Code Number: 055-127-E
Division Number: VII
Professional Group: Library Theory and Research
Joint Meeting with: -
Meeting Number: 127
Simultaneous Interpretation:   No  

The emergence of alternative medicine as a new field: a bibliometric study of a rapidly - growing field

Abstract

An on-line search, not limited to English-language publications, conducted on eighteen databases specializing in medical and scientific fields, retrieved close to 4000 publications, mostly articles, of which (80%) appeared between 1985 and 1994. This demonstrates the rapid growth which occurred in the number of publications dealing with alternative medicine throughout the twenty years period checked. This growth was accompanied by a parallel increase in the number of different keywords used in the process of indexing these publications. Most of these publications appeared in two countries : the US and the UK, whose combined share has increased from 44% in 1975-1979 to 70% in 1989-1994, with a recent considerable shift in favor of the US. About 87% were published in only eight 'top' countries out of the fifty found in the study. Caution is needed, however, when drawing conclusions concerning the accurate situation world-wide, since findings are probably strongly biased in favor of the western countries and the English language, two biases which prevail in the eighteen databases searched.

Paper

INTRODUCTION

OBJECTIVES

METHODOLOGY

LITERATURE REVIEW

The "Law of Exponential Growth" has been further dealt by Tague and others (1981), Wolfram and Xin lu (1990), Ravichandra Rao and Meera (1992), Egghe and Ravichandra Rao (1992) and many others. In most articles it is argued that the literature grows exponentially and it is described mathematically by the exponential function Yt =a.ebt where Yt represents the size at time t, a is the initial size, and b is the continuous growth rate which is related to the annual percentage growth rate r, as: r =100(eb-1). Egghe and Ravichandra Rao (1992) claim, however, that the power model (with exponent >1) is the best growth model for sciences and technology fields, while the Gompertz S-shaped distribution fits better databases of the social sciences and the humanities.

FINDINGS

Although the 18 databases searched cover other types of publications (like books, dissertations, conference proceedings, etc.) the vast majority of the publications retrieved in this study were articles in various kinds of journals and serials. Evidently, as shown in Table 1, a rapid growth occurred in the number of publications referring to alternative medicine, when summed up in two 10-year groups : their number increased almost five fold, from only 638 in the first decade (1975-1984) to 3096 in the next decade (1985-1994) an increase of close to 400%.

TABLE 1
Number of Publications Found in Databases According to Year of Publication

Year	No. of Publications	Change in %	Year	No. of Publications	Change in %
1975	31	---	1985	139	49.5
1976	19	-38.7	1986	213	53.2
1977	13	-31.5	1987	126	-40.8
1978	40	207.7	1988	183	45.2
1979	92	130.0	1989	199	8.7
1980	87	-5.4	1990	426	114.1
1981	90	3.4	1991	259	-39.2
1982	75	-16.7	1992	329	27.0
1983	98	30.7	1993	516	56.8
1984	93	-5.1	1994	706	36.8
Total	638		Total	3096

To be sure, comparing each year to the preceding one, clearly shows that the annual rate of growth fluctuated and was neither identical nor linear.

On the average, the percentage (r) by which the size of the literature increased each year was 30.9% , and the formula r = 100 (eb - 1) yields b = 0.269 (doubling time d = 2.6 years). But, substituting this b figure in the commonly used exponential growth function Yt = a ebt should have yield 5140 publications in the 19th year (1994), which obviously is not the case here.

However, calculating r and b in a different way, by substituting Yt in the exponential growth function with Y19 = 706 (number of publications in 1994) yields : b = 0.16 , r = 17.35% and d = 4.33 years, which are much more reasonable.

As already mentioned, Price (1963, 1975) found a growth rate for scientific literature of approximately 5% over the past two centuries, corresponding to a doubling time of 15 years. Ravichandra Rao and Meera (1991) found a growth rate of 6.6% for mathematics, and growth rates of 12% to 16.5% for eight scientific databases for 1968 -1987. Compared to these figures, and especially to the 15% growth rate found for MEDLINE, the rate found here for the field of alternative medicine is rather impressive. A comparison to former growth studies in other fields reveals that the field of alternative medicine is still on the steep part of the growth curve, not yet reaching the saturation stage.

Grouping the twenty year period checked (1975-1994) into four five-year periods (shown in Table 2) the rapid growth can be seen from another angle : the number of publications rose from 195 in the first period (1975-1979) to 443 in the second one (1980-1984), then to 860 in 1985-1989, jumping to 2236 in the last period (1990-1994).

TABLE 2
Number of Publications and Rates of Increase According to Five-year Periods

Period	No. of Publications	% of Total	Increase in %(*)	"Bradford's Multiplier"
1975-1979	195	5.2	-	-
1980-1984	443	11.9	127%	2.27
1985-1989	860	23.0	94%	1.94
1990-1994	2236	59.9	160%	2.60
Total	3734	100.0%	-	-

* When compared to preceding figure (not cumulative)

Comparison between periods shows that 60% of the literature in this subject area was published during the last period checked (1990-1994) and about one-fourth during the preceding five years (1985-1989). Altogether, 83% of the literature was published during the second decade (1985-1994), while only 17% during the first decade checked (1975-1984).

Comparing each five-year period to the preceding one, one may see that the growth rate was almost 100% or even higher, or if the four periods are looked at as analogous to the classic Bradfordian Zones, the Multiplier fluctuates between 1.94 and 2.60, averaging 2.27. The relatively high rate in the second period (127%) characteristic to many areas in their beginning stage, is not surprising, since the basic figure (195 publications) was relatively low, even a small addition of 248 publications creates an increase of over 125%. Although 417 publications were added during the third period (1985-1989) the increase rate is lower, due to the higher base figure (443). The striking finding is the relatively high increase rate of 160% during the last period (1990-1994) expressing the new 2236 publications added throughout these five years.

It may be argued that the relatively high increase rates may have been affected not only by real growth in the number of publications, but also by changes in coverage policy by some of the 18 databases searched, done throughout the twenty-year period. A decision to widen the coverage and to include certain journals and/or topics uncovered yet, may of course, contribute to growth rates higher than those that existed in actuality. At any rate, even if the 'real' growth rates were in fact slightly lower, the abovementioned argument still supports our underlying assumption that the field of alternative medicine has undergone a significant transformation during the twenty years checked, emerging into a widely-recognized field, and undergoing a process of growing consolidation with the conventional medical-scientific establishment.

2. Number of Keywords

A quantitative analysis showed that the number of different keywords in article titles grew from 36 in the first period (1975-1979) to 51 in the last one (1990-1994) while the parallel increase in the number of different descriptors was significantly higher, from 30 to 103. Altogether, the total number of different keywords, whether appearing in titles or in descriptors, was about 350.

However, after eliminating synonyms and closely-related words, it appeared that the 'net' increase was more than double, from about 43 different keywords and descriptors in the first period to about 113 in the last one, meaning an addition of about 70 new keywords during fifteen years.

3. Place of Publication

Table 3 gives the frequency distributions and ranking of place of publication of all items retrieved in the study, in descending order of the figures of the first period, for each of the four periods checked. Clearly, The US and the UK are significantly higher than all other countries throughout all four periods. Their share grew steadily from about 44% in the first period (1975-1979) to more than 53% in the second, dropping to 50% in the third and rising considerably to 70% in the last one (1990-1994), being 62% for the whole period of twenty years. A significant shift occurred, however, in the proportions of the two leading countries: while in the first period both had an equal number of publications, with UK's share even slightly higher, the opposite is true for the other three periods, especially for the last one, in which the ratio has drastically changed and the percentage of publications published in the US (46.3%) was double that of the UK (23.6%).

For most other countries, generally speaking, there have been decreases, mainly in the last period : Germany, Switzerland, Italy, Scandinavia, Japan, India, China. The proportion for France was on the rise during the first three periods, but dropped in the last one, while for Holland the opposite is true. It is also noteworthy that France, Germany and Switzerland maintained their places, with slight variations, among the seven leading countries during all four periods checked.

Apparently, one would expect that much higher proportions of the literature in the area of alternative medicine to be published in countries like : China, India, and Japan. Many publications probably appear in China each year in this area, but they are either not in English, not covered by the various databases, or not designated as alternative medicine.

TABLE 3
Place of Publication of Publications Dealing with alternative medicine
(In descending order of the first period, 1975-1979)

Period	1975-1979			1980-1984			1985-1989			1990-1994			Total
Country	No	%	Rk	No	%	Rk	No	%	Rk	No	%	Rk	No	%
UK	43	22.1	1	94	21.2	2	200	23.3	2	527	23.6	2	864	23.1
USA	42	21.5	2	142	32.1	1	228	26.5	1	1036	46.3	1	1448	38.8
Germany	31	15.9	3	54	12.2	3	59	6.7	5	149	6.7	3	293	7.8
Switzerland	15	7.7	4	50	11.3	4	66	7.7	4	64	2.9	5	195	5.2
Holland	11	5.6	5	16	3.6	6	23	2.7	8	73	3.3	4	123	3.3
Italy	6	3.1	6	3	0.7	12.5	18	2.1	9	11	0.5	16	38	1.0
France	5	2.6	8	22	5.0	5	68	7.9	3	48	2.1	7	143	3.8
Norway	5	2.6	8	2	0.4	16.5	10	1.2	13	31	1.4	10	48	1.3
Sweden	5	2.6	8	5	1.1	10	16	1.9	11	17	0.8	13	43	1.2
Denmark	3	1.5	11.5	1	0.2	19.5	48	5.6	6	33	1.5	9	85	2.3
Austria	3	1.5	11.5	3	0.7	12.5	8	0.9	15.5	15	0.7	14	29	0.8
Japan	3	1.5	11.5	3	0.7	12.5	3	0.3	17.5	10	0.4	17	19	0.5
India	3	1.5	11.5	2	0.4	16.5	2	0.2	19.5	7	0.3	18	14	0.4
Australia	2	1.0	14.5	7	1.6	7	16	1.9	11	34	1.5	8	59	1.6
Spain	2	1.0	14.5	2	0.4	16.5	3	0.3	17.5	2	0.1	20	9	0.2
Belgium	1	0.5	18	3	0.7	12.5	2	0.2	19.5	12	0.5	15	18	0.5
Canada	1	0.5	18	6	1.4	8.5	24	2.8	7	63	2.8	6	94	2.5
S. Africa	1	0.5	18	6	1.4	8.5	8	0.9	15.5	25	1.1	11	40	1.1
N. Zealand	1	0.5	18	1	0.2	19.5	16	1.9	11	20	0.9	12	38	1.0
China	1	0.5	18	2	0.4	16.5	9	1.0	14	3	0.1	19	15	0.4
Other	11	5.6	---	19	4.3	---	33	3.8	---	56	2.5	---	119	3.2
Total	195	100%	---	443	100%	---	860	100%	---	2236	100%	---	3734	100%

Similarly, it is very likely to assume that many more than the 13 items found in this study, were published in the former Soviet Union throughout this period of twenty years, but they were not included in the 18 databases searched. It is likely to explain the very low proportions found here by the strong western and English language biases of the 18 databases searched.

TABLE 4
Place of Publication of Publications Dealing with alternative medicine
(for the Whole Period : 1975-1994)

Considering the proportions published in other English-speaking countries (Canada, Australia, etc.) one may estimate that at least 68% of all publications found in this study are in English. Admittedly, this high proportion does not reflect the real situation in the field and is probably affected by the abovementioned English-language bias of the 18 'source' databases, from which the sample was drawn. This inherent bias probably explains to some extent the negligible number of items in the sample originating in non-western countries.

However, since the bibliographical databases has become the main tool for scientists for literature update, it seems that the English language has become the 'lingua franca' of the field , as has happened in many other fields during recent decades.

CONCLUSIONS

1. There has been a rapid growth in the number of publications dealing with alternative medicine throughout the period of twenty years checked : 1975-1994. This growth was accompanied by a parallel increase in the number of different keywords used in the process of indexing these publications.
2. Most of these publications were published in two countries, the US and the UK, whose combined share has increased from 44% in 1975-1979 to 70% in 1989-1994. However, the division between the two countries has been drastically changed from equal proportions in 1975-1979 to the US share doubling during 1989-1994. Along with six other countries, 87% of all publications appeared in only eight 'top' countries out of the fifty found in the study.
3. Our findings are probably strongly biased in favor of the western countries and the English language, since these biases prevail in the eighteen databases searched in the current study. Therefore, one should be very cautious when drawing conclusions concerning the accurate situation world-wide.

BIBLIOGRAPHY

1. Egghe, L and Ravichandra Rao, IK (1992) "Classification of Growth models based on growth rates and its applications", Scientometrics 25 (1), pp. 5-46.
2. Goffman, W (1966) "Mathematical approach to the spread of scientific ideas", Nature 212, pp. 449-452.
3. Goffman, W (1971) "Mathematical model for analyzing the growth of a scientific discipline", Journal of the Association of Computing Machinery 18 (2), pp 173-185.
4. Humphreys, BL and McCutcheon, DE (1994) "Growth patterns in the National Library of Medicine's serials collection and its Index Medicus journals, 1966-1985", Bulletin of the Medical Library Association 82 (1), pp. 18-24.
5. Orr, RH and Leeds, AA (1964) "Biomedical literature : volume, growth and other characteristics", Federation Proceedings 23 (6), pp. 1310-1331.
6. Price, DJ deSolla (1975). Science Since Babylon. Enlarged edition. New Haven, Yale University Press, pp. 165-186.
7. Price, DJ deSolla (1963) Little Science, Big Science. NY, Columbia University Press.
8. Ravichandra Rao, IK and Meera, BM (1991) "Growth and obsolescence of literature : an empirical study", INFORMETRICS '91 : Selected Papers from the Third International Conference on Informetrics, 9-12 August 1991, Bangalore (ed. IK Ravichandra Rao) Bangalore, 1992, pp. 377-394.
9. Tague, J a.o. (1981) "The law of exponential growth: evidence, implications and forecasts", Library Trends 30 (1), pp. 125-145.

Appendix

1. AIDSLINE (157)
2. AMA Online (442)
3. BIOSIS PREVIEWS (5)
4. BNA Daily News from Washington (655)
5. CANCERLIT (159)
6. DIGENES (158)
7. Dissertation Abstracts
8. EMBASE (73)
9. Health Periodical Database (149)
10. Health Planning and Administration (151)
11. INSPEC (2)
12. International Pharmaceutical Abstracts (74)
13. MEDLINE (155)
14. PASCAL (144)
15. PMS (457)
16. SciSearch (434)
17. SPORT (48)
18. TOXLINE (156)