There are four different ways to calculate scores using PORPUS. The first way uses equal weights for all ten questions; the other three use unequal weights to reflect the fact that patients might find that different problems impair Quality of Life more than others. For example, a patient might find that leaking urine is worse than having poor communication with the doctor.
PORPUS-P is a psychometric instrument. It has the most straightforward calculation: each question is weighted equally. The PORPUS-P has scores ranging between 0 (worst Quality of Life) to 100 (best Quality of Life.)
The scoring formula for the PORPUS-P is:
In this formula, the original scores on each item are X1 to X10, where, for example, X2=3 if a subject answered "3" on question 2. The maximum possible scores on each item (i.e., the number of response categories for each item) are M1 to M10, and COMPLETE is the number of items out of the 10 that are completed.
For a fully complete questionnaire, COMPLETE will be equal to 10 but this formula allows for adjustment of the observed score to a full score out of 100. We suggest that at least 8 items are completed for a PORPUS-P score to be computed. For a fully completed questionnaire, each item contributes between 0 and 10 points to the final score and the sum of the 10 item-scores is subtracted from 100 so that higher scores represent better QOL.
PORPUS-U is an indirect utility instrument. It weights each question according to pooled ratings from men with prostate cancer. This allows the PORPUS to be scored like the EQ-5D3 and Health Utilities Index (HUI)4, but specific to prostate cancer.
The pooled ratings were obtained from 234 men with prostate cancer, including newly diagnosed and metastatic, who completed personal interviews at the Prostate Centre at Princess Margaret Hospital in Toronto, Ontario, Canada. These men gave rating scale and standard gamble utilities for several of the health states described by the PORPUS.
We followed the methodology used by the developers of the HUI8 to estimate the multi-attribute utility function for the PORPUS. This was used to develop an formula to score the PORPUS-U, a prostate cancer-specific indirect utility instrument.
The scoring formula for the PORPUS-U is:
Response to Item (1 = Best; Highest = Worst)
(3) Social Support
(4) MD communication
(6) Urinary frequency
(7) Urinary leakage
(8) Sexual function
(9) Sexual interest
(10) Bowel problems
Look up the score for the responses to each item in the table above. Call these scores Z1 to Z10. For example, if a patient gave the responses coloured in grey above the values for Z1 to Z10 would be:
The PORPUS-U score for someone giving the responses above is
= 3.642 – 2.826
The PORPUS-U score calculated using this formula should be rounded to two decimal places (e.g., 0.82).
3. PORPUS-URS and 4. PORPUS-USG
PORPUS-URS and PORPUS-USG are direct utility instruments. They weight each question according to the individual ratings of the man answering the questionnaire. Weights can be elicited using the rating scale (RS) or the standard gamble (SG.)
PORPUS-URS and PORPUS-USG require the PORPUS questionnaire, two colour-coded clinical marker state cards which describe mild (white) and severe (orange) impairment in health-related quality of life in terms of the items and levels of the PORPUS, Feeling Thermometer and Chance Board visual props for the rating scale and standard gamble, a trained interviewer, and an interview script.
Patients first complete the PORPUS questionnaire, thus providing a disease-specific profile of their current health. This profile is recorded on a card resembling those used for the marker states. Patients then rank the description of their own health state and the 2 marker states in order of preference between full health and dead. Full health is described as the highest level on each of the 10 attributes of the PORPUS. Utilities for the patients' own health profile and the marker states are then obtained using the rating scale and standard gamble methods6.
The use of the PORPUS and the marker states may improve the reliability and validity of the rating scale and standard gamble7. However, these direct methods of utility elicitation require face-to-face interviews. PORPUS-U offers a less individualized but also less time-consuming alternative.
The reliability, validity, and responsiveness of the PORPUS-P and PORPUS-U
The reliability and validity of the PORPUS-P and PORPUS-USG was tested in 141 prostate cancer patients who completed the PORPUS instruments, generic utility instruments and prostate cancer-specific psychometric instruments. The PORPUS-P was shown to have good test-retest reliability (intraclass correlation coefficient = 0.81) and construct validity (correlations = 0.24 – 0.579. The PORPUS-USG also had good test-retest reliability (intraclass correlation coefficient = 0.66) and good construct validity with the HUI3 (correlation = 0.35), but not with other generic utility instruments or measures of symptoms.
Another study assessed the responsiveness of the PORPUS instruments10. Responsiveness is the ability of an instrument to detect change when a change in clinical status has occurred. In this study with 248 prostate cancer patients, the PORPUS-P, PORPUS-URS, PORPUS-USG, and PORPUS-U were consistently more responsive than generic utility instruments. In a subset of 142 newly diagnosed prostate cancer patients who were assessed before treatment and 3 months after treatment in this study, PORPUS-P scores decreased by a mean of 11 (scale range = 0-100), and PORPUS-U scores decreased by a mean of 0.09 (scale range = 0-1.0), reflecting treatment-related decrements in quality of life, whereas mean HUI score decreased by only 0.01 (scale range = 0-1.0). Although the scoring algorithm for the PORPUS-U has changed since the reporting of these results, we expect the revised PORPUS-U to perform well.
Published studies using the PORPUS
Differences between utilities derived from patients using the PORPUS and those from generic utility instruments have been reported611. The PORPUS has been used to assess the effects of androgen deprivation therapy12. The long-term effects of radical prostatectomy and radiation therapy have also been assessed using the PORPUS13. The PORPUS can be administered efficiently in a busy clinic using a handheld personal digital assistant14.
The PORPUS-U has been successfully used in a mail survey to obtain high quality utility data from nearly 700 prostate cancer survivors in three regions of the province of Ontario in Canada15, 16. These real-life utility data will be used in a cost-effectiveness model to inform policy on prostate cancer treatment.
Adaptations of the PORPUS to special populations
The PORPUS is being adapted for use in low-literacy patients17. The PORPUS-P has been translated into German and validated in a group of newly-diagnosed prostate cancer patients in Germany18.
Where to find more info
- this website
- Patient-Reported Outcomes and Quality of Life Instruments Database (www.proqolid.org).
The research on the PORPUS was supported by the following:
Grant #94081 from the Canadian Cancer Society
Grant #006169 from the Canadian Institutes of Health Research
Zeneca Canada Incorporated
Grant (NCIC 011393) from the Canadian Cancer Society
Grant #93258 from the Canadian Institutes of Health Research
Dr. Krahn has been supported by an Arthur Bond career award from Physicians' Services Incorporated Foundation, and a Scientist Award from the Canadian Institutes of Health Research. He is currently supported by the F. Norman Hughes Chair in Pharmacoeconomics, University of Toronto.
Other investigators who have contributed towards the PORPUS are Dr. Andrea Bezjak, Dr. Jane Irvine, Dr. Gary Naglie (supported by the Mary Trimmer Chair in Geriatric Medicine Research), Dr. Paul Ritvo, Dr. George Tomlinson, and Dr. John Trachtenberg.