Although erectile dysfunction (ED) has been considered a complication of other medical conditions, clinicians and researchers suggest that ED may serve as a clinical marker of vascular health. This retrospective claims study examined the prevalence of predefined comorbid conditions in men with ED (N=301 994) in the 12 months before and the 6 months following ED diagnosis. Consistent with previous research, comorbid conditions were prevalent among men with ED. Comorbid conditions were most often diagnosed before an ED diagnosis, although new diagnoses in the 6 months following an ED diagnosis were common and occured more frequently than new diagnoses in a matched sample of men without ED during the same period. Differences by age, region and insurance coverage for ED medications were also examined. Findings support previous research that suggests ED may serve as a marker for previously undetected comorbid conditions.
Erectile dysfunction (ED), defined as the persistent inability to attain and/or maintain an erection sufficient for satisfactory sexual performance,1 is a prevalent condition in men. The recent multinational Men's Attitudes to Life Events and Sexuality (MALES) study found an overall prevalence rate of self-reported ED among men aged 20–75 years to be 16%, with rates ranging from 10 to 22% across countries.2 The Massachusetts Male Aging Study found even higher rates of ED among men aged 40–70 years, with 52% of men reporting minimal, moderate or complete ED.3
Numerous studies have documented the association of ED with various risk factors and comorbid conditions, including increasing age, smoking and medical conditions such as diabetes mellitus, cardiovascular disease, hypertension, hyperlipidemia and depression.2, 3, 4, 5, 6 Although ED has often been considered a complication of other medical conditions, more recently clinicians and researchers have suggested that ED may serve as an early clinical marker of vascular health and underlying comorbid conditions, particularly cardiovascular disease.5, 6, 7, 8, 9, 10, 11
Therefore, the diagnosis of ED can be an important point to intervene and potentially improve men's health through the encouragement of behavioral or lifestyle changes, and/or the diagnosis and management of comorbid conditions. Many men, however, are still reluctant to seek treatment and/or to initiate discussions about ED with their physicians.10 In addition, physicians may be reluctant to initiate inquiries and discussions about ED with patients for a variety of reasons, including concerns about time, perception of ED as a low priority condition, and level of comfort discussing sexual issues.12 The consequences are that ED may be underdiagnosed and undertreated, and perhaps more importantly, that underlying medical conditions may not be pursued, diagnosed and treated. This is important not only for improving men's health. Early detection and treatment of ED-related conditions such as diabetes, heart disease and hypertension is important as these conditions and their complications present a substantial economic burden in terms of both health and work loss costs.13, 14, 15 Thus, as ED may be a marker of other serious conditions, it is important for physicians to screen men for ED and associated conditions.16, 17
This retrospective database study examined the extent to which men in the US were newly diagnosed with comorbid conditions subsequent to an ED diagnosis. We examined the presence of diagnosed comorbid conditions in men with ED in the 12 months before ED diagnosis and in the 6 months following ED diagnosis. Specific research questions were: (1) among members with a claim for any health care utilization and an ED diagnosis, what percentage of men had a diagnosis of each ED-related comorbid condition in the year before their ED diagnosis? (2) among members with a claim for any health care utilization and an ED diagnosis, what percentage of men had a new diagnosis of each ED-related comorbid condition in the 6 months post-ED diagnosis? and, (3) for men with any claim for health care utilization and a newly diagnosed comorbid condition in the 6 months after their ED diagnosis, what is the mean number of days between the first ED diagnosis and the first diagnosis of the individual ED-related comorbid conditions? We also examined whether diagnoses of comorbid conditions before and after an ED diagnosis differed by age, region of the country or type of health insurance coverage for phosphodiesterase type 5 (PDE5) therapies (i.e., sildenafil citrate; tadalafil; vardenafil hydrochloride) for the treatment of ED.
Materials and methods
This study used an extract from the PharMetrics Patient-Centric Database (PharMetrics Inc.), a large, longitudinal integrated database of medical and pharmaceutical claims from over 40 million commercially insured patients in over 70 health plans, and two billion patient observations from across the US. The data set included paid claims and enrollment information that spanned from 16 February 1995 to 31 December 2003. The database is completely deidentifed and compliant with the Health Insurance Portability and Accountability Act. An additional data source consisted of formulary database information regarding insurance coverage for ED medications that was merged with the patient files described above. Not all patients in the claims database had insurance coverage information.
Identification of sample
For the analyses of men with ED described in this paper, patients had to be: (a) male; (b) 18 years of age or older; (c) have a diagnosis of ED; (d) have at least one health care utilization claim record identified in the time period ranging from 12 months before the date of their ED diagnosis to 6 months post-ED diagnosis. For analyses regarding insurance status, patients had to have at least 12 months of the same type of insurance coverage for ED medications. ED was defined as a claim that had an ED diagnosis code based on the ninth version of the International Classification of Diseases Clinical Modification (ICD-9CM); codes used were 607.84 for ED of organic origin and 302.72 for ED of nonorganic origin. Insurance coverage for ED medications (specifically, any PDE5 inhibitor) was classified as none (i.e., not reimbursed), partial (i.e., covered only with prior authorization) and full (i.e., covered with variable co-pay, no prior authorization).
We were also interested in knowing whether the prevalence of newly diagnosed conditions in men following an ED-diagnosis differed from the prevalence of new conditions seen in men without ED during the study period. To examine this, men in the ED cohort were matched 1:1 without replacement to a non-ED cohort based upon age and region of residence. Individuals in the non-ED cohort were then randomly assigned a starting date that ranged from 16 February 1995 to 31 December 2003 (i.e. same starting date range as the ED cohort).
Identification of comorbid conditions
Diagnosed hypertension, diabetes mellitus, high cholesterol, depression, cardiovascular disease and prostate cancer were identified using ICD-9CM codes. The following codes were used: 401–405 for hypertension; 272.0–272.4, 272.8 and 272.9 for high cholesterol; 250 for diabetes mellitus; 410–414, 415–417 and 420–429 for cardiovascular disease; and 185, 233.4, 222.2 and 236.5 for prostate cancer. Owing to methods used for reimbursement of benign prostatic hyperplasia/lower urinary tract symptoms (BPH/LUTS), this condition was defined using a combination of ICD-9CM codes (600.0, 600.2, 601.1, 601.9, 602.9, 788.1–788.6), procedure codes based on Current Physician Terminology codes (52647, 53850, 53852–53, 84152–54, 51725–26, 51785, 51792, 51772, 51736, 51741, 51795, 51797), and National Drug Classification codes for doxazosin mesylate, dutasteride, finasteride, tamsulosin hydrochloride, terazosin hydrochloride and alfuzosin hydrochloride. Comorbid conditions were flagged if the patient had a claim for any diagnosis in the 18-month study period (12 months before the date of their ED diagnosis to 6 months post-ED diagnosis).
All analyses were conducted using SAS® statistical software version 8.1. (SAS Institute Inc., Cary, NC, USA). Analyses were conducted for all patients, and stratified by age group and region. Stratification by type of insurance coverage for ED medications was conducted for the subsample of men with ED for whom there was data on their insurance coverage status. Differences among groups were examined using χ2 for categorical variables and analysis of variance for continuous variables.
Characteristics of men with ED
A total of 301 994 men met the inclusion criteria of ED for this study and was matched to an individual in the non-ED cohort (Table 1). The majority of individuals were in age groups 45–<55 (33.90%) and 55–<65 (34.44%), and resided in the Midwest (38.38%) and South (33.62%). Of individuals for whom there was insurance information (n=118 804), only a relatively small percentage (1.09%) were classified as not having any insurance coverage for ED medications.
Prevalence of comorbid conditions in men with ED
Table 2 lists the overall prevalence of seven comorbid conditions among men with ED during the study period, conditions existing in the year before the ED diagnosis and conditions newly diagnosed in the 6 months following the ED diagnosis. In the periods before and after ED diagnosis, hypertension, high cholesterol and BPH/LUTS were the most common comorbid conditions. Conditions were most often diagnosed before a diagnosis of ED; however, there were new diagnoses of these comorbid conditions in the 6 month following a diagnosis of ED. For example, over one-quarter of men with ED and depression (26.7%) and men with ED and cardiovascular disease (26.2%) received their diagnosis of the comorbid condition in the 6 months after their ED diagnosis.
We examined whether there were any differences in the rates of comorbid conditions in the time periods before and after ED diagnoses by age group, region or insurance status for ED medications.
Differences by age
In the year before ED diagnosis, there were significant differences in rates of each comorbid illness across age groups (Table 3). For hypertension, high cholesterol, diabetes, BPH/LUTS and prostate cancer, rates of occurrence increased by age group until approximately age 75 and then began to decline. For cardiovascular disease, rates of occurrence increased with each age group, including those 75 and older. For depression, rates of occurrence were slightly higher in the younger age groups (18–<45 years) and declined starting with the 45–<55 age group. This same age-group pattern was found for conditions newly diagnosed in the 6 months following an ED diagnosis, with the exception that prostate cancer rates increased consistently by age group.
Differences by region
Table 4 shows comorbid conditions existing in the year before ED diagnosis and newly diagnosed in the 6 months following ED diagnosis, by region. For all conditions there were small, although significant, regional differences in the rates of comorbid conditions in the year before ED diagnosis. However, there does not appear to be an easily discernible pattern-linking region to rates of comorbid illnesses in the year prior to ED diagnosis.
In the 6 months post-ED diagnosis, the highest rates of new diagnoses of hypertension, high cholesterol, diabetes, BPH/LUTS, cardiovascular disease and prostate cancer were found in individuals in the East, whereas the lowest rates for these conditions were generally found in individuals in the West and Midwest. On the contrary, a slightly higher rate of newly diagnosed depression was found in individuals in the West. These findings suggest potential differences in medical care accompanying a diagnosis of ED in these regions of the country.
Differences by insurance status for ED medications
Among the subgroup with insurance coverage information for ED medications (n=118 804), there were significant differences in the rates of all comorbid conditions based on insurance coverage in the year before ED diagnosis. Individuals with no ED coverage had the highest rates of hypertension, diabetes, cardiovascular disease and prostate cancer, and the lowest rates of high cholesterol, depression or BPH/LUTS. Individuals with full ED insurance coverage had the highest rates of BPH and the lowest rates of hypertension and diabetes. Those with partial ED coverage had the highest rates of high cholesterol and depression, and lowest rates of cardiovascular disease and prostate cancer. As with regional differences, there does not appear to be any discernible pattern between type of ED coverage and rates of comorbid illnesses in the year before ED diagnosis. Also, it is important to remember that results may be affected by the relatively small size of the cohort with no ED medication coverage.
In the 6 months post-ED diagnosis, there were small but statistically significant differences in the rates of newly diagnosed comorbid conditions by insurance status, with the exception of prostate cancer. Individuals with no ED insurance coverage had the highest rates of hypertension, high cholesterol, diabetes and cardiovascular disease and the lowest rates of depression and BPH/LUTS. Conversely, individuals with full ED coverage had the highest rate of BPH/LUTS and the lowest rates of hypertension, high cholesterol, diabetes and cardiovascular disease. These differences in rates of comorbid conditions may reflect systematic differences among plans that do and do not provide coverage for ED medications and/or among individuals who have these different types of coverage.
Average time from ED to new diagnoses of comorbid conditions
For men who had a newly diagnosed comorbid condition in the 6 months following their ED diagnosis, we examined the average number of days from ED diagnosis to diagnosis of the comorbid condition. Table 5 shows that the average time between diagnosis of ED and new diagnosis of comorbid condition is between 2 and 3 months. New diagnoses for diabetes, prostate cancer and BPH/LUTS came earliest after ED diagnosis (about 65.1–66.5 days), whereas new diagnoses for cardiovascular disease, hypertension and depression came latest (76.1–77.8 days). We also examined time between ED diagnosis and new diagnoses of comorbid conditions by age groups, region and insurance coverage for ED medications and found no notable differences in time to diagnoses among these variables.
Comparison of prevalence of medical conditions in men with and without ED
Table 6 shows the prevalence of the medical conditions among men without ED during the study period. For both groups of men, the most common medical conditions were hypertension, high cholesterol and BPH/LUTS. In both study periods, men with ED were significantly more likely than men without ED to have a diagnosis of each medical condition.
ED is a prevalent condition in men and may serve as a clinical marker of vascular health and underlying comorbid conditions.5, 6, 7, 8, 9, 10, 11 This retrospective database study examined the presence of seven predefined comorbid conditions in men with ED in the 12 months before ED diagnosis and in the 6 months following ED diagnosis. We found that these comorbid conditions were common among men with ED and occurred more often in men with ED compared with a matched cohort of men without ED. We should note that for conditions to be flagged, the patient had to have a health care utilization claim for one of the conditions during the 18-month study period (12 months before the date of their ED diagnosis to 6 months post-ED diagnosis). Therefore, the actual prevalence of these conditions among men with ED could be higher, as claims for these conditions that occurred outside of the defined 18-month period would not be captured. However, the prevalence rates of hypertension, high cholesterol, diabetes and cardiovascular problems in our study were similar to those found among men with ED in the recent MALES study.2
BPH/LUTS was the most prevalent comorbid condition found among men with ED in our study. As noted, this condition was identified differently than the other conditions examined in the study. We used a combination of diagnosis codes, procedure codes and drug codes to identify BPH/LUTS, whereas we used only diagnosis codes to identify other conditions. Although this method of identifying BPH/LUTS may have overestimated the percentage of men with this condition, other studies have shown a strong association between BPH/LUTS and ED,18, 19 and The Cologne Male Survey found the prevalence of LUTS to be 72.2% in men with ED aged 30–80.20
We found that for each comorbid condition, men were more likely to have a diagnosis for that condition in the 12 months before the first diagnosis of ED. This is not surprising given that ED has often been considered a consequence of diseases such as hypertension, cardiovascular disease and diabetes.10 It may be that clinicians were more likely to evaluate men for ED once they were diagnosed with another related condition. Despite this, this study demonstrated that a substantial proportion of men were being newly diagnosed with comorbid conditions in the 6 month period after their ED diagnosis, suggesting that ED may indeed serve as a marker for previously undetected conditions. Further, the average time between the ED diagnosis and the comorbid diagnosis was between two and two and a half months. Although we cannot demonstrate that the diagnosis of the comorbid condition was a result of the patient pursuing medical care for ED, this possibility is important to consider. Recent practice guidelines developed by the Minority Health Institute are based on the rationale that ED is an early clinical manifestation of systemic vascular disease, and suggest that all men aged 25 years and older should be asked about their sexual function and those with ED should be assessed for cardiovascular risk and vascular disease.9 Thus, it is important for physicians to view ED as a manifestation of other comorbid conditions in many cases and as an opportunity for prevention. Indeed, men with ED were more likely than a matched sample of men without ED to be newly diagnosed with the predefined medical conditions, suggesting that physicians may be more likely to screen for certain comorbid conditions among men who are diagnosed with ED.
Similar to other studies, we found an association between increasing age and prevalence of comorbid conditions among men with ED, with the exception of depression, although we did not find any distinctive pattern in terms of region. As an exploratory analysis, we examined the potential influence that health care insurance coverage for PDE5 therapies might have on the diagnosis of comorbid conditions. Our hypothesis was that health insurance coverage for PDE5 inhibitor therapy may act as an incentive for men to seek treatment for ED, thus increasing the potential that previously undiagnosed comorbid conditions may be detected sooner. In our study, there were few men who did not have health insurance coverage for PDE5 treatment, and in general those men had higher rates of diagnosed hypertension, diabetes, cardiovascular disease, and lower rates of depression and BPH/LUTS. It is possible that these differences in rates of comorbid conditions may reflect different characteristics of plans that do and do not provide coverage for ED medications and/or individuals who have different types of coverage. Further, it is difficult to assess the potential influence of health care insurance coverage for medications on health-seeking behavior due to the high percentage of patients who had at least some insurance coverage for these therapies. Further research is needed to assess any association.
The use of a claims database does have some limitations. One limitation is that data may not be generalizable to men who are not covered by commercial insurance plans (i.e., men who are uninsured or insured through Medicare or Medicaid). In addition, the claims database only captures information that has been submitted for reimbursement, therefore it may not adequately represent all men with ED as many are reluctant to seek treatment.21, 22 We should also note that the seven conditions examined in this study were prespecified by the authors based on the previous literature and may not represent the most prevalent ED-related conditions in this sample.
Further prospective research is needed that examines the extent to which men who are diagnosed with ED are subsequently diagnosed with, or experience health conditions related to vascular health. Such research should be conducted with broad samples of men with ED and should examine which conditions are diagnosed most commonly following ED diagnosis. In summary, this study contributes to the growing literature that suggests ED may serve as a marker for previously undetected medical conditions. Further, it suggests that clinicians are following clinical guidelines for men who are diagnosed with ED.
NIH Consensus Development Panel on Impotence. Impotence. JAMA 1993; 270: 83–90.
Rosen RC, Fisher WA, Eardley I, Niederberger C, Nadel A, Sand M . The multinational Men's Attitudes to Life Events and Sexuality (MALES) study: prevalence of erectile dysfunction and related health concerns in the general population. Curr Med Res Opin 2004; 20: 607–617.
Feldman HA, Goldstein I, Hatzichristou DG, Krane RJ, McKinlay JB . Impotence and its medical and psychosocial correlates: results of the Massachusetts Male Aging Study. J Urol 1994; 151: 54–61.
Fung MM, Bettencourt R, Barrett-Connor E . Heart disease risk factors predict erectile dysfunction 25 years later. J Am Coll Cardiol 2004; 43: 1405–1411.
Seftel AD, Sun P, Swindle R . The prevalence of hypertension, hyperlipidemia, diabetes mellitus and depression in men with erectile dysfunction. J Urol 2004; 171: 2341–2345.
Sun P, Swindle R . Are men with erectile dysfunction more likfely to have hypertension than men without erectile dysfunction? A naturalistic national cohort study. J Urol 2005; 174: 244–248.
Baum N . Erectile dysfunction and cardiovascular disease. Clin Geriatr 2004; 12: 21–24.
Blumenthals WA, Gomez-Caminero A, Joo S, Vannappagari V . Should erectile dysfunction be considered a marker for acute myocardial infarction? Results from a retrospective cohort study. Int J Impot Res 2004; 16: 350–353.
Levine LA, Kloner RA . Importance of asking questions about erectile dysfunction. Am J Cardiol 2000; 86: 1210–1213.
Billups KL, Bank AJ, Padma-Nathan H, Katz S, Williams R . Erectile dysfunction is a marker for cardiovascular disease: results of the minority health institute expert advisory panel. J Sex Med 2005; 2: 40–52.
Ganz P . Erectile dysfunction: pathophysiologic mechanisms pointing to underlying cardiovascular disease. Am J Cardiol 2005 (in press).
Fisher WA, Rosen RC, Eardley I, Niederberger C, Nadel A, Kaufman J et al. The Multinational Men's Attitudes to Life Events and Sexuality (MALES) Study Phase II: understanding PDE5 inhibitor treatment seeking patterns, among men with erectile dysfunction. J Sex Med 2004; 1: 150–160.
Druss BG, Marcus SC, Olfson M, Tanielian T, Elinson L, Pincus HA . Comparing the national economic burden of five chronic conditions. Health Aff 2005; 20: 233–241.
Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J . Global burden of hypertension: analysis of worldwide data. Lancet 2005; 365: 217–223.
Hodgson TA, Liming C . Medical care expenditures for hypertension, its complications, and its comorbidities. Med Care 2001; 39: 599–615.
Sadovsky R, Mulhall JP . The potential value of erectile dysfunction inquiry and management. Int J Clin Pract 2003; 57: 601–608.
Sadovsky R . Integrating erectile dysfunction treatment into primary care practice. Am J Med 2000; 109: 22S–28S.
Frankel SJ, Donovan JL, Peters TI, Abrams P, Dabhoiwala NF, Osawa D et al. Sexual dysfunction in men with lower urinary tract symptoms. J Clin Epidemiol 1998; 51: 677–685.
Rosen RC, Altwein J, Boyle P, Kirby RS, Lukacs B, Meuleman E et al. Lower urinary tract symptoms and male sexual dysfunction: The Multinational Survey of the Aging Male (MSAM-7). Eur Urol 2003; 44: 637–649.
Braun MH, Sommer F, Haupt G, Mathers MJ, Reifenrath B, Engelmann UH . Lower urinary tract symptoms and erectile dysfunction: co-morbidity of typical ‘aging male’ symptoms? Results of the ‘Cologne Male Survey. Eur Urol 2003; 44: 588–594.
Kubin M, Wagner G, Fugl-Meyer AR . Epidemiology of erectile dysfunction. Int J Impot Res 2003; 15: 63–71.
Shabsigh R, Perelman MA, Laumann EO, Lockhart DC . Drivers and barriers to seeking treatment for erectile dysfunction: a comparison of six countries. BJU Int 2004; 94: 1055–1065.
About this article
Cite this article
Cameron, A., Sun, P. & Lage, M. Comorbid conditions in men with ED before and after ED diagnosis: a retrospective database study. Int J Impot Res 18, 375–381 (2006). https://doi.org/10.1038/sj.ijir.3901439
- retrospective claims study
- comorbid conditions
- erectile dysfunction
- vascular health
- managed care
Can written information material help to increase treatment motivation in patients with erectile dysfunction? A survey of 1188 men
International Journal of Impotence Research (2007)
International Journal of Impotence Research (2007)