|
|
Post by straightly on Jun 25, 2017 19:37:40 GMT -5
I am of the opinion that, whatever huddles hindering our Nrx, would also be stopping our refills. Now Nrx is coming around, which means we figure something out, we shouls start to see the refills compounding.
These in the know: what rate of retention should one expect? In other words, assuming our retention rate is terrible right now, how/when will know we have turned refill around also, or that we have a retention problem? 90 days from now assuming 90 days is the expected interval? Should we expect 159+385 refill then? How does one calculate?
|
|
|
|
Post by rockstarrick on Jun 26, 2017 0:15:20 GMT -5
Total rx - New rx = Refill, (return rx)
Assuming all scripts are for the same time period, (30 days or 90 days for example).
|
|
|
|
Post by matt on Jun 26, 2017 8:37:38 GMT -5
It is difficult to make an accurate calculation; you sort of need to track it for a period of time and then do some trend analysis. TRx is just the total number of scripts. Some are new and some are refills. NRx is the number of newly written scripts. Some are truly new (patients who have never been on Afrezza) and some that are renewal of expired prescriptions (more like refills in concept). The reporting services that aggregate the overall script numbers don't break out the truly new patients from the renewed script number (they will do that for an extra fee). The other complication is that some physicians will write 90 day fills with three refills (a year of product), some will only give six months, and most just starting a patient on a new medication will give just 30 days because they want to monitor the patient more closely at first. Likewise, some insurance companies that cover Afrezza, mostly on tier 3 or 4, will limit supplies to 30 days at a time. Given the mix of new patients, old patients with a new script, fill rates of 30, 60, and 90 days at a time, it is hard to come up with a simple calculation. Sanofi stated publicly that they saw a drop-out rate of 65% and that this was one of the reasons to discontinue the marketing relationship. That number is affected by free samples, co-pay programs, and other factors that hide the real economics to the patient. I don't think there is enough reliable data to know what the present drop-out rate is, and we won't know that for some months to come. |
|
|
|
Post by casualinvestor on Jun 26, 2017 12:21:12 GMT -5
Looking at Liane's spreadsheet, for the past few months (excluding the most recent week), the numbers have essentially been TRx is twice NRx.
There are outliers. When NRx jumps steeply, then of course TRx is less than 2x NRx. Going further back, during the Sanofi ramp-up the TRx numbers were 1.5x NRx, and when Sanofi sales started slacking, TRx were around 2.5-3x NRx
But in general, I'd say that people are getting either a no-refill perscription, or a perscription with 2 refills. The jump in sales dollars per script I think is covered by the larger titration packs. If I remember, the # of units per script went up 50-100%, depending on which pack.
If we assume that most of the scripts being written are 1-month with no-refill or 1-month with 2 refills, then about 50% of of scripts written are 1-month no-refill. Some of those are new users on Afrezza, some of them are insurance/doc limited to 1-month-no-refill.
But the other 50% would be 1-month-with-2-refills. Which would suggest that our retention rate is over 50%
The big assumption being that a large majority of scripts written are 1-month-no-refill or 1-month-with-2-refills
|
|
|
|
Post by mnholdem on Jun 26, 2017 12:26:58 GMT -5
What would be interesting is for the members who post the scripts to post an example of Novolog or Humalog and compare how their weekly refill % currently looks. Of course, this kind of exercise might be depressing once you see the amount of scripts written every week by the makers of RAA prandial insulin.  |
|
|
|
Post by babaoriley on Jun 26, 2017 12:34:33 GMT -5
It just seems to me that there ought to be a cumulative effect on total Rx, and I just don't see that, either during the SNY period or now ours. I worry that many users are not re-upping. If it's insurance, that's not good, if it's that it doesn't work well for them, that is bad.
|
|
|
|
Post by peppy on Jun 26, 2017 12:51:37 GMT -5
It just seems to me that there ought to be a cumulative effect on total Rx, and I just don't see that, either during the SNY period or now ours. I worry that many users are not re-upping. If it's insurance, that's not good, if it's that it doesn't work well for them, that is bad. I think it is working for them. Laura K still calling it a miracle and new users on instagram showing glucose reading on monitors. MNKD has the nurses now to teach dosing and the dosing guide quiz. Vdex sending people out with a cgm for a week. eric pounding with video's and sam as charming as ever, and living his life.
The place where I have seen the wear and tire, insurance, sweedee's dad and then the user using for 3 years, insurance now denying. Additionally a lady that was hammering on twitter regarding blue cross blue shield is happy, her husband covered. Matt wrote parapharing, if the PBM do not deliver to the manufacter the # , the PBM do not get their bonus rebate so to speak. It may be right there, why the man using for 3 years no longer approved.
Speaking to myself, it is the insurance stupid. Insulin in general, and meal time insulin specifically has always been very difficult to use, as shown in the glucose monitors on my avatar. Subq insulin is not walk in the park, users glucose levels go way high then way low. Afrezza doses differently. and then a walk in the park. It would help if people didn't have to work blind.
I remember when my dad was dying. I went down to the cardiac echo. an ultrasound. The picture so clear now I watch the vibration of the fibrillation of his heart values. I could see.
|
|
|
|
Post by straightly on Jun 26, 2017 19:27:34 GMT -5
It just seems to me that there ought to be a cumulative effect on total Rx, and I just don't see that, either during the SNY period or now ours. I worry that many users are not re-upping. If it's insurance, that's not good, if it's that it doesn't work well for them, that is bad. My guess is the dosage. When it was wrong, people would think AND say it does not work. Now this is fixed, we SHOULD see refill up in a few weeks. I do not have to know, of course, but I just wonder how do I ascertain that Mike really KNOW. |
|
|
|
Post by InvesterSam on Jun 27, 2017 2:53:21 GMT -5
I used a trial-and-error procedure to estimate the retention rate of Afrezza. Being an engineer, I first plot all data and found out there are trends with too much noise. I decided to use 5 week moving average (MA) (2 weeks before and after the current week). Patient group from one week would not have much connection to patient group from the next week; one group could be a very large size and the next group could be very small and keep refilling at different weeks. Averaging 4 or 5 weeks could include all patient group and minimizing the variation. The trend of the moving average seemed much clearer with less noise. I assumed that a patient who decided to use Afrezza would refill every 4 weeks. Thus,
Refill This Week = (NRx 4 weeks ago) x (Retention Rate) + Refill 4 weeks ago
By adjusting the retention rate, I had a pretty good match between the calculated and actual refill for 2015 data with retention rate of 12% (1 in 8 patients continued to refill). For recent data (August 2016 to June 2017), refill counts matched pretty well for the retention rate 3% or 4%. However, this model works well only for the steady enrollment and steady dropout. We had a significant changes in dropout rate and the retention rate 3% or 4% would not mean much anything. So I decided to match the slope of refill versus week to estimate the retention rate for the given 4-week prior NRx numbers.
For 1/27/2017-4/14/2017, the refill was flat only gaining 0.26/week on average with significant scatter of data (r-square = 9%). This slope (0.26/week) was attained with 1% retention rate for the NRx data.
For 4/14/2017-6/2/2017, the refill had a clear upward trend, gaining 3.1/week with r-square = 92%. The retention rate of 12% provided 3.1/week slope for the NRx data.
Refill = 3.09 x (week #) – 227
Week #1 = 1/30/2015, Week #125 = 6/16/2017, so on
I also plotted 2017 TRx and NRx (5 week MA) showing very clear trends;
1/6/2017-2/24/2017: No change in TRx and NRx
2/24/2017-6/2/2017: VERY VERY clear upward trend
TRx = 4.40 x (week #) – 240 r-square = 95% (15 data)
NRx = 2.83 x (week #) – 195 r-square = 92% (15 data)
Week #1 = 1/30/2015, … Week #125 = 6/16/2017, so on
Something happened on 2/24/2017 (or 2 week before that) and TRx and NRx started to increase. Then from 4/14/2017, the refill count started to increase.
Prediction: Using the best-fit equations above, I made prediction for the next week and the end of the year.
Week 126 (6/23/2017)
TRx = 314 (or = NRx + Refill = 323)
NRx = 162
Refill = 161 (or = TRx – NRx = 152)
Week 153 (12/29/2017)
TRx = 432 (or = NRx + Refill = 483)
NRx = 238
Refill = 245 (or = TRx – NRx = 194)
Possibility of Trend Change
For 6/16/2017, we saw an exceptionally high jump in NRx. In the previous 16 weeks, the maximum difference between the actual NRx and the predicted NRx is less than 32. For 6/16/2017, the actual NRx is greater than the predicted value by 67, more than twice of the previous maximum. Few more week’s data will tell us if it is an outlier or the start of a new trend.
It is my first message board post with any significant length. I tried to include several graphs using attachment method but I failed. Anyone who can generate 5 week moving average and simple EXCEL plots can see the trend. It is so clear.
|
|
|
|
Post by promann on Jun 27, 2017 5:16:17 GMT -5
Sam, Respectfully i think your end of year week #153 prediction is way to low. Not sure how that was figured. But I don't need any fancy math equations to know that it will be much larger then that. It's impossible to predict using any equations when we may be at a inflection point and tv advertising will bring even more awareness next month. Also Label change will be huge I think that will be a eye opener.
Would you care to predict end of year numbers by just using a good guess
Thanks.
|
|
|
|
Post by InvesterSam on Jun 27, 2017 9:58:14 GMT -5
Making predictions at inflection point (or transition) is almost impossible. I am not knowledgeable in sales and market either. The prediction next week (NRx = 162) was based on "No Inflection" or change of trend. So if we get NRx around 162 this Friday, it is more likely that we do not have change in trend. If NRx comes in 260-300, it is more likely we do have change in trend. For meaningful prediction, we still need at least couple more data (weeks).
Just simple math for Dec 29 2017 prediction only using the last two data points would be
NRx (12/29/2017) = (last change, 81/wk) x (28 more weeks) + (last NRx, 226) = 2494
With 50% discount of last change (40/wk) would give us NRx = 1346
Considering NRx change for 2/24/2017-6/2/2017 was 2.83 scripts/week, 40-80 scripts/wk is a huge change. Maybe doable if each dedicated sales rep makes one new script each week.
Improving the retention rate from the current 12% estimate could make another huge difference in TRx. It would take couple months to know for sure. We will see.
|
|
|
|
Post by casualinvestor on Jun 27, 2017 10:57:12 GMT -5
Improving the retention rate from the current 12% estimate could make another huge difference in TRx. It would take couple months to know for sure. We will see. I think that your estimated retention rate is ignoring the fact that NRx can include a retained person who has gone back to the doc for a new script. Which is something they need to do every 1-3 months, depending on doc or insurance policy. The lack of breakout between a "true" new person, and someone who has gone back to the doc for a new script after using all their refills, is what keeps us from getting an accurate retention #. |
|
|
|
Post by sportsrancho on Jun 27, 2017 10:59:38 GMT -5
For what's it's worth I know no one who gets monthly prescriptions. They are for 3 months.
So I just asked some people I know on Twitter. They said 3 months also.
Tom's kids have always been 3 months. I can't find anyone who's script is for a month ...
Sorry what I meant was the scripts are for a year and re-fill is every 3 months.
|
|
|
|
Post by compound26 on Jun 27, 2017 11:27:25 GMT -5
For what's it's worth I know no one who gets monthly prescriptions. They are for 3 months. So I just asked some people I know on Twitter. They said 3 months also. Tom's kids have always been 3 months. I can't find anyone who's script is for a month ... Sorry what I meant was the scripts are for a year and re-fill is every 3 months. I think Sam Finta used to refill monthly, but have recently switched to every 3 month. I think Spiro gets it monthly. Rick also refills monthly. stocktwits.com/rckhrrr |
|
|
|
Post by sportsrancho on Jun 27, 2017 12:09:35 GMT -5
For what's it's worth I know no one who gets monthly prescriptions. They are for 3 months. So I just asked some people I know on Twitter. They said 3 months also. Tom's kids have always been 3 months. I can't find anyone who's script is for a month ... Sorry what I meant was the scripts are for a year and re-fill is every 3 months. I think Sam Finta used to refill monthly, but have recently switched to every 3 month. I think Spiro gets it monthly. Rick also refills monthly. stocktwits.com/rckhrrr Jun. 25 at 1:55 PM rckhrrr @nwrong My Educated Guess is that Many New Scripts are Now being Written for a Minimum of 90 Days to START With! Better Results that Way! 👍 Bullish 1 5 via StockTwits for iOS Jun. 25 at 12:01 PM rckhrrr @nwrong No! I refill every 30 Days with 180 Pack! However You can get it every 90 days & that would Be (3) 180 Packs on one Pick Up! Bullish Thanks:-) |
|
