EBM SPECIAL TOPIC

Outcomes Article

Survey Research
Amy K. Alderman, M.D.,
Summary: Survey research is a unique methodology that can provide insight
M.P.H. into individuals’ perspectives and experiences and can be collected on a large
Barbara Salem, M.S.W., M.S. population-based sample. Specifically, in plastic surgery, survey research can
Ann Arbor, Mich. provide patients and providers with accurate and reproducible information to
assist with medical decision-making. When using survey methods in research,
researchers should develop a conceptual model that explains the relationships
of the independent and dependent variables. The items of the survey are of
primary importance. Collected data are only useful if they accurately measure
the concepts of interest. In addition, administration of the survey must follow
Downloaded from [Link] by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 10/14/2021

basic principles to ensure an adequate response rate and representation of the

intended target sample. In this article, the authors review some general
concepts important for successful survey research and discuss the many
advantages this methodology has for obtaining limitless amounts of valuable
information. (Plast. Reconstr. Surg. 126: 1381, 2010.)

T
he basic concept of survey research involves Survey research is a unique methodology that
capturing beliefs, attitudes, or outcomes that can provide insight into individuals’ perspectives
can be generalized to a population from and experiences. Successful outcomes in plastic
which the sample was selected. The sample can surgery are often measured by improvement in a
include any population of interest, such as physi- patient’s quality of life rather than by mortality
cians, health care administrators, patients, and pa- rates, which are used by other surgical areas. As
tients’ significant others. The outcomes of interest such, survey research is a valuable tool for plastic
in surgery can range from national epidemiolog- surgeons to consider when interested in under-
ical trends in surgical care to physicians’ beliefs standing the impact of surgery on patient-re-
about the surgical management of a disease to ported outcomes. With the large variety of surgical
patient-reported surgical outcomes. For example, techniques available in plastic surgery, choosing
suppose that you are interested in rheumatolo- the best procedure for a particular patient can by
gists’ attitudes toward hand surgery for rheuma- a daunting task, even for experienced surgeons
toid arthritis. A random sample of several hun- and highly educated patients. Patient-reported
dred rheumatologists in the United States could outcome measures collected through survey re-
be mailed a survey that addresses the specific search can provide patients and physicians with
attitudes you want to study. The responses from reliable and useful information to assist in this
the rheumatologists are then coded into a stan- decision-making process. In particular, patient sat-
dardized form that can be recorded electroni- isfaction and health-related quality of life data of-
cally in a quantitative manner. Responses are fer patients a means of evaluating and comparing
then subjected to an aggregated analysis to options based on previous patients’ experiences
describe the attitudes of the rheumatologists and perspectives. Furthermore, survey research is
toward hand surgery and determine correla- a reliable approach to obtaining data on what is
tions among different responses. Conclusions effective, not just efficacious, by examining pa-
reached are then generalized to the entire U.S. tient-reported outcomes on a large sample of pa-
population of rheumatologists.1–3 tients treated by a many different surgeons in dif-
ferent treatment settings.
From the Section of Plastic Surgery, Department of Surgery, Our purpose is to describe the following re-
The University of Michigan Medical Center; and Division garding survey research: (1) the clinical question
of General Medicine, Department of Internal Medicine, Uni-
versity of Michigan.
Received for publication November 9, 2009; accepted Jan-
uary 15, 2010. Disclosure: The authors have no conflicts of inter-
Copyright ©2010 by the American Society of Plastic Surgeons est to declare.
DOI: 10.1097/PRS.0b013e3181ea44f9

[Link] 1381
 Plastic and Reconstructive Surgery • October 2010

that is suitable for survey research, (2) key ele- significant others, friends, and news media. The
ments for conducting a high quality survey, (3) survey could be piloted on a small sample and
advantages and disadvantages of this methodol- the feedback used to refine the measures for the
ogy, (4) how to report survey results, and (5) in- final survey instrument.
terpreting and deriving evidence from survey re-
search. WHAT ARE THE ESSENTIAL ELEMENTS
FOR CONDUCTING A GOOD SURVEY?
WHAT TYPES OF CLINICAL QUESTIONS
ARE BEST SUITED FOR SURVEY 1. Conceptual Model
RESEARCH? The most important step in survey research is
Clinical questions best suited for survey re- the development of the survey, which should be
search include questions that are: (1) descriptive, based on a conceptual framework that explains
(2) explanatory, or (3) explorative.3 One use for the relationships of the independent and depen-
clinical research is to make descriptive assertions dent variables. The conceptual model is a working
about a population. In this case, the clinical ques- strategy containing the major concepts and their
tion is not directed at why an observed distribution relationships. The model formulates the research
exists but rather what the distribution is.3 For ex- questions and hypotheses.5 Researchers can mod-
ample, you may be interested in knowing how ify preexisting conceptual models or develop their
many mastectomy-treated breast cancer patients own. An example would be a researcher who was
receive breast reconstruction. In this example, interested in studying the patient decision-making
one is not interested in patient and system-level process for postmastectomy breast reconstruction.
factors influencing the use of postmastectomy The first step is to design a conceptual model to
breast reconstruction, but rather the research help explain the decision-making process. In this
question is limited to describing epidemiological case, the researcher could use a conceptual model
trends in receipt of surgery treatment. from social science research called the Transtheo-
Survey research can also be used to make ex- retical Model, Stages of Change Construct (Fig. 1).6
planatory assertions about a population. This re- This model proposes that patients move through
search design almost always requires the use of a series of steps or stages when making decisions
multivariate analysis, which is the simultaneous and taking actions about breast reconstruction:
examination of two or more variables.3 An exam- precontemplation (lack of knowledge of or desire
ple would be if you were interested in knowing why for breast reconstruction); contemplation (think-
the use of postmastectomy breast reconstruction ing about reconstruction); preparation (action-
across the United States was low with large geo- oriented activities, such as presurgical consulta-
graphical variations. A survey could be designed to tion); and action (receipt of reconstruction). Each
study general surgeons’ referral patterns for post- box represents a concept that must be measured
mastectomy breast reconstruction. A multivariate in the survey, such as (1) knowledge and attitudes
analysis could be performed that looked at the toward reconstruction; (2) external influences,
association between surgeon and treatment facil- such as family and friends; and (3) enabling fac-
ity characteristics and referral practices for tors, such as health insurance. Some concepts do
reconstruction.4 not have previously validated measures, such as
Survey research can also be explorative, for knowledge and attitudes toward breast recon-
which the study question is less defined than in the struction. In this case, ad hoc questions must be
previous examples. In this case, it can be used to designed. When available, however, validated
initiate an inquiry into a particular area. A loosely measures should be used, such as when assessing
structured survey can be designed in the initial patients’ decisional satisfaction,7 decisional regret,8
stages of the development of a research project, and health-related quality of life.9 Issues related
and the results can be used to inform and sub- to measure validation are discussed in the next
stantially revise the research design and survey section.
instrument.3 An example would be if you were
interested in better understanding outside fac- 2. Survey Items
tors that influence a patient’s decision regard- The questionnaire items are of primary im-
ing postmastectomy breast reconstruction. A portance in survey research. Collected data are
loosely structured survey could be designed that only useful if it accurately measures the concepts
asked about a variety of influences on the sur- of interest. In other words, a good question-and-
gical decision, ranging from health providers, answer process is one that produces answers that

1382
Volume 126, Number 4 • Survey Research

Fig. 1. Conceptual model.

provide meaningful information about what is trying Questionnaire for hand function18 and the
to be described.10 In addition, the measurement pro- Breast-Q for quality of life and patient satisfac-
cess must produce consistent results.10 When devel- tion after breast surgery19 (see Table 1 for ad-
oping a survey research project, a researcher is con- ditional examples20 –33).
fronted with a set of abstract concepts that are It is important to realize that these instruments
thought to help explain a clinical area of interest. lose validity if they are changed in any way. If
These abstract concepts must be converted into modifications are deemed necessary, the instru-
questions in a survey instrument to collect empirical ment must undergo repeat validation testing. In-
data that can help the researcher better under- vestigators should resist the temptation to use in-
stand the clinical question.3 Accurately designing house, nonvalidated instruments for assessing
an instrument is a rigorous process that involves patient-reported outcomes. This practice de-
several stages, such as item generation, item re- creases the research quality and eliminates the
duction, pretesting, field management, and at- ability to compare the results with other studies.11
tribute testing.11–13 For more information on sur- Sometimes the researcher is interested in under-
vey development, we recommend reading Health standing concepts that do not have previously val-
Measurement Scales: A Practical Guide to Their Devel- idated measures. In these cases, ad hoc questions
opment and Use by David Streiner and Geoff Nor- can be designed by the researcher but must be
man, Oxford Medical Publications, 2003. This carefully reviewed by the research team along with
thorough process of instrument development pilot testing to ensure that the questions accu-
helps ensure measurement quality by testing the rately measure the concept of interest.
reliability (i.e., consistent responses) and valid-
ity (i.e., corresponds to the “true value”) of the
items.3,10 When possible, it is best to use mea- 3. Question Construction
sures that have been previously validated with Several guidelines can be helpful when devel-
proven reliability, validity, and responsiveness to oping questions in the case where previously val-
change.14 –17 Examples of validated patient-re- idated instruments are not available. We recom-
ported measures include the Michigan Hand mend reading: Improving Survey Questions: Design

1383
 Plastic and Reconstructive Surgery • October 2010

Table 1. Examples of Validated Instruments for Survey Research*

Health Condition Instrument
Hand-related conditions Australian/Canadian Hand Osteoarthritis Index (AUSCAN) Osteoarthritis Hand Index20
Carpal Tunnel Questionnaire (CTQ)21
Disabilities of the Arm and Shoulder (DASH)22
Michigan Hand Questionnaire (MHQ)23
Breast-related conditions BREAST-Q19
Functional Assessment of Cancer Therapy-Breast Quality of Life Instrument (FACT-B)24
European Organization for Research and Treatment of Cancer Quality of Life
Instrument, European Organization for Research and Treatment of Cancer (EORTC)
Quality of Life Questionnaire (QLQ)-C30 and Breast Cancer Module, QLQ-BR2325,26
Health-related quality of life Body Image and Relationships Scale27
Sexual Function Scale28
EuroQol-5D in burn patients29
Medical Outcomes Short Form-36
Decisional quality in medical O’Connor’s Decisional Regret Scale8
decision-making Rovner’s Satisfaction with Decision Scale7
Optimism/coping styles Life Orientation Test30,31
Brief COPE Instrument32
Beck Hopelessness Scale33
*This is only an example of validated survey instruments and is not meant to be an inclusive list of available questionnaires.

and Evaluation by Floyd J. Fowler, Jr. (Sage Pub- whether there is an association between the var-
lications, 1995), and The Survey Kit, which is a series ious items. For example, rather than asking if a
of set books edited by Arlene Fink (Sage Publica- respondent regrets having surgery because of
tions, 1995; especially “How to Conduct Self-Ad- complications (this is both double barreled and
ministered and Mail Surveys” by Linda B. Bourque causal), one could ask a series of questions about
and Eve P. Fielder) for a thorough discussion of satisfaction with various aspects of the surgery
effective survey questions. We will address a few of and another series of questions about the oc-
the more salient points here. When possible, currence of various problems that could have
closed-ended questions are preferred because been associated with the surgery. Table 2 pro-
they provide greater uniformity of responses and vides examples of poorly stated questions.
are more easily analyzed. Open-ended questions Normalizing statements can be an important
can be very difficult to code for data entry and tool for encouraging respondents to answer ques-
analysis.3 When using closed-ended questions, tions accurately rather than answering with per-
however, the responses must be both exhaustive ceived socially desirable responses. For example,
and mutually exclusive.3 Double-barreled ques- in a physician survey about surgeon–patient com-
tions should also be avoided.3 Look for this situ- munication, a series of questions about potential
ation when the word and is present in the ques- areas of conflict could be introduced with the
tion. An example would be: “A patient with stage following: “Involving your patients in treatment
4 breast cancer should be treated with a mastec- decisions can be difficult for many reasons. We
tomy and should delay breast reconstruction.” Sur- want to know your opinions about the challenges
vey items should be clear, short statements without to actively engaging your patients in treatment
bias. Negative items should also be avoided.3 Avoid decision making.” The appearance and language
questions that ask about more than one thing at level of the survey is also significant. Wording of
a time. For example, rather than asking: “Did the questions should be aimed at the lowest edu-
you get information about treatments such as cation level of the anticipated respondent popu-
radiation and chemotherapy?” these should be lation. The layout should be clean, directions easy
asked as two separate questions. Questions that to follow, and pages should not be too crowded so
ask respondents to report on other people’s be- as to avoid overwhelming respondents and lower-
liefs or experiences should also be avoided as ing their likelihood of participating.
respondents do not tend to be accurate report- A Likert scale is the psychometric scale most
ers for others. Researchers should also avoid commonly used to measure response options in
questions that ask respondents to assign causal- survey research. The scale measures a respon-
ity to a particular situation. Instead, the com- dent’s level of agreement with a statement.34 Scales
ponents to be studied should be asked sepa- can vary in the number of response options pro-
rately, and the researcher can then calculate vided. A five-point scale is most commonly used,

1384
Volume 126, Number 4 • Survey Research

Table 2. Examples of Poorly Worded Questionnaire Items

Questionnaire Item Correction
Double-barreled questions
How much do you agree with the following statement? How satisfied are you with the overall results of your surgery?
I regret having surgery because of the complications. How much of a problem was the length of your recovery?
How much of a problem were surgical complications?
How much of a problem was scarring?
Please answer true or false to the following: I am Do you provide immediate reconstruction? (Y/N)
unable to provide immediate reconstruction because If no: Why don’t you provide immediate reconstruction?
of conflicts with scheduling and coordination of Please check all that apply.
care. Scheduling conflicts
Difficulty with coordination of care
Negative questions
Please answer true or false to the following: Plastic Please answer true or false to the following: Plastic surgeons
surgeons should not be required to offer procedures should be able to choose whether or not to offer
with limited reimbursement. procedures with limited reimbursement.
Questions with ambiguous wording
Have you had another episode of breast cancer since Have you had another episode of breast cancer since your
your diagnosis? (Y/N) If yes, where was it found? diagnosis? (Y/N) If yes, on what part of the body was it
found?
How accessible is your doctor? How long does it take to get to your doctor’s office?
Are you able to reach your doctor on the telephone when
you have a question?
How long do you wait to see your doctor when you have a
doctor’s appointment?

although a four-point scale could be administered cern for question order bias, multiple versions of
if the researchers want to force the respondent the survey can be developed with different order-
into a level of agreement or disagreement, as the ing of items and piloted on a sample to determine
middle neutral response is eliminated.35 Table 3 the effects of the item order.3 It is also important
displays a variety of commonly used responses for to consider placing the most important items early
the Likert scale. in the questionnaire in case the respondent does
not complete the entire survey.
4. Ordering Questions in the Survey
The order of questions in the survey can 5. Pilot Studies
greatly influence responses. An example would be Conducting pretests of the study administra-
if the first question in a survey asked about phy- tion, survey, and analysis is extremely important.
sicians’ attitudes toward reimbursement for post- Before survey administration, intensive individual
mastectomy breast reconstruction. Any subse- interviews should be performed with the survey
quent questions, such as “What is the most questions to evaluate the responder’s understand-
important health policy issue in plastic surgery?” ing of the questions.10 For example, a researcher
would be influenced by the first financial question is interested in understanding rheumatologists’
related to breast reconstruction. The safest way to management of rheumatoid hand disease. Previ-
guard against such potential bias is to be sensitive ously validated measures are not available. The
to the issue. Random ordering of questions creates researcher designs questions aimed at under-
a chaotic survey that respondents will not want to standing physicians’ beliefs about medical versus
complete. If the researchers have substantial con- surgical treatment. The researcher should then

Table 3. Commonly Used Responses for Five-Point Likert Scales

Responses with a neutral middle option and equal but opposite extremes at each end
1 Strongly disagree Disagree Neither agree nor Agree Strongly agree
disagree
2 Very dissatisfied A little dissatisfied Neither satisfied nor A little satisfied Very satisfied
dissatisfied
Responses that begin at one extreme and increase at approximately equal intervals to the opposite extreme
3 Poor Fair Good Very good Excellent
4 Not at all A little Somewhat Quite a bit A lot
5 Never Rarely Some of the time Usually Most of the time
6 Not at all likely A little likely Somewhat likely Quite likely Very likely

1385
 Plastic and Reconstructive Surgery • October 2010

have individual rheumatologists take the survey ings (an original and two follow-ups) produce the
and explain what they believe each question is best response rates. Time between mailings is 2 to
asking. This process helps ensure the content va- 3 weeks3,41 The follow-up mailings can be limited
lidity of the survey items. A representative sample to a letter of encouragement for participation.
of the target population should be administered The most effect method, however, is to enclose
the final survey. Data collection, cleaning, coding, a new copy of the survey with the incentive gift,
and analysis should representative of the final re- along with follow-up letters and phone calls.
search methods and analysis. The survey can then Other ways that have been shown to improve
be revised based on information received about response rates include university sponsorship,
missing data, variances in responses, and internal short questionnaires, personalized letters, re-
validation along with respondent feedback on corded delivery with stamped return envelopes,
question clarity and questionnaire flow and increased number of contacts with the respon-
format.3 Pretesting does add time and cost to the dent, and an advance letter notifying the re-
research; however, valuable information will be spondent that a survey will follow.3,39,42
gained through this process, and the final deliv- Achieving a high response rate is an essential
erables from the research will be greatly improved. goal of survey research. The higher the response
rate, the lower the chance for a nonresponse bias
and the greater likelihood that the responses rep-
6. Survey Administration resent the target population. In general, a re-
Surveys can be administered in several ways, sponse rate of at least 50 percent is considered
such as telephone interviews and electronic mail. adequate for analysis and reporting. A response
Electronic methods have the advantage of lower rate of 60 percent is considered good, and 70
costs and faster survey administration, and can be percent is considered very good.3 Response rates
easier for the respondent when skip patterns are are generally calculated by dividing the number of
present because the questionnaire can be elec- completed survey by the net sample size. The net
tronically formatted so that the respondent only sample size is derived from the initial sample size
sees the questions they are supposed to answer.36 minus subjects that could not be administered a
Examples of email survey software include Survey- survey due to death, bad addresses, etc.3
Monkey, SurveyTracker, QuestionPro, and Sur-
veyShare. Many, however, are concerned about 7. Importance of a Statistician
sample representativeness due to exclusion of par-
ticipants from lower socioeconomic status that Consider consulting with a statistician early in
might not have access to a computer.37 Response the process of questionnaire development and
rates are often lower with electronic surveys com- study design. Statisticians can help assess the re-
pared with mailed surveys.38 This review article will liability of ad hoc questions from pilot question-
focus on self-administered mailed surveys, which naires. They can assist with sample size calcula-
are one of the most common distribution meth- tions to ensure that you have enough statistical
ods. The basic method for self-administered power to analyze the data appropriately. It is also
mailed surveys is to send the respondent the ques- important to discuss how the measures will be
tionnaire along with a letter of explanation and a analyzed and reported before survey administra-
return envelope with either business-reply postage tion so that adjustments can be made as appro-
or stamps.3 When surveying a group, it can be priate in questionnaire response categories.
beneficial to privately ask members of the group
whose opinion they respect most regarding the WHAT ARE THE ADVANTAGES AND
research topic. Response rates may improve if you DISADVANTAGES OF THIS
have that person explain in the cover letter why METHODOLOGY?
the responder’s opinion is important. The outgo- Survey research has many advantages. A lim-
ing survey should have an identification number itless amount of desirable and useful information
that will link the questionnaire to the respondent, can be collected from large population-based sam-
which will allow for follow-up mailings to the non- ples. The information can be used to describe
responders. Often an incentive is provided to en- characteristics of a large population, which few
courage response rates, such as a small cash gift or other research methodologies can provide. For
gift card.39,40 Returns must be closely monitored, example, survey research can allow for the exam-
and follow-up mailings to nonresponders must ination of surgical outcomes across multiple pro-
proceed in a timely fashion. In general, three mail- viders and multiple health systems. This popula-

1386
Volume 126, Number 4 • Survey Research

tion-based approach provides useful information between the covariates and the dependent variable
on what treatment is effective in the general pop- of interest, and (3) regression analyses that shows the
ulation under “real world” circumstances rather independent association of covariates on the depen-
than what is efficacious under ideal treatment con- dent variable while controlling for other important
ditions, such as those in a clinical trail.43 In addi- clinical and demographic characteristics that may
tion, if a large sample size is used, this will increase confound the results.
the statistical power of the study and allow for the The methods section of the manuscript should
analysis of multiple covariates. Surveys can also be also include detailed information that is specific to
administered from remote locations. Use of stan- survey research, such as: (1) the survey instrument;
dardized or previously validated measures will en- (2) sampling technique, inclusion and exclusion cri-
sure precise and reliable measurement and allow teria, and power analysis when appropriate; (3) sur-
for comparisons with other studies. vey administration, including incentive gifts and
Survey research has some disadvantages. This number of contacts; (4) response rate along with
method can be costly, depending on the incentive descriptive analyses between responders and nonre-
gift and postage. The use of previously validated sponders; (5) analytic coding for variables (e.g., con-
instruments may be too general for the research tinuous, categorical); and (6) a well-described ana-
questions of interest. For example, the Short Form lytic process that relates the variables of interest to
Health Survey measures general health states not the study’s hypotheses.
specific enough for most surgical outcomes. The
accuracy of responses can also be an issue in survey HOW DOES ONE INTERPRET AND
research. Although self-reporting is consistently DERIVE EVIDENCE FROM SURVEY
more accurate than proxy-reporting,10 the accu- RESEARCH?
racy of self-reporting can be affected by time to
One must critically appraise the study’s sam-
recall an event, inadequate knowledge regarding
pling techniques, respondent population, and re-
a topic, and the respondent’s willingness to report
porting methods when evaluating the validity of
on a given topic.10 For example, recall is more
the results. How representative is the study sample
likely to be accurate the more recent the event and
to the population of interest? Would the nonre-
the greater impact of the event.10,44 When survey
sponders’ answers to the survey vary significantly
instruments obtain information subject to recall
from the answers by those who completed the
bias, statistical methods should be employed to
questionnaire? Did the study subjects have ade-
help limit this bias. These methods include con-
quate knowledge to accurately answer the ques-
trolling for the time between the event and the
tions? Has there been too much time between the
survey completion and controlling for factors that
event in question and the time to complete the
may affect an individual’s ability to recall infor-
survey that might result in inaccurate responses
mation, such as age. Medical diagnosis is a good
from the study subjects? It is also important to
example of inadequate knowledge, for which sev-
consider whether the instruments used were pre-
eral studies have shown a mismatch between the
viously validated or ad hoc questions. In cases of
medical conditions patients report and the con-
ad hoc questions, is there adequate information
ditions recorded in the medical record.10,45 Lastly,
about how the questions were tested for validity
but most importantly, the validity of the study re-
and reliability in the study population of interest?
sults is highly dependent on the response rate, as
Lastly, it is important that the results address the
a poor response rate can result in a nonresponse
study’s original hypotheses and that the analytic
bias.
process adequately controlled for possible con-
HOW SHOULD SURVEY RESULTS BE founding factors through regression techniques.
REPORTED?
It is beyond the scope of this article to address SUMMARY
in detail the process for data analysis. How to Report Survey research is a unique methodology that
Statistics in Medicine by Thomas Lang and Michelle can provide valuable insight into individuals’ per-
Secic (American College of Physicians, Philadel- spectives and experiences, and can be collected on
phia, 1997) provides a comprehensive summary of a large population-based sample. The information
how to report statistical information for clinical from survey research in plastic surgery can provide
research. In general, the analytic portion of the patients and providers with accurate and repro-
manuscript should provide: (1) general descriptive ducible information to assist with medical deci-
data of the study sample, (2) bivariate comparisons sion-making.

1387
 Plastic and Reconstructive Surgery • October 2010

Amy K. Alderman, M.D., M.P.H. 18. Chung KC, Hamill JB, Walters MR, Hayward RA. The Mich-
Plastic Surgery igan Hand Outcomes Questionnaire (MHQ): Assessment of
University of Michigan responsiveness to clinical change. Ann Plast Surg. 1999;42:
2130 Taubman Center 619–622.
1500 East Medical Center Drive 19. Pusic AL, Reavey PL, Klassen AF, et al. Measuring patient out-
Ann Arbor, Mich. 48109-0340 comes in breast augmentation: Introducing the BREAST-Q
aalder@[Link] Augmentation module. Clin Plast Surg. 2009;36:23–32, v.
20. Bellamy N, Campbell J, Haraoui B, et al. Clinimetric prop-
erties of the AUSCAN Osteoarthritis Hand Index: An eval-
ACKNOWLEDGMENT uation of reliability, validity and responsiveness. Osteoarthritis
This work was supported by a career development Cartilage 2002;10:863–869.
award from the Robert Wood Johnson Foundation to Dr. 21. Levine DW, Simmons BP, Koris MJ, et al. A self-administered
Alderman. questionnaire for the assessment of severity of symptoms and
functional status in carpal tunnel syndrome. J Bone Joint Surg
Am. 1993;75:1585–1592.
REFERENCES 22. Changulani M, Okonkwo U, Keswani T, Kalairajah Y. Out-
1. Alderman AK, Ubel P, Kim H, Fox D, Chung K. Surgical come evaluation measures for wrist and hand: Which one to
management of the rheumatoid hand: Consensus and con- choose? Int Orthop. 2008;32:1–6.
troversy among rheumatologists and hand surgeons. J Rheu- 23. Chung KC, Hamill JB, Walters MR, Hayward RA. The Mich-
matol. 2003;30:1464–1472. igan Hand Outcomes Questionnaire (MHQ): Assessment of
2. Alderman AK, Chung K, Kim H, Fox D, Ubel P. Effectiveness responsiveness to clinical change. Ann Plast Surg. 1999;42:
of rheumatoid hand surgery: Contrasting perceptions of 619–622.
hand surgeons and rheumatologists. J Hand Surg (Am.) 2003; 24. Brady MJ, Cella DF, Mo F, et al. Reliability and validity of the
28:3–11. Functional Assessment of Cancer Therapy-Breast quality-of-
3. Babbie E. Survey Research Methods. 2nd ed. Belmont, Calif.: life instrument. J Clin Oncol. 1997;15:974–986.
Wadsworth; 1998. 25. Niezgoda HE, Pater JL. A validation study of the domains of
4. Alderman AK, Hawley ST, Waljee J, et al. Correlates of re- the core EORTC quality of life questionnaire. Qual Life Res.
ferral practices of general surgeons to plastic surgeons for 1993;2:319–325.
mastectomy reconstruction. Cancer 2007;109:1715–1720. 26. Hjermstad MJ, Fossa SD, Bjordal K, Kaasa S. Test/retest study
5. Riegelman R. Studying a Study and Testing a Test. Philadelphia: of the European Organization for Research and Treatment
Lippincott Williams & Wilkins; 2000.
of Cancer Core Quality-of-Life Questionnaire. J Clin Oncol.
6. Glanz K, Lewis FM, Rimer BK. Health Behavior and Health
1995;13:1249–1254.
Education. San Francisco: Jossey-Bass; 1997.
27. Hormones JM, Lytle LA, Gross CR, Ahmed RL, Troxel AB,
7. Holmes-Rovner M, Kroll J, Schmitt N, et al. Patient satisfac-
tion with health care decisions: The satisfaction with decision Schmitz KH. The Body Image and Relationships Scale: De-
scale. Med Decis Making 1996;16:58–64. velopment and validation of a measure of body image in
8. Brehaut JC, O’Connor AM, Wood TJ, et al. Validation of a female breast cancer survivors. J Clin Oncol. 2008;26:1269 –
decision regret scale. Med Decis Making 2003;23:281–292. 1274.
9. Sprangers MA, Groenvold M, Arraras JI, et al. The European 28. Lasry J. Effect of Cancer on Quality of Life. Boca Raton, Fla.: CRC
Organization for Research and Treatment of Cancer breast Press; 1991.
cancer-specific quality-of-life questionnaire module: First re- 29. Oster C, Willebrand M, Dyster-Aas J, et al. Validation of the
sults from a three-country field study. J Clin Oncol. 1996;14: EQ-5D questionnaire in burn injured adults. Burns 2009;35:
2756–2768. 723–732.
10. Fowler FJ. Improving Survey Questions: Design and Evaluation. 30. Scheier MF, Carver CS. Optimism, coping, and health: As-
London: Sage; 1995. sessment and implications of generalized outcome expect-
11. Bindra RR, Dias JJ, Heras-Palau C, et al. Assessing outcome ancies. Health Psychol. 1985;4:219–247.
after hand surgery: The current state. J Hand Surg (Br.) 2003; 31. Svelnikas K. Optimism and Pessimism as Regulators of Self Mo-
28:289–294. tivation [unpublished bachelor’s thesis]. Tartu, Estonia: Univer-
12. Meadows KA. So you want to do research? 5: Questionnaire sity of Tartu; 1998.
design. Br J Community Nurs. 2003;8:562–570. 32. Carver CS. You want to measure coping but your protocol’s
13. Rattray J, Jones MC. Essential elements of questionnaire too long: Consider the brief COPE. Int J Behav Med. 1997;
design and development. J Clin Nurs. 2007;16:234–243.
4:92–100.
14. Dias JJ, Bhowal B, Wildin CJ, Thompson JR. Assessing the
33. Beck AT, Weissman A, Lester D, Trexler L. The measure-
outcome of disorders of the hand: Is the patient evaluation
ment of pessimism: The hopelessness scale. J Consult Clin
measure reliable, valid, responsive and without bias? J Bone
Psychol. 1974;42:861–865.
Joint Surg Br. 2001;83:235–240.
15. Angst F, Goldhahn J, Pap G, et al. Cross-cultural adaptation, 34. Likert R. A technique for the measurement of attitudes. Arch
reliability and validity of the German Shoulder Pain and Dis- Psychol. 1932;140:1–55.
ability Index (SPADI). Rheumatology (Oxford) 2007;46:87–92. 35. Dawes J. Do data characteristics change according to the
16. Bot SD, Terwee CB, van der Windt DA, et al. Clinimetric number of scale points used? An experiment using 5-point,
evaluation of shoulder disability questionnaires: A systematic 7-point, and 10-point scales. Int J Market Res. 2008;50:61–77.
review of the literature. Ann Rheum Dis. 2004;63:335–341. 36. Bachmann D, Elfrink J. Tracking the progress of email versus
17. Cohen J. Applied Multiple Regression/Correlation Analysis for the snail-mail. Market Res. 1996;8:31–35.
Behavioral Sciences. Hillsdale, N.J.: Lawrence Erlbaum Asso- 37. Dillman D. Mail and Internet Surveys: The Tailored Design
ciates; 1983. Method. New York: John Wiley & Sons; 2000.

1388
Volume 126, Number 4 • Survey Research

38. Andreson S, Gansneder BM. Using electronic mail surveys and 42. Fox R, Crask MR, Kim J. Mail survey response rates. Public
computer monitored data for studying computer mediated Opin Q. 1988;52:467–491.
communication systems. Soc Sci Comput Rev. 1995;13:33–46. 43. Wennberg DE, Lucas FL, Birkmeyer JD, et al. Variation in
39. Edwards P, Roberts I, Clarke M, et al. Increasing response rates carotid endarterectomy mortality in the Medicare popula-
to postal questionnaires: Systematic review. BMJ. 2002;324:1183. tion: Trial hospitals, volume, and patient characteristics.
40. Halpern SD, Ubel PA, Berlin JA, Asch DA. Randomized trial JAMA. 1998;279:1278–1281.
of 5 dollars versus 10 dollars monetary incentives, envelope 44. Cannel C, Marquis K, Laurent A. A Summary of Studies.
size, and candy to increase physician response rates to mailed Vol. 69. Washington, D.C.: Government Printing Office;
questionnaires. Med Care 2002;40:834–839. 1977.
41. Hoddinott S, Bass MJ. The Dillman Total Design Survey 45. Cannell C, Fisher G, Bakker T. Reporting of Hospitalization in
Method: A Sure-Fire Way to Get High Survey Return Rates. the Health Interview Survey. Vol. 6. Washington, D.C.: Govern-
Can Fam Physician 1986;32:2366–2368. ment Printing Office; 1965.

Instructions for Authors: Key Guidelines

Manuscript Length/Number of Figures
To enhance quality and readability and to be more competitve with other leading scientific journals, all
manuscripts must now conform to the new word-count standards for article length and limited number of
figure pieces:
• Original Articles and Special Topics/Comprehensive Reviews are limited to 3000 words and 20 figure
pieces.
• Case Reports, Ideas & Innovations, and Follow-Up Clinics are limited to 1000 words and 4 figure pieces.
• Letters and Viewpoints are limited to 500 words, 2 figure pieces, and 5 references.

1389