Access to all articles, new health classes, discounts in our store, and more!
The Nature/Nurture Debate: An Update
Published in International Journal of Alternative and Complementary Medicine, Vol. 12, No. 8, pp. 29-31, August 1994.
* * *
Introduction
Four items serve as an excellent prelude to and justification for this report on the role of genetics versus the environment in the genesis of health and sickness. Firstly, it is abundantly clear that the controversy exists. Two major medical mouthpieces, Nature1–4 and the New England Journal of Medicine5-6 continue to bring into focus the subject by means of important reports, editorials and letters. Secondly, the prevailing conclusion is that most major chronic diseases probably result from the accumulation of environmental factors over time in genetically susceptible persons. We are still currently unclear regarding the relative contributions of the environment and inheritance. It is essential to know the nature/nurture ratio. Obviously, the quotient would dictate the direction of subsequent research efforts and clinical pursuits. Thirdly, it is interesting that the present conclusions are based on conventional (and cumbersome) familial models including parents and children, twins, siblings, the orphaned and adopted. Finally, the least studied matrix, with the greatest possible contribution, is spouse-likeness. Clearly, married couples are rarely kin and obviously environmentally very close.
This report is intended to restudy the nature/nurture argument7 through an examination of the simplest and least expensive familial prototype, the husband and wife. Interest in this particular model is further heightened by the recent observations8 in undergraduate psychology students provided with separate male and female photographs of couples, freshly married and of long-standing. It is noteworthy that these inexperienced students, almost without exception, could not match newly married couples.
In contrast, they readily identified wedded couples of longstanding. This simple experiment demonstrates the old adage that spouses (and even their pets) begin to look alike with time. It also sharpens the role of the husband/wife relationship in the gene/environment argument.
Additionally, this is another in our series of essays on Medical Ignorance: Myths and Magics in Modern Medicine.
What We Know Today
Over the past three decades there has been an interest in the overall problem of spouse-likeness. Laced throughout the scientific literature for the past 50 years, with two notable exceptions, are approximately 50 reports generally confirming similarities in diverse patterns of married couples. For example, Winkelstein and Sackett9-12 as well as others13-23 have looked into the problem of family aggregation as it relates to hypertension. There has also been considerable interest by a number of investigators in carcinomatosis.24-31
The first of the two exceptions, Garn and his colleagues32-49 have provided a series of consistent, structured and productive contributions extending over a period of approximately 20 years.
They have examined weight and weight change, urinary and blood vitamin studies (vitamins A and C, riboflavin, thiamine) serum cholesterol and triglycerides, dental caries, bone haemoglobin, hematocrit and diet (calories, calcium, protein, alcohol). It is safe to conclude that all of these parameters correlate significantly in husband/wife groups. In other words, with advancing time (length of cohabitation) these parameters become increasingly similar.
We, the other exception, here at the University of Alabama Medical Center, have also been studying familial aggregation in a group of dental practitioners and their spouses. We have published approximately 20 papers over a period of 12 years.50-69
Experimental Design
The model we have employed for spousal similarities is shown in Table 1. Two hundred and sixty one couples were studied in terms of their serum cholesterol (line 1). These same dental practitioners’ scores were compared to age and sex matched unrelated (line 2). Finally, the two female groups were compared (line 3). This format provided us with the opportunity of raising (and hopefully answering) three questions.
- What is the relationship of serum cholesterol in married couples?
- How does the husband-wife correlation compare with the pattern in the husband and an age-paired unrelated female?
- Is the connection a function of time?
Line 1 shows a highly statistically significant correlation of serum cholesterol (r = +0.361, p<0.01). Hence, in answer to the first question, the evidence suggests that husbands and wives seem to demonstrate similar serum cholesterol levels.
In contrast, an examination of the men compared with the women age-paired against their wives (line 2) showed no such significant relationship (r = 0.075, p<0.05). Therefore, with regard to the second query, there is no convincing correlation coefficient between serum cholesterol concentrations in men and women unrelated by marriage. This is heightened by the fact that the two women groups do show a remarkable similarity (r = 0.189, p<0.01).
Finally, the question arises as to whether couples, consciously or otherwise, select each other on the basis of serum cholesterol. Lines 4, 5 and 6 show the correlation coefficients of the married groups in terms of advancing age. Line 4 pictures the men less than 40 years of age and their spouses. In this, the relatively youngest combination, there is so statistically significant correlation (r = +0.176, p<0.05).
In contrast, in the next age group (the men in the 40s), the relationship sharpens and becomes significant (r = +0.464, p<.01).
In answer to the third and final question, this clearly suggests that the combination is predominantly environmentally inspired.
The cholesterol model has been utilised in the study of all of the parameters listed in Table 2. In the interest of expedition and clarification, it should be pointed out that, in several instances, a parameter was studied initially (line 2) in a small sample and subsequently re-examined (line 1) in a larger group. The initial report is identified as the preliminary document; the re-examination as the final report. This obtains in the case of SGOT (lines 6 and 13) and refined carbohydrates (lines 4 and 7).
Four points are worthy of emphasis. First, there is a considerable spread of correlation coefficients from a high of +0.896 (line D) to a low of +.185 (line 2).Secondly, in all but one case (line 19), SGOT, the relationships are statistically impressive.At the present time, there is no explanation for the uniqueness of SGOT except that it might well be a problem in sample size. This is borne out by the lack of significance in the young group (Table 3, line 20) and the statistical consequence in the older group (Table 4, line 5). Thirdly, the most significant correlations (and some are unbelievably high) occur in the enzyme groups (Tables 3 and 4, lines 1-3). Fourth, it is noteworthy that the results of the preliminary and final experiments are surprisingly constant. Lastly, it should be underlined that these findings were not viewed in the light of time of cohabitation.
Tables 3 and 4 are designed to emphasize the importance of time of marriage. Table 3 summarises the correlation coefficients for all of the studied parameters in the relatively younger groups. Three points become evident. First, there is a considerable spread of correlation coefficients from a high of +0.948 (Line 1) for LDH to a low of -0.259 for blood glucose (line 21). Secondly, the majority of correlations are significant. This last point raises the interesting possibility that it might well be that partners do indeed tend to select themselves on the basis of some or all of these characteristics. On the other hand, it may be that couples begin to assume similar patterns early in marriage.
In contrast, Table 4 summarises the characteristics of relatively older married couples. We note, firstly, that there are many more significant correlations in older than younger couples. In point of fact, in this group there are all but two that show a statistical meaningfulness (lines 19 and 21). Secondly, in the main, the correlations sharpen as the groups get older. In other words, in more cases than not, the r is higher in the older than the younger. Thus, the overall evidence suggests that married couples, more often than not, become more similar with time of cohabitation.
Comments
In general, the consensus seems to be as judged from the literature, that environmental factors play a very important role in spousal-likeness. The most consistent comparisons can be made from the work of Garn and his colleagues and Cheraskin and his group. Both of these investigators, in some cases, studied the same or similar parameters. For example, this was surely the case with regard to diet. It is safe to conclude that the results are consistent. In other instances, Garn examines some characteristics not reported by Cheraskin and Cheraskin and his cohorts looked at areas not studied by Garn. Obviously, comparisons are not possible. However, the parameters still seem to follow predicted patterns. There is no question but that genetics play an important role (Chinese parents still seem to have Chinese children). However, within the limits of these studies, it would appear that the environment plays a cardinal (possibly even primary) role in the genesis of health and sickness. Even if it is no more important than genetics, it becomes a more serious practical consideration since it is easier to modify nurture than nature.
While the spouse-likeness mould is relatively simple, inexpensive and highly convincing in differentiating inheritance from the environment, it is not without its problems. For example, there have been enough incontrovertible studies to show that, even in marriage, there are factors which must be viewed as social and which dictate familial aggregation. For example, it is well known that the selection of a mate is in part a function of height, religion, economics and geography.
Summary and Conclusions
Even in this day and age, and with the expenditures of large amounts of money, time and energy the general consensus is that both inheritance and environment dictate health and sickness What has still eluded us is a more precise estimate of their relative contributions of nature versus nurture. As but one current example, in the case of breast malignancies, one major point is now evident from the Nurses’ Health Study (12-year look at 117,988 middle aged women).70 The contributions of inheritance are relatively small (approximately 2.5 per cent). This would tend to corroborate the cardinal role of the environment.