Another 2 months of CMCO?

We have passed week 8 of the CMCO and can see the first sign of the decline in the number of weekly cases - refer to Chart 1.

We all know that the CMCO is not going to be as effective as the MCO.

For those who have forgotten, it took week 5 of the MCO to see the first sign of weekly decline.

Mathematically you can say that the CMCO is taking 60 % longer to bring the number of cases down

Chart 1

Positivity rates

The other good news is that after so many weeks, we are beginning to see a drop in the test positivity rates as can be seen in Chart 2.

The current weekly rate of 6.3% is close to the Mac 6.7% but at least is it coming down. 

I hope that this is the start of a downtrend and not some temporary decline like that for the week of 29 Oct - 4 Nov. 

Chart 2

The positivity rate together with a decline in the number of weekly cases is supposed to be a leading indicator.  So we are likely to see a weekly decline from now onwards

Keep our fingers crossed.

Regional Performace

If you breakdown the number of weekly cases by regions as shown in Chart 3 below, you can see that for Sabah & Kedah, the decline from the peak has been slow.  

At the projected rate as shown by the blue trend line,  it will probably take another month for the number of weekly cases in Sabah & Kedah to get below 500 cases per week.

Chart 3

If you follow the same trendline for the Other region, it would another 2 to 3 months before the Other regions reach a stage of fewer than 500 cases per week.

If correct we will have the CMCO ending only around Chinese New Year 2021. Is this too long?

The point is that for the Peninsular part of the country, we seemed to have new hot spots.  For example, Johore is coming up as the next area to be hit.

And for the central region of Selangor, KL, and Putrajaya, the 7 days moving average as can be seen from Chart 4 below has yet to show any real decline.

Chart 4

Conclusion

If you believe that numbers don't lie, then you would conclude as follows

• For the central region, I expect the CMCO to be extended for at least another month 

• For the whole country, expect to see more than 500 cases per week for another 2 months.

Note that during the Mac wave, we went from MCO to CMCO 8 weeks after the MCO was implemented.

If we accept that the current CMCO is taking 60% longer to bring the virus under control, then we would expect the virus to be under control after 13 weeks from the CMCO period ie around mid-Jan. 

OK, the above is a simplistic analysis but I have yet to see anyone providing a statistical or another mathematical model to predict how long the CMCO will last.

I am sure it will not end on 9 Dec for Selangor and KL

- - - - - - - - - - - - - - - - - - - - - - - - - - -

Be fearless to get more to Stay Safe, Share this Blog

PS: This blog is for me to better understand COVID 19 as this will impact my investments. If you are also into equities, follow me at i4value.asia

Disclaimer:  I am not an epidemiologist, healthcare worker, pharmacist, or staff in the Ministry of Health, but rather is someone with a strong interest in numerical analysis.  The content is an attempt to understand what is happening in the battle against COVID 19 from a data-based perspective. The opinions expressed here are based on information extracted from readily available public sources but I do not warrant its completeness or accuracy and should not be relied on as such.

We have dumb luck

This post is about how lucky we are to be staying in Malaysia during the Covid-19 pandemic because of the following:

• While the CMCO has not been as effective as the MCO in controlling the virus, it has managed to bring the number of cases in Sabah to a plateau.  Better than going up!

• While the number of cases for the rest of the country are still increasing, at least most of the economy is still able to function. It would have been sad to have a total lockdown with the cases still increasing. 

• Although the positivity rates are rising, it has yet to hit the Mac 2020 high.  So the spread within the community is not so bad comparatively. 

• During the Oct period while the number of cases were increasing, we had the ability to “create” beds to ensure that this does not become a medical capacity bottleneck.  We apparently did not have this ability during the Mac spike. 

• Despite the large number of sick patients, those that needed ventilation is very low.  So we can cope with even larger number of sick patients. 

If the whole purpose of the CMCO is to enable the economy to function as much as possible while keeping a lid on the spread of the virus, we have achieve this aim.

Whether the goal is to bring the virus down to single digits (excluding imports) is something interesting to watch as there is no evidence yet that this is working. 

I present my analysis below to support the above views.

Number of cases not really coming down

In my previous post, I compared the current weekly number of Covid-19 cases with those in Mac/April 2020.

An update of the comparison is shown below.

 

If it obvious that in terms of the time taken to bring the virus under control from the start of lockdown, the CMCO has not been as effective as the MCO.

• We are into week 5 of the current CMCO and the number of weekly cases is still increasing

• In Mac/Apr MCO lockdown, by the respective week 5, the virus had definitely been brought under control

But if you look at the performance of specific regions, you can see that 

• It does not appear to be increasing in Sabah/Kedah.  But looking at the chart, I would not conclude yet that it is trending down. It appears to be at a plateau.  

• It is growing in other areas

If you look at the chart for the Central region as per below, you would agree that there is no sign of any downtrend yet.

I don't think anyone should be surprised that the number of cases is still not coming down.  After all, the CMCO was originally a relaxation from the MCO.

We may have to learn to live with the virus while having the CMCO till the end of the year.  My view is because the positivity rates are still increasing. 

Positivity rates still increasing

The positivity rate is computed by dividing the number of Covid-19 cases detected by the number of tests carried out for the same period.

As can be seen from the positivity chart, the positivity rate has been trending upwards since the end of Sep 2020.

I would interpret this to mean that the number of cases within the general public is still increasing.  The "good news" is that

• It is lower than the 6.7% we had in the second half of Mac this year

• It is much lower than the 20% in some states in the US

• It is still below the WHO standard of 10%

MOH has been reporting that the R0 is coming down, but I have yet to hear any news about bringing the positivity rates down.

Mathematically, I would have thought that you could not have R0 coming down without the positivity rate trending in the same direction.

But then I am not an epidemiologist.  Sadly no one is mentioning this.

Maybe it is not important

Sabah bed magic

The objective of any lockdown is to flatten to curve so as not to stress the medical capacity.  One of my concern at the onset of this round of infection was the number of beds available. 

It seems that my worry is unfounded as we now have this "magic formula" that was not apparent during the Mac wave. 

There are 8,600 people under treatment currently in Sabah.  I am surprised that we have been able to cope because on 7 Oct when the CMCO was coming into effect, MOH had reported that there were 2,600 Covid beds with about 46% utilization.  

We now have about 8 times more patients than previously and there has not been any news about bed shortage.  Hats off to those that created the additional 6,000 odd beds.  Where did these come from? 

• On 13 Oct MOH reported that in Sabah there were 1081 beds at the 9 hospitals with 71 % utilized.  There are another 4,161 beds at 19 low risk quarantine and treatment centres. 

• Code Blue reported on 23 Oct that “Sabah has even broken away from federal MOH protocol on admitting confirmed coronavirus cases into a hospital, or quarantine and treatment centres for low-risk patients, as the state began allowing early-stage Covid-19 patients to be treated at home."

• On 28 Oct, the Sabah Local Govt and Housing Minister had told the press that the state had 7,273 beds for Covid-19 patients across 9 hospitals and 35 quarantine centres.  Source: Code Blue

• On 2 Nov, MOH reported that in total for Sabah there are 8,539 beds, which includes 1,111 at the nine hospitals and 7,428 at the low-risk centres.

Reading in between the lines from the above  reports, I would think that the additional beds are non-hospital ones. 

I guess these could be like those “temp” centres created in China or New York at their peak to cope with the number of patients.

So what does this all mean?

With the ability to "create" beds at low risk treatment centres, we will not run out of beds in the country.

So the challenge is PPE, ICU beds and ventilators should the number of cases in the rest of the country spike. 

Ventilator mystery

Secondly with respect to ventilators, we are lucky as the current usage has been very low.

Nationwide, during the month of Oct, the highest number of patient requiring ventilation was less than 40 when the number of patients ranged from 10,000 to 11,000.

I am not a doctor but the comparative numbers for Mac/April seems strange.  At that juncture we had about 50 to 60 ventilators in use when the number of people in treatment ranged from 2,000 to 3,000.

So do we have a different strain of virus, or are the people being infected today more healthy? 

The moral of the story? We only need to worry about staff and may be PPE when it comes to medical capacity.

- - - - - - - - - - - - - - - - - - - - - - - - - - -

Be fearless to get more to Stay Safe, Share this Blog

PS: This blog is for me to better understand COVID 19 as this will impact my investments. If you are also into equities, follow me at i4value.asia

Disclaimer:  I am not an epidemiologist, healthcare worker, pharmacist, or staff in the Ministry of Health, but rather is someone with a strong interest in numerical analysis.  The content is an attempt to understand what is happening in the battle against COVID 19 from a data-based perspective. The opinions expressed here are based on information extracted from readily available public sources but I do not warrant its completeness or accuracy and should not be relied on as such.

How effective is the CMCO compared to the MCO for Malaysia?

We closed the current week (22 to 28 Oct) with 6,484 cases.  This is higher than the 5,415 cases for the previous week 

This is not the first time that we have "lockdowns".

• We had the Mac MCO for the whole country although certain areas had the more stringent EMCO

• Currently, we have the CMCO for Sabah, Kedah, Selangor, KL, and Putrajaya.  OK a few areas are also under EMCO although we now call it TECMO (for temporary, as if we needed to remind people that it is not permanent)

The CMCO is not as stringent as the MCO in that most of the economic activities can continue under the CMCO.

Given the less stringent CMCO and that not all states are involved, does it mean that it will take a longer time c/w with Mac/April to bring the virus under control?

To get an answer to this question we compared the current pace of infections with those in Mac/April

Specifically, I used trendline to illustrate my point

When we talk of bringing the virus under control, there are two perspectives

• the first is just to get the number of cases to start trending down

• the second is to get the number of weekly cases to be below 10% of the number of cases at the peak week

At this stage, we are just looking at the former perspective

Synchronizing MCO and CMCO - Chart 1

Malaysia had started its MCO in the week of 12 to 18 Mac 2020 (week 1) and as can be seen from the Mac/Apr chart below, the number of weekly cases started to come down in week 5 (9 to 15 April).

The CMCO for Selangor, KL, and Putrajaya was implemented on week 15 to 21 Oct 2020. To synchronize the current situation with that of Mac, I have numbered this as week 1 in the Oct/Nov chart as per below.  

Based on this, we have just completed week 3 (22 to 28 Oct).

If we assume that we have the Mac trajectory, then the current weekly number of cases would only come down in week 5 ie (week 5 to 11 Nov) 

But if the CMCO type of lockdown is not as effective as the MCO type of lockdown, then it would take longer.

Do the statistics so far give any clue on whether it would take longer?

Chart 1: Comparative charts

Trendline projections - Chart 2

To illustrate the pace of the infection, I used trend lines to show how the number of cases changed from week to week.

Referring to the Mac/Apr chart below

• The red line shows the trend from week 1 to week 2

• The green line shows the trend from week 2 to week 3

You will notice that the green line is less steep than the red line implying that the pace of infection has come down. 

You will also see that for week 3 to 4 of Mac/Apr, we have the same green trendline. 

It meant that from week 2 to 4, the increase in the number of cases per week was about the same.  For those who slept thru maths class, it meant that the pace of infection has reduced.  

The same increase in the number of cases but a slower pace?  How can this be?  Mathematically it is correct as the pace is based on dividing the increase in the number of cases by the previous week's total number of cases. A bigger total for the previous week will result in a slower pace.

To compare what is happening currently, I have plotted similar trendlines for the Oct/Nov chart as below. 

Reading from the green trendline in the Oct/Nov chart

• If the virus is controlled as effectively as in Mac/Apr, then for next week we would expect to have 7,500 cases

• If it is more, then we can conclude that the CMCO is not going to be as effective and we can see more lockdown extensions

• If on the other hand, the number of cases is less than 7,500, we can be confident that the CMCO is working better. 

Chart 2: Comparative charts with trendlines 

Sabah TEMCO

You may ask why don’t I start plotting the lines from when the TEMCO started in Sabah?

I think that even with using simple tools like trendlines we ought to be comparing apples to apples.

The Sabah TEMCO is only confined to Sabah and thus would not be reflective of what is happening nationwide.

But if they had ban flights from Sabah or had the quarantine for those flying in from Sabah, then I would have counted the TEMCO as week 1 and the virus would probably be under control by now. 

But this is with hindsight and does not help.  But if we have another state election in Sarawak, this could be an important lesson. 

Conclusion

The real question is if the number of cases next week is less than 7,500 does it mean that the number of cases would start to come down in week 5 or would it take longer?

In theory, if the CMCO is less effective, it would take longer.

At this juncture, the trendlines are not able to give any insight into this question.

My forecast is that the week of 29 Oct to 4 Nov would still have a higher number of weekly cases ie > 6,484.  This will be the peak and thereafter it will start to decline.  

But it will take longer than week 6 to get below the peak. We should not be surprised as the CMCO is not as effective as the MCO

But if the number of cases really comes down, then we have a way to control the virus by keeping the economy going.

My view of any "lockdown" is to bring the virus to a level that is below the capacity of the medical system.

For Peninsular Malaysia, we are still operating below our medical capacity.  But for Sabah, I think it is up to their neck. 

- - - - - - - - - - - - - - - - - - - - - - - - - - -

Be fearless to get more to Stay Safe, Share this Blog

PS: This blog is for me to better understand COVID 19 as this will impact my investments. If you are also into equities, follow me at i4value.asia

Disclaimer:  I am not an epidemiologist, healthcare worker, pharmacist, or staff in the Ministry of Health, but rather is someone with a strong interest in numerical analysis.  The content is an attempt to understand what is happening in the battle against COVID 19 from a data-based perspective. The opinions expressed here are based on information extracted from readily available public sources but I do not warrant its completeness or accuracy and should not be relied on as such.

Will the current CMCO for Selangor, KL and Putrajaya really end on 27th Oct 2020?

This was a dramatic week for Malaysia in the context of Covid-19

Firstly, we have ended the week of 8 to 14 Oct with a record number of weekly cases. As can be seen from the chart below, we had 3,547 new Covid-19 cases for the week whereas the previous peak of Mac/April was 610 cases.

The other dramatic things that happened this week were:

• The CMCO was implemented for Selangor, KL, and Putrajaya.  This kicked in on 14th Oct and is supposed to end on 27th Oct 2020.

• The govt finally decided to implement CMCO for the whole of Sabah after weeks of implementing TEMCO for certain districts

The CMCO for the whole of Selangor was of course challenged by the State Govt - the question was why the whole state and not confined it to just the red zones. 

The reason given by our PM was that the other districts in the state also had an alarming increase and this was a precautionary measure.

If you listen to the news about what is happening to the US, there is a benefit of taking measures early.

The only problem with the PM reasoning was that we did not have a consistent policy.

• Sabah should have been placed under lockdown much earlier. 

• Travel restrictions to and from Sabah should have also been implemented earlier rather than from 12th Oct.

But I would have thought the implementing it for KL, Putrajaya and most of the Klang Valley makes sense because people stay and work in different places in these regions.  It would be a big challenge to control traffic if only certain districts were under the CMCO.

While the CMCO is not as restrictive as the MCO in that most businesses are allowed to operate, the question on my mind is whether the CMCO would actually end on 27th Oct.

In other words, what is the likelihood that it will be extended?  

 

What will happen in 2 weeks time? 

If we have got the virus under control in 2 weeks' time,  we will revert back to the RMCO. 

Unfortunately, I am not confident about this. 

If you look at the historical performance as per the above chart you will note that it will take more than 2 weeks to bring the virus under control as follows:

• From the Mac/Apr 2020 peak, it dropped to about 40% of the peak after 2 weeks

• For the week of 21 to 27 May peak, you will see that after 2 weeks, it is still around half of the peak. 

So if we assume the 40 % drop,  we can expect the number of cases to drop to about 1,400 cases in 2 weeks' time.

This is even higher than the peak numbers in the Mac/April wave. 

If we have these 1,400 weekly cases by the end of Oct, I wonder what would be the justification for ending the CMCO. 

If there is to be any hope that the current restriction would be eased for Selangor or Sabah, it would be for a dramatic drop in the number of cases next week.

So let us see the actual number next week.  If it has not dropped to 40% of the peak, I am quite confident that the “restrictions” would be extended. 

If you want to look at precedent, we went from CMCO to RMCO on 9th June 2020.  The announcement was made on 7th June

• During week 4 to 10 June, we had 368 new Covid-19 cases nationwide.  So we could use the 368 as one metric

•  Another thing is that one week later ie 11 to 17 June we had 177 new Covid-19 cases. It would appear that after lifting the restrictions the number of cases dropped by half.  So another metric to look at is to see that the number of Covid-19 cases is dropping dramatically ie to half of the current week numbers. 

Do you think that we can meet both these criteria on 27th Oct?  Unlikely if the current path follows the Mac/April path

• It will take more than 2 weeks for the national number of weekly new cases to come down to 368 cases

• I have no confidence that the rate of decline in the number of cases would touch 50%

Regional level

Given that the majority of the nationwide cases are from Sabah/Kedah, maybe we have to look at this at the district or even at the state level when looking at Selangor, KL, and Putrajaya.

The chart below shows the number of cases by regions

The number of weekly cases for Sabah/Kedah continues to grow at 15% compared to last week whereas for the other parts of the country the growth was 92%

I would describe the situation for the other parts of the country as “exponential growth.”

Yes, the rate of growth is slowing down but at 92% compared to the 233 % growth rate the week before it is still exponential growth.

Since the main areas in the “others” category are Selangor, KL, and Putrajaya, I wanted to see whether a more granular analysis might give a better picture of what is going to happen in 2 weeks' time. 

I look at 2 statistics

• The number of daily new cases

• The number of active cases

Let see the statistics for Selangor, KL, and Putrajaya during the last time we went from CMCO into RMCO.  

• On 6 Jun, Selangor had 230 active cases with 13 local cases (ie excl imported) for the day.

• On  6 Jun, Putrajaya had 4 active cases with 0 daily cases

• I could not find the equivalent statistic for KL

• During the week, the 3 areas had 310 cases in total

The table summarizes the current position compared with the historical situation. 

Period	No of weekly cases (excl imported)	No of active case
4 to 10 Jun	310	234 (as of 6 June)    (a)
1 to 7 Oct	253	262 (as of 7 Oct)     (a)
8 to 14 Oct	444	502 (as of 14 Oct)

Note (a) I could not find any statistics for KL

So based on the historical performance, I would say that the CMCO for Selangor, KL, and Putrajaya could end on 27 Oct if

• The number of new cases for the week for the region is about 280 

• The number of active cases for the region on 27th Oct is around 250 cases 

The target numbers are based on the average respective numbers of the first week of Jun and Oct respectively.

I think that the other indicator is that there should not be a red zone in any of the districts within these 3 areas. 

For Malaysia, MOH has classified each district based on the number of active Covid-19 cases in the district

• Red zones are those with more than 41 active cases or more

• Green zones are those with zero cases

• Yellow zone are those with 1 to 40 cases

Since the CMCO just started for Selangor, KL, and Putrajaya, it is likely that the number of cases for the week of 8 to 14 Jun is not the peak ie I expect the number for the coming week to be higher.

If so, and given that it will take more than 2 weeks for the numbers to be halved,  it is unlikely for these regions to come out of the restrictions come 27th Oct. 

Another way to predict the next CMCO or TEMCO

We have precedents that if the number of active cases in a particular district goes about 70,  it is likely to be placed under TEMCO.

We now have the example that if several districts within a state have more than 70 active cases, the whole state would be placed under CMCO eg Sabah and Selangor.

Are there other indicators?

I would think that looking at the number of clusters may give a clue. 

When Malaysia went into its Mar MCO, we could say that the virus was new and the MCO was to flatten the curve so as not to overwhelm our medical capacity.

At that juncture, I suspect that we were still guessing the actual need for beds, ICU facilities, and ventilators.  I remember MOH mentioning about building up more capacity during the first few weeks of the lockdown

Today we have a better picture of the treatment needs and it would appear that the current lockdown is not about flattening the curve to below our treatment capacity.

Rather my reading is that it is our contact tracing capacity that is being overwhelmed.

Our Covid-19 strategy is to contact trace and isolate all contacts. This is why MOH reports the progress made for each cluster.

So if we get to a stage where we have difficulty in contact tracing, having the CMCO would help.

My guess is that since we had 54 active clusters at the time the CMCO was announced, this can be another metric to monitor in the future when trying to forecast whether we will go into another restriction. 

Of course, this is a national figure but it can give a picture of how serious the Covid-19 situation is.

Conclusion

• I don't think the CMCO for Selangor, KL, and Putrajaya will end on 27th Oct.  This is because it will take more than 2 weeks to bring the numbers down to the target numbers

• I see the following as the target numbers for Selangor, KL, and Putrajaya
• 280 new Covid-19 cases for the week 22 to 28 Oct
• 250 active Covid-19 cases on 27th Oct
• No red zones and the number of cases are falling

Let us monitor the situation over the next 2 weeks to see whether there is any hope of ending the restriction come 27th Oct

- - - - - - - - - - - - - - - - - - - - - - - - - - -

Be fearless to get more to Stay Safe, Share this Blog

PS: This blog is for me to better understand COVID 19 as this will impact my investments. If you are also into equities, follow me at i4value.asia

Disclaimer:  I am not an epidemiologist, healthcare worker, pharmacist, or staff in the Ministry of Health, but rather is someone with a strong interest in numerical analysis.  The content is an attempt to understand what is happening in the battle against COVID 19 from a data-based perspective. The opinions expressed here are based on information extracted from readily available public sources but I do not warrant its completeness or accuracy and should not be relied on as such.

Will the rest of Malaysia see exponential growth of Covid-19 cases?

So we have another record day with 691 Covid-19 cases. 

The current week cases (from 1 Oct to 6 Oct, based on 6 days) compared to the rest of the year can be seen from the chart below.  You can see that even though there is still one more day to go for the current week, it is already a record week.

The media is calling the current situation the 3rd wave as they consider the first wave as the Jan cases and the 2nd wave as the Mac cases

Either I have something wrong with my eyesight or I have to go back to “curve drawing” class

• The highest weekly number of cases in Jan was just 7 cases

• By this definition of a wave, then the week of 21 to 27 May should also be another wave

• Accordingly, what we are having this week should be the 4th wave

I hope the media is not taking the counting cue from the authorities.

If we have a counting problem, then we should not be surprised that nobody is analyzing the numbers to find out the real situation. 

Technically we already have exponential growth for the country even based on 6 days for the current week as can be seen from the table.

Week	No of week cases	% growth c/w previous week
10 to 16 Sep	448
17 to 23 Sep	474	6 %
24 to 30 Sep	719	52 %
1 to 6 Oct (6 days)	2280	217 %

Our PM has just told people not to panic even though we can expect a higher number over the next couple of days. 

Actually, our PM is right because

• The majority of the current wave is in Sabah and Kedah where EMCO has been put in place

• The sporadic cases in the rest of the country are still below my exponential trigger point of 30 cases per day

So it is not time yet to panic over an exponential growth for the rest of the country (excluding Sabah and Kedah)

Details of my rationale are presented below.

Sabah and Kedah 

The chart below shows the local cases for Sabah and Kedah c/w the other states in the country. 

We have to remember that a big portion of the increase is from Sabah and Kedah where TEMCO has been instituted.

• From 1 to 6 Oct, Sabah and Kedah accounted for 85 % of the total number of cases.  The week before the 2 states accounted for 83 % of the total number of cases

• Given that part of Sabah and Kedah are under TEMCO, we would expect the number of cases in these 2 states to be brought under control within the next 2 to 3 weeks. 

• The numbers in these states grew because of the delay in implementing the TEMCO. We can argue the reasons for the delay but it is academic now since the lockdown has been executed.  

The issue is how long it would take to bring the number of cases down.  Given our history, I would expect it to go down to under 5% of its current level within a month.

The real question is that what are the risk to the rest of the country in the meantime. 

Flying in from Sabah

Over the past 6 days (1 to 6 Oct) we have an average of 28 cases daily due to those to coming back from the high-risk area (Sabah) as reported by MOH.  

I remember reading a comment from one of the Ministers that the number of cases due to people coming back from Sabah is not significant.

This is of course not accurate if you compare the 28 daily cases with the 57 total daily cases excluding Sabah and Kedah.  You can see that about half of the non-Sabah and Kedah cases are due to people coming back from Sabah.  

The only way it can be insignificant is to look at the 28 daily cases in the context of all the cases in the country.  OK, so we now know some of the ministers can’t count. 

It would appear that there are not many people flying back from Kedah.  Of course, there are no figures on people traveling interstate from Kedah.  But I think we can assume that there will not be those from the TEMCO areas in Kedah going to the rest of the country.

So the risk is still from people flying in from Sabah.

Until we have the mandatory quarantine for interstate travel from Sabah, we will continue to see cases due to travelers from the state.  

You can understand my disappointment when PM 

• Did not announce some border checks for traffic from Kedah

• Announced a delay in the ban of flights from Sabah till 12 Oct.  

Clusters

The chart below shows the breakdown in the number of cases from 1 to 6 Oct into

• Those from clusters

• Imported ones (there are actually only 8 cases so it was not significant enough to show up in the chart)

• Those from misc sources

About 1/3 of all the cases from 1 Oct to 6 Oct are from misc sources.  The clusters account for most of the balance. 

As of 6 Oct, we have 39 active clusters for the whole country.  

• Of these, 3 (2 in Sabah and 1 in Kedah) account for 79 % of all the cum cases for these 39 clusters.  

• Note that one each in Sabah and Kedah are prison clusters where the spread has been high due to the closed nature of the prison. 

Leaving aside the prisons, the MOH track record for clusters is that general almost all are “de-activated” ie not grown any more within 1 month of being identified. 

So I would imaging that these 39 clusters would not be a source of exponential growth.

The other thing about the cluster is the most of them don’t really exhibit significant growth.  

• In fact of the 122 clusters in the country from the start of the year, only 24 have more than 50 total cases each by the time the cluster is “deactivated”.

• It is the top 10 clusters that account for 55% of the total cases in the country. 

The main point is that we don’t really have to worry about clusters.  The ones that got us into trouble are due to Sri Petaling (our first experience) and the prison/detention ones (confined space)

Sporadic cases

The risk of exponential growth then comes from the non-cluster. 

Over the period from 1 to 6 Oct we had a national average of 117 new cases under the misc category.  I consider these as “sporadic” cases

Excluding the sporadic cases from Sabah and Kedah, we had about 31 cases per day for the rest of the country due to sporadic cases 

• As per my previous post, this is just at my exponential trigger point of 30 daily cases.

• Whether these 31 daily cases will cause exponential growth will depend on out contact tracing and isolation efforts.  

The important question then is whether in the coming week, 

• the 31 daily cases will lead to new clusters and grow exponentially before they are “extinguished” by either EMCO or the normal contact tracing and isolation 

OR

• there will be new sporadic cases 

I am more concerned about the latter.  If over the next one week we actually have double the number of new sporadic cases ie not due to the clusters from the current week's cases, it will be time to panic. 

If it remains around or below the trigger point, then I think we don’t need to worry.

Countrywide MCO

I have often said that unless the number of cases is going to exceed our capacity to handle them - from the treatment and contact tracing perspectives, we should avoid another nationwide MCO.

By all means, lockdown specific locations with the TEMCO.  But I don’t see why the rest of the country should be lockdown if the number of cases is within our medical capacity.

I think the PM articulated the same point in his address on 6th Oct.

So I think we are safe from another countrywide lockdown. 

The risk is localized TEMCO.  So, monitor the number of cases within your locality if you want to project what is going to happen to your residential area. 

- - - - - - - - - - - - - - - - - - - - - - - - - - -

Be fearless to get more to Stay Safe, Share this Blog

PS: This blog is for me to better understand COVID 19 as this will impact my investments. If you are also into equities, follow me at i4value.asia

Disclaimer:  I am not an epidemiologist, healthcare worker, pharmacist, or staff in the Ministry of Health, but rather is someone with a strong interest in numerical analysis.  The content is an attempt to understand what is happening in the battle against COVID 19 from a data-based perspective. The opinions expressed here are based on information extracted from readily available public sources but I do not warrant its completeness or accuracy and should not be relied on as such.

Will Klang Valley be the next EMCO zone?

My last post was on 10 Sep when I concluded that the Sabah spike will not lead to another wave in the country.

I had expected the pattern to be similar to the ones we had over the Semenyih detention centre since the Sabah spike was also in a “captive” environment.

For reasons yet to be explained by the authorities, the “containment” was not as effective as the Semenyih detention cluster and so we had an increase in the number of weekly cases since then as can be seen from the weekly chart below.

The current pattern is different from the last spike (shown by the red oval).

In the chart, I have also shown by the dotted line the number of weekly cases when we just entered into the MCO.

You can see that 

• The current number of weekly cases has gone above the level when we just went into the MCO

• Unlike the last spike, this current one did not “fade” away. Rather it has been building up for the past 3 weeks and I suspect that the numbers will continue to be high for the next 2 weeks until MOH brings the situation under control.

I also know that this is not a normal situation by looking at the number of daily tests reported for Sep 2020.  As you can see from the chart below, there has not been any reports for the past week. 

No of daily tests for Malaysia as reported by Worldodometer

By right there should be daily reports as we know that Covid-19 tests are being carried out daily.  

In the past, we have had times when the tests for a couple of days were not reported.  This situation arises when the private labs have yet to report in their test results.

As you can see from the chart, usually there would be missing reports for 2 or 3 days and then all the results for the past few days get reported at one go and you have the jump in the number of tests.  When this happens I just average it out to get the daily average.

The issue is not the daily average. The issue here is that there has not been any test report for the past 8 days.  

I would interpret that the failure in providing the test reports to mean that we suddenly have an overwhelming number of tests being conducted.  

We should not be surprised by the increased number of tests as MOH protocol is to contact trace and tests all.  Given the increase in the number of positive cases, there would be an increased in the number of contacts to be tested.

The real question is whether the number of cases being reported reflects the current situation ie are there under-reporting since we don't have all the test results?

What does this all mean?

Is the country facing another exponential growth?

In my last post, I said that it is time to worry if the “sporadic cases” ie those due to unknown causes reaches 30 cases daily.

So far all the large number of cases have been confined to Sabah and certain regions there have been placed under EMCO.

The EMCO has been effective in preventing the spread of the virus in Kedah.  No doubt the number of cases would be high for a week or two before the containment effect kicks in.  

No matter what you may think about whether it is right to restrict the movement of groups of people, the EMCO together with contact tracing and treatment has been an effective way to control the spread of the virus.

So I am confident that by mid-Oct, the number of cases in Sabah would be brought under control.

  

As of 29 Sep, there are 888 active cases in Sabah.  We have yet to hear any news about whether the hospitals or healthcare workers in Sabah are facing capacity issues. 

Based on a conservative 50 cases per day for the next 2 weeks (ie assume 2 weeks treatment) I would expect the total number of active cases to peak at 1,500.  If this was the Klang Valley, I would not worry about capacity issues. 

But I don’t know enough about Sabah to say whether this would be a problem.

If MOH or the new state govt is transparent, they should state whether they have the capacity to handle all of them.

Spread to the Klang Valley

Leaving aside the question of Sabah, the real concern is the people travelling between Sabah and other parts of Malaysia.

I think that hopefully with the EMCO, the people within the EMCO zones in Sabah would not be able to travel.

But then there are others in Sabah who are travelling.  The news has reported about the congestion in KLIA due to people coming from Sabah.

MOH and the govt are practising double standards - they have mandatory quarantine (in quarantine centres) for all those coming in from other countries whereas, for those coming from Sabah, the mandatory quarantine seem to apply to only those tested positive.

I think the Klang Valley is heading for trouble.

• To get immediate results at KLIA, they must be using some rapid anti-gen testing.  But this does not detect those who just caught the virus and have yet to show any symptoms.  So the likelihood is that we have potentially sick people slipping through

• All those coming back who do not show any signs of the virus are told to quarantine at home.  I think we already have a history that home quarantine does not work.  

We went from mandatory quarantine at specialized centres to home quarantine for a while.  And then back to specialized centres when the authorities realized that we don’t really have civic conscious people.

 

Now we are making the same mistake of having home quarantine.  I hope this is not politically motivated as many of those coming back from Sabah are politicians who went over to campaign for the recent elections. 

The conclusion is that we are allowing potential sick people into the country.

So I expect a spike in the number of cases all over the country.

MOH has already reported that the Rt, ie the reproduction number at a particular time t, for Selangor is higher than that for Sabah.  I find this surprising but I suspect it is cos Selangor is starting with a small base number and an increase in the number of cases would lead to a faster increase in Rt.

Will we have exponential growth in the Klang Valley?

It depends on how widespread the infection is.  The Sri Petaling event led to the nationwide MCO because of the sheer numbers to test and contact trace.

By the same analogy, if we have 15,000 people or more coming back from Sabah over the few days, and there are many who are sick, it could overwhelm our contact tracing capability.

I think this is the weak point - I have not seen any reports on the numbers coming back but it must be substantial as the airport had to arrange for additional counters. 

You must remember that while we have data on the number of beds, ventilators and even the healthcare workers, I have not seen any data on the contact tracing capacity.

Yes, we are better off today compared with the Sri Petaling time due to my Sejahtera and other contact tracing SOP.

But the system has yet to be tested for a spike situation like now. 

The only good point is that the authorities seem to be willing to have locations based EMCO. 

For the general public, the unknowns are

• Will your residential area be subjected to EMCO?  I am sure that if there are more than a dozen cases within your residential areas over a short period of time, you can expect an EMCO.  You don't have control.  The only control you have is to get your provisions ready. 

• How safe are you when going out?  When the number of cases in a particular zone is very low, you can ignore the SOP eg wearing mask, social distancing without any real danger.  But when the cases are on the rise and especially since we know that there are many asymptomatic cases, you ignore the SOP at your peril.  So avoid going out and if you have to, imagine everyone is a potential case.  My favourite story of everyone being a zombie is now more relevant than ever

I don't think there will be exponential growth...more likely expect some EMCO

- - - - - - - - - - - - - - - - - - - - - - - - - - -

Be fearless to get more to Stay Safe, Share this Blog

PS: This blog is for me to better understand COVID 19 as this will impact my investments. If you are also into equities, follow me at i4value.asia

Disclaimer:  I am not an epidemiologist, healthcare worker, pharmacist, or staff in the Ministry of Health, but rather is someone with a strong interest in numerical analysis.  The content is an attempt to understand what is happening in the battle against COVID 19 from a data-based perspective. The opinions expressed here are based on information extracted from readily available public sources but I do not warrant its completeness or accuracy and should not be relied on as such.

Will there be spikes but no new waves?

Note:  Over the past few months there have been several posts that looked at surges in the number of cases.  For example, the one on 5 May viewed the surge from 5 different angles to assess whether the surge would turn into another wave.  This post is another perspective of assessing why the spike would not turn into another wave. 

I guess I should start by clarifying what I meant as "spike" and "wave"

• I consider a spike as a sudden short-lived increase in the number of cases.  Under this scenario, there is unlikely for any exponential growth

• A wave would be a prolonged increase in the number of cases.  There is a danger that if not managed well, this could lead to exponential growth requiring possibly another lockdown

The chart below illustrates the difference

Over the past few days, there has been a spike in the number of cases

• It was 62 cases on 7 Sept up from 6 cases the day before

• It went further up to 100 cases on 8 Sept

• But it came down to 24 on 9 Sept

The sudden surge was due to 2 new clusters - Benteng Lahad Datu and Sungai.

It is not the end of the story for these 2 clusters because as of 9 Sep because there are still people waiting for test results 

• 1307 people from the Benteng cluster

• 619 people from the Sungai cluster

And this is assuming that there will not be another generation of infection from these clusters

So will the current surge in the number of cases be a spike or the start of another wave?

Second surge

This is the second major spike in the number of cases since the Tabligh event in Mac of this year.  

As can be seen from the chart of weekly cases since the start of the infection, the last spike was towards the end of May due to the infection among the detainees. 

The current spike is slightly different as it is due to two separate clusters happening together

• Those in the detention centres - Benteng Lahad Datu cluster

• A large cluster in Kedah - Sungai cluster

The May detention spike did not lead to an increase in the number of cases among the general public as it was confined to those in the detention centres. I suspect that the contact tracing and isolation of the staff in the detention centres also help to contain the spread.

Along the same lines, there is unlikely to have any large transmission from the Lahad Datu cluster to the general public since most of the cases were among the prisoners - of the 66 tested positive, 64 were prisoners.

The Sungai cluster is more challenging as the index case was a healthcare worker at a private medical centre.  It managed to spread to the index case families and colleagues.

There are two questions with the current spike

• Will we continue to see new spikes in future - my answer is yes

• Will the spikes be contained especially if it is not due to cases in a detained environment - again it is likely to be contained.  In other words, I don't expect an exponential growth

Further Spikes in Future

My rationale on why we would continue to see spikes in the future is based on the following

• The virus has not been eliminated in Malaysia.  Accordingly, there will always be some local who gets infected (hereinafter referred to as index case)

• Malaysia has a targeted testing programme.  As such unless the index case is part of the targeted testing group (eg those in the old folk's homes, healthcare workers, those undergoing surgery) there is the possibility that the index case would infect other before he/she falls sick enough to seek treatment.  Worst would be for the index case to be asymptomatic as more people would be infected before the cluster is detected. 

• Given the 1 to 2 weeks incubation period and/or the 2 to 3 weeks treatment period (for the asymptomatic cases), it is likely that a cluster would form

• Eventually, someone from this cluster would seek treatment.  At this juncture, MOH would get into the picture and commence contact tracing and testing of all the contacts.

• The cluster would be identified and the number of people in the cluster would depend on the time the index case became infected to the time testing of the contacts was done.  Historically, it can be as few as 3 cases to as high 100 cases.

• The high numbers would cause the spike

In summary, we would have periodic spikes because of our targeted testing programme.  

Having said that, there is a mystery on why the healthcare worker in the Sungai index case did not get detected earlier as healthcare workers are part of the targeted group.

Containing the spike

Before answering this question, it is instructive to see why we had the MCO in March. 

The MCO was instituted due to several reasons

• The number of index cases due to the Tabligh event was too great for our contact tracing team then to handle.  Besides the authorities had an initial problem identifying all the people who attended the event.

• At the same time, there was concern that there could be a new wave from those returning to Malaysia after going overseas for the school holidays.

• In March the virus was still new and I suspect the authorities were still trying to address the medical treatment capacities. The MCO was seen as a way to flatten the curve to be below the treatment capacities. 

We managed to bring the virus under control with the MCO.  But don’t forget that during the MCO, there was a lot done to trace those infected and to quarantine them and their contacts.

At the same time, the new wave from those coming back did not arise

Today, we are unlikely to see another exponential growth because

• We have the contact tracing SOP (eg recording of visitors, MySejarhtera, etc) and so we should not have any issues with contact tracing

• Secondly, there are still restrictions for large gathering so any infections would be limited to small groups.  This makes contact tracing and isolating the infected more manageable

• The authorities have used the EMCO to restrict movements in critical areas to assist testing.  I expect this will still be used in a future surge in cases. 

• Our protocol of isolating all - index cases, contacts - will limit the transmission from the cluster to the general public.

Misc

There are two other practices that will help to keep the number of cases in Malaysia low and indirectly reduces the prospects of infections - ie no infections = no spike or wave. 

• We have the 14 days quarantine measure for incoming visitors. Hopefully, this will reduce new sources of infections

• We test enough as shown by the low positivity rates. The positivity rates for the current spike is lower than those for the May spike.  

Item	Period	No of cases	No of tests	Positivity
May spike	14 - 27 May	840	248,223	0.34
Current spike	27 Aug - 9 Sep	262	141,417	0.19

If you want a comparison, the current positivity rates in the US is double digits % whereas in Malaysia we are talking of less than 0.5 %.

Conclusion

Malaysia is one of the few countries that have brought the number of cases below 5% of the peak.

Having achieved this, we will have occasional spikes because we do not conduct mass testing.

Nevertheless, when a cluster is detected, MOH contact trace and test all and this protocol will ensure that the surge in cases does not turn into a wave ie we will not have exponential growth.

This will mean that while there will be the occasional spikes, there will not be another wave. 

- - - - - - - - - - - - - - - - - - - - - - - - - - -

Be fearless to get more to Stay Safe, Share this Blog

PS: This blog is for me to better understand COVID 19 as this will impact my investments. If you are also into equities, follow me at i4value.asia

Disclaimer:  I am not an epidemiologist, healthcare worker, pharmacist, or staff in the Ministry of Health, but rather is someone with a strong interest in numerical analysis.  The content is an attempt to understand what is happening in the battle against COVID 19 from a data-based perspective. The opinions expressed here are based on information extracted from readily available public sources but I do not warrant its completeness or accuracy and should not be relied on as such.