NHIS - Injury and Poisoning Information
Overview of the Injury and Accident Questions
Historical Note: NHIS used the term “accident” throughout most of this period to refer to all types of incidents that were not intentional. Although this term is no longer used in our reports, for the sake of simplicity, this term is used here generically.
From its inception, the NHIS focused on conditions and impairments, including those caused by injury, and the effects of these conditions on the U.S. population. Beginning with the first NHIS in 1957, questions on accidents and injuries and their effects were included in the core questionnaire each year and, in some years, these questions were expanded into both minor and substantial supplements to the NHIS core. There were no questions asked specifically about poisoning, although injury caused or inflicted by a “poisonous substance swallowed” was included as a response category from the beginning.
The definition of what constituted an injury prior to 1997 differs from the more recent years. Prior to 1997, data were collected on injury conditions occurring in the two weeks prior to interview that met at least one of two criteria: requiring medical attention or causing at least one disability day, such as a bed day, in the previous two weeks. Disability days are days on which the person cuts down on his or her usual activities for at least half of the day because of illness or injury. Disability days occurring in the previous two weeks due to injuries were included in estimates, regardless of the date of the accident. For example, a broken leg occurring a month prior to the interview that still restricted going to work in the past two weeks was included in the disability days estimates but not in the estimates of injuries to persons.
1957-1968 (Fiscal Years)
Although the NHIS first went into the field in July of 1957, no microdata files exist before 1963; however, copies of the questions in these early years still exist and are included in the Injury and Poisoning Questions section of this report. In addition, there are reports (i.e. NCHS Publications Series B and C reports) that focus on injury related data from these early years such as: numbers of persons injured, persons injured by class of accident, and persons injured in motor vehicle accidents and associated disability.
Core conditions: During 1957-1968, the core NHIS used the so-called “condition” approach; data were collected on all conditions, both chronic and acute, that caused or were related to disability days, doctor visits, or hospital visits, including visits to an emergency room. Although injuries were part of the acute condition focus, based on a two week reference period prior to the interview, the survey designers’ interest in acute conditions and effects was demonstrated by including questions in the condition section specific to injuries. “In accordance with the condition approach, the questionnaire was designed to elicit reports of all acute conditions experienced during the 2-week period prior to the week of interview…” “The questions, designed primarily to encourage the reporting of acute conditions, were concerned with sickness, accidents, or the utilization of medicine during the 2 weeks preceding the interview.” (1)
“Additional detailed questions were then asked about each condition reported in response to the probe questions. These additional questions were aimed at obtaining the best diagnostic description of the condition, at determining if the condition was medically attended, and at obtaining information on the number of short-term disability days (restricted activity, bed-days, time loss from work or school) and the onset of this condition.” (1)
Each condition mentioned within the interview generated a condition record specific to that condition. Within the condition section of the questionnaire each year, and on the condition records, there were questions specific to injuries. During the interview, when any condition was identified, the interviewer had to determine whether or not the condition was the result of an injury in order to determine if the extra injury questions needed to be asked. If the interviewer did not already know based on the interview that the condition was caused by an accident, an additional question was asked to determine whether or not the condition resulted from an injury. As an example, an arthritic knee may have resulted from a war injury many years before. In 1959 and 1960, the injury questions were more detailed and were divided into motor vehicle and non-motor vehicle sections with some questions asked about all injuries. From 1961 through 1968 (using the condition approach), the core questions eliciting injury conditions were substantially unchanged.
More detailed information on these early years as well as the definitions of terms used in reports can be found in the appendix of each year’s Current Estimate report.
1968 (Calendar Year) – 1981
Beginning with calendar year 1968, the core NHIS was revised to reflect what was termed the “person” approach and the time period covered by the survey changed to coincide with the calendar year. This new approach was intended to “build a person-data foundation and then generate the condition information…questions are asked to determine if the person had…suffered any disability or if he had received medical attention during the 2 weeks prior to the week of the interview. If a person…had some short-term disability days, or has sought medical attention, the condition or conditions causing these phenomena were then obtained” (1). As before, a specific question on injuries in the two weeks preceding the interview was asked. Injury conditions were treated like other conditions, but once injuries were identified, additional questions were asked about the circumstances and details of the incident causing the injury. Each injury condition with a different ICD code was recorded in a separate condition record even if multiple injuries occurred to the same person in the same incident.
The follow-on questions and response categories for injuries were slightly modified during this period. There were occasional extra questions in selected years. As an example, in 1971, if a motor vehicle was involved, there were questions about what the respondent was doing at the time of the accident (outside the vehicle, getting in/out, a passenger, the driver) and what types of vehicles were involved.
The questions regarding the “current effects” of injuries as well as other conditions remained essentially the same during this time period as did the questions specific to injuries occurring during the previous two-week period. However, after the revision of the NHIS questionnaire in 1982, the specific questions relating to injuries “from an accident or other cause” appeared much later in the questionnaire than previously.
In the condition section of the NHIS core questionnaire, injury conditions were treated like other conditions; however, once injuries were identified, respondents were asked additional questions about the circumstances and details of the incident causing the injury conditions. These questions remained essentially unchanged throughout 1982-1996.
One of the earliest supplements to the NHIS was on motor vehicle accidents and injuries. In 1968, the questions generating the separate motor vehicle person file concerned involvement in a “motor vehicle accident either as a driver, passenger or pedestrian” in the previous 12 months, whether or not the person was injured. Only those records pertaining to motor vehicle accidents occurring in the previous six months were kept in the file. Subsequent research, however, indicated that more accurate estimates could be made using only the previous three months of accidents – specifically those occurring during the 91 days prior to interview (2).
In 1975, there was a set of extra questions for any accidents or injuries occurring to family members within the previous six months that received medical attention or caused disability days. Questions were asked about a list of 14 categories of injuries, including gunshot wound, suffocation, electric shock, poisoning from swallowing, breathing, or coming in contact with a poisonous substance, and, for the first time, a reaction to medication or cosmetics. Data were placed in the 1975 NHIS Accident file.
Several supplements to the NHIS addressed injuries that had occurred in specific settings such as the workplace. There were also supplements that included questions on the prevention of injuries of various types. For example, in 1985, 1990, and 1991, questions on child safety addressed safety seats and seat belt use. Car safety questions were asked of adults in 1985, 1990, 1991, and 1993. Questions on workplace safety issues were asked in 1985, 1988, and 1992. In 1994, questions were asked about firearm safety in the home. In 1992, the NHIS also fielded a survey called the Youth Risk Behavior Survey for persons aged 12-21 years; eight of those questions were related to behavior likely to result in injury.
Family Core Injury Section: 1997-1999
Starting in 1997, a redesigned NHIS questionnaire was fielded that contained a section devoted entirely to injuries and poisonings. The Family Core Injury Section (FIJ) asked detailed questions about medically attended injuries and poisonings that occurred to any member of the family within the previous three-month reference period. Responses to these questions provided information about who in the family was injured and/or poisoned, the number of times the person was injured and/or poisoned, the cause of the injury or poisoning episode, how the injury happened, the body part injured, the type of injury, what the person was doing at the time of the injury episode, the date and place of occurrence, the elapsed time between the date of the injury or poisoning episode and the date of the interview, whether the person was hospitalized, whether the person missed any days from work or school due to the injury or poisoning, and whether the injury episode caused any limitation of activity. All injury and poisoning information was provided by the family respondent. This section remained unchanged from 1997 to 1999.
Family Core Injury Section: 2000-2003
As with the 1997-1999 NHIS, the Family Core Injury Section of the 2000 survey included questions about medically attended injury and poisoning episodes that occurred to any member of the family within a three-month reference period. All injury and poisoning information continued to be provided by the family respondent.
However, in 2000, the Family Core Injury Section of the NHIS was redesigned. These changes included combining the injury and poisoning questions, revising response categories for some existing questions, adding a question about where medical advice or treatment was received, adding a question about animal/insect bites, and deleting questions about drowning and/or water-related injuries and firearm-related injuries. Specific questions about drowning and/or water-related injuries and firearm-related injuries were no longer included in the survey because they were rare events and it was not possible to calculate reliable estimates given such small frequencies.
When changes to the section were made in 2000, a key question was inadvertently reworded. The question asking respondents “How many different times in the past three months were you injured or poisoned seriously enough to seek medical advice or treatment?” was mistakenly changed to “How many times in the past three months did you seek medical advice because you were injured or poisoned?”. As a result, respondents gave the number of times they sought medical advice for an injury or poisoning during the three months prior to the interview, rather than the number of times they were injured or poisoned during the three months prior to the interview. In the 2001 survey, the wording for this question was changed back to the wording that was originally intended in 2000. Except for this question, this section remained unchanged between 2000 and 2003.
Family Core Injuries & Poisoning Section: 2004-present
For data years 2004 and beyond, the section of the NHIS that included questions about injuries and poisonings was called the Family Core Injuries & Poisoning Section. This section included questions about injuries and poisonings that occurred to any member of the family within a three-month reference period for which medical providers were consulted. All injury and poisoning information continued to be provided by the family respondent.
However, in 2004, many changes were made to the Family Core Injuries & Poisoning Section of the NHIS. Questions that previously required verbatim responses (body part injured and kind of injury) now had specific response categories. For people who did not know the exact month, day, and year when the injury/poisoning occurred, additional questions were asked about the date of the injury/poisoning. Several questions (type of vehicle that hit pedestrian, cause of burn, type of animal bite, and whether the injury/poisoning caused any limitation of activity) were eliminated from the survey. In addition, several questions had response category changes.
Injury and Poisoning Prevention Questions: 1997-present
In addition to the injury and poisoning questions found in the Family Core Injury Section and the Family Core Injuries & Poisoning Section of the 1997-present NHIS, questions pertaining to injury and poisoning prevention were occasionally added to the survey via supplements or as additional questions inserted in the Sample Adult Core and the Sample Child Core. In 1998, the NHIS included an Adult Prevention Module and a Child Prevention Module that contained questions about preventing injuries and poisonings. In the Adult Prevention Module, these questions covered the topics of seat belt usage, working smoke detectors in the home, testing of the home for lead and radon, the availability and use of workplace information on injury prevention and prevention of back injuries, family discussions about safety and things you can do to prevent injuries, and firearm safety. In the Child Prevention Module, these questions covered the topics of child safety seat and seat belt usage and the use of mouth guards and protective headgear when participating in organized sports. In 2002, both the Adult Core and the Child Core contained a question about wearing eye protection when participating in activities that can cause eye injury. The 2007 the Sample Adult questionnaire contained two questions on the use of ear plugs or ear muffs when exposed to loud sounds.