This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Koroukian, S. M. Articles by Madigan, E. Search for Related Content PubMed PubMed Citation Articles by Koroukian, S. M. Articles by Madigan, E.

Comorbidity, Disability, and Geriatric Syndromes in Elderly Cancer Patients Receiving Home Health Care

Siran M. Koroukian, Patrick Murray, Elizabeth Madigan

From the Division of Health Policy, Department of Epidemiology and Biostatistics, School of Medicine, and Frances P. Bolton School of Nursing, Case Western Reserve University, and the Center for Health Care Research and Policy, MetroHealth Medical Center, Cleveland, OH

Address reprint requests to Siran M. Koroukian, PhD, Assistant Professor, Department of Epidemiology and Biostatistics, School of Medicine, Case Western Reserve University, 10900 Euclid Ave, Cleveland, OH 44106-4945; e-mail: skoroukian{at}case.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Appendix Authors' Disclosures of... Author Contributions REFERENCES

 
Purpose: To assess the prevalence of comorbidity, disability, and geriatric syndromes, or a combination thereof, in elders with cancer receiving home health care (HHC).

Patients and Methods: Using the Ohio Cancer Incidence Surveillance System, we identified Ohio residents 65 years of age or older who were diagnosed with incident breast (n = 952), prostate (n = 324), or colorectal cancer (n = 1,276) during the 28-month study period, August 1999 through November 2001. We used the Outcome and Assessment Information Set, a database compiling comprehensive assessment forms completed for all HHC patients, to group individuals in independent and overlapping categories of comorbidity, disability, and geriatric syndromes on the basis of the patients' clinical condition 14 days before the date of the assessment.

Results: The proportion with no comorbidity, disability, or geriatric syndromes was 26.4% in breast cancer patients, 12.0% in prostate cancer patients, and 14.0% in colorectal cancer patients. The proportion of patients presenting all three entities at once was 11.7%, 24.7%, and 15.7%, respectively, in three cancer sites. As expected, the proportion of patients with no comorbidity, disability, or geriatric syndromes declined gradually with increasing age, and that of patients with all three entities was highest among patients 85 years or older.

Conclusion: The proposed taxonomy will help us gain a more nuanced understanding of older cancer patients' clinical presentation and may lead to a more accurate identification of older patients who might benefit from standard cancer treatment, and those who might experience adverse outcomes.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Appendix Authors' Disclosures of... Author Contributions REFERENCES

 
The burden of cancer is borne disproportionately by the elderly¹: whereas cancer is one of the leading causes of death in the younger population, it is the leading cause of death in women aged 40 to 79 years and in men aged 60 to 79 years.² Cancer prognosis is determined by cancer type and stage at diagnosis as well as adequacy of cancer treatment and follow-up care. A number of recent studies have documented cancer-related disparities by age and comorbid conditions.^3-5

Recent studies of cancer-related outcomes in the elderly have focused on the role of functional limitations and geriatric syndromes in addition to that of comorbidities. It is widely accepted that while cancer treatment can be life saving in fully functional individuals, it can worsen geriatric syndromes⁶ or may even be life threatening to patients with reduced functional reserve⁷—hence, the importance of frailty in decisions to proceed with cancer therapy.⁸ The differential age-related cancer outcomes that have been described may result from either standard treatment that overwhelms the biologic reserves or from less than adequate treatment because an older person is erroneously perceived to be susceptible to the side effects of the treatment. Finding the correct balance between these competing issues has been a challenge to clinicians.

In this study, we build on the framework proposed by Balducci and Extermann to describe the frail elderly on the basis of the presence of any of the following criteria: age 85 years, dependence in one or more activity of daily living (ADL), one or more comorbidities, and one or more geriatric syndromes including delirium, dementia, depression, osteoporosis, incontinence, falls, neglect and abuse, and failure to thrive. We also adopt an approach detailed by Fried et al,⁹ in which they demonstrate that in a cohort of patients with cardiovascular disease, comorbidity, disability, and frailty can present independently or in conjunction with each other. We similarly posit that there may be significant overlap in older cancer patients across comorbidity, disability, and geriatric syndromes and that a more nuanced description of their clinical presentation may assist researchers in determining when less aggressive approaches to cancer treatment in the elderly are appropriate and when they may be deleterious to outcomes. The purpose of this study is to describe the rates and overlap of comorbidity, disability, and geriatric syndromes in a cohort of elderly patients with cancer receiving home health care (HHC).

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Appendix Authors' Disclosures of... Author Contributions REFERENCES

 
This is a cross-sectional study using the linked Ohio Cancer Incidence Surveillance System (OCISS) and home health care (HHC) Outcome Assessment Information Set (OASIS) database. The study protocol has been approved by the University Hospitals of Cleveland (Cleveland, OH), the Centers for Medicare & Medicaid Services (Baltimore, MD), and the Ohio Department of Health (Columbus, OH), which administers the OCISS.

Data Sources
The OCISS. Established in 1991, the OCISS is administered by the Ohio Department of Health. It includes data reported by all hospitals as well as clinics and private physician offices. With the exception of basal and squamous cell carcinoma of the skin and carcinoma-in-situ of the cervix, all primary cancers diagnosed on or after January 1, 1992 are required to be reported to the OCISS. The registry contains patient identifiers, including first name, last name, social security number (SSN), date of birth, sex, race, county, address of residence, and zip code, as well as cancer-specific data, including date of diagnosis, anatomic site, stage, and grade at diagnosis. The quality of the OCISS data has been improving since its inception. According to the assessment by the North American Association of Central Cancer Registries, the OCISS data represent from 91% to 97% of the total number of cancers expected to be diagnosed in Ohio for any given diagnosis year.

In addition to identifying incident cases of cancer in Ohio during the study period for the anatomic sites of interest, the OCISS files were used to retrieve the date of cancer diagnosis as well as the Surveillance, Epidemiology, and End Results (SEER) summary stage. The presence of patient identifiers in the OCISS files allowed linkage with the OASIS files.

HHC OASIS. The Medicare HHC benefit is designed to address postacute and other health care needs of beneficiaries who are homebound or confined to their residence.¹⁰ To qualify for Medicare's HHC, beneficiaries must be under the care of a physician and must require physical therapy, speech therapy, continued occupational therapy, or intermittent skilled nursing care.¹⁰ The OASIS assessments, which are mandatory, are made on admission to and discharge from HHC and at 60-day intervals, if the patient is not discharged from HHC.

The OASIS includes patient identifiers, including SSN, date of birth, and sex. It also includes the date of assessment, a variable indicating whether the assessment had been performed on admission, follow-up care, or discharge, variables describing the patient's status in the 14 days before the date of assessment, and variables describing the current patient status. The OASIS record is quite extensive, including variables referring to the patient's urinary and bowel incontinence, behavioral and cognitive status, including functional status, as detailed through the elements of the ADL, and Instrumental Activities of Daily Living. In addition, the OASIS carries diagnosis codes as retrieved from medical charts, making it possible to assess the patient's comorbidities.

The quality of data collected through the HHC OASIS has been assessed for some of the pertinent variables. Inter-rater reliability has been reported elsewhere.^11,12 Using the Landis and Koch definitions for kappa scores above 0.60 as substantial,¹³ all of the current ADL and Instrumental Activities of Daily Living items, frequency of confusion, urinary incontinence, and bowel incontinence had substantial inter-rater reliability.

Study Population
Patients were identified from the OCISS files, and the corresponding records were linked with that of the OASIS on a year-by-year basis, using SSN and sex. The study population in this cross-sectional study includes all residents of the state of Ohio, 65 years or older, diagnosed with incident breast, prostate, or colorectal cancer during the 30-month period, August 1999 through November 2001, and receiving HHC in the 30-day period before or after initial date of cancer diagnosis.

OASIS assessments were included in the study only if the first home health assessment of a given patient in that year occurred within 30 days from the date of cancer diagnosis, as retrieved from the OCISS. A total of 2,682 cancer patients were identified in both the OCISS and OASIS files; 130 patients were excluded because of incomplete data in the OASIS portion of the record, leaving the study population at 2,552 patients.

Study Variables
Variables retrieved from the OCISS. The OCISS file was used to identify cancer site (ie, breast, prostate, or colorectal), the SEER summary stage (ie, in situ, local, regional, distant, and unstaged/unknown stage), date of cancer diagnosis, and demographics, including age (categorized in 5-year increments), race (African American and all others), and sex for colorectal cancer patients.

Variables derived from OASIS: Comorbidity. We opted to use a listing of comorbid conditions developed by the National Institute of Aging and the National Cancer Institute (NIA/NCI). This list includes 27 major medical categories and four classes of severity: (1) current medical management or diagnosis problem, (2) condition noted, no medical management, (3) history of condition only, and (4) condition noted, but unknown whether it is a current or past condition.^3,5 Analysis revealed that conditions were documented in OASIS only if they were symptomatic and required medical management. Relative to the NIA/NCI list of comorbidities, therefore, all comorbidity conditions would fall under severity category (1) of current medical management or diagnosis problems.

Comorbidities according to the NIA/NCI list of conditions were identified in OASIS using International Coding of Diseases (9th Revision, Clinical Modification; ICD-9-CM; Table A1).

Of note is that given that the conditions of depression, delirium, incontinence, and failure to thrive in the NIA/NCI list are considered as geriatric syndromes by Balducci and Extermann, we accounted for them in the latter category.

Variables derived from OASIS: Disability. Disability was defined as the need for assistance in one or more ADLs (Table A2).

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Appendix Authors' Disclosures of... Author Contributions REFERENCES

 
We identified a total of 34,261 patients with incident female breast (n = 10,471), prostate (n = 12,510), or colorectal cancer (n = 11,280). Of those, 952 breast cancer patients (9.1%), 324 prostate cancer patients (2.6%), and 1,276 colorectal cancer patients (11.3%) also received HHC in the 30 days preceding or following their initial date of cancer diagnosis. The mean age in each of the HHC patient groups was 76.6 years, 79.3 years, and 77.9 years, respectively. Although the mean age of HHC patients was significantly higher than that of non-HHC patients, this difference was clinically meaningful only in prostate cancer patients (79.3 years v 74.6 years, respectively).

The distribution of OCISS–OASIS and OCISS–non-OASIS older cancer patients by demographic characteristics and cancer stage by cancer site is presented in Table 1. The most notable difference across the cancer sites and between OCISS–OASIS and OCISS–non-OASIS patients is the greater representation of older patients among OCISS–OASIS patients with prostate cancer. Also, the proportion of prostate cancer patients with distant stage at diagnosis was markedly higher in the OCISS–OASIS than in the OCISS–non-OASIS group (17.6% v 4.0%; P < .01). Additionally, the proportion of unstaged patient cases or those of unknown stage was considerably lower in OCISS–OASIS breast and colorectal cancer patients than their OCISS–non-OASIS counterparts. No other meaningful differences could be noted between OCISS–OASIS and OCISS–non-OASIS patients.

The findings reported, hereafter, are relevant to OCISS–OASIS patients only. The prevalence of comorbid conditions in the study population was 60.8%, 75.0%, and 70.5% in breast, prostate, and colorectal cancer patients, respectively (Table 2). Consistently across all patients, the conditions with highest prevalence were hypertension, cardiovascular disease, diabetes, and arthritis. Chronic obstructive pulmonary disease was highly prevalent among prostate cancer patients (11.7%), and anemia was documented in approximately 13% of colorectal cancer patients. More than 13% of breast and colorectal cancer patients were identified as obese.

The prevalence of geriatric syndromes in the study population was lowest in breast cancer patients (34.7%) and highest in prostate cancer patients (51.2%). The syndromes with highest prevalence were urinary incontinence, dementia, and depression. All three conditions had the highest prevalence in prostate cancer patients. Disability was present in one-third of the study population, and, similar to comorbidities and geriatric syndromes, the prevalence was lowest in breast cancer patients (24.7%) and highest in prostate cancer patients (45.4%).

Findings from additional analysis grouping patients in distinct and overlapping entities of comorbidities, disability, and geriatric syndromes indicated that 26.4% of breast cancer patients, 12.0% of prostate cancer patients, and 14.0% of the colorectal cancer patients presented with no comorbidities, no disability, and no geriatric syndromes (Figs 1 to 3). At the other end of the spectrum, 11.7%, 24.7%, and 15.7% of the patients, respectively, in each of the cancer sites presented all three entities simultaneously. The distribution of patients by age and by the independent and overlapping categories of these three entities, within each cancer site, is presented in Table 3. Consistent across the three cancer sites, we note the gradual decline in the proportion of patients from the older age groups in the category of no comorbidities, no disability, and no geriatric syndromes and an increase in the proportion of older patients in the category represented by the overlap of the three entities.

View larger version (15K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 1. Distribution of Ohio Cancer Incidence Surveillance System–Outcome Assessment Information Set (OCISS–OASIS) patients by independent and overlapping categories of comorbidity, disability, and geriatric syndromes for breast cancer patients. Patients with no comorbidity, no disability, and no geriatric syndromes: 251 (26.4%).

View larger version (14K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 2. Distribution of Ohio Cancer Incidence Surveillance System–Outcome Assessment Information Set (OCISS–OASIS) patients by independent and overlapping categories of comorbidity, disability, and geriatric syndromes for prostate cancer patients. Patients with no comorbidity, no disability, and no geriatric syndromes: 39 (12.0%).

View larger version (15K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 3. Distribution of Ohio Cancer Incidence Surveillance System–Outcome Assessment Information Set (OCISS–OASIS) patients by independent and overlapping categories of comorbidity, disability, and geriatric syndromes for colorectal cancer patients. Patients with no comorbidity, no disability, and no geriatric syndromes: 179 (14.0%).

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Appendix Authors' Disclosures of... Author Contributions REFERENCES

Before being used in future applications, however, this taxonomy must undergo serious evaluation to assess its abilities to predict relevant outcomes in this population. In applying the criteria to identify comorbidities, disability, and geriatric syndromes, we made some important decisions that must be reassessed in subsequent studies. For example, rather than considering a gradient in ADL dependence, we identified patients with disability even if they had the most minimal level of need for assistance in their ADLs. Furthermore, this categorization was dichotomous and failed to capture a gradient in disability. Similar low thresholds were assumed in identifying patients with comorbidities or geriatric syndromes. Of note, we accounted for every condition in the above-referenced NIA/NCI list of comorbidities. Our finding indicating that the prevalence of some of the conditions such as Parkinson's disease and multiple sclerosis was quite low in our study population favors a greater focus on the more commonly observed conditions such as heart disease and diabetes. The effects of such nuances in defining these clinical entities will be assessed in future sensitivity analyses.

Findings from preliminary analyses assessing the association between age and the independent and overlapping categories lend credence to the soundness of the approach used in this study. The results indicate that the prevalence of all three entities at once increases in older age groups and is highest in the oldest old. The association between greater age and the increased prevalence of only one condition or a combination of some two entities at once is not as consistent, however. For example, the percentage of breast cancer patients presenting with disability and geriatric syndromes is relatively small (0.7% to 1.7%) until age 85+ years when there is a large increase to 10.2% (Table 6). Conversely, the prevalence of geriatric syndromes alone decreases with age, after an increase in the age group 70 to 74 years, to reach a low of 4.7% in patients 85 years of age and older. These findings, which vary across patients of different cancer sites, support our contention that using age alone to make treatment decisions would not be the desired approach. Of note, however, is that older patients with higher prevalence of disability and/or geriatric syndromes with more complex health care needs are likely to be institutionalized, and, therefore, under-represented in this HHC-only study population.

There are several limitations that should be borne in mind in interpreting these findings: First, elderly patients with cancer, receiving HHC, may present more complex health care needs than their counterparts not requiring or receiving HHC, and, therefore, constitute a vulnerable segment of the geriatric oncology patient population. This is evidenced by the findings indicating that HHC patients were more likely to be older than their non-HHC counterparts. Further, patients with more advanced stages of cancer were more likely than others to be represented in the HHC subgroup of cancer patients. It is noteworthy, however, that the non-HHC group is likely to reflect a wide range of mix in the complexity of their health care needs, including both community-dwelling patients and those residing in nursing homes.

Second, while the availability of OASIS data presented a unique opportunity to develop this taxonomy and to identify patients in clinically relevant categories, there are limitations that may be directly related to the quality of the data documented in OASIS. For example, when the depression items in OASIS are compared with other depression screening instruments (CES-D), the OASIS was shown to under-report depressive symptoms.¹⁴

Third, we used lists of comorbid conditions and geriatric syndromes that have not been validated previously. We opted to use the NIA/NCI listing of comorbid conditions because it is far more comprehensive than the one proposed by Charlson. The translation of the NIA/NCI conditions into ICD-9 codes will make it possible to employ them in large scale studies using population-based administrative data and to compare its performance relative to Charlson or other indices in studying health care outcomes. With regard to geriatric syndromes, we limited ourselves to the syndromes listed by Balducci and Extermann. A review of the literature, coupled with empirical analyses in future studies will help determine whether the listing of geriatric syndromes used in this study is comprehensive, or whether additional syndromes should be considered as well.

In closing, we feel that describing older cancer patients in independent and overlapping categories of comorbidity, disability, and geriatric syndromes may help us gain a more nuanced understanding of older cancer patients' clinical presentation using secondary data. The development of such taxonomies and their use in future studies will also help to more accurately identify patients who might benefit from standard cancer treatment, and those who might experience deleterious effects. From a methodologic standpoint, this study will pave the way for the refinement of risk-adjustment approaches in cancer-related health services research.

	Appendix

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Appendix Authors' Disclosures of... Author Contributions REFERENCES

 

	Authors' Disclosures of Potential Conflicts of Interest

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Appendix Authors' Disclosures of... Author Contributions REFERENCES

	Author Contributions

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Appendix Authors' Disclosures of... Author Contributions REFERENCES

 

Conception and design: Siran M. Koroukian, Patrick Murray, Elizabeth Madigan Financial support: Siran M. Koroukian Administrative support: Siran M. Koroukian Provision of study materials or patients: Siran M. Koroukian Collection and assembly of data: Siran M. Koroukian Data analysis and interpretation: Siran M. Koroukian, Patrick Murray, Elizabeth Madigan Manuscript writing: Siran M. Koroukian, Patrick Murray, Elizabeth Madigan Final approval of manuscript: Siran M. Koroukian, Patrick Murray, Elizabeth Madigan

 

	ACKNOWLEDGMENTS

 
We thank Ms Georgette Haydu, MS, of the Ohio Cancer Incidence Surveillance System ODH, for her review of earlier drafts of this article.

	NOTES

 
Supported by a National Institutes of Health Cancer-Aging Research Development Grant (P20 CA103736; S.M.K., pilot project investigator); and by a Career Development Grant from the National Cancer Institute (K07 CA096705 to S.M.K.).

Presented in part at the 3rd Annual Meeting of the Geriatric Oncology Consortium, Washington, DC, September 15-17, 2005.

Cancer incidence data were obtained from the Ohio Cancer Incidence Surveillance System (OCISS), Ohio Department of Health. Use of these data does not imply that the Ohio Department of Health either agrees or disagrees with any presentation, analyses, interpretations, or conclusions. Information about the OCISS may be obtained at odh.state.oh.us/ODHPrograms/CI_SURV/ci_surv1.htm.

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION Appendix Authors' Disclosures of... Author Contributions REFERENCES

 
1. Yancik R, Ganz PA, Varricchio CG, et al: Perspectives on comorbidity and cancer in older patients: Approaches to expand the knowledge base. J Clin Oncol 19:1147-1151, 2001[Abstract/Free Full Text]

2. Jemal A, Murray T, Ward E, et al: Cancer Statistics 2005. CA Cancer J Clin 55:10-30, 2005[Abstract/Free Full Text]

3. Yancik R, Wesley MN, Ries LAG, et al: Effect of age and comorbidity in postmenopausal breast cancer patients aged 55 years and older. JAMA 285:885-892, 2001[Abstract/Free Full Text]

4. West DW, Satariano WA, Ragland DR, et al: Comorbidity and breast cancer survival: A comparison between black and white women. Ann Epidemiol 6:413-419, 1996[CrossRef][Medline]

5. Yancik R, Wesley MN, Ries LAG, et al: Comorbidity and age as predictors of risk for early mortality of male and female colorectal carcinoma: A population-based study. Cancer 82:2123-2134, 1998[CrossRef][Medline]

6. Naeim N, Reuben D: Geriatric syndromes and assessment in older cancer patients. Oncology (Williston park) 15:1567-1580, 2001[Medline]

7. Balducci L, Extermann M: Management of the frail person with advanced cancer. Crit Rev Oncol Hematol 33:143-148, 2000[Medline]

8. Grunfeld ES, Ramirez AJ, Maher EJ, et al: Chemotherapy for advanced breast cancer: What influences oncologists' decision-making? Br J Cancer 84:1172-1178, 2001[CrossRef][Medline]

9. Fried LP, Ferrucci L, Darer J, et al: Untangling the concepts of disability, frailty, and comorbidity: Implications for improved targeting and care. J Gerontol A Biol Sci Med Sci 59:255-263, 2004[Medline]

10. US General Accounting Office. Medicare Home Health Care: OASIS Data Use, Cost, and Privacy Implications. Report No. GAO-01-205. Washington, DC, January 2001, p 3

11. Madigan EA, Fortinsky RH: Interrater reliability of the outcomes and assessment information set: Results from the field. Gerontologist 44:689-692, 2004[Abstract/Free Full Text]

12. Hittle DF, Shaughnessy PW, Crisler KS, et al: A study of reliability and burden of home health assessment using OASIS. Home Health Care Serv Q 22:43-63, 2003[CrossRef][Medline]

13. Landis JR, Koch GG: The measurement of observer agreement for categorical data. Biometrics 33:159-174, 1977[CrossRef][Medline]

14. Fortinsky RH, Madigan EA, Tullai-McGuinness SL: How to obtain meaningful and reliable results with OASIS data. National Association for Home Care, 20th Annual Meeting, Las Vegas, Nevada, October 15, 2001

Submitted June 17, 2005; accepted January 6, 2006.

This article has been cited by other articles:

				D. Papamichael, R. Audisio, J.-C. Horiot, B. Glimelius, J. Sastre, E. Mitry, E. Van Cutsem, M. Gosney, C.-H. Kohne, and M. Aapro Treatment of the elderly colorectal cancer patient: SIOG expert recommendations Ann. Onc., October 15, 2008; (2008) mdn532v1. [Abstract] [Full Text] [PDF]

				C. T. Cigolle, K. M. Langa, M. U. Kabeto, Z. Tian, and C. S. Blaum Geriatric Conditions and Disability: The Health and Retirement Study Ann Intern Med, August 7, 2007; 147(3): 156 - 164. [Abstract] [Full Text] [PDF]

				M. Extermann and A. Hurria Comprehensive Geriatric Assessment for Older Patients With Cancer J. Clin. Oncol., May 10, 2007; 25(14): 1824 - 1831. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Koroukian, S. M. Articles by Madigan, E. Search for Related Content PubMed PubMed Citation Articles by Koroukian, S. M. Articles by Madigan, E.