SEARCH

SEARCH BY CITATION

Keywords:

  • Behavior;
  • Endocrinology;
  • Feline;
  • Owner;
  • Psychometry

Abstract

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

Background: Success in management of diabetes mellitus (DM) is defined as improvement of blood glucose concentrations and clinical signs. However, the psychological and social impact of DM and its daily treatment regimen on quality of life (QoL) of both animal and owner is uncertain.

Hypothesis/Objectives: To design, validate, and apply a diabetic pet and owner-centered, individualized measure of impact of DM (DIAQoL-pet).

Animals/Subjects: Two hundred and twenty-one owners of insulin-treated diabetic cats were recruited to complete the DIAQoL-pet.

Methods: Discussions and pilot surveys with clinicians and owners of diabetic cats led to the design of 29 specific DM-associated QoL questions. Owners of diabetic cats completed the finalized survey. Each item was scored according to impact frequency and perceived importance. An item-weighted impact score (IWIS) for each item was calculated, as was an average-weighted impact score (AWIS) by averaging all IWISs. Principal component analysis and Cronbach's α calculation assessed the measure's reliability. Two overview questions measured overall QoL and diabetes-dependent QoL.

Results: The DIAQoL-pet showed high reliability (Cronbach α 0.83). The AWIS was −1.76 ± 2.4 (mean ± SD). Areas reported as most negatively impacting QoL included: “boarding difficulties” (IWIS ± SD: −4.67 ± 5.3), “owner wanting more control” (−4.34 ± 4.7), “difficulties leaving cat with friends or family” (−4.21 ± 4.7), “worry” (−4.10 ± 3.9), “worry hypo” (−3.67 ± 3.5), “social life” (−3.48 ± 3.9), “costs” (−3.04 ± 3.8), and “work life” (−3.03 ± 3.7). Forty-one percent of owners believed their cat's life would be “a little better” without DM.

Conclusions and Clinical Importance: The DIAQoL-pet proved robust and identified specific areas most negatively impacting on diabetic cats and their owners' QoL. This tool warrants further investigation for use in clinical or research settings.

Abbreviations
AWIS

average-weighted impact score

DM

diabetes mellitus

DSH

domestic short hair

HBGM

home blood glucose monitoring

IWIS

item-weighted impact score

PCA

principal components analysis

QoL

quality of life

VIN

Veterinary Information Network

Veterinary clinicians often focus on control of blood glucose concentrations in cats with diabetes mellitus (DM). Treatment success is usually defined as obtaining close to normal serial blood glucose concentrations, improving fructosamine or glycosylated hemoglobin values, or both, in conjunction with the resolution or at least amelioration of the key clinical signs generally attributed to DM (polyuria, polydipsia, polyphagia, and weight loss).1 However, evaluating treatment success in diabetic pets should probably also include recording the psychological and social effects of the impact of DM and its relatively complex daily treatment regimen on quality of life (QoL) of both animal and owner, as well as the owner's perceptions of the disorder and its management. Indeed, should an owner perceive a particularly negative impact on his or her QoL or the QoL of their pet as a result of the DM, a decision to cease treatment could be the inevitable outcome regardless of the apparent clinical success. In spite of QoL being acknowledged as an important outcome in its own right in human diabetology,2 assessment of pet and owner QoL has not been reported in any great detail in veterinary diabetology. Additionally, the importance of interacting biopsychosocial factors in the management of chronic disorders is recognized in humans and probably should not be disregarded in animals.3

A diabetic animal and owner-centered, individualized measure of the impact of DM was designed and subsequently applied to describe owner-perceived QoL of their pet with DM and measure their individual feelings about the impact of its diagnosis and treatment on the pet's and owner's everyday lives. This tool, subsequently named DIAQoL-pet, was designed to identify and address specific and concrete areas in life (items) affected by DM and enabled application of a rating of these items in order to qualify and quantify their importance in the individual animal's and owner's lives.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

Design of the DIAQoL-Pet

Qualitative research was conducted to ensure that the items were diabetes centered and valid to diabetic pets and their owners. Detailed discussions were held with veterinary surgeons (n = 21) and nurses (n = 33) (both in primary practice and in referral practice), 2 human DM QoL survey designers, a human diabetology lecturer and consultant, a clinical epidemiologist and owners of diabetic pets (n = 23). Studies using questionnaires for children with diabetes were also used as a basis for the tool's design.2,4

This initial phase allowed the identification of 29 specific DM QoL issues or items, leading to the design of 29 corresponding specific DM-associated QoL questions and corresponding multiple-choice answers (DIAQoL-pet).

Subsequently, the DIAQoL-pet was digitalized and publicized online by the software package Questionmark Perception Manager.a An initial pilot trial was conducted among 19 veterinary surgeons, 11 veterinary nurses, and 9 owners of diabetic companion animals in order to identify areas of confusion and assess the questions' validity. Feedback was used to finalize the DIAQoL-pet, before subsequent application of this final version (Table 1) in the larger population of owners of diabetic cats.

Table 1.   Overview of DIAQoL-pet items and abbreviations.
Item NumberAbbreviationItem
  • Items A and B: separate overview questions; items 1–29: DIAQoL-pet items.

  • a

    Positive items.

  • QoL, quality of life.

AGeneral QoL petI feel that the quality of my pet's life is
BDiabetes-dependent QoL petIf your pet did not have diabetes, his/her quality of life would be:
1WorryDo you worry about your pet's diabetes?
2No treatsDo you ever feel you want to give your pet treats but you don't because of the diabetes?
3Injections restrict lifeDo you feel your life is restricted because of the daily insulin injections?
4Injection painDoes your pet ever react annoyed or in pain when injected?
5Injection worriesDo you ever worry about whether you have given the insulin correctly?
6Resent injectDo you resent having to give your pet insulin injections?
7Restrict your activitiesDo you ever find the diabetes of your pet restricts or limits what you are doing or what you want to do, like going on holidays, away for weekends, away for the day/night, working?
8aExtra thingsDo you ever give your pet extra things, like snacks, treats, extra attention or extra walks because of the diabetes?
9More controlDo you ever feel you want to take more control of the diabetes on your own, without the help from vets and other people?
10Pet's moodsDo you think the diabetes affects your pet's moods?
11Pet unwellDoes your pet ever feel unwell, tired or in any other way negatively affected since treatment with insulin was started?
12Boarding kennelsDo you ever choose not to put your pet into boarding kennels because of the diabetes?
13Friends and familyDo you ever choose not to leave your pet to stay with friends or family because of the diabetes?
14HypoglycemiaDoes your pet ever show signs of a low blood sugar? (eg, wobbliness, collapse)
15Active dayDo you ever choose not to take your pet with you on an active day (eg, walking longer distances, going to the beach, etc.) because of the diabetes?
16DrinkingDoes your pet still drink more than before the diagnosis?
17HungryIs your pet still hungrier than before the diagnosis?
18UrinateDoes your pet still urinate more than before the diagnosis?
19Weight lossIs your pet still losing weight since treatment has begun?
20Future careDo you ever feel worried you will not be able to take care of your pet in the future because of the diabetes?
21Worry hypoDo you ever feel worried about your pet suffering from an episode of low blood glucose?
22Worry DKADo you ever feel worried about your pet suffering from an episode of ketoacidosis?
23Worry visionDo you ever worry about your pet getting vision problems because of cataracts or did you worry about this before your pet suffering from such problems?
24Play lessAre you less inclined to play with your pet now that he/she has diabetes?
25aPlay moreAre you more inclined to play with your pet now that he/she has diabetes?
26Social lifeDo you ever find you need to fit your pet's diabetes into your social life? (eg, carrying needles, food, insulin, providing food on time)
27Working lifeDo you ever find you need to fit your pet's diabetes into your working life? (eg, having to makespecial arrangements when you need to work late or need to start working earlier)
28aSpecial bondDo you feel you have a more special bond with your pet now that you are managing his/her diabetes?
29CostsDo your ever worry about how much money your pet's diabetes costs you and your family?

Description of the DIAQoL-Pet Survey

To ensure an individualized and quantitative character of the tool, each item was scored according to frequency at which it impacted on owner's and pet's lives and how important the item is in the individual owner's and pet's lives (Fig 1).

image

Figure 1.  Example DIAQoL-pet question with corresponding multiple choice answers. All the time = 3, often = 2, occasionally = 1, never = 0. Very important = 4, important = 3, moderately important = 2, low importance = 1, not at all important = 0.

Download figure to PowerPoint

Multiplying frequency and importance ratings for each item provided a so-called item-weighted impact score (IWIS).3 In case of a particular item or issue never occurring in response to questions about the owner's, pet's, or either, life (ie, score 0) this automatically resulted in an IWIS of 0, regardless of the chosen importance rating of the item. Similarly, in case of an item being regarded as “not at all important” by an owner, a score of 0 would also be reached, regardless of a possible high frequency of occurrence of the issue.

Table 1 contains the description of each item and shows the abbreviations that will be used in the remainder of this paper. Most items represent areas of the owner's and pet's lives potentially negatively impacted by the DM and their frequency scores are therefore negative or 0 (range −3 to 0). However, 3 items with possible positive impact were also entered in the DIAQoL-pet and these were therefore given positive frequency scores or 0 (range 0–3).

An average-weighted impact score (AWIS) was calculated to provide a single quantitative measure of diabetes-dependent QoL by dividing all IWISs by the number of items, ie 29.

Two separate overview questions measured current overall QoL, as well as diabetes-dependent QoL (Table 1). These were not included in calculating the AWIS and were analyzed separately.

Additionally, the 3 following hypothetical questions were asked: “If your pet had not been insured for the treatment of diabetes, would you have been willing or able to go ahead with treatment”?; “In the unfortunate event that another one of your pets were to get diabetes, would you treat this pet with insulin too”?; and “Would you recommend friends and family with pets diagnosed with DM to go ahead with insulin injection treatment on the basis of your experiences with your pet”?

Finally, a free comments section served for cognitive feedback and queried if owners had anything else they would like to report about their experiences of life with a diabetic pet receiving insulin injections.

Recruitment of Respondents

The aim was to recruit a large number of owners of diabetic cats whose pets were receiving insulin injections at the time the survey was completed. The survey-based tool DIAQoL-pet was made available online through the URL http://www.rvc.ac.uk/diabetes. In order to prevent malicious participation influencing the results of the surveys, the web link was purposely not advertised openly on the World Wide Web. Diabetic pet owners were only indirectly contacted through their veterinary clinicians and not directly through open-access websites designed for diabetic pet owners. The internet protocol (IP) address of each respondent was recorded with each entry allowing subsequent identification and deletion of duplicate entries by the same IP address, a possible sign of malicious participation. A collaboration was established with the world-wide online veterinary community “Veterinary Information Network” (VINb; http://www.vin.com) allowing the survey links to be sent directly to all VIN members (>40,000). Subsequently, the survey introduction and survey link were also sent to the UK-based Vetsurgeon.org online community (approximately 2,500 members), as well as being advertized through various other electronic and nonelectronic veterinary media outlets, veterinary interest groups, national and international congresses, and continuous education events. Also over 400 veterinary practices in the United Kingdom, Belgium, and the Netherlands were individually approached via email. Finally, within the United Kingdom, the charity practice chain the People's Dispensary for Sick Animals was approached and consented to advertizing the survey among their hospitals. The presented data were voluntarily provided by owners of diabetic cats in the full knowledge that this information facilitated the study of QoL of diabetic pets and their owners; as such, protocol at the Royal Veterinary College does not necessitate additional specific approval of its ethics and welfare committee.

Statistical Analysis

Principal components analysis (PCA) was used to validate the DIAQoL-pet by assessing the unidimensionality of the survey, ie, does it measure diabetes-related QoL or does it measure additional latent variables.5,6 It was also used to identify critical and noncritical items within the scale, thereby providing the possibility of reducing the number of items without negatively impacting on the function of the tool, as well as reassessing the meaning of each item or factor in light of its relation with the latent variable. As part of the PCA, extraction communalities were calculated. Extraction communalities indicate the amount of variance in each variable that is accounted for by all other variables or items. A small value indicates that the item does not correspond well to the other items and should possibly be dropped from the analysis unless further arguments exist to retain the item. Low communalities across the set of items indicate the variables are little related to each other and indicate a poor quality measure.

As an additional component of the PCA, a factor matrix was generated, recording the factor loadings for each variable. The factor loadings represent the correlation between the item and the latent variable. Small values indicate items that do not correspond well to a latent variable. Kline suggested 0.3 as an appropriate cut-off level.6 In the present study inclusion/exclusion decisions were based on a subjective decision advised by a combination of theory behind the item, communalities, loading factors, and reliability analysis.

Cronbach's α, a statistic, commonly used as a measure of the internal consistency of a psychometric instrument, was also determined. Cronbach's α measures how well a set of variables or items measures a single, unidimensional latent construct. An acceptable minimum α can be 0.7–0.8.7

Cronbach's α is defined as

  • image

with N representing the number of components (items), inline imagethe variance of the observed total test scores, and inline image the variance of item i.

Additionally Cronbach's α was calculated when each item was left out of the scale in order to identify items that markedly negatively influence the scale's reliability as a whole.

Finally, corrected item-total correlations were calculated and correlations were thought acceptable if ≥ 0.2.5

Correlations between DIAQoL-pet and answers to general and diabetes-related QoL questions (item A and B, Table 1) were assessed by performing a 2-tailed nonparametric Spearman's ρ, and correlations were considered to be significant when P-value was <.05.

All statistical analyses were performed by the statistical software package SPSS Statistics for Windows 17.0.c

Results

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

Demographical Data

Two hundred and twenty-one owners of insulin-treated diabetic cats completed the survey. Respondents originated from the United Kingdom (n = 49), United States (n = 137), Canada (n = 15), Australia (n = 3), the Netherlands (n = 4), France (n = 4), Germany (n = 3), Finland (n = 1), or unknown (n = 5). The cats' mean (±SD) age was 12.4 ± 3.1 years (range 2–16 years); body weight 5.69 ± 1.9 kg (range 2–15 kg); 98 (44%) were recorded as domestic short hairs (DSH), 39 (18%) as cross breeds, 22 (10%) as domestic long hair, 10 (5%) as Maine Coon, 9 (4%) as British short hair, 4 (2%) as Burmese, 4 (2%) as Siamese, 24 (11%) as “other,” and 11 (5%) as “unknown.” Average insulin dose was 0.55 ± 0.5 IU/kg (range 0.02–4.3) and 182 (89%) received insulin 12 hourly, the remainder, 23 (10%) once daily. Eighty-seven were receiving PZI, 50 were receiving glargine, 42 vetsulin/caninsulin, 27 “other,” 12 Novolin N/humulin N, 9 Insuvet Lente, 2 Humulin L, and 8 “don't know” or blank. Twenty-three (10%) owners reported injecting varying numbers of units each day: 8 of whom specifically reported determining the daily dose of insulin according to daily home blood glucose measurements. One hundred and thirty-two cats were receiving a special diabetic diet (59.7%). One hundred and fifty-three owners (70%) reported “never” checking urine glucose concentration at home, 52 (24%) checked “occasionally,” 10 (5%) “often,” and 3 (1%) checked “all the time.” One hundred and seventy-five owners (58%) reported “never” checking urine ketone concentration at home, 87 (29%) checked “occasionally,” 31 (10%) “often,” and 9 (3%) checked “all the time.” Sixty owners (27%) reported “never” checking blood glucose concentration at home, 26 (12%) checked “occasionally,” 26 (12%) “often,” and 109 (49%) checked “all the time.” Seven cats (3%) were completely covered by insurance for their disease; 7/221 (3%) were partially insured for this condition, while the majority (207/221; 94%) were not insured.

Descriptive Statistics

All negative items confirmed negative weighted impact of diabetes, whereas positive items indicated positive impact of diabetes (Table 2, Fig 2). Areas reported as most negatively impacting QoL (based on IWIS) were: “boarding difficulties,”“owner wanting more control over DM,”“difficulties leaving cat with friends/family,”“worry about cat's DM,”“worry hypo,”“adapting social life,”“DM-related costs,” and “adapting work life.”

Table 2.   Descriptive statistics of the DIAQoL-pet items applied to a population of diabetic cats and their owners.
Item NumberAbbreviation% Never% All the Time% Very ImportantIWIS (Mean ± SD)
  • a

    Positive item; possible range IWIS −12 to 0 (negative items) and 0 to +12 (positive items).

  • IWIS, item-weighted-impact-score.

1Worry8.121.717.2−4.10 ± 3.91
2No treats49.86.313.6−1.58 ± 2.54
3Injections restrict life16.35.47.2−2.26 ± 2.59
4Injection pain53.40.921.3−1.10 ± 1.77
5Injection worries45.22.725.3−1.97 ± 2.60
6Resent inject78.32.312.7−0.61 ± 1.83
7Restrict your activities14.09.012.2−2.81 ± 3.09
8aExtra things30.311.312.2+3.02 ± 3.60
9More control37.121.731.2−4.24 ± 4.66
10Pet's moods23.55.420.4−2.87 ± 2.88
11Pet unwell35.30.528.1−2.32 ± 2.36
12Boarding kennels43.441.236.7−4.67 ± 5.30
13Friends and family33.931.731.7−4.21 ± 4.65
14Hypoglycemia64.70.051.6−1.28 ± 1.88
15Active day86.97.75.9−0.58 ± 2.26
16Drinking32.66.823.5−2.43 ± 2.90
17Hungry39.89.522.2−2.37 ± 3.27
18Urinate28.59.526.7−2.78 ± 3.24
19Weight loss62.91.831.2−1.49 ± 2.44
20Future care52.55.435.7−2.29 ± 3.31
21Worry hypo27.67.245.2−3.66 ± 3.55
22Worry DKA41.66.836.2−2.70 ± 3.32
23Worry vision57.93.618.1−1.75 ± 2.86
24Play less93.70.523.5−0.24 ± 1.18
25aPlay more35.38.120.8+3.07 ± 3.49
26Social life26.716.319.5−3.48 ± 3.93
27Working life33.512.719.5−3.03 ± 3.71
28aSpecial bond12.745.242.1+6.62 ± 4.72
29Costs29.011.819.9−3.04 ± 3.78
image

Figure 2.  Mean item-weighted impact scores (IWIS) of the 29 items of the DIAQoL-pet for diabetic cats and their owners. *Positive items.

Download figure to PowerPoint

The least severe negative impact of diabetes was felt for: “play less,”“active day,”“resent injection,”“injection pain,” and “hypoglycemia.” Diabetes had positive impact on: “special bond,”“play more,” and “extra things.”

Analysis of answers to the separate overview questions revealed that although most owners of diabetic cats (94.5%) rated their cat's QoL as “fairly good” (17%), “good” (26%), or “as good as it could possibly be” (52%), 69% reported a negative impact of the DM on QoL with 8% reporting their pet's life would be “a great deal better,” 20%“quite a lot better,” and 41%“a little better” without the DM (Table 3).

Table 3.   Descriptive statistics of DIAQoL-pet overview items for diabetic cats and their owners.
 Item NameMean ± SDRange
  • a

    Maximal range −3 to 3.

  • QoL, quality of life.

AGeneral QoL pet+1.62 ± 1.38−2 to 3a
BDiabetes-dependent QoL pet−1.55 ± 1.14−3 to 2a

PCA

All items had extraction communalities ≥0.5, indicating all items were reasonably related to each other (Table 4). Factor matrix analysis revealed 5 items (“resent injection,”“more control,”“boarding kennels,”“hypoglycemia,” and “play less”) to have loadings below 0.3 (cut-off suggested by Kline6) (Table 4). However, all 5 factors contributed to overall scale reliability according to Cronbach's α (see below and Table 4), indicating that inclusion would not deter significantly from the validity of the DIAQoL-pet.

Table 4.   Principal components analysis (extraction communalities and loadings) and reliability analysis of the DIAQoL-pet.
Item NumberItem NameCommunalities (Extraction)Factor Matrix (Loading)Corrected Item − Total CorrelationCronbach's α if Item Deleted
  • a

    Positive item.

1Worry0.6050.6900.5740.792
2No treats0.5820.5300.4260.801
3Injections restrict life0.6830.5590.5250.798
4Injection pain0.7170.4870.3870.804
5Injection worries0.6270.6190.5380.797
6Resent inject0.5640.2320.1820.808
7Restrict your activities0.7440.6060.5560.795
8aExtra things0.487−0.354−0.2000.826
9More control0.7130.2520.1640.814
10Pet's moods0.6550.6530.5310.797
11Pet unwell0.7090.5840.4820.800
12Boarding kennels0.5200.2620.2000.815
13Friends and family0.6150.4390.3830.802
14Hypoglycemia0.6570.2200.1850.808
15Active day0.5890.3460.3350.804
16Drinking0.7060.4010.3700.802
17Hungry0.7010.4080.3380.803
18Urinate0.7620.5060.4650.798
19Weight loss0.5650.3480.3130.805
20Future care0.6100.6160.5160.796
21Worry hypo0.6380.6160.5130.795
22Worry DKA0.6400.6440.4770.797
23Worry vision0.5400.4940.3950.802
24Play less0.6410.2840.2070.808
25aPlay more0.934−0.444−0.2030.826
26Social life0.6580.5540.4750.797
27Working life0.6190.6260.5740.792
28aSpecial bond0.934−0.444−0.2030.826
29Costs0.4890.4970.4520.798

Internal Consistency Reliability

Cronbach's α proved satisfactory at a level of 0.83 (N = 221). All corrected item-total correlations were >0.2, apart from factors 6 (“resent injection”), 9 (“more control”), and 14 (“hypoglycemia”), which showed a correlation of 0.182, 0.164, and 0.185, respectively (Table 4). However, deletion of these factors from the scale did not result in an increase in the α coefficient and the factors were therefore left in. In fact, deletion of none of the 29 factors resulted in an increase in α.

AWIS

AWIS was −1.88 ± 2.04, consistent with a negative effect of the DM and its treatment on QoL of cat and owner.

Correlation of DIAQoL-Pet and General Measures of QoL

There was a moderate correlation between DIAQoL-pet AWIS and the overview item A (general QoL pet; ρ= 0.36, P < .001). The correlation with overview item B was higher (diabetes-dependent QoL; ρ= 0.48, P < .001).

Additional Questions

When asked “If your pet had not been insured for the treatment of diabetes, would you have been willing or able to go ahead with treatment”? Three of 221 (1%) of respondents answered “no,” 14/221 (6%) “don't know,” 36/221 (16%) did not answer this question, and 168/221 (76%) answered “yes.”

When asked “In the unfortunate event that another one of your pets were to get diabetes, would you treat this pet with insulin too”? Eighteen of 221 (8%) of respondents answered “probably,” 3/221 (1%) “not sure,” 1/221 (0.5%) “definitely not,” 0/221 (0%) did not answer, and 199/221 (90%) reported “without a doubt.”

Finally, when asked “Would you recommend friends and family with pets diagnosed with DM to go ahead with insulin injection treatment on the basis of your experiences with your pet”? Twenty-three of 221 (10%) of respondents answered “probably,” 4/221 (2%) “not sure,” and 194/221 (88%) “without a doubt.”

Free Comments Section

One hundred and fifty-two owners of diabetic cats provided an entry into the free comments section. Comments mostly emphasized areas already covered by the DIAQoL-pet. Three areas were mentioned by 3 or more respondents and were not directly covered by the DIAQoL-pet. Twenty-three respondents reported lack of veterinary advice or support, as well as poor understanding or knowledge of their veterinarian, particularly with regards to diet and home monitoring of blood glucose. Forty-two owners made additional comments on blood glucose testing at home with 19 owners of insulin-treated diabetic cats specifically reporting being less worried now that they performed such monitoring; while 3 owners of diabetic cats emphasized the burden of the monitoring and stabilization process because of the time and money investment. Eight owners reported finding support on various websites helpful in dealing with the DM of their cat.

Discussion

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

The DIAQoL-pet was developed to quantify the perceived impact of DM and its treatment on QoL of both diabetic pets and their owners. Robustness (validity, unidimensionality, reliability) of the tool was proven by PCA, Cronbach's α assessment and large-scale testing.

Although some items showed lower communality than others, this coefficient should be interpreted carefully. For instance, the communality of “costs” (0.489) in the DIAQoL-pet is one of the lowest of all items, yet proves quite meaningful when looking at its impact on the latent variable (IWIS: −3.04 ± 3.78), indicating necessity of inclusion in the QoL tool. Similarly, some items with relatively low loadings were purposely left in the scale. For instance, the loading of “boarding kennels” (0.262) is relatively low, yet has the highest IWIS (−4.67 ± 5.30) of all items. This is a commonly accepted procedure where arguments exist to retain the low-scoring item. Such arguments usually relate to the theory behind the initial item's inclusion and this is generally deemed to overrule any arbitrary cut-off level of loading scales, as long as the item does not significantly lower the tool's reliability.2

Data obtained in any study should be interpreted in light of the demographics and characteristics of the studied population. Indeed, the majority of cat owners originated from the United States and United Kingdom and represented owners of uninsured diabetic cats. Extrapolation to other populations of diabetic pets with different characteristics should therefore be performed cautiously.

The current study's results could have been influenced by the survey being only available online. The participating owners had to have internet access to complete the survey, thereby potentially lowering the proportion of elderly or underprivileged owners completing the survey. Conversely, the United Kingdom, the United States, as well as all other countries of origin are societies with high internet access availability. Seventy percent of households in the United Kingdom had internet access in 2009 and 74% of the population in North America were online regularly in 2008.8,9 Furthermore, it is likely that owners of diabetic cats who are willing and able to treat their pet are probably more likely to belong to a more privileged socioeconomic subset of society with also higher internet access levels.

The signalment of diabetic cats was similar to recent reports.10,11 Most animals were middle aged to elderly DSH. Most cats were treated with twice daily injections, according to the latest recommendations. A high proportion of owners of diabetic cats (49%) practiced home blood glucose measurement. This relatively high proportion might be explained by this survey being more likely to be completed by pet owners who are more proactive and involved than those who do not necessarily engage as intensively with their cat's DM and leave measurement of blood glucose concentration to their veterinary team.

The top 10 items with the most negative impact are a testament to the importance of recording the impact of treatment on the owner's life in addition to the, usually more attention receiving, animal's QoL. Indeed, 9 of these 10 items were more associated with the owner's QoL than the pet's QoL. Interestingly, items specifically related to injecting insulin did not feature in the top 15 most negatively impacting items. In fact the 1st item of this nature documented the restriction exerted by insulin injections on the owner's life rather than the actual injecting itself, while items such as “injection worries,”“injection pain,” and “resent injections” featured even lower on the list. All items related to classical clinical signs associated with DM (“drink,”“hunger,”“urinate,” and “weight loss”) featured outside the top 10 of most negatively impacting items. This may suggest most animals in the current study were fairly well controlled (judging from the low frequency ratings for these signs). Perhaps their IWIS would have been higher if a group of uncontrolled animals and their owners had been surveyed.

A discrepancy was identified between the perceived high impact of worrying about hypoglycemia (“worry hypo”: −3.67 ± 3.5) and the perceived lower impact of hypoglycemia actually occurring (“hypo”: −1.28 ± 1.88). It is understandable that owners are concerned about the possible occurrence and consequences of overdosing and hypoglycemia, and 35% of owners did report hypoglycemic episodes to possibly have been occurring at least occasionally. This information suggests a cautious approach to the current trend toward application of more aggressive, remission focused, insulin protocols, because, if associated with an increased frequency of hypoglycemia, the associated anxiety could discourage owners from treating the DM.

Additionally, clinicians might, in light of this discrepancy, want to consider to more actively advising owners to further reduce the chances of occurrence of these episodes, as well as to promote owners' confidence in early recognition and adequate management. These measures could markedly reduce anxiety levels among owners and increase QoL. For example, individual owners who experience a major level of anxiety over this particular issue could be introduced to home blood glucose monitoring (HBGM). Entries into the free comments section of the DIAQoL-pet, confirm the anxiety-reducing potential of HBGM, although some owners might in fact perceive this as an unwanted increase in responsibilities.

In recent years, and in contrast with the situation in dogs with DM, various different treatment protocols have been developed in feline DM treatment, instigated particularly by the advent of glargine insulin, HBGM devices and protocols, the increasing popularity and perceived benefits of low carbohydrate diets, and the desire to induce diabetic remission.12–16 Additionally, more and more owners are also suffering from DM, or know people with DM, and therefore notice the difference in treatment approaches in human and veterinary clinical diabetology. Both these developments have led to different schools of thought on the ideal protocol for managing a diabetic cat, both within and outside the veterinary community and have undoubtedly led to seemingly confusing and, at times, contradictory advice being offered to owners by veterinarians, fellow diabetic pet owners, and certain self-help veterinary, as well as nonveterinary, feline diabetes websites and groups. This might well be reflected by both the item “more control” achieving a high score in the DIAQoL-pet (cat) (2nd place; IWIS −4.24 ± 4.66), as well as the relatively frequent negative comments with regards to the “support from the veterinary team” in the free comments section.

In view of the comments made in the free comments section, consideration will be given to include items regarding “support from the veterinary team,”“HBGM,”“monitoring and stabilization process,”“concurrent disease,” and “support from non-veterinary sources, such as websites” into future adaptations of the QoL tool.

In conclusion, the DIAQoL-pet was able to reliably quantify diabetes-dependent QoL of diabetic cats and their owners and identified specific areas impacted by DM and its treatment. The tool could be used by clinicians and diabetic pet owners to assess and seek to improve QoL and treatment success, alongside clinical signs, BG, and fructosamine measurement. Application of the DIAQoL-pet in a population of diabetic dogs and their owners is also planned. The DIAQoL-pet may also prove useful in clinical trials of new diabetic treatment options, alongside measures of biological effect, as is routinely the case in human DM research.17,18

Footnotes

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

aCopyright Questionmark Corporation, London, UK

bVIN, Davis, California

cSPPS Inc, Chicago, IL

Acknowledgment

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

Grants: The work was not supported by an external grant.

References

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References
  • 1
    Briggs CE, Nelson RW, Feldman EC, et al. Reliability of history and physical examination findings for assessing control of glycemia in dogs with diabetes mellitus: 53 cases (1995–1998). J Am Vet Med Assoc 2000;217:4853.
  • 2
    McMillan CV, Honeyford RJ, Datta J, et al. The development of a new measure of quality of life for young people with diabetes mellitus: The ADDQoL-teen. Health Qual Life Outcomes 2004;2:61.
  • 3
    Christie MJ Living with asthma: Contemporary perspectives—An editorial review. In: ChristieMJ, FrenchDJ, eds. Assessment of Quality of Life in Childhood Asthma. Chur, Switzerland: Harwood Academic Publishers; 1994:311.
  • 4
    Wilson RJ, Christie MJ, Bradley C. A qualitative investigation to inform the design of quality of life measures for children with diabetes. Diabet Med 1998;15:122123.
  • 5
    Kline P. A Handbook of Test Construction, 1st ed. London: Routledge; 1993:1211.
  • 6
    Kline P. An Easy Guide to Factor Analysis, 1st ed. London: Routledge; 1994:1139.
  • 7
    Todd C, Bradley C. Evaluating the design and development of psychological scales. In: BradleyC, ed. Handbook of Psychology and Diabetes: A Guide to Psychological Measurement in Diabetes Research and Practice. Chur, Switzerland: Harwood Academic Publishers; 1994:1390.
  • 8
    Office for National Statistics. Internet Access Level USA, Internetworldstats.com. Available at: http://www.statistics.gov.uk/cci/nugget.asp?ID=8. Accessed December 16, 2009.
  • 9
    Internetworldstats.com. Internet Access Level USA, Internetworldstats.com. Available at: http://www.internetworldstats.com/stats.htm. Accessed December 16, 2009.
  • 10
    Rand JS, Fleeman LM, Farrow HA, et al. Canine and feline diabetes mellitus: Nature or nurture ? J Nutr 2004;134:2072S2080S.
  • 11
    McCann TM, Simpson KE, Shaw DJ, et al. Feline diabetes mellitus in the UK: The prevalence within an insured cat population and a questionnaire-based putative risk factor analysis. J Feline Med Surg 2007;9:289299.
  • 12
    Rand J. Editorial: Glargine, a new long-acting insulin analog for diabetic cats. J Vet Intern Med 2006;20:219220.
  • 13
    Reusch CE, Kley S, Casella M. Home monitoring of the diabetic cat. J Feline Med Surg 2006;8:119127.
  • 14
    Weaver KE, Rozanski EA, Mahony OM, et al. Use of glargine and lente insulins in cats with diabetes mellitus. J Vet Intern Med 2006;20:234238.
  • 15
    Slingerland LI, Fazilova VV, Plantinga EA, et al. Indoor confinement and physical inactivity rather than the proportion of dry food are risk factors in the development of feline type 2 diabetes mellitus. Vet J 2009;179:247253.
  • 16
    Hall TD, Mahony O, Rozanski EA, et al. Effects of diet on glucose control in cats with diabetes mellitus treated with twice daily insulin glargine. J Feline Med Surg 2009;11:125130.
  • 17
    Al Mazroui NR, Kamal MM, Ghabash NM, et al. Influence of pharmaceutical care on health outcomes in patients with type 2 diabetes mellitus. Br J Clin Pharmacol 2009;67:547557.
  • 18
    Holbrook A, Thabane L, Keshavjee K, et al. Individualized electronic decision support and reminders to improve diabetes care in the community: Compete II randomized trial. Can Med Assoc J 2009;181:3744.