riers moderately lower, and quality of life moderately higher for inhaled ... life, and barriers to insulin adherenceâ ...... routine: a nationwide quality-circle expe-.
Clinical Care/Education/Nutrition O R I G I N A L
A R T I C L E
Satisfaction and Quality of Life With Premeal Inhaled Versus Injected Insulin in Adolescents and Adults With Type 1 Diabetes MARCIA A. TESTA, MPH, PHD1 DONALD C. SIMONSON, MD, MPH, SCD2
OBJECTIVE — We sought to compare and evaluate the impact of inhaled versus injected insulin on potential mediators of patient acceptance of insulin therapy while maintaining comparable A1C levels. RESEARCH DESIGN AND METHODS — During a noninferiority efficacy trial conducted in 40 centers in the U.S., we surveyed treatment satisfaction, quality of life, and adherence barriers at weeks ⫺4, ⫺1, 6, 12, 20, and 24 in adolescents aged 12–17 years and adults with type 1 diabetes who received premeal regular plus twice-daily NPH insulin during a 4-week run-in; then, subjects were randomized to premeal inhaled human insulin plus twice-daily NPH (adults, n ⫽ 102; adolescents, n ⫽ 60) (inhaled) or remaining on run-in therapy (n ⫽ 105 and 60, respectively) (subcutaneous injection). RESULTS — Overall treatment satisfaction (0 –100) increased by 13.2 ⫾ 1.1 units for inhaled insulin (baseline ⫽ 63.3 ⫾ 1.2) compared with 1.7 ⫾ 0.8 for subcutaneous insulin injection (baseline ⫽ 64.1 ⫾ 1.2, P ⬍ 0.0001). All 12 satisfaction subscales favored inhaled insulin (all P ⬍ 0.01), and effects did not vary by age or sex. Despite similar baseline-adjusted end point A1C for inhaled (7.7 ⫾ 0.1%) and subcutaneous (7.9 ⫾ 0.1%) regimens, quality-of-life scales of mental health, symptoms, health status, cognitive functioning, and adherence barriers during treatment were more favorable for inhaled insulin (all P ⬍ 0.05). Greater satisfaction was associated with fewer barriers to insulin adherence (rho ⫽ ⫺0.78, P ⬍ 0.0001) and a greater reduction in A1C (rho ⫽ ⫺0.18, P ⬍ 0.001). CONCLUSIONS — Treatment satisfaction was substantially more favorable, adherence barriers moderately lower, and quality of life moderately higher for inhaled compared with subcutaneous regimen. It remains to be demonstrated whether these patient-reported outcomes will translate into improved adherence and glycemic control. Diabetes Care 30:1399–1405, 2007
aintaining A1C levels ⬍7% in individuals with type 1 diabetes has been shown to reduce microvascular and some macrovascular complications (1,2). Reaching A1C goals by intensifying the insulin regimen often re-
M
quires either an insulin pump or a basal/ bolus multiple-injection insulin regimen. However, use of the pump has been limited by its cost, required technical expertise, and the relative paucity of health care providers trained in its use, and the ac-
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
From the 1Department of Biostatistics, Harvard School of Public Health, Boston, Massachusetts; and the 2 Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts. Address correspondence and reprint requests to Marcia A. Testa, MPH, PhD, Department of Biostatistics, Harvard School of Public Health, 677 Huntington Ave., Boston, MA 02115. E-mail: testa@hsph. harvard.edu. Received for publication 20 July 2006 and accepted in revised form 16 February 2007. Published ahead of print at http://care.diabetesjournals.org on 2 March 2007. DOI: 10.2337/dc06-1497. Clinical trial reg. no. NCT00424333, clinicaltrials.gov. M.A.T. and D.C.S. have received research grants and honoraria from Pfizer. Additional information for this article can be found in an online appendix at http://dx.doi.org/10.2337/ dc06-1497. Abbreviations: FPG, fasting plasma glucose. A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion factors for many substances. © 2007 by the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
DIABETES CARE, VOLUME 30, NUMBER 6, JUNE 2007
ceptance of basal/bolus insulin regimens is hindered by the burden of multiple injections (3– 6). Clinical trials have demonstrated that inhaled insulin is comparable with injected insulin in lowering A1C in individuals with type 1 (7,8) and type 2 (9) diabetes. However, the clinical rationale for its use is based on the belief that multiple inhalations will be more acceptable to patients, thereby promoting increased adherence to intensive insulin regimens. However, evidence of this causal relationship has not been empirically established during clinical trials of efficacy since adherence to treatment and A1C goals are strictly enforced for all patients. Additionally, real-world clinical experience with the only form of inhaled insulin approved by the U.S. Food and Drug Administration is lacking since it has just recently become available. While empirical data documenting the relationship between greater acceptance and superior glycemic control is not available, mediators in the pathway— namely, treatment satisfaction, quality of life, and barriers to insulin adherence— can be evaluated. Such measures could provide valuable insight when weighing the risks and benefits of incorporating inhaled insulin delivery as part of intensive insulin regimens. The purpose of our study was to compare the impact of inhaled versus injected insulin on potential mediators of adherence during a clinical trial (8) that previously reported that inhaled human insulin has comparable safety and efficacy to injected insulin in type 1 diabetic patients currently using multiple daily injections. RESEARCH DESIGN AND METHODS — A detailed description of the clinical results of the parent protocol was published previously (8). Briefly, this open-label, randomized, multicenter trial consisted of a screening visit, 4-week lead-in, and 24-week treatment phase. A total of 419 individuals with type 1 diabetes for at least 1 year were screened at 40 centers in the U.S. and Canada. Inclu1399
Satisfaction with inhaled insulin in type 1 diabetes Table 1—Demographic and clinical characteristics of study groups at randomization Inhaled insulin regimen
n Male Married Caucasian Occupation Paid employment Retired Student Age (years) Duration of diabetes (years) BMI (kg/m2) A1C (%) C-peptide (pmol/ml)
Subcutaneous regimen
Adolescents
Adults
Adolescents
Adults
60 31 (51.7) 0 (0.0) 52 (86.7)
102 53 (52.0) 66 (64.7) 87 (85.3)
60 30 (50.0) 0 (0.0) 54 (90.0)
105 59 (56.2) 65 (62.5)* 98 (93.3)
0 (0.0)* — 59 (100.0)* 13.8 ⫾ 1.4 (12–17) 5.5 ⫾ 3.5 (1.0–13.7) 22.4 ⫾ 3.4 (16.9–30.0) 8.3 ⫾ 0.9 (6.3–10.6) 0.08 ⫾ 0.06 (0.07–0.40)
82 (81.2)* 2 (2.0)* 5 (5.0)* 38.2 ⫾ 10.6 (19–65) 17.0 ⫾ 10.6 (2.2–50.0) 25.4 ⫾ 2.9 (18.0–32.0) 7.8 ⫾ 0.9 (6.2–10.5) 0.07 ⫾ 0.03 (0.07–0.20)
1 (1.7) — 58 (96.7) 14.1 ⫾ 1.7 (11–17) 6.1 ⫾ 3.5 (1–13.6) 22.0 ⫾ 3.2 (15.0–30.0) 8.3 ⫾ 0.9 (6.3–10.6) 0.08 ⫾ 0.03 (0.07–0.20)
90 (85.7) 3 (2.9) 4 (3.8) 38.6 ⫾ 11.0 (19–65) 19.4 ⫾ 11.2 (1.5–49.0) 25.7 ⫾ 3.3 (17.4–35.0) 7.8 ⫾ 1.0 (5.9–10.5) 0.07 ⫾ 0.03 (0.07–0.23)
Data are n (%) or means ⫾ SD (range). *Missing data on one individual.
sion criteria were age 12– 65 years, stable on an insulin regimen of at least two injections daily for 2 months before screening, baseline A1C between 6 and 11% inclusive, fasting plasma C-peptide ⱕ0.2 pmol/ml, and BMI ⱕ30 kg/m2. Exclusion criteria included poorly controlled asthma, recent smoking, and significant laboratory abnormalities. The protocol was approved by the institutional review board, and informed consent was obtained from all subjects. To standardize baseline assessments during the 4 weeks before randomization, all patients were switched to subcutaneous premeal regular insulin plus twicedaily NPH insulin administered in four subcutaneous injections daily (subcutaneous regimen). Target glucose ranges were 80 –120 mg/dl (4.4 – 6.7 mmol/l) before meals and 100 –140 mg/dl (5.6 – 7.8 mmol/l) before bedtime. A total of 162 individuals were randomized to insulin human (rDNA origin) Inhalation Powder (Exubera; Pfizer and Nektar Therapeutics) and twice-daily NPH insulin (inhaled regimen); 165 individuals were instructed to continue on their prerandomization subcutaneous regimen. Inhaled insulin was administered as one to two inhalations within 10 min of starting each meal. The insulin powder was packaged in foil blisters of 1- and 3-mg doses, which are approximately equivalent to 3 IU and 8 IU of subcutaneous insulin, respectively. Insulin doses were adjusted weekly by the investigator to achieve target premeal glucose levels, and patients adjusted doses according to diet and exercise requirements. 1400
Assessments Clinical. A1C was measured at baseline (week ⫺1 to 0) and end point (week 24 or last visit), and fasting plasma glucose (FPG) was recorded at each visit. Patient-reported questionnaires. At week ⫺4 (screening); week ⫺1 (baseline); weeks 6, 12, 20, and 24; and early withdrawal, questionnaires were sent directly from clinical sites to the central survey laboratory. Validated instruments of treatment satisfaction, quality of life, and barriers to insulin adherence were used (10,11) (see online appendix available at http://dx.doi.org/10.2337/dc06-1497). These measures are responsive to the effects of therapeutic interventions (12– 14), the impact of side effects of medication (15), symptoms of diabetes (16 –18), and changes in A1C (19 –21). An age-appropriate parallel version of the adult questionnaire was used for adolescents aged 12–17 years. All scales were coded such that higher scores reflected the more favorable response, i.e., higher satisfaction, better quality of life, and reduced barriers to adherence.
Statistical methods Treatment satisfaction and quality of life during treatment at weeks 6, 12, 20, and 24 were evaluated using linear mixed models with fixed (treatment) and baseline covariate effects. In addition, changes from baseline to end point (week 24 or last visit) satisfaction were compared using ANCOVA (adjusted for baseline). All P values report nominal significance levels at ␣ ⫽ 0.05. For reference, a partial
Bonferroni’s adjustment for type I error for multiple end points is provided (22,23). Correlation analysis (Pearson r and Spearman’s rho) and multiple regression were used to examine the relationships between treatment, glycemic control, satisfaction, adherence barriers, and quality of life. Data are given as means ⫾ SE unless specified otherwise.
RESULTS Study accrual, withdrawals, and baseline characteristics In total, 120 adolescents aged 12–17 years were randomized to the inhaled (n ⫽ 60) and subcutaneous (n ⫽ 60) regimens, and 207 adults were randomized to inhaled (n ⫽ 102) and subcutaneous (n ⫽ 105) regimens. For the 160 patients on inhaled regimen who completed baseline questionnaires, 154 (96.3%) completed the week 24 questionnaire and 159 (99.4%) completed week 12 or later; for the 162 subcutaneous regimen baseline completers, corresponding values were 152 (93.8%) and 155 (95.7%), respectively. Baseline demographics, clinical characteristics, and insulin treatments are summarized in Table 1. Baseline satisfaction scores indicated moderate satisfaction with the run-in subcutaneous insulin treatment (Table 2) (scale range was 0 [greatest dissatisfaction] to 100 [highest satisfaction]). The one exception was the low score of 35 for the preference scale, suggesting patients had a desire to seek new treatments. DIABETES CARE, VOLUME 30, NUMBER 6, JUNE 2007
Testa and Simonson Table 2—Diabetes treatment satisfaction scores: baseline, end point (last observation carried forward), and change from baseline for inhaled insulin and subcutaneous regimen treatment groups Baseline Scaled (n of items) Regimen Burden (14) Convenience (6) Flexibility (4) Hassle (8) Interference (11) Pain (3) Social (9) Outcomes Perceived efficacy (3) Side effects (5) Net benefit General satisfaction (5) Preference (2) Advocacy (2) Total score Overall satisfaction (72)
End point
Change
Tests
Group
n
Mean
SE
Mean
SE
Mean
SE
P*
P†
INH SC INH SC INH SC INH SC INH SC INH SC INH SC
159 158 160 159 160 159 160 159 160 159 160 159 158 159
69.6 71.9 61.4 60.8 60.5 62.0 67.3 67.7 70.8 71.0 70.9 72.7 72.5 74.7
1.6 1.4 1.5 1.5 1.6 1.7 1.6 1.6 1.7 1.8 1.7 1.7 1.7 1.6
79.4 74.2 74.0 62.0 72.7 63.6 77.0 72.0 80.6 76.1 85.2 73.5 82.4 76.2
1.4 1.6 1.5 1.6 1.6 1.8 1.5 1.7 1.5 1.8 1.4 1.6 1.4 1.7
9.9 2.3 12.6 1.2 12.2 1.6 9.7 4.4 9.8 5.1 14.2 0.8 9.9 1.5
1.3 1.0 1.6 1.1 1.4 1.3 1.2 1.2 1.2 1.2 1.5 1.2 1.2 1.0
⬍0.0001 0.018 ⬍0.0001 0.27 ⬍0.0001 0.21 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 0.521 ⬍0.0001 0.141
⬍0.0001
INH SC INH SC
160 159 159 159
54.2 55.3 73.4 73.2
1.5 1.4 1.2 1.3
68.1 58.1 76.8 70.6
1.5 1.7 1.4 1.5
13.9 2.9 3.4 ⫺2.6
1.7 1.5 1.2 1.2
⬍0.0001 0.051 0.004 0.034
⬍0.0001
INH SC INH SC INH SC
160 159 158 159 158 156
58.7 59.2 34.7 35.2 65.2 66.3
1.7 1.8 1.4 1.9 1.6 1.9
77.0 59.4 62.0 36.9 81.5 66.5
1.7 1.9 1.8 1.9 1.5 2.0
18.3 0.3 27.4 1.7 16.3 0.2
2.2 1.6 2.1 1.5 2.0 1.8
⬍0.0001 0.855 ⬍0.0001 0.251 ⬍0.0001 0.922
⬍0.0001
INH SC
160 159
63.3 64.1
1.2 1.2
76.4 65.8
1.2 1.4
13.2 1.7
1.1 0.8
⬍0.0001 0.036
⬍0.0001
⬍0.0001 ⬍0.0001 0.003 0.005 ⬍0.0001 ⬍0.0001
⬍0.0001
⬍0.0001 ⬍0.0001
All scales range from 0 to 100, and higher scores reflect better or more favorable satisfaction. n is less than total due to missing values and/or questionnaires. *Nominal P value for paired t test (baseline ⫺ end point) ⫽ 0. †Nominal P value for mean treatment differences between changes from baseline by ANCOVA with baseline score as covariate. Treatment effects for the corresponding linear mixed-model analyses (weeks 6, 12, 20, and 24 assessments) adjusted for baseline (week ⫺1) were significant at P ⬍ 0.0001 for all scales. For the 12 baseline to end point treatment satisfaction difference scores with mean correlation coefficient r ⫽ 0.59, compare nominal P values to a reduced Bonferroni-corrected ␣ of 0.018/0.0036/0.0003 to achieve significance at ␣ ⬍ 0.05/0.01/0.001. INH, inhaled insulin; SC, subcutaneous.
Changes in glycemic control Baseline-adjusted A1C at final visit was 7.7 ⫾ 0.1% for inhaled insulin and 7.9 ⫾ 0.1% for subcutaneous regimens (baseline ⫽ 8.0 ⫾ 0.1% for both groups, P ⫽ 0.066 between treatment-adjusted A1C baseline to end point change). ANCOVA showed an overall age effect (P ⫽ 0.002); however, tests of the interactions indicated that the treatment impact on A1C did not vary by age (P ⫽ 0.59), sex (P ⫽ 0.92), or age and sex (P ⫽ 0.94). The 24-week change in FPG for inhaled insulin was ⫺1.9 mmol/l (⫺35 mg/dl), whereas the subcutaneous group increased by 0.2 mmol/l (⫹4 mg/dl), with an adjusted treatment group difference of ⫺2.2 mmol/l (⫺40 mg/dl) (95% CI ⫺3.2 mmol/l [⫺58 mg/dl] to ⫺1.2 mmol/l [⫺22 mg/dl]). DIABETES CARE, VOLUME 30, NUMBER 6, JUNE 2007
Patient-reported outcomes Diabetes treatment satisfaction. Baseline to end point changes across all 12 satisfaction subscales were more favorable for inhaled insulin (Table 2). For the seven regimen-related subscales, improvements for inhaled insulin were substantially greater than for subcutaneous regimen. The two outcome-related subscales, satisfaction with glycemic control and side effects (primarily weight gain and hypoglycemia), also were more favorable for inhaled insulin. The three “net benefit” measures of preference (desire to seek alternative diabetes treatment), advocacy (would recommend treatment), and general satisfaction confirmed that patients found inhaled insulin more acceptable after balancing regimen processes and outcomes. For inhaled insulin,
the overall treatment satisfaction scale increased to 75.3 ⫾ 1.1 by week 6 and remained constant at that level until end point, whereas subcutaneous regimen did not change. Subgroup treatment comparisons by age-group and sex are shown in Fig. 1. None of the two- and three-way interactions from ANCOVA were statistically significant, indicating that differences between inhaled insulin and subcutaneous regimens were comparable across these demographic subgroups. Quality of life during treatment. Between screening and baseline there was 0.4% decrease in A1C (P ⬍ 0.001) with a corresponding improvement in overall quality of life (P ⫽ 0.024) for both groups combined. Overall quality of life (absolute scale range of 500 units and operative range of ⬃64 units [17]) improved at end 1401
Satisfaction with inhaled insulin in type 1 diabetes
Figure 1—Mean baseline-adjusted change in overall treatment satisfaction for inhaled insulin regimen ( ) and subcutaneous regimen (䡺) evaluated at baseline score of 63.7. §P ⬍ 0.0001 and ‡P ⬍ 0.01 versus inhaled insulin. A1C treatment changes were not statistically significantly different between inhaled insulin and subcutaneous regimens within the age- and sex-specific subgroups.
point by 12.2 ⫾ 3.3 units for inhaled insulin and 2.6 ⫾ 3.4 units for subcutaneous regimens (P ⫽ 0.043), representing a 15% difference in the operative range (0.31 ⫾ 0.15 SD responsiveness units). This difference is within the range of values that are likely to be clinically relevant (10,11,17,21). As shown in Table 3, treatment differences in the quality-of-life scales were more favorable for inhaled insulin, including symptom distress, symptom interference in daily activities, cognitive functioning, and general health. The more positive mental and emotional health scores were driven primarily by improvements in anxiety and behavioral and emotional control. Both age and sex modified the treatment effect on mental health (treatment interactions with agegroup [P ⫽ 0.001], sex [P ⫽ 0.002], and age-group by sex [P ⬍ 0.0001]). Adult female subjects showed the largest inhaled insulin regimen gain in mental health (⫹10.9 units) in contrast to adult male (⫺3.6 units), adolescent male (⫹5.9 units), and adolescent female (⫹5.4 units) subjects. Adherence barriers to insulin use. As shown in Table 3, the inhaled insulin group scored more favorably on six of the seven questions relating to barriers to insulin adherence, and these effects did not 1402
vary by age. There was no difference with regard to self-consciousness when using insulin away from home. Six questions addressing barriers specific to injectable insulin revealed that inhaled insulin rated fewer barriers with insulin (regardless of delivery) (total score 74.2 ⫾ 1.4 for inhaled insulin vs. 67.0 ⫾ 1.6 for subcutaneous regimens [higher score reflects fewer/lower barriers], P ⫽ 0.001). Answers to the 12 inhaler device–specific insulin questions confirmed positive endorsement for the inhaler device itself, with the highest endorsement rating of 94.9 ⫾ 1.0 (on a scale of 0 –100) for reduced pain. All ratings were significantly higher than a neutral endorsement score of 50.0 (P ⬍ 0.0001), and the average total rating was 78.3 ⫾ 1.4 (P ⬍ 0.0001 vs. 50.0). Inhaled insulin patients rated the bother and hassle associated with their inhaler device consistently lower at weeks 6, 12, 20, and 24 ([mean ⫾ SE] 83.3 ⫾ 1.9, 82.2 ⫾ 2.0, 80.3 ⫾ 2.1, and 82.4 ⫾ 2.0, respectively) than their NPH injections (68.6 ⫾ 2.0, 68.4 ⫾ 2.1, 66.1 ⫾ 2.1, and 67.6 ⫾ 2.3, respectively, P ⬍ 0.001). Symptom distress. Of the 53 patientreported symptoms, the inhaled insulin group reported less symptom distress and severity overall (Table 3, symptom dis-
tress scale). Forty symptom scores were less distressing for the inhaled insulin group. For symptom distress treatment differences ⬎0.10 – 0.19 SD responsiveness units, 13 were more favorable for inhaled insulin and 1 was more favorable for subcutaneous regimen. All seven symptom distress treatment differences ⬎0.20 SD responsiveness units, a typical threshold for clinically important differences (10,11,17,21), favored inhaled insulin, including general weakness or fatigue (⫹0.21), sugar in urine (⫹0.23), feelings of high blood glucose (⫹0.27), impaired or worsening vision (⫹0.27, P ⬍ 0.05), lethargy, no energy to do things (⫹0.36, P ⬍ 0.001), tired, feeling weary (⫹0.36, P ⬍ 0.001), and feeling overweight (⫹0.37, P ⬍ 0.01). Adult female subjects on inhaled insulin had a decrease in weight of 0.4 ⫾ 0.3 kg, while those on subcutaneous regimen gained 0.6 ⫾ 0.3 kg (difference of ⫺1.0 ⫾ 0.4 kg, P ⫽ 0.01). The reduction in distress with feeling overweight was attributed primarily to adult female subjects (treatment difference of ⫹0.71 SD units), and actual weight loss was associated with less distress (r ⫽ 0.30, P ⬍ 0.001). Comparative treatment preference and associations between treatment satisfaction, quality of life, adherence barriers, and glycemic control. Preference choices are shown in Table 3, and, except for “easier to adjust dose,” all favored inhaled insulin. Analyses between end point overall treatment satisfaction and insulin adherence barriers demonstrated greater satisfaction with lower barriers (rho ⫽ ⫺0.78, P ⬍ 0.0001). A1C and overall treatment satisfaction correlation analysis indicated that better glycemic control was only weakly associated with higher satisfaction (rho ⫽ ⫺0.18, P ⬍ 0.001) since the lack of change in A1C in this noninferiority design limited the ability to detect an association. CONCLUSIONS — For individuals with type 1 diabetes treated with two or more daily injections of regular/NPH insulin and moderately satisfied with their current therapy, there is opportunity to improve their experience with insulin therapy. Inhaled insulin patients reported substantial and stable improvement in their satisfaction ratings between weeks 6 and 24 in contrast to no change for those remaining on the run-in subcutaneous regimen. These findings are consistent with other recent studies suggesting that ease of use, convenience, social comfort, DIABETES CARE, VOLUME 30, NUMBER 6, JUNE 2007
Testa and Simonson Table 3—Quality of life, barriers to insulin adherence, and comparative preference during treatment (weeks 6, 12, 20, and 24 assessments) adjusted for baseline (week ⴚ1) for the inhaled insulin and subcutaneous insulin groups
QOL scales and subscales (n of items)* Overall mental health Psychological distress Psychological well being (total subscale) General health status (3) Symptom distress (53 Items) Cognitive functioning (total scale) (100–500) Diabetes symptom interference (7) Barriers to insulin adherence‡ I find it difficult to take every dose of insulin recommended. I find it difficult to take insulin away from home. I find it easy to travel for a few days and take all my doses of insulin. I find it convenient to take insulin. The time I spend taking each dose of insulin is acceptable. I am self-conscious taking insulin away from home. Overall, I am satisfied with my current way of taking insulin. Insulin adherence barriers (total score)
Baseline week ⫺1 covariate score†
Inhaled insulin
Subcutaneous insulin
482.4 515.9 426.4 500.1 580.6 422.6
491.1 ⫾ 1.8 523.4 ⫾ 1.8 437.7 ⫾ 2.5 509.4 ⫾ 2.4 583.8 ⫾ 0.8 427.1 ⫾ 1.4
486.2 ⫾ 1.8 523.4 ⫾ 1.8 433.7 ⫾ 2.5 502.3 ⫾ 2.4 579.1 ⫾ 0.8 421.3 ⫾ 1.4
546.5
556.9 ⫾ 2.6 547.7 ⫾ 2.7
Mean treatment difference (⫹, favors inhaled P treatment insulin) main effect† ⫹5.1 (0.8 to 10.2) ⫹5.5 (0.5 to 10.4) ⫹4.0 (⫺2.8 to 10.9) ⫹7.1 (0.3 to 13.9) ⫹4.7 (2.4 to 7.0) ⫹5.8 (2.0 to 9.6)
0.047 0.031 0.248 0.039 ⬍0.0001 0.003
⫹9.2 (1.8 to 16.5)
0.015
75.6
82.9 ⫾ 1.8
74.4 ⫾ 1.8
⫹8.4 (3.4 to 13.4)
0.001
70.1
76.4 ⫾ 2.0
70.7 ⫾ 2.0
⫹5.7 (0.3 to 11.2)
0.04
66.9
76.8 ⫾ 2.0
66.9 ⫾ 2.0
⫹9.9 (4.4 to 15.4)
0.0005
56.4 69.3
70.0 ⫾ 2.0 77.5 ⫾ 1.5
60.1 ⫾ 2.0 68.7 ⫾ 1.5
⫹9.9 (4.4 to 15.5) ⫹8.8 (4.5 to 13.1)
0.0005 0.00010
59.3
62.8 ⫾ 2.4
65.1 ⫾ 2.4
⫺2.4 (⫺4.4 to 9.2)
0.49
60.1
75.4 ⫾ 1.8
61.7 ⫾ 1.8
⫹13.7 (8.7 to 18.7)
⬍0.0001
65.5
74.7 ⫾ 1.3
66.7 ⫾ 1.3
⫹8.0 (4.4 to 11.6)
⬍0.0001
Comparative preference ratings§
Adolescents 关% (n/total)兴储
Is easier to use? Is more convenient? Allows more flexibility in daily activities? Causes fewer episodes of low blood sugar? Is more comfortable to use in public? Is easier to adjust dose? Allows easier disposal of medication supplies? Allows more flexibility in scheduling meals? Is more convenient when traveling? Gives you better glucose control? Would you choose if both were the same price? Makes you feel better about yourself? Overall is the one you prefer?
93.0 (53/57) 87.7 (50/57) 83.9 (47/56) 80.0 (44/55) 82.1 (46/56) 39.3 (22/56) 83.9 (47/56) 83.6 (46/55) 84.2 (48/57) 70.4 (38/54) 86.0 (49/57) 90.6 (48/53) 87.7 (50/57)
P¶ ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 0.14 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001
Adults 关% (n/total)兴储 83.5 (81/97) 77.1 (74/96) 81.1 (77/95) 71.6 (63/88) 79.4 (77/97) 27.4 (26/95) 95.9 (93/97) 78.7 (74/94) 81.1 (77/95) 69.3 (61/88) 82.5 (80/97) 64.6 (77/91) 82.5 (80/97)
P¶ ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001 0.0004 ⬍0.0001 ⬍0.0001 ⬍0.0001
Data are means ⫾ SE or mean (95% CI) unless otherwise indicated. Higher scores reflect better quality of life, less symptom distress, and lower barriers to adherence. *All scaled 100 – 600 unless noted otherwise, and higher scores represent higher quality of life (QOL). †Treatment mean differences adjusted at baseline covariate score using linear mixed-models analysis. Baseline-adjusted treatment means (weeks 6, 12, 20, and 24) are calculated at mean baseline score (week ⫺1). For the 10 mutually exclusive quality-of-life scores tested (sexual functioning, general symptom life interference, and health limitation were not statistically different and are not shown) with mean correlation coefficient r ⫽ 0.49, compare nominal P values with a reduced Bonferroni-corrected ␣ of 0.0003/0.003/0.016 to achieve significance at ␣ ⬍0.001/0.01/0.05, respectively. For the seven adherence barriers, compare with 0.0004/0.004/0.018. ‡All scaled 0 –100, and higher scores represent lower barriers and great ease. §Percent in the inhaled insulin group reporting a preference for the inhaler compared with injections at week 24. 储Total number who declared a preference. ¶Binomial test, P ⫽ 0.5.
and flexibility of the treatment process are important issues to both type 1 and type 2 diabetic patients for insulin administration (24,25). Our findings also documented that more positive perceptions of DIABETES CARE, VOLUME 30, NUMBER 6, JUNE 2007
glycemic control, side effects, cognitive function, and physical and psychological well being accompanied the higher satisfaction ratings for patients on the inhaled insulin regimen. It is important to note
that our comparative findings may not apply to other insulin regimens. Although inhaled insulin is not approved for individuals aged ⱕ17 years, our study demonstrates its impact on sat1403
Satisfaction with inhaled insulin in type 1 diabetes isfaction and quality of life in a younger adolescent population with an average age of 14 years. Using age-appropriate instruments, improvements with inhaled insulin were found to be similar to adults among both male and female subjects. In this age-group, most patient-reported outcome studies evaluating intensive insulin regimens have focused on comparing insulin pumps with multiple daily injections. Results have varied depending on the responsiveness and sensitivity of the patient-reported measures used and the study design used, but improvement in patient satisfaction and ability to adhere to intensive regimens are typically higher for the pump. Our study used very sensitive measures and a longitudinal design during which patients answered nearly 2,400 questions over 6 months. As such, we were able to detect improvements in many domains of satisfaction, quality of life, and barriers to adherence for the inhaled insulin group. While the stability and durability of the satisfaction scores argue against a novelty effect, we cannot rule out that openly positive or negative reactions of the clinic staff to the inhaler device could have impacted patient attitudes. The reasons for the positive effects of inhaled insulin on the quality-of-life outcomes are uncertain; however, it is possible that the rapidly acting premeal inhaled insulin resulted in a more favorable metabolic profile that mediated these improvements. Also, FPG was lower by 40 mg/dl, and the inhaled insulin group reported “better glycemic control” even though both groups had statistically comparable A1C. A1C might be too insensitive a measure to detect quality-of-life improvements that are important and salient to patients. Indeed, the higher symptom distress reported by patients on the subcutaneous regimen were for thirst, general weakness or fatigue, impaired or worsening vision, lethargy, no energy to do things, tiredness, and feeling weary, all of which are associated with hyperglycemia. One of the most intriguing findings was that out of 53 symptoms, the symptom that was ameliorated most by inhaled insulin was “feeling overweight.” Although it was previously reported that body weight increased comparably in both groups (8), the gains were predominant in the growing adolescent subjects, which masked the treatment differential in adult female subjects who were most distressed by weight gain. Weight gain 1404
with insulin is a significant barrier to adhering to insulin regimens, especially in women. Because of the noninferiority design of this trial, during which adherence was purposely and strictly enforced in both groups, there was little opportunity to observe differences in A1C resulting from differences in adherence. As such, we chose to focus assessment on patient ratings of barriers to insulin adherence such as “how much difficulty they had taking every dose of insulin as recommended.” It was not surprising that individuals who expressed the most difficulty had the lowest overall satisfaction with treatment. Since it has been previously observed that clinical trials substantially underestimate discontinuation with drug therapy and overestimate adherence and that quality of life may better predict outcomes in actual practice (26,27), we might anticipate that the more favorable satisfaction and quality-of-life outcomes observed for inhaled insulin might translate into greater acceptance and better adherence with insulin therapy in clinical practice. The barriers to effective diabetes management are multifactorial, but our study supports the belief that psychological, behavioral, and attitudinal factors play a role in patient satisfaction with insulin therapy. The use of inhaled insulin in type 1 diabetic patients should proceed with caution since the long-term consequences (both positive and negative) will not be known until real-world use, and experience yields the long-term safety and efficacy data that only postmarketing surveillance can provide. Communication between physicians and patients pertaining to these patient-centered factors, especially in individuals failing to intensify or maintain their insulin regimen, might better inform the clinical decision process when weighing risks and benefits of alternative methods of delivering intensive insulin regimens. Acknowledgments — We thank the investigators (see ref. 8 for a complete list), Ralph R. Turner, PhD, MPH; Johanna F. Hayes, ScM; Sharon Murphy; and the staff at Phase V Technologies, Inc., for managing the survey data.
References 1. Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mel-
litus. N Engl J Med 329:977–986, 1993 2. Reichard P, Nilsson BY, Rosenqvist U: The effect of long-term intensified insulin treatment on the development of microvascular complications of diabetes mellitus. N Engl J Med 329:304 –309, 1993 3. Zgibor JC, Songer TJ, Kelsey SF, Weissfeld J, Drash AL, Becker D, Orchard TJ: The association of diabetes specialist care with health care practices and glycemic control in patients with type 1 diabetes: a crosssectional analysis from the Pittsburgh Epidemiology of Diabetes Complications Study. Diabetes Care 23:472– 476, 2000 4. Muller UA, Femerling M, Reinauer KM, RisseA, Voss M, Jorgens V, Berger M, Muhlhauser I: Intensified treatment and education of type 1 diabetes as clinical routine: a nationwide quality-circle experience in Germany: ASD (the Working Group on Structured Diabetes Therapy of the German Diabetes Association). Diabetes Care 22 (Suppl. 2):B29 –B34, 1999 5. Zambanini A, Newson RB, Maisey M, Feher MD: Injection related anxiety in insulin-treated diabetes. Diabetes Res Clin Pract 46:239 –246, 1999 6. Mollema ED, Snoek FJ, Heine RJ, van der Ploeg HM: Phobia of self-injecting and self-testing in insulin-treated diabetes patients: opportunities for screening. Diabet Med 18:671– 674, 2001 7. Quattrin T, Belanger A, Bohannon NJ, Schwartz SL, Exubera Phase III Study Group: Efficacy and safety of inhaled insulin (Exubera) compared with subcutaneous insulin therapy in patients with type 1 diabetes: results of a 6-month, randomized, comparative trial. Diabetes Care 27:2622–2627, 2004 8. Skyler JS, Weinstock RS, Raskin P, Yale JF, Barrett E, Gerich JE, Gerstein HC, Inhaled Insulin Phase III Type 1 Diabetes Study Group: Use of inhaled insulin in a basal/bolus insulin regimen in type 1 diabetic subjects: a 6-month, randomized, comparative trial. Diabetes Care 28:1630 – 1635, 2005 9. Hollander PA, Blonde L, Rowe R, Mehta AE, Milburn JL, Hershon KS, Chiasson JL, Levin SR: Efficacy and safety of inhaled insulin (Exubera) compared with subcutaneous insulin therapy in patients with type 2 diabetes: results of a 6-month, randomized, comparative trial. Diabetes Care 27:2356 –2362, 2004 10. Testa MA, Simonson DC: Assessing quality-of-life outcomes. N Engl J Med 334: 835– 840, 1996 11. Testa MA: Interpretation of quality of life outcomes: issues that affect magnitude and meaning. Medical Care 38:II-166 –II174, 2000 12. Testa MA, Turner RR, Krafcik MB, Simonson DC, Calvo C, Luque-Otero M, the Nifedipine GITS Study Group: Quality of life and calcium channel blockade with DIABETES CARE, VOLUME 30, NUMBER 6, JUNE 2007
Testa and Simonson
13.
14.
15.
16.
17.
18.
nifedipine GITS versus amlodipine in hypertensive patients in Spain. J Hypertens 16:1839 –1847, 1998 Testa MA, Anderson RB, Nackley JF, Hollenberg NK: Quality of life and antihypertensive therapy in men: a comparison of captopril with enalapril. N Engl J Med 328: 907–913, 1993 Testa MA, Hollenberg NK, Anderson RA, Williams GH: Assessment of quality of life by patient and spouse during antihypertensive therapy with atenolol and nifedipine GITS. Am J Hypertens 4:363–373, 1991 Anderson RB, Nackley JF, Testa MA: Symptom distress check lists as a component of quality of life measurement: comparing symptom reports with responses to multiple choice questionnaires. Drug Inf J 29:1689S–1707S, 1995 Testa MA, Simonson DC: Measuring quality of life in hypertensive patients with diabetes. Postgrad Med J 64 (Suppl. 3):50 – 58, 1988 Testa MA, Simonson DC, Turner RR: Valuing quality of life and improvements in glycemic control in people with type 2 diabetes. Diabetes Care 21 (Suppl. 3): C44 –C52, 1998 Anderson RB, Testa MA: Symptom
DIABETES CARE, VOLUME 30, NUMBER 6, JUNE 2007
19.
20.
21.
22.
23.
distress checklists as a component of quality-of life measurement: comparing prompted reports by patient and physician with concurrent adverse event reports via the physician. Drug Inf J 28:89 – 114, 1994 Testa MA, Turner RR, Simonson DC: Effects of lower blood glucose and reduced inter-day glucose variability on quality of life in type 2 diabetes (Abstract). Diabetes 52 (Suppl. 1):A420, 2003 Testa MA, Hayes JF, Turner RR, Simonson DC: Improved quality of life is associated with improved glycemic control in type 2 diabetes: an international, multicultural, multi-center, placebo-control clinical trial (Abstract). Diabetes 49 (Suppl. 1):A73, 2000 Testa MA: Quality-of-life assessment in diabetes research: interpreting the magnitude and meaning of treatment effects. Diabetes Spectrum 13:36 – 41, 2000 Dubey SD: Adjustment of p-values for multiplicities of intercorrelating symptoms. Proceedings of the VIth International Society for Clinical Biostatisticians, Du¨sseldorf, Germany, 1985 Armitage P, Parmar M: Some approaches to the problem of multiplicity in clinical trials. Proceedings of the XIIth Interna-
24.
25.
26.
27.
tional Biometrics Conference, Seattle, Washington, 1986 Weintrob N, Benzaquen H, Galatzer A, Shalitin S, Lazar L, Fayman G, Lilos P, Dickerman Z, Phillip M: Comparison of continuous subcutaneous insulin infusion and multiple daily injection regimens in children with type 1 diabetes: a randomized open crossover trial. Pediatrics 112:559 –564, 2003 Herman WH, Ilag LL, Johnson SL, Martin CL, Sinding J, Al Harthi A, Plunkett CD, LaPorte FB, Burke R, Brown MB, Halter JB, Raskin P: A clinical trial of continuous subcutaneous insulin infusion versus multiple daily injections in older adults with type 2 diabetes. Diabetes Care 28: 1568 –1573, 2005 Croog SH, Levine S, Testa MA, Brown B, Bulpitt CJ, Jenkins CD, Klerman GL, Williams GH: The effects of antihypertensive therapy on the quality of life. N Engl J Med 314:1657–1664, 1986 Andrade SE, Walker AM, Gottlieb LK, Hollenberg NK, Testa MA, Saperia GM, Platt R: Discontinuation of antihyperlipidemic drugs: do rates reported in clinical trials reflect rates in primary care settings? N Engl J Med 332:1125–1131, 1995
1405
