The 2005 STCA Health Survey is a census of the Scottiesh Terrier breed. As such it serves as a snapshot of the breed health. The results have been divided in a set of smaller documents for easier reading.
2005 Survey Owner/Breeder
The number of responses to the 2005 Health Survey was roughly equivalent to the response to the 1995 survey. Table 1 below provides a high level summary of the records that are included in the database delivered to the Health Trust. Specifically, surveys were received from 487 different individuals, but 7 of them involved either Litter or Health and Disease surveys that did not have an associated General Information form, so there are 7 blank entries in this portion of the database. Although there were a total of 203 Reproductive Information forms recorded in the database, there are 20 of these forms that have a litter count of zero, meaning that these individuals are not really Scottie breeders with experience pertinent to the other areas of the form. And finally, there were Litter survey forms received from 73 different sources, but 5 of these sources had all of the litters they submitted eliminated from the survey because the whelp date that they provided was outside on the two year window specified in the survey instructions.
The Owner/Breeder Survey actually spans two survey forms; the General Information form and the Reproductive Information form. The following tables are grouped into categories of similar questions.
General Information Survey Data:
Table 2 shows a comparison between the 1995 and the 2005 survey in terms of the numbers of responses and some of the demographic type information. There were roughly the same number of responses to both the 1995 and the 2005 survey. Responses to the 2005 survey included 272 that indicated they were a member of the STCA and 243 that were a member of a Regional Scottish Terrier Club. Although it is not shown in the table, 171 of the responses were from individuals that are a member of both the STCA and a Regional Club. It should also be noted that 135 responses (27.7%) indicated that they were not a member of either the STCA or a regional club
Table 2 also compares years of experience and numbers of Scotties owner by each respondant. With roughly the same number of surveys being submitted, the 2005 data shows owner experience (e.g. an average of 21.1 years that they have owned Scotties) was up by 23% over the numbers reported in 1995. At the same time, the number of Scotties (e.g. an average of 2.8 Scotties per response) was down 9% overall and 12% in the average survey household. Future analysis of the survey data may choose to investigate some of these changing demographics and the impact that they may have on breed health and education objectives.
Table 3 shows the types of activities that individuals are participating in along with their Scotties. Similar data was not presented in the 1995 survey, so it is not possible to offer a comparison. In this instance, the data suggests that the vast majority of respondents consider their Scotties to be companions. Many of the “Other” activities also include non-AKC event activities such as Therapy Dog, walking and running companion, rescue and just plain “pet.” This may be an area where further analysis of the data using statistical techniques may be able to shed some additional light. Even knowing the spread and variation in STCA membership across the various activities could be interesting.
Table 4 provides some insight into the age statistics of our breed. While the number of dogs and the average number of dogs in each household are down slightly from 1995, there does not appear to be any information to suggest a change (either positively or negatively) in terms of the average lifespan of Scotties. I need to point out a significant problem, both with the design of this question and the analysis of the data. First, more than one respondent obviously had a problem with the question because they included the ages of living dogs in their response – it is my understanding that only deceased Scotties should be included in a lifespan analysis. Secondly, from a purely mathematical (or statistical) standpoint it is not proper to average a number that is already an average – the results of this calculation do not produce an answer with clear meaning.
[Ed. In those cases, (where it was clearly obvious) the author did not record an answer that pertained to living dogs in the household. A low number (like 3 years) was included if it was related to the age at death of a Scottie. This is an example of where this low number (for one dog) was recorded as an equal to another household that responded with an answer of 10 years that represented many dogs.]
As an alternative to the General Survey question, a look at the Health and Disease data for deceased dogs provides a much more specific answer to the average lifespan question. The data shows a total of 266 deceased dogs, with an average lifespan of 10.7 years with maximum and minimum values of 17 and 0.8 years respectively.
Table 5 reports on the responses to two related questions on the survey. Both questions asked for information on whether individuals had either bred or owned Scotties in the past two years that exhibited six specific traits and genetic diseases. A comparison with data from the 1995 survey was not possible because that survey did not include a similar question. Furthermore, the numbers from the 2005 survey do not in all cases agree with the data obtained from the Health and Disease survey. Specifically, there were only 30 dogs with Scottie Cramp and 22 dogs with CA included in the more detailed Health survey. It is difficult to tell whether those individuals who answered this question included only dogs they owned in the last two years of if they may have included all dogs that they have ever owned.
Table 6 addresses the use of heartworm preventatives among Scottie owners. While the numbers of owners that reported using a monthly heartworm preventative appears to have increased, the usage among breeders with active breeding stock appears to have dropped.
Table 7 provides data on those diseases that respondents reported as their foremost concern; either among their own dogs or within the general Scottie population as a whole. The answers to these two survey questions were obviously based on opinion and not fact. The two answers often had some similarities, but overall they included some significant differences. In the first case, many of the issues listed in the first response were also listed by these owners in their responses to the Health and Disease Survey questions, so there was definitely some correlation between “My major health concerns” and the diseases that were reported for individual dogs in the health survey. I would also point out that an answer of “Cancer, cancer, cancer” was counted only once in this section, while each dog with cancer in the health & disease section was counted individually.
Another difference between these two lists is the appearance of several non disease issues in the list of major Scottie health issues. Specifically, there were references to the STCA standard and various breeding practices as being a major health issue.
2005 Survey Appendixes
Appendix 1 : Listing of Illness Codes
The following list of Illness Codes were used in preparation of this survey report. The list is pages long, and was prepared with the assistance of the Health Trust and Dr. Marcia Dawson.
Figure 1 shows a scatter plot of the Guesstimate of Carrier Frequency values versus the Frequency of Occurrence data as documented in the report on the 1995 survey. This plot is essentially the Hardy-Weinberg Law in graphical form. Figure 2 shows a similar scatter plot of the numbers documented in this report on the 2005 survey.
This figure confirms that the values used in both reports are consistent with each other.
2005 Survey Health and Disease Summary
Table 20 offers some summary information about the dogs included in the Health and Disease survey database. The first section shows the spread of coat colors across dogs and bitches. Noteworthy is that black is the most prevalent coat color and there are more bitches than dogs in the survey.
Table 20 also provides a summary of the number of healthy and diseased dogs by sex and their current living or deceased status. As can be seen, over 80% of the dogs in the database are still living. Over half of these living dogs were reported to have no health issues.
One of the major accomplishments of the 1995 Health Survey was to produce a listing of genetic traits and diseases and then use the survey data to help make a prediction of the frequency and carrier rates for each of these diseases. Table 21 offers a review of the 1995 data along with new numbers from the 2005 survey data. The 2005 survey data includes much more information, because this table lists only those diseases and genetic traits that showed up on both surveys.
Dr. George A. Padgett, DVM was the one that helped compile and calculate the original statistics. The data and his methods were based on the Hardy-Weinberg Law was also published in his book titled; Control of Canine Genetic Diseases. The data in table 21 is based on this same approach and provides very similar data.
The right hand column in Table 21 indicates how some of the carrier frequencies may have fluctuated up or down between the two surveys. Before anyone tries to put too much importance on these fluctuations in the numbers let me paraphrase a couple of comments from Dr. Padgett’s book.
Surveys are not accurate. Data is not always correct. Modes of inheritance are not always known. Hardy-Weinberg does not really apply. Guesstimates are not absolute numbers, They are just the best numbers we have today!
[Ed. The 1995 Survey report did not include an analysis of the formula used by Dr. Padgett to calculate his Guesstimate of Carrier Frequency values. However a plot of the data reveals the basis for his values and permits us to make new guesstimates based on the new data from the 2005 survey. Appendix 2 shows plots of the data from both surveys and confirms that they are both based on the same analysis.]
Table 22 offers some insight into the most frequently reported illnesses. The table lists any illness that had 20 or more occurrences in the database. You should ask, exactly how did all of the diseases that were reported get sorted? Many of the common diseases like TCC, Cushings and hypothyroid problems were pretty easy to identify and sort into these lists. But the data contained hundreds of alternative names, similar names and unrecognized names for reported illnesses.
To sort out all of these reported illnesses, the Disease and Illness index that was distributed along with the survey was used as a starting point. Each item on the list was assigned a unique three digit code that can be easily sorted by the computer. Each reported illness was then assigned to one of these illness codes. After attempting to assign codes to as many reported illnesses as possible, the complete list of illnesses and the coded index was reviewed by Dr. Marcia Dawson, DVM. Working with Dr. Dawson a few new illness codes were identified that needed to be added to the index. Dr. Dawson also provided invaluable assistance because she verified that all of the reported illnesses had been assigned to an appropriate illness code. The database file contains a complete list of the illness codes that were used as well as the original illness data from each survey form. Table 22 is therefore a report on the most frequently assigned illness codes.
In the 1995 survey analysis, Sue Martin noted that several of the related illness categories might be combined together, and that this would change the order of the most frequent illnesses. Using the 2005 data, similar results can be noted.
Allergies: Combining the 5 related Allergy codes and the general “Skin” code would result in a total of 95 occurrences and raise this to the number 2 health issue.
Periodontal: There were separate codes for periodontal disease and gingivitis, which if combined would result in 38 occurrences and raise this issue from number 12 to number 6 on the list.
Temperament: There were three codes dealing with temperament; Aggression(g), Fearfulness, and Instability that when combined would total 31 and put this in the list of top ten items.
Table 23 addresses the variety of cancers that were noted in the 2005 survey. There were over 30 different codes dealing with very specific cancers, general cancers, and abnormal tumors or growths. If all of these codes are combined together the count would be over 250, so instead, table 23 offers a listing of the top 10 cancers reported in descending order that were noted in the survey data.
2005 Survey Conclusion
The 1995 Health Survey results were documented by Susan Martin, who worked in collaboration with Dr. Padgett, in a series of Bagpiper articles that formed the bases for several other studies. Because many of the 2005 survey questions were so similar, it was only appropriate that this report should follow the same format. The following links to the 1995 Health Survey reports are offered as additional background for anyone that may be interested.
The nature of many surveys and the reports on their results is that they tend to be biased, either in terms of the questions used or the interpretation of the answers. The author hopes that this report on the 2005 Health Survey offers "Just The Facts" and that any interpretations have been left to the reader or future authors.
2005 Survey Introducton
The following article, which discusses responses to the 2005 Health Survey, is raw data. The reader is strongly cautioned that the results here presented represent a unique snapshot of a small number of dogs reported upon by a small number of owners at a particular moment in time. The results may or may not hold true for the larger population of all Scottish Terriers or for any dogs of other breeds. While the observations reported may hold true in the overall population of Scottish Terriers, no probability calculations have been done.
/s/ Louis A. Mitchell, M.D. Chairman, STCA Health Trust Fund
Using a compare and contrast format, this report provides an update to the article “Report On The State Of Health Of The Scottish Terrier” written by Susan Martin for the 1995 survey. Additional reports may be commissioned by the STCA Health Trust to look further into the survey data in the hopes of developing a complete understanding of the health of our breed.
The Data Transformation Process
As was the case in 1995, every member of the Scottish Terrier Club of America (STCA), U.S., Canadian, received a copy of the 2005 Health Survey. The Health Trust sent 50 sets of the survey forms to each regional club for distribution to non-STCA members. These local clubs were also encouraged to have members give copies to puppy customers, grooming customers and anyone they knew who owned a Scottie. The 2005 Health Survey forms were also posted on the STCA Website where anyone could download a complete set of forms and instructions.
Similar to the process used in 1995, everyone was instructed to send their completed survey forms to Leah Deckner who was designated as the blind recipient for all survey responses. In an effort to maintain complete privacy, Leah stripped the envelope and any incidental material from each survey. Groups of survey forms were then boxed and mailed to Jim Orsborn. Upon receipt, each stack of survey forms was numbered and prepared for processing.
[Ed. Because each survey response was totally anonymous, it is not possible to track any problems by geographical distribution. However, because a unique record number has been assigned to each set of survey forms it may be possible to correlate some Health and Litter Survey data with information on the General Information forms (e.g. litters by STCA Members versus non-members). However, this does NOT mean that it will be possible to link any survey data to either kennel names or to individuals submitting the survey data.]
All of the survey responses were recorded in a database program, called FileMaker Pro, which is a full functioned, relational database application. For this project, FileMaker’s biggest advantage was the ease with which the user interface (data entry pages) can be designed so that they mimic each of the survey forms. Data transcription errors have hopefully been minimized because both the paper and the computer input forms were almost identical. Figure 1 below shows an example of how closely the data entry form for the General Information survey resembles the actual survey form.
Figure 1. FileMaker Pro Data Entry Screen
Although the initial data recording went very well, it was necessary to make sure that the resulting data package was as accurate as possible. So once the data was recorded, a second pass through the survey forms was performed to check for errors. This second pass was not a full review, but rather a sample check of approximately 10% for the survey forms. When reviewing the following tables be aware that everyone did not return a complete set of survey forms. In several cases, respondents also failed to answer all questions. There were also instances where the answer to one question (e.g. the number of dogs listed on the General Information form) did not match the data provided on other forms (e.g. dogs listed on the Health and Illness form). The process of transcribing data from the handwritten paper forms into a computerized database requires an interpretation of each response and there were some responses that had to be reformatted. When transcribing the survey data, a consistent process was used that did not intentionally modify any of the responses while recording data. It should be noted that all numerical values were converted to a consistent unit (e.g. months versus years) for each question.
Transcribing the Health and Disease Survey forms involved a unique set of challenges beyond the obvious difficulty of reading and interpreting the handwritten inputs. The full database file turned over to the Health Trust includes each response as transcribed from the original forms. In order to provide an initial report on the data, the data needed to be organized into categories. Working with Dr. Marcia Dawson, who helped develop the Disease and Illness list which accompanied the 2005 Survey, each response was assigned to one of these categories. The complete data set clearly identifies the original response and the illness category code that it was assigned to. It was impossible to fit some problems into an established category, those problems have been listed as “unclassified” in the following analysis.