Test, Test, Test and Detect More!

India’s COVID-19 testing invited a lot of criticism as Indian Council for Medical Research (ICMR) hold on to a rigid and limited testing criteria, which was at best can be described as ‘don’t test, don’t find’. Finally ICMR forced to scale up their testing numbers, widen their laboratory network and also expanded the testing criteria. At present, there are 143 Government laboratories, 67 private laboratories and 3 collection sites, totalling 213 spread across 35 States and Union Territories of India except Lakshadweep Islands.

As per ICMR press release dated 13^th March 2020, they disclosed that they have tested more than 6500 samples from 5900 individuals and out of that 78 were confirmed as positive cases. As per the bulletin published by ICMR on 9^th April, 144910 samples from 130792 individuals were tested and out of that 5705 are detected as positives. It is also mentioned that within the last 24 hours, 16002 samples were tested, which is an appreciable achievement considering the earlier testing figures of an average 1806 tests between 13^th March to 31^st March and an average of 9349 tests between 1^st April to 5^th April. At present, India tests 107.6 people per million of population (ppm), which is still low compared with many countries around the world. India's abysmal low testing figures in comparison with other countries was discussed in my recent article in the WIRE as well as latest issue of FRONTLINE.

MORE TESTS MORE CASES

The following table illustrates how India’s testing figures changed between 13^th March and 9^th April.

Date	Total People Tested (cumulative)	Total Positives	% of Positives wrt People tested	Tested People per Million of Population
13-Mar	5900	78	1.32%	4.4
09-Apr	130792	5705	4.36%	97.1

As India’s testing density increased from 4.4 ppm to 107.6 ppm, we finally started to detect 3.94 percentage of positive COVID-19 from the total samples in comparison with 1.32 percentage as on 13^th March. More tests bring out more percentage of positive cases. This data buttressed the need for more testing and underlines that unless we test more, we are not going to find more infections. Many experts already cautioned India that social distancing and personal hygiene should be followed by a robust testing strategy. This is important to detect and isolate the infected share of population and it should be at the heart of India's strategy to 'flatten the curve'.

It is regretted to state that ICMR’s data outflow to the public from the very beginning was quite frugal, as their bulletins seldom disclose about State wise or laboratory wise figures. ICMR bulletins, until yesterday, messed up with the testing figures and the number of people tested, as a mysterious difference of 890 between samples vs individuals repeated for past many days. This this mysterious 890 number is really bewildering because this constant difference created a very fundamental question of how our laboratories certified the people, once they are cured without repeated tests while the number remained constant! This author raised this question through a twitter thread on 4^th April. Another problem noted with the ICMR website was no access to the archives of past bulletins, unless we religiously download all bulletins every day before ICMR purges those bulletins, we will lose all the past data!

NEED FOR STATE WISE DATA

It is very important to understand that how the testing is spread across various States of India, at least to grapple whether the State Governments are testing enough people and detecting and isolating the virus infected population. This author made a prolonged exercise to collect the above data for the past two weeks, by trawling to various State Governments' Health Department/Public Relation Department websites, dedicated Corona dash boards, the social media postings at Twitter and Facebook by the official handles. All the data thus scrapped from these sources finally put into an excel sheet to make the analysis. 

It was not an easy task at all, as many States were also displaying the similar stingy attitude of the Union Government, as they failed to update their websites regularly. Some States do not care to disclose the testing data in their websites. It is perplexing to note that certain State Governments simply abandoned their websites which was created for exclusively for COVID-19  and the started afresh new websites without giving any link in the old websites! All the South Indian States were very transparent from the very beginning to publish the data in public domain barring the lone case of Telangana Government, which abruptly stopped the disclosure of testing figures after 26^th March. When this author published the article on the WIRE, only 20 States data were detected by him, out of the total 36 States/UTs of India.

As of today, after a tedious chase and search, the testing data from 33 states/UTs are now being collected regularly. No testing figures pertaining to the three States/UTs viz. Lakshadweep, Meghalaya and Dadra and Nagar Haveli & Daman and Diu (DNHDD) are able to find out in the public domain. From the above 33 States/UTs, data pertaining to 30 States/UTs as of 9^th April is up to date. Telangana State, which did not disclose the testing figures after 26^th March, is definitely a huge disappointment. As the data pertaining to 9^th April is not available for Puduchery & Ladakh, their data are not considered for the analysis and discussion (Puduchery tested 201 people @131 ppm as on 7th April while Ladakh tested ~264 people @895 ppm as on 6th April).

While analysing the data, the sum of positive cases and negative cases were used to find out the total number of people tested, instead of taking the total samples sent/tested. These figures are available in most of the State Governments' press releases. A report titled “Report of the Technical Group on Population Projections, November 2019” published by National Commission on Population by Government of India is relied for the figure of 2020 population, which is used to find tested people per million population (ppm) data.

HIGH VARIANCE IN TESTING

The analysis revealed that Maharashtra, which had tested 30229 is at the top of number of tests done table, followed by Rajasthan (18211), Kerala (11826), Delhi (9363), Uttar Pradesh (8308) and Karnataka (7373). The sum of testing figures from the above six States represents 59 percent of the total people tested in India (1,44,910).

But the above testing numbers alone are not enough to reach a conclusion that these States are testing a fair representation of their population. Hence it is imperative to consider how much people are tested per million of the population and how many positive cases were detected with respect to the total tests done, both of which are important indicators to understand the nuances.

HIGH POPULOUS STATES

First of all, examine how the high populous states (population above 7.5 crores) had done testing of their people. The seven States coming in this category represents 60 percentage of the total population of India.

Name of State	Tests done	PPM	% +VE
Maharashtra	30229	245	4.51%
Rajasthan	18281	234	2.35%
Uttar Pradesh	8308	36	4.94%
Tamil Nadu	6658	88	12.53%
Bihar	4907	40	1.04%
Madhya Pradesh	4400	53	9.34%
West Bengal	1889	19	5.45%

Maharashtra and Rajasthan tested a representative sample of 245 ppm and 234 ppm, more than 200 percentage of the national average of 107.6 ppm. Meanwhile West Bengal conducted an abysmally low testing rate of 19 ppm, which is followed by Uttar Pradesh (36 ppm), Bihar (40 ppm), Madhya Pradesh (53 ppm) and Tamil Nadu (88 ppm) – all these figures are below the national average, which is really a worrying factor for this big States. Unless the last four big populous States test more people, we will be grapple in the dark about the status of infected people and definitely will be left in the lurch.

Another important factor is to be considered is the percentage of positive cases out of the total tested population. Here, Tamil Nadu is leading the tally with 12.53 percentage and Madhya Pradesh (9.34%), West Bengal (5.45%), Uttar Pradesh (4.94%) and Maharashtra (4.51%) are following, as all of them show the percentage of positive cases much above the national average of 3.94 percentage. Bihar shows an abysmal low of 1.04 percentage of positive cases is a curious case when considers their low representative testing data.

Nothing wrong in making a statistical assumption that Rajasthan with a high ppm and low positive rate is containing the virus better than Maharashtra with a similar ppm rate but high positive rate. The above may be a conjecture considering limited testing data and can be brushed aside as premature, but it is a good statistical indicator that how these States differ in testing and detectioaan.

MEDIUM POPULOUS STATES

In this category, the States with a population between 2 crores to 7 crores are considered. Twelve States are coming under this category, which represents 36 percentage of the population of India. So together with the high populous States discussed above, both category together represent 96 percentage of the total population of India. As no data from Telangana is forthcoming for the last two weeks, that State is not considered for this analysis (Telangana tested 1638 people @44ppm as of 26^th March).

Name of State / UT	Tests done	PPM	% +VE
Kerala	11826	335	3.02%
Delhi	9363	464	7.69%
Karnataka	7373	111	2.67%
Gujarat	6199	90	4.23%
Andhra Pradesh	5960	114	6.09%
Punjab	2907	97	4.47%
Odisha	2841	65	1.55%
Chhattisgarh	2630	90	0.68%
Assam	2195	63	1.28%
Haryana	1839	63	9.19%
Jharkhand	1239	33	1.05%

Here Kerala, who had tested 11826 people tops the list, followed by Delhi (9363), Karnataka (7373), Gujarat (6199) and Andhra Pradesh (5960). But if representative testing is considered, Delhi with 464 ppm lead the pack, followed by Kerala (335 ppm), Andhra Pradesh (114 ppm), Karnataka (111 ppm), Punjab (97 ppm) and Chattisgarh (90 ppm). The rest of the States show abysmally low representative testing figures.

When considering the percentage of positive cases, Kerala shows better control than Delhi, as it has only 3.02 percentage of positive cases while Delhi has 7.69 percentage of positive cases. Though Delhi Government squarely put this blame on a cluster due to Delhi conference, this author have a different take that Delhi's initial representative testing figures were low in comparison with Kerala's figures in the same period. Another factor influenced this was the number of people both States kept at institutional and home quarantine. While Delhi keeps 2664 people at institutional and 47,208 people at home quarantine, Kerala keep 723 people at institutional and 1,35,472 people at home quarantine. Moreover, Kerala is one of the rarest States in India, which shows the decrease of percentage of positive cases as the number of tests increase. The percentage of positives peaked to 3.81 percentage on 27^th March, which is now decreased to 3.02 percentage and Kerala’s infection curve started to show the flattening tendency, while Delhi's curve is on an exponential rising trajectory.

When consider Karnataka, Gujarat and Andhra Pradesh (population varying from 5.2 crores to 6.8 crores), it is evident that Karnataka is having a better control than the other two States. Karnataka has less percentage of positive cases (2.67%) than Andhra Pradesh (6.9%), when both have almost similar representative figures, ie, Andhra Pradesh (114 ppm) vs Karnataka (111 ppm). Meanwhile Gujarat with 90 ppm and 4.23% positive cases remains a cause of concern, as their death rate is highest in the India. The failure of early detection of cases and poor health infra of Gujarat should be attributed for their dismaying statistics.

Chhattisgarh shown a better control in comparison with Punjab and Haryana, all three is having more or less equal population of around 3 crores. Punjab among these three States, tested a high representative testing of 97 ppm, but the percentage positives is 4.47 percentage. The similar statistics for Haryana is 63 ppm and 9.19% positives indicates that they must do more rigorous testing, as their neighbouring Delhi tests at 464 ppm.

SMALL STATES & UNION TERRITORIES

Among the remaining small States and Union Territories, Jammu & Kashmir had done maximum tests of 2649 while Andaman & Nicobar Islands done India’s record representative testing of 1311 ppm. High percentage of positive cases at Chandigarh and Jammu & Kashmir are a cause of concern. Most of the North Eastern States are relatively insulated for the time being from the spread of COVID-19.

Meghalaya, Nagaland and Sikkim are not yet reported any infection cases along with Lakshadweep Islands.

Name of State / UT	Tests done	PPM	% +VE
Jammu and Kashmir	2649	199	6.95%
Uttarakhand	1270	113	2.76%
Himachal Pradesh	716	97	3.91%
Andaman & Nicobar	523	1311	2.10%
Goa	333	215	2.10%
Tripura	274	68	0.36%
Arunachal Pradesh	186	122	0.54%
Chandigarh	186	156	9.68%
Manipur	185	59	1.08%
Mizoram	75	62	1.33%
Nagaland	65	30	0.00%
Sikkim	48	72	0.00%

TESTING CRITERIA 

Many experts, journalists and civil society continuously criticised ICMR’s low-testing strategy in the beginning. All urged for more testing quoting the WHO Chief, who made an emotional appeal for more testing. A handful of health reporters repeatedly questioned this scant testing strategy during the joint press conferences held by the administrators of Ministry of Health and Family Welfare (MoHFW) along with the scientists of ICMR. Instead of being more transparent and proactive, the Government of India obtained a gag order from the Supreme Court of India against media!

Meanwhile, West Bengal Chief Minister Smt. Mamata Banarjee and Chattisgarh Health Minister Sri. T.S. Singh Deo raised the lack of availability of testing kits and also questioned ICMR’s rigid testing criteria.

All this forced ICMR to relax their rigid testing strategy in stages and also resulted in expansion of network of testing laboratories from 13 to 210.

The latest testing criteria was issued as of 9^th April (version 4.0), which fairly covers a good number of population. As per this latest criteria, the following persons should be tested.

• All symptomatic individuals who have undertaken international travel in the last 14 days
• All symptomatic contacts of laboratory confirmed cases
• All symptomatic health care workers
• All patients with Severe Acute Respiratory Illness (fever AND cough and/or shortness of breath)
• Asymptomatic direct and high-risk contacts of a confirmed case should be tested once between day 5 and day 14 of coming in his/her contact

Further ICMR decided to do rapid anti-body testing in hotspots/cluster (as per MoHFW) and in large migration gatherings/ evacuees centres for all symptomatic ILI (fever, cough, sore throat, runny nose). Hope with this expanded testing criteria and also with the introduction of rapid anti-body test kits, in addition to the RT-PCR testing, will definitely lead to more testing  and more detection of the COVID-19 infections.

CONCLUSION

Apparently, no state-wise testing data figures are being compiled and published in the rudimentary website maintained by the MoHFW. It is a tedious exercise to trawl at various State Government sources for this simple information of state wise figures of testing, which can be easily provided by either ICMR or MoHFW. It is regretted to state that MoHFW is not bother to take a lead role to create a baseline reporting time among the States and a standard format for reporting the data. The earlier mentioned messing of testing data by ICMR, mysterious 890, is quite damaging to their credibility as an institution. The corrupted data reporting by ICMR for many days reminds the very data credibility crisis faced by Reserve Bank of India, during the demonetisation fiasco. Across the globe, most of the institutional and governmental websites disseminate reliable and detailed information in a professional manner, that too in real time basis. Also many State Governments in India are running much superior dash boards than MoHFW website. Some examples are Maharashtra, Kerala, Karnataka, Gujarat, Rajasthan, Andhra Pradesh etc.

When people’s lives are at stake, it is quite distressing to witness the slapdash information management and poor data dissemination by the MoHFW. On the one hand, the Indian government touts Digital India at every turn, and greedily grabs every bit of data on its citizenry; on the other, they are frugal with the reciprocal of the vital information for the consumption of its citizenry. This is quite disquieting and disheartening!