Compliance of sildenafil treatment for erectile dysfunction and factors affecting it


To assess the compliance of treatment, its affecting factors, and reasons for dropout, a questionnaire was mailed to a cohort of 2139 subjects who received sildenafil prescriptions for erectile dysfunction (ED) at our institution from 1999 to 2002. A total of 726 subjects (34%) with a mean age of 67 years answered the questionnaires. The response rate for sildenafil treatment was 67%. Of these sildenafil responders, 43% reported that they continued using sildenafil while 57% did not, in a mean follow-up of 3 years. Common reasons for discontinuation were effect below expectations, high cost, loss of interest in sex, and inconvenience in obtaining sildenafil. The continuers showed a higher rate than the discontinuers (P<0.05) of having tried other treatments, dose titration, and a dose higher than 50 mg. The discontinuers reported having a lower mean responding dose and improvement score post sildenafil treatment than the continuers. In conclusion, effect below expectations was the leading reason for discontinuation of sildenafil treatment. How ED subjects tried the medication and the adequacy of education in the initial treatment period may impact the compliance of sildenafil treatment.


The launch of sildenafil (Viagra®), the first oral form of phosphodiesterase-type V (PDE-5) inhibitor, in 1998 leads the treatment of erectile dysfunction (ED) to a new era. It has attracted many men with ED to seek help, which was evidenced by a tremendous growth in sales, since its introduction.1, 2 Besides sildenafil, two novel PDE-5 inhibitors, tadalafil (Cialis®) and vardenafil (Levitra®) were introduced into markets recently. Numerous publications advocate their efficacy and safety in the treatment of ED.

In a prior follow-up study of sildenafil users, the refill rate in the responders was only 66%.3 This figure sparked our curiosity to investigate their compliance and the reasons for their dropout. Meanwhile, we also compared how the continuers and discontinuers used the medication.

Patients and methods

We used the computer system to search for the subjects who had been diagnosed with ED and received sildenafil prescriptions in outpatient settings at our institution from March 1999, when it was marketed in Taiwan, to the end of 2002.

A questionnaire containing questions about marital status, ED duration, frequency of sexual intercourse, history and current status of Viagra® use, improvement score post sildenafil treatment, reasons for discontinuation (multiple choices), and other treatments adopted for ED was mailed to all the subjects in September 2003. Responders to sildenafil were defined as those replying ‘yes’ to the question ‘Did you have satisfactory sexual intercourse with Viagra® treatment?’ Based on the response to the question ‘Did you take Viagra® in recent 3 months?’, the responders who answered ‘yes’ to it were classified as continuers and those who answered ‘no’ as discontinuers. The sildenafil nonresponders were not assessed for the compliance of treatment.

Additionally, the efficacy was assessed by using five questions concerning erection confidence, penetration ability, erection firmness, maintenance ability, and intercourse satisfaction resulting from sildenafil treatment. In each question, the response was graded on 5 scales ranging from +2 to −2: with +2 representing much improvement after sildenafil treatment; +1, some improvement; 0, no change; −1, some regression; and −2, much regression. The total score of the five questions with a range from 10 to −10 represented the improvement score post sildenafil treatment.

Data were analyzed by descriptive statistics, χ2 test, t-test. The level of significance was considered as P<0.05.


A total of 2139 subjects received sildenafil prescriptions for the treatment of ED at our institution in a 4-year period. A total of 726 (34%) subjects (respondents) with mean age of 66.8±9.8 years (medium 70 years and a range of 28–89 years) answered the questionnaires. Mean age of the respondents was 6 years more than that of the nonrespondents (60.5±13.2 years, medium 64 years and a range of 22–88 years, P<0.001).

Of 726 respondents, mean duration of ED was 60 months (median 36 months) and their marital status was as follows: 84% were married; 7%, divorced or separated; 7%, widowers; and 2% were single. Of the respondents, 76% still had sexual intercourse with a medium frequency of two times per month (mean 3.0±2.7 times/month), while the remaining 24% did not have sexual intercourse for now. In addition to Viagra®, 44% of respondents reported adopting other treatments for ED, including food remedy, phytotherapy, and physical rehabilitation, etc. (Table 1).

Table 1 Treatments adopted for ED other than sildenafil in 726 respondents

The response rate to sildenafil achieved 67% (444/661) after excluding 65 subjects, including 19 not answering the question and 46 never taking sildenafil, out of the 726 respondents. Of 661 subjects, 47% experienced one or more adverse effects (AEs), in whom 34% had facial flushing, 12% had visual disturbance, 12% had palpitation, 6% had headache, 6% had nasal stuffiness, 6% had chest discomfort, 5% had dyspepsia, and 3% had others. The responders had a shorter mean duration of ED (50 vs 74 months, P<0.01) and a higher improvement score post sildenafil treatment than the nonresponders (6.2 vs −0.1, P<0.01). There was no significant difference in mean age between the responders and nonresponders.

All of the 444 responders started on sildenafil more than 3 months before the survey. After excluding 10 subjects not answering the question about the discontinuation of sildenafil out of 444 responders, 43% (188/434) reported that they continued using it within 3 months prior to the survey, while 57% (246/434) discontinued it in a mean follow-up of 3 years. There was no significant difference in the discontinuation rate among the new patients of each year. Of the 188 continuers, sildenafil was refilled from physicians in 71%, over the counters in 38% and from friends or partners in 8%. Over-the-counter purchase was attributed to greater convenience in 47% of the subjects and a lower cost in 30%. The reasons why sildenafil responders dropped out from treatment are listed in Table 2. Effect below expectations was the leading cause of discontinuation.

Table 2 Reasons for discontinuation in 246 sildenafil responders

Table 3 shows the comparison between the continuers and discontinuers in age, ED duration, improvement score post sildenafil treatment, rate of having a responding dose greater than 50 mg, dose titration, having tried doses greater than 50 mg and adopted other treatments for ED, and incidence of AE. The continuers were more likely to have a shorter mean ED duration, a higher improvement score post sildenafil treatment, dose titration, a higher dose trial, and other treatments than the discontinuers (P<0.05). Of the continuers 46% reported their responding dose was higher than 50 mg, while only 37% of the discontinuers did so (P=0.08) and a higher incidence of AE was reported in the continuers than in the discontinuers (63 vs 47%, P<0.05). However, palpitation occurred less in continuers than in discontinuers (12 vs 15%, P<0.05).

Table 3 Comparison between the continuers and discontinuers of the 434 sildenafil responders

Of the discontinuers, 78% (192/246) reported remaining sexually active, of whom 46% (88/192) received other treatments for ED, but only 14% (27/192) adopted the active treatments that were vacuum erection device, intracavernous injection, penile prosthesis implantation or psychotherapy.


Only a few studies investigated the compliance of sildenafil treatment for ED. The discontinuation rates of sildenafil responders were reported to range from 29 to 35% with a duration of follow-up ranging from 6 to 12 months and the mean age ranging from 50 to 60 years.4, 5, 6, 7 Diversity of study subjects, durations of follow-up and methods of study might account for the different results of various studies. In our study, the discontinuation rate of sildenafil in a mean follow-up of 3 years reached 57%, which might be attributed to an older mean age of study subjects, a longer duration of follow-up, and use of a retrospective method.

Several reasons were reported to be responsible for discontinuation: ineffectiveness, lack of opportunity or desire for sexual intercourse, lack of emotional readiness for sexual life after long-term abstinence, partner reluctance, high cost, AE, fear of AE, recovery from ED, illness, and difficulty with physicians.4, 5, 6, 7 In studies with a shorter duration of follow-up (6 months), lack of opportunity or desire, or unreadiness for sexual life was the leading cause of discontinuation,6, 7 while in studies with a longer duration of follow-up (9.7–12 months), the main reason became ineffectiveness of treatment.4, 5

In our study, 42% of discontinuers who reported having a satisfactory sexual intercourse with treatment dropped out from it on account of effect below expectation. It seemed unreasonable that effectiveness became a main problem for discontinuation in responders. Several possible reasons may explain it. The responders were defined in a qualitative manner. However, difference in the improvement score post sildenafil treatment existed among the responders by quantitative measure. According to the improvement score, the discontinuers did have a poorer effect of sildenafil treatment than the continuers (4.8 vs 7.8, P<0.05). In addition, subjects may have an unrealistic expectation from the oral medication. Cure was valued as the most important variable in defining the success of ED treatment.8 Furthermore, ED is an aging process and as severity progresses9 the effectiveness of treatment may become worse.

The responding dose greater than 50 mg was less commonly used in the discontinuers than the continuers. Over half of the discontinuers did not titrate the dose and never tried a dose greater than 50 mg. These figures indicate that there is much room for the discontinuers to improve the efficacy of sildenafil treatment. Harrold et al.10 reported that those who had previous treatment for ED were more likely to refill the sildenafil. Our data also demonstrated that the continuers had a higher rate of adopting other treatments than the discontinuers and the continuers made more efforts to optimize the sildenafil treatment. This reflects that aggressive behaviors in seeking treatment are related to a better result.

High cost and loss of interest in sex were common reasons for discontinuation. However, 78% of discontinuers remained sexually active of whom 46% adopted treatments other than PDE-5 inhibitors. These figures indicate that mainly discontinuers were interested in treating their erectile problems. Unfortunately, most of the treatments adopted were not considered effective except vacuum erection device, intracavernous injection, penile prosthesis implantation, and psychotherapy.

Hackett et al.4 reported that difficulties with the general practitioner stood as the fifth common reason for discontinuation. In our study, 29% of the discontinuers blamed inconvenience in obtaining sildenafil for discontinuation. Even for the continuers here, 38% of them obtained sildenafil over the counter. In Taiwan, over 70% of sildenafil was distributed from drugstores, possibly without prescription.2 This indicated that many Taiwanese men with ED did not seek professional help due to inconvenience, embarrassment or not considering ED as a disease.

Since mean dose used by the continuers was higher than that of the discontinuers, the continuers also had a higher incidence of AE than the discontinuers. This seems to indicate that having AE is not an important factor for discontinuation. Interestingly, there is one exception; palpitation occurs more frequently in the discontinuers. This might be due to the impression projected by the mass media that taking sildenafil might cause a lethal cardiovascular accident. Therefore, when palpitation occurred, the fear of death kept men from using it. Casabe et al.5 as well as Son et al.7 also reported fear of AE as a cause for discontinuation.

A low respondent rate of 34% was a flaw in this study. However, for this kind of questionnaire concerning a personal problem, it was difficult to have the respondent rate to meet a high standard. One favorable point was the relatively large study number. Nevertheless, there was concern that men with a better response were more likely to answer the questionnaire. However, the fact that the respondents were 6 years older on average than the nonrespondents, and the response rate was lower than that of a prior study on similar subjects (67 vs 72%),3 excluded the possibility of this concern. Our result was confined to representing the old age group around 70 years also probably because the group was more cooperative.


The discontinuation rate in the sildenafil responders was 57% in 3 years. Common reasons for discontinuation were effects below expectations, high cost, loss of interest in sex, and inconvenience in obtaining sildenafil. Of the discontinuers, 78% were still engaged in sexual activities and almost half of them even tried other treatments, but the truly effective treatments for ED were relatively less chosen.

Making more treatment efforts of trying a higher dose and titrating dose were related to a higher compliance. Seeking professional help should be addressed to ED sufferers not only to improve the efficacy and compliance of treatment but also to offer further treatment options.


  1. 1

    Wysowski DK, Swann J . Use of medications for erectile dysfunction in the United States, 1996 through 2001. J Urol 2003; 169: 1040–1042.

    Article  Google Scholar 

  2. 2

    Jiann BP, Yu CC, Su CC . Impact of introduction of sildenafil on other treatment modalities for erectile dysfunction: a study of nationwide and local hospital sale. Int J Impot Res 2004; 16: 527–530.

    CAS  Article  Google Scholar 

  3. 3

    Jiann BP, Yu CC, Tsai JY, Wu TT, Lee YH, Huang JK . What to learn about sildenafil in the treatment of erectile dysfunction from 3-year clinical experience. Int J Impot Res 2003; 15: 412–417.

    Article  Google Scholar 

  4. 4

    Hackett GI, Milledge A . A 12-month follow up of 260 patients taking sildenafil. NHS clinical experience. In: Fourth Congress of the European Society for Sexual and Impotence Research (ESSIR) 2001, 30 September–3 October 2001 (poster 171).

  5. 5

    Casabe A, Cobreros C, Bechara A, Roletto L, Cheliz G, Hochman S et al. Drop out reason in responders to sildenafil. Int J Impot Res 2001; 13(Suppl 2): S5, (moderated poster 9).

    Google Scholar 

  6. 6

    Klotz T, Mathers M, Klotz R, Sommer F . Why do patients with erectile dysfunction abandon effective therapy with sildenafil (Viagra®)? Int J Impot Res 2005; 17: 2–4.

    CAS  Article  Google Scholar 

  7. 7

    Son H, Park H, Kim SW, Paick JS . Reasons for discontinuation of sildenafil citrate after successful restoration of erectile function. Asian J Androl 2004; 6: 117–120.

    CAS  PubMed  Google Scholar 

  8. 8

    Hanson-Divers C, Jackson SE, Lue TF, Crawford SY, Rosen RC . Health outcomes variables important to patients in the treatment of erectile dysfunction. J Urol 1998; 159: 1541–1547.

    CAS  Article  Google Scholar 

  9. 9

    Seftel AD . Erectile dysfunction in the elderly: epidemiology, etiology and approaches to treatment. J Urol 2003; 169: 1999–2007.

    CAS  Article  Google Scholar 

  10. 10

    Harrold LR, Gurwitz JH, Field TS, Andrade SE, Fish LS, Jarry PD et al. The diffusion of a novel therapy into clinical practice: the case of sildenafil. Arch Intern Med 2000; 160: 3401–3405.

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to B-P Jiann.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Jiann, B., Yu, C., Su, C. et al. Compliance of sildenafil treatment for erectile dysfunction and factors affecting it. Int J Impot Res 18, 146–149 (2006).

Download citation


  • compliance
  • discontinuation
  • sildenafil
  • erectile dysfunction
  • phosphodiesterase type 5 inhibitors

Further reading