ORIGINAL RESEARCH–OUTCOMES ASSESSMENT: Evaluation of the Psychometric Properties of the Life Satisfaction Checklist as a Screening Tool for Erectile Dysfunction

Authors


Ignacio Moncada, MD, Department of Urology, Hospital Gregorio Marañón, Dr. Esquerdo, 46, Madrid 28007 Spain. Tel: +34629057026; Fax: +34913833059; E-mail: ignacio@moncada.name

ABSTRACT

Introduction.  There is a need for a more feasible and acceptable screening tool for erectile dysfunction (ED) in our health context. The Fugl-Meyer's Life Satisfaction Checklist (LISAT-8) has been shown to be a simple and good instrument for assessing the satisfaction of ED patients in different aspects of their lives and has also shown acceptable psychometric properties to be used in the Spanish population with ED. Furthermore, this checklist has been used as a screening tool in patients with and without ED, showing valid and reliable properties for detection of ED.

Aim.  To evaluate the validity and feasibility of LISAT-8 checklist as a screening tool to detect ED.

Methods and Main Outcome Measures.  A cross-sectional, multicenter, and observational study was conducted including 6,986 Spanish men aged over 18 years. Patients completed the International Index of Erectile Function (IIEF), Sexual Health Inventory for Men (SHIM) and LISAT-8 checklists. Questions about the simplicity of SHIM and LISAT-8 and time to complete them were used for feasibility evaluation. We used logistic regression analysis to select the best predictors for ED among the different items of LISAT-8 and the receiver operating characteristic (ROC) curve analysis to select the best cutoff value discriminating ED vs. non-ED subjects. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), along with kappa agreement coefficient, were also estimated.

Results.  A total of 6,986 patients were included. Among the patients, 71.2% preferred LISAT-8 to SHIM. Time to completion of checklist was 0.5 minute less with LISAT-8 than with SHIM (P < 00001). Items 2 (sexual life), 5 (contacts with friends and acquaintances), and 3 (partner relationship) of LISAT–8 were selected as predictors for ED. ROC curve analysis showed a cutoff point ≥15 with a sensitivity of 81.7% (95% confidence intervals: 80.5–82.9), a specificity of 79.2% (77.5–80.8), and PPV of 88.7% (87.6–89.8%) and NPV of 72.0% (70.0–73.0%). Kappa agreement coefficients were 0.57 (LISAT-8 vs. SHIM) and 0.60 (LISAT-8 vs. IIEF).

Conclusions.  LISAT-8 and its composite variable (formed by items 2 + 3 + 5) showed acceptable psychometric properties, and it could be a valid screening tool for ED in daily clinical practice. Moncada I, Micheltorena CF, Martínez-Sánchez EM, and Rejas Gutiérrez J. Evaluation of the psychometric properties of the LISAT-8 checklist as a screening tool for erectile dysfunction. J Sex Med 2008;5:83–91.

Introduction

Erectile dysfunction (ED) is a public health problem affecting millions of citizens all over the world [1,2]. Only in Spain, approximately two million men aged from 25 to 70 have some degree of ED [3]. The impact of ED often is transferred to other areas of life of the ED patient. Thus, some studies have shown a significant correlation between ED, quality of life, and satisfaction. ED appears to have a clear negative influence on the quality of life perception in men with this problem [4–7]. Despite the controversial data coming from the literature and the complexity of these disorders, and according to the National Institutes of Health panel for the study of ED, this male problem may be associated with depression, low self-esteem, poor self-concept, fear, and mental stress, all of which have a negative effect on sexual relations with partners, on family, and friends [1,8]. ED may also lead the patient to withdraw from sexual and social relationships, causing impairment in his overall health status [1].

The results of the Spanish study on ED (Epidemiología de la Disfunción Erectil Masculina [EDEM] study) found a lower prevalence of ED when it was reported by the patient answering a general and simple self-assessment question than when the disorder was evaluated by the administration of a specific checklist [3,9] for ED. Self-administered checklists have been shown to be a good alternative way of assessing sexual function to specific interviewer-administered questionnaires [10,11]. Multidimensional scales are also considered more valid and reliable from the psychometric point of view than one-dimensional scales, to assess a dysfunction and treatment outcomes [12].

For this purpose, the Sexual Health Inventory for Men (SHIM) is a widely used scale for screening ED, whose cross-cultural validity has been shown, and which has high sensitivity and specificity to detect ED [13]. However, SHIM has a hard and direct way of exploring the patient's erectile function, using a technical wording that limits its theoretical self-checklist capability; it also needs to be administered by a physician or medical personnel familiar with this topic. In fact, and because of cultural constraints, Spanish subjects have important concerns when filling/scoring the questionnaire as a result of the previous comments, and they are also very reluctant to talk about their sexual problems in an open way. Thus, in the routine of daily clinical evaluation in Spain, it is impractical to ask additional questions, so SHIM is not widely used.

Among the existent questionnaires to evaluate the sexual function, we have found that some of the existing tools are not fully validated into Spanish (i.e., Massachusetts General Hospital [14]), or they explore different aspects of sexual life such as sexual dysfunction secondary to psychopharmacological treatment (psychotropic drugs), that is the case for the Arizona Sexual Experience Scale, Sexual dysfunction secondary to psychotropic drugs questionnaire (SALSEX), or Changes in Sexual Functioning Questionnaire (CSFQ) instruments [15–17], or it analyzes the changes in the sexual functioning (CSFQ [17]), or they are not specific for male sexual dysfunction. This means that we are missing a feasible and acceptable screening tool for ED in our health context.

The Fugl-Meyer's Life Satisfaction Checklist (LISAT-8) [18,19] has been shown to be a simple and good instrument for assessing the satisfaction of ED patients in different aspects of their lives [4,9]. The Spanish version of the LISAT-8 has shown acceptable psychometric properties to be used in the Spanish population with ED [20,21]. In addition, this checklist has been used as a screening tool in a previous study [22] in patients with and without ED, showing valid and reliable properties for detection of ED. Hence, this study was designed to evaluate the psychometric properties of LISAT-8 as a screening tool for ED, to be used in the routine daily clinical evaluation.

Subjects and Methods

This was a cross-sectional, multicenter, and observational study to evaluate the validity and feasibility of LISAT-8 as a screening tool for detecting ED. The physicians participating were all primary care physicians (PCP) and urologists from all over Spain. The participants were selected at random from an investigator pool database, and they were geographically representative of the distribution of the Spanish population according to its regional weight in the whole country. PCPs and urologists selected consecutively the first 5 and 10 patients, respectively, who attended the health center or the urology office for any cause. The enrollment process was finished when the foreseen number of patients to be included was completed. Patients were asked to participate in the study if they were eligible and if they agreed to be part of it.

Target population consisted of Spanish men aged over 18 years, agreeing to participate in the study, without mental disease according to clinical routine practice, and able to read, understand, and complete patient-reported outcome tools written in Spanish. Written consent was requested from all the patients. The patients were asked about sociodemographic data including age, smoking habits, alcohol consumption, educational level, and concomitant diseases: cardiovascular disease (ischemic heart disease, hypertension, dyslipemia), diabetes mellitus, depression, benign prostatic hypertrophy (BHP), and sleep disorder.

Feasibility of both the LISAT-8 checklist and the SHIM inventory was assessed by means of questions about the simplicity of these two instruments to be filled out, time to completion, percentage of missing form, and ceiling and floor effects.

ED was detected using the Spanish version of the International Index of Erectile Function (IIEF) [23–25]. The SHIM [26] was used as a screening tool for ED. The patients also completed the LISAT-8 [27]. The IIEF index is a simple (15-item), reliable, multidimensional, self-administered tool, which has shown its linguistic and psychometric validity for examining erectile function [23,24] in different languages, including Spanish [25]. This instrument explores five domains: erectile function, orgasmic function, sexual desire, satisfaction with intercourse, and overall satisfaction with sexual activity.

SHIM is a brief five-item version of the IIEF, validated, multidimensional, and self-administered inventory, which is a sensitive indicator of changes in erectile function and treatment outcomes [26]. Each item is scored from 1 to 5:1, never/occasionally; 2, less than half the time; 3, sometimes/half the time; 4, more than half the time; and 5, almost always or always. Total score can range from 5 to 25, with higher scores indicating better sexual health. Patients with a score of 21 or less may have evidence of ED.

LISAT-8 [27] is a self-administered checklist (figure 1) with eight items, which is scored on a Likert-type scale: 1, very dissatisfied; 2, dissatisfied; 3, rather dissatisfied; 4, rather satisfied; 5, satisfied; and 6, very satisfied. The items measure satisfaction with eight different aspects of the patient's life: life in general, sexual life, relationship with partner, family life, relationship with friends, leisure, professional, and financial situation. Total score can range from 8 to 48. All items can be grouped in three dimensions: satisfaction with social life (items 1 and 4–6), satisfaction with affective life (items 2 and 3), and satisfaction with financial life (items 7 and 8) [21].

Figure 1.

Life satisfaction checklist [4].

Statistical Analysis

A descriptive statistical analysis was performed for all efficacy measures, including measurement of central tendency and dispersion for the quantitative variables, and absolute and relative frequency for nominal variables, with their 95% confidence interval in both cases. All bivariate analyses for continuous variables were performed using t-test. The chi-squared test was applied for nonordinal categorical variables, while the Mantel–Haenszel chi-squared test was used for analysis of ordinal categorical variables. McNemar test was used to compare proportions of both low and upper extreme responses (floor and ceiling effects, respectively) between LISAT-8 checklist and SHIM inventory.

A logistic regression analysis was carried out (stepwise model) in order to determine which items of the LISAT-8 were the best predictors for ED. Items remained in the composite variable according to the Wald statistic and the odds ratio values. A significant (P < 0.05) level for the Wald statistic was required along with a 95% confidence interval of the odds ratio excluding the value 1. Receiver operating characteristic (ROC) curve analyses were carried out to estimate the cutoff value that would better discriminate ED using the composite variable. The gold standard used as a reference for the ROC curve analysis was the diagnosis of ED or non-ED according to the IIEF index. Validity of the composite variable was assessed by determining the values of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Also, the kappa agreement coefficient was calculated to test concordance of diagnosis of ED—non-ED vs. the SHIM inventory score and the composite variable of the LISAT-8. Because this study intended to analyze the ability of the LISAT-8 checklist as a screening tool, high sensitivity and PPVs were used as criteria to choose the best combination of items as the estimator of ED. All statistical analyses and outputs were performed using the SAS (version 8.2) statistical package (SAS, Cary, North Carolina, USA).

Results

Five-hundred nine PCPs and 386 urologists from all over Spain took part in this study. A total of 6,986 patients were included, 68.7% from the primary care setting and 31.3% from the urological setting. Six-thousand five hundred ninety out of 6,986 recruited patients completed the IIEF index, and 4,114 (62.4%) patients showed to have ED to some degree: 2,038 (30.9%) mild, 1,101 (16.7%) moderate, and 975 (14.8%) severe according to the classification of the IIEF.

Table 1 compares the sociodemographic and clinical characteristics of the ED and non-ED groups according to the IIEF index, including drug (alcohol and tobacco) consumption and educational levels. The age of the patients was higher (P < 0.0001) in the ED group 51.3 ± 15.4 years vs. non-ED group 35.8 ± 11.8 years. The educational level of the majority of patients included in the study was primary and secondary. The proportion of subjects with higher education was greater (P < 0.0001) in the non-ED group (37.5%) than in the ED group (17.6%). Half the patients in both groups were smokers, declaring an average number of 19.5 ± 9.4 cigarettes per day among the patients with ED and 17.8 ± 8.5 in the subgroup without ED. The main comorbidities in the study sample were hypertension followed by dyslipemia, BHP, and diabetes, being the corresponding proportions of comorbidities significantly higher in the ED group than in the non-ED group (P < 0.0001 in all cases). Table 2 shows the mean scores for each LISAT-8 dimension in patients without ED and with different levels of ED according to SHIM and IIEF questionnaires. The mean scores obtained by subjects with ED were significantly lower than in subjects without ED (P < 0.0001), consistently throughout all dimensions of the LISAT-8 checklist when patients were classified according to SHIM or IIEF criteria as well.

Table 1.  Sociodemographic and clinical characteristics
Age (%)Non-EDEDP value
  1. The values are given as mean (standard deviation) and relative frecuency %.

  2. SP = systolic pressure; DP = dyastolic pressure; BHP = benign prostatic hypertrophy; BMI = body mass index.

 18–27 years31.99.6<0.0001
 28–36 years30.412.7 
 37–49 years21.918.7 
 48–62 years12.929.9 
 63 to >85 years2.929.1 
Age (years)35.8 (11.8)51.3 (15.4)<0.0001
Weight (kg)79.3 (11.0)80.4 (11.9)0.0004
Height (cm)175 (6.7)172 (7.5)<0.0001
BMI (kg/m2)25.9 (3.2)27.2 (3.4)<0.0001
Blood pressure (mm Hg)
 SP126.4 (12.4)136.3 (15.2)<0.0001
 DP75.5 (8.7)81.1 (9.5)<0.0001
Smoker (%)43.647.10.0066
Alcohol consumption (%)
 No20.723.20.0010
 Occasional59.849.0 
 Moderate19.527.8 
Educational levels (%)
 No educational certificates1.27.9<0.0001
 Primary education20.340.2 
 Secundary education41.034.3 
 College studies37.517.6 
Concomitant desease (%)
 Hypertension10.743.3<0.0001
 Dyslipidemia15.537.6<0.0001
 BHP3.626.4<0.0001
 Diabetes3.917.5<0.0001
 Ischemic heart disease1.27.6<0.0001
 Depression3.311.2<0.0001
Table 2.  Total and dimension scores of Life Satisfaction Checklist in patients with and without ED by SHIM or IIEF criteria
 Total (8–48)Affective life (2–12)Social life (4–24)Financial life (2–12)
  1. *Non-ED vs. ED: P < 0.0001; **P < 0.0001 between groups.

  2. ED = erectile dysfunction; IIEF = International Index of Erectile Function; SHIM = Sexual Health Inventory for Men.

SHIM
 ED33.4 (6.3)7.3 (2.2)17.8 (3.3)8.2 (2.1)
 Non-ED39.0 (4.4)*10.1 (1.4)*20.1 (17.8)*8.6 (2.0)*
IIEF********
 Severe30.0 (6.9)5.4 (2.0)16.7 (4.0)7.8 (2.3)
 Moderate31.4 (5.8)6.6 (1.6)17.0 (3.3)7.8 (2.1)
 Mild35.9 (4.8)8.5 (1.6)18.8 (2.5)8.5 (1.9)
 No dysfunction38.9 (4.5)10.1 (1.4)20.1 (2.3)8.6 (2.0)

Feasibility/Acceptability of LISAT-8

The 99.7% of patients completed correctly the LISAT-8 checklist, while the corresponding value for the SHIM inventory, 97.2%, was significantly lower (P < 0.0001). The 71.2% of subjects considered the LISAT-8 checklist easier to complete than the SHIM, while the opposite was observed in 27.3% only (P < 0.0001). Table 3 shows the time to completion of both instruments, overall and in different subgroups of patients according to age, educational level, and severity of ED. The average time to complete the instruments was 0.5 minutes less for the LISAT-8 than for the SHIM (P < 0.00001). Consistently, time to completion of LISAT-8 checklist was significantly lower than the one for SHIM inventory throughout the different age, educational level, and severity of ED subgroups. This was of particular relevance in the oldest age group (up to 62 years), showing the greatest difference, and in the lowest educational level (no certificates): P = 0.0001 and P = 0.0004, respectively.

Table 3.  Mean time to completion (in minutes) of SHIM inventory and LISAT-8 according to age, educational level, and severity of ED
 SHIMLISAT-8P
  1. Values expressed as mean (standard deviation).

  2. ED = erectile dysfunction; LISAT-8 = Life Satisfaction Checklist; SHIM = Sexual Health Inventory for Men.

Overall sample4.3 (4.6)3.7 (3.9)<0.0001
Age subgroup
 18–27 years3.6 (3.3)3.2 (3.3)0.0044
 28–36 years3.6 (3.4)3.2 (3.4)0.0012
 37–49 years3.8 (3.2)3.4 (3.9)0.0218
 48–62 years4.7 (6.9)4.4 (4.1)0.0092
 Up to 63 years5.2 (4.1)4.4 (3.8)0.0001
Educational levels
 No educational certificates6.3 (4.7)5.1 (3.8)0.0004
 Primary education4.7 (6.0)4.1 (4.0)0.0002
 Secondary education4.0 (3.7)3.5 (3.9)0.0001
 College studies3.4 (3.4)3.0 (3.3)0.0032
Severity of ED
 No ED3.2 (5.2)2.8 (3.3)0.0017
 Mild4.4 (3.6)3.9 (3.9)0.0001
 Moderate5.2 (4.7)4.5 (4.3)0.0004
 Severe5.5 (4.5)4.9 (4.3)0.0019

Table 4 shows the floor and ceiling effects of LISAT-8 checklist, composite variable, and the SHIM inventory. Floor and ceiling effects were not observed in the LISAT-8 checklist or in the composite variable. All items had bottom and upper extreme frequency responses less than 50%. The SHIM inventory did not show floor or ceiling effect either; however, all of the items, except item 1, showed upper extreme frequency responses considerably higher when compared with LISAT 8. Hence, the total scale ceiling effect of the SHIM inventory (13.13%) was significantly higher than those of both the LISAT-8 checklist (0.91%) and the composite variable (5.07%), P < 0.0001 in both cases. Statistically, differences were also found when floor effect was compared between scales, although these differences were very small (Table 4).

Table 4.  Floor and ceiling effects of LISAT-8, composite variable (items 2 + 3 + 5), and SHIM inventory, and product–moment Pearson's correlation coefficients with erectile domain scoring of the International Index of Erectile Function
 Floor (%)Ceiling (%)Pearson rP value
ItemLISAT-8 checklist
11.010.30.4455<0.0001
25.410.20.8073<0.0001
32.715.90.5283<0.0001
41.018.10.3125<0.0001
50.519.30.2988<0.0001
60.912.70.3114<0.0001
72.710.30.1644<0.0001
81.46.60.1336<0.0001
Total0.04*0.91*0.5550<0.0001
Composite (2 + 3 + 5)0.16*5.07*0.7875<0.0001
ItemSHIM inventory
  • *

    P < 0.0001 vs. corresponding value for SHIM total (paired McNemar test).

  • LISAT-8 = Life Satisfaction Checklist; SHIM = Sexual Health Inventory for Men.

18.318.00.8159<0.0001
22.132.50.8548<0.0001
32.730.00.8299<0.0001
42.640.20.8388<0.0001
52.334.60.8226<0.0001
Total1.2413.130.8324<0.0001

Properties of LISAT-8 as a Screening Tool for ED

Table 4 shows the binary correlations between each of the eight items of LISAT-8 with the ED domain of the IIEF. Pearson's correlation coefficients were positive in all of the items and statistically significant, P < 0.0001, in all cases, although very weak, as expected, for items 7 and 8.

The stepwise multivariate logistic model (Table 5) carried out to explore the ability of each of the LISAT-8 items as predictors of ED showed statistical significant association with ED for items 2, 3, and 5 only, all of them with a P value below 0.05, therefore fulfilling the criteria to be included in the composite variable. Table 6 shows the results of the ROC curve analysis using the composite variable formed with the sum of items 2, 3, and 5 of the LISAT-8. A cutoff point of >15 was the best predictor value, showing a sensitivity of 81.7% (95% confidence interval: 80.5–82.9) and a specificity value of 79.2% (77.5–80.8). Thus, for men who had ED, 81.7% of them were correctly classified by the composite variable of LISAT-8, and for men who did not have ED, 79.2% of them were classified correctly as not having ED. Overall, the PPV and NPV of the study population were 88.7% (87.6–89.8%) and 72.0% (70.0–73.0%), respectively. Figure 2 displays the ROC curve for the composite variable of LISAT-8. The estimated area under the ROC curve (AUC), which shows the level of well-classified diagnosis, was 0.863. Hence, from a randomly selected pair of men, one with ED and one without ED, there was about an 86% chance of correctly identifying the man with ED and the man without ED based on their LISAT-8 composite variable. Moderate concordance (according to Landis criteria) was observed between LISAT-8 and SHIM instruments (kappa agreement coefficient = 0.57) and between LISAT-8 and IIEF (kappa = 0.60). Meanwhile, concordance was better between SHIM and IIEF (kappa = 0.79).

Table 5.  Regression logistic analysis of Life Satisfaction Checklist items selected
 EstimateSEWald chi-squarePOR (CI 95%)
  1. Estimates using analysis of maximum likelihood.

  2. SE = standard error; OR = odds ratio; CI = confidence interval.

Item 2−1.73350.05451,012.20580.00010.177 (0.159–0.197)
Item 3−0.11690.04556.61900.01010.890 (0.814–0.973)
Item 5−0.18230.049213.73490.00020.833 (0.757–0.918)
Table 6.  Cutoff values for validity for screening of erectile dysfunction of the composite variable (items 2 + 3 + 5) of the the Spanish version of Life Satisfaction Checklist
Cutoff valueSensitivityEspecificityPPV 5%PPV 10%PPV 15%PPV 20%NPV 5%NPV 10%NPV 15%NPV 20%
  1. PPV = positive predictive value; NPV = negative predictive value.

30.00100.000.000.000.000.000.000.000.000.00
40.1299.880.050.100.150.200.020.010.010.00
50.4499.840.130.230.330.410.030.010.010.01
61.2899.840.300.470.590.670.030.010.010.01
72.8999.800.430.620.720.780.040.020.010.01
85.7499.590.420.610.710.780.080.040.020.02
910.2799.510.520.700.790.840.090.050.030.02
1019.4299.060.520.700.780.840.180.100.060.04
1128.1598.410.480.660.760.820.300.170.110.08
1239.6897.180.430.610.710.780.470.300.210.16
1353.4893.830.310.490.600.680.720.540.430.35
1467.6189.010.240.410.520.610.870.750.660.58
1581.6879.170.170.300.410.500.960.910.870.82
1693.8042.810.080.150.220.290.990.990.980.97
1798.1123.900.060.130.190.241.001.001.000.99
1899.689.970.060.110.160.221.001.001.001.00
Figure 2.

Receiver operating characteristic curve analysis for screening of erectile dysfunction of the composite variable (items 2 + 3 + 5) of the Life Satisfaction Checklist.

Trials to explore other possible composite variables with items 2 + 5 and only-2 item were carried out performing the corresponding ROC curve analysis. However, none of them showed better sensitivity (79.9 and 80.4%, for 2 + 5 and only-2, respectively) or PPVs (88.6 and 90.1%, for 2 + 5 and only-2, respectively) than the figures observed for composite variable (see previous discussion), while maintaining similar values for the area under the curve and kappa's coefficient of agreement with diagnosis of ED using the IIEF index (AUC 0.86 and 0.88, and kappa 0.60 and 0.64, respectively).

Discussion

The present study was designed to evaluate the psychometric properties of the LISAT-8 checklist as a screening tool for ED in the daily clinical evaluation of patients, including the whole of the representative age range of the Spanish population with ED. This disorder may be diagnosed or screened by several validated tools [19,22]. The SHIM is recommended as the best instrument for the detection of patients with possible ED and has become the instrument of choice for screening ED in most settings all over the world [13]. Nevertheless, in our society, many physicians and patients find it problematic to raise the issue of sexual function, and an important part of doctors may be reluctant to incorporate it into their clinical routine practice because of the intimate nature of the topic the instrument explores [24]. In fact, the SHIM investigates patients' erectile function in a rough and direct manner, and it is written in a rather technical language, making it inappropriate for some patients, who sometimes refuse to complete it. This fact limits its theoretical capacity as a self-administered tool; hence, it has to be frequently given under direct physician supervision.

Fugl-Meyer's LISAT-8 [14,15,23] has been shown to be a simple and good instrument for assessing the satisfaction of ED patients in different aspects of their lives [4,9]. The Spanish version of this checklist showed good psychometric properties to be used in the Spanish subjects with ED as a self-administered patient-reported-outcome instrument [16,17]. This tool has also been used as a screening tool in a previous study in patients with and without ED under routine medical practice at PCP settings (the Investigación de la Disfunción Eréctile en Atención Primaria [IDEAP] study) [18]. The Spanish version of the LISAT-8 is written in a simple language, having shown good feasibility properties to be used in the Spanish population with ED [16]. Our results are in agreement with those findings, showing that the LISAT-8 checklist is a more feasible tool than SHIM, easier to complete for the majority of patients included in the present study; these results may support the routine use of this checklist. Physicians may then find it helpful to ask patients to complete a tool like the LISAT-8, in which sexual life is explored together with other different aspects of the patient's life in a friendly and more colloquial manner. This may provide an opportunity to open a dialogue about sexual dysfunction and treatment options.

A composite variable of LISAT-8 used as an ED screening tool has been reported previously [18]. Our composite variable includes, now, item 5, which could have a possible sexual significance for some patients in the context the current study was carried out. In addition, and in agreement with results of the logistic regression model performed, the three-item composite variable showed statistically significant ability to predict ED with higher sensitivity and PPVs equal or even higher than using only item 2 or the combination of items 2 + 5. This means that the composite variable of LISAT-8 was found to be valid, with sensitivity and PPVs over 80% for a cutoff point of ≥15. Although these values recorded here could be considered slightly poorer than the previous study [18], they have been observed in a wider and more representative sample of the Spanish male population.

On the other hand, even though multidimensional scales are considered more valid and more sensitive from the psychometric point of view than unidimensional scales to assess a dysfunction [12], other investigations have used patients' self-assessment of ED, answering a single question, and observed a moderate-to-high correlation and agreement with SHIM [25–27]. A direct measure of ED, assessment by a single item, it is recommended when it is impractical to ask additional questions of a more clinical nature [25], although it cannot be expected to have the same level of sensitivity for screening ED as the SHIM [25]. Psychometric properties of our composite variable of LISAT-8 are slightly poorer than SHIM [22], although patients preferred to answer the LISAT-8 checklist and needed less time for completion. In addition, the SHIM inventory showed higher ceiling effect than that observed for the composite variable of the LISAT-8. Physicians should evaluate whether they prefer easy, quick screening in which a lot of patients are included, or a more limited but reliable screening.

This study has some limitations. First, we found a high prevalence of ED, very different from the results of other authors [27,28]. A possible explanation for this could be that in an observational study like ours, there is more chance for physicians to administer the checklist to patients with suspected ED. We administered the LISAT 8 checklist as a whole and, because of this, we have no data on the psychometric properties of the use of the composite variable as an ED screening tool. Further research should be carried out to test the ability of this composite variable when administered alone. The validated version of this checklist is applicable for Spanish (Spain) people only. Additional validation works should be performed to adapt the instrument to other languages. As commented previously, the proportion of subjects who reported some comorbidities was higher in the ED group than in the non-ED group, and this fact could interfere with the results of the study.

In conclusion, the results of this work show that a composite variable formed with items 2 + 3 + 5 of the Spanish version of LISAT-8 checklist possess acceptable psychometric properties to be used as a screening tool for ED, and that this could be a feasible alternative to the SHIM inventory in routine medical practice in Spain.

Conflict of Interest: None declared.

Ancillary