International Construction Specialty Conference of the Canadian Society for Civil Engineering (ICSC) (5th : 2015)

Development of a cost normalization procedure for National Health Care Facility Benchmarking Sharma, Vivek; Yun, Sungmin; Oliveira, Daniel P.; Mulva, Stephen P.; Caldas, Carlos H. Jun 30, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52660-Sharma_V_et_al_ICSC15_290_Development_Of_A.pdf [ 950.33kB ]
52660-Sharma_V_et_al_ICSC15_290_Development_Of_A_slides.pdf [ 459.53kB ]
Metadata
JSON: 52660-1.0076452.json
JSON-LD: 52660-1.0076452-ld.json
RDF/XML (Pretty): 52660-1.0076452-rdf.xml
RDF/JSON: 52660-1.0076452-rdf.json
Turtle: 52660-1.0076452-turtle.txt
N-Triples: 52660-1.0076452-rdf-ntriples.txt
Original Record: 52660-1.0076452-source.json
Full Text
52660-1.0076452-fulltext.txt
Citation
52660-1.0076452.ris

Full Text

5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   DEVELOPMENT OF A COST NORMALIZATION PROCEDURE FOR NATIONAL HEALTH CARE FACILITY BENCHMARKING  Vivek Sharma1,2,3,4, Sungmin Yun3, Daniel P. Oliveira3 Stephen P. Mulva3, Carlos H. Caldas1 1 Department of Civil, Architectural, and Environmental Engineering, University of Texas at Austin, USA 2 Department of Engineering Technology, Texas State University, USA  3 Construction Industry Institute, University of Texas at Austin, USA 4 vivek.sharma@cii.utexas.edu Abstract: This paper presents a cost normalization framework for the National Health Care Facilities Benchmarking program developed by Construction Industry Institute (CII). Since 2009, The CII has been engaged in developing this benchmarking program with government and industry participants.  In the effort, the methodology and process were developed to measure health care capital project performance in terms of cost, schedule, change, space, and best practice. Hospitals are complex building systems that are becoming more challenging with ever changing codes and regulations. Reliable comparison of hospitals built in different regions requires unique normalization approach tailored specifically for health care facilities. A single cost index had not achieved the desired results; therefore a combination of indices was employed to normalize various factors for proper benchmarking. In addition to location, time and currency, space was included for proper benchmarking and performance assessment. This paper covers the issues and challenges of normalizing the costs and spaces associated with health care capital projects, and present a practical example on how cost normalization is applied to a health care capital project. Challenges and considerations which are associated with cost indices applicable to cost normalization are also discussed. This study contributes to a better understanding of cost normalization amongst health care capital projects. Keyword: Health Care, Benchmarking, Normalization, Capital Projects 1 INTRODUCTION A methodology adopted for normalization for a first of its kind National Health Care Capital Facility Benchmarking program was developed at Construction Industry Institute (CII). CII has been one of the leading organizations in capital project benchmarking for the last 17 years. CII’s Performance Assessment (previously the CII Benchmarking and Metrics Program) has a database of more than 2300 projects worth over $300 billion of total installed cost (TIC) (CII, 2014).The Health care Benchmarking program is a collaborative effort amongst The University of Texas at Austin’s (UT) Construction Industry Institute (CII), the U.S. Department of Defense (DoD) / Defense Health Agency (DHA), and the U.S. Department of Veteran Affairs (VA) among other health care industry leaders. CII’s health care benchmarking program is one of the many industry specific performance assessment programs.     290-1 According to Noah (Kahn, 2009), health care construction has been less affected by economy’s fluctuations than residential and non-residential construction sectors. He also mentioned, however, that recent upheaval in economy has forced everyone to re-evaluate health care system capital plans. Therefore, health care industry is looking at means to improve project delivery and process controls (Kahn, 2009). Health care costs continue to grow faster than the economy, and health care is forecasted to be 19.3 percent of Gross Domestic Product by year 2023, up from 17.4 percent in 2013 (CMMS, 2015).  Health care projects are the most complex facilities to design, construct, and operate in Architectural/Engineering/Consultants industry (Enache-Pommer et al., 2010). The proposed CII health care benchmarking research first developed a metrics framework to allow meaningful comparisons of buildings from different areas, regions, and climate/code zones. One of the key components to allow meaningful comparisons in the benchmarking program is normalization methodology. The relative metrics such as cost growth do not require cost normalization for comparison; however in order to benchmark absolute metrics (TIC per square foot) cost needs to be normalized for currency, location and time. The Health Care Facility Benchmarking program has 223 health care specific metrics including 102 absolute metrics. This paper highlights challenges associated with cost normalization for health care facility benchmarking, and describes the procedure associated with cost normalization. 2 LITERATURE REVIEW It is essential to select appropriate indices for cost normalization in benchmarking. Previous studies have outlined various merits and demerits of published cost indices (Dai et al. 2012, McCabe et al. 2002, and Remer et al. 2008, Nasir et. al., 2014). After comprehensive literature review, two types of indices were identified and categorized as input-based cost indices and output-based price indices.  Input-based indices measure various construction process inputs, such as materials, equipment and labor hours usually referred as change in prices of a fixed basket of inputs. An input index measures the changes in the cost of resources to a contractor and it does not record the change in price to a client. It doesn’t account for technological innovations, productivity changes, contractor’s overhead and profit margins (Mohammadian and Seymour, 1997). In spite of input cost index disadvantages it can be used for identifying trends in resource costs and cost fluctuation in contracts.  Output-based price indices represent price to a client, and therefore is a direct measure of inflation. Output-based price indices are developed through model price index, hedonic price index, and bid/unit price index (Mohammadian and Seymour, 1997). The output-based indices under model price index compare the construction cost of a hypothetical structure by location and/or time (Dai et al., 2012). For example, a Producer Price Index (PPI) for an industry is a measure of changes in prices received for the industry's output sold outside the industry (Bureau of labor Statistics, 2015). Bureau of labor Statistics (BLS) producer price index for new health care building construction (NAICS code 236224) follows North American Industry Classification System (NAICS) index codes is a model price index (Bureau of labor Statistics, 2015). Model price index allows for construction heterogeneity by modeling different common building types, and is more sensitive to market change conditions.   Hedonic price index includes quality of the final product as a measure to construct an index (Mohammadian and Seymour, 1997). In United States, the hedonic price index is used to construct indices for single family homes. Hedonic price indexes may be considered a type of component pricing where the component prices are estimated from a cross-section regression. The indices based on hedonic characteristics for other construction sectors have been largely unsuccessful (Pieper, 1991). Considering the indices characteristics, the model price index under output-based price indices synchronizes well with the construction industry.   Time and location adjustment will cover for escalation and inflation adjustment. Various indices provide multiple options to accomplish time and location adjustment. According to the study on construction cost sources (Table 1) conducted by McCabe et al. (2002), the selection and use of location adjustment 290-2 indices (input or output) sometimes appears to be unreliable, resulting in significant variation in cost adjustment.  For time adjustment indices, few studies have reviewed industry specific approach for cost normalization, and in particular there are no past studies for health care facilities cost normalization. Therefore, this study reviewed and compared the existing indices to be used for health care facilities’ cost normalization including Turner Cost Index (TCI, 2015), R.S. Means (RSM-CCI, 2014) Construction Cost Index, Engineering News Record Construction Cost Index (ENR-CCI, 2014), Engineering News Record Building Cost Index (ENR-BCI, 2014) and Rider Levett Bucknall (RLB, 2014) Construction Cost Index. Table 1 summarizes the nature of indices along with process inputs considered by different indices.  Table 1: Summary of Cost Sources (Adopted from McCabe et al., 2002)  Cost Source Cost Type Labor  Input Material Input Equipment Input Assemblies Input Location  Indices ENR Yes Yes Yes No Input Hanscomb Yes No No Yes Input and Output RS Means Yes Yes Yes Yes Yes RLB Yes Yes Yes Yes Output  3 COST NORMALIZATION METHODOLOGY APPROACH In CII’s experience of developing similar systems for other industry, the key to success has been participation by industry. The development of health care system followed the same principle. The industry leaders come together to develop a system that is defined by the industry for the industry (Mulva et al., 2014). The normalization process also followed the same model as shown in Figure1.                         Figure 1: Normalization Methodology Approach Healthcare Steering Committee  CII PA Committee  Literature Review  Normalization Methodology 290-3 ENR-CCI and ENR-BCI showed very little variation to economic downturn. ENR uses year 1913 as its reference year. RLB Construction Cost Index presents comparative cost of construction in 12 cities on a quarterly basis. According to RLB (2014), comparative cost index is built on bid cost comprised of labor, material, contractor/subcontractor overhead cost and profit. This index also includes sales and use tax for construction contracts (RLB, 2014). The reference year for RLB index is 2001. CII Performance Assessment (PA) committee and health care steering committee members provide their feedback and share their organization experience on the normalization methodology. In a similar exercise, steering committee participated in a study comparing the behaviour of various indices in relation to 2007-2009 economic slowdown. Based on the literature review, comparison of cost indices and industry experts’ feedback, the cost normalization procedure was developed.  The ability to adjust location to 930 cities, and a relatively recent reference year of 1993 made RSMeans a prime index for location adjustment. It was also used in past studies by CII with success. The time adjustment was preferred through RLB with a reference year of 2001. Its ability to include profit and sales tax meant market conditions were also adjusted with time.   3.2 Cost Normalization Procedure for Health Care Facilities  The comprehensive approach is outlined in a flow chart in Figure 3. The approach to a location that is not covered in the selected index is similar to Hwang et al. (2008). The closest city was chosen based on industry experience. During the data validation process the selection is verified again. The closest city approach is also recommended by RSMeans. RSMeans being an input index, using mid-point of construction for location adjustment is an appropriate approach. In other words, it moves composite average of various construction costs to the new location. RSMeans City Cost Index (RSMCCI) is used for location adjustment in health care normalization.  After the selection of RLB index by steering committee, Bureau of labor Statistics producer price index (BLS-PPI) for new health care building construction was launched. BLS-PPI was presented to the steering committee as an option since it is a health care specific index. BLS-PPI, an output-based model price index based on health care construction was immediately preferred over other methods.  The steering committee agreed on BLS-PPI for health care, but PPI index had just come into existence in the year 2012. Therefore a hybrid index was created by CII research team with base year as 2012 for normalizing cost for time.   The time adjustment is done using a hybrid index of RLB Construction Cost Index and BLS-PPI for new health care building construction NAICS 236224. Among all industry experts in the steering committee, there was a consensus on hybrid approach with RLB for the years before 2012.   3.2.1 Currency Conversion The health care benchmarking system was designed for global benchmarking for health care facilities. Owners are required to provide currency exchange rate to allow for conversion to US currency. Exchange rate is also verified during validation process. This part is designed to accommodate future expansion of health care benchmarking to include foreign health care owners and contractors. National Health Care Benchmarking program project cost is adjusted from the project location to Chicago, which is the baseline location for the health care system. The reason for selecting Chicago as the reference city for health care normalization is to maintain consistency with CII’s use of Hanscomb Means International Construction Cost index (HMICCI) for international projects in all its previous benchmarking systems. HMICCI also uses Chicago as its reference city. This will allow seamless expansion of health care benchmarking program to include foreign projects.  290-5 3.2.1 Location Adjustment The health care benchmarking system employs an online Performance Assessment System (PAS) for data entry. The mid-point of the construction phase is a mandatory data field before the data entered can be submitted. Without the mid-point of the construction phase, normalization is not possible as the system would not allow benchmarking associate to submit projects into database. However, there are exceptions to mid-point of the construction phase that are only discussed during the validation process. And such exceptions are beyond the scope of this paper. The second mandatory data field is the location of a project job-site. In the absence of a location, the costs are normalized to the combined national average of RLB/BLS-PPI construction cost indices. For a case where mid-point and location is available, RSMeans CCI is used for location adjustment. 3.2.2 Time Adjustment As discussed earlier, the steering committee for health care benchmarking and CII decided to use RLB/BLS-PPI for time adjustment for health care normalization. The BLS-PPI for health care was launched in the year 2012, and therefore, most health care organizations are not familiar with it. The index for NAICS 236224, New Health Care Building Construction, is the latest measure developed and published by PPI as a part of its Nonresidential Building Construction (NRBC) initiative (BLS.gov, 2015). Under NRBC initiative indices yield a national weighted average of output price changes. According to BLS website, BLS-PPI includes an array of health care buildings such as hospitals, mental hospitals, infirmaries, hospital infrastructure (buildings for radiology, CT/MRI, radiation therapy, etc.), medical clinics, medical offices, medical labs, doctor and dentist offices, outpatient clinics, research labs (non-manufacturing, non-educational, or non-hospital), nursing homes, hospices, orphan homes, sanatoriums, drug clinics, rehabilitation centers, rest homes, and adult day-care centers. BLS-PPI is a type of model price index where these models represent typical health care buildings constructed in each of the four major census regions. According to BLS website, multiple health care models were developed to accommodate regional variations in building design. CII health care benchmarking program also follows the same four census regions to categorize projects geographically.  4 NORMALIZATION PROCEDURE EXAMPLE A hypothetical example is presented for better understanding of the cost normalization procedure for health care facilities. Amongst diverse cost types, TIC is used in an example to present normalization procedure. The different types of costs such as procurement cost, management cost etc. are adjusted with different indices. The currency conversion is excluded as health care benchmarking program is focused within the U.S.  A hypothetical health care project (TIC= $50,000,000) in Indianapolis is considered for this example. Assuming the project was built in 2007, and associated costs need to be normalized to year 2014. The breakdown of the cost is presented as shown in Table 2. Table 2: Breakdown of Cost  Cost Category Local Currency Total Installed Cost (TIC) $50,000,000 Capital Medical Equipment $10,000,000 Total A/E and Construction management Cost $5,000,000  Total installed cost (TIC) is defined as the total actual project cost (excluding the cost of land) from Programming/Front-end Planning through commissioning, including capitalized amounts expended for in-house salaries, overhead, travel, etc.    290-7 • Step 1: Location Adjustment (from Indianapolis to Chicago in 2007)  The procurement and management cost are subtracted from TIC before location adjustment (Table 3). The assumption is that equipment and management cost doesn’t vary with location as the equipment procured and the design teams are generally not located at the location of the construction site of health care facility.  Table 3: TIC Adjustment for Location Cost Category Amount in 2007 Indianapolis TIC $50,000,000 Capital Medical Equipment $10,000,000 Total A/E and Construction management Cost $5,000,000 Net Cost to be normalized for location ($50,000,000-$10,000,000-$5,000,000) = $35,000,000  $35,000,000 out of TIC of $50,000,000 will be adjusted for location (Table 4). The rest is capital medical equipment and management cost. These cost items are only adjusted for time.   Table 4: TIC Adjustment for Location Cost Category Amount (2007) Net Cost to be normalized for location (Indianapolis2007) $35,000,000 Location Index Indianapolis - 2007 158.5 Location Index Chicago - 2007 191.9 Location Adjustment =35000000 x (191.9/158.5) Adjusted TIC (Chicago 2007)) = $42,375,394   • Step 2: Time Adjustment (from 2007 to 2014 in Chicago using RLB/BLS-PPI)   With this normalization procedure, the health care project database allows for meaningful comparison among projects within the health care framework. The health care framework and algorithms attributed to selecting similar projects are essential in comparisons of health care projects constructed in different location and/or time. Time and location adjustment of various costs is also critical for comparisons. The current process has been validated by the steering committee for health care benchmarking which is comprised of industry experts. Table 5 presents three costs that need time adjustment. They are adjusted TIC (Chicago 2007), capital medical equipment and management cost which are still in Indianapolis dollars (2007).             290-8 Table 5: TIC Adjustment for Time Cost Category Total Amount 2007 Adjusted TIC (Chicago 2007) $42,375,394 Capital Medical Equipment (Indianapolis 2007) $10,000,000 Total A/E and Construction management Cost (Indianapolis (2007) $5,000,000 Total Cost to be normalized for Time ($42,375,394 +$10,000,000+$5,000,000) = $57,375,394.32   Total cost $57,375,394.32 in 2007 needs to be adjusted to year 2014. The adjusted TIC (Chicago 2007), capital medical equipment and management cost can also be adjusted for time separately and added later. It will provide the same result.   Table 6: TIC Adjustment for Time Cost Category Amount in 2007 Indianapolis Amount in 2014 Chicago Net Cost to be normalized for Time $57,375,394.32   RLB/PLS-PPI Index Chicago 94.8 104.5 Time Adjustment = $57,375,394.32 x (104.5/94.8)  $63,246,083.40  As shown in Table 6, $63,246,083.40 is an adjusted TIC normalized to year 2014and city Chicago. The project’s actual TIC was $50,000,000 in the year 2007 constructed in Indianapolis.    5 CONCLUSION AND PATH FORWARD Normalization is essential for meaningful comparison. Through literature review on existing cost indices and feedbacks from industry experts, the cost normalization procedures was designed to normalize costs associated with health care projects for the first time. Its efficacy and accuracy can only be validated over time. As stated earlier, the initial validation from health care industry experts have been positive. Health care normalization procedure is different than other industry specific benchmarking systems developed by CII. The published indices have inherent challenges based on their accuracy, compilation method, and reference years. Indices are also not able to capture cost variations due to the regulatory environment, site conditions, and change in code requirements.   Current normalization procedure is susceptible to these drawbacks. However, there are improvements over the past in the procedure developed for the health care benchmarking. The use of RLB/BLS-PPI allows for market conditions to be captured for the first time. Health care benchmarking system collects costs by Construction Specification Institute’s (CSI) divisions. It uses the RSMeans - CSI cost index.  Using the CSI index, health care benchmarking system could accommodate productivity data for numerous trades. As in the past, it was a challenge and a shortcoming for normalization procedures as they lacked indices supporting productivity data. According to Goodrum (2001), the CSI index provides unit labor costs, unit equipment costs and physical output data. Labor costs are based on the average wage rates from 30 US cities and the equipment costs on rental rates plus operating costs (Goodrum, 2001).  The use of RSMeans - CSI index will allow for productivity factors in normalizing costs for the first time in CII’s history. The adjustment of space for the final metric calculations is also unique to health care 290-9 benchmarking program. As previous studies indicated that the selection of an appropriate indices are critical to valid methodology. A shift in model price index for time adjustment is a positive step in health care normalization methodology. The current scope of this paper did not cover the methodology for CSI and space based adjustment in detail. Study will next focus on challenges associated with CSI and space based adjustment for health care normalization methodology.  References Bureau of Labor Statistics (BLS), 2015, Producer Price Indexes. Retrieved from http://www.bls.gov/ppi/ppinaics236224.htm  (last accessed on February 12, 2015) Centers for Medicare and Medicaid Services (CMMS), 2015. NHE Fact Sheet                           http://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData/NHE-Fact-Sheet.html (last accessed on February 03, 2015) Construction Industry Institute (CII), 2014. 2013 Annual Report, Retrieved from https://www.construction-institute.org/cii_ar13.pdf (last accessed on February 03, 2015) Dai, J., Mulva, S., Suk, S. J., Kang, Y., 2012. Cost Normalization for Global Capital Projects Benchmarking, Construction Research Congress, ASCE 2012, PP. 2400-2409 ENR Building Cost Index (ENR-BCI) 2014. Quarterly Building Cost Index. Dodge Data and Analytics.  Two Penn Plaza, 9th Floor, New York, NY. ENR Construction Cost Index (ENR-CCI) 2014. Construction Cost Annual Average. Dodge Data and Analytics. Two Penn Plaza, 9th Floor, New York, NY. Enache-Pommer, E., Horman, M., Messner, J., and Riley, D. 2010. A Unified Process Approach to Healthcare Project Delivery: Synergies between Greening Strategies, Lean Principles, and BIM. Construction Research Congress 2010: pp. 1376-1405. Goodrum, P. 2001. The impact of equipment technology on productivity in the US construction industry, PhD Dissertation, The University of Texas. Hwang, B., Thomas, S.R., Degezelle, D. and Caldas, C.H. 2008, Development of a Benchmarking Framework for Pharmaceutical Capital Projects, Taylor & Francis, Construction Management and Economics, 26(2), 177-19. Kahn N. 2009. National Health Care Capital Project Benchmarking--an owner's perspective. HERD, Volume 3 pp. 5, ISSN 1937-5867 McCabe, B.Y., O’Grady, J., and Walker, F. 2002. A Study of Construction Cost Sources,  Annual Conference of the Canadian Society for Civil Engineering, Montréal, Québec, Canada, June 5-8, 2002. Mohammadian R. and Seymour S., 1997. An Analysis of Some Construction Price Index Methodologies, Price Division, Statistics Canada, Catalogue No. 62F0014MPB, No. 2, ISBN: 0-660-59261-4 Mulva, S., Sharma V., and Yun S., 2014. National Health Care Facility Benchmarking Summary Report, 2011-2013, Construction Industry Institute, The University of Texas, Austin, TX. Nasir H., Ahmed H., Haas C. & Goodrum P. M., 2014. An analysis of construction productivity differences between Canada and the United States, Construction Management and Economics, 32:6, 595-607 Pieper, P., 1991. The Measurement of Construction Prices: Retrospect and Prospect, Department of Economics, University of Illinois at Chicago, IL, U.S.A. Remer, D.S., Lin, S., Yu, N. and Hsin, K., 2008. An update on cost and scale-up factors, international inflation indexes and location factors. International Journal of Production Economics, 114, pp 333-346. Rider Levett Bucknall (RLB), 2014. Riders Digest 2014, USA Edition, 4343 East Camelback Road, Suite 350, Phoenix, Arizona 85018.  R.S. Means (RSM-CCI) 2014. Quarterly Means Construction Cost Index. Reed Construction Data.  700 Longwater Dr., Norwell, MA. Turner Cost Index (TCI), 2015. Retrieved from http://www.turnerconstruction.com/cost-index. (last accessed on February 07, 2015)   290-10  5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   DEVELOPMENT OF A COST NORMALIZATION PROCEDURE FOR NATIONAL HEALTH CARE FACILITY BENCHMARKING  Vivek Sharma1,2,3,4, Sungmin Yun3, Daniel P. Oliveira3 Stephen P. Mulva3, Carlos H. Caldas1 1 Department of Civil, Architectural, and Environmental Engineering, University of Texas at Austin, USA 2 Department of Engineering Technology, Texas State University, USA  3 Construction Industry Institute, University of Texas at Austin, USA 4 vivek.sharma@cii.utexas.edu Abstract: This paper presents a cost normalization framework for the National Health Care Facilities Benchmarking program developed by Construction Industry Institute (CII). Since 2009, The CII has been engaged in developing this benchmarking program with government and industry participants.  In the effort, the methodology and process were developed to measure health care capital project performance in terms of cost, schedule, change, space, and best practice. Hospitals are complex building systems that are becoming more challenging with ever changing codes and regulations. Reliable comparison of hospitals built in different regions requires unique normalization approach tailored specifically for health care facilities. A single cost index had not achieved the desired results; therefore a combination of indices was employed to normalize various factors for proper benchmarking. In addition to location, time and currency, space was included for proper benchmarking and performance assessment. This paper covers the issues and challenges of normalizing the costs and spaces associated with health care capital projects, and present a practical example on how cost normalization is applied to a health care capital project. Challenges and considerations which are associated with cost indices applicable to cost normalization are also discussed. This study contributes to a better understanding of cost normalization amongst health care capital projects. Keyword: Health Care, Benchmarking, Normalization, Capital Projects 1 INTRODUCTION A methodology adopted for normalization for a first of its kind National Health Care Capital Facility Benchmarking program was developed at Construction Industry Institute (CII). CII has been one of the leading organizations in capital project benchmarking for the last 17 years. CII’s Performance Assessment (previously the CII Benchmarking and Metrics Program) has a database of more than 2300 projects worth over $300 billion of total installed cost (TIC) (CII, 2014).The Health care Benchmarking program is a collaborative effort amongst The University of Texas at Austin’s (UT) Construction Industry Institute (CII), the U.S. Department of Defense (DoD) / Defense Health Agency (DHA), and the U.S. Department of Veteran Affairs (VA) among other health care industry leaders. CII’s health care benchmarking program is one of the many industry specific performance assessment programs.     290-1 According to Noah (Kahn, 2009), health care construction has been less affected by economy’s fluctuations than residential and non-residential construction sectors. He also mentioned, however, that recent upheaval in economy has forced everyone to re-evaluate health care system capital plans. Therefore, health care industry is looking at means to improve project delivery and process controls (Kahn, 2009). Health care costs continue to grow faster than the economy, and health care is forecasted to be 19.3 percent of Gross Domestic Product by year 2023, up from 17.4 percent in 2013 (CMMS, 2015).  Health care projects are the most complex facilities to design, construct, and operate in Architectural/Engineering/Consultants industry (Enache-Pommer et al., 2010). The proposed CII health care benchmarking research first developed a metrics framework to allow meaningful comparisons of buildings from different areas, regions, and climate/code zones. One of the key components to allow meaningful comparisons in the benchmarking program is normalization methodology. The relative metrics such as cost growth do not require cost normalization for comparison; however in order to benchmark absolute metrics (TIC per square foot) cost needs to be normalized for currency, location and time. The Health Care Facility Benchmarking program has 223 health care specific metrics including 102 absolute metrics. This paper highlights challenges associated with cost normalization for health care facility benchmarking, and describes the procedure associated with cost normalization. 2 LITERATURE REVIEW It is essential to select appropriate indices for cost normalization in benchmarking. Previous studies have outlined various merits and demerits of published cost indices (Dai et al. 2012, McCabe et al. 2002, and Remer et al. 2008, Nasir et. al., 2014). After comprehensive literature review, two types of indices were identified and categorized as input-based cost indices and output-based price indices.  Input-based indices measure various construction process inputs, such as materials, equipment and labor hours usually referred as change in prices of a fixed basket of inputs. An input index measures the changes in the cost of resources to a contractor and it does not record the change in price to a client. It doesn’t account for technological innovations, productivity changes, contractor’s overhead and profit margins (Mohammadian and Seymour, 1997). In spite of input cost index disadvantages it can be used for identifying trends in resource costs and cost fluctuation in contracts.  Output-based price indices represent price to a client, and therefore is a direct measure of inflation. Output-based price indices are developed through model price index, hedonic price index, and bid/unit price index (Mohammadian and Seymour, 1997). The output-based indices under model price index compare the construction cost of a hypothetical structure by location and/or time (Dai et al., 2012). For example, a Producer Price Index (PPI) for an industry is a measure of changes in prices received for the industry's output sold outside the industry (Bureau of labor Statistics, 2015). Bureau of labor Statistics (BLS) producer price index for new health care building construction (NAICS code 236224) follows North American Industry Classification System (NAICS) index codes is a model price index (Bureau of labor Statistics, 2015). Model price index allows for construction heterogeneity by modeling different common building types, and is more sensitive to market change conditions.   Hedonic price index includes quality of the final product as a measure to construct an index (Mohammadian and Seymour, 1997). In United States, the hedonic price index is used to construct indices for single family homes. Hedonic price indexes may be considered a type of component pricing where the component prices are estimated from a cross-section regression. The indices based on hedonic characteristics for other construction sectors have been largely unsuccessful (Pieper, 1991). Considering the indices characteristics, the model price index under output-based price indices synchronizes well with the construction industry.   Time and location adjustment will cover for escalation and inflation adjustment. Various indices provide multiple options to accomplish time and location adjustment. According to the study on construction cost sources (Table 1) conducted by McCabe et al. (2002), the selection and use of location adjustment 290-2 indices (input or output) sometimes appears to be unreliable, resulting in significant variation in cost adjustment.  For time adjustment indices, few studies have reviewed industry specific approach for cost normalization, and in particular there are no past studies for health care facilities cost normalization. Therefore, this study reviewed and compared the existing indices to be used for health care facilities’ cost normalization including Turner Cost Index (TCI, 2015), R.S. Means (RSM-CCI, 2014) Construction Cost Index, Engineering News Record Construction Cost Index (ENR-CCI, 2014), Engineering News Record Building Cost Index (ENR-BCI, 2014) and Rider Levett Bucknall (RLB, 2014) Construction Cost Index. Table 1 summarizes the nature of indices along with process inputs considered by different indices.  Table 1: Summary of Cost Sources (Adopted from McCabe et al., 2002)  Cost Source Cost Type Labor  Input Material Input Equipment Input Assemblies Input Location  Indices ENR Yes Yes Yes No Input Hanscomb Yes No No Yes Input and Output RS Means Yes Yes Yes Yes Yes RLB Yes Yes Yes Yes Output  3 COST NORMALIZATION METHODOLOGY APPROACH In CII’s experience of developing similar systems for other industry, the key to success has been participation by industry. The development of health care system followed the same principle. The industry leaders come together to develop a system that is defined by the industry for the industry (Mulva et al., 2014). The normalization process also followed the same model as shown in Figure1.                         Figure 1: Normalization Methodology Approach Healthcare Steering Committee  CII PA Committee  Literature Review  Normalization Methodology 290-3 ENR-CCI and ENR-BCI showed very little variation to economic downturn. ENR uses year 1913 as its reference year. RLB Construction Cost Index presents comparative cost of construction in 12 cities on a quarterly basis. According to RLB (2014), comparative cost index is built on bid cost comprised of labor, material, contractor/subcontractor overhead cost and profit. This index also includes sales and use tax for construction contracts (RLB, 2014). The reference year for RLB index is 2001. CII Performance Assessment (PA) committee and health care steering committee members provide their feedback and share their organization experience on the normalization methodology. In a similar exercise, steering committee participated in a study comparing the behaviour of various indices in relation to 2007-2009 economic slowdown. Based on the literature review, comparison of cost indices and industry experts’ feedback, the cost normalization procedure was developed.  The ability to adjust location to 930 cities, and a relatively recent reference year of 1993 made RSMeans a prime index for location adjustment. It was also used in past studies by CII with success. The time adjustment was preferred through RLB with a reference year of 2001. Its ability to include profit and sales tax meant market conditions were also adjusted with time.   3.2 Cost Normalization Procedure for Health Care Facilities  The comprehensive approach is outlined in a flow chart in Figure 3. The approach to a location that is not covered in the selected index is similar to Hwang et al. (2008). The closest city was chosen based on industry experience. During the data validation process the selection is verified again. The closest city approach is also recommended by RSMeans. RSMeans being an input index, using mid-point of construction for location adjustment is an appropriate approach. In other words, it moves composite average of various construction costs to the new location. RSMeans City Cost Index (RSMCCI) is used for location adjustment in health care normalization.  After the selection of RLB index by steering committee, Bureau of labor Statistics producer price index (BLS-PPI) for new health care building construction was launched. BLS-PPI was presented to the steering committee as an option since it is a health care specific index. BLS-PPI, an output-based model price index based on health care construction was immediately preferred over other methods.  The steering committee agreed on BLS-PPI for health care, but PPI index had just come into existence in the year 2012. Therefore a hybrid index was created by CII research team with base year as 2012 for normalizing cost for time.   The time adjustment is done using a hybrid index of RLB Construction Cost Index and BLS-PPI for new health care building construction NAICS 236224. Among all industry experts in the steering committee, there was a consensus on hybrid approach with RLB for the years before 2012.   3.2.1 Currency Conversion The health care benchmarking system was designed for global benchmarking for health care facilities. Owners are required to provide currency exchange rate to allow for conversion to US currency. Exchange rate is also verified during validation process. This part is designed to accommodate future expansion of health care benchmarking to include foreign health care owners and contractors. National Health Care Benchmarking program project cost is adjusted from the project location to Chicago, which is the baseline location for the health care system. The reason for selecting Chicago as the reference city for health care normalization is to maintain consistency with CII’s use of Hanscomb Means International Construction Cost index (HMICCI) for international projects in all its previous benchmarking systems. HMICCI also uses Chicago as its reference city. This will allow seamless expansion of health care benchmarking program to include foreign projects.  290-5 3.2.1 Location Adjustment The health care benchmarking system employs an online Performance Assessment System (PAS) for data entry. The mid-point of the construction phase is a mandatory data field before the data entered can be submitted. Without the mid-point of the construction phase, normalization is not possible as the system would not allow benchmarking associate to submit projects into database. However, there are exceptions to mid-point of the construction phase that are only discussed during the validation process. And such exceptions are beyond the scope of this paper. The second mandatory data field is the location of a project job-site. In the absence of a location, the costs are normalized to the combined national average of RLB/BLS-PPI construction cost indices. For a case where mid-point and location is available, RSMeans CCI is used for location adjustment. 3.2.2 Time Adjustment As discussed earlier, the steering committee for health care benchmarking and CII decided to use RLB/BLS-PPI for time adjustment for health care normalization. The BLS-PPI for health care was launched in the year 2012, and therefore, most health care organizations are not familiar with it. The index for NAICS 236224, New Health Care Building Construction, is the latest measure developed and published by PPI as a part of its Nonresidential Building Construction (NRBC) initiative (BLS.gov, 2015). Under NRBC initiative indices yield a national weighted average of output price changes. According to BLS website, BLS-PPI includes an array of health care buildings such as hospitals, mental hospitals, infirmaries, hospital infrastructure (buildings for radiology, CT/MRI, radiation therapy, etc.), medical clinics, medical offices, medical labs, doctor and dentist offices, outpatient clinics, research labs (non-manufacturing, non-educational, or non-hospital), nursing homes, hospices, orphan homes, sanatoriums, drug clinics, rehabilitation centers, rest homes, and adult day-care centers. BLS-PPI is a type of model price index where these models represent typical health care buildings constructed in each of the four major census regions. According to BLS website, multiple health care models were developed to accommodate regional variations in building design. CII health care benchmarking program also follows the same four census regions to categorize projects geographically.  4 NORMALIZATION PROCEDURE EXAMPLE A hypothetical example is presented for better understanding of the cost normalization procedure for health care facilities. Amongst diverse cost types, TIC is used in an example to present normalization procedure. The different types of costs such as procurement cost, management cost etc. are adjusted with different indices. The currency conversion is excluded as health care benchmarking program is focused within the U.S.  A hypothetical health care project (TIC= $50,000,000) in Indianapolis is considered for this example. Assuming the project was built in 2007, and associated costs need to be normalized to year 2014. The breakdown of the cost is presented as shown in Table 2. Table 2: Breakdown of Cost  Cost Category Local Currency Total Installed Cost (TIC) $50,000,000 Capital Medical Equipment $10,000,000 Total A/E and Construction management Cost $5,000,000  Total installed cost (TIC) is defined as the total actual project cost (excluding the cost of land) from Programming/Front-end Planning through commissioning, including capitalized amounts expended for in-house salaries, overhead, travel, etc.    290-7 • Step 1: Location Adjustment (from Indianapolis to Chicago in 2007)  The procurement and management cost are subtracted from TIC before location adjustment (Table 3). The assumption is that equipment and management cost doesn’t vary with location as the equipment procured and the design teams are generally not located at the location of the construction site of health care facility.  Table 3: TIC Adjustment for Location Cost Category Amount in 2007 Indianapolis TIC $50,000,000 Capital Medical Equipment $10,000,000 Total A/E and Construction management Cost $5,000,000 Net Cost to be normalized for location ($50,000,000-$10,000,000-$5,000,000) = $35,000,000  $35,000,000 out of TIC of $50,000,000 will be adjusted for location (Table 4). The rest is capital medical equipment and management cost. These cost items are only adjusted for time.   Table 4: TIC Adjustment for Location Cost Category Amount (2007) Net Cost to be normalized for location (Indianapolis2007) $35,000,000 Location Index Indianapolis - 2007 158.5 Location Index Chicago - 2007 191.9 Location Adjustment =35000000 x (191.9/158.5) Adjusted TIC (Chicago 2007)) = $42,375,394   • Step 2: Time Adjustment (from 2007 to 2014 in Chicago using RLB/BLS-PPI)   With this normalization procedure, the health care project database allows for meaningful comparison among projects within the health care framework. The health care framework and algorithms attributed to selecting similar projects are essential in comparisons of health care projects constructed in different location and/or time. Time and location adjustment of various costs is also critical for comparisons. The current process has been validated by the steering committee for health care benchmarking which is comprised of industry experts. Table 5 presents three costs that need time adjustment. They are adjusted TIC (Chicago 2007), capital medical equipment and management cost which are still in Indianapolis dollars (2007).             290-8 Table 5: TIC Adjustment for Time Cost Category Total Amount 2007 Adjusted TIC (Chicago 2007) $42,375,394 Capital Medical Equipment (Indianapolis 2007) $10,000,000 Total A/E and Construction management Cost (Indianapolis (2007) $5,000,000 Total Cost to be normalized for Time ($42,375,394 +$10,000,000+$5,000,000) = $57,375,394.32   Total cost $57,375,394.32 in 2007 needs to be adjusted to year 2014. The adjusted TIC (Chicago 2007), capital medical equipment and management cost can also be adjusted for time separately and added later. It will provide the same result.   Table 6: TIC Adjustment for Time Cost Category Amount in 2007 Indianapolis Amount in 2014 Chicago Net Cost to be normalized for Time $57,375,394.32   RLB/PLS-PPI Index Chicago 94.8 104.5 Time Adjustment = $57,375,394.32 x (104.5/94.8)  $63,246,083.40  As shown in Table 6, $63,246,083.40 is an adjusted TIC normalized to year 2014and city Chicago. The project’s actual TIC was $50,000,000 in the year 2007 constructed in Indianapolis.    5 CONCLUSION AND PATH FORWARD Normalization is essential for meaningful comparison. Through literature review on existing cost indices and feedbacks from industry experts, the cost normalization procedures was designed to normalize costs associated with health care projects for the first time. Its efficacy and accuracy can only be validated over time. As stated earlier, the initial validation from health care industry experts have been positive. Health care normalization procedure is different than other industry specific benchmarking systems developed by CII. The published indices have inherent challenges based on their accuracy, compilation method, and reference years. Indices are also not able to capture cost variations due to the regulatory environment, site conditions, and change in code requirements.   Current normalization procedure is susceptible to these drawbacks. However, there are improvements over the past in the procedure developed for the health care benchmarking. The use of RLB/BLS-PPI allows for market conditions to be captured for the first time. Health care benchmarking system collects costs by Construction Specification Institute’s (CSI) divisions. It uses the RSMeans - CSI cost index.  Using the CSI index, health care benchmarking system could accommodate productivity data for numerous trades. As in the past, it was a challenge and a shortcoming for normalization procedures as they lacked indices supporting productivity data. According to Goodrum (2001), the CSI index provides unit labor costs, unit equipment costs and physical output data. Labor costs are based on the average wage rates from 30 US cities and the equipment costs on rental rates plus operating costs (Goodrum, 2001).  The use of RSMeans - CSI index will allow for productivity factors in normalizing costs for the first time in CII’s history. The adjustment of space for the final metric calculations is also unique to health care 290-9 benchmarking program. As previous studies indicated that the selection of an appropriate indices are critical to valid methodology. A shift in model price index for time adjustment is a positive step in health care normalization methodology. The current scope of this paper did not cover the methodology for CSI and space based adjustment in detail. Study will next focus on challenges associated with CSI and space based adjustment for health care normalization methodology.  References Bureau of Labor Statistics (BLS), 2015, Producer Price Indexes. Retrieved from http://www.bls.gov/ppi/ppinaics236224.htm  (last accessed on February 12, 2015) Centers for Medicare and Medicaid Services (CMMS), 2015. NHE Fact Sheet                           http://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData/NHE-Fact-Sheet.html (last accessed on February 03, 2015) Construction Industry Institute (CII), 2014. 2013 Annual Report, Retrieved from https://www.construction-institute.org/cii_ar13.pdf (last accessed on February 03, 2015) Dai, J., Mulva, S., Suk, S. J., Kang, Y., 2012. Cost Normalization for Global Capital Projects Benchmarking, Construction Research Congress, ASCE 2012, PP. 2400-2409 ENR Building Cost Index (ENR-BCI) 2014. Quarterly Building Cost Index. Dodge Data and Analytics.  Two Penn Plaza, 9th Floor, New York, NY. ENR Construction Cost Index (ENR-CCI) 2014. Construction Cost Annual Average. Dodge Data and Analytics. Two Penn Plaza, 9th Floor, New York, NY. Enache-Pommer, E., Horman, M., Messner, J., and Riley, D. 2010. A Unified Process Approach to Healthcare Project Delivery: Synergies between Greening Strategies, Lean Principles, and BIM. Construction Research Congress 2010: pp. 1376-1405. Goodrum, P. 2001. The impact of equipment technology on productivity in the US construction industry, PhD Dissertation, The University of Texas. Hwang, B., Thomas, S.R., Degezelle, D. and Caldas, C.H. 2008, Development of a Benchmarking Framework for Pharmaceutical Capital Projects, Taylor & Francis, Construction Management and Economics, 26(2), 177-19. Kahn N. 2009. National Health Care Capital Project Benchmarking--an owner's perspective. HERD, Volume 3 pp. 5, ISSN 1937-5867 McCabe, B.Y., O’Grady, J., and Walker, F. 2002. A Study of Construction Cost Sources,  Annual Conference of the Canadian Society for Civil Engineering, Montréal, Québec, Canada, June 5-8, 2002. Mohammadian R. and Seymour S., 1997. An Analysis of Some Construction Price Index Methodologies, Price Division, Statistics Canada, Catalogue No. 62F0014MPB, No. 2, ISBN: 0-660-59261-4 Mulva, S., Sharma V., and Yun S., 2014. National Health Care Facility Benchmarking Summary Report, 2011-2013, Construction Industry Institute, The University of Texas, Austin, TX. Nasir H., Ahmed H., Haas C. & Goodrum P. M., 2014. An analysis of construction productivity differences between Canada and the United States, Construction Management and Economics, 32:6, 595-607 Pieper, P., 1991. The Measurement of Construction Prices: Retrospect and Prospect, Department of Economics, University of Illinois at Chicago, IL, U.S.A. Remer, D.S., Lin, S., Yu, N. and Hsin, K., 2008. An update on cost and scale-up factors, international inflation indexes and location factors. International Journal of Production Economics, 114, pp 333-346. Rider Levett Bucknall (RLB), 2014. Riders Digest 2014, USA Edition, 4343 East Camelback Road, Suite 350, Phoenix, Arizona 85018.  R.S. Means (RSM-CCI) 2014. Quarterly Means Construction Cost Index. Reed Construction Data.  700 Longwater Dr., Norwell, MA. Turner Cost Index (TCI), 2015. Retrieved from http://www.turnerconstruction.com/cost-index. (last accessed on February 07, 2015)   290-10  2015-11-10Cost Normalization Procedure for National Health Care Facility BenchmarkingSungmin Yun, Ph.D.Construction Industry InstituteThe University of Texas at Austin1ICSC 2015 - The CSCE International Construction Specialty ConferenceVivek Sharma, Sungmin Yun*, Daniel P. Oliveira, Stephen P. Mulva, Carlos H. Caldas• National Health Care Benchmarking Program• A Challenge• Literature Review• Methodology• Health Care Normalization Method• Normalization Procedure Example• Conclusion and Path Forward2OutlineCII National Health Care Facility Benchmarking• The Health care Benchmarking program is a collaborative effort amongst : The University of Texas at Austin’s (UT) Construction Industry Institute (CII) The U.S. Department of Defense / Defense Health Agency (DHA) The U.S. Department of Veterans Affairs (VA)GOAL: To develop a methodology and process were to measure health care capital project performance in terms of cost, schedule, change, space, and best practice.• Along with various industry partners (Round1):3A Challenge!Medical Office Building (MOB) in San Francisco, US by Company AVs.Medical Office Building (MOB) in Vancouver, Canada by Company B Can we compare square foot cost (an absolute metrics) of these health care projects in a meaningful way?Hospitals are complex building systems that are becoming more challenging with ever changing codes and regulations.Cost Normalization Framework 4• Normalization is needed for cost, time, and location(Other healthcare project-specific areas such as shell space)• Key: Selection of most appropriate indices– What type of index is suitable for health care specific industry?• Input vs. Output vs. Hedonic– What does each index represent?• Materials, equipment, labor hours, market conditions, productivity, overhead and profit, hypothetical structure by location and/or time etc.5Literature Review6Literature ReviewCost IndexCost TypeLabor InputMaterialInputEquipmentInputAssembliesInputLocation IndicesENR Yes Yes Yes No InputHanscomb Yes No No Yes Input and OutputRS Means Yes Yes Yes Yes YesRLB* Yes Yes Yes Yes OutputReview of Cost Indices* Rider Levett Bucknall Construction Cost Index (Adopted from McCabe et al., 2002)7Normalization MethodologySelection of cost indices:The cost indices reviewed in the literature review were evaluated and compared over time with reference year 2008 to measure their sensitivity to economic changes during recession.In CII’s experience of developing similar systems for other industry, the key to success has been participation by industry. The industry leaders come together to develop a system that is defined by the industry for the industry (Mulva et al., 2014).8Currency ConversionLocation AdjustmentTime AdjustmentHealth Care Cost Normalization ProcedureAggregation• A hypothetical example is presented in the paper for better understanding of the cost normalization procedure for health care facilities. • Amongst diverse cost types, Total Installed Costs (TIC) is used in an example to present normalization procedure. • Case project: Hospital built at Indianapolis, IL in 20079Cost Normalization ExampleCost Category Local CurrencyTotal Installed Cost (TIC) $50,000,000Capital Medical Equipment $10,000,000Total A/E and Construction management Cost $5,000,000Cost BreakdownCost Category Amount in 2007 IndianapolisTIC $50,000,000Capital Medical Equipment $10,000,000Total A/E and Construction management Cost $5,000,000Net Cost to be normalized for location ($50,000,000-$10,000,000-$5,000,000) = $35,000,00010Cost Breakdown & Net Cost CalculationCost Category Amount (2007)Net Cost to be normalized for location (Indianapolis2007) $35,000,000Location Index Indianapolis - 2007 158.5Location Index Chicago - 2007 191.9Location Adjustment =35000000 x (191.9/158.5)Adjusted TIC (Chicago 2007)) = $42,375,394Cost Category Total Amount 2007Adjusted TIC (Chicago 2007) $42,375,394Capital Medical Equipment (Indianapolis 2007) $10,000,000Total A/E and Construction management Cost (Indianapolis (2007)$5,000,000Total Cost to be normalized for Time ($42,375,394 +$10,000,000+$5,000,000) =$57,375,394.32 Cost Category Amount in 2007 Indianapolis Amount in 2014 ChicagoNet Cost to be normalized for Time $57,375,394.32 RLB/PLS-PPI Index Chicago 94.8 104.5Time Adjustment = $57,375,394.32 x (104.5/94.8) $63,246,083.40Location Adjustment for Net Cost Combining Net Cost and Other CostsTime Adjustment from 2007 to 2014Location AdjustmentIndianapolis  Chicago• Normalization is essential for capital project benchmarking. • This method is the first cost normalization procedure specialized in health care projects considering market condition and productivity factors.• Methodologies for CSI* and Space adjustment are not included in this paper.• No index is perfect, but choose the best appropriate one for normalization 11Conclusion and Path Forward* Construction Specification Institute12

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52660.1-0076452/manifest

Comment

Related Items