1	BEYOND RCTs: USING LARGE CLNICAL DATASETS TO EVALUATE THE EFFECTIVENES OF ADVANCED TECHNOLOGIES Caroline E. Fife, MD
2	Disclosures for Dr. Fife Speaker’s Bureau: KCI, RW Johnson, Organogenesis Grant funding: KCI Owner of Intellicure, Inc. which licenses the software used to collect the EMR data in this study.
3	2005 MCAC Review: The Center for Medicare Services (CMS) posed questions regarding the “usual care” of chronic wounds. No data existed regarding the demographics of patients treated at “wound centers:” co-morbid conditions the utilization of “advanced therapeutics” the average duration of treatment for a given wound.
4
5
6	Medicare and “Chronic Wounds” CMS Definition of “chronic”: wounds which take longer than 30 days to completely heal. 2005 MCAC review was restricted to 3 types: venous ulcers, pressure ulcers and diabetic ulcers.
7	Efficacy vs. Effectiveness Efficacy refers to whether the intervention can be successful when it is properly implemented under controlled conditions Efficacy best determined by prospective, controlled, randomized clinical trials Effectiveness is the capability of producing an effect. Effectiveness refers to the impact in real world situations (whether the intervention typically is successful in actual practice). RCTs are problematic because they may not relate well to “real-world” populations.
8	How well do wound center patients resemble patients in RCTs?
9	RCTs in Wound Healing Research 3 clinical models utilized for wound-healing investigations: venous stasis ulcerations well-vascularized neuropathic diabetic foot ulcerations pressure sores To avoid “beta type” error in which study design causes a product to appear to be ineffective, patients with significant co-morbid conditions are excluded from nearly all wound-healing studies.
10	Evaluating Whether RCTs Reflect Typical Wound Center Patients Electronic searches were conducted on PubMed for the years 1996-2006 using the MeSH terms venous stasis ulceration, diabetic foot ulcer, pressure sore, and wound healing (English-only, RCT) Initial set~ 3,000 abstracts (many duplications) reduced by eliminating: papers that dealt with infection those that did not report “n” RCTs that used < 50 patients selecting only those papers that used “high-technology” wound products those which were duplicates of the same study.
11	Evaluating Whether RCTs Reflect Typical Wound Center Patients Left with 17 RCTs over last 10 years evaluating advanced wound healing technologies. Exclusion criteria were compared by wound-care model (venous, diabetic foot, and pressure ulcer) to patients in the Intellicure Research Consortium (IRC). The IRC consists of de-identified, pooled data obtained from the electronic medical records of patients seen at out-patient wound centers in 11 states.
12	Basic exclusion criteria for all wound related RCTs 1996 - 2006 For DFU studies, no ulcers > Wagner Grade II The presence of Diabetes as a co-morbid condition for any study other than DFU Venous stasis if you are studying DFU Alcohol/drug abuse Anticoagulant treatment Cellulitis or local wound infection Cancer or recent cancer treatment Collagen vascular disease/connective tissue disease Rheumatoid arthritis/autoimmune disease of any type Scleroderma/lupus or any autoimmune disease Charcot foot changes in diabetics Corticosteroid treatment Deep venous thrombosis/pulmonary embolus Gastrointestinal disease/Celiac disease/Cholecystitis/Crohn’s disease/Gastric ulcer/Inflammatory bowel disease/Gastro esophageal reflux disease /Hepatic disease/Liver cirrhosis/Liver failure/Liver disease/Alcoholic liver disease/Hepatitis/Esophageal varices/Hepatitis Renal impairment/End stage renal disease/Renal dialysis/Renal transplant Any organ transplant In diabetics, HbA1c > 8-10 Nutritional impairment/Albumin < 3.0 mg/dl Osteomyelitis Vascular disease/Peripheral arterial disease/Peripheral arterial occlusive disease
13	RCTs vs. Usual Wound Center Patients These exclusion criteria were compared to the complete medical data from 8,611 wound center outpatients Also analyzed were typical parameters including wound “age” (difference between reported onset of wound and data of consultation) and typical wound size allowed within clinical trials.
14	RCTs and Wound Center Patients Among 8,611 wound center out-patients, > 50% would have been excluded from the wound related RCTs at the “first pass,” (exclusions on the basis of co-morbid conditions, previous surgeries or medications, even before further tests are performed). “First Pass” exclusion included: 8.4% of patients were on steroids 5% on renal dialysis (many others had renal insufficiency) 10% of patients with various forms of PVD (low estimate based on reported prior vascular surgery, transcutaneous oximetry or ABI results in charts) 26% of wounds that were not diabetic foot ulcers were found in patients who had diabetes.
15	RCTs and the Usual Wound Center Patient Subsequent use of “utility scores” (estimating the frequency of these diseases in the general population) suggest that many wound related RCTs do a poor job of representing the population at large, much less the wound care population. The restrictiveness of inclusion/exclusion criteria might limit the generalizability of RCTs to the usual wound care patient While RCTs are useful in establishing efficacy under ideal circumstances for a variety of wound-healing interventions, they do not provide the necessary information to establish effectiveness in the “usual” compromised wound-center patient. So where do we go from here?
16
17	The Goals of the MCAC Review: To define what is the “usual care” of chronic wounds To identify what are the common modalities used across the different wound types To identify what unique modalities are noted for a particular wound type.
18	How to get information on “usual care?” Analysis of large numbers of patients Must have complete medical information Must be obtained in uniform manner to facilitate analysis (must be in computerized format) = “Electronic Medical Record” (EMR)
19	METHODS: EMR with “Point of Service” Documentation Physician does documentation in the room while interviewing the patient. All paperwork is finished when the physician leaves the room. 100% of medical information, wound documentation, tests, and treatments are collected in EMR. EMR generates all necessary documents (H&P, nursing notes, follow up letters, procedure notes, home nursing orders, etc.) The legal chart is the electronic one.
20	METHODS: Intellicure Research Consortium Retrospective analysis of de-identified data contained in the electronic medical records (EMR) of 18 wound centers from 11 states. All utilized the Intellicure EMR system. Permission obtained from individual facilities (“Intellicure Research Consortium”) Queries written to answer specific questions posed by CMS.
21	How are advanced technologies used in real world practice?
22	The Usual Care of Chronic Wounds: The Use of Advanced Technology (AT) 8,611 patients 15,499 wounds, 112,000 patient visits 1.8 wounds per patient Average age: 60.4 (1 - 104) Mean wound size =8.3 cm2 Surgical wounds were the largest on average (20 cm2) Arterial ulcers the smallest (3.7 cm2).
23	The Usual Care of Chronic Wounds: n= 15,499 wounds Wound distribution: Pressure ulcers 19% Diabetic foot ulcers 11.5% Venous stasis 11.2% Surgical complications 9.3% Flaps and grafts 2% Arterial ulcers 1.8%
24	The Usual Care of Chronic Wounds Payer statistics: 32.9% Medicare 25.7% Private insurance 4.4% Medicaid 1.2 % Self-pay 0.9% Worker’s compensation Mean number of physician visits over the total length of care = 9.5 Mean number of facility visits = 12.7 Patients were seen on average once every 12 days 74.9% of all visits involved a physician
25	Wound Duration (“Age”) at Consultation: Average = 189 days
26	The Usual Care of Chronic Wounds: The Use of Advanced Technology (AT) Average number of Co-morbid conditions = 6 32% hypertension 5.0%, renal dialysis 10.7% smokers 6.3% CHF 8.4% of patients were on steroids 26% of wounds which were not specifically diabetic foot ulcers were in patients who had diabetes 39% of patients developed a new wound during their course of care.
27	The Usual Care of Chronic Wounds: n= 15,499 wounds 52% of patients were placed on antibiotics during their course of care 45% of patients received prescriptions for pain medications during their course of care In 2% of patients the goal of therapy was listed as palliative care 3.7% of wounds were resolved by amputation 11% of patients (n= 955) died during treatment or within 3 months of the last clinic visit 90.8% of all wounds were healed
28	The Usual Care of Chronic Wounds: n= 15,499 wounds Diabetic Foot Ulcers: 14% had no recorded Wagner grade 16% Grade 1 45% Grade 2 25% Grade 3 Pressure Ulcers: 20% Unstageable 11% Stage I 48% Stage II 21% Stage III.
29	Healed or Not Healed with Moist wound Care: Associated Factors
30	Definitions: “Advanced Tehnology” vs. “Moist Wound Care” AT: Negative Pressure Wound Therapy (Exclusively provided with the VAC Therapy System in this dataset) Hyperbaric Oxygen Therapy (HBOT) Becalpermin (Regranex) Skin Substitutes Apligraf Dermagraft Moist Wound Care: All dressing products in common use in all categories including silver containing products (e.g. alginates, hydrocolloids, foams) There were almost no saline dressings
31	Rates of Use of Advanced Technology (n =15,499 wounds) Moist wound care (MWC) was employed exclusively in 71.0% of wounds. AT was used in 29% of wounds NPWT =5.81% (n = 1146) HBO2T = 9.48% (n = 1870) Bioengineered skin: 3.26% (n Apligraf = 520; n Dermagraft = 123) Becalpermin = 10.49 % (n = 2069)
32	NPWT Use by Pressure Ulcer Stage
33	NPWT Use by Pressure Ulcer Depth
34	NPWT Among 15,499 wounds Used in only 10.7% of Pressure Ulcers (PUs) Stage 4 PUs had significantly higher usage than any other PU wound stage (18.4%) 89% of potentially eligible PU patients did not get NPWT (wounds deeper than 0.3 cm). Average NPWT treatment time in PUs overall was 59 days (range: 1 – 692 days). In diabetic foot ulcers (DFUs), 84 % of potentially eligible patients did not receive NPWT, based on the parameters of those who did receive the VAC (i.e., Wagner II or III wounds > 2 cm2).
35	Rate of wound closure in Pressure Ulcers (NPWT vs. MWT)
36	Factors affecting response to treatment with NPWT Oral steroids Diabetes Infection PVD Active smoking Maximum wound depth Wound volume 60-day gaps in care
37	% Volume Reduction with NPWT: Associated Factors
38	Response to NPWT in relation to wound size and days of use
39	“Lag time” to initiating ATs Average “time to heal” all moist wound care patients = 148 days. Healing rate for moist wound care = 91.2% Healing rate for wounds treated with AT = 85.7% “Lag time” is the time between the initial visit and the time that AT is initiated The average “lag time” to initiating NPWT for all wounds in all categories (date of consult to date of first recorded application) is 115 days. Thus, the average lag time to starting NPWT is slightly less than the average time to heal with moist wound care!
40	What did we learn about “usual care?” The average wound center patient is elderly and has multiple co-morbid conditions, some of which were statistically related to a negative effect on healing: Smoking gaps in care >60 days PVD having a very deep or very large wound Oral steroid administration Infection Diabetes Almost half of wound center patients are on pain medication and more than half receive antibiotics. 39% of patients develop a new wound during their course of care. The average wound is 2.5 to 6 months old at the time of presentation. Nevertheless, 90.3% of these wounds are healed in an average time of 148 days.
41	What did we learn about “usual care?” This impressive healing rate occurs in these compromised patients using primarily conservative, moist wound care techniques (72.7%). Advanced therapeutics is used in only 27.3% of patients. In 7,097 patients with 16,579 wounds, the average “time to heal” for all non-VAC patients was 68 days. This “time to heal” in non-VAC patients is SHORTER than the lag time to starting the VAC in most wound/ulcer categories. In other words, in this dataset, the VAC is usually initiated at a time AFTER non-VAC wounds would have already healed. Despite this, the healing rate for wounds receiving ATs is 85.7%.
42	Areas for future research In most wounds treated with AT, moist wound care is required to fail before AT is begun. This raises issues of Cost-effectiveness. Cost-effectiveness How soon should the treatment be started? How long should the treatment last? What are the criteria for treating a wound with AT first instead of moist wound care?
43
44	“Real World Practical Trials” “CMS doesn't fund randomized trials . . .but they can waive current payment structure . . . as long as you're collecting data and it's new information, we will pay for it . . . They also have what is called quality improvement programs . . . They also . . . . pay for performance . . . .and all of those would lend themselves to some real world practical trials.”
45	Thank You! For a copy of today’s presentation, please visit my website at: www.Intellicure.com/SAWC