ORIGINAL ARTICLE

Validation of a Computer Version of the Patientadministered Danish Prostatic Symptom Score

Questionnaire
Henrik L. Flyger,1 Else B. Kallestrup2 and Svend O. Mortensen3
From the 1 Department of Surgery, Urology Section, Hiller d Hospital, Hiller d, 2Department of Urology, Herlev Hospital, Herlev,
and 3 Department of Surgery, Urology Section, Glostrup Hospital, Glostrup, Denmark
(Submitted November 12, 2000. Accepted for publication November 28, 2000)

Scand J Urol Nephrol 35: 196199, 2001

Objective: To validate a computer version of the Danish Prostatic Symptom Score (DAN-PSS) questionnaire and compare it
with the paper version.
Material and methods: Ninety-three male patients (aged 2587 years), referred to a department of urology for lower urinary
tract symptoms (LUTS), lled in a personal computer (PC) version and a paper version of the DAN-PSS questionnaire.
Subsequently they answered a questionnaire concerning their preferences and computer experience.
Results: A fair correlation between the total scores from the paper and PC versions was found. The differences were
independent of both total score and age. In the PC version all LUTS questions were answered while 9.8% were left blank in
the paper version. The sexual questions were answered by 71% of patients in the paper version and by 87.5% in the PC
version. For the questions in the PC version a learning curve was observed in terms of the time taken to answer the questions.
Previous computer experience did not in uence answering time, difference in score between paper and PC versions or total
score. Almost all patients preferred the PC version to the paper version.
Conclusion: The PC scores are reasonably reliable judging by comparison with previously validated traditional paper scores.
The PC questionnaire had a higher response rate and was preferred by the majority of patients.
Key words: benign prostatic hyperplasia, DAN-PSS, personal computer, questionnaire, validation.
Dr. Henrik L. Flyger, Department of Surgery F118, Herlev Hospital, DK-2730 Herlev, Denmark

Questionnaires have become an important tool in the

assessment of lower urinary tract symptoms (LUTS).
Their use has been recommended by The International
Consultation on Benign Prostatic Hyperplasia (BPH)
(1). The original questionnaires of Madsen and Iversen
(2) and Boyarsky et al. (3) were administered by
clinicians and were therefore time-consuming. Later
questionnaires, such as IPSS (4) and The Danish
Prostatic Symptom Score (DAN-PSS) (5), were selfadministered, but needed to be evaluated and calculated by the clinician. By using a computer version we
wanted to (1) reduce the incidence of missing data; (2)
reduce the effort involved in calculating scores; (3)
make the data collection and processing more ef cient;
and (4) keep costs as low as possible.
The DAN-PSS system for evaluation of BPH symptoms and bother from symptoms is a well-established
score system in patients with LUTS (6). The questionnaire is patient-administered and deals with 12 items
relating to voiding, storage and postoperative problems.
Additionally it contains three optional items on sexual
function. Each item is scored on a scale between zero
2001 Taylor & Francis. ISSN 00365599

and three for both presence of symptom and bothersomeness. These scores are then multiplied to give the
total score for the item. All scores for the 12 LUTS items
are then added to give the total score. The same
procedure is followed for the three questions on sexual
function (6). Until now the questionnaire has only been
administered in a paper version. It is usually mailed to the
patient or issued at the rst consultation. The physician
calculates the item and total scores by hand. To avoid
misunderstandings and mistakes in the answers (e.g.
indicating bother from a symptom not present), to
obviate the need to calculate the total score by hand
and to obtain computer registration of the data a patientadministered computerized questionnaire [for use on a
personal computer (PC)] was designed by one of the
authors (S.O.M.). Furthermore, we had noticed, in the
evaluation of the DAN-PSS system, a low response rate
to the three sexual questions, which we hoped to improve
by using a computer.
This study was performed to nd out whether the
computer model was reliable compared with the
previously validated paper version.
Scand J Urol Nephrol 35

PC version of the DAN-PSS questionnaire

197

median age of the patients was 69 years (range 2587

years). The computer-based version was answered rst
by 25 patients, whereas 68 answered the traditional
paper version rst.

RESULTS

Fig. 1. Difference between PC and paper scores by rank (n = 93).

MATERIAL AND METHODS

Methods
A computerized version of the DAN-PSS questionnaire
was programmed to run on an IBM-compatible, DOS
3.2 PC so that each of the 12 items of the questionnaire
were presented to the patient one at a time. The DANPSS items are split into two parts: presence of symptom
and an associated bothersomeness of the symptom. In
the PC version this was achieved by using two
consecutive screens. The patient interface was limited
to the computer screen and a modi ed keyboard, made
to resemble the keys of a telephone.
After a screen of instructions, each patient was asked
to key in his/her social security number, which in
Denmark contains information on birth date and sex.
The 12 DAN-PSS items followed. At the end, three
optional questions on the patients sexual function were
asked. The computer recorded the time taken to
complete each item, produced a printed record for
both the patient and the hospital and saved the results in
an anonymous form.
All patients also completed the traditional paper
version of the questionnaire, and nally answered a
short paper questionnaire comprising three questions
concerning how they liked the computer version, their
preference for the computer or paper version and their
previous experience with computers. These questions
were answered using a visual analogue scale (VAS).
As the DAN-PSS items are scored on an ordinal
scale, the statistical methods employed are nonparametric except where noted.
Patients
Male patients referred for LUTS who had not
previously been subjected to the DAN-PSS questionnaire were eligible. PC and traditional paper questionnaires were administered to 93 volunteers. The

There was a fair correlation between the total scores

from the PC and paper versions although some
differences were noticed. Figure 1 shows the difference
between the paper and PC scores by rank. The
differences were judged to be independent of both
total score (Fig. 2) and age (Fig. 3). An approximately
linear drop in the time used on each item was observed,
with the patients using a median of 37 s (95%
con dence interval [CI] 2268 s) for the rst question
and a median of 10 s (95% CI 035 s) for the last
question. The time spent was age dependent, with the
older patients taking a longer time. The fastest time to
complete the questionnaire was 2.8 min (38-year-old
patient) and the slowest was 19 min (75-year-old
patient).
The paper version of the questionnaire was not
completely lled in: 9.8% of the LUTS question elds
were left blank. The computer version does not allow
blank elds. The answering rate for the three optional
questions on sexual functions was 71.0% for the paper
version and 87.5% for the PC version. Both those
patients who answered the paper version rst and those
who answered the PC version rst preferred the PC
version to a considerable degree: 91% and 100%,
respectively. Previous computer experience was scored
by the patients on a VAS from one to 10. The median
score was 4.5 (95% CI 25). Experience with
computers did not appear to in uence answering
time, difference in score between paper and PC version
or total score.

Fig. 2. Difference between PC and paper scores vs mean of PC and

paper scores.
Scand J Urol Nephrol 35

198

H. L. Flyger et al.

Fig. 3. Difference between PC and paper scores vs age.

DISCUSSION
A computerized questionnaire about LUTS has previously been used (7) in an interactive multimedia
prostate education program (MMP). In that study
patients rst completed a paper version of the IPSS
and then attended an education program before
completing a computer-administered IPSS. The study
showed a signi cant decrease in mean IPSS score as a
result of computerization of the questionnaire. This
study cannot be compared with our study because the
preconditions for the answers were different for the two
models. We did not see any difference in the answering
pattern depending on whether the paper or PC version
of the questionnaire was completed rst.
We found that all patients could handle the telephone keypad of the computer quickly and easily. A
likely explanation for this is that the rest of the
keyboard was covered so that patients only had 11
keys to choose from. This is corroborated by Maitland
& Mandel (8), who replaced the keyboard with a
numeric keypad, which the majority of their subjects
rated easy to use. Furthermore, they found that the
mean time to produce a computer report was signi cantly lower than that for a paper report (39 s vs 309 s).
We discovered that the older patients took longer to
answer but that the quality of the answers, judging by
the difference between paper and PC scores, was ageindependent; however, there was a large variation in
the results. Overall we observed a learning curve,
where the patients were three times faster on the last
question than on the rst (median 37 s vs 10 s).
Unfortunately it was not possible to register the time
used to answer the paper version.
In this study almost all subjects rated the computer
version favourably. This tendency is generally found in
studies where patients assess sensitive personal data
such as drug, alcohol and smoking habits (9), HIV/
AIDS risk (10) and sexual behaviour (11).
It seems that computer-based assessments are highly
Scand J Urol Nephrol 35

acceptable to almost all patients. In this unselected

patient sample only one patient preferred the paper
version to the computer version and ve patients found
them equally acceptable. Bock et al. (12) found a
computer-based assessment of smoking habits among
an ethnically diverse, low-income population highly
acceptable, as did Shakeshaft et al. (9) among clients in
a community-based drug and alcohol clinical setting
and Williams et al. (10) among not-in-treatment drug
users. In our study half of the patients rated themselves
as unfamiliar with computers, corroborating the assertion that previous knowledge of computers is not
necessary for use in this setting.
As described by Millstein & Irwin (11) adolescent
girls preferred computer interviews concerning their
general health and sexual behaviour to face-to-face or
self-administered questionnaires. The same tendency
was observed in this study, where LUTS patients were
more likely to answer questions about their sexual
function in the PC version than in the paper version.
This may be caused by the fact that it is not possible to
complete the PC version if an answer is omitted,
whereas answers can be omitted in the paper version.
Furthermore it seems more private and less embarrassing to give the data to the computer where the answer
is hidden as soon as the patient has entered the answer
via the keyboard.
In conclusion we nd that the PC scores are
reasonably reliable judging by comparison with previously validated traditional paper scores, that there
was a higher response rate for the three optional
questions on sexual function with the PC questionnaire
and that patients preferred the PC questionnaire to the
traditional paper version.

REFERENCES
1. Mebust W, Roizo R, Schroeder F, Villiers A. Correlation
between pathology, clinical symptoms and the course of
the disease. In: Cockett ATK, Aso Y, Chatelain C, et al.,
eds. The International Consultation on Benign Prostatic
Hyperplasia (BPH). SCI, 1991: 5162.
2. Madsen PO, Iversen P. A point system for selecting
operative candidates. In: Hinman F Jr., ed. Benign
prostatic hypertropy. New York, Springer Verlag, 1983:
7635.
3. Boyarsky S, Jones G, Paulson DF, Prout GR Jr. A new
look at bladder neck obstruction by the food and drug
administration regulators: Guide lines for investigation
of benign prostatic hypertrophy. Trans Am Assoc
Genitourin Surg 1977; 68: 2932.
4. Barry MJ, Fowler FJ, OLeary MP, Bruskewitz RC,
Holtgrewe HL, Mebust WK, et al. The American
urological association symptom index for benign prostatic hyperplasia. J Urol 1992; 148: 154957.
5. Hald T, Nordling J, Andersen JT, Bilde T, Meyhoff HH,
Walter S. A patient weighted symptom score system in

PC version of the DAN-PSS questionnaire

6.

7.

8.
9.

the evaluation of uncomplicated benign prostatic hyperplasia. Scand J Urol Nephrol Suppl 1991; 138: 5962.
Hansen BJ, Flyger H, Brasso K, et al. Validation of the
self-administered Danish prostatic Symptom Score
(DAN-PSS-1) system for use in benign prostatic
hyperplasia. Br J Urol 1995; 76: 4518.
van Schaik P, Ahmed T, Suvakovic N, Hindmarsh JR.
Effect of an educational multimedia prostate program on
the international prostate symptom score. Eur Urol 1999;
36: 369.
Maitland ME, Mandel AR. A client-computer interface
for questionnaire data. Arch Phys Med Rehabil 1994; 75:
63942.
Shakeshaft AP, Bowman JA, Sanson-Fisher RW.

199

Computers in community-based drug and alcohol

clinical settings: are they acceptable to respondents ?
Drug Alcohol Depend 1998; 50: 17780.
10. Williams ML, Freeman RC, Bowen AM, Saunders L.
The acceptability of a computer HIV/AIDS risk assessment to not-in-treatment drug users. AIDS Care 1998;
10: 70111.
11. Millstein SG, Irwin CE Jr. Acceptability of computeracquired sexual histories in adolescent girls. J Pediatr
1983; 103: 8159.
12. Bock B, Niaura R, Fontes A, Bock F. Acceptability of
computer assessments among ethnically diverse, lowincome smokers. Am J Health Promot 1999; 13: 299
304.

Scand J Urol Nephrol 35

Copyright of Scandinavian Journal of Urology & Nephrology is the property of Taylor & Francis Ltd and its
content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.