Lesson 5: Public Health Surveillance
Public health importance of chlamydia
Estimated to be 2.8 million new cases each year in the United States.
Approximately 40% of infected, untreated women experience pelvic inflammatory disease. Five-fold increase in risk among women of experiencing HIV infection, if exposed.
Mortality caused by chlamydia
Ectopic pregnancy, a potential complication of chlamydial infection, can cause death, but frequency is unknown.
Complications of chlamydial infections among women have impact on their reproductive ability and can cause chronic illness among certain women, resulting in an undue burden on them, their families, and the health-care system.
Passed person to person through sexual contact or from mother to baby during birth.
Potential for an outbreak
Varies with population sexual activity and practices, as well as underlying prevalence.
Public perception and concern
Not described in fact sheet provided. Different readers might have differing perceptions of the level of public concern.
Ability to prevent, control, or treat chlamydia
Preventable by sexual abstinence, sexual contact with uninfected partners, and use of latex male condoms.
Control measures and treatment
Secondary prevention through annual screening for chlamydia, which is recommended for all sexually active women aged ≤25 years. An annual screening test is also recommended for women aged ≥25 years who have risk factors for chlamydia (e.g., a new sex partner or multiple sex partners). All pregnant women should have a screening test for chlamydia. Chlamydia is highly responsive to antibiotic treatment. Sexual partners should be evaluated, tested, and treated, if infected.
Capacity of health-care system to implement control measures for Chlamydia
Speed of response
Disease is often asymptomatic, resulting in delayed diagnosis. Annual screening for chlamydia is recommended for all sexually active women aged ≤25 years and for women aged ≥25 years who have risk factors for chlamydia.
Treatment is typically through the health-care system, and the costs are paid by insurers, employers, or the government. Follow-up of patients to identify contacts is the responsibility of the health department. Given the frequency of the disease, this might require substantial resources that are not be available in certain places.
Availability of resources
Dependent on location.
What does surveillance for this event require?
Screening and diagnosis of men and women with chlamydial disease and then reporting of disease by health-care providers to the state health department by using a standard form. The percentage of women who actually receive recommended screening is unknown. Surveillance can also be conducted by using reporting of positive diagnostic tests by laboratory facilities.
- Surveillance provides an estimate of the true prevalence of this important but often overlooked condition.
- Infection is treatable, and transmission is preventable.
- Untreated chlamydial infection is a major cause of pelvic inflammatory disease and infertility.
- Surveillance can be conducted through routine laboratory reporting of all positive tests for chlamydia, which might reduce the reporting burden on health-care providers.
- Clinicians might ignore the requirement to report chlamydia, even if it is added to the list of notifiable diseases, if they believe the list is already too long. They might believe they should only be required to report communicable diseases with statistically significant morbidity or mortality that can lead to immediate intervention by the health department.
- Clinicians might not adhere to screening recommendations, and therefore, recognition of disease might be low.
- Adding chlamydia to the list will not lead to better diagnosis and treatment, because the majority of infections are asymptomatic.
Asthma is a chronic illness that can vary in severity. Using just one source of data or just one dataset to monitor it provides limited knowledge of its extent and the potential effect of treatment and other interventions on it. Thus, using multiple sources of data with information on asthma's incidence, prevalence, morbidity, and mortality is the best way to conduct surveillance for this illness.
- Self-reported asthma prevalence or attacks provides information on its occurrence among the entire population, even those who might not seek or receive medical care for it.
- The majority of cases of asthma requiring medical attention are observed in physician offices, emergency departments, or outpatient clinics. Thus, obtaining information from these sources provides optimal knowledge of its occurrence and morbidity among the majority of persons.
- Severe episodes of illness can require hospitalization and be an indicator that routine treatment in outpatient settings is not being delivered effectively to the whole population. Thus, data on hospitalizations caused by asthma is helpful in monitoring effectiveness of interventions.
- Deaths from asthma are similar to hospitalizations and might represent a failure of the health-care system to deal effectively with the illness.
In addition to the usefulness of different sources, as described previously, certain advantages and disadvantages of different methods of gathering data from these sources are described in the following sections.
- More control over the quality of the data.
- More in-depth data possibly collected on each case than is usually possible with notifications.
- Can identify the spectrum of illness, including cases that do not warrant medical care.
- More accurate assessment of true incidence and prevalence.
- More costly to perform because surveys usually require development of de-novo data-collection systems and hiring of interviewers who require training and supervision.
- Might represent only a single point in time ("snapshot''), if survey is not periodically repeated; might miss seasonal trends, rare diseases, or rapidly fatal diseases.
- Recall bias more likely to affect results because data collected retrospectively (notifications are usually prospective).
Notifications (Reporting of illness by health-care providers)
- Cheaper (for the health department).
- Typically use existing systems and health-care personnel for collecting data.
- Allows monitoring of trends over time.
- Ongoing data collection might allow collection of an adequate number of cases to study those at risk. With surveys, an event might be too infrequent to gather enough cases for study; with notifications, the observation period can be extended until sufficient numbers of cases are collected.
- Might not provide a representative picture of the incidence or prevalence unless care is taken in selecting reporting sites and ensuring complete reporting.
- Data that can be collected are limited by the skill, time, and willingness of the data collectors, who usually have other responsibilities.
- Quality control might be a major problem in data collection.
- The quality of data might vary among collection sites.
- As a result, notifications usually provide a substandard estimate of the true incidence and prevalence.
An alternative to notification might be to enroll interested and appropriate health-care providers and clinics in a sentinel system to gather case numbers of asthma.
Factors that influence the choice of one source of data or one dataset over another include severity of illness (e.g., hospitalization and mortality); need for laboratory confirmation of diagnosis; rarity of the condition; specialization, if any, of the health-care providers who commonly examine patients with the condition under surveillance; quality, reliability, or availability of relevant data; and timeliness of the data in terms of need for response.
Listeriosis: A wide spectrum of nonspecific clinical illness and a low case fatality rate exists (except among newborns and immunocompromised persons). Therefore, surveillance should be based on morbidity rather than mortality data; diagnoses should be confirmed in the laboratory. Possible sources of surveillance data include laboratory reports, hospital discharge data (although patients with listeriosis are often not hospitalized), or adding listeriosis to the notifiable disease list.
Spinal cord injury: This is a severe health event with substantial mortality; almost all persons who sustain a spinal cord injury are brought to a hospital. Therefore, surveillance would most logically be based on hospital records and mortality data (e.g., death certificates or medical examiner data). Special efforts might be directed to obtaining data from regional trauma centers. Using data from emergency medical services and rehabilitation centers might also be explored.
Lung cancer among nonsmokers: Similar to spinal cord injury, lung cancer is a severe health event with high morbidity and mortality. Unfortunately, hospital discharge records and vital records do not routinely provide smoking information. For this condition, cancer registries might provide the best opportunity for surveillance, if smoking information is routinely collected. Alternatively, surveillance might be established by using interested internists, oncologists, and other health-care providers likely to interact with lung cancer patients.
Possible explanations for the sudden increase include those listed in the following. Each possibility should be investigated before deciding that the increase is a true increase in incidence.
- Change in surveillance system or policy of reporting.
- Change in case definition.
- Improved or incorrect diagnosis.
- New laboratory test.
- Increased physician awareness of the need to test for tuberculosis, new physician in town, and so forth.
- Increase in publicity or public awareness that might have prompted persons or parents to seek medical attention for compatible illness.
- New population subgroup (e.g., refugees) in state A who have previous recent vaccination against tuberculosis using the bacille de Calmette-Guérin (BCG) vaccine.
- New or untrained staff conducting testing for tuberculosis and incorrect interpretation of skin reaction to tuberculin.
- Increase in reporting (i.e., improved awareness of requirement to report).
- Batch reporting (unlikely in this scenario).
- True increase in incidence.
No right answer exists, but one set of tables for health department staff might be as follows:
Table 1 addresses disease occurrence by place. Tables 2 and 3 address disease occurrence by time. Together, these tables should provide an indication of whether an unusual cluster or pattern of disease is occurring. If such a pattern is detected, person characteristics might then be explored.
A report for health-care providers does not need to be distributed as frequently and does not need to include all of the notifiable diseases. One approach might be to distribute a report every 6 months and include notifiable diseases that have demonstrated substantial change since the last report, with a discussion of possible causes for the change. Maps of the geographic distribution and figures illustrating the trends over time of selected diseases might be more appealing and informative to health-care providers than tables of frequencies. Information on the diagnosis and treatment of highlighted diseases might also be of interest to health-care providers.
Reports for the media and public typically should be issued to inform them of outbreaks, of new diseases, or of diseases of particular concern. These reports should include basic information about the diseases, the location and frequency of their occurrence, and information on recognition, prevention, and treatment of the diseases.
State health department newsletters do not always go to all those who have a need to know. Even among those who receive the newsletter, some do not read it, and many others skim the articles and ignore the tables. In addition, depending on the timing of the laboratory report and publication deadlines, the information might be delayed by weeks or months.
This information about finding rabid raccoons in a residential area is important for those who might be affected and for those who might be able to take preventive measures, including the following:
- Other public health agencies (e.g., neighboring local health departments or animal control staff) — Contact and inform by telephone or e-mail message.
- Health-care providers serving the population in the affected area — Contact and inform through a special mailing.
- Veterinarians — Inform through a mailing so that they can be on alert for pets that might have come into contact with rabid wildlife; veterinarians can provide specimens, as appropriate, of both wild animals and pets to the state laboratory for testing for rabies.
- The public — Inform by issuing press release to the media asking the public to avoid wild animals and to have their pets vaccinated.