Obstructive sleep apnea: screening and home testing news

- Kenny Lin, MD, MPH

According to a recent article in AFP, obstructive sleep apnea (OSA) is present in 2 to 14 percent of the general adult population, with a higher prevalence in older and obese persons. Most people are unaware of their diagnoses, either because they do not recognize symptoms or do not report them to physicians. Since it is hard to make an asymptomatic person feel better, is there any good reason to screen for OSA in asymptomatic adults? Screening advocates suggest that treating patients with moderate to severe OSA with continuous positive airway pressure (CPAP) reduces hypoxic episodes that could trigger cardiovascular events in patients with known vascular disease. A POEM in the January 15th issue summarized a randomized trial that seemed to refute this hypothesis. After almost 4 years of follow-up, the group that received CPAP reported slightly less daytime sleepiness, but had the same frequency of cardiovascular events as the control group.

The U.S. Preventive Services Task Force (USPSTF) considered this study and others in issuing a new recommendation statement on January 24th that concluded "the current evidence is insufficient to assess the balance of benefits and harms of screening for OSA in asymptomatic adults." The USPSTF found inadequate evidence that validated questionnaires (such as STOP-BANG) accurately identify who will benefit from polysomnography (PSG) in asymptomatic populations (as opposed to those with suspected OSA). The Task Force also could not determine if CPAP or mandibular advancement devices improve health outcomes (mortality, cognitive impairment, motor vehicle crashes, and cardiovascular or cerebrovascular events) other than sleep-related quality of life.

Although an insufficient evidence statement is not necessarily a recommendation to not screen, Drs. Sachin Pendharkar and Fiona Clement argued in an editorial in JAMA Internal Medicine that the costs of screening for OSA (not considered by the Task Force) warrant clinicians doing just that. Based on the sensitivity and specificity of one widely used screening tool, and an OSA prevalence of 26% in the Medicare population, the authors estimate that $21 billion would be wasted on negative PSG tests, or $4.4 billion if less expensive home-based sleep studies were used instead.

On a related note, the comparative accuracy of home-based tests versus laboratory PSG in diagnosing OSA has been an actively studied topic. A 2014 practice guideline from the American College of Physicians recommended that portable sleep monitors (limited-channel sleep studies) only be used to diagnose OSA when PSG was not available. However, a randomized non-inferiority trial published last week in Annals of Internal Medicine found that patient outcomes after limited-channel studies were similar to those after PSG. This finding may be a blow to the for-profit sleep testing industry, but it is undoubtedly good news for our patients.

After emergency contraception: what next?

- Jennifer Middleton, MD, MPH

Developing a regular, ongoing contraception plan when women request emergency contraception (EC) makes intuitive sense, and the updated Centers for Disease Control and Prevention (CDC)'s Practice Recommendations for Contraceptive Use, as described in the January 15 issue of AFP, includes several points for physicians to consider when doing so. One important discussion point involves the risks and benefits of simultaneously providing a ulipristal (ella) prescription and initiating long-acting hormonal contraceptive methods.

Women desiring EC in the U.S. currently have three oral medication options, in addition to the copper IUD, to choose from: the Yupze method and levonorgestrel are approved up to 72 hours after unprotected intercourse, and ulipristal is approved up to 120 hours after unprotected intercourse. Patients requesting EC are often willing to initiate a regular contraceptive method at the same visit. Initiating hormonal contraception at the same time as levonorgestrel or the oral contraceptives used in the Yupze method poses no drug-drug interaction risk, but how ulipristal's antiprogestin effect might impact outcomes is less clear.

Hormonal contraceptive methods, regardless of delivery mode (oral, implant, or IUD) may decrease ulipristal's efficacy, and, conversely, ulipristal may also decrease the initial efficacy of a regular hormonal method. The CDC recommends waiting at least 5 days after taking ulipristal before beginning a hormonal contraceptive method. This delay, however, can be inconvenient for women and can increase the risk of them not initiating a regular contraceptive method at all. Discussing these risks and benefits with patients at the time of providing EC is a must.

Ulipristal has definite positives; it's the most effective oral medication for EC, it only requires one dose, and it works up to 5 days after unprotected intercourse. The potential negative of these interaction risks, however, drives the CDC to encourage transparent discussion with patients. Patient-centered decision making is one framework well-suited to such conversations:

The health care provider's role includes provision of information, facilitating the identification of patient preferences, ensuring that preferences are not based on misinformation, helping patients to think about how their preferences relate to the available options, and coming to a mutually acceptable decision.

Women want their preferences included in discussions of contraceptive choice, and they also want to have the final decision in what method they will use. Discussions about EC should include options for initiating a regular form of contraception along with information about ulipristal's effectiveness and possible interactions. 

Family physicians should not dismiss ulipristal as an option for EC given its convenience and efficacy, but considering the possible decreased effectiveness of both ulipristal and whatever new contraceptive method patients choose is important. Providing this information to women in the context of patient-centered decision making will allow them to choose both an EC method and a regular contraceptive method that best fit their priorities and wishes. If you'd like to read more about ulipristal, there's an AFP STEPS article that outlines its use, and this Update on Emergency Contraception describes use of the Yupze method, levonorgestrel, and the copper IUD. There's also an AFP By Topic on Family Planning and Contraception that contains in-depth information about a variety of contraceptive methods.

How do you counsel women about EC?

What's in a name? Obesity, ABCD, and prediabetes

- Kenny Lin, MD, MPH

A recent position statement from the American Association of Clinical Endocrinologists and the American College of Endocrinology proposed replacing obesity with the term "adiposity-based chronic disease," or ABCD for short. The authors argued that this new term emphasizes that most persons with obesity will struggle with weight gain for their entire lives; encourages a complications-centric as opposed to body mass index-based management approach; and "avoids the stigmata [sic] and confusion" associated with obesity in popular culture. They also asserted that ABCD is more amenable to interventions based on the Chronic Care Model, which explicitly recognizes that screening and office-based management need to be adapted to the patient's unique environment.

None of these concepts should surprise family physicians, though, and after reading through the AACE/ACE statement, I was not sold on the benefits of the new term. Some patients with body mass indexes above 30 don't like the obesity label, but would they respond any more positively to the disease acronym ABCD? There are potential harms to consider, too. One of my AFP physician colleagues felt that the new term was "intimidating" and "not at all patient centered," while another thought that it "only hides the issue [of obesity] instead of confronting it."

This discussion brought to mind another medical term often associated with overweight and obese patients: prediabetes. On one hand, being classified as "prediabetic" or at risk for this exceptionally common diagnosis may motivate obese patients to lose weight through improved diet and physical activity. On the other, the term prediabetes is misleading: many of these patients will not develop diabetes, and the diagnostic accuracy of the most common screening tests (hemoglobin A1c and fasting glucose levels) is poor, according to a systematic review published in the BMJ. Due to the tests' low sensitivity and specificity, some persons are incorrectly diagnosed with prediabetes, and others who might actually benefit from interventions to prevent diabetes are falsely reassured. Therefore, the review authors concluded, "'screen and treat' policies alone are unlikely to have substantial impact on the worsening epidemic of type 2 diabetes."

For all its limitations, obesity is a diagnosis with well-established clinical utility. It is less clear how many patients have been helped (or harmed) by being diagnosed with prediabetes. With more study, adiposity-based chronic disease might someday become a useful term, but the current case for more widespread use is unconvincing.

Guest Post: I have a new patient

- Donna J. Schue, MD

I've been in a busy rural private practice for 15 years. Outside of new obstetrical patients, the children I deliver, and the occasional close relative or close friend of an existing patient, my practice has been full.

I received a phone call on a Saturday from one of our Emergency Department physicians. Our practice was listed that day to provide an ED follow-up visit for those patients who were seen and had no physician. My colleague had just seen a delightful and independent 80 year-old woman who had come to the ED, not having seen a physician for over 25 years. She had discovered a breast lump about two years before. She had witnessed her husband die of cancer 15 years before despite all attempts at treatment, and had decided that she would continue to have good days until she wasn't having good days anymore. The ED evaluation had revealed an obvious large breast cancer, extensive ascites, and a large pleural effusion. A CT scan showed scattered bony metastases. She did not want to see an oncologist and did not want to be admitted to the hospital, but knew she would need care in the months to come. I told him I would accept her as a patient and saw her in my office on Monday.

Now, four months later, my new patient has undergone a thoracentesis and several paracenteses. She is beginning to have discomfort related to metastases, and we are managing her pain. She has completed a Medical Order for Life-Sustaining Treatment form and prepared her legal affairs. She has also continued to live independently, attend daily Mass, volunteer at a hospice home and a clothing donation center, and has visited out of town family three times. She has told no one of her diagnosis. She does not want those around her to worry. She tells me repeatedly that she is grateful that it is winter and her sweaters can hide the changes to her body that would be more obvious otherwise.

The clinician in me wonders what might have been the outcome if she had presented to care earlier. However, having had two patients die in the last year from opportunistic infections while immunocompromised due to chemotherapy, I remind myself that treatment does not guarantee longer survival.

We talk at each visit about telling her family. As I lost my own mother due to side effects of cancer treatment a few years ago, I share with her that as a daughter I would want to know that this was happening to my mother. She agrees that at some point she will tell them, but not yet. She mentions that she may need help with that when the time comes.

I remain impressed by her fortitude to continue her daily routines. I realize, again, that sometimes we family physicians are called to comfort and not cure. I see how filling her remaining days by helping others continues to bring her a sense of purpose. I have learned a great deal from her in a short time and am grateful that I accepted a new patient.