Therapeutic Practice in Colonial New England

Demographic Properties of the Four Practices

Of the 7095 patient visits included in the samples from the four physicians’ ledgers, fifty-two percent were Bartlett’s, sixteen percent were Rowe’s, twenty-one percent were Aspinwall’s, and eleven percent were Townsend’s. As Table I shows, adult men constituted the majority among the three practices for which data pertaining to age and sex are available. Children were less numerous in Dr. Rowe’s practice than in Dr. Bartlett’s, both in small New Hampshire towns. Over the thirty-seven years of Bartlett’s ledgers, the proportions of women and children increased; perhaps they were artificially low during the first years of his career, before his skills became recognized and accepted. Because both he and Rowe were the only physicians in their respective communities, we wonder who took care of the children in East Kensington or whether, in fact, their parents even sought professional medical help when the children were ill.8 Perhaps another physician in Rowe’s area “specialized” in the care of sick children, an inference suggested by the fact that Rowe saw only about two-thirds as many patients each day as did Bartlett. It is most likely that Rowe’s patients were drawn from a smaller population than Bartlett’s, whose patients came not only from Kingston but also from surrounding towns, like Dover, Hampton, and Newmarket. Unfortunately, it is not possible to estimate the number of patients seen daily by the two Boston physicians.

table i

Demographic properties of the practices of four colonial New England physicians, collated from data in their ledgers.

Sources: See text.

* Among patients whose age and sex can be ascertained from the ledger entries.

† All four physicians often saw more than one patient during each house call.

As Table II shows, the populations of both Kingston and Kensing ton were relatively stable during Bartlett’s and Rowe’s lifetimes. After the turn of the century, Kingston’s population even fell by a modest amount. By contrast, Boston’s population was increasing rapidly. Although the town’s prewar population had fallen by about two-thirds at the end of the siege of 1775–1776, by the time of the Aspinwall and Townsend ledgers it was increasing rapidly, by about a third over the last decade of the eighteenth century.

table ii

Populations of Kingston, Kensington, and Boston in the late eighteenth century.

Sources: Jeremy Belknap, The History of New-Hampshire, 2nd ed., 3 v. (Boston, 1813), iii, 228; Timothy Dwight, Travels in New England and New York, 4 v. (1821–22; rev. rpt. ed., Cambridge, Mass., 1969), I, 301; Frank C. Mevers, ed., Microfilm Edition of the Papers of Josiah Bartlett (1729–1795) (Concord, N.H., 1976), frame 3075, and Guide, 11; Lemuel Shattuck, The Vital Statistics of Boston (Philadelphia, 1841), xii–xvii

Bartlett has provided us with one tabulation of the population of his own town, one that he made in his capacity as a Kingston selectman in 1773:

Although Bartlett’s arithmetic failed him (the actual total is 989), the age and sex distribution of Kingston’s population was like that of all New Hampshire. The town’s doctor served a population of about 1,000, plus his patients who lived in neighboring communities. This physician-to-population ratio is like that found in Portsmouth (whose population was 5,000) at the same time, and probably in eighteenth-century America as a whole.9 Boston’s population was about twenty times larger than Kingston’s, and boasted an even greater ratio of physicians to population, as Eric H. Christianson has shown elsewhere in this volume.

Graph 2 shows fluctuations in the average number of patients that Bartlett saw each day. His practice increased most during epidemics, reaching its peak work load just after the French and Indian War, when both poverty and disease were widespread.10 Bartlett’s growing involvement with public service and then in revolutionary politics eventually curtailed his opportunities to see patients, although he did see some throughout the entire period of his ledgers. During the last years of his practice, Bartlett was seeing fewer patients than Rowe was at the same time (Table I), but both men were making similar numbers of house calls each day. It was not possible to determine the daily number of house calls in the two Boston practices, but it was probably like those of the two rural practitioners.

Bartlett saw patients most frequently in the autumn months, September through November, and least often in January and February (as determined by the mean number of patients seen each month during the thirty-seven years during which he kept the ledgers). This peak incidence of disease in the autumn was also felt in Portsmouth, where deaths were most numerous in the same months, but the fewest deaths occurred there in the spring, from March through June.11 However, many sporadically epidemic diseases with high morbidity have low mortality, and vice versa, factors which could well influence the value of this statistic (see, e.g., Table XV). Data for the other three physicians’ monthly practices were insufficient to permit making further comparisons.

table iii

Mean patient fees charged (although not necessarily collected) by four colonial New England physicians. Part of the large variation (as reflected in the Standard Deviations) can be attributed to the charges per mile superimposed on standard fees when making house calls outside each physician’s own town, and to the varying numbers of drugs prescribed by each physician.

Sources: See text.

* Both Aspinwall and Townsend charged twenty-eight shillings for attendance at a delivery; prenatal care seems not to have been included in that standard fee

table iv

Frequencies of bleeding, tooth extraction, application of dressings, opening abscesses, obstetrical deliveries, and other surgical procedures in four colonial New England medical practices. See Table V for a detailed list of the other operations performed by these physicians.

Sources: See text.

*Dr. Bartlett bled ten children by cutting the undersides of their tongues, all in only the first two quarters of the ledger pages sampled.

† Surgical and accidental wound dressings are collated here, because it was not possible to differentiate between them in most instances.

Graph 2 also shows that Dr. Bartlett’s professional fees increased at an average rate of about 2.4 shillings a year over the first thirteen years of his practice, until they fell suddenly, by ninety-five percent, in early 1765. Afterwards, he and Rowe charged similar fees, while their Boston colleagues charged about twice as much (Table III), presumably because the cost of living was greater in urban than in rural areas. During the entire period of his ledgers, Bartlett kept his accounts in pounds sterling, but he accepted Continental and, later, United States, money at current rates. For instance, on 31 December 1786, fifty dollars were credited to one patient’s account as four shillings and three pence.

The precipitous decline in Bartlett’s professional fees in early 1765 seems to be associated with the economic decline prevalent throughout the colonies. Most of this is attributable to the recession that followed the French and Indian War, along with Parliament’s ill-advised attempts to raise more revenue from the colonists. The Currency Act of 1764, which prohibited the printing of any more paper money in the colonies, and other acts regulating the colonial trades, probably combined with the acute money shortage, caused by the forcible removal of ready cash to England, to lower Bartlett’s fees. He could not have expected more from his patients, whose ability to pay seems to have been limited even in the best of times. Kingston was described, shortly after his death, as “chiefly a pine plain. It contains a scattered hamlet near the church. The houses throughout the town are indifferent. The soil is light and thin, the cultivation very ordinary, and the whole appearance of the country unpromising.”12

“Surgical” Treatments

Table IV shows the extents of non-drug treatments employed by the four New England physicians, treatments which, for convenience, will be called “surgical” here. Dr. Aspinwall used surgical techniques the least often, and Dr. Rowe the most often, although the ledgers provide no definitive clues that might explain the fourfold difference between the two practitioners. Presumably because Bartlett and Rowe were the only practitioners in their towns, they used surgical techniques more often than their colleagues in Boston, where opportunities for even primitive degrees of specialization were greater and more easily exploited.

The two rural physicians bled about five percent of their patients, while the two Boston doctors almost never did. In no instance was any one patient bled more than once for the same illness (i.e., within a year), but the amounts removed were not recorded in the ledgers. From one-third to three-fourths of the patients who were bled also received drugs at the same time, but there was no discernible association of specific drugs with bleeding. Bartlett bled women disproportionally more frequently (not tabulated) than expected from the proportion of women among his patients, but his reasons for doing so are not identifiable. On the other hand, both Rowe and Bartlett bled children in direct proportion to their numbers among those physicians’ practices (also see Table I), and Rowe also bled women in proportion to their numbers in his practice.

The Boston doctors extracted teeth much less often than did their counterparts in New Hampshire, where, presumably, physicians were called upon to render a greater diversity of professional services. None of the four physicians administered drugs to his dental extraction patients. In all four practices, the sex and age group of the dental patients were consistent with those of the entire practices. Tooth extractions may have been more common in America than elsewhere. Timothy Dwight remarked that “It is supposed that our teeth more generally decay at an untimely period than those of Europeans.”13

Again, the rural doctors opened abscesses more often than the city doctors did, but the rural doctors seem not to have delivered any babies, perhaps because the families of pregnant women provided whatever assistance was needed. There is no evidence of midwives practicing in Kingston or Kensington. Other kinds of nonmedical (i.e., non-drug) therapies were performed about equally frequently by the four physicians studied.

table v

Numbers of nonmedical treatments (excluding bleeding and tooth extractions, for which see Table IV) provided by four colonial New England physicians.

Sources: See text.

*Dr. Rowe tapped his father, also a doctor, sixty-seven times, during which 2368 pounds of fluid were removed over twelve months in 1786–1787. The father survived another three years.

Dr. Aspinwall, in Boston, applied dressings to his patients, for whatever reasons, much less frequently than did the other three physicians. It is unlikely that this difference can be attributed to the availability of a “specialist,” but there may have been some other practical reason for it. The age and sex of patients who were dressed by all four physicians were in proportion to the age and sex groups of their entire patient populations, as shown in Table I. Drs. Rowe and Townsend, who used the most dressings, were also the least likely to give those same patients drugs at the same time.

The other surgical therapies the four doctors provided are listed in Table V. Although major operations such as couching for cataract (i.e., dislocating the opacified lens into the lower part of the eyeball) and amputations were being performed in Portsmouth and Boston at the same time, none of the four physicians seems to have done them. Smallpox inoculation was well known throughout New England, yet there is little evidence for it in the four ledgers. Aspinwall was regarded as a prominent inoculator, yet his ledgers show only two inoculations. However, inoculations were not usually given to private patients unless an epidemic threatened, so Aspinwall probably kept separate inoculation ledgers. Bartlett never even had himself inoculated until he went to Philadelphia in 1775, when Benjamin Rush immunized him for four pounds ten shillings. The ledger data confirm the memory of Bartlett’s son Levi, also a physician, who wrote that his father’s “feelings were too humane & tender to attempt operations in Surgery, or to shine as a Surgeon.” Because of this, it is likely that surgeons were called from Portsmouth or other towns when needed for major operations.14

The Medical Milieu of New Hampshire and Boston as Reflected in Mortality Data

Table VI shows crude death rates and ratios of deaths to births in colonial New Hampshire towns and in Boston. The data indicate that mortality rates were greater in the most densely populated areas, either when measured as crude death rates or when taken in conjunction with birth rates. In Boston, births just balanced deaths, whereas in rural New Hampshire each death was replaced by two births. (The high infant mortality rates would have lessened the actual ratio, but the infant mortality would have been even greater in Boston, resulting in greater net gains in the rural areas.) Throughout the eighteenth century, New Hampshire’s population was increasing more rapidly than that of all thirteen colonies combined, while the Massachusetts population was among the most slowly growing on the continent.15

table vi

Crude death rates and proportion of deaths to births in colonial New Hampshire towns and Boston in the eighteenth century.

Sources: Daniel Adams, ed., Medical and Agricultural Register, 1 (1806–1807), 284–285; Jeremy Belknap, History of New-Hampshire, 2nd ed., 3 v. (Boston, 1813), III, 179–184; J. P. Brissot de Warville, New Travels in the United States of America 1788, trans. Mara Soceanu Vamos and Durand Echeverria (Cambridge, Mass., 1964), 288; J. Farmer and J. B. Moore, Collections, Topographical,... New Hampshire, 3 v. (Concord, 1822), I, 24, 80–83; Ora Pearson, “Bill of Mortality for Kingston, N.H.,” Collections of the New Hampshire Historical Society, 5 (1837), 250–252; Lyman Spalding and Richard Thurston, Bills of Mortality for Portsmouth, New Hampshire (1801–1811, 1818–1820), in Boston Medical Library; J. Worth Estes, “‘As Healthy a Place as Any in America’: Revolutionary Portsmouth, N.H.,” Bulletin of the History of Medicine, 50 (1976), 536–552; Jabez Dow, Bill of Mortality for Kensington, N.H. (Dover, N.H., 1802).

Table VII shows that there were more deaths in Josiah Bartlett’s Kingston than in Benjamin Rowe’s Kensington during the middle six decades of the eighteenth century, when Kingston was presumably a little larger. Not until the last two decades did the numbers of deaths in the two towns become approximately equal, as were their populations by then. There is no reason to believe that Kensington’s patients received medical care that was substantially different from that offered in Kingston; the throat distemper epidemic which affected much of New Hampshire in the first half of the century was responsible for many more deaths in Kingston than in Kensington.

table vii

Mean annual number of deaths (with Standard Deviations) during decennial periods in Kingston and Kensington, New Hampshire, throughout the eighteenth century. See Table II for the populations of the two towns.

Sources: Ora Pearson, “Bill of Mortality for Kingston, N.H.,” New Hampshire Historical Society Collections, 5 (1837), 250–252; Jabez Dow, Bill of Mortality for Kensington, N.H. (Dover, N.H., 1802).

* When the deaths attributed to the 1735 epidemic of throat distemper are subtracted, this value falls to 13.9 ± 9.6.

table viii

Recorded causes of death in colonial New Hampshire and Boston.

Sources: Jeremy Belknap, History of New-Hampshire, 2nd ed., 3 v. (Boston, 1813), III, 179–187; Jabez Dow, Bill of Mortality for Kensington, N.H. (Dover, N.H., 1802); Lyman Spalding and Richard Thurston, Bills of Mortality for Portsmouth, New Hampshire (1801–1811, 1818–1820), in Boston Medical Library; Lemuel Shattuck, The Vital Statistics of Boston (Philadelphia, 1841).

During the years of Bartlett’s ledgers for which data are available, the people of Kingston suffered an average of about thirteen deaths per year, slightly less than for the entire period. Rowe’s entire practice spanned the last three decades of the eighteenth century; during the few years covered in his ledger, the mean annual number of deaths in Kensington was about nine, just slightly more than that for his entire career.16

The reported causes of death in Hampton and Dover, where some of Bartlett’s patients lived, in Kensington, where most of Rowe’s patients lived, and in Portsmouth and Boston are listed in Table VIII. Although it is notoriously difficult to equate eighteenth-century diagnoses with those in use today, the earlier diagnoses can be categorized into certain general classes. For instance, infectious diseases, especially among children, were the leading causes of death in our four physicians’ practices. Of these, consumption, most cases of which were probably caused by the tubercle bacillus, was most conspicuous. Some diagnoses, such as convulsions, fever, cough, diarrhea, dysentery, and putrid fever were too nonspecific to be easily interpretable today. But other diseases, such as bilious colic, measles, scarlet fever, smallpox, whooping cough, and perhaps childbed fever are less likely to be misunderstood in terms of twentieth-century concepts of their etiology.

In general, the data in Table VIII show considerable uniformity among the attending physicians’ diagnostic standards. For the New Hampshire towns at least, and probably in Portsmouth and Boston, the diagnoses given in their bills of mortality were originally made by doctors in those communities, not by nonprofessionals (like those who had contributed to the earlier London bills, for instance). The one major difference is for the throat distemper, which appeared from 1735 to 1740 in Kingston and spread to many rural New England towns, while inflicting relatively little damage on the large seaport towns. Probably a combination of diphtheria and scarlet fever, the throat distemper’s effects are still seen in the Kingston cemetery, in the midst of which is a large swath of small children’s headstones, most of them dated 1735 (at which time Josiah Bartlett was only six years old).17 It is unlikely that even the few deaths attributed to the throat distemper in Portsmouth and Boston after the turn of the century were, in fact, caused by the same combination of bacteria that had attacked rural New England many years earlier.

In all the ledger pages sampled in this study, only three deaths were noted, all in William Aspinwall’s Boston practice. Of these, the cause was given for only one, “in childbirth.” Both Aspinwall and Bartlett noted that a single patient had measles; presumably both patients survived.

Because the Kensington bills of mortality are given by year, it is possible to determine just which patients died during Benjamin Rowe’s practice in that town and, consequently, which diagnoses he must have made both throughout his career and during the period of his ledger. Mortality among infants dropped from sixteen percent of Kensington deaths before Rowe began his practice to eleven percent afterwards (although, as noted earlier, Rowe seems to have few pediatric patients), while deaths attributed to consumption climbed from five percent to eighteen percent (probably as the town grew in population), and those to dropsy from three percent to five percent. Otherwise, Rowe’s diagnoses of causes of death do not differ remarkably from those made by his father, who had had the entire practice of Kensington before the son opened his own practice in adjacent East Kensington.

Pride in local longevity was endemic throughout early America, an early form of chauvinism that was consistent with the very concept of a New World. Although the Kingston bills of mortality list no diagnoses, names and ages are given for persons who died at ages greater than ninety. Of sixteen nonagenarians dying in Kingston between 1751, when Bartlett began his practice, and 1810 (only five were men), virtually all were or had been his patients, as can be determined from his ledgers.

table ix

Life expectancies, in years, for colonial New Hampshire towns and Boston. The total life expectancy experience is measured as the area under the life expectancy curve, here compared with that for Portsmouth as the standard.

Sources: Jeremy Belknap, History of New-Hampshire, 2nd ed., 3 v. (Boston, 1813), III, 178–187; J. P. Brissot de Warville, New Travels in the United States of America 1788, trans. Mara Soceanu Vamos and Durand Echeverria (Cambridge, Mass., 1964), 288; J. Farmer and J. B. Moore, Collections, Topographical,. . . New Hampshire, 3 v. (Concord, 1822), I, 24, 81–83; Lyman Spalding and Richard Thurston, Bills of Mortality for Portsmouth, New Hampshire (1801–1811, 1818–1820), in Boston Medical Library; Lemuel Shattuck, The Vital Statistics of Boston (Philadelphia, 1841); “Bill of Mortality for Rochester, N.H.,” Collections of the New Hampshire Historical Society, 1 (1824), 283–284; J. Worth Estes, “‘As Healthy a Place as Any in America’: Revolutionary Portsmouth, N.H.,” Bulletin of the History of Medicine, 50 (1976), 536–552.

The crude death rate and consumption mortality data (in Tables VI and VIII) are again reflected in life expectancies in four rural New Hampshire towns, Portsmouth, and Boston, as seen in Table IX. Because there were no additions to physicians’ therapeutic tools in the last decade of the eighteenth century that might have substantially altered the death rates or longevity for entire populations, the data for the two larger communities can be considered to reflect at least the same situation which must have been found there throughout the last half of the century.

The data in Table IX, like those in Tables VI and VIII, support the hypothesis, well known to contemporary writers such as Malthus, that mortality and presumably morbidity were proportional to population density. That is, inhabitants of the smallest and least heavily populated communities could expect to live longer than those in the more populous cities and towns. Aspinwall and Townsend, then, would have had fewer older patients, and would have seen higher morbidity and mortality rates among their patients than would Bartlett and Rowe.

Drug Therapies

About ten percent of patients in the four colonial New England practices received no drugs at all. Dr. Rowe used fewer drugs than his three colleagues, and Dr. Townsend was the most likely to administer no therapy of any kind at all. Bartlett gave “Directions” for self-care to 5.3% of his patients, Rowe to 0.5%, Aspinwall to 1.0%, and Townsend to 0.1%. Bartlett “advised” other physicians as to the care of their patients four times, and Aspinwall twice. Paid professional consultations, therefore, seem to have been as rare in the city as in the country, but Bartlett may have distinguished himself for the number of instructions to patients for their own care.

Colonial physicians seem not to have practiced polypharmacy to any great extent, as seen in Table X. The average number of drugs prescribed for all patients, including “surgical” patients, ranged from about 1.0, in Boston, to about 1.5 in rural New Hampshire. The median number of drugs administered to nonsurgical (i.e., “medical”) patients was even lower, ranging from about 0.5 in Boston to a little over 1.0 in New Hampshire. Bartlett prescribed more drugs in the late 1760’s than during the rest of his practice, but the reason is not now obvious. In general, the Boston physicians used fewer drugs than the country doctors, but multiple drug use was nowhere common. In fact, the Appendix shows that even those drugs that had more than one ingredient were few, and in the vast majority the other ingredients were not even then thought to be active drug principles.

table x

Numbers of drugs prescribed for their patients by four colonial New England physicians.

Sources: See text.

* Excluding patients who received only surgical treatments (given in Table IV).

table xi

Drugs administered by four colonial New England physicians, grouped according to the classification of James Thacher (for which see the Appendix), and arranged by decreasing frequency of Josiah Bartlett’s usage of each drug class. Within the classes, drugs are listed alphabetically after those of unknown composition but with names which necessitate their inclusion in the class. Individual drugs are identified in the Appendix. Each drug was assigned to the first drug class given for it by Thacher; when he ascribed other attributes to any drug, its secondary class assignments are given in square brackets in this Table. Alternative drug names used by any of the four physicians are given in parentheses.

DRUGS AND DRUG CLASSES [other classes to which drugs also assigned by Thacher]

Sources: Frank C. Mevers, ed., Microfilm Edition of the Papers of Josiah Bartlett (1729–1795), 7 rolls (Concord, N.H., 1976), rolls 6 and 7, every tenth frame from frames 3663 to 4562; every tenth double page of the ledger of Dr. Benjamin Rowe, Jr., a 10% random sampling of well-filled legible pages of the ledger of Dr. William Aspinwall, and a 30% sampling of similar pages of the ledger of Dr. David Townsend, all in the Boston Medical Library, James Thacher, The American New Dispensatory (Boston, 1810), passim; P. S. Brown, “Medicines Advertised in Eighteenth-Century Bath Newspapers,” Medical History, 20 (1976), 152–168; James McKittrick Adair, Medical Cautions ... (London, 1787), 85; A. C. Wootton, Chronicles of Pharmacy, 2 v. in 1 (1910; rpt. ed., Boston, 1971); J. Worth Estes, “John Jones’s Mysteries of Opium Revcal’d (1701): Key to Historical Opiates,” Journal of the History of Medicine and Allied Sciences, 34 (1979), 200–209; The Merck Index, 7th ed. (Rahway, N.J., 1960); John Woodall, The Surgion’s Mate (London, 1617).

table xii

Presumed active ingredients, as listed by James Thacher, of drugs administered by four colonial New England physicians. Modern names for chemical compounds, and for plant and animal species, genera, and families are used. Because some drugs had more than one active ingredient, and because some drugs listed in Table XI had no active ingredients, according to their eighteenth-century users, the sums of all ingredients used by each practitioner do not add up to 100%. The agents listed are arranged by decreasing frequency of Josiah Bartlett’s use of each.

Sources: Same as for Table XI.

* Imported, according to Thacher.

† Probably imported.

Table XI lists all the drugs prescribed or, in most instances, dispensed by Drs. Bartlett, Rowe, Aspinwall, and Townsend. The 225 drugs of known composition or therapeutic aim listed included one hundred different ingredients that Thacher considered to be their active components, as listed in Table XII. Some of the drugs listed in Table XI contained two, and a few even three, “active” ingredients (see Appendix I) and some were recognized to contain no substances that had any specific therapeutic effect, only local palliative actions. In Table XI the drugs are assigned to the first of the drug classes listed for each in Thacher’s compendium of current drug usage (see Appendix I for Thacher’s classification), and the classes are given in descending order of Josiah Bartlett’s usage of them.

When the frequencies with which each of the four physicians used the drug classes are ranked, it is clear that the four physicians used the drug classes in similar relative orders of frequency among their patients.18 Although Dr. Thacher provided no data pertaining to the frequency of use of the several classes, the data in Table XI permit us to infer that his Dispensatory does, in fact, reflect a generalized picture of pharmacological indications for drug usage in colonial and federalist New England.

Close inspection of variations in the usage of individual drugs by the four physicians shows that each had a favorite drug for a given situation, such as Bartlett’s remarkable dependence on Elixir of Camphor among the Narcotics, or Aspinwall’s even more remarkable dependence on Spirits of Mindererus as a laxative among the Cathartics. But even discordances such as these do not substantially alter the relative rankings of drug class usage by the two doctors.

In general, Thacher’s classification and, apparently, the four physicians’ patterns of drug use were well grounded on concepts of humoral medicine. That is, the drugs were exploited in attempts to restore imbalances of the humors, by different routes or by different mechanisms of action. Most of the drugs used seem to have been considered as possessing “positive” qualities; most of them were “stimulants” to various portions of weakened or “debilitated” bodies.

Of the drugs listed, the Cathartics are the most prominent in terms of both sheer number and frequency of use. This drug class occupies only a minor role among prescription medicines today. Unfortunately, we do not know just what cathartic agents were used to compound a substantial proportion of many of the preparations given by any of the four physicians. Nevertheless, their supposed pharmacological function is clear. A few of the agents listed were obviously not used only because they promoted evacuations from the bowel, drugs such as Van Swieten’s Antivenereal and Essence Veneris, but it is difficult to believe that there were many patients with syphilis or gonorrhea in the two rural practices where those drugs were used.

Bartlett did not rely on Elixir of Camphor, the Narcotic that he used most often, for its analgesic properties, but for its putative sedative effect. The opiate analgesics were used with approximately equal frequencies by all four physicians. Thacher’s definition of Narcotics (see Appendix) included the use of opiates for their constipating, or antidiarrheal, property. The specific opiates used confirm my earlier hypothesis that more potent forms of the same agent tend to supplant less potent ones, the reverse of Gresham’s Law. In these four practices, Paregoric (and its equipotent analog Elixir Asthmaticus) were used much more frequently than the older, less potent, opiates such as Diacodium and Theriac, which Thacher recognized were unreliable.19

The two country doctors prescribed Tonic drugs more frequently than their urban colleagues. Of these drugs, the Virginia Snakeroot (Radix Serpentina) was the most commonly administered by all four physicians. Bartlett was the only one to rely heavily on the patented Stoughton’s Elixir. It seems likely that the Tonics may be the most “typically New England” of all these ancient drug classes. They rank second in mere numbers of drugs assigned to each class, and third in frequency of use. More important, the word “tonic” remains in common use—in New England, at least—today. The most successful nineteenth-century patent medicines were those advertised as having tonic properties, which carried implicit promises of benefits that would be desirable to the expanding American population. Perhaps it is not too cynical to suggest that Americans, uniquely among national groups, find special merit in agents that are supposed to provide additional tone—extra strength—to the spirit as well as to the body.

Cortex Peruviana was used, for its Astringent properties, more often than any other active principle (except for Camphor in Josiah Bartlett’s practice). The Peruvian Bark was also used in more different preparations than any other agent except, perhaps, opium. Only Aspinwall seems to have preferred another Astringent, Cascarilla, probably as a substitute for the Bark.

Because the Astringent Calybeate was virtually always prescribed for young girls with chlorosis, (a form of iron-deficiency anemia characterized by greenish skin color), the data in Table XI permit the inference that iron deficiency was more common in the country practices than in the city. In general, the four physicians used similar amounts of Astringents throughout their practices.

The country doctors were more likely to prescribe Antispasmodics than their Boston colleagues. Among these drugs Bartlett preferred Volatile Drops, probably an extract from the Cajeput tree, while Rowe preferred Valerian Root. The Boston physicians seldom used any Antispasmodic, and they used Epispastics less frequently than Bartlett and Rowe. Although Aspinwall used Emetics only sparingly, his colleagues used them often, especially Townsend. If the “Emetic” often administered by all but Aspinwall is the “Emetic Weed” (Lobelia inflata) described by Thacher, it is one of the few native plants that found major use in early American medicine. Aspinwall used so many available drugs so seldom, and Spirits of Mindererus so often, that one is tempted to think that he employed that remedy as a sort of panacea, that he thought it to be effective in a wide spectrum of disease.

We have already seen that dropsy, even in the eighteenth century recognized as the chief indication for Diuretic therapy, was diagnosed in about two to five percent of New Hampshire and Boston deaths. The frequency with which the four physicians prescribed Diuretics is entirely consistent with the mortality data in Table VIII. Of these drugs, Squill was regarded as the drug of choice, although Bartlett preferred Ol Terebinth. (Digitalis had been introduced to America through New Hampshire in 1787, but, because the leadership of the Massachusetts Medical Society remained unconvinced of the new wonder drug’s virtues, it would not gain general acceptance until after the turn of the century; consequently, digitalis does not appear in any of the ledgers studied.20)

We are not particularly surprised at the infrequent use made of the remaining classes of internal medications described by Thacher. Worms, uterine diseases, stones, and probably gastric hyperacidity (however it was diagnosed) were not common diseases. And the infrequent use of drugs such as Escharotics, Sialagogues, Expectorants, Errhines, and Diluents suggests that those classes had for some time been recognized as not very useful.

Dr. Townsend, in Boston, used external medications more often than the other three practitioners, of whom Aspinwall used them the least often. Some of these drug preparations may have been administered for the treatment of internal diseases, not only skin conditions. Therefore, it is difficult to ascertain patterns of use among the Emollients and Plasters.

Over the long period of Josiah Bartlett’s practice only one new drug appears for the first time after the first quarter of ledger pages. He first used Sal Nitre in the second quarter (1758–1764) and was using that Refrigerant in 1.6% of his patients by the end of his career. Over the four quarters of Bartlett’s ledger pages, the only drugs whose usage steadily increased were Castor Oil (1.4% to 4.7%), Radix Serpentina (0.6% to about 3.5%), Cremor Tartar (0.6% to 1.4%), Spermaceti (0.1% to 1.2%), and Opiates (2.3% to 5.0%). During the same thirty-seven years, Bartlett’s use of Cantharides and other Epispastics fell (2.0% to 0.5%) as did his use of Lavender (3.5% to 0.5%). He used the Alexipharmical Tincture increasingly often over the first three quarters (2.3% to 7.5%), but in the last quarter he used it relatively rarely (0.6%). And he provided his patients with Directions for self-care increasingly often over the entire period (0.9% to 6.4%). He used all the other drugs at similar frequencies throughout his entire practice. However, at least for Sal Nitre, Castor Oil, Radix Serpentina, Cremor Tartar, Spermaceti, and Opiates, there is some evidence that Dr. Bartlett became increasingly persuaded, perhaps through firsthand observation and experience, of their clinical efficacy. Similarly, he must have found little evidence that either Cantharides or Lavender should be used as often as he had been using them. And he must have begun to realize, after the first few years of his practice, that his patients were able to provide some of their own care, without the necessity (and, presumably, the expense) of further professional visits. Thus, Bartlett seems to have been satisfied with the therapeutic results of most of the drugs he used, for whatever reasons, because he continued to use virtually all of them in similar proportions for almost forty years, and introduced no major new therapies into his regular practice.

The frequencies with which the four practitioners prescribed the drug components that Thacher presumed to be the drugs’ pharmacologically active ingredients are arranged in the order with which Bartlett prescribed them in Table XII.21

Drs. Bartlett and Rowe were the most alike in their uses of these agents.22 Although Bartlett’s drug use was also proportional to those of Drs. Aspinwall and Townsend in Boston, the two rural New Hampshire physicians were more alike in their uses of drugs than they were like their Boston colleagues, as measured by the absolute values of the correlation coefficients.23 The two Boston physicians did not use the drugs they used for similar relative frequencies of their patients.24 Because Dr. Aspinwall used both Aloes and Ammonium Acetate (as Spirits of Mindererus, originally extracted from Aloes), in extraordinarily high proportions of his patients (25.45% and 24.91%, respectively), the physicians’ uses of drugs were compared after omitting those two drugs. When their prescriptions for Aloes and Ammonium Acetate are not taken into account, Dr. Aspinwall’s drug use is highly correlated with those of Dr. Bartlett and Dr. Townsend,25 although Dr. Townsend still seems to have differed somewhat in his use of pharmacologically active agents as the four physicians understood them.

The data in Table XII show that they differed more in their uses of the drugs that any of them used the most often than in their uses of drugs that any of them used least often (in fewer than about 1.0% of patients). Thus, we can infer that drugs which seldom found use in medical practice were used about equally infrequently by all the physicians studied, but that they differed most in their “favorite” drugs, those that they felt were most beneficial to the most patients, even for the same disease state, as seen in Table XI. We can only, hope that discovery of more detailed physicians notes, including information pertaining to both diagnosis and treatment, will someday permit us to understand better how physicians chose drug therapies for patients with demonstrably similar or identical illnesses.

Of the one hundred ingredients listed in Table XII, sixty-eight percent were of botanical origin, twenty-six percent were of chemical or mineral origin (including a few that could be made in eighteenth-century laboratories although originally they had been extracted from plants), and six percent were animal products (including two from insects).

table xiii

Distributions of patients, drugs, and active ingredients of drugs among practices of four colonial New England physicians.

Sources: Tables I, XI, and XII.

table xiv

Occurrence of drugs and active ingredients of drugs among four colonial New England medical practices.

Sources: Tables I, XI, and XII.

As shown in Table XIII, Bartlett administered a larger number of different drugs in his practice than did his three colleagues, and he used more drugs that were not used by the other three than did any of them. Because more than half of all the patient visits represented in the total sampling were Bartlett’s, and because his practice spanned the longest time, the greater diversity among the drugs he employed is not surprising.

Table XIV shows that only fourteen percent of all the drugs, but about a third of their putative active ingredients, were administered by all four of the physicians studied. Only about one-fourth of all these ingredients were employed by only one of the four physicians, and these chiefly among the least frequently used preparations (see Table XII). These data are consistent with the multiplicity of the most commonly used drugs made from the same raw materials, and with the evidence that each physician tended to rely on different preparations within major drug classes. As is clear in Table XII, those active ingredients that were used the most often by any one of the four physicians were used by all the other three as well.

Sources of Drugs

The microfilm edition of Josiah Bartlett’s papers also includes forty-three receipts or orders for drugs that he obtained from eight “wholesalers” in northeastern Massachusetts and southeastern New Hampshire from 1750 to 1772. The forty-three orders comprise eighty-nine different drugs or drug ingredients ordered a total of 502 times. Only three items he ordered do not appear among the drugs Bartlett prescribed or dispensed for his patients.

In order to assess the extent to which the drugs Bartlett ordered reflect his usage of them, correlation coefficients were computed. For all eighty-nine different drugs he ordered, the correlation between the frequencies of his orders and his prescriptions for the same drugs was low but significant.26 When Elixir of Camphor is omitted from the correlation computation (because it appears only twice among 502 drug items ordered, although Bartlett prescribed it for over twelve percent of his patients), the degree of correlation between drug orders and drug prescriptions increases markedly.27 And, for five other drugs there were marked (five to 76-fold) discordances between orders and prescriptions; when they are omitted, the correlation increases still further.28 These data lead to the inference that although the drug orders do indeed reflect Bartlett’s prescribing patterns, the orders do not provide a complete picture of the drugs he used in providing therapy for his patients. Most of the discordance is, of course, attributable to the incompleteness of the drug orders that have survived. Nevertheless, in the absence of a proven complete set of orders, they cannot be considered to reflect a physician’s everyday practice to the same extent that his ledgers can.

Towards the end of the century Jeremy Belknap solicited from physicians throughout New Hampshire their opinions about the medically useful indigenous plants of the state. His final list was (the Latin names in parentheses are Belknap’s, and the asterisks indicate names that also appear among the active ingredients listed in Table XII):

• * Sassafras (Laurus sassafras)
• Spice-wood, or Fever-bush (Laurus benzoin*)
• Oil-nut, or Butter-nut (Juglans cathartica)
• *Pitch pine (Pinus taeda), for Ol Terebinth
• *Larch (Pinus larix), for Ol Terebinth
• *Fir (Pinus balsamea)
• *Hemlock (Pinus abies)
• *Juniper (Juniperus sabina), an Anthelminthic
• White willow (Salix alba), a Cortex Peruviana* substitute
• Black poplar (—), like Balsam of Peru*
• Ginseng (Panax trifolium)
• *Prickly ash (—), for chronic rheumatism
• *Black elder (Sambucus niger)
• Red elder (Viburnum opulus)
• Maiden-hair (Adianthus pedatuus)
• Sarsaparilla (Aralia)
• Snakeroot (Polygala senega)
• Poke, Indian tobacco (—), with strong Narcotic quality
• Witch hazel (Hamamelis), for inflammation
• *Lobelia [L. inflata], a strong Emetic
• Lobelia cardinalis, “employed in the cure of a disease, with the name of which I will not stain my page” [i.e., syphilis]
• Buckbean (Menyanthes), of “singular use in medicine”
• Skunk cabbage (Arum), for asthmatic complaints

Poisons, also used in the cure of spasmodic diseases:

• Hemlock (Cicuta)
• Thornapple (Datura stramonium)
• Henbane (Hyoscyamus niger)
• Nightshade (Solanum nigrum)

Other poisons, without stated medical uses:

• *Ivy (Hedera helex [sic])
• Creeping ivy, or Mercury (Rhus radicans)
• Swamp sumach (Rhus toxico dendrum)
• Water elder (Viburnum opulus) (same as Red elder above)
• Herb Christopher (Actea spicata)
• Stinking snakeweed (Cliffortia trifoliata)
• White hellebore (Veratrum album)29

Comparing this list of medically useful plants indigenous to New Hampshire with the drugs and active ingredients listed in Tables XI and XII shows that, with the possible exception of Lobelia inflata, if that species was the “Emetic” listed so often by Drs. Bartlett, Rowe, and Townsend, the native vegetable productions of New Hampshire, and perhaps all New England, were insufficient to furnish most of the medical profession’s therapeutic needs.

Of the one hundred active drug ingredients included in Table XII, forty were imported. These forty (marked with an asterisk in the table) constitute fifty-five percent of all the drugs administered by Bartlett and Rowe, forty-one percent of those given by Townsend, and thirty-two percent of those given by Aspinwall. Of the other sixty drug ingredients, about half are inorganic chemicals that were easily prepared by physicians or their drug suppliers. Most of the other half were native plants, but they constitute only a small proportion of all the drugs administered. In general, New England physicians had to rely on imported materials for the drugs they most often prescribed. It would seem, then, that native American plants had not contributed substantially to the list of effective medicines available to colonial New England physicians, and that they were still dependent on drugs procured in other countries long after the Revolution.

Of all the drugs prescribed by the four physicians, 19.4% were inorganic chemicals or purified organic extracts. They constituted only fifteen percent of Bartlett’s practice, but from twenty-four percent, of Rowe’s, to thirty-two percent, of Aspinwall’s (because of his reliance on Ammonium Acetate). These proportions suggest that drugs in the Paracelsian tradition were more important among the urban practices than in rural areas, but that colonial physicians gave Galenic botanical preparations to the majority of their patients.