Health Economics and Compliance of Vials/Syringes Versus Pen Devices: A Review of the Evidence

Abstract

Background:

The goal of this review was to assess the state of the published literature on health economics and compliance of vials/syringes versus pen devices.

Methods:

A literature search was performed using the Embase search engine for publications that linked drug terms (insulin and insulin lispro) to disease terms (insulin-dependent diabetes mellitus, non–insulin-dependent diabetes mellitus) and other terms (accuracy, article, clinical trial, controlled clinical trial, controlled study, cost benefit analysis, drug delivery system, drug dosage form, drug dosage form comparison, drug dose comparison, drug preference, equipment design, force, glycemic control, healthcare cost, human, insulin treatment, needle, patient attitude, patient compliance, patient safety, torque) along with author keywords (Diabetes, Dose accuracy, FlexPen^® [Novo Nordisk, Bagsvaerd, Denmark], Insulin, Next Generation FlexPen).

Results:

The search yielded 39 articles, of which five articles met our study criteria. The focus of the critical outcomes was patient adherence to insulin pen devices versus insulin vials (syringes), hypoglycemic events, emergency department visits due to hypoglycemic events, and costs associated with diabetes and health care. The observation period, mean age of patients, and data sources differed across the studies. The studies indicated that there was an improved adherence with insulin pen devices as opposed to insulin vials (syringes) and that the associated healthcare resource utilization and costs associated with them were found to decrease with the use of pen devices, compared to vials.

Conclusions:

The use of pen devices improves the health economics benefits and adherence to insulin therapy.

Introduction

D iabetes is a metabolic disorder characterized by a failure of cells to utilize glucose as an energy form. The disorder is caused either by a lack of endogenous insulin (type 1 diabetes), by an insufficient response to endogenous insulin (type 2 diabetes), or by aspects of both. The prevalence of diabetes among U.S. residents has been estimated to be 23.6 million (7.8% of the population) in 2007 and appears to be increasing.¹ Initial pharmacotherapy for diabetes depends on the patient's diagnosis. For example, type 1 diabetes is treated with exogenous insulin, whereas type 2 diabetes is initially treated with an oral antihyperglycemic agent, such as metformin, with additional agents being added on as needed, including sulfonylureas, insulin, thiazolidinediones, gliptines, and glucagon-like peptide 1 agonists.²

Untreated diabetes can lead to major and costly microvascular and macrovascular complications, including chronic kidney disease, retinopathy, neuropathy, stroke, myocardial infarction, and death. Pharmacotherapy has been demonstrated to decrease the occurrence of these adverse outcomes.³ However, the full potential clinical benefit of treatment is often not realized because of suboptimal adherence. The conventional mode of administering insulin has been through a vial (syringe) via the subcutaneous route. In 1987, Novo Nordisk (Bagsvaerd, Denmark) was the first to launch the pen devices for insulin therapy, which encompasses both the cartridge and syringe in one unit. Pen devices may either be reusable (by means of a changeable insulin cartridge) or disposable.⁴ Advantages of insulin pen devices compared to vials/syringes include (1) ease of use, (2) patients' preference, (3) dosing accuracy, (4) improved adherence, and (5) reduced healthcare resource utilization and costs.^4
–6

Numerous studies have shown that patients exhibit low levels of adherence with prescribed insulin and oral antidiabetes regimens (ranging from 36% to 96%).⁷ The impact of poor adherence on outcomes in diabetes has not previously been well quantified.

Compliance has been defined by the International Society for Pharmacoeconomics and Outcomes Research as the extent to which a patient acts in accordance with the prescribed interval and dose of a dosing regimen, and a related term, persistence, is defined as the duration of time from initiation to discontinuation of therapy.⁸ The term adherence may be considered an umbrella term that encompasses both concepts. There are numerous particular methods for estimating patient adherence, including directly asking patients about their medication-taking behavior, conducting pill counts, and analyzing refill data.

The goal of this review was to assess the state of the published literature on health economics (HE) and compliance of vials/syringes versus pen devices. Quantifying the relationship between patient adherence and outcomes is a difficult task in any disease state, particularly in observational studies. Although reason dictates that the benefits of taking medications cannot be realized if patients do not take them, we cannot say for certain that the clinical outcomes demonstrated by patients who take their medications are entirely attributable to the drugs. This is because patients who take their medications may also be more likely to adhere to other health-promoting behaviors such as improving their diets, exercising, smoking cessation, preventive health screenings, and immunizations. Caution is imperative when interpreting results of observational studies that link patient adherence to improved outcomes. The benefits of taking medications should not be discounted, but differences in the level of certain undetermined for patient characteristics (e.g., motivation, self-efficacy, etc.) may affect the generalizability of the findings of any individual study. In light of this difficulty, we planned to pay particular attention to any well-conducted observational studies in which the minimum standard sources of confounding were controlled for (patient characteristics, comorbid conditions, and disease severity) in addition to indicators of other health-seeking behaviors.

Methods

We searched both Embase and PubMed on November 30, 2009 for publications that linked drug terms (insulin and insulin lispro) to disease terms (insulin-dependent diabetes mellitus, non–insulin-dependent diabetes mellitus) and other terms (accuracy, article, clinical trial, controlled clinical trial, controlled study, cost benefit analysis, drug delivery system, drug dosage form, drug dosage form comparison, drug dose comparison, drug preference, equipment design, force, glycemic control, healthcare cost, human, insulin treatment, needle, patient attitude, patient compliance, patient safety, torque) along with author keywords (Diabetes, Dose accuracy, FlexPen^® [Novo Nordisk], Insulin, Next Generation FlexPen). Only Embase search results were retrieved as they covered the PubMed search results. All study design types were considered eligible, including randomized controlled trials, meta-analyses, and observational studies (retrospective and prospective studies with cohort, cross-sectional, and case-control designs). Eligible patient populations included adults and children receiving any drug treatment for the management of either type 1 or type 2 diabetes. The search was not restricted to any particular time span or language, but after the initial search, articles were excluded that were not written in English, that did not have an abstract available, or that had publication type variables indicating that they were review articles, editorials, or letters.

Embase citations were downloaded into a word file to facilitate abstract screening and documentation of findings. Three stages of review were planned: stage I, abstract screening; stage II, article screening; and stage III, data extraction. In the first stage (abstract screening), publications were passed to the next level if reviewers could rule in that they met two criteria: (1) empiric data on some measure of HE and/or patient adherence to diabetes pharmacotherapy were analyzed, and (2) the methods for measuring HE and/or patient adherence were described in the article. Publications not excluded at stage I were retrieved, at which time reviewers rescreened the articles for the presence of the first two criteria, listed above. Articles that were found to meet both criteria ultimately progressed to stage III data extraction and were included in the final review.

Results

Articles

The Embase search retrieved 39 articles. Reviewers then screened abstracts for all 39 articles, and 34 were excluded because they did not meet both stage I criteria. The five articles that remained progressed to stage II screening. None of the articles was eliminated. A total of five articles met all of the criteria, underwent data extraction, and were included in the review. Table 1 contains a summary of the characteristics of all studies included.

Table 1.

Characteristics of Studies Included in This Review

PMID	Source data	Observation period	Study design	Mean age (years)	Diabetes type	n	Data sources	What was studied	Outcome types
18593681⁶	Journal	July 2005–June 2006	Randomized controlled trial, noninferiority	Insulin pen group: 57.5 years Vial/syringe group: 57.1 years	Type 1 or type 2	75 total Insulin pen group: n = 35 Vial/syringe group: n = 40	Patients admitted to medical-surgical units during the observation period	Patient satisfaction, hypoglycemic/hyperglycemic events, postdischarge insulin use, hospital costs	Self-administered survey results, telephone survey, hypoglycemic/hyperglycemic event rate, cost-saving analysis
18046929⁹	Journal	September 24, 2001–July 18, 2006; patients were followed for a minimum of 2 years	Observational: Comparison Group 1 compared patients switched from vial/syringe to pen vs. vial/syringe; Comparison Group 2 compared patients initiated on pens vs. vial/syringe	Comparison Group 1: pen group, 40.8 years; vial/syringe group, 43.1 years Comparison Group 2: pen group, 52.5 years; vial/syringe group, 59.9 years	Type 2	Comparison Group 1: pen group, n = 560; vial/syringe group, n = 9,988; pen group was then matched using propensity score (both groups n = 560) Comparison Group 2: pen group, n = 168; vial/syringe group, n = 1,162	North Carolina Medicaid program patient claims database	Adherence, healthcare cost	MPR, total healthcare costs (excluding drug costs), total hospital cost, total diabetes-relates cost, total outpatient costs
17594200¹⁰	Journal	Patient data 6 months prior to biphasic insulin aspart 70/30 pen use and followed for at least 2 years after pen started. Patients included if they began using a pen July 1, 2001–December 31, 2002. Data collected January 1, 2001–April 30, 2005	Retrospective, longitudinal observational analysis	45.1	Type 2	486	PharMetrics national database (57 insurance companies)	Adherence, hypoglycemic events, healthcare costs. The cohort included patients who had been on vial and syringe insulin and then were initiated on an insulin pen (NovoLog Mix 70/30 FlexPen).	MPR, hypoglycemic events, and payments made by third-party payer to providers (physician visit, emergency department, hospitalization, outpatient visit, pharmacy costs)
17157128¹¹	Journal	Patient data 6 months prior to insulin analog pen use and followed for at least 2 years after pen started. Patients included if they began using a pen July 1, 2001–December 31, 2002. Data collected January 1, 2001–April 30, 2005	Retrospective, longitudinal observational analysis	45.4	Type 2	1156	PharMetrics national database (57 insurance companies)	Adherence, hypoglycemic events, relationship between hypoglycemic events and adherence, and associated healthcare costs. The cohort consisted of patients receiving human insulin vial and syringe or analog insulin and syringe: 670 patients were started on the insulin aspart pen, and 486 were started on the biphasic insulin aspart 70/30 pen.	MPR, hypoglycemic events, and payments made by third-party payer to providers (physician visit, emergency department, hospitalization, outpatient visit, pharmacy costs)
9663364¹²	Journal	Novolin prefilled syringe: patients were surveyed at initiation of using the pen and 3 weeks later. NovoPen 1.5: patients were surveyed after using the pen for 3 weeks	Survey	Novolin prefilled syringe: <35 years (19%), 36–55 years (37%), 56–65 years (21%), >66 years (24%) NovoPen 1.5: no age data collected	Type 1 and type 2	Novolin prefilled syringe: n = 803 NovoPen 1.5: n = 507	Novolin prefilled syringe: patients surveyed in 24 U.S. centers NovoPen 1.5: patients surveyed in 64 U.S. centers	Insulin history, current insulin use, compliance with insulin, patient comfort, impact on lifestyle	All measures were assessed with patients' survey responses

NovoLog^® 70/30, Novolin^®, and NovoPen^® are manufactured by Novo Nordisk.

The five studies that met the inclusion criteria for this review were observational studies except for the study conducted by Davis et al.,⁶ which was a randomized controlled noninferiority trial. Two^6,9 of the five studies controlled for potential bias, whereas the remaining articles are associated with selection bias (except for the study by Lee et al.,¹¹ which did not account for bias). Critical outcomes that the five studies focused on were patient adherence to insulin pen devices versus insulin vials (syringes), hypoglycemic events, emergency department visits due to hypoglycemic events, and costs associated with diabetes and health care.

Table 2 gives the overall summary of the association between adherence and vial/syringe utilization for all included studies.

Table 2.

Association Between Adherence and Vial/Syringe Utilization

Reference	Controlled for sources of bias	Compliance measure	Compliance findings	Effect of compliance on glucose control
Davis et al.⁶	Yes	Patient self-administration rates of insulin while in hospital	More patients with a pen self-administered at least one dose compared to the vial/syringe group (77.1% vs. 12.5%, P = 0.001).	No significant differences in hyperglycemic/hypoglycemic events between the two groups. Total insulin costs per patient were higher in the pen group compared to the vial/syringe group ($159.39 ± $91.41 vs. $108.04 ± $62.30). When costs were adjusted to compare equivalent amounts of insulin dispensed, the pen group found a cost savings of $36 per patient.
		Postdischarge use of pen vs. vial/syringe	Postdischarge: 12 (35%) patients in the pen group continued using pens, while 14 (41%) switched to vial/syringe; 23 (58%) of patients in the vial/syringe group continued using vial/syringe, while 4 (10%) switched to pens
Pawaskar et al.⁹	Yes	MPR	Comparison Group 1: patients in the pen group had lower diabetes-related compliance (45% vs. 56%, P < 0.05). Overall medication compliance was higher in the pen group (92% vs. 90%, P < 0.05).	Comparison Group 1: no significant difference in the total healthcare costs (excluded drug costs) between the pen and vial/syringe groups ($11,476.42 vs. $10,755.31)
			Comparison Group 2: There was no significant difference in compliance between the pen and vial/syringe groups (53% vs. 50%).	Comparison Group 2: the pen group had significantly less total healthcare costs (excluding drug costs) ($14,857.42 vs. $31,764.78, P < 0.05). This was due to lower costs found in hospital costs ($1,195.93 vs. $4,965.31, P < 0.05), diabetes-related costs ($7,324.37 vs. $13,762.21, P < 0.05), and outpatient costs ($7,795.98 vs. $13,103.51, P < 0.05).
Cobden et al.¹⁰	Selection bias: controlled for some patient demographics; no data included on income or race	MPR	An increased MPR was associated with the pen (68 ± 31% vs. 59 ± 30%, P < 0.01). The percentage of patients with >80% MPR was higher in the pen group compared to the vial/syringe group (56.2% vs. 35.4%, P < 0.01).	Reduction in risk of hypoglycemic event (OR 0.40, 95% CI 0.27–0.61, P < 0.05)
				Reduction in ED visits due to hypoglycemia (OR 0.36, 95% CI 0.16–0.84, P < 0.05)
				All-cause healthcare costs decreased ($16,004 vs. $14,256, P < 0.01).
				Cost of insulin increased $339 ($2,278 vs. $2617, P < 0.01).
				Cost of oral antidiabetes drugs decreased $772 ($1,790 vs. $1,018, P < 0.01).
Lee et al.¹¹	—	MPR	MPR was higher in the pen group compared to the vial/syringe group (69% vs. 62%, P < 0.01). The percentage of patients with MPR >80% was greater in the pen group compared to the vial/syringe group (54.6% vs 36.1%, P < 0.01).	Reduction in risk of hypoglycemic event (OR 0.50, 95% CI 0.37–0.68, P < 0.05)
				Reduction in ED visits due to hypoglycemia (OR 0.44, 95% CI 0.21–0.92, P < 0.05)
				All-cause healthcare costs decreased ($16,359 vs $14,769, P < 0.01).
				Diabetes-associated pharmacy costs decreased $629 ($3,690 vs. $3,061, P < 0.01).
Graff and McClanahan¹²	Patients were notified of NovoNordisk sponsorship and investigation site honorarium prior to the survey. Patients were approached for the survey during a provider visit.	Patient-reported compliance	Novolin prefilled syringe: 85% of patients reported not missing a dose in the 3-week period compared to 72% of patients saying they did not miss a dose with vial/syringe insulin.	N/A
			NovoPen 1.5: N/A

CI, confidence interval; ED, emergency department; N/A, not available; OR, odds ratio.

Measures of adherence

Of the five studies included in the review, three studies^9
–11 used the medication possession ratio (MPR) as a measure of adherence. The remaining studies used the patient self-administered rates of insulin while in the hospital, postdischarge use of pen versus vial/syringe,⁶ and patient-reported compliance as the measures of adherence.¹² The MPR was defined as the days of supply dispensed divided by the sum of the number of days between the first fill and last fill. As shown in Table 2, except for one study,⁶ all the studies found a higher compliance with pen devices versus vial/syringe. In those studies^9
–11 that included MPR as a measure of adherence, MPR ranged from 53% to 69% for pen devices versus 50% to 62% for vials/syringes (included just Comparison Group 2 in the study of Pawaskar et al.⁹). In the studies conducted by Cobden et al.¹⁰ and Lee et al.,¹¹ the percentages of patients with MPR >80% were, respectively, 56.2% and 54.6% for pen devices versus 35.4% and 36.1% for vials/syringes. In addition, these studies^8,9 reported the compliance and percentage of patients with MPR >80% to be significantly different (P < 0.01) between pen devices and vials/syringes. Although there was no significant difference reported in compliance for pen devices versus vials/syringes for Comparison Group 2 in the study by Pawaskar et al.,⁹ the overall medication compliance was discovered to be significantly higher (P < 0.05) in Comparison Group 1.

Measures of HE

Among the five studies considered for the review, four studies^6,9
–11 evaluated the economic benefits and healthcare utilizations between pen devices versus vials/syringes. The studies by Cobden et al.¹⁰ and Lee et al.¹¹ reported a significant reduction (P < 0.05) in hypoglycemic events (odds ratio, 0.40 and 0.50, respectively) and emergency department visits due to hypoglycemic events (odds ratio, 0.36 and 0.44, respectively) for pen devices as opposed to vials/syringes. Also, these two studies showed a significant decrease (P < 0.01) in all-cause healthcare costs for pen devices ($16,004 and $16,359, respectively) compared to vials/syringes ($14,256 and $14,769, respectively). The study by Davis et al.⁶ calculated a total insulin cost saving of $36 per patient after adjusting the amounts of insulin dispensed for pen devices versus vials/syringes. Similarly, the studies by Pawaskar et al.⁹ and Cobden et al.¹⁰ demonstrated a significant reduction (P < 0.05) in diabetes-related costs and a significant decrease (P < 0.01) in oral antidiabetes drug costs. In addition, the study by Cobden et al.¹⁰ found a significant increase (P < 0.01) in insulin costs. Furthermore, the study by Pawaskar et al.⁹ reported no significant difference in total healthcare cost for Comparison Group 1 and a significant difference (P < 0.05) in total healthcare costs for Comparison Group 2.

Discussion and Conclusions

Insulin is the most effective medication to control glucose. “Early” rather than “later” use in type 2 diabetes has been advocated in hopes of decreasing hyperglycemia and its related complications. Certainly all persons with type 1 diabetes and many persons with type 2 diabetes require insulin. Overall, it is estimated that about 27% of all persons with diabetes use insulin.¹³ Although it is well recognized that the most effective course of action is to achieve a target hemoglobin A1C of <7%, 45% of all persons do not reach this target.¹⁴

Although insulin administration technology has evolved into very convenient and more acceptable dosage forms, a large number of persons omit insulin injections. A recent survey of 502 individuals with diabetes found that 57% reported omitting or missing insulin injections, and, surprisingly, 20% reported regular insulin injection omission.¹⁵ In this survey, among other reasons, persons with type 2 diabetes reported embarrassment or pain as reasons for insulin omission. Hence, insulin pen devices may be an attractive alternative to some of these persons to help diminish injection omission. Persons with decreased manual dexterity may also be candidates for insulin pen. For individuals who use lower doses of insulin, pens may provide less waste than larger-dose vials, due to expiration date issues.

An important message is that providers should determine whether patients who are not at goal hemoglobin A1C should explore whether the individual is actually using his or her prescribed insulin and whether he or she is omitting insulin doses. This is especially problematic in persons with type 2 diabetes who may still have residual insulin production and are not using insulin consistently. Providers should also then explore reasons for patients' reluctance to use insulin. If embarrassment is a factor, then a pen is a more discreet device to administer insulin in the work place or in social situations. “Psychological insulin resistance” has been reported to be a factor and is basically the term used to describe insulin avoidance by patients.¹⁶ Other related factors include a fatalistic belief that having to use insulin signifies a person's failure with his or her therapy, as shown in the Diabetes Attitudes Wishes and Needs study.¹⁷ Other factors have included injection pain, the belief that insulin use is the “last resort,” the belief that weight gain may occur, and that it is inconvenient, more time consuming, and is more costly.¹⁸ Perhaps, a pen device may help to diminish psychological resistance and other factors associated with reluctance to use insulin.

For all of the convenience that pen devices may offer, some individuals may wish to continue vials and syringes because of habit and possibly because they may not be covered by their health insurance programs. However, pen devices have been shown to have increased patient acceptance and lower associated costs as demonstrated in this article. An important message should be to increase insulin acceptability so that it is prescribed earlier and used more consistently by patients to reach goal hemoglobin A1C in hopes of diminishing diabetes-related complications.

Footnotes

Acknowledgments

We would like to thank Michael Alwan for his kind assistance in helping us search the literature.

Author Disclosure Statement

The authors have no commercial associations or competing financial interests that might create a conflict of interest in connection with this manuscript. C.V.A. was a paid consultant for a previous literature review for Novo Nordisk on compliance and health economics of pens versus syringes A–C. In addition, C.V.A. has participated in a University of Utah study grant from Novartis Pharmaceuticals where he is the principal investigator in a diabetes-related study.

References

National Diabetes Information Clearinghouse (NDIC). 2007. National Diabetes Statistics. Updated June 2008. http://diabetes.niddk.nih.gov/DM/PUBS/statistics/#what. 2009 December 1.

Nathan

, Buse

, Davidson

, Ferrannini

, Holman

, Sherwin

, Zinman

. American Diabetes Association; European Association for Study of Diabetes. Medical management of hyperglycemia in type 2 diabetes: a consensus algorithm for the initiation and adjustment of therapy: a consensus statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care, 2009; 32:193–203.

Saenz

, Fernandez-Esteban

, Mataix

, Ausejo

, Roque

, Moher

. Metformin monotherapy for type 2 diabetes mellitus. Cochrane Database Syst Rev, 2005; 3:CD002966.

Kroon

. Overview of insulin delivery pen devices. J Am Pharm Assoc (2003), 2009; Aug 18:118–131.

Molife

, Lee

, Shi

, Sawhney

, Lenox

. Assessment of patient-reported outcomes of insulin pen devices versus conventional vial and syringe. Diabetes Technol Ther, 2009; 11:529–538.

Davis

, Christensen

, Nystrom

, Foral

, Destache

. Patient satisfaction and costs associated with insulin administered by pen device or syringe during hospitalization. Am J Health Syst Pharm, 2008; 65:1347–1357.

Cramer

, Roy

, Burrell

, Fairchild

, Fuldeore

, Ollendorf

, Wong

. Medication compliance and persistence: terminology and definitions. Value Health, 2008; 11:44–47.

Cramer

. A systematic review of adherence with medications for diabetes. Diabetes Care, 2004; 27:1218–1224.

Pawaskar

, Camacho

, Anderson

, Cobden

, Joshi

, Balkrishnan

. Health care costs and medication adherence associated with initiation of insulin pen therapy in Medicaid-enrolled patients with type 2 diabetes: a retrospective database analysis. Clin Ther, 2007; 29:1294–1305.

10.

Cobden

, Lee

, Balu

, Joshi

, Pashos

. Health outcomes and economic impact of therapy conversion to a biphasic insulin analog pen among privately insured patients with type 2 diabetes mellitus. Pharmacotherapy, 2007; 27:948–962.

11.

Lee

, Balu

, Cobden

, Joshi

, Pashos

. Medication adherence and the associated health-economic impact among patients with type 2 diabetes mellitus converting to insulin pen therapy: an analysis of third-party managed care claims data. Clin Ther, 2006; 28:1712–1725.

12.

Graff

, McClanahan

. Assessment by patients with diabetes mellitus of two insulin pen delivery systems versus a vial and syringe. Clin Ther, 1998; 20:486–496.

13.

Centers for Disease Control and Prevention, Department of Health and Human Services. Diabetes Data & Trends. www.cdc.gov/diabetes/statistics/meduse/dtltable2.htm. 2010 February 28.

14.

Hoerger

, Segel

, Gregg

, Saaddine

. Is glycemic control improving in U.S. adults? Diabetes Care, 2008; 31:81–86.

15.

Peyrot

, Rubin

, Kruger

, Travis

. Correlates of insulin injection omission. Diabetes Care, 2010; 33:240–245.

16.

Polonsky

, Fisher

, Guzman

, Villa-Caballero

, Edelman

. Psychological insulin resistance in patients with type 2 diabetes; the scope of the problem. Diabetes Care, 2005; 28:2543–2545.

17.

Peyrot

, Rubin

, Lauritzin

, Snoek

, Matthews

, Skovlund

. Psychosocial problems and barriers to improved diabetes management: results of the Cross-National Diabetes Attitudes, Wishes and Needs (DAWN) Study. Diabet Med, 2005; 22:1379–1385.

18.

Meece

. Dispelling myths and removing barriers about insulin in type 2 diabetes. Diabetes Educ, 2006; 32,1 Suppl:9S–18S.