <8.0); 92 (34.3%) had poor diabetic control ([HbA.sub.1c][greater than or equal to]8) (Table 1).

Comparison of Patient Characteristics

Chi-square statistics were calculated to assess differences in patient characteristics between cases and controls. Among controls, 110 patients (62.9%) were black, 63 (37.1%) were white, and 2 (1%) were another race ([X.sup.2] = 2.3, P= .31). Among cases, 48 (53.3%) were black, 41 (45.6%) were white, and 1 (1.1%) was another race. More young people were in the case group than in the control group. Among controls, 10 patients (5.6%) were aged 20 to 49, 31(17.6%) were 50 to 59, 48 (27.3%) were 60 to 69, and 87 (49.3%) were 70 or older. Among cases, 19 (20.6%) were 20 to 49, 29 (31.5%) were 50 to 59, 25 (27.2%) were 60 to 69, and 19 (20.6%) were 70 or older ([X.sub.2]) = 30.4, P=.001. More overweight patients were in cases than in controls, with 61 overweight patients (72.6%) (BMI 28) in the case group and 101 (60.8%) in the control group ([X.sub.2] = 3.4, P = .07). Thirty-one patients (38.8%) in the case group and 34 (23.3%) in the control group consumed alcohol ([X.sub.2] = 6.0, P = .014). There was no statist ical difference in cigarette smoking between the two groups, with 22 patients (27.2%) in the control group and 44 (29.7%) in the case group being smokers ([X.sub.2] = 0.2, P=.68). There was also no statistical difference in ADA diet between the two groups; in the control group, 45 (48.9%) were on the ADA diet and in the case group, 92 (52.6%) were on that diet ([X.sub.2] = 0.3, P=.57). For diabetic medication, patients were more likely to be taking insulin and metformin. Forty-seven patients (65.2%) in the case group and 73 (41.5%) in the control group were taking insulin; 42 patients (45.7%) in the case group and 54 (30.7%) in the control group were taking metformin (Table 2).

Results of Multiple Logistic Regression Models with Diabetic Control as Dependent Variable

Poor diabetic control ([HbA.sub.lc] [greater than or equal to] 8.0) was the outcome dependent variable for the analysis. Independent variables were tested univariately in a logistic model, controlling for age and race. To assess the relationship of individual independent variables while controlling for the effects of other variables, a series of analyses were undertaken using multiple logistic regression methods. The two controlling variables in the final logistic regression model are age and race. The eight independent variables in the final model are alcohol consumption, BMI, exercise, diet, insulin use, sulfonylurea use, metformin use, acarbose use, current smoking status, and history of having ever smoked.

Table 3 illustrates that after adjustment for age, race, smoking, BMI, diet, and various diabetic medications, patients without regular exercise were 2.71 times more likely than patients with regular exercise to have poor diabetic control (P = .004). Patients who were using alcohol were 2.10 times more likely than patients who were not using alcohol to have poor diabetic control (P = .044). Patients with poor diabetic control were 6.31 times more likely to be taking insulin (P = .001), 2.29 times more likely to be on sulfonylurea (P .09), and 2.35 times more likely to be taking metformin (P = .02) than patients with good diabetic control.

DISCUSSION

In this report, the association of exercise with diabetic control among diabetic patients proved to be independent of age, BMI, race, smoking, alcohol intake, diet, and diabetic medication. This result is consistent with the theory that exercise at least partly contributes to glucose control. Lower adiposity improves insulin sensitivity and glucose tolerance and reduces free fatty acid levels. (3-7,14) Other possible mechanisms by which physical activity may improve insulin sensitivity and glucose tolerance include increased skeletal muscle mass, increased muscle blood flow, greater insulin receptor density, increased glucose transporter protein levels, enhanced skeletal muscle glucose disposal, and improved muscle fiber type and capillary density. (14)

Many studies have been conducted to assess the possible causal relationship between lack of exercise (physical inactivity) and type 2 diabetes. (15-24) Relative risk for physically inactive versus active participants typically have ranged from 1.2 to 2.5, though some studies found that the association was statistically insignificant in certain race groups and between sexes. (18,20) Some studies (15,16,19,21,22) have reported a dose-response relationship between the amount of physical activity and the degree of reduction in diabetes risk, while others have not. (17,24) Consensus panels have concluded that regular vigorous physical activity affords the greatest health benefits, but even moderate physical activity may offer significant benefits. (25,26)

The possible benefits of exercise for the patient with diabetes are substantial, and recent studies strengthen the importance of long-term exercise programs for the treatment and prevention of this common metabolic abnormality and its complications. (27) Several long-term studies have shown a consistent beneficial effect of regular exercise training on carbohydrate metabolism and insulin sensitivity, which can be maintained for at least 5 years. These studies used exercise regimens at intensity of 50% to 80% of maximum oxygen utilization ([VO.sub.2max]) three to four times a week for 30 to 60 minutes each session. Improvements in [HbA.sub.1c] were generally 10% to 20% of baseline and were most marked in patients with mild type 2 diabetes and in those who are likely to be the most insulin-resistant. Our study results are consistent with those of previous studies. Most of previous studies are limited by inadequate randomization and controls and confounding factors such as life-style. This study adjusted for oth er possible confounding factors, and after adjusting for these factors, our results indicate that patients without regular exercise are almost three times more likely to have poor diabetic control.

Study participants were randomly selected from a pool of 1,632 diabetic patients encountered in 1996 at VAMCNO. The study population is representative of the veteran diabetic patient population seen in the VAMCNO. One physician conducted all data extractions, and he was blinded to the patient's case-control status. This minimized the likelihood of information bias. The case-control status was classified based on mean [HbA.sub.1c] level done at the clinical lab at VAMCNO. This also minimized the likelihood to encounter classification bias and selection bias.

Some limitations of this study deserve attention. Fourteen percent of the patients failed to respond to the physical activity questionnaire offered at VAMCNO clinics. The physical activity questionnaire was brief and did not ask about frequency of moderate and vigorous physical activities or the duration of usual physical activities. The questionnaire was completed at different clinics and was not standardized.

In conclusion, our findings suggest that exercise offers benefits in diabetic control to the veteran diabetic patient population after adjusting for all other factors such as medication, diet, BMI, smoking, and alcohol consumption. Future studies in this area should include more objective and quantifiable measures of exercise. Further studies should use a cohort study design to study physical activity and diabetic control among diabetic patients.

References

(1.) National Diabetes Data Group: Diabetes in America. Bethesda, Md, National Institutes of Health, 2nd Ed, 1995

(2.) American Diabetes Association: Nutrition recommendation. and principles for people with diabetes mellitus. Diabetes Care 1995; 18(suppl 1):16-19

(3.) Ronnemaa T, Mattila K, Lehtonen A, et al: A controlled randomized study on the effect of long-term physical exercise on the metabolic control in type 2 diabetic patients. Acta Med Scand 1986; 220:219-224

(4.) Holloszy JO, Schultz J, Kusnierkiewicz J, et al: Effects of exercise on glucose tolerance and insulin resistance: brief review and some preliminary results. Acta Med Scand 1986; 711 (suppl):55-65

(5.) King DS, Dalsky GP, Staten MA, et al: Insulin action and secretion in endurance-trained and untrained humans. J Appl Physiol 1987; 63:2247-2252

(6.) Folsom AR, Jacobs DR Jr, Wagenknecht LE, et al: Increase in fasting insulin and glucose over seven years with increasing weight and inactivity of young adults: the CARDIA study. Am J Epidemiol 1996; 144:235-246

(7.) Mayer-Davis EJ, D'Agostino RD Jr, Karter AJ, et al: Intensity and amount of physical activity in relation to insulin sensitivity: the Insulin Resistance Study. JAMA 1998; 279:669-674

(8.) US Department of Health and Human Services: Physical Activity and Health: Report of Surgeon General Executive Summary. Atlanta, Ga, centers for Disease Control and Prevention, 1996, pp 1-14

(9.) Ford ES, Herman WH: Leisure-time physical activity patterns in the US diabetic population. Diabetes Care 1995; 18:27-33

(10.) Schneider SH, Ruderman NE: Exercise and NIDDM (technical Review). Diabetes Care 1990; 13:785-789

(11.) Wasserman DH, Zinman B: Exercise in individuals with IDDM (technical Review). Diabetes Care 1994; 17:924-937

(12.) Diabetes and exercise: the risk-benefit profile. The Health Professional's Guide to Diabetes and Exercise. Devlin JT, Ruderman N (eds). Alexandria, Va, American Diabetes Association, 1995, pp 3-4

(13.) SAS Institute: SAS/STAT User's Guide, Version 6. Cary, NC, SAS Institute Inc, 4th Ed, 1990

(14.) Ivy JL: Role of exercise training in prevention and treatment of insulin resistance and non-insulin-dependent diabetes mellitus. Sports Med 1997; 24:321-336

(15.) Helmrich SP, Ragland DR, Leung RW, et al: Physical activity and reduced occurrence of non-insulin-dependent diabetes mellitus. N Engl J Med 1991; 325:147-152

(16.) Manson JE, Rimm EB, Stampfer MJ, et al: Physical activity and incidence of non-insulin-dependent diabetes mellitus in women. Lancet 1991; 338:774-778

(17.) Manson JE, Nathan DM, Krolewski AS, et al: A prospective study of exercise and incidence of diabetes among US male physicians. JAMA 1992; 268:63-67

(18.) Lipton RB, Liao Y, Cao G, et al: Determinants of incident non-insulin-dependent diabetes mellitus among blacks and whites in a national sample: the NHANES I Epidemiologic Follow-Up Study. Am J Epidemiol 1993; 138:826-839

(19.) Burchfiel CM, Sharp DS, Curb JD, et al: Physical activity and incidence of diabetes: the Honolulu Heart Program. Am J Epidemiol 1993; 138:826-839

(20.) Monterrosa AB, Haffner SM, Stern MP, et al: Sex difference in lifestyle factors predictive of diabetes in Mexican-Americans. Diabetes Care 1995; 18:448-456

(21.) Perry IJ, Wannamethee SG, Walker MK, et al: Prospective study of risk factors for development of non-insulin-dependent diabetes in middle aged British men. BMJ 1995; 310:560-564

(22.) Lynch J, Helmrich SP, Lakka TA, et al: Moderately intense physical activities and high levels of cardiorespiratory fitness reduce the risk of non-insulin-dependent diabetes mellitus in middle-aged men. Arch Intern Med 1996; 156:1307-1314

(23.) Physical Activity and Health: A Report of the Surgeon General. Atlanta, Ga, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, 1996

(24.) Paffenbarger RS Jr, Lee IM, Kampert JB: Physical activity in the prevention of non-insulin-dependent diabetes mellitus. World Rev Nutr Diet 1997; 82:210-218

(25.) Pate RR, Pratt M, Blair SN, et al: Physical activity and public health: a recommendation from the Centers for Disease Control and Prevention and American College of Sports Medicine. JAMA 1995; 273:402-407

(26.) NIH Consensus Development Panel on Physical Activity and Cardiovascular Health: Physical activity and cardiovascular health. JAMA 1996; 276:241-246

(27.) American Diabetes Association: Clinical practice recommendations 2000. Diabetes mellitus and exercise. Diabetes Care 2000; 23(suppl 1): S1-S116

TABLE 1.

Selected Characteristics of Study Subjects (N=258)


Age                  20-49 (29, 10.8%)    50-59 (60, 22.4%)
Race                 White (104, 38.8%)   Black (158, 59.0%)
Body mass            Normal weight (<28)  Overweight ([greater than
 index               (88, 35.2%)           or equal to]28)
                                           (162, 64.8%)
Alcohol              Yes (65, 24.7%)      No (161, 61.0%)
Exercise             Yes (116, 43.3%)     No (114, 42.5%)
Smoke (current)      Yes (66, 24.7%)      No (163, 61.0%)
Smoke (ever smoked)  Yes (144, 53.7%)     No (85, 31.7%)
[HbA.sub.lc]         Under control (<8)
                      (176, 65.7%)



Age                  60-69 (73, 27.2%)          70-99 (106 39.6%)
Race                 Other (6, 2.2%)
Body mass
 index

Alcohol
Exercise
Smoke (current)
Smoke (ever smoked)
[HbA.sub.lc]         Not under control
                      ([greater than or
                       equal to]8) (92, 34.3%)


Age                  Missing 0
Race                 Missing 0
Body mass            Missing 0
 index

Alcohol              Missing 42 (14.2%)
Exercise             Missing 38 (14.2%)
Smoke (current)      Missing 38 (14.2%)
Smoke (ever smoked)  Missing 39 (14.6%)
[HbA.sub.lc]         Missing 0
TABLE 2.

Characteristics of Case patients and Control Patients with Type
2 Diabetes

                                                        [chi
Characteristic                 Cases (%)  Control (%)  square]  P Value

Race
 Black                         48 (53.3)  110 (62.9)
 White                         41 (45.6)   63 (37.1)
 Other                          1 (1.1)     2 (1.1)      2.3      .31

Age
 20-49                         19 (20.6)   10 (5.6)
 50-59                         29 (31.5)   31 (17.6)
 60-69                         25 (27.2)   48 (27.3)
 [greater than or equal to]70  19 (20.6)   87 (49.3)    30.4      .001

Body mass index
 <28 (normal weight)           23 (27.4)   65 (39.2)
 [greater than or equal to]28
 (overweight)                  61 (72.6)  101 (60.8)     3.4      .07

Alcohol
 Yes                           31 (38.8)   34 (23.3)
 No                            49 (61.3)  112 (76.7)     6.0      .014

Current Smoker
 Yes                           22 (27.2)   44 (29.7)
 No                            59 (72.8)  104 (70.3)     0.2      .68

ADA Diet
 Yes                           45 (48.9)   92 (52.6)
 No                            47 (51.1)   83 (43.4)     0.3      .57

Insulin
 Yes                           60 (65.2)   73 (41.5)
 No                            32 (34.8)  103 (58.5)    13.6      .001

Sulfonylurea
 Yes                           36 (39.1)   83 (47.2)
 No                            56 (60.9)   93 (52.8)     1.6      .21

Metformin
 Yes                           42 (45.7)   54 (30.7)
 No                            50 (54.4)  122 (69.3)     5.9      .02
TABLE 3.

Results of Multiple Logistic Regression Model with Diabetic
Control as Dependent Variable (N=225)

Independent                             [chi
Variables                        Beta  square]   SE   P Value   OR

ETOH
  No                                                           1.00
  Yes                            0.74   4.07    0.36   .044    2.10

Race
  White                                                        1.00
  Black                          0.36   1.04    0.35   .307    1.43

BMI
  <30                                                          1.00
  [greater than or equal to]30   0.88   1.44    0.73   .231    2.42

Exercise
  Yes                                                          1.00
  No                             1.00   8.34    0.34   .004    2.71

Diet
  ADA diet                                                     1.00
  Not on ADA                    -0.29   0.71    0.34   .399    0.748

Age
  <60                                                          1.00
  [greater than or equal to]60   0.80   5.03    0.34   .025    2.24

Insulin
  No                                                           1.00
  Yes                            1.84   14.09   0.49   .001    6.31

Sulfonyurea
  No                                                           1.00
  Yes                            0.83   2.83    0.49   .09     2.29

Metformin
  No                                                           1.00
  Yes                            0.85   5.44    0.36   .02     2.35

Acarbose
  No                                                           1.00
  Yes                            1.55   6.25    0.61   .01     4.70

Smoker (current)
  No                                                           1.00
  Yes                           -0.51   1.30    0.46   .25     0.60

Ever smoke
(current and former)
  No                                                           1.00
  Yes                            0.09   0.05    0.40   .80     1.10

Independent
Variables                        95% C.I

ETOH
  No
  Yes                           1.02-4.32

Race
  White
  Black                         0.72-2.86

BMI
  <30
  [greater than or equal to]30  0.57-10.29

Exercise
  Yes
  No                            1.38-5.32

Diet
  ADA diet
  Not on ADA                    0.38-1.47

Age
  <60
  [greater than or equal to]60  1.11-4.51

Insulin
  No
  Yes                           2.42-16.51

Sulfonyurea
  No
  Yes                           0.87-6.02

Metformin
  No
  Yes                           1.15-4.80

Acarbose
  No
  Yes                           1.40-15.85

Smoker (current)
  No
  Yes                           0.24-1.45

Ever smoke
(current and former)
  No
  Yes                           0.50-2.41

RELATED ARTICLES: KEY POINTS

* Type 2 diabetes mellitus is a common disease in the veteran population.

* Lifestyle modification correlates with glycemic control in this population.

* Self-reported regular exercise correlates positively and significantly with glycemic control as measured by [HbA.sub.1c] level.