Knowledge Capability and Capacity in Biomedical Field
Generation of knowledge, capability and the capacity in the biomedical area confronts a complex milieu. This milieu is defined by the notion of and the practice of health and the structures and institutions of the biomedical market, which in turn depends upon the dynamics of contests between three major types of economic and innovative institutions of – the drugs, the medical devices and the physicians. Most often capability in developing drugs has alone been considered as the indicator of capability in the biomedical, leaving aside two other important facts – (1) the contests between physicians, medical devices and the drugs to control innovation outcome and also the fact that (2) physicians and devices constitute very important factors contributing to the formation of significant portion of the market of health.
In the subsequent sections generation of knowledge, capability and capacity in the bio-medical area has been examined in the light of the contests between these three types of institutions characterised first by the practices pertaining to discovering or developing solutions based upon drug, and second by the healing practices of physicians and the third by diagnostics as based upon devices (this entire account is from Banerjee, 2007). Such contests define the current endowment of resources, flow as well as the allocation of future resources, formation of coveted assets and their values, and the formation of capabilities and capacities.
Resources are distributed over several markets that in turn are governed by distinct and different sets of rules. There are multiple institutions that compete between them for getting access to as also in engendering greater resources and its distribution within one select institution. Organization of corporation including that of public R&D, unfortunately is a recipient of resources, which the organization cannot based-upon its internal strategic governance gather. Resources are generated through institutions and allocated as well as controlled through contests between institutions. Resources would then flow past one institution to be available with another. The circuit of governance is thus much larger and it goes much beyond the inside of the firm or public R&D organization to that outside where a firm/organization compete for negotiating power.
Innovation is about how to shape asset-outcomes in multiple institutions. Institutional contests in the case of biomedical for example, remain innovative for both rendering more valuable assets under own control while devaluing assets under the control of contesting institutions. This appears to be strange. Innovation studies particularly in relation to biomedical often remain delimited by specific sector or by specific firm. Others trace the difference in networking between academia and industry. Biomedical innovation offers strange circumstances inviting involvement by multiple institutions. Simply put, biomedical innovation is the outcome of contests between institutions. Innovation is the key instrument in arranging a contest. Innovation refers to only such changes that attract or command greater or most voting. The regulator, the legislator or the judges and bureaucrats oversee voting. Regulator has no or little role in typically Schumpeterian innovation. Contrarily, regulator has the most important role in the context of institutional contests.
Contests between the three institutions related to drugs, devices and the physicians take place through sets of institutionalised practices/rules. These are: the clinical observational and interactive practices undertaken by the physician; next is the theory-driven and theory-building open and not-for-profit knowledge practices of the university; and the third is the closed-door for-profit organized science-research often controlled by the pharmaceutical firms. There are finer subdivisions within these sets. The devices research while in practice shares the same institution with the drug firms, in market these are fiercely competing for the grab and they compete to define different sets of practices for the patient. Institution of devices maps a body differently from the mapping by the drug firm. We can, however, overlook this since Indian biomedical system has no capability in devices and till date device-intensity in medical setting is far lower compared to advanced countries.
In short we can call the first institution as clinical or hospital, and the second as university, and the third as organized research. Clinical institution has features uniquely differentiating this from other modes of practices and other streams of knowledge’s. University offers contrast. Disciplines of knowledge undergird practice in university. Disciplinary groups and sharing practice as well substantial part of the accumulated knowledge mostly within the boundaries of disciplines have spawned of turfs. Organized research is instrumental and goal oriented. When undertaken by business firm great secrecy is necessarily maintained. Public R&D organization often resides in the interface between university and organized research.
Institutional contests for domination refers to the power to: define agenda; influence and shape practices; therefore decide definition of what constitutes an asset or contrarily to render assets formed disjoint and beyond integration within that institutional practices; hence disallow accumulation; thus weaken reproduction of the field of practices; achieve above through innovations defined as novel mode to force re-ordering of fields, practices, assets and their accumulation as well as values; induce flow of most of funds and other resources to own institution.
In the following sections contests between hospitals and universities as well as the organized research has been discussed to exhibit both the processes of knowledge generation and the building up of capacity and capability. Further in the global context especially in the context of China as a comparator country, this analysis explores what remains as potential gains for India and what Indian policies could potentially alter the contemporary rules of contests for innovations.
Background: A brief statistical account of the current states of affairs would be useful before detailed discussions on R&D capacities and capabilities are taken up. The number of medical colleges under allopathy has been steadily progressing in the country. In 2006 there were 242 such medical colleges while the numbers were 189 and 125 in the years 2001 and 1987 respectively. All such colleges are teaching and healthcare facilities. Large number of physicians are private general practitioners and nearly equal numbers are retained with government agencies, 65,576 physicians were retained by the latter around 2003-04. Most medical expenses are out of pocket. NSSO estimations suggest that in rural area the monthly per capita expenditure on health rose from 4.2% in 1987-88 to 6.6% in 2004-05 of total monthly consumption of Rs. 158.1 and Rs. 559 respectively in current prices (NSSO Reports 458, 490, and 515). Similarly the urban consumer’s monthly per capita expenditure on health rose from 3.3% to 5.2% respectively. Medical insurance provision is emerging and currently covers a small section of the population. Very large number of the population depends upon public provisioning of healthcare and out of pocket expenses.
Around 2006 the provisional figures on rural health facilities indicate that on average a single rural Primary Health Centre (PHC) covers 32,754 and a Community Health Centre (CHC) covers 1,90,000 persons. The 3,910 CHCs had in 2006 total number of 3,979 doctors under the specialists categories of surgeons, pediatricians, and similar others. Total number of beds under the hospital system was 6,65,639 in 1998 with 4,818 government hospitals providing for 4,18,008 beds and the rest provided for by the private hospitals. Total number of reported and treated cases under ‘diarrheal’ category was largest in 2004 and was 95,75,112 from the number of 81,56,688 in 1997, similarly cases under ‘malaria’ were 9,11,293 and 26,60,057 in 2004 and 1997 respectively, and under ‘acute respiratory infection’ were 2,55,71,757 and 1,38,27,662 in years 2004 and 1997 respectively, or under ‘tuberculosis’ were for the same set of years 7,31,952 and 8,76,956 respectively. Number of persons per 1000 of population in ailment over 30 days were (derived by the NSSO 52nd Round, 1995-96) 86 in rural areas and 84 in the urban areas, and from this it could be appreciated that an average physician examines very large number of patients afflicted with multiple diseases. This provides the Indian doctors with extremely rich repertoire of clinical knowledge. Further, genetic diversity in India is perhaps the highest in the world. Doctors thus possess the richest source of knowledge for advancing innovations in biomedical areas.
However, institutionalized system of innovation refers to development of drugs as the principal objective. Our study below indicates that the two research systems of development of drugs and clinical healing practices shared very little, if at all, respective knowledge and experiences. Further, and given the fact that domestic manufacturing of medical devices remains very weak, the sharing of knowledge and research between devices and clinical practices have been perhaps nil.
The R&D in drugs and pharmaceuticals in India has been dominated by the chemical approaches and routes. Biochemistry knowledge in this country has been very strong and so have been several areas of chemical. Relatively laggard both biology and biotechnology are yet to catch up with the capability level of chemical routes. Most business firms and public R&Ds as well university R&Ds in drugs and pharmaceuticals have therefore developed upon the relatively cheaper and surer chemical routes. Other sections of this Report provide detailed discussions on the R&D by firms and public facilities, hence a brief introduction alone would be provided here.
The business of drugs really received the fillip with the opening up of the market for generics in the USA. In parallel the domestic market grew up. Firms could take up manufacturing generics and later even discovering new molecules of drugs because the public R&D facilities, especially from the CSIR system, provided them with molecules, detailed knowhow and pharmaceuticals knowledge. However, following WTO and TRIPS and changes in business strategy of very large global drug firms of taking generics into fold have over the last few years forced domestic drug firms and domestic R&D systems rethinking on the sustainability of the previous research as well as business strategy. Currently large global firms have initiated an acquisition (as well as mergers where applicable) strategy. In fact, over last two years a few important Indian firms have been acquired by global firms. The generics market of USA and most OECD countries have offered greater resistances to Indian firms’ entry modes through intensified litigation strategy, raised surveillance and higher regulatory controls. In Indian market the global large firms are in the process of securing patents’ rights and are pushing new drugs. Both global devices companies and drug companies in alliances with companies from insurances business are redefining health-market through (1) acquisitions of generics business, (2) acquisitions of new molecules, (3) shifting the market to vaccines and public procurement, (4) tie-ups with new insurance products and enhanced public coverage of cost of health, (5) redefining regulations through enhanced global controls, (6) joint strategy of devices and drugs/vaccines/generics companies, (7) digitizing and personalizing medicines, and similar others.
Innovativeness of typical drug firms is typically observed through such indicator as R&D expenditure to sales, and efficiency of expenditure through say R&D expenditure per Rupees to sales of new drugs; or otherwise ratio of new drug sales to total drug sales, or R&D expenditure per new drug, and similar indicators. However, under the present circumstances when global market of health is getting restructured and dominance of global large drugs/devices companies is extraordinary – the Indian strategy for innovation in biomedical or in other words, in the health sector, need look beyond these indicators of innovation.
In its early years, the domestic drug firms with the support of domestic public R&D took upon cost containment strategy. Cost containment offered the opportunity to export old, off-patent formulations that faced severe price competitions in markets abroad. There has been little revenue in this route therefore. Volume has increased since 1980 in two decades by more than 100 times in value terms and export has never been lower than 55% often much more than that. The country imported net of exports in value terms till late eighties, and since then the country has remained net exporter. In bulk drugs, volumes in value terms increased by nearly 300 times over this two decades and exports have increased over the years from a low of about 24% to a high of about 45%. By 2003 there was 400 bulk drugs manufacturer with output valued at Rupees 78 billion. In generics or formulations there are 60,000 items spread over 60 therapeutic categories, and total export has crossed Rupees 141 billion by 2003. A large number of pharmaceutical firms are small or medium, most often unlisted and the total of firms is about 10,000 out of which only about 300 are in the listed organized category. Again amongst these listed firms most are of medium size with only a handful of firms that are large by Indian standards. A large number of these listed firms have US FDA approved GMP certificates; a few have the WHO certified GMP. By 2003 number of Drug Master Files (DMF) filed with US FDA was 126. With industry-total of R&D expenditure at Rupees 6.6 billion by 2002-03 with industry turnover of Rupees 226 billion, only a handful of firms claimed spending about 6% of sales turnover on R&D with the rest below 2% in early years the new millennium. Few domestic firms have been developing new molecules for new discovery drugs. Currently more than 10 molecules are under various stages of development. In the last few years the R&D structure of Indian companies have been changing – (1) a few firms have raised R&D as % of turnover to 12% or above while a larger numbers stopped reporting for R&D expenses, (2) most R&D portfolios have been hived off from the revenue earning main business lines, (3) R&D on generics has come down and for new molecules has often been given up, (4) alliances with large global firms have increased, and (5) the stock market has reacted negatively in general to innovation based value propositions.
A typical discovery route can now be exhibited. The stages are:
(1) Target identification (through contributions from biosciences), to
(2) Target validation, to
(3) Biochemical assay (application of biochemistry), to
(4) Strategy based drug design (using often high throughput screening approach), to
(5) Actives, to
(6) Secondary assays, to
(7) Hits (compounds), to
(8) Optimization, to
(9) Leads (through Lead identification campaign, called LI), to
(10) Drug metabolism and pharmacokinetics (DMDK), to
(11) Optimization (Lead optimization, or called LO phase) to
(12) Toxicological tests, to
(13) Candidate drug compound, to
(14) Rigorous toxicological studies, to
(15) Investigative new drug (called IND), to
(16) Phases I, II and III clinical trials, to
(17) Registering as drug.
The time that these stages take is long, spreading beyond 10 years in general, and the cost particularly in the phases of clinical trials is very large unaffordable to the domestic drug firms. As a result in all cases of new molecules local firms have entered into alliances with large global firms.
Several other firms have entered the new fields of contract research and contract manufacturing. Large part of the former is devoted to the undertaking of clinical trials, a field that has recently opened up and is currently experiencing high growth. However, in most cases these two types have left out the very large and distributed knowledge base with the doctors. Doctors, however, are the real repositories of ground data in India. Handful of these practitioners ever gets involved in drug-testing related clinical trials. Therefore most publications from hospitals of India are unrelated to the segment of research production and standardization of practices that appears to be most active in the USA and other advanced countries. This distinction of the doctors in India is remarkable especially when compared to the degree of integration of a researcher from major organizations with its global counterparts. In the domain of practices Indian doctors, however, can be classed into a few groups. Those from research hospitals even while distanced from regular research are in close interactions with regular disciplinary researchers, such as from genetics. Their practice-set naturally differs from those who in distant cities practice in neighborhoods. In between belongs another group based in large cities and who sometimes through medical representatives or through occasional interactions with researcher-physicians develop semblances of research-informed practices. Typically bundling activities of these three groups of doctors differ significantly.
Journals rarely reach a doctor. Evidence based practices are most often unknown. Regulations by the medical association are limited in their reach. A doctor retains, however, close network linkages with her local peers and her potential and current transactions partners. Reputation asset remains therefore highly localized. Network-routed information sends signals to a practitioner on un-theorized practices including surgical methods, on local epidemiological trends, on the local portfolio of drugs and devices, and similar others. The important asset with this doctor is her knowledge of the local bodies, and her knowledge of how to investigate a local body through dialogue for example.
The worlds of organized research and that of the university are very far from the world of clinical practice. Only very indirectly large research can influence this practice. Large research, as a global institution, however, affects immediately and directly the Indian organized research. Networks between Indian major/research university and the Indian organized research on the one hand and the global majors on the other hand, are very active. Therefore, even in the absence of very large programs in India the domestic counterparts become a partner in the global program, somewhat as a laggard and as the supplier of human resources. The same are however, not true for both the institutions of distributed innovations and the clinical practice. Distributed innovation processes and the bundling by doctors differ across regions being more directly controlled by global processes in research-hospitals or large private for-profit hospitals in the large cities to regions that are indirectly related or unrelated to the global brand-devices institutionalization. Clinical practice enjoys highest freedom or in other words, from the perspective of research-intensive for-profit biomedical assets Indian clinical practice lags by decades, perhaps. Clinical practice institution has two faces then. In one its practices are most genuine to the patient. In this face the doctor and patient together unravels body. The other face of clinician informs us that little contemporary advances in genetics, for example, reaches her. Biomedical innovation in India must stand on these three pillars and the associated institutions of university and organized research. Can it however, follow as a laggard the path treaded by USA and other advanced countries? Could India adopt a strategy of late entrant? Should India, knowing that a laggard can never reap innovator’s profit tread the same innovational and institutional strategies as have been followed by USA and other advanced countries?
Research from hospitals: evidence
In order to estimate the extent of research and capability of research of both medical and non-medical streams we collected data from a public database, the PubMed MEDLINE database (via http://www.ncbi.nlm.nih.gov/PubMedOld/medline.html). This is run by the National Library of Medicine of USA and covers about 10 million articles from about 4,000 scientific and medical journals. This database is possibly the most reputed and most frequently visited, and we analysed the data available there.
There are several publication types in which articles in the database have been classified. These are called publication types (PT) (Ojasoo, Maisonneuve and Dore, 2001). Examples of these PT are Clinical Trial (CT), Randomised Controlled Trial (RCT), Controlled Clinical Trial (CCT), Multicenter Study (MS), Review of Reported Cases (RRC), Case Reports (CR), Practice Guideline (PG), Clinical Conference (CC), and similar others.
A CT publication refers to pre-planned clinical study of diagnostic, therapeutic or prophylactic drugs, devices or techniques on humans. Similarly, RCT publication refers to such a clinical trial as involves at least one test treatment and one control treatment, and also involves randomized selection of administered treatment. The CCT refers to a clinical trial involving one or more test treatments, at least one control treatment and pre-specified measure for evaluation. A MS refers to controlled study conducted in several cooperating organizations. This group refers to clinical trial or studies relating to administration of drugs or a new procedure or devices. Therefore, this group reflects research in which PH/RP (a Physician-healer, PH, has the goal of healing and is the node for all data pertaining to the patient; a RP is researcher-physician with the goal of research on diseases and necessarily can access limited data) is involved in research directed or influenced by non-medical such as pharmaceutical firm or equipment firm. Such studies identify major role of a PH first as RP with the rider that the RP acts as an intelligent provider of health services. Coordination of research is another important characteristic here.
PG caters to principles or directions to support physicians with patient care decisions regarding diagnostic, therapeutic or other clinical procedures. CC tells us about conference of physicians on their observations of patient at bedside. RRC refers to review literature on all known cases of a disease. Finally, CR reports on patient cases. This group of research refers to dominant role of physician as researcher, and the RP relates directly to the patient as a healer and much less as a provider of certain drug or technique or device.
The change in relative importance of each of the above publication types has been presented in Table 1. The first group of research where physician acts more as provider of clinical services, has grown faster overall compared to the overall growth in the second group where physician acts more as healer-researcher. One caution we should note that several publications are reported simultaneously under different publication types, for example, a single publication under CT, CCT and MS is quite common. It is better that we look at growth of group (that is CT, RCT, CCT and MS in 1st group; PG, CC, RRc and CR in the 2nd group) rather than on individual segment. In the second group share of PG and CC are negligible, and the share of RRC too is very small. As a result PG grew fast on a very small base, so is the case with CC. However, RRC with relatively larger base grew less and growth of CR with a very large base was minimum. This was not the case with the first group. Considering the limitation on first group that several publications have been reported under multiple headings, the overall growth is above the second group. In particular, the multi-centered study (MS) grew fastest. General impression from Table-1 is that physicians have been dominated by non-medical approaches to research. The second group has another novel aspect. Few publications here may come under multiple heads of publication types. Considering this aspect, the overall impression that physicians are turning out as providers could prove to be mistaken. On the balance, we could draw a tentative conclusion that PH is increasingly becoming a RP; and RP is increasing the Mode-2 type of research with physician as the key to research while possibly, and may be equally significantly, systemic pressures are increasing on physician to cooperate often as the second fiddle on research agenda set by non-medical business interest.
Table 1: Global growth across types of publications in MEDLINE
1999 (March) |
248238 |
113363 |
48254 |
27826 |
3664 |
2568 |
35098 |
917175 |
2005 (Feb.) |
395409 |
195709 |
67268 |
61393 |
8583 |
4301 |
51165 |
1154745 |
2006 |
36021 |
17150 |
2654 |
10489 |
1139 |
297 |
|
54902 |
2007 |
37568 |
18398 |
2031 |
12049 |
1012 |
271 |
|
56573 |
NOTE: CT= Clinical trial; RCT= Randomized controlled trial; CCT= Controlled clinical trial; MS= Multicenter study; PG= Practice guideline; CC= Clinical conference; RRC= Review of reported cases; CR= Case reports
Source: Ojasoo, Maisonneuve & Dore, 2001; except for CR
Perhaps looking at the country break-up of research publication types would reveal both variations in and relative importance of research participation by physicians. Table 2 provides data on India, China, Israel, USA and UK in year 2005 on distribution of publications over types. Looking at the first group (CT, RCT, CCT & MS) we observe that while the relative distribution of publications from a country over four types remain similar, in both USA and China the Multicenter study (MS) occupies importance. MS represents the degree to which hospitals or clinical research centers can be organized by either the country bureaucracy (as in China) or by the large and powerful business interests from pharmaceuticals or devices firms. In India MS occupies a weak position. If we compare the relative contribution of a country in the global total of type-publication (such as in CT) both India and China contribute less than on average 0.5%, Israel contributes about 1%, USA about 15% (however, in MS the contribution by USA is above 25%) and UK about 4%.
Table 2: Distribution of publication types across countries in 2008 (February) in MEDLINE
Country |
CT |
RCT |
CCT |
MS |
PG |
CC |
RRC |
CR |
India |
3574 |
1969 |
417 |
280 |
16 |
36 |
619 |
12821 |
China |
6011 |
3619 |
553 |
735 |
7 |
21 |
257 |
5819 |
Israel |
5038 |
2284 |
645 |
878 |
25 |
59 |
851 |
11655 |
UK |
25158 |
13523 |
2570 |
6565 |
509 |
92 |
1200 |
29544 |
USA |
100957 |
50463 |
8117 |
27288 |
1907 |
2032 |
7426 |
125829 |
Table 2, above, when read for the second group of publication types (PG, CC, RRC & CR) reveals a picture somewhat different from the above. Overall, the contributions from India and China are better in this group compared to the contribution from USA. Further, the relative distribution of publications from a country over four types do not remain similar, and in contrast to the picture from the first group here both CC and RRC are fairly high especially for India. In fact Practice guideline (PG) often represents a set of guidance’s from a business firm over how to manage health care, and in PG India as well as China is relatively weak. However, in both Clinical conferences (CC) and Review of reported cases (RRC) India’s relative contribution is higher and in fact much ahead of China. The Case reports (CR) too suggest a similar strength. Moreover, in terms of share in the global total of publication USA contributes less than 10% in CR and about 10% in RRC, India is close to 1% in both. CC in particular represents a domain of the physicians and USA shares about 30% of global reporting, suggesting both the dominance of organized health delivery in USA and the relative autonomy of the physicians’ discourse.
CR and RRC (in particular the latter) represent several aspects: relative autonomy of the discourse of physicians, coexistence of both unorganized and organized modes of health delivery, the relative abundance and importance of Mode-1 research, and the importance of public sphere amongst the physicians. Given these, Indian physicians appears to enjoy higher autonomy, larger public sphere and less occupied to serve organized health delivery. This reading is further supported when the figures for first group are compared with figures for the second group for each country. In both India and China autonomy of practice and public space of discourse for the physicians enjoy relative strength. Conversely physicians here are less subjected to the institutional compulsions to undertake clinical trials or follow guidelines suggested by non-medical systems of knowledge. Israel has certain strangeness; it shows coexistence of both organized and directed clinical practice together with less-organized and relatively autonomous actions by physicians. USA and UK exhibit dominance of organized and directed physician practice, however, USA simultaneously represent the existence of spheres of autonomy.
Table 3: Distribution of total (1991-2001) publication over types of institutions
Country |
Medical/Hospital |
University |
Others/Research |
India |
50.2 |
35.1* |
14.7 |
China |
48.8 |
43.4 |
7.9 |
* This includes research institution’s contribution at 10.3%, and university’s at 24.8%
Another important aspect is the emergence of data as the strongest component in biomedical research. All the types of publications, in particular CT, RCT, CCT, CR and RRC, are thick in data and research here is often driven by and founded upon data. It appears more than one million publications belong to this category. Growth of this segment as shown in first Table, over a period of six years has been about 40%. Largeness of data and overwhelming importance of data-driven research has been determining the course of innovation and research in biomedical. USA enjoys strong dominance in this data based research; however, some other OECD countries too are strong contenders.
More often than not research in biomedical is conducted by physicians on patients and as part of ongoing clinical investigation or healing. Few such researches are conducted outside the world of physicians’ practice. A typical stand-alone investigation or experiment as is the wont in chemical or biological areas can be budgeted for under R&D and can be conducted in a separate entity. However, typically medical or most of biomedical investigations would be field-based. As a result if we have to appreciate the extent of research in biomedical it would be preferable to consider statistics not of R&D expenditure instead it should be output. Output again could be published as publications or patented. Biomedical research is data driven, and several forms of property rights including copyright, in particular of digital form data, ensures that little of such output gets patented, and most gets published as research publications. For this reason precisely we consider research publication output extensively in order to appreciate nature of research in biomedical.
Another doubt might still prevail; this is about why we should consider global output and global comparison while investigating capability of a country institution in biomedical innovation! An analogy might clarify better this choice of method. Very similar to globalization of finance, the globalization of knowledge in particular of data makes possible appropriation of the factor of knowledge even while the knowledge-generating person remains stationery. Appropriation of this factor would depend firstly on the nature of property rights, and secondly on the degree to which data of a particular research could be de-contextualised. High context specificity implies that data pertains to an individual or a small group of individual and to a particular context. The group one type of research such as the clinical trial de-contextualises data while much of case reports asserts context specificity. Generalization of context-specific data, such as through the publication type Meta analysis (where independent studies are synthesized through quantitative or statistical techniques), achieves liquidity of data.
In Mode-1 research inferences based on past data are made by a physician on another new specific context. In this case the physician without resorting to framing a general rule makes inferences on applicability of past data and learning on the context of another new patient or disease. Each Mode-1 inference therefore generates another context-specific data, and the field of applicability of data increases. In Mode-2, however, rule formation is the principal objective. Past data are now made amenable to all future contexts. Typically, business firms from pharmaceuticals took to this latter route. This generalization generates a property right that can be appropriated at any future space and time. In contrast, Mode-1 generates a property right delimited by specific space and time.
Globalization of data therefore could lead to generation of two broad modes of property rights: Mode-1 engendering small parcels of variable properties; Mode-2 engendering most liquid property. The former can generate business model that transacts in rights to property without severance of ownership of property, while the latter transacts not in rights but in alienation of ownership, thereby increasing the size of property. A physician ordinarily is happy with the former while the pharmaceutical firm looks for the latter.
Both types of property, however, refer to data generated beyond the local. Traditionally, a multinational pharmaceutical firm alone used to look for global data. Local physicians could care little for global data because the data was out of reach. However, with increasing digitization reach of both group are increasing and as a result it is necessary for us to care for global comparisons of research and data-generation.
In order to appreciate the extent and variety of biomedical research we approached once again the MEDLINE database. We downloaded data for eleven years (1991-2001) of research output for seven countries, namely, UK, Israel, India, Singapore, Malaysia, China and Chile; and seven years (1995-2001) of data on output from USA. Data on US production for years 1991-95 was not reliable. Later the publication data was upgraded to recent years. Because of the extensive coverage that MEDLINE has we believe that our data is fairly representative of country publications. For countries such as China, however, there remains a very large amount of data in native language. Chinese data appears to under-represent the capability of the country. Another important point – unlike the data in the above Tables where single entry could appear in multiple reporting of publication types, in this part of data we dealt directly with entries, and therefore this data would not over-represent the capability of a country.
Next two Figures represent output comparisons between these eight countries. The overwhelming dominance of output of USA can be appreciated from the fact that six years output of USA when related to the total output of seven countries for eleven years, the former alone retains a share of 69.8%, followed by the UK with 17.7%, followed subsequently by Israel with 4%, China 3.8%, India 3.1%, Singapore 0.7%, Chile 0.6% and Malaysia 0.2%. Growth rates of countries, however, differ remarkably. China changed the course around 1995 with steep rise followed subsequently by very steep rise since 1999. China was behind India till 1996, and currently China is on the top (Figure 2). Growth in Indian output can be noticed since 1997, however, because of lower growth rate the early privilege of India lost to Chinese publications in English alone. Israel experienced nearly no growth, although with initial starting figure Israel remained close to top and much above Malaysia, Chile and Singapore who too experienced marginal growth. Next Figure shows growth differences as well as the base-differences between USA and UK. USA excepting for one dip experienced higher growth rate over a very high initial starting figure of publication, and as a result USA increased the differences between its publication figures and the figures for all other countries.
Fig 1: Trends for Number of Papers (Medline-Database) for different Countries
Fig 2: Trends for Number of Publications (Medline-Database) for different Countries
Institutional distribution of biomedical knowledge
Geographic distribution of assets compounded by the institutional distribution followed by organizational compounding characterizes the distribution of assets in a country, and precisely this aspect we studied in the previous section. Often hospital research because of the addresses used has been classified under the university system – this system with 43.4% of country total (including part from hospital) is largely distributed across universities in this coastal region.
Table 4: Medical and non-medical authors in country publications of India, China and Israel
Country/Author |
Total author |
Number of authors from medical |
Number of authors from non-medical |
India |
3808 |
1742 |
2066 |
China |
8029 |
1761 |
6240 |
Israel |
458 |
36 |
422 |
Source: Scidirect
Distribution of biomedical research from the university institution of India exhibits characteristics similar to China. Indian research output, however, always is in English, and as a result this figure is more representative of Indian states of affairs than the corresponding figure on China. In several cases of Indian universities, such as with the Banaras Hindu University (BHU) close to the East, the output is indeed from the hospital system and is not from the university departments. BHU tops the list of Indian university institution whose contribution to the country total at lower than 24.8% is much smaller than its Chinese counterpart. Second highest contribution is from the IISc at Bangalore in South, followed by two universities of Delhi from North, followed by Punjab university in North, followed by in close succession two universities from Kolkata in East, and similar others. Regionally as well as spatially somewhat evenly balanced research assets in biomedical is also characterized by non-dominance of any particular university organization. A related issue is the long-tailed distribution of research assets in the university organizations reflected in the contribution by the ‘others’ category at 33.5%. In the case of China this category is much smaller with 19% of country total coming out of multiple smaller universities.
Fig 3: Publications percent in Universities (24.8%) of India
Fig 4: Publications percent in Institutions (10.3%) of India
Fig 5: Publications percent in Medical Colleges/Centers & Hospitals (50.2%) of India
Research assets in biomedical contributed by public organized research are, however, geographically skewed. Public organized research (with contribution to country total at 10.3%) excluding contributions from Hyderabad, half-way in South, and the little that appeared from Bangalore in South. The CCMB from Hyderabad contributes a figure comparable to the CDRI, and contributions from Hyderabad follows in close succession the highest contribution made by Lucknow. CDRI, followed by the ITRC both from Lucknow (sharing both North and East of India) top the list, followed as third the BARC from Mumbai in West, followed in close succession two organizations the Bose Institute and the IICB from Kolkata in East, and similarly. Lucknow alone, situated halfway north, contributes 31.2% of total of organized public research. Delhi and Chandigarh, in the north, contribute 15.7%; Mumbai and Pune in the west contribute 28.7%.
Hospital institution contributes most to biomedical research in India; at above 51% of country total, the distribution of research assets across hospital organizations and regions is, however, skewed. The specialty research-based hospitals contribute a very large proportion of research at above 32% of country total. AIIMS in Delhi in North tops the list, followed again from North the PGI at Chandigarh, followed closely as fourth the SGPGI from Lucknow, which again is rather close to North. From South a large contribution appears from Vellore based CMC, from West Mumbai based TMH too contributes a bit. However, from amongst these major research-intensive hospitals there are few from East, such as from Kolkata the contribution is less than 2% and from Kerala in South less than 1%. The ‘others’ segment is little above half of the country total, and to this segment belong the large number of hospitals, spatially well distributed (excepting hospitals from the North-East of India).
Assets distributed over geography, types of institutions, and organizations have similar features in India and China. A quick review of four other countries could be enlightening. Distribution in Israel shows all three types of institutions have intensively cultivated research. Hospitals are not too many in numbers; however, their share is substantial at little under 50%. Universities are few but each contributing fairly good amount of share to the country pool, often at about 5% each, such as Tel Aviv University, Ben Gurion University, and Hadassah University. The organized research at the Technion Institute and at the Weizmann Institute together while not contributing a substantially large share would appear large when one considers that much of university research in Israel could be classified under organized research. Geography of Israel is small compared to the two countries discussed above; however, much of the research is concentrated in a few select regions only. Geographic concentration is visible in Malaysia as well. Hospital institution’s research capability appears to be weak because its share in country total is much smaller than from the university system. Organized research possibly is the weakest in Malaysia. University institution dominates Malay contribution. University Sains Malaysia contributes 15.5%, University Kabangsaan another 18.8%, University Malaya another 22%, or the National University of Malaysia 3.5%, and similar others. Organizational dominance is clearly in evidence. A point of caution needs to be noted: much of Malay university contribution is indeed from hospitals.
University institution is strong in the UK as well. In fact over the years share of universities has been increasing. A quick review reveals that no single organization dominates, with the highest share from the King’s College at 8.4% and the subsequent shares such as from University of Oxford at 7.3%, Imperial College at 5.1%, University College London at 7.4%, and University of Cambridge at 7.7%, and the rest below this level – a group of about twenty-five universities share the output. Tiny contributions are rare and as it appears a contributing university has a minimal share. Those with no contribution although are plenty. Geographically, however, the share is more skewed. The domination is acutely reflected. The London-Cambridge axis has the clear dominance. Central England follows this; with the Scottish concentration around Edinburgh the rest of UK has little to offer. Hospital institution in UK too reveals a similar picture.
Before concluding our discussion on cross-country comparisons on asset distribution we must take a look at the corresponding distribution in the USA. China evidenced sharp coastal-geography based assets. The UK too evidenced three major geographies. Israel and Malaysia too exhibited geographic patches. India alone possibly remained most distributed. In USA, as in China, there are two strong geographies in the Northeast (NE) and in the Southwest (SE), although Texas evidences strong research assets. The NE is dominant. Most dominant cities are New York, Baltimore, Bethesda, Houston and San Francisco. The dominant geographic axis is Baltimore-New York-Boston; next in line is around San Francisco, followed by the patch around Houston. Recalling our previous discussion on the institutional dominance in USA, organized research dominates other institutional forms of assets. California with 16% contribution tops the list of hospitals, followed by hospital from Massachusetts at 10%, then from Texas at 10%, followed by from New York at 9%, then Maryland at 7% and such others. The ‘others’ category contributes only 18%. The distribution of university-based research in biomedical too exhibits similar geographic propensities. Overall three geographies in US have concentrated most of the research assets belonging to any of three institutions. Moreover concentration of assets in organizations is acute too. A likely result of such geographic and institutional agglomeration is intensification of spillovers not just across organizations but also more importantly across institutions.
The received literature has indicated possibility of spillover in geography across organizations. The received argument does not differentiate between different researches agenda that different institutions would have. As evidence to the cumulative dimension of spillovers, these arguments cited research outputs, in particular of patents. We could contrarily argue that hastened patenting does not indicate cumulative spillovers and in contrast indicates intensified competition across competing research agenda, set primarily by competing institutions. Similarly, hastened publications could indicate intensified competitions across institutions, and possibly negotiations between competing research agenda and paradigms on the relative desirability of the two, namely non-conforming exclusively institution-based agenda and the conforming mixed research agenda set up as collaborative ventures between competing institutions.
Concentrated geography while reducing transaction costs and hastening spillovers, does make it easier for institutions and respective affiliating organizations to compete with each other. Competing research agenda might in lieu of becoming cumulative be non-cumulative and dissimulative. Possibly more important is that close and intensified geography makes it easier to compete, and especially for an institution and its powerfully member organization to be in a position to effectively influence the research agenda, the research processes and the flow of finance as well as the generation and flow of human capital to the competing institution and its member organization. In short, geographic corporate governance appears feasible.
Geographic corporate governance busies itself with resources and agenda of things that are not within an organization, but are more involved with what remains outside the organization and the institution including resources that could be tapped from total country or the global locales. Geographic corporate governance is thus about influencing research agenda and similar others of competing institutions and its organizations. Intense negotiations across these institutions appear in the form of increasing research collaborations, joint authorship of research, joint patenting and the formation of patent pools. Such a pool of research projects, papers or patents serves twin purposes of (a) inter-institution/organization negotiation, and (b) deterrence to any new and novel proposal on departure in research or research assets. The pool from USA being quantitatively too large in comparison with the corresponding output from for example, India, can exercise greatly restraining power on any departure in agenda and also effective influences on the undertaking of research. No wonder that multinational R&D investments in both India and China are increasing. Geographic nearness and not the simple economic attractiveness of the manpower attract a multinational enterprise to this country location, and geographic nearness makes geographic corporate governance possible.
No less important for the governance function to operate smoothly are the following factors: the cost of monitoring; concentration of assets in both organization and in geography; credibility and increased acceptability of mechanisms to negotiate research agenda and inter-institutional negotiations; and, of course the credibility as well as increasing acceptability of the mechanisms of appropriation of research assets. Concentration of assets is reflected first in organizational dominance, institutional and geographic dominance. In some country cases we observed that all three types of dominance co-existed. In India while there are patches of dominance such as by the AIIMS in hospital institution, the broad patterns suggested that evenness in distribution was the principal characteristic. China exhibited less dominance than India, for example. Global governance would therefore be happy if these states of affairs in India and China changed towards increased concentration. For a country governor from India for example, it would be easier to monitor funding and increase output per funding if organizations were more concentrated. Possibly more importantly, the large and diffused research base such as in Indian hospital institution could prove a threat as well as act as deterrence to the dominant global paradigm of organized research. With increasing manipulation of research agenda in another institution such as in hospital or the university, the organized research would cherish enhanced organizational dominance, such as through opening more research university or research hospitals that could suck off most of the finances available.
Research from hospital or the clinical system has distinct difference with the research from the university system or from the research laboratories. There are several points of departures, and possibly the most important difference refers to the fact that (1) a physician even while performing a research must remain primarily a healer. This concern for the well being of the patient is the core ethical guideline that a physician must follow. University professor or a researcher may not be guided by such a code. This difference sets in a specific methodology of inquiry, specific goals of research and important constraints on research. Other differences between a physician and a professor would include; (2) physician is driven by practice while a professor would be driven more by theory; (3) research goals of a physician cannot often be set beforehand and the goals are in general emergent, while goals are pre-determined for a professor; (4) medical theories can be built when feasible from data, and often in lieu of theories there would be a plethora of guidelines, rules or standards and observation-cases would engender large multitude of variegated and irrefutable claims – however, non-medical research is least driven by data and facts and are directed to refuting or revisioning of existing theory-pieces; (5) much of data generation and of observation is dependent upon the skill of a physician and this is so even after healing practices been made dependent upon medical devices, and thus medical research presents a classical locus of serendipity, however, non-medical research has increasingly removed any element of serendipity.
Finally, hospital represents a system which unlike a business corporation (based upon fordistic assembly principle or post-fordistic methods of organization) or the department and discipline-based separations as in vogue in the university – generates and perpetuates a distinct mode of autonomous inquiry dependent much on serendipitous formulation of the ‘problem’ in the patient. Devices surely have partly robbed off such autonomy. However, much of the future of research or health-care pertaining to a patient depends upon this first formulation of problem. This research refers least to problem solving approach. Reputation plays great role in medical profession and in its research, and much of reputation builds upon this faculty to formulate a problem that can be solved.
There is another facet of difference. A physician is at the center of multiple relationships. The physician relates to the patient, to the medical profession and to the hospital or clinic, to the pharmaceutical company and to the payment system in practice in the country. Several parties to the relations can be ingratiated and the physician has often been accused of opportunistic behavior. A professor does not have many relations to be ingratiated, and scope for opportunistic behavior is extremely limited for this professor or researcher. This, however, contrasts to the professors’ zeal and opportunities in taking up the help of property rights (in USA, for example, under Bayh-Dole Act 1980, or in Europe the Copyrights Act or the Directive on legal protection of databases). Medical research till date is less affected directly by incentives to opportunistic behavior as framed through extant systems of property rights. Much of property rights in medical research are generated through pharmaceutical firm initiated clinical research and only occasionally through generation of data claiming privileges of property rights. These differences in economic incentives set up different patterns and different rights to access in these two systems of research – research coming out of hospital and the research coming out of university and research centers.
Fig 6: Publications percent in Medical & Hospitals of different countries ( Medline Database)
Fig 7: Publications percent in Universities of different countries (Medline Database)
Out of the total publications from the medical and hospital systems of the eight countries USA alone accounted for 67.8%, UK accounted for 17.6%, Israel for 4.9%, China for 4.7% and India for 3.9%. Other countries accounted for negligible share of the total. Similarly, out of the total publications from the university systems including from other research bodies, USA contributed 54.2%, UK for 24.9%, Israel for 6.9%, China for 6.6% and India for 4.3%. Other countries such as Malaysia or Chile contributed insignificant share. Remarkably for USA the relative contribution in biomedical research by universities is significantly lower than the contribution from the hospital system of USA. Contrarily, in UK the university system contributes significantly higher proportion of output. For both Israel and China university system’s contribution is growing and for India contribution from the university system is marginally above that from hospitals.
The university system, in some countries such as in India, takes up extra-ordinarily high teaching load and consequently research takes a back seat. In fact, in most countries including in India, the hospitals too being teaching in nature take up large teaching load. Contrarily, research system is far from teaching except in few instances of doctoral teaching. In fact private research system does not ordinarily offer supports to students below post-doctor.
Fig 8: Distribution of Publications over various Universities, Institutions, Hospitals and Medical centers of India (1991-2001)
In UK, hospitals contributed nearly twice the output from university or other research bodies. All the three components grew during 1991-2001, with modest growth output from hospitals grew 100% while with rapid growth outputs from university grew about 500% and from other research bodies by 300%. In China, growth has been spectacular. Output from universities grew more than 1500% while the output from research bodies declined, and from hospitals grew comparatively modestly by about 500%. India, however, experienced little growth segment-wise. In fact all the segments grew, with output from hospitals increasing by about 160%, from universities by about 200% and from other research bodies by about another 200%. However, because of the early lead that hospitals had, the 2001 contribution shared by hospitals still remained higher than from the university. US data shows as in Figure 8 a different picture. All the three segments grew about the same, namely by nearly 160% during 1995-2001 period. However, the other research bodies had the early lead with their output above the hospitals both during 1995 and 2001. Share contributed by universities in biomedical research remained comparatively low during this period. In fact the dominance in US research output by other research bodies could be gauged by the fact that what these bodies produced in 1995 in USA was above the total production of all three segments of UK during 2001.
Cross-country Comparison in capability
Role of hospital in India and China: Hospital provides for the scope to undertake integrated research, and research on a very large number of patients with all possible types of diseases. Both India and China have very large population and this alone provides for a very large research pool. However, the genetic pool in India is much diverse and India offers a virtual microcosm of global genetic abundance. More importantly in both the countries hospital is the most important provider of health care. Chinese healthcare is more organized around hospital system than Indian healthcare, which is distributed over several small clinics, general physicians as private practitioner, specialized centers and private hospitals. Hospital system of India has contributed at a marginally higher level than its counterpart in China. However, the definition of hospital varies, and rather often data indicated by the address shows a hospital as university, or as research organization.Organized research under publicly funded system has contributed significantly to Indian total. Figure 9 for example, shows a detailed break up of contribution by different types of institutions. Universities contributed at 37.9%, organized public research contributed 15.6% and higher research-teaching establishments such as the Indian Institutes of Technology (IIT) and the Indian Institute of Science (IISc) contributed 9%. Private research and private medical research contributed 0.3% and 1.7% respectively while pure hospital-based research contributed 32.5%. This contribution by the organized publicly funded research could be further sub-divided into a component contributed by a few major organizations. Figure 10 exhibits a break-up. Bhabha Atomic Research Centre (BARC) contributed 1.2%, Central Drug Research Institute (CDRI) 1.7%, Institute of Microbial Technology (IMTECH) 0.5%, National Institute of Immunology (NII) 1%, Bose Institute 1%, Indian Institute of Chemical Biology(IICB) 1%, Institute of Virology 0.3%, and similar others.
Fig 9: Publications percentage in different sectors in India
However, rather often a publication of the CDRI or of IICB or BARC would be jointly accomplished with partners from local hospitals such as the St. John’s, the Tata Memorial, the Safdarjung Hospital, and several others. Thus the output of organized research and of the hospitals including from the universities (especially in large affiliating universities, such as the Banaras Hindu University(BHU), the Calcutta University, Madras University) are often joint. Apart from making the data regarding origination of research a suspect, the major role of such joint research is in influencing the research. Role of organized research system is super-dominant in the USA. This sector alone contributed more than 40% of total output. In China much of research is organized around the universities. In USA the university system contributed about one quarter of all publications. In India university system (when medical universities are taken out) contributes far less, and even the non-medical organized research pursuing such problems as genomics, proteomics, drug design, and other chemical-biological problems in biomedical research contributes far less compared to both USA, and UK.
Biomedical knowledge assets in Indian cities
Medical practice and medical case-based knowledge assets exhibit excellently the characteristic of severalty. In case accumulation of quantity of knowledge asset is possible severalty of assets would be undervalued while large quantity amassed or the efficiency of operation would be valued. Possibly a good amount of non-serendipitous research knowledge in areas such as chemistry or biotechnology belong to these classes.
In the type of assets where severalty is valued, the spatial distribution and spatially unique characteristic of for example medical knowledge receives high value. This space could refer to physical geographical space, the genetic space, or the temporally observed series of facts on a group of subjects belonging to a single locality. Contrarily, knowledge of organic chemistry for example, would have the least reference to similar spatial aspects. It follows then distribution characteristics of assets within institutions of knowledge are important. Spatial distribution of medical knowledge generation should be considered highly desirable. Spatial concentration of chemical or biotechnological knowledge could in contrast be desired, especially because both spillovers would be higher and the transactions costs lower in concentrated space. A related aspect would be concentration of most such knowledge for example in chemical, inside an organization, and a market in such a case would be constituted by transactions between organizations. In the other type of medical knowledge assets would better remain dispersed, often even through physicians’ networks. Market in this latter case could be least through inter-organizational transactions.
Spatial distribution of medical knowledge assets is thus an important value feature. The following Figure presents the distribution of cumulative biomedical publications over eleven years across several cities, and a detailed annual break up of publications over thirty-five smaller cities from India exhibits interesting pattern. Publication is an asset in so far as the institution can appropriate it. Publishing personnel are otherwise the assets, and publication as a dummy can represent distribution of active asset-holding personnel. Moreover, the person publishing a research result acquires a reputation, which acts as valued asset. A paper as such cannot be transacted for commercial benefits although. The institution of hospital contributes the larger part of the output from the large cities such as Delhi, Bangalore, Kolkata, Mumbai, Lucknow or Hyderabad and Madras. In fact, considering the fact that all organized research and most of universities are located in large cities, the contribution from non-major cities for example Patna, Mangalore, Coimbatore, Guahati and similar others are largely due to hospital. Some of the non-major cities have university and as a result part of the output is from the university system. However, in most cases we observed collaborations between the physicians and the university personnel. Organized research is simply absent in all such distributed small locations.
Fig 10: Publications percent in Cities of India (Medline Database 1991-2001)
Cumulative publications as assets have small values in most such locations. Importantly, annual publications for even smaller cities often have zero values. In major cities the cumulative values for a handful of organizations are very high in contrast. The small city hospital it appears fails in governing resources. Organizational corporate governance enables mobilization of resources for the goal of the organization. In the language of corporate governance, the following three aspects: strategic control, organizational integration and financial commitment, determine the capability to govern resources for innovation and sustenance. These issues appear to be crucial for private for profit organizations, and are unlikely to prove important in public organizations. A public organization has multiple and sometimes conflicting goals.
Human resources owe their allegiance more to the profession than to the organization, and in particular for public hospitals the physicians are employed by the hospital system and not by a hospital, resulting into routine transfers of physicians across several hospitals within a state. In hospitals from small cities organizational corporate governance is missing. A physician would conduct research only if institutional norms prevailed over and only if some formal or informal network operated. The private gain to the physician has not been an organizational gain to the small city hospital. Given the fact that a physician would get transferred across several public hospitals within a state, the nature of distribution of asset-holders would largely remain independent of the organizational governance of a small city hospital. This, however, would not hold true of large city based reputed hospitals and especially of the specialty research hospitals, such as the AIIMS. Therefore, distribution over small cities is more important than identifying specific organization across small cities. A corollary to this should be: governance of institution and the governance of asset generation through formal and informal or normative network is important for the medical component of research.
The institution of hospital dominates biomedical research from non-major cities of India. Two hospitals from Agra and none other belonging to other institutions; four hospitals, one university and one research organization from Ahmedabad; similarly in the proportions of 1:0:0 from Ajmer, 1:1:0 from Aligarh, 1:1:1 from Berhampur, 1:1:0 from Bhopal, 0:2:0 from Coimbatore, 2:0:1 from Cuttack, 1:0:0 from Guntur, 7:0:0 from Jaipur, 4:0:0 from Jodhpur, 8:2:2 from Pune, 2:1:0 from Ranchi, 4:0:0 from Rourkela, 2:1;) from Surat, 1:2:0 from Tirupati, and 2:1:0 from Wardha – where the order in the proportion is hospital, university and organized research respectively. In none of the cities the total contribution is large, and in none any single organization continued research in sustained manner to accumulate sizeable quantity of assets. Overall, however, in all places including where the university has contributed, institution of hospital dominates either by way of being the only provider of research or else by being the dominant source of research.
Asset distribution in major cities
Corporate governance issue would, however, remain important for organizations belonging to organized research institution in major cities. University institution has multiple goals in general, excepting a few privileged organizations such as the IISc in Bangalore whose charters enable them to undertake primarily research function. Hospitals in major cities do not enjoy any special privilege vis-à-vis hospitals in smaller cities excepting receipt of larger financial allocations. In research organizations, goals are limited and manpower remains tied with the organization. University system does not transfer academicians across several universities; however, teaching function draws away most resources. More importantly, strategic control of researchers and academicians, and organizational integration of disparate academic disciplines in separated departments prove near impossible resulting in the collapse of governance within and by the organization. A research organization by definition is dedicated to single function. However, there are multiple nuances in the meanings of the term ‘dedicated research’ under public system. All the types of organizations exhibit weak governance to various degrees, and the weaknesses in organizational corporate governance have not always been substituted with strong institution-based corporate governance.
A city, however, differs from other geographies. Small cities often would be having only one or two organizations at the most. A major city has in contrast more than twenty-five such organizations. Major cities enjoy other infrastructures, huge pool of relevant manpower and above all much higher inflow of funds. Such a city can possibly enjoy spillovers and increasing return limited to the geography. Contrarily, small cities are unlikely to enjoy similar spillovers. Spillovers and increasing return, if any, could be enjoyed by organizations in dispersed geographies possibly through the institutional mode alone. Major cities, however, have different proportions of organizations belonging to different institutions. The following series of Figures exhibit specific endowments of the cities Bangalore, Mumbai, Hyderabad, Kolkata, Lucknow, Chennai, and Delhi respectively. Bangalore is dominated by the presence of the IISc, which supported by a few other research-driven teaching systems for example JNCASR, and including the Bangalore University contribute ore than sixty percent of the total publication output from the city. Hospitals in Bangalore, when compared to hospitals from Mumbai have little voice and presence. St. John’s hospital contributes 6.9% and all hospitals put together contribute less than 15%. No less important is the low number of hospitals participating in research.
Fig 11: Publications percent in Institutes of Bangalore (1991-2001)
Fig 12: Publications percent in Institutes of Mumbai (1991-2001)
Fig 13: Publications percent in Institutes of Hyderabad (1991-2001)
Mumbai offers sharply contrasting features. Three remarkable characteristics are: (1) large number of hospitals participating in research process, (2) total contribution by hospitals is dominant, and (3) non-dominance of organized research. More than thirty-two hospitals out of forty non-major contributors from Mumbai (such as the Tata Memorial hospital, the Edward, the Nair, and similar others with contributions above 2% of the city total) have participated in research during the last decade. University has low participation, and organized research from the BARC, the TIFR and others put together contribute little above 25%. No single organization in Mumbai has contributed more than 15% of total city publication, while IISc in Bangalore alone has contributed about 49%.
Fig 14: Publications percent in Institutes of Chennai (1991-2001)
Hyderabad and Kolkata exhibit features intermediate between Mumbai and Bangalore. Unlike the dominance by the IISC, no single organization from Hyderabad and Kolkata contributes above 29% and 14% respectively of total city publications. University in Hyderabad contributes close to 30%, organized research from public and private about 40%, and research-intensive hospitals about 17%. Similar figures for Kolkata are more distributed: university close to 15%, organized research from public about 34%, and large hospitals about 18%. Less research-intensive hospitals are few in both the cities. There are eight and seven such non-major hospitals in both the cities out of a total of fifteen non-major participants in biomedical research. Distributions in both these cities are not sufficiently long-tailed.
Chennai has, however, a long tail, quite close to what we observe for Mumbai, and this distribution is very different from what we see at Lucknow. Research in Lucknow has been parceled between a handful of organizations most of whom are from public organized research, which contributes about 52% of total city publication. Lucknow University has very low participation in biomedical research. Medical research, too, is concentrated in the specialty research-based hospitals, the Sanjay Gandhi and the Kasturba Gandhi, the former contributing above 31% and the latter 10.6%. In the tail end at Lucknow there are only two minor hospitals out of a total of five minor organizations. Chennai, however, has unique features. Its tail end is long though smaller than Mumbai. It has 19 minor hospitals participating in hospitals out of a total of 24 minor organizations. The university has fairly large research base, although most of it coming from hospitals under the university, a feature very similar to Kolkata. Organized research is relatively weak with about 13% contribution to the city total. Hospitals are many, and none dominating the research, a feature, which even Mumbai cannot display. Most research-intensive hospital contributes about 5% of city total, and there are four of them. Large hospitals contribute about 30% of city total of publications.
Fig 15: Publications percent in Institutes of Kolkata (1991-2001)
Fig 16: Publications percent in Institutes of Lucknow (1991-2001)
Delhi has some other kind of uniqueness. Its tail end is as long as Mumbai, however, its distribution is more skewed than Mumbai. The AIIMS, a specialty research hospital, alone contributes the dominant share of 33.6%, followed by the next most-contributing hospital the Maulana Azad with only 3.1%, followed further by the third most-contributing hospital the G. B. Pant at only 1%. There are twenty-nine minor participants from the medical area out of a total of forty-four minor organizations. University system in Delhi contributes a little above 12%, with nearly the total output shared by the two major universities of the city, the Delhi University and the Jawaharlal Nehru University. Organized research contributes only about 7% of city total.
Fig 17: Publications percent in Institutes of Delhi (1991-2001)
The ‘others’ category of each city provides an insight. Delhi has the highest figure, at 40.7% of city total, followed by Kolkata at 35.5% of city total, followed by Chennai at 26.7%, then Mumbai, Bangalore and Hyderabad at par, and Lucknow with the lowest figure of 4.4% of city total being contributed by the ‘others’. Any organization contributing lower than 0.3% in general could not be exhibited in these figures and were clubbed under this ‘others’ category. This category can be compared with the other extreme of a distribution, the dominance. Bangalore is most dominated, followed by Delhi, which is followed by Hyderabad seconded by Lucknow. Kolkata and Mumbai do not have any significant peak. If we look at this pair, Dominance-Others, Delhi is the strangest with two peaks at two ends.
However, the length of distributions too is important. ‘Others’ represents a length such that no participant enjoys any significance in research. Another dimension of length is represented by organizations that feature in the figures exhibited above, contributing in general above 1% of city total. Mumbai has the longest chain in this dimension, followed by Chennai, followed by Delhi and Kolkata, then Bangalore, and then Hyderabad. The least long chain can be seen in Lucknow. The number of non-major organizations is also indicative: Delhi and Mumbai both at 44, Chennai at 24, Bangalore at 21, Kolkata and Hyderabad at 15, and finally Lucknow at 5.
Another way of characterizing these cities would be along the dominance of institution. Bangalore is dominated by organized research, undertaken mostly at research university of the IISc. Hence Bangalore is doubly dominated, first by institution then by an organization. Hospital institution dominates Mumbai. However, no single organization dominates the medical research there. Hospital institution dominates Delhi too, however, with one significant departure from Mumbai that in Delhi we experience organizational domination as well. Chennai is equally shared between institutions, although when one takes into account medical research under university as belonging to hospital institution, the latter as institution dominates. This being similar to Mumbai, differences between these two cities persist. General clinical research is stronger at Mumbai with cancer, for example receiving higher attention. In Chennai, however, general clinical research has low presence. This city has specialized along a few disease areas around diabetes and vision, for example.
The institution of organized research dominates Kolkata. In fact, most part of the university contribution comes from the hospital system; and the rest of the university research is close in terms of research paradigm employed with the organized research. Hyderabad too shares similarity with Kolkata, but with one difference that university institution is strong at Hyderabad. Two institutions, and two organizations, one each from each institution dominate Lucknow, the CDRI from public organized research and the Sanjay Gandhi from hospital institution.
These cities are distributed along the entire geography of India. Mumbai is from West, Chennai and Bangalore from South, Hyderabad from Central, Delhi and Lucknow from North and Kolkata from the East keep the geographical balance. Each geography offers distinct knowledge or asset pool. This, however, is not status of some other countries. China shows a distinct geographical concentration. Research assets in China appear to be highly concentrated in the coastal ring, which contributes above 81% of China total. The adjacent areas, just behind the coastal, contributes about 7%, the next hinterland ring contributes about 5%, and the rest about 8% by the rest part of China. A similar result to be discussed later can be observed in London-Cambridge output cornering about 50% of national output, and for the USA, there is close competition between the two poles of biomedical research - the North-Eastern pole competes with the South-Western pole, and who together produces the majority of US research.
References:
- P.Banerjee. 2007. Biomedical Innovation in India. New Delhi: Har-Anand
Go back to Home Page of 'India, Science & Technology:2008'