PREFACE

            Official Disability Guidelines^TM (ODG) links together four U.S. government databases to provide length-of-disability experience data that can be used to manage employee productivity.  These four databases are the following:

ICD-9-CM — The International Classification of Diseases, 9th Revision, Clinical Modification.  This is the principle coding system used worldwide for the diagnosis of any medical condition.  The main section of ODG  includes verbatim the complete "Volume 1, Diseases: Tabular List" of the current Official ICD-9-CM publication.

CDC NCHS NHIS — The National Health Interview Survey (NHIS) is conducted annually by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC).  ODG  uses data from every year beginning in 1987 until the most current.

OSHA BLS OII — The Bureau of Labor Statistics (BLS) reports annually on Occupational Injuries and Illnesses (OII) from forms submitted by employers to the Occupational Safety and Health Administration (OSHA).  ODG  uses data from the latest available year.

HCUP — The Healthcare Cost and Utilization Project (HCUP) is a family of health care databases and related software tools and products developed through a Federal-State-Industry partnership and sponsored by the Agency for Healthcare Research and Quality (AHRQ) to create a national information resource of patient-level health care data. HCUP includes the largest collection of longitudinal medical care data in the United States. These databases enable research on a broad range of health policy issues, including cost and quality of health services, medical practice patterns, access to health care programs, and outcomes of treatments at the national, State, and local market levels.  The latest available year is included in ODG.

            From the beginning, the first edition of Official Disability Guidelines (ODG) in 1996 provided lost time guidelines using actual experience data from these federal government databases, specifically OSHA BLS (Occupational Safety and Health Administration – Bureau of Labor Statistics) and CDC NCHS NHIS (Centers for Disease Control and Prevention, National Center for Health Statistics, National Health Interview Survey).  The raw data is presented graphically so users can compare it directly with their own experience, and it is designed to enhance a timely and appropriate return-to-work for workers suffering from illness or injury.  From the beginning, ODG was based on actual experience, not “expert” opinion.  This made ODG fair to employees and defensible by employers.  With changes to the Federal Rules of Evidence, the ODG guidelines also became the most likely to stand up in court.  As a result of recent U.S. Supreme Court decisions, the Federal Rules of Evidence were recently amended in December 2000.  The new rules state that statistical studies will be admissible under the Federal Rules of Evidence, and that such methods generally satisfy important aspects of the “scientific knowledge” requirement articulated in the Daubert Decision.[1]  Furthermore, it states that “courts have described surveys as the most direct form of evidence that can be offered, and several courts have drawn negative inferences from the absence of a survey.”[2]  

            Official Disability Guidelines is based on actual reported data from the annual CDC National Health Interview Survey (NHIS), the BLS Survey of Occupational Injuries and Illnesses (SOII), and over 2 million medical records from actual workers’ compensation claims. This includes actual observed case data - rather than government survey "patient recollection" data. All data is tracked by ICD-9-CM code and not just general body part. Since 2003 all of the ODG disability duration data has been validated and enhanced by actual client claims data, and this is reflected in the Return-To-Work Summary Guidelines (Claims data Midrange and At-Risk) as well as the Return-To-Work "Best Practice" Guidelines, the RTW Claims Data (Calendar-days away from work by decile), and the RTW Post Surgery (Calendar-days away from work by decile). Official Disability Guidelines also includes client data, based on over 2 million claims from WLDI’s multi-year multi-state workers comp database, covering almost 50 million paid invoices on medical encounters for those claims. These medical costs represent a total of $10.0 billion dollars in actual incurred costs, and the indemnity costs represent a total of $7.2 billion dollars in actual incurred costs, for a total of over $17 billion of workers’ compensation costs, and they are presented in the table, entitled “Workers’ Comp Costs per Claim.” A detailed Methodology Outline, covering both the Treatment Guidelines as well as the Return-to-Work Guidelines, has been posted on the Web at http://www.odg-disability.com/methodology_outline.pdf. There is also a detailed Methodology Description using the AGREE Instrument (Appraisal of Guidelines Research and Evaluation), posted online at http://www.odg-disability.com/ODG_AGREE.htm.

RETURN-TO-WORK "BEST PRACTICE" GUIDELINES 
  
            The next step in the evolution of ODG  was the identification of pathways for each condition, based primarily on drilling down into the raw data in the NHIS, which has a wealth of detail on type of therapy, type of job, demographics, comorbidities and severity.  These pathways provided the different treatment options with their resultant time out of work, including considerations for severity and type of job.  When different types of jobs made a difference in disability duration, job considerations specific to that diagnosis are identified.  For example, “light duty” is not the same for carpal tunnel as for back strain as for depression.  With different return-to-work pathways for each type of job, modified duty opportunities can be identified, and the appropriate time frames determined.  These treatment pathways are titled “Return-To-Work Best Practice Guidelines”, and in effect, they brought Official Disability Guidelines halfway to offering treatment guidelines.

The term “Best Practice” describes the use of these pathways to manage disability.  For example, for carpal tunnel syndrome (ICD9 354.0) the first entry in the “Best Practice” Guidelines is “Conservative treatment, modified work (no repetitive use of hand/wrist): 0 days”.  The “Best Practice” is to try to follow this initial treatment pathway, and, if it is followed, the actual data support a norm of 0 days.  (Also under carpal tunnel syndrome, the CDC bar chart shows that in 43% of the cases reported to CDC, where a physician has diagnosed carpal tunnel syndrome, no time is missed.  This is the data that supports that pathway.)

The “Best Practice” guidelines were first launched in the 1997 edition of ODG, but they have been expanded in each subsequent annual edition.  Currently, Official Disability Guidelines has “Best Practice” guidelines for four times as many conditions as in 1997, and the average number of treatment options per condition is more than double what it was in 1997.  With over 16,000 clients, the “Best Practice” guidelines in ODG are nationally recognized.  Since they are based on actual experience data from the federal government, they are scientifically valid and outcome-based.  New users of lost time guidelines have gravitated toward ODG because the “Best Practice” guidelines, the heart of ODG, identify what makes a difference in return-to-work.  Rather than looking at an average or a median for all cases for a particular condition, ODG allows comparison among like cases.  Within diagnoses, some cases should return earlier than others because they are on a different pathway.  Trying to make them all adhere to an overall median will not only let some cases be out too long, but will also force some cases back to work too soon.  As ODG became a focal point facilitating communication among all parties in the return-to-work process, including patients, it has been assisting all parties with regard to the appropriate treatment and management of work related injuries and illnesses.  The framework of the “Best Practice” guidelines has established elements against which aspects of care can be compared, and allowed identification of treatments and services that are reasonable and medically necessary for treatment of a particular injury.

            The "Best Practice" disability duration data is contained in boxed bold type under each diagnosis.  These durations are what can be achieved through management of the disability case, based on analyzing the raw data and comparing findings with the experience of clients of Work Loss Data Institute.  The "Best Practice" Guidelines also reflect the experience of the members of the ODG Editorial Advisory Board, comprised of about eighty medical professionals, typically medical directors at large corporations or other employers with significant disability experience, who review the “Best Practice” guidelines every year to identify new return-to-work pathways and compare the durations with their own experience.

            Separate data is provided for whatever factors would significantly affect the disability duration.  For example, choice of therapy may be identified, including different procedures, along with target disability durations for each of those.  Each of these procedure names is also indexed in the Keyword Index for easy access to durations on procedures.  In many cases there are significant differences between cases treated surgically versus medically (e.g., drug therapy).  For diagnoses where co-morbidities are a significant factor, these may also be identified in the "Best Practice" Guidelines section. These "Best Practice" Guidelines focus on return-to-work, and are not as detailed as clinical best practice guidelines, but they should include the various paths of treatment suggested by clinical best practice guidelines.

            Where type of job makes a difference, that is shown also.  This may be clerical work versus manual work, but it may also be other factors such as sedentary versus standing, or use of a particular body part.  Where clerical/modified work shows a shortened disability duration, the "modified" work may be an opportunity to return the worker to restricted duty, before he or she returns to their normal duties.  Where other factors affect duration, such as non-dominant versus dominant arm, they are also identified.  For each diagnosis, there may be specific job characteristics that affect length of disability.  These characteristics may not correspond to the five job classifications in the Department of Labor's Dictionary of Occupational Titles.  Where they do apply, "sedentary" corresponds to class 1 (sitting, up to 10 pounds of force), "clerical" is class 2 (up to 20 pounds), "manual" is class 3 (up to 50 pounds), and "heavy manual" covers class 4 or 5 (up to 100 pounds and over 100 pounds).

            The graphs shown for each database of experience data may show different clusters of data.  The "Best Practices" section helps explain what causes those clusters, e.g. medical treatment versus surgery, hospitalization versus non-hospitalization, light-duty versus heavy manual work, etc.

            Throughout the "Best Practice" Guidelines there is consistency in the definition of days.  Return-to-work durations are always in calendar days, as opposed to workdays, so they can be applied to workers on different shifts, full time versus part time, etc.  Length of disability of 7 days is equal to one week.  A partial day missed is treated as one day if the employee would be expected to be out for most of the day (e.g., for a colonoscopy).  Time off for an hour or two, say for routine diagnostic examination, physical therapy, or limited chemotherapy, would be treated as zero days.  Each of the treatment options under the "Best Practice" Guidelines generally has its own disability duration.  For example, the time out of work for initial conservative treatment would be separate from time off for surgery, if the conservative treatment is chosen but it turns out to be unsuccessful, and then surgery is selected.  On the other hand, a diagnostic procedure may be also part of a more extensive therapeutic procedure.

            These guidelines are meant to be used to identify cases that are out of the norm, where questions may be asked, such as what makes them different.  Especially where there is a great variation in severity, for example, for some cancers, additional information may be requested and the additional time out of work may be justified. If the patient has co-morbidities that are not specifically identified in the guidelines, application of the guidelines is more difficult.  The final opinion regarding any medical condition and the ability of a patient to return to work should rest with the physician treating that patient.  Where the "Best Practice" disability duration guidelines indicate "by report", variances in the data made it impossible to select a benchmark number of days, and the report by the evaluating physician should guide the amount of time off work.

            It should also be noted that achieving the best practice guidelines disability durations typically requires appropriate job descriptions and availability of altered work.  Depending on the type of work, some injuries will have a residual chronic pain syndrome that will require accommodation.  It is recommended that these guidelines be achieved in a setting that includes modified duty work as well as case management.  Some employers have found that with aggressive Return-To-Work modified duty programs, disability schedules can be considerably shortened compared to the "Best Practice" guidelines.  On the other hand, modified duty policies are quite variable among employers, and the clinician needs to acknowledge that the level of RTW function they approve may not be accommodated by company policy.

            Some physicians consider the return-to-work dates in the "Best Practice" guidelines to be aggressive, and there may be some cases that do not meet these guidelines.  This may result in disagreement between case managers and physicians.  The best practices take into account the best circumstances.  Some patients can return to work earlier than the best practices suggest, and others later than suggested.  Such variables as age, co-morbid conditions, severity, job type and other items can impact disability duration and must always be taken into account.  When patients fall outside these values, most notably if the projected disability duration exceeds "Best Practice" estimates, the case manager should consult the treating physician as to why the case might not fit the "Best Practice" guidelines.

            One of the challenges in disability management is what to do when a person has recurrent problems.  For instance, when someone has headaches, rheumatoid arthritis, osteoarthritis, or cancer that has recurrent symptoms, it is very difficult to determine a "Best Practice" disability duration.

            The Return-to-Work “Best Practice” Guidelines, comprising the most important feature in ODG, are now closer to the top within each condition, in the second box.   These still show estimated days out of work (based on national norms) for typical cases within each condition depending on severity, type of treatment and type of job, including modified duty.  They are indispensable to effective case management, by identifying up front what “pathway” a case is likely to follow.  In addition, multipliers for many common comorbidities have been added, based on the raw experience data.

            Multiple incidences of disability duration: When managing multiple incidences of disability duration for the same worker on a prospective basis, ODG users should consider each one separately when creating return-to-work expectations using the “Best Practice” Guidelines. The disability duration data used to derive the pathways in the Return-to-Work “Best Practice” Guidelines is based on single incidences of missed work, because that is the only way to isolate the specific factors affecting return-to-work in each of the pathways. (On the other hand, when benchmarking claims on a retrospective basis, users should be aware that the RTW Summary Guidelines, as well as the ODG claims data by decile, include all instances of absence for each claim.) And, as with all disability durations in ODG, the length of disability is calculated by taking the return-to-work date minus the last day of work less one-day. When using durations in the “Best Practice” Guidelines to manage a specific worker’s absence on a prospective basis, the expected disability duration should generally “reset to zero” if the worker has returned to work for a period, but then misses work again at a later time. Furthermore, because of the potential for abuse from multiple incidences of absence, users should probably “reset” the duration only for the second instance of absence for the same condition within a year (not the third, fourth, etc.), and only if the time during which the worker returned to work is significant, i.e., it exceeds the disability duration that preceded it. Specific absence situations that exceed these guidelines will need to be reviewed on a case-by-case basis. The return-to-work period need not be job specific, so it may be regular work that could have been impacted by the medical condition, or either regular work or modified duty where the condition should have a limited effect. In addition, please note that disability duration pathways in ODG that refer to surgery are calculated from the date of surgery, and not from the last day of work. These general principles should also apply to recommendations regarding number of treatment visits, for example, physical therapy, even though these visits may or may not be during an absence from work. In general, a second incidence of physical therapy visits after a substantial time back at work may represent a recurrence of the original condition that might allow another series of physical therapy visits. Without a disability-duration to trigger this, the “substantial time back at work“ might be considered anything greater than the number of “At-Risk” days for that condition. And, physical therapy visits post surgery should be considered separately from visits used up in an attempt at conservative treatment that might have avoided surgery. Again with respect to retrospective benchmarking, as opposed to prospective claims management, disability duration data in ODG that is used for this benchmarking, for example, the Return-to-Work Summary Guidelines or the RTW Claims Data, includes all incidences of disability duration for a single claim with that primary diagnosis over the previous year. 
   

            Additional note on co-morbidities: With respect to co-morbidities, in most cases the expected disability duration will be driven by the most severe diagnosis. In fact, when disability durations are calculated in ODG from actual experience data, and a case has multiple ICD9 diagnosis codes, all of the lost workdays for that case are assigned to only one ICD9 code, the one with largest number in the ODG Return-To-Work Summary Guidelines (i.e., Midrange, All absences). Consequently, when using ODG to determine an expected return-to-work date, the disability durations should not generally be added together when the worker has multiple injuries or illnesses. Unless there are instructions that incorporate those specific co-morbidities, users should just take the longest duration and go with that. The healing time of the less serious condition should fall within that. For example, if a worker has a back sprain and a disc disorder, the return-to-work (RTW) date should be driven by the disc disorder diagnosis, and the healing time for the back sprain should fall within that. This is similar for RTW after surgery. For example, if someone has a spinal fusion and a discectomy, the return-to-work date would be driven by the fusion.

RETURN-TO-WORK SUMMARY GUIDELINES

            Based on input from users, the 2003 edition of Official Disability Guidelines underwent a major re-design to facilitate finding the right information quickly. With the new format you can efficiently locate the number you need to reserve a claim, or you can get the in-depth backup information necessary for more extensive case management.

            Return-to-Work Summary Guidelines show estimated days out of work (based on national norms) for each condition in summary, for those who just want to select a target, and you can now cost justify case management efforts by “beating the guideline” using the At-Risk date.  These are followed by ODG’s well-respected “Best Practice” Guidelines, which have proven indispensable to effective case management by identifying up front what “pathway” a case is likely to follow.  This new “Summary Guideline” box brings to the front of each diagnosis, experience data that was previously contained in the ODG “decile table”, using the 50% number for “Midrange” and the 90% number for “At-Risk”.

            Please note: An important distinction needs to be made between the Return-To-Work Summary Guidelines and the Return-To-Work “Best Practice” Guidelines.  The Summary Guidelines were designed primarily for retrospective benchmarking of claims, requiring only a diagnosis, plus a disability duration.  On the other hand, the “Best Practice” Guidelines were designed primarily for prospective case management, when more details about a case are known, for example type of therapy, type of job, severity, co-morbidities, etc.   The At-Risk date in the Summary Guidelines should NOT be used for prospective case management – it may be too late to begin management at that point.  This is the point when the case has already become an outlier and is at risk of never returning to work, no matter how effective additional case management may be.  Instead, the At-Risk date may be used as a consistent measure across different operating units to determine how effective case management efforts have been against national norms in “beating the guideline”, since unmanaged cases will tend to become outliers, and hit the At-Risk date.

            Beginning with the 2005 edition, a new row was added to the Return-to-Work Summary Guidelines.  This “Summary Guideline” box brought to the front of each diagnosis experience data that is contained in the ODG “decile table” (the RTW Claims Data - Calendar-days away from work by decile), using the 50% number for “Midrange” and the 90% number for “At-Risk”, and this “decile table” includes only cases that were out more than 7 days, so that the data is consistent with and comparable to the claims data that most ODG clients use when benchmarking.  Now these numbers continue to be shown in the Summary Guidelines in the first row, entitled “Claims data”.

In recent years there has been increased focus on “incidental absence”, those cases typically out for 7 days or less, that may never become claims under most workers’ compensation rules or under the eligibility requirements of most disability benefit programs.   Furthermore, many employers and their vendors have moved to early reporting of absence, in order to improve early return to work.  Because of this, they are picking up cases in their case mix that never would have been in their database of reported absence in the past.  In order to provide benchmarking data for these clients, a new row has been added, entitled “All absences”.  This row uses the 50% number for “Midrange” and the 90% number for “At-Risk”, covering all absences, and not just cases that were out more than 7 days.  Because this data includes the shorter duration cases, these “Midrange” and “At-Risk” numbers will generally be shorter than the previous numbers (which are still being displayed in the “Claims data” row).

When deciding which numbers to apply in benchmarking, users will need to ask themselves whether or not their own dataset generally includes cases out for 7 days or less.  If so, to be consistent in their application of the national norms, they should use the row labeled “All absences”.  For some conditions, such as a broken leg, there will not be significant differences in the numbers in the two rows.  For other conditions where a significant percentage of cases miss less than 8 days, such as colds or flu, the differences will be substantial.

As before, these Summary Guidelines are followed by ODG’s well-respected “Best Practice” Guidelines, which have proven indispensable to effective case management by identifying up front what “pathway” a case is likely to follow

Some of the ways these upfront Summary Guidelines numbers (typically the At-Risk number) are used is as follows:

·         Reserves: Estimating duration for purposes of setting conservative reserves.

·         Targets: Selecting a duration as “the number to hit” or the “the number to beat”.

·         Pre-authorization rules:  Some workers’ compensation systems use the “At-Risk” date to trigger pre-authorization requirements, making providers submit approved treatment plans prior to payment, for cases that have exceeded this limit.

·         Budgeting: Making an initial prediction of disability duration, and keeping that prediction, unmodified, in a database in order to compare the eventual actual duration against that first estimate.

·         Performance reviews: Using duration as a QA performance standard, as part of the qualitative evaluation of a few selected case manager files.  A powerful way to evaluate the effectiveness of case managers or teams is to compare actual vs. estimated durations across whole caseloads or other large groups of claims.

·         Organizational benchmarking: Using duration as an aggregate system benchmark for median duration of disability across the whole book of business, and providing top management a report every month tracking the actual performance of each operating unit against those benchmarks.  In addition, innovations in claim and case management can be tracked for their effectiveness in reducing median durations.

·         Client benchmarking: Claims organizations can measure overall performance, or individual performance, by adding up the actual durations for all cases and dividing that by the sum of the “at-risk” numbers.  This can also indicate "total days saved”.

·         Grading performance:  Sum up all internal claims durations, sum up corresponding At-Risk durations from ODG Summary Guidelines (with an ICD9 coded At-Risk date corresponding to each claim), divide the sum of the At-Risk dates minus the sum of your internal claims durations by the sum of the At-Risk dates, and multiply the result by 100 to get a percentage score.  For more details on this technique, which is also described in a CE article in the February 2005 issue of the AAOHN Journal, request a copy of “Benchmarking Medical Absence” from Work Loss Data Institute, or find it on the Web at http://www.disabilitydurations.com/benchmarking_lost_time.htm.

OTHER KEY FEATURES

            After the Summary Guidelines and Return-To-Work “Best Practice” guidelines are the following additional features, which appear under each condition where they apply and there is sufficient data available:

            Capabilities & Activity Modifications

Activity Modifications shows condition-specific modifications for each level of job identified in the “Best Practice” Guidelines.  These are meant to be used in conjunction with the “Best Practice” Guidelines to determine what level of job is appropriate and for how long.  For example, if “Severe, clerical/modified work: 0-3 days” appears, then look at the definition of “clerical/modified work” for job modifications used to prevent re-injury.  Then,  “Severe, manual work: 14-17 days”, about two weeks later the worker may transition to work defined under “manual work”.  Physicians can copy & paste these restrictions into a RTW form for use by employers, and all parties (doctors, patients, employers, and insurers) use them as a communication tool to create shared expectations.  They can also facilitate return to modified duty, which is often a critical first step in the return-to-work process. 

            Description and Other Names

            For most common injuries and illnesses there is a description of the diagnosis, along with common symptoms, causes, and complications, using terminology understandable to non-medical personnel.  Common names for this diagnosis are also provided.

ICD-10 Codes

ICD-10 Codes are next, providing complete ICD-10 translations for each condition (Web version only).

            Procedure Summary (from ODG Treatment)

            This section lists procedures and other topics relevant to this diagnosis, as they appear in the Procedure Summary of ODG Treatment in Workers' Comp.  (In the Web version each procedure is a hyperlink going directly to that entry.)  The Procedure Summary is the most important section in ODG Treatment, and the first two sections, the Treatment Protocols and the Codes for Automated Approval, are based on the conclusions from the evidence in the Procedure Summary.  The Procedure Summary lists all possible therapies and diagnostic methods, as well as other issues that apply for each condition, and provides a summary of the latest evidence from the highest quality medical studies.  The studies providing this evidence are referenced so that they can be consulted directly, and if necessary, copied into a claims report.  For each condition, there may be as many as 100 separate listings covered in this fashion.  Many of these procedures are being performed regularly, but are not supported by the quality medical evidence as summarized in this guideline, and in some cases, are proven to be harmful.  When patient selection is important to the success of a procedure, the criteria for patient selection is also outlined, and the appropriate study is referenced.  In supporting decisions to approve or deny medical services, users of ODG Treatment can go beyond quoting a set of guidelines, and copy and paste the results of the actual study, taking "evidence based medicine" to its logical end point.  Clicking on the hyperlinks (containing author name and study year) in the Web version of the Procedure Summary will take the user directly to the studies supporting that statement.  These reference summaries, including an abstract, plus the WLDI evaluation and rating of the reference, are in alphabetical order for those who want to browse them all, and important points in the study are highlighted.  WLDI uses a proprietary rating system to evaluate the quality of the studies, ranging from 1a to 11c.  Within the Procedure Summary there are specialized guidelines for various topics that stand out because they are highlighted in light blue.  For surgical procedures that may be supported by high quality medical studies, ODG Treatment presents a decision matrix called “ODG Indications for Surgeryä” that itemizes the decision-making process and patient selection criteria for successful outcomes from the surgery.  Also within the Procedure Summary there is another specialized guideline for various topics that stands out in light blue.  Contained in this section, where appropriate under imaging procedures, such as Radiography, Magnetic resonance imaging (MRI), or Ultrasound, are the recommended criteria for those modalities.

            Causality Likelihood

Based on the raw data, causality likelihood indicates what percentage of total lost workdays were occupational (Web version only).  The data sets used for this calculation are OSHA lost workdays per 100 workers (for cases meeting the requirements as an OSHA recordable injury or illness), and CDC NHIS lost workdays per 100 workers (for all cases, including non-occupational illnesses and injuries) from these two ODG databases through the 2004 edition.  (Methodology differences make this comparison difficult using data after 2004.)  When the causality likelihood percentage is large, cases with that diagnosis are likely to be occupational in nature.  This indicator may be used as an aid in evaluating causality, but any definitive determination of causality requires analysis of the specific details of each case.

Hospital Costs

Hospital Costs are next, showing average hospital costs for each condition, including total number of cases per sample (Web version only).  Average costs are shown for each condition where there is sufficient data, and the number of cases is shown for the most recent year.  These costs are from the Healthcare Cost & Utilization Project (HCUP), produced by the U.S. government Agency for Healthcare Research and Quality (AHRQ).  The costs only cover cases that were hospitalized.  For conditions where there are not enough cases of hospitalization to estimate average costs, no medical costs are shown.

            Hospital Length Of Stay

Also from the Healthcare Cost & Utilization Project (HCUP), average Hospital Length Of Stay is shown for the most recent year (Web version only).  Note: when using these numbers as benchmarks, it is important to use the most current edition of ODG, because there has been a continued decline in hospital length of stay, but a significant increase in average medical costs.  For selected workers’ comp conditions, average hospital length of stay will also be shown for surgical procedures done as an inpatient, along with the ICD9 procedure code for those procedures.  (For hospital procedures, the ICD9 procedure coding system is used, whereas for physician office procedures, the CPTâ coding system is used.)

Procedure Codes

Procedure Codes commonly performed for each condition are listed (Web version only).

            Case Management Triage

Case Management Triage priority indicators are next (Web version only).  Each condition has a heading, “CM Triage”, which uses algorithms applied to the raw data to label each condition with priority indicators:  Level 1 - "Low Touch", Level 2 - "Case Management", or Level 3 - "Long Term Planning".  Level 1 conditions (66%) don't require initial management.  If they reach their inflection point without resolution, they become Level 2 and require CM.  Level 2 (17%) benefit immediately from CM, and Level 3 (17%) cases are anticipated to be long-term, with a large percentage out for a long period, allowing advance planning with respect to paperwork, reserves, SSDI, etc.  Copies of the algorithms are available in electronic versions of ODG.

            Physical Therapy Guidelines

Physical Therapy Guidelines, showing recommended frequency and duration of PT visits are next.  Only appropriate conditions have physical therapy guidelines.  These guidelines provide evidence-based benchmarks for the number of visits with a physical or occupational therapist and the period of time during which these visits take place.  (Note: These guidelines do not include work hardening programs.)  The physical therapy guidelines do not describe the type of therapy required, and the number of visits does not include physical therapy that the patient should perform in their own home or work site, after proper training from a clinician.  Unless noted otherwise, the visits indicated are for outpatient physical therapy, and the physical therapist's judgment is always a consideration in the determination of the appropriate frequency and duration of treatment.  Support for the physical therapy guidelines is relevant medical literature and actual experience data, combined with consensus review by experts.  The most important data sources are the high quality medical studies that are referenced in the treatment guidelines, ODG Treatment in Workers’ Comp, within the Procedure Summaries of each relevant chapter, summarized under the entry for “Physical Therapy.”  For clinical trials that show effectiveness for these therapies, the number of visits required to achieve this are isolated from each study and combined with the same information from other successful studies to arrive at the benchmark number of visits in ODG.

There are a number of overall physical therapy philosophies that may not be specifically mentioned within each guideline: (1) As time goes by, one should see an increase in the active regimen of care, a decrease in the passive regimen of care, and a fading of treatment frequency; (2) The exclusive use of "passive care" (e.g., palliative modalities) is not recommended; (3) Home programs should be initiated with the first therapy session and must include ongoing assessments of compliance as well as upgrades to the program; (4) Use of self-directed home therapy will facilitate the fading of treatment frequency, from several visits per week at the initiation of therapy to much less towards the end; (5) Patients should be formally assessed after a "six-visit clinical trial" to see if the patient is moving in a positive direction, no direction, or a negative direction (prior to continuing with the physical therapy); & (6) When treatment duration and/or number of visits exceeds the guideline, exceptional factors should be noted.

            Generally there should be no more than 4 modalities/procedural units in total per visit, allowing the PT visit to focus on those treatments where there is evidence of functional improvement, and limiting the total length of each PT visit to 45-60 minutes unless additional circumstances exist requiring extended length of treatment. Treatment times per session may vary based upon the patient's medical presentation but typically may be 45-60 minutes in order to provide full, optimal care to the patient. Additional time may be required for the more complex and slow to respond patients. While an average of 3 or 4 modalities/ procedural units per visit reflect the typical number of units, this is not intended to limit or cap the number of units that are medically necessary for a particular patient, for example, in unusual cases where co-morbidities involve completely separate body domains, but documentation should support an average greater than 4 units per visit. These additional units should be reviewed for medical necessity, and authorized if determined to be medically appropriate for the individual injured worker.

As described above, for more detail users should refer to ODG Treatment in Workers’ Comp, within the Procedure Summaries of each relevant chapter, for recommendations about specific treatments and modalities, along with supporting links to the highest quality relevant medical studies, which have been summarized, rated, and highlighted. In these Procedure Summaries ODG covers many different types of treatments that can be supported by the medical evidence, and it also identifies the maximum number of visits that can be justified by the evidence; however, this does not mean that a provider should do every possible treatment that may be recommended (actually, this would be highly unlikely since different specialties would be required), or always deliver the maximum number of visits, without taking into account what was needed to cure the patient in a particular case. Furthermore, duplication of services is not considered medically necessary. While the recommendations for number of visits are guidelines and are not meant to be absolute caps for every case, they are also not meant to be a minimum requirement on each case (i.e., they are not an “entitlement”).  Any provider doing this is not using the guidelines correctly, and provider profiling would flag these providers as outliers. This applies to all types of treatment, and not just physical therapy. Furthermore, flexibility is especially important in the time frame recommendations. Generally, the number of weeks recommended should fall within a relatively cohesive time period, between date of first and last visit, but this time period should not restrict additional recommended treatments that come later, for example due to scheduling issues or necessary follow-up compliance with a home-based program. When there are co-morbidities, the same principles should apply as in the ODG guidelines for return-to-work. See Additional note on co-morbidities at the end of the description of the Return-To-Work "Best Practice" Guidelines. In estimating the maximum number of treatment visits for workers with multiple diagnoses, users should use the number from the diagnosis with the longest number of visits. This assumes that whatever separate therapy, if any, that the lesser diagnosis requires, it can be done during the same visits addressing the more serious problem. If there are reasons why these therapies cannot be concurrent, documentation should support medical necessity. For example, in unusual cases where co-morbidities involve completely separate body domains, requiring separate treatments that would be difficult to combine, either additional visits or additional time for a visit may be justified. [For the purpose of this discussion, we would assume there could be only three separate body domains: (1) spine and pelvis; (2) upper extremity and hands; & (3) lower extremity and feet.] Of course, each billed treatment should require one-on-one patient contact with the licensed therapist and not include modalities/exercises that the patient has learned to do on their own without supervision, and there should also be some economies of scale such that the involvement of two body domains should not require either a doubling of the number of visits or a doubling of the modalities (or time) per visit. Also see Multiple incidences of disability duration in the same section for recommendations regarding number of treatment visits, for example, physical therapy, in these situations. And physical therapy visits post surgery should be considered separately from visits used up in an attempt at conservative treatment that might have avoided surgery.

            Physical medicine treatment (including PT, OT and chiropractic care) should be an option when there is evidence of a musculoskeletal or neurologic condition that is associated with functional limitations; the functional limitations are likely to respond to skilled physical medicine treatment (e.g., fusion of an ankle would result in loss of ROM but this loss would not respond to PT, though there may be PT needs for gait training, etc.); care is active and includes a home exercise program; & the patient is compliant with care and makes significant functional gains with treatment.

Chiropractic Guidelines

Chiropractic Guidelines are next, showing recommended frequency and duration of chiropractic care.  These guidelines provide evidence-based benchmarks for the number of visits with a chiropractor and the period of time during which these visits take place.  Support for the chiropractic guidelines is relevant medical literature and actual experience data, combined with consensus review by experts.  The most important data sources are the high quality medical studies that are referenced in the treatment guidelines, ODG Treatment in Workers’ Comp, within the Procedure Summaries of each relevant chapter, summarized under the entry for “Manipulation.”  For clinical trials that show effectiveness for manipulation, the number of visits required to achieve this are isolated from each study and combined with the same information from other successful studies to arrive at the benchmark number of visits in ODG.  Another major source was the “Mercy Guidelines”, the consensus document created by the American Chiropractic Association in conjunction with the Congress of State Chiropractic Associations, entitled Guidelines for Chiropractic Quality Assurance and Practice Parameters, Proceedings of the Mercy Center Consensus Conference.  Many of the general philosophies described above under “Physical Therapy Guidelines” should also apply to the chiropractic guidelines.  More specifically, in addition to a “six-visit clinical trial”, every six visits thereafter the treating physical or occupational therapist/chiropractor should validate improvement in function as it relates to the patient’s essential job functions, hours working, health related quality of life indicators (e.g. Oswestry) or a standard pain scale in order for treatment to continue. Pain reduction should be accompanied by improved function and/or reduced medication use. For other general guidelines that may apply to chiropractic care, also see Physical Therapy Guidelines.

            Workers’ Comp Costs per Claim

            Indemnity costs, medical costs and total costs per claim for over 2000 ICD9 diagnosis codes seen in workers’ comp are provided. Within each cost category ODG shows the cost distribution by quartile (25%, 50%, and 75%), the mean (or average) costs, and the percentage of claims with no costs in that category, plus total number of claims that the cost data is based.  It includes almost 2 million claims from WLDI’s multi-year multi-state workers’ comp database, and it covers almost 50 million paid invoices on medical encounters for those claims.  The medical costs cover multiple categories, including office visits, surgeries, PT, pharmaceuticals, hospital, durable medical equipment, etc.  They are from medical provider bills that were approved and paid, but not bills from MCO's, so they do not include the cost of managed care services (bill review, case management, UR, etc).  When there are multiple ICD9 diagnostic categories in a claim, all costs for that claim are assigned to the most severe ICD9 code, using the ODG disability duration database to identify the most severe ICD9 code.  These medical costs represent a total of $10.0 billion dollars in actual incurred costs, and the indemnity costs represent a total of $7.2 billion dollars in actual incurred costs, for a total of over $17 billion of workers’ compensation costs.

            Age Adjustment Factors

Age Adjustment Factors are next, where there is sufficient raw data, in a boxed table providing condition-specific multipliers important for the aging workforce, plus the At-Risk date pre-adjusted by the multipliers.  Most experts believe it is reasonable to modify return-to-work by multiplying the value from the appropriate category, ideally using the “Best Practice” Guidelines, by the value for the corresponding age of the patient.

            RTW Claims Data

RTW Claims Data for benchmarking is the table formerly called “RTW Raw Data by Decile (with 7-day waiting period)”, showing days away from work by decile (10 percent of claims back by tenth day, etc.), including mean.  This boxed table displays the disability duration data by decile for only those cases with over 7 lost workdays.  The 7-day cut-off was chosen so the data would be comparable to the most common reporting systems used for short-term disability, which have a 7-day waiting period.  Showing calendar days off by percentile allows meaningful benchmarking of disability claims experience data, to identify opportunities for improvement.  On the other hand, the bar chart for Integrated Disability Durations raw data shows all the disability duration data, starting at 1 day missed, and the footnote to that bar chart even identifies cases with no missed work.

            RTW Post Surgery

            This table is the same format as the RTW Claims Data above, but it shows disability duration data after selected surgical procedures that may be commonly done for this condition.  These durations only include cases where the specific procedure was performed when the primary diagnosis was the ICD9 diagnosis code indicated above.  Consequently, the post-surgical disability durations in ODG may vary for the same procedure when it appears under different ICD9 diagnosis codes.

            Integrated Disability Durations Raw Data

Integrated Disability Durations raw data is next, including Length of Disability Data from CDC NCHS (Centers for Disease Control National Center for Health Statistics), charting disability duration data for all cases from the National Health Interview Survey.  The “Impact on Total Absence”, based on this data, is under this table showing impact on total absence for each condition (total incidence and prevalence data).

            The two length-of-disability databases are each provided in a similar format, with summary information plus a graphical representation of the actual data using bar-chart format.  The bar charts for each of the two different databases are of two different widths so they can be quickly distinguished from each other.  Typographical differences are used to distinguish the disability databases from the ICD9 database, which serves as the organization of ODG, and the framework upon which the disability databases rely.  Whereas the ICD9 data is in serif type, left-justified format with tabs denoting its hierarchical structure, the disability data is in smaller san serif type and centered.

            The first database to appear is labeled Integrated Disability Durations.  These include cases of calendar days away from work.  Four data summaries are shown -- median (mid-point), mean (average), mode (most frequent), and calculated rec.  This data is reflective only of those cases that report at least one day of lost work within the previous year (cases with no lost work have been excluded from calculations).  The calculated recommendation data is a calculation that takes into account the other three summary data points, giving extra weight to the median.  Following these summary calculations is a bar chart showing "Percent of Cases".  The bar chart always has 14 bars.  Depending on the data, there may be a bar for each length, e.g., "1 day", "2 days", etc., with the last bar showing "14 and more days".  If 14 days is not long enough to show the detail, a different scale is used, saying "Range of Days (up to)".  For example "3" means "1 to 3 days", then "6" means "4 to 6 days", etc.  There may be a footnote to the bar chart if any cases were reported with no lost work, saying "cases with no lost workdays" with the percent of total cases, including the cases indicated graphically on the bar chart.

            This data is also the basis for several other presentations of raw data.  The heading "Impact on Total Absence" provides incidence and prevalence data for each diagnosis, as well as for higher-level groups of diagnoses.  The prevalence data provides the frequency of a diagnosis in percent, by dividing the total lost workdays for that diagnosis by the total lost workdays for all diagnoses.  The incidence data equates this prevalence rate to total lost days per year per 100 full time equivalent workers, using base absence data described below.  With this data, users can easily target those conditions that have the largest impact on productivity and profitability.  Total lost-work days, for any unscheduled absence due to illness or injury (including sick leave, short and long term disability, and workers’ compensation), add up to 1,050 days per 100 workers (or 10.5 days per worker, equating to an absence rate of 4.2% assuming 250 work-days annually).  This does not include scheduled absences (e.g., vacation, holidays, certain leave-of-absence) or unscheduled absences due to other causes (e.g., personal reasons, care of a family member, “no-shows”, absences caused by an “entitlement” mentality).  Other sources of absence data vary somewhat from this benchmark of 4.2% -- some are higher and some lower.  A survey of very large employers by Mercer resulted in an estimate of 4.4%[3], and this estimate includes workers’ comp medical costs and vendor administrative charges, plus the average employer size was over 5,000 employees.  BNA’s annual survey shows a much lower rate, 1.7%[4], but this is based only on absences reported to the human resources department, and it covers a cross section of U.S. employers, including smaller companies without a rich benefit structure.  The results of CCH’s annual survey were slightly higher than BNA’s at 2.2%[5], with a similar methodology to the BNA survey.  The highest rate of all was reported by the annual Watson Wyatt survey done in conjunction with the Washington Business Group on Health, showing 6.3%[6], but this survey is very much weighted toward the largest employers and those with the most generous benefit structures.  It should be noted that, despite their widely differing estimates, the above studies are all based on surveying employers.  On the other hand, ODG is based primarily on actual data reported to the federal government.

            Occupational Disability Durations Raw Data

Occupational Disability Durations raw data is next, including OSHA DAW Data (Occupational Safety and Health Administration Days Away from Work), providing lost time statistics on work-related disabilities as reported to OSHA, and calculating estimated workers’ compensation indemnity costs for each condition.  The “Impact on Occupational Absence”, based on this data, is under this table showing impact on occupational absence (occupational incidence and prevalence data).

This second database is based on reports by employers of missed workdays for occupational related injuries and illnesses.  Only a median is provided for this data and the bar charts always have 7 columns.  Beginning with the 2005 edition of ODG, OSHA now uses calendar days in its surveys.  At the request of ODG users, a benchmark indemnity cost estimate is provided along with all OSHA tables.  These costs are based on the State Average Weekly Wage (SAWW) used by the Texas Workers' Compensation Commission ($539).  These costs do not include medical costs, but only the indemnity (lost work) portion of workers' comp costs.  The total benchmark indemnity costs include both direct costs, the actual wages paid, as well as indirect costs, which include the costs of replacement, rehiring, retraining, overtime, down time, lost productivity, and infrastructure costs to manage current systems.  Total costs including indirect costs have been estimated to be five times the direct costs, based on industry studies (Kalina, AAOHN Journal, August, 1998, page 385, and Guidotti, American Medical Association, Occupational Health Services, 1989).  Of course, the percentage of indirect costs will vary considerably from industry to industry.  The OSHA data is the basis for the data following the heading "Impact on Occupational Absence".  The prevalence rate is the total occupational lost workdays for that condition divided by the total occupational lost workdays for all conditions.  The incidence rate is based on applying the incidence to total lost work-days for workers' comp cases based on data from NIOSH (the National Institute for Occupational Safety & Health).

FINDING THE RIGHT CONDITION

            The framework for the main section of ODG,  "Disability Guidelines", is the same as the ICD-9-CM, and each of the other three databases which provide disability durations are linked to the appropriate ICD9 for which they apply.

            Official ICD9 data is organized by code number.  These codes are hierarchical and may be three, four, or five digits, or may be at the group level, depending on the detail required, and they are grouped together for similar conditions.  The "At a Glance" Tables of Contents indicate where each major section begins.

            There are three ways to find the correct section of "Disability Guidelines" using ICD9 codes, as follows:

            1) Medical Record Virtually all medical records and billing forms should have an ICD9 diagnostic code for the patient.  These codes may not have a decimal point, and they may be padded out to 5 digits on the right with zeros.  For example code "57200" would actually be "572.0" in the list of ICD9 codes, which are presented in numerical order by code.

            2) Tables of Contents Using the "At a Glance" Table of Contents for Major Categories in ODG, it is possible to identify the major classification, e.g., "Diseases of the Digestive System".  Then, the "At a Glance" Table of Contents for 3-digit Group Level Ranges will reveal the major sub-heads, such as "Appendicitis (540-543)".  Reading the detail within the sub-head will help locate the correct code because inclusions and exclusions are spelled out.

            3) Keyword Index At the end of ODG is an "ICD9 Keyword Index", which takes every word in the description plus the inclusions for an ICD9 code, and identifies that code along with its page number in the main section of ODG.  The Keyword Index also contains procedures referred to in the "Best Practices" section of each diagnosis.  Reviewing the information in the main section will confirm that the code is correct, as well as allowing access to the disability duration data for that code.

            It should be noted that, whereas the main section of ODG includes ICD-9-CM "Volume 1, Tabular List" verbatim, the ODG  "Keyword Index" is not the same as the ICD-9-CM, "Volume 2, Index to Diseases and Injuries".  The ODG index allows more powerful searching; for example, the word "rotator" will help locate "rotator cuff tendonitis" in ODG and in Volume 1 of ICD9, but there is no such heading in Volume 2 of ICD9.

EVALUATING THE DATA

            The ICD9 headings in ODG are complete -- every bit of detail in the official version is included.  The raw data appears wherever there is collected data for that ICD9 code.  This means some codes may have information from all three, from two, from only one, or not from any of the three databases.  Of course, if no data appears that probably means that the diagnosis is very rare, it is unlikely that you would ever look up information for that diagnosis, and if you did, there is probably not enough experience to develop a reasonable estimate.

            It is also possible that the data may not be available at the detailed diagnosis level, and a higher level, broader diagnosis should be checked.  For example, HCUP hospital length-of-stay data are usually available at the 4-digit level, but the disability duration data may not be.  It may be necessary to go up to the 3-digit level to find disability duration data.  Sometimes the disability duration data is clustered at a 4-digit level ending in "9", for "unspecified".  It may even be necessary to go back to a group level (a range of 3-digit codes) to find OSHA data, because the OSHA form is not always as specific in the diagnosis as ICD9 codes allow.

            In evaluating the data that does appear, it is important to note the number of cases.  Data is presented whenever it exists, so some sample sizes are large and some are small.  The size of the sample is valuable for determining the relative frequency of a diagnosis.  If it is very small, the diagnosis is rare.  Perhaps that is not the best diagnosis to select, and there is a better, more common one not far away.  The number of cases is also important in evaluating how well that data should predict disability durations in every case.

            The bar charts showing detail are valuable in determining how to apply the data in your own situation.  For example, the data may clump around several different points, indicating that there may be other variables, which could make the duration, be either "short" or "long".  The most common point, the highest bar or the "mode", is also useful.  For example, a disease may have a median duration of 2 days, but 1 day is the most likely duration for most cases.

            Differences in the three databases for the same or similar conditions are also worth noting. For some diagnoses, for example Carpal Tunnel Syndrome, the OSHA days are longer than the self-reported days, possibly because incentives under Workers’ Comp may delay how quickly workers return to work when their injury is related to their work, than if it might have been caused outside of work.  Since the OSHA data is occupationally related, the incidences are much higher for injuries and illnesses that are "work-related".

            The HCUP Hospital length of stay is provided as another benchmark for comparison with the two disability duration databases.  For conditions in which hospitalization is the norm, the hospital length of stay may show how much of the total length of disability may be spent in the hospital.  For many conditions this will not be true since the hospital data only shows cases where admission to hospital occurred, whereas the disability duration data shows all cases.  In situations where the number of HCUP hospital cases is much smaller than the number of disability duration cases, it would appear that hospitalization is unlikely.

OTHER VERSIONS OF ODG

            Official Disability Guidelines is also available in a "Top 200 Conditions" version.  This book covers those diagnoses that represent about 75% of all lost workdays.  With its smaller size, it is more portable than the complete edition, and it is easier to find the more common conditions.  Of course, information on the vast majority of conditions is missing.  Raw data from ODG is also available for licensing on computer tape or CD-ROM.  With electronic versions of Official Disability Guidelines, users can link into the ODG disability duration data as they bring up each of their computerized cases.  Of course, a Web-based version is available, so that users can access ODG from any location using a secure login and password.   ODG is also available as an integrated treatment disability duration guideline, ODG Treatment in Workers’ Comp (ODG – TWC).

            For assistance in using this publication, or information on other services, please call 1-800-488-5548.

COPYRIGHT PAGE

Copyright © 2008 by Work Loss Data Institute

Printed and bound in the United States

“Official Disability Guidelines” and “ODG” are trademarks of Work Loss Data Institute

Work Loss Data Institute is an independent organization dedicated to providing comprehensive evidence-based disability information.  All rights reserved.  No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without prior written permission from the publisher.  While the publisher and the publisher's editorial contributors have attempted to provide accurate information, they freely admit the possibility of error, and will not accept liability for incorrect information.  The publisher is not engaged in rendering medical, legal or other professional advice.  These publications are guidelines, not inflexible proscriptions, and they should not be used as sole evidence for an absolute standard of care.  Guidelines can assist clinicians in making decisions for specific conditions and also help payors make reimbursement determinations, but they cannot take into account the uniqueness of each patient's clinical circumstances.

[home]

Send mail to our webmaster with questions or comments about this web site.
Copyright © 2009 Official Disability Guidelines
Last modified: May 25, 2010