Diabetic peripheral neuropathy: resource utilization and burden of illness

Abstract

Introduction:

Diabetic peripheral neuropathy (DPN) is a debilitating complication of diabetes and accounts for significant morbidity by pre-disposing the foot to ulceration and lower extremity amputation. Using a large US commercial claims database, this study analyzes the drug class usage and co-morbidities associated with DPN as well as estimates the associated economic burden.

Methods:

Patients older than 18 and diagnosed with DPN were followed longitudinally for 2 years pre- and post-diagnosis date. Patients were analyzed for age, gender, hospital visits, ER and doctor’s office visits, pharmacy claims, co-morbidities, and drug classes prescribed pre- and post-DPN diagnosis. The economic impact post-diagnosis of DPN was compared to the patients’ pre-diagnosis resource use.

Results:

In total, 10,982 incident DPN patients were identified, with a median age of 61 years, and an equal gender distribution. Post-DPN diagnosis, there was a 20% increase in the number of patients visiting hospitals and a 46% increase in the number of visits to hospitals. Further, there was a 46% increase in the annual cost per patient associated with visits to the hospitals, emergency room (ER), doctor’s office, and pharmacy claims. As per the analysis presented in this study, increase in the number of visits, cost per visit, and number of patients visiting hospitals, ER and doctor’s offices added up to a 46% increase in aggregated cost associated with Medical Resource Utilization (MRU) owing to DPN, with the highest increase (60%) in costs associated with hospitalization of patients with DPN.

Conclusion:

This study highlights the high economic burden associated with DPN. The results indicate that resource use significantly increases post-diagnosis of DPN, which leads to an increase in costs for payers. A noticeable proportion of patients with DPN had a pain co-diagnosis signifying the need for treatments that can effectively manage painful DPN.

Keywords::

• Diabetic peripheral neuropathy
• DPN
• Medical resource utilization
• Economic burden
• Claims database

Introduction

Diabetic peripheral neuropathy, or peripheral neuropathic pain in patients suffering from diabetes, is defined by the International Association for the Study of Pain as ‘pain arising as a direct consequence of abnormalities in the peripheral somatosensory system in people with diabetes’ (Ref. 1: p. 1630, Ref. 2: p. 2286).1,2 Peripheral neuropathy is the most common co-morbidity associated with diabetes mellitus3. Up to 50% of patients suffering from diabetes have peripheral neuropathy associated with the condition4–10. DPN affects the lower limbs first and then moves up to the upper limbs in a ‘stocking and glove’ pattern of distribution9. In some patients, occurrence of hypoesthesia is symptomatic of DPN; however, in half the cases, DPN is asymptomatic and such patients are at risk of injury to their feet and developing foot ulcers4,11, making DPN a major factor associated with foot ulceration, amputation, and death4,5,12,13. In symptomatic patients, symptoms of DPN can be positive (prickling, pain) or negative (loss of sensation and strength)14,15.

The American Academy of Family Physicians algorithm for management of DPN is shown in Figure 116. On top of analgesic treatment and treatments for management of diabetes, therapies for DPN management include tri-cyclic antidepressants (TCAs), serotonin norepinephrine re-uptake inhibitors (SNRIs) such as duloxetine, and anticonvulsants such as pregabalin and gabapentin, the use of which in painful DPN is supported by clinical evidence17,18. Although there are a number of treatment options available for patients with symptomatic DPN, pain management is often inadequate8, either due to lack of efficacy or due to adverse events associated with the use of such medications, the latter leading to curtailment of dose and, thereby, limiting their effectiveness.

Figure 1. Algorithm for management of symptomatic DPN16. Reproduced from Lindsay TJ, et al. Treating diabetic peripheral neuropathic pain. American Family Physician 2010;82(2):151–158, with permission of the American Academy of Family Physicians.

Untreated or under-treated chronic painful DPN has been shown to have a debilitating impact on quality-of-life8,19,20, as it results in sleep disturbance, anxiety, and depression8,21–24. Gore et al.8 assessed the burden associated with DPN and published a study based on patients with DPN recruited by community-based physicians. In this study, patients with DPN were found to be suffering from co-morbidities such as depression (27.8%), anxiety (26.7%), and peripheral vascular disease (24.3%), at rates greater than the normal population (National Comorbidity Survey Replication study reported 12 month period prevalence of any mood disorder including depression at 9.5%, and any anxiety at 18.1%, while Hiatt et al.25 have reported peripheral arterial disease to be affecting 12–14% of the general population). Additionally, other co-morbidities such as microvascular complications of diabetes including retinopathy (13.7%) and nephropathy (17.3%), and chronic musculoskeletal pain conditions such as osteoarthritis (33.7%) and nociceptive low back pain (26.7%) were found to be common among patients with DPN, along with other chronic neuropathic pain conditions such as carpal tunnel syndrome (13.7%) and low back pain with neuropathic involvement (13.3%).

These co-morbidities associated with DPN, along with polypharmacy and medical resource utilization, generate a considerable economic burden. Annual cost associated with DPN in the US alone was estimated to be $10.3 billion in 20034. Direct medical costs associated with patients with DPN are 5-times higher than the for patients suffering from diabetes mellitus alone10, and DPN contributes 27% to direct medical cost of diabetes4.

The objective of this study was to use real world data to characterize patients with DPN, specifically their demographics, co-morbidities, and treatment patterns, and define medical resource utilization (MRU) before and after diagnosis of DPN, and, thus, contribute to existing literature by confirming and quantifying the unmet need and high burden associated with DPN.

Methods

Data source

This retrospective study is based on a large US commercial database, the LifeLink (formerly PharMetrics) Health Plan Claims Database (IMS, Inc., Watertown, MA). The LifeLink database contains adjudicated medical and pharmaceutical claims data from more than 98 managed care organizations throughout the US (Midwest 34%, Northeast 22%, South 29%, West 15%). The database consists of more than 61 million patients and ∼16 million covered lives per year and includes information on inpatient and outpatient diagnoses (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] format) and procedures (Current Procedure Terminology, 4th Edition and Healthcare Common Procedural Coding System formats), as well as demographic variables, provider specialty, prescription records, National Drug Code numbers, days’ supply of prescriptions, and quantity dispensed of prescriptions. The database also provides costs of Medical Services such as hospitalization, emergency room (ER) visits, physician visits, procedures, home care, and medical tests26.

Study sample

All patients with one or more healthcare encounters with an associated diagnosis of DPN (ICD-9-CM code 250.6x and 357.2) starting January 1, 2009 through December 31, 2009 were identified. A patient’s first DPN diagnosis in 2009 was used as the baseline date (‘Baseline Date’) for each patient. Each patient was then historically evaluated to check for any previous diagnosis of DPN, for at least a period of 2 years, or where data was available for longer, for the entire history of that patient. Patients with no prior diagnosis of DPN were termed ‘incident DPN’ patients. Additional eligibility conditions that were considered before patients were included in the study were that the year of birth for the patient should have been 1992 or earlier (at least 18 years old), and that the patient had continuous enrolment of at least 24 months prior and 24 months post-baseline date (n = 10,982, Figure 2).

Figure 2. Algorithm used to identify incident patients with DPN using LifeLink database.

Pain co-morbidities and medical resource utilization (MRU)

Pain co-morbidities, specifically post-herpetic neuralgia, back and neck pain, and osteoarthritis, along with co-morbidities associated with pain such as depression, sleep disorder, and other relevant pain co-morbidities (including atypical facial pain, phantom limb pain, trigeminal neuralgia, reflex sympathetic dystrophy, nerve root and plexus disorders, other disorders of soft tissues, and inflammatory and toxic neuropathy) were analyzed at the patient level to determine if patients after a diagnosis of DPN had a greater occurrence of these co-morbidities.

Analyses were also conducted to examine MRU both in terms of costs (on the basis of the ‘Allowed’ field in the LifeLink data) and number of visits, including hospitalizations, ER visits, office visits and prescription filling, to determine whether DPN diagnosis was associated with greater healthcare utilization and costs. Hospital, ER, and office visit costs and pharmacy costs were added to derive and compare total costs associated with incident patients pre- and post-DPN diagnosis. Percentage (%) class possession days for a drug was calculated as number of days a drug was prescribed out of 730 study days (2 years) pre- and post-DPN diagnosis.

Statistical analysis

The t-test (two-tailed) was used to analyze the difference in parameters before and after being diagnosed with DPN. The parameters that were compared are mean occurrences of co-morbidities, cost per patient, cost per visit, and number of visits per patient across different MRU categories. The Z-test was used to compare occurrence of different co-morbidities and drug/drug class usage, between pre- and post-DPN diagnosis. The level of significance was fixed at 5%. The Bonferroni correction was applied to adjust for multiple comparisons and only values less than 0.00156 (p < 0.05/n, n = 32) were considered statistically significant

Results

Characteristics of patients with DPN

The included study sample of 10,982 patients (Table 1) comprised similar proportions of males (49.7%) and females (50.3%). Their mean age was 62 years (median 61 years), with the mean age of male patients being 62 years (median = 61 years) and female patients 61 years (median age = 60 years). While 54.7% of these incident patients were between 45–64 years old, 27.9% were between 65–80 years old (Table 1). Out of 10,982 incident patients, only 57.2% of patients were being treated with any one of the following classes: opioids, anti-convulsants, anti-depressants, and NSAIDs in the baseline month (the calendar month of the Baseline Date), which went up to 68% over the 2-year follow-up period post-DPN diagnosis.

Table 1. Characteristics of incident patients with DPN included in the analysis.

	Total
Number of patients	10,982
Male (%)	49.7
Mean age (years)	61.6
Group distribution (age in years)
18–44 (%)	8.7
45–64 (%)	54.7
65–80 (%)	27.9
80+ (%)	8.7
Patients treated* at baseline (%)	57.3
Patients treated* over 2-year -p (%)	68.0

*Treated refers to treatment with drugs analyzed for usage in Table 6.

Medical resource utilization (MRU) and associated costs

The number of patients visiting the hospitals and ER was 20% and 8% higher, respectively, after their DPN diagnosis compared to pre-diagnosis (Table 2). A similar trend was seen in office visits and pharmacy claims. The number of office visits was 24% higher (Table 3) and number of pharmacy claims 20% higher (Table 3) post-DPN diagnosis. The mean number of hospital visits per patient increased significantly from 1.8 to 2.2 post-DPN diagnosis, and the mean number of ER visits per patient from 2.3 to 2.5, resulting in an overall increase in the number of hospital visits of these patients by 45.6% and ER visits by 20.3% post-DPN diagnosis (Table 3). This increase in number of visits, as well as an increase in cost per visit of 9.2% for hospital visits and 21.5% for ER visits (Table 4), contribute to an increase in total costs associated with hospitalization (59% increase) and ER (47% increase) (Figure 3). Similarly, an ∼20% increase in number of office visits and number of pharmacy claims, along with an 11.6% increase in the cost per office visit and 8.3% increase in the cost per pharmacy claim (Table 4) post-DPN diagnosis, contribute to an increase in the total direct cost of office visits and pharmacy claims by 39% and 30%, respectively (Figure 3). Combined cost of hospital visits, ER visits, office visits, and treatments lead to a 46% increase in the aggregated costs associated with MRU post-DPN diagnosis (Figure 3).

Figure 3. Total cost ($) for different types of visits. ^Treatments only include pharmacy claims.

Table 2. Number and percentage of patients that make hospital, ER, and office visits.

	Pre-diagnosis period		Post-diagnosis period
	# patients	% patients	# patients	% patients	% Increase (post DPN diagnosis)	p-Value H0: Mpre = Mpost H1: Mpre < > Mpost
Hospital visits	3247	29.6%	3884	35.4%	19.6%	<0.001
ER visits	4548	41.4%	4920	44.8%	8.2%	<0.001
Office visits	10,871	99.0%	10,950	99.7%	0.7%	<0.001
Treatments*	10,330	94.1%	10,456	95.2%	1.2%	<0.001

*Treatments only include pharmacy claims.

H₀, Null Hypothesis, H₁, Alternative Hypothesis; M_pre, Percentage of patients that have a relevant claim pre-DPN diagnosis, M_post, Percentage of patients that have a relevant claim post-DPN diagnosis.

Table 3. Count, mean, median, and standard deviation of hospital, ER, and office visits.

	Pre-diagnosis period			Post-diagnosis period
	Total visits	Mean1^a	Mean2^a	Total visits	Mean1^a	Mean2^a	% Increase in visits/claims (post-DPN diagnosis)	p-value H0: M1pre = M1post H1: M1pre < > M1post	p-Value H0: M2pre = M2post H1: M2pre < > M2post
Hospital visits	5818	1.8	0.53	8470	2.2	0.77	45.6%	<0.001	<0.001
ER visits	10,285	2.3	0.94	12,373	2.5	1.13	20.3%	<0.001	<0.001
Office visits	262,614	24.2	23.91	325,945	29.8	29.68	24.1%	<0.001	<0.001
Treatments^b	502,499	48.6	45.76	602,685	57.6	54.88	19.9%	<0.001	<0.001

^aMean1 = Number of visits per patient with any relevant visit/claim; Mean2 = Number of visits per all patients.

^bTreatments only include pharmacy claims.

H₀, Null Hypothesis; H₁, Alternative Hypothesis; M1_pre, Mean1 pre-DPN diagnosis, M1_post, Mean1 post-DPN diagnosis; M2_pre, Mean2 pre-DPN diagnosis; M2_post, Mean2 post-DPN diagnosis.

Table 4. Cost per visit and annual cost per patient.

	Pre-diagnosis period		Post-diagnosis period
	Mean cost per visit	Annual cost per patient	Mean cost per visit	Annual cost per patient	% Increase (post DPN diagnosis)	p-value H0: Mpre = Mpost H1: Mpre < > Mpost	p-Value H0: ACpre = ACpost H1: ACpre < > ACpost
Hospital visits	17,677	15,837	19,302	21,046	9.2%	0.022	<0.001
ER visits	729	824	886	1114	21.5%	<0.001	<0.001
Office visits	91	1100	102	1513	11.6%	<0.001	<0.001
Treatments*	148	3606	161	4629	8.3%	0.004	<0.001

*Treatments only include pharmacy claims

H₀, Null Hypothesis, H₁, Alternative Hypothesis; M_pre, Mean cost per visit in pre-DPN diagnosis, M_post, Mean cost per visit in post-DPN diagnosis; AC_pre, Annual cost per patient with relevant visit in pre-DPN diagnosis, AC_post, Annual cost per patient with relevant visit in post-DPN diagnosis.

Increase in the number of visits and cost per visit also contributed to a significant increase (46%) in the average annual cost per patient post-DPN diagnosis ($13,874) (Table 4).

Pain co-morbidity analysis

Pain co-morbidities and pain-associated co-morbidities were analyzed post-DPN diagnosis. There was a 6% increase (Table 5) in the number of patients suffering from various co-morbidities analyzed. In particular, there was a 37% increase in patients suffering from other pain co-morbidities such as atypical facial pain, phantom limb pain, trigeminal neuralgia, reflex sympathetic dystrophy, and nerve root and plexus disorders. There was also a 23% increase in patients suffering from depression and a 16% increase in patients suffering from osteoarthritis in the post-diagnosis period (Table 5).

Table 5. Number and pecentage of patients suffering from co-morbidities.

	Pre-diagnosis period		Post-diagnosis period
	# Patients	% of all patients^a	# Patients	% of all patients	% increase in # of patients (post-DPN diagnosis)	p-Value H0: Ppre = Ppost H1: Ppre < > Ppost
Post-herpetic neuralgia	234	2.1%	251	2.3%	7.3%	0.42
Cancer	1158	10.5%	1318	12.0%	13.8%	<0.001
Osteoarthritis	3170	28.9%	3679	33.5%	16.1%	<0.001
Back & neck pain	2487	22.6%	2765	25.2%	11.2%	<0.001
Low back pain with neuropathic component	3194	29.1%	3509	32.0%	9.9%	<0.001
Other pain co-morbidities^b	1639	14.9%	2246	20.5%	37.0%	<0.001
Other diabetic complications^c	5475	49.9%	6528	59.4%	19.2%	<0.001
Depression	1355	12.3%	1671	15.2%	23.3%	<0.001
Sleep disorder	1883	17.1%	1946	17.7%	3.3%	0.26
Total^d	9079		9620		6.0%	<0.001

^a% for any co-morbidity is calculated by number of patients suffering from the co-morbidity/total no. of patients (10,982).

^bOther pain co-morbidities include atypical facial pain, phantom limb pain, trigeminal neuralgia, reflex sympathetic dystrophy, nerve root and plexus disorders, other disorders of soft tissues, and inflammatory and toxic neuropathy.

^cOther diabetic complications include diabetic retinopathy/cataracts, nephropathy/Dialysis/renal failure/ESRD, peripheral vascular disease, gangrene/diabetic foot/diabetic foot ulcer and other diabetic complications.

^dTotal no. of patients suffering from any of the above-mentioned co-morbidities.

H₀, Null Hypothesis, H₁, Alternative Hypothesis; P_pre, Proportion of patients suffering from a co-morbidity pre-DPN diagnosis, P_post, Proportion of patients suffering from a co-morbidity post-DPN diagnosis.

Drug usage in patients with DPN

In the analysis of various drug classes, viz. opioids, anti-convulsants, anti-depressants, and NSAIDs before and after DPN diagnosis, it emerged that NSAIDs were the most used drugs in analyzed patients irrespective of DPN diagnosis. The number of patients using NSAIDs stayed approximately the same pre- (31%) and post-DPN (30%) diagnosis. However, usage of other drug classes increased post DPN diagnosis (Table 6). The increase in usage was highest for anti-convulsants, especially gabapentin (56%), followed by long acting opioids, duloxetine, and short acting opioids (37%, 36%, and 26%, respectively) (Table 6). Additionally, there was an increase in the percentage of class possession days (as described in the Methods) for all treatments analyzed post-DPN diagnosis, with the highest absolute increase in the class possession days for the two anti-convulsants, pregabalin (10%) and gabapentin (6%).

Table 6. Drug/class usage by number of patients, percentage of patients, and percentage of class possession days.

	Pre-diagnosis period		Post-diagnosis period
Drug/class usage	# of patients	% of patients^a	% of class possession days^b	# of patients	% of patients^a	% of class possession days^b	% Increase in number of patients (post-DPN diagnosis)	p-value H0: Ppre = Ppost H1: Ppre < > Ppost	p-Value H0: CPDpre = CPDpost H1: CPDpre < > CPDpost
Gabapentin	1267	12%	0.34	1977	18%	0.40	56%	<0.001	<0.001
Pregabalin	782	7%	0.27	921	8%	0.37	18%	<0.001	<0.001
Other ACONs	697	6%	0.48	847	8%	0.50	22%	<0.001	0.43
Duloxetin HCL	457	4%	0.44	621	6%	0.48	36%	<0.001	0.21
Other ADEPs	2684	24%	0.59	2941	27%	0.64	10%	<0.001	<0.001
NSAID	3393	31%	0.22	3263	30%	0.23	4%	0.06	0.07
ODPN	2125	19%	0.28	2441	22%	0.31	15%	<0.001	0.008
OP2L	286	3%	0.43	393	4%	0.43	37%	<0.001	0.78
OP2S	1569	14%	0.12	1983	18%	0.14	26%	<0.001	0.04
OP34	4784	44%	0.14	5057	46%	0.17	6%	<0.001	<0.001

^a% of patients is calculated out of total patients (10,982).

^b% class possession days = Sum of days supply for patients that have used this drug-class out of a total of 730 days of potential drug class usage

H₀, Null Hypothesis; H₁, Alternative Hypothesis; P_pre, Proportion of patients taking a drug/class pre-DPN diagnosis, P_post, Proportion of patients taking a drug/class post-DPN diagnosis; CPD_pre, Percentage of class possession days of a drug/class, pre-DPN diagnosis; CPD_post, Percentage of class possession days of a drug/class, post-DPN diagnosis.

ACON, Anticonvulsants; ADEP, Antidepressants; ODPN, Drugs, other than those belonging to the specified classes, typically used for DPN patients (e.g., anti-arrhythmics, benzodiazepines, lidocaine); NSAID, Nonsteroidal Anti-inflammatory Drugs; OP2S, Opioid Schedule 2 Short-Acting; OP2L, Opioid Schedule 2 Long-Acting; OP34, Opioid Schedule ¾.

Discussion

Neuropathy associated with diabetes is a growing healthcare priority due to its impact on patients’ quality-of-life and the additional resource burden as a result of the condition5. This study is an attempt to quantify the burden associated with DPN.

In this study, the mean age of identified DPN cohort was 61 years, which is similar to several other published studies27–30, but is higher than the mean age of 51 years reported by another study by Le et al.10 based on the same database. This difference is likely due to exclusion of all patients co-diagnosed with depression in the study by Le et al.10

Various factors which contribute to costs associated with MRU and, hence, economic burden of DPN such as treatment cost, cost per visit, and total cost associated with visits to the hospitals/ER/doctor’s office were analyzed. This study shows that diagnosis of DPN is associated with a 60% (∼$60 million) increase in the hospitalization costs; of this, $11 million was found to be directly associated with pain owing to DPN diagnosis and $49 million was associated with diagnosis of other pain co-morbidities analyzed. This increase in hospitalization costs is contributed to by an increase in the number of patients visiting hospitals and a higher rate of hospital visits post-DPN diagnosis. In further analysis of 3884 patients who visited hospitals post-DPN diagnosis, it was found that 1148 patients visited hospitals owing to pain associated with DPN and 2545 visited hospitals owing to a pain diagnosis associated with the co-morbidities analyzed (Table 5). Similarly, out of 8470 hospital visits post-DPN diagnosis, 1486 visits were found to have a DPN claim associated with the visit and 4651 had a claim associated with the pain co-morbidities analyzed, contributing to a 46% increase in the number of hospital visits and a 60% increase in the costs associated with hospitalization post-DPN diagnosis. In this study, annual cost per patient post-DPN diagnosis was calculated to be $13,874, which was 46% more than the annual cost per patient pre-DPN diagnosis ($9498). This cost ($13,874) is in agreement with the previously reported cost of $14,062 for patients with DPN31, and is 3–5-fold higher than the annual cost per patient reported for pain conditions such as OA ($2600–$4833)32,33. These results underscore the chronic nature of pain in DPN and its contribution towards a heavy patient burden associated with the condition. While we cannot attribute causation (i.e. the increase in resource use post-diagnosis was solely due to pain), the only major change in all of these patients was the fact that they had a DPN diagnosis leading us to believe that most of the increased resource usage is attributed to the DPN diagnosis.

Inadequate pain management could be one of the reasons that contribute to recurrent visits by patients to the hospital. A high number of patients with DPN (42%) were found to be untreated with any of the following classes: opioids, anti-convulsants, anti-depressants, and NSAIDs at the baseline month which decreases to 32% over the 2-year follow-up period. For treatment, NSAIDs was found to be the most prescribed drug class, but there was no increase in its usage post-DPN diagnosis. This is consistent with the fact that NSAIDs are largely ineffective in the treatment of DPN8. The database only captures prescription drug use, so it is difficult to say if the patients were using other over-the-counter NSAIDs to manage their pain, unlike other classes of drugs which are mostly prescription-based and, hence, allowed a more accurate analysis of their usage.

Gabapentin, an anticonvulsant which is an emerging first line treatment for DPN5, showed the highest increase in usage (56%) post-DPN diagnosis. However, in terms of total usage, only 18% of patients with DPN were prescribed gabapentin compared to NSAIDs (30%). Among opioids, Schedule III/IV opioids were the most prescribed (46%), followed by Schedule II short acting opioids (18%) and Schedule II long acting opioids (4%). Although several clinical trials have established a role for opioids in the treatment of different types of neuropathic pain34–37, it is interesting to note that tapentadol extended release, a centrally acting analgesic which is a mu-opioid agonist and norepinephrine re-uptake inhibitor, is the only opioid that has been approved in the US by the Food and Drug Administration for the management of neuropathic pain associated with DPN. This analysis shows a high use of opioids in patients with DPN, which suggests that a majority of patients with DPN are perhaps refractory to first line treatments34 such as NSAIDs and anti-depressants and, hence, were subsequently prescribed opioids for palliation or symptom relief. High use of opioids in patients with DPN may also be attributed to a large number of such patients suffering from pain co-morbidities such as osteoarthritis (33%) and low back pain (32%), which require opioids for pain management.

Analysis of pain co-morbidities associated with DPN showed that co-morbidities such as post-herpetic neuralgia, osteoarthritis, and back and neck back pain, showed an upward trend post-DPN diagnosis, and contributed to the economic burden associated with DPN, as they affect daily functioning of patients and impair their quality-of-life. A recent study showed that one out of four patients with DPN had anxiety, depression, and sleep disturbance8. While there was no significant increase in the number of patients suffering from sleep disorder post-DPN diagnosis (p-value = 0.26), there was a significant increase in the number of patients suffering from depression post-DPN diagnosis (p-value <0.001) (Table 5). This insignificant increase in the number of patients suffering from sleep disorder post-DPN diagnosis could be due to multiple reasons: (1) other co-morbidities associated with DPN overshadow sleep disorder and, hence, are being overlooked by physicians, (2) direct and indirect effect of prescribed medicines17, and/or (3) sedatives prescribed to treat sleep disorder prior to DPN diagnosis.

Since this study was carried out using a commercial claims database, it does not represent the entire US population and it suffers from the limitations associated with any claims data analysis, including coding accuracies and non-availability of clinical and diagnostic information, including any indicator of severity of pain, or its frequency of occurrence. Further, although the study shows an association of DPN diagnosis with an increase in MRU costs, only a percentage of these costs may actually be related to DPN diagnosis. However, given that the data analyzed is pre–post for the same patients, and the fact that the only common change is the diagnosis of DPN, makes us believe that the DPN diagnosis is a large contributor to the change in resource use. Lastly, costs have not been adjusted for inflation, and thus the comparison of cost between the pre-DPN and post-DPN periods may be viewed and inferred from accordingly.

Conclusion

The study uses data from a large US-based commercial claims database to evaluate drug class use as well as resource use and costs associated with DPN. Results of the analysis indicate that resource use significantly increases post-diagnosis of DPN, which could eventually lead to a significant increase in costs for payers. A noticeable proportion of patients with DPN have a pain co-diagnosis and, with tapentadol being the only approved opioid for treating DPN, there is a significant need for more treatments that can effectively manage DPN as well as other associated pain conditions.

Transparency

Declaration of funding and financial/other relationships

Janssen Global Services is the sponsor of the study, and procured the commercial claims data from IMS, Inc. The data analysis for the study was carried out by SmartAnalyst, Inc, who were financially remunerated by Janssen Global Services for their services.

Acknowledgments

The authors thank Kala Hill, MS, Janssen Global Services, LLC, and Nitesh Garg, BTech, SmartAnalyst, Inc. for providing analytical support for this study